Impacts of local circulations and their interactions on ozone in Suzhou on 5 June 2019 are investigated using surface observations and three-dimensional air quality simulations. It is found that cities on the southeastern shore of Lake Taihu, such as Suzhou, Jiaxing, and Huzhou, have higher ozone concentrations compared with other cities in the Yangtze River Delta, although their emission levels are relatively low. Local circulations and their interactions are found to be the primary causes of the episode. The coupling of lake-land breeze circulation (LLBC), sea breeze circulation (SBC), and urban heat island circulation (UHIC) in regions on the southeastern shore of Lake Taihu such as Suzhou and Jiaxing create wind convergence zones, influencing the spatiotemporal distribution of ozone in the boundary layer. During the daytime, the LLBC plays a dominant role in maintaining the wind convergence zone. It transports pollutants to the surrounding regions of Lake Taihu, resulting in reduced pollution concentrations over the lake and worsen air quality in surrounding areas. The UHIC decouples with SBC at noon, leading to a stronger SBC, thus contributing to a reduction in pollution in upwind regions but hinders downwind diffusion. The UHIC enhances the convergence zones of secondary circulations, making it easier for the accumulation of ozone pollutants in areas between Lake Taihu and lakeside cities.

The health effects of air pollution have become more important and apparent in recent years. Exposure to air pollutants is associated with increased mortality hospitalization rates. However, it should be noted that the health effects of air pollution may vary in different regions. Here we focus on the health risks of air pollution in specific areas and the underlying attribution analysis. Considering that Day-to-day (DTD) temperature variability is an important characteristic of air temperature, which significantly affects human health, we examined the long-term trends of DTD over the past 26 years in the United States, and its association with changing air pollution. By using the observed data and The Coupled Model Intercomparison Project Phase 6 (CMIP6) multi-model simulations, We demonstrate that the positive trend of the DTD index in US can be attributed to the anthropogenic aerosols, while the negative trend of which can be attributed to the natural forcing and greenhouse gas forcing. The observed DTD enhancement over 1997-2022 is dominated by the effect of anthropogenic aerosols, while natural forcing and GHGs partially counteract the effect of anthropogenic aerosols. Based on climate modeling experiments, we demonstrate that the reduced aerosol emissions in US can contribute to the enhanced trend of DTD in USA, Which may indicate that the effect of improved air pollution on health risks in US is not entirely positive.

We employed a modified the Cavity Attenuated Phase Shift NO₂ monitor and Chemical Ionization Mass Spectrometer, coupled with the eddy covariance method, to quantify NO_x and ClNO₂ fluxes in urban Beijing. We assessed the temporal and spatial emission characteristics of NO_x, and evaluating the anthropogenic NO_x emission inventory. Through ClNO₂ flux measurements, we also analyzed the impact of deposition on ClNO₂ concentration and yield. Our findings highlight NO_x flux exhibiting spatial and temporal heterogeneity in urban Beijing. Comparison results indicate that the current emission inventory overstates NO_x emission in the city of Beijing. For ClNO₂, our study reveals a significant influence of deposition on its concentration and yield.

Atmospheric amines are gaining more and more attention in the field of atmospheric chemistry owing to their important roles in new particle formation and growth. In this study, aerosol aminiums over a coastal city (Shanghai) and the Yellow and East China seas (YECS) were characterised. The concentrations of ammonium, dimethylaminium (DMAH) and trimethylaminium plus diethylaminium (TMDEAH) over Shanghai were all found to be higher in the winter of 2018 than in the summer of 2019, suggesting their non-negligible terrestrial contributions. DMAH and TMDEAH concentrations over the YECS in summer were closely correlated and linked to surface phytoplankton biomass, implying that marine biogenic sources might be a predominant contributor to aminiums at this time. Aminiums over Shanghai generally showed a bimodal distribution with a main peak in droplet mode and a secondary peak in condensation mode, suggesting the notable contribution of aqueous-phase or heterogeneous reaction to the formation of aminiums. In contrast, aminiums over the YECS often showed a unimodal distribution, which may be caused by the competition between amines and NH3 for reaction with acidic compounds. We estimated the contributions of marine biogenic sources, 73.6% to DMAH and 80.1% to TMDEAH over the YECS, using methanesulfonate/non-sea-salt SO₄^2– as an indicator. Our results suggest that marine biogenic emission of amines from China’s marginal seas may have a potential impact on coastal cities, and this source should be considered in modelling new particle formation and air quality in coastal areas.

Atmospheric ammonia emissions are deleterious due to their role in multifaceted environmental pollution. It has been demonstrated that volatile ammonia (NH₃), through interacting with sulphur dioxide and nitrogen oxide gaseous emissions, contributes to the formation of airborne fine particulate matter contributing to air quality degradation, reduced visibility, and regional haze. Further, NH₃ released by nitrogen (N) fertilizer applications (FA) worldwide leads to substantial economic loss in the agriculture sector. India, after China, stands as the world's second-largest producer and consumer of synthetic N-fertilizer. Despite the significance of NH₃ in the global nitrogen (N) cycle, notable inaccuracies and uncertainties persist in India regarding its atmospheric emissions. Current assessments of emissions from fertilizer applications rely on 'fixed' emission factors, introducing limitations and uncertainties in the estimation process. This study aimed to enhance the accuracy of NH₃ emission estimations by correcting the fixed emission factor with different modification factors (MFs). These MFs are functions of crucial influencing variables, including fertilizer type, soil pH, application method (basal and top dressing), application rate, and temperature. In 2018-19, the overall NH₃ emissions in India reached approximately 3.15 tera-grams (Tg). The considered fertilizer types encompass Urea (2959 Giga-grams (Gg)), DAP (Di-ammonium phosphate (112.6 Gg), AS (15.30 Gg), and NPK (65.96 Gg). Considering all fertilizer types, the agroclimatic regions of India characterized by the highest NH₃ emissions are the ‘Trans Gangetic Plain Region (525.61 Gg)’ followed by the ‘Upper Gangetic Plain Region (466.83 Gg)’. The total NH₃ emissions were found to peak during the monsoon season (Jul-Sep; 1217.35 Gg) followed by winter (Dec-Mar; 926.36 Gg). The current study addresses the exigency to minimize NH₃ emissions from FA and helps in implementing various stringent mitigation strategies and regulatory actions for sustainable environmental management.

Nitrate has become an increasingly important component of fine particulate matter (PM_2.5) in some major cities in China. It didn’t decrease as rapidly as the precursor nitrogen oxides (NO_x) did (or even increased), which has attracted wide interest. However, previous research mainly focused on elevated nitrate in cold seasons, leaving seasonal disparities in the trends of nitrate unclear. Here we investigated the interannual trends in nitrate at an urban site in Beijing for all seasons over 2015-2019 and explored the drivers by assembling in-situ measurements, machine learning, and chemical transport model (CTM) simulations. The results show that nitrate would increase in all seasons with the impacts of meteorology removed (i.e., de-weathered), at rates ranging from 1.1% per year in winter to 9.8% per year in summer. Such emission-induced increases in nitrate were inhibited by meteorology in the fall and winter, as a result of decreased humidity, while enhanced in the spring and summer. In addition, the de-weathered nitrate concentrations increased nonlinearly with NO₂, with steeper slopes of nitrate to the precursor in 2018-2019 than in 2015-2017, reflecting a higher nitrogen oxidation ratio (NOR) in more recent years. The reduced NOx emissions in spring, fall, and winter increased nocturnal NO₃ radicals, which could facilitate the heterogeneous formation of nitrate at night and promote NOR. In summer, NOR increased as well, but this might primarily be related to increased daytime OH radicals and enhanced NO₂+OH nitrate production. This study highlighted the distinct impacts of emission changes and meteorology on nitrate in four seasons, and the necessity for season-specific nitrate control strategies.

Nitrous acid (HONO) is a significant primary source of OH radicals in the troposphere. Thus it affects the oxidizing capacity of the atmosphere and the formation of ozone (O₃) and fine particulate matter (PM_2.5). However, the abundance of HONO is always not constrained in the chemical transport models (CTMs), indicating the sources have not been well recognized. In this study, we updated the parameterizations of HONO formation, particularly considering the impacts of light intensity and relative humidity on heterogeneous reactions of NO₂ on ground and aerosol surfaces, and added new sources (photolysis of nitrate, the heterogenous formation of HONO on soot, etc.) in the air quality model CMAQv5.3.2. We conducted simulations and investigated major sources of HONO at a rural site in the Yangtze River Delta region in the fall of 2020. The improved HONO chemistry case led to a substantial increase in HONO and reduced the gap between the simulation and observation (NMB from -0.83% to -0.40%). A sensitivity test indicated that negative biases of modeled NO₂ (most likely due to emissions) explained the remaining model-observation discrepancies. Overall, the main formation pathway of HONO at the site was the heterogeneous reaction on ground surfaces, accounting for approximately 90% of HONO formation at night and ~80% during the daytime. Meanwhile, we found a 2× increase in the daytime average OH concentration and a 1.8× increase of HO₂ with improved HONO simulation, resulting in enhanced nitrate, sulfate, ammonium, and SOA by factors of 1.3, 0.3, 0.8, and 1.1, respectively. This study emphasized the role of HONO chemistry in the formation of secondary pollutants, and future work needs constraint of radicals in the CTMs as well.

Volatile chemical products (VCPs) have been considered one of the major sources of reactive organic carbon (ROC, including volatile organic compounds (VOCs), and semi/intermediate volatile organic compounds (S/IVOCs)) in megacities. Among VCPs, emissions from domestic volatile chemical products (DVCPs, for instance, personal care products, cleaning products, and insecticides) and the associated impacts are not well understood, particularly in China. In this study, we compiled an emission inventory for DVCPs and explored the impacts of DVCP emissions on the atmospheric environment over the Yangtze River Delta (YRD) using an air quality model. The newly estimated emissions from DVCPs are four times higher than those in the traditional inventory and accounted for 10% of total residential emissions in the YRD. The simulation with updated DVCP emissions compared well with measurements in terms of DVCP tracers (e.g., siloxanes, isopropanol) in urban Shanghai. With improved gas-phase chemistry for oxygenated VOCs as well as parameterization for secondary organic aerosol (SOA) from compounds that had been overlooked earlier, DVCPs contributed to the daily maximum 1-hour O₃ and SOA by up to ~3 ppb and ~1 μg/m³ in November of 2019. The impacts were higher in the urban center of Shanghai and under unfavorable meteorological conditions (e.g., stagnation). This study revealed that DVCP emissions are sources that cannot be overlooked in urban centers. Considering the dynamic nature of DVCP emissions, which evolve alongside changes in formulas and our lifestyle, they need more attention in the future.

Since the residential type in urban areas have moved to the high-altitude along with increasing the number of the tall buildings, the public demand for information about atmospheric pollutants at those levels including ozone also have increased. As the first air quality monitoring station (AQM) at high altitude, the AQM at the top of POSCO Tower Songdo (37.39 °N, 126.64 °E, 305m a.s.l.) has been measured since December 2022. Because POSCO Tower Songdo is located in the upwind region of the Seoul Metropolitan Area (SMA) along with industrial complexes and large point source nearby, it is suitable location for identifying the effects from both local emission and long-range transport of ozone and precursors. During ozone alert periods (from May to June 2023), gaseous pollutants (O₃, NOx, SO₂, CO) and non-methane VOCs (NMVOCs) were measured using Air Quality Monitoring System (AQMS, T series, Teledyne API.) and Selected Ion Flow Tube Mass Spectrometer (SIFT-MS, Voice 200, Syft Tech.), respectively. Also, the nitrogen oxides (HONO and HNO₃) were obtained by Monitor for AeRosols and Gases in ambient Air (MARGA, 2060, Metrohm) We analyzed ozone formation characteristics at high altitude in SMA using the BOX Modeling eXtensions (BOXMOX) model by conducting the sensitivity test. We believed that these results could be scientific evidence for formulating ozone management policies.

This study emphasizes on the initial crucial step towards mitigating the particulate matter (PM_2.5) by scientifically quantifying and pinpointing its sources. The need for the accurate emission inventories is of paramount importance, and among the diverse methods available, ground-based activity data stands out for its high precision. However, preparing such emission inventories poses challenges as it is very resource and time intensive. This study focusses on the preparation of a detailed emission inventory for PM2.5, with a resolution of 300m × 300m, using ground-based activity data for eight non-attainment cities in Punjab, India (Jalandhar, Amritsar, Patiala, Mandi Gobindgarh, Khanna, Naya Nangal, Dera Bassi, and Dera Baba Nanak). Furthermore, using this in-house emission inventory, Emissions Database for Global Atmospheric Research (EDGAR) which is a widely acclaimed database, is updated and WRF-Chem simulations are performed, the results of which are also a part of this study. The findings of this study reveal Jalandhar is the highest contributor to PM2.5 emissions, estimated 4296 tons/year, respectively, while Dera Baba Nanak emerges as the least contributor with total PM2.5 emissions of 92 tons/year. The primary sources identified across all cities are predominantly vehicular, road dust, industrial, and domestic emissions, making them the top four contributors. Another important insight from this study is the identification of the hotspots within each city from the gridded total emission maps. The results of this study hold critical importance for policymakers, providing a basis for informed decisions and targeted actions to reduce emissions and protect public health. Ultimately, the study emphasizes the importance of thorough emission quantification and source identification in developing effective strategies for reducing particulate matter.

The OH reactivity (OHR), the reciprocal of the OH lifetime, indicates the total amount of reactive trace gases loading in the atmosphere. We examine the characteristics of OHR and the environment in South Korea during the Korea-United States Air Quality (KORUS-AQ) campaign by using the comprehensive NASA DC-8 airborne measurements. The data present that anthropogenic emissions control OHR in South Korea with significant contributions of CO, OVOCs, NO_X, and hydrocarbons showing the increase of observed total OHR (tOHR) near and downwind regions of the primary emission sources. However, we found that a 3-D chemical transport model (CTM), GEOS-Chem, underestimates the concentrations of CO and OVOCs compared with observations, thus causing significant OHR underestimation. To reconcile the discrepancy between observed and simulated OHR (missing OHR), we examine the model's processes, such as emission inventories and chemistry schemes. We first adopt the observationally constrained VOC emission inventory based on the preceding studies using airborne and satellite observations. The adopted emission inventories lead to the increase of simulated OVOCs (~16%), aromatics (~23%), and O₃ (~4%). Second, we update tropospheric chlorine chemistry to the model, resulting in the increases of simulated NO_X (~12%), O₃ (~2%), OH (~5%), and HO₂ (~2%). Furthermore, we quantify the impact of CO underestimation in the model on the NO_X-O₃-HO_X cycle and simulated OHR. As a result, we reconcile some parts of the missing OHR from ~52% to ~40%, but OVOCs and O₃ underestimation in the model still need to be figured out.

Secondary inorganic aerosols, including sulfates, nitrates, and ammonium, are known as the primary constituents of fine particulate matter (PM_2.5). Recently, the proportion of sulfates has diminished while that of nitrates has increased with the reduction in sulfur emissions due to stricter regulations. To investigate the impacts of nitrates on PM_2.5 concentration growth for pollution periods, we conducted field intensives at the Yongdang Campus of Pukyong National University, located in the coastal area of Busan, from January 2021 to June 2022. The measurement site is adjacent to a port and a small industrial complex. Measured components included water-soluble inorganic ions (PILS-IC, Metrohm Eco IC 883, Metrohm) and precursor gases (NO, NO₂, NH₃, O₃; TEI instruments). The aqueous liquid water content (ALWC) and pH were calculated using the ISORROPIA II thermodynamic equilibrium model. First, we found that nitrates played a significant role in the accumulation of high PM_2.5 concentrations, regardless of the season. This study attempted to elucidate the nitrate formation mechanism that contributed to the growth of PM_2.5 concentrations under highly polluted conditions. We further analyzed temporal variations and correlations among the concentrations of inorganic particles, excess ammonium, humidity, ALWC, and Ox (NO₂+O₃) to support the nitrate formation processes we propose here. The tentatively proposed nitrate formation process that increases PM_2.5 concentrations in this coastal site is as follows. Initially, the ALWC increased through the sea breeze system due to increasing specific humidity in the daytime, in turn, increasing relative humidity at night. Then, N₂O₅ reacted with the surface of aerosol H₂O to form HNO₃ (aq) at night, leading to the formation of nitrates. Thus, the heterogeneous formation increased the concentration of nitrates at night, which subsequently led to an increase in the PM_2.5 levels.

In urban coastal environments, mesoscale atmospheric circulation can potentially influence air quality by affecting the chemical processes of air pollutants. In this study, we measured air pollutant concentrations (NO₂, NO, and O₃) in Incheon and Ulsan, developed urban coastal areas in South Korea. Based on highly spatially resolved observations, we investigated the spatiotemporal variations of pollutants affected by local emission environments. Additionally, we examined the effects of sea breeze on surface ozone concentrations and the associated chemical processes. To verify emission environments in Incheon and Ulsan, we referred to land cover maps. Incheon is primarily traffic-oriented due to its proximity to cargo terminals, ports, and an airport, while Ulsan features extensive industrial areas along the coastline. We installed air quality sensors in representative areas for each microenvironment and conducted measurements for two weeks. We found that the variation of the air pollutant concentrations varied depending on the emission environments. Notably, an inverse correlation was observed in the NOx and O₃ changes during the diurnal pattern on weekdays, and this variation was more distinct in traffic and industrial areas. These variations of NOx and O₃ concentration were also evident during sea breeze periods. Measured O₃ near the coast was relatively high during sea breeze periods. However, in the industrial areas with high NOx emissions, O₃ concentrations showed a decreasing gradient. This suggests that NOx emissions from inland areas contributed to the reduction in O₃ concentrations. Based on the O₃ concentration gradient, we evaluated the impact of sea breeze on O₃ concentrations through quantifying O₃ advection rates, and examined approaches to the chemical budget of O₃ concentrations.

Although carbonaceous aerosols are higher health risk to humans than PM_2.5, there are still uncertainty in Physicochemical properties, transportation, and spatiotemporal distribution. This study used one year semi-continuous PM_2.5 chemical composition from the Gyeonggi Air Quality Research Center in Ansan, Korea (37.31°N, 126.80°E) to identify the major sources of carbonaceous aerosols in PM_2.5. Because Ansan is a representative industrial and urban mixture area nearby Seoul Metropolitan Area, it is suitable to monitoring the various type of PM_2.5 characteristics. Carbonaceous aerosols were analyzed using sunset OCEC analyzer based on NOISH (National Institute for Occupational Safety and Health) thermal optical transmittance (TOT) analysis protocol. Then, we classified the concentrations of organic carbon (OC; from OC1 to OC4) and elemental carbon (EC; from EC1 to EC6) detected depending on the analysis temperatures (310 °C to 870 °C). Additionally, the hourly-based 12 heavy metal components (Si, Ti, V, Mn, Fe, Ni, Cu, Zn, As, Se, Br, Pb) in PM_2.5 was also measured using XRF (X-Ray Fluorescence spectrometer) and the major emission sources of carbonaceous aerosols will be identified using statistical approaches (such as principal component analysis (PCA) and/or a receptor model). Therefore, we believe that this research will be helpful to formulating effective regulation policies for the carbonaceous aerosols by estimating the contribution of emission sources, quantitatively.

To improve air quality, it is important to establish effective reduction and regulation policies for major pollutants by estimating their contributions and impacts on human health. For these purposes, comprehensive semi-continuous PM_2.5 measurements from the Gyeonggi Air Quality Research Center in Ansan, Gyeonggi-do (37.3202 ºN, 126.8283 ºE, 13 m above sea level), which is located in industrial and urban mixture areas, has been conducted since 2020 to identify characteristics of complicated industrial areas. We selected hourly-based chemical species of PM_2.5 from January 1 to December 31 in 2020, including inorganic ions from an ambient ion monitor (9000D AIM), trace elements from an online x-ray fluorescence spectrometer (XRF Xact 625i), and carbonaceous aerosol (organic and elemental carbon) from thermal optical transmittance analyzer (Semi-continuous OCEC Analyzer) based on a modified NOISH (National Institute for Occupational Safety and Health) Method. To estimate the contributions of sources, the Positive Matrix Factorization (PMF) receptor model was applied to PM_2.5 chemical speciation data which passed the strict QA/QC criteria. The highest source contribution was ammonium nitrate as 39.6% and followed by ammonium sulfate (21.3%), Gasoline (10.9%), Diesel (9.8%), Roadway emission (7.7%), Industry (3.8%), Coal Combustion (2.6%), Oil Combustion (1.8%), Soil (1.4%), and Nonferrous metal (1.0%). Additionally, conditional probability function (CPF) analysis was performed to understand the influence of local emissions. Most emission sources were located in the west direction where an industrial area and a highway are located. However, the diesel emission source was highly contributed from the east direction, where Ansan Bus Terminal and major commercial facilities are located. We believe the evidence of local contributions will be useful in formulating policies that improve the air quality in industrial/urban mixed areas in Gyeonggi Province, Korea.

NO₃ radical was found to play a critical role as a dominant oxidant in the nocturnal atmospheric chemistry, especially in the absence of the OH radical. It mainly oxidizes volatile organic compounds (VOCs) and NOx in an urban air mass during the night. Furthermore, N₂O₅, produced by the reaction with NO₃ radical and NO₂ in thermal equilibrium, contributes to the formation of nighttime secondary organic aerosol and the subsequent increase in daytime ozone production through a series of heterogeneous reaction. The chemistry of these reactive nitrogen oxides is significantly influenced not only by its source and sink mechanisms but also by complex weather conditions, including local and region circulations. Therefore, it is necessary to continuously monitor the chemical processes of NO₃ radical and N₂O₅ for all seasons. We performed to measure NO₃ radical in the summertime using a home-made Long-Path Differential Optical Absorption Spectroscopy (LP-DOAS) in Seoul. Currently, we also successfully developed Incoherent Broadband Cavity-Enhanced Absorption Spectroscopy (IBB-CEAS) to observe N₂O₅. This study aims to simultaneously measure and characterize the NO₃ radical and N₂O₅ in the wintertime at Korea University in Seoul during the Asia-AQ campaign. Subsequent analysis will be conducted to assess the wintertime temporal variations and chemical mechanisms based on the production and loss of NO₃ radical and N₂O₅ in Seoul with the simultaneous measurements of key trace gases, particulate pollutants and meteorological parameters. The detailed data analysis will be presented during the conference.

Characterizing inflow structure is important to better represent tropical cyclone impacts in numerical models. While much research has considered the impact of storm translation on the distribution of inflow angle, comparatively less research has examined its distribution relative to the environmental wind shear. This study analyzes data from 3655 dropsondes in 44 storms to investigate the radial and shear-relative distribution of surface inflow angle. Emphasis is placed on its relationship with intensity change. The results show that the radial variation in the inflow angle is small and not significantly dependent on the shear magnitude or intensity change rate. In contrast, the azimuthal distribution of the inflow angle shows a significant asymmetry, with the amplitude of the asymmetry increasing with shear magnitude. The maximum inflow angle is located in the downshear side. The degree of asymmetry is larger in the outer core than in the eyewall. Intensifying storms have a smaller degree of asymmetry than steady-state storms under moderate shear.

Forecasting tropical cyclone (TC) activities has been a topic of great interest and research. Taiwan island is one of the key regions that is highly exposed to TCs in the western North Pacific. Here we utilize the mainstream reanalysis datasets in 1979–2013 and propose an effective statistical seasonal forecasting model, namely the Sun Yat-sen University (SYSU) Model, for predicting the number of TC landfalls on Taiwan island based on the environmental factors in the preseason. The comprehensive predictor sampling and multiple linear regression shows that the 850-hPa meridional wind over west of the Antarctic Peninsula in January, the 300-hPa specific humidity at the open ocean southwest of Australia in January, the 300-hPa relative vorticity at the west of the Sea of Okhotsk in March and the sea surface temperature in April at the South Indian Ocean are the most significant predictors. The correlation coefficient between the modeled results and observations reaches 0.87. The model is validated by the leave-one-out cross validation and recent 9-year observations (2014–2022). The prediction of the SYSU Model exhibits a 98% hit rate in 1979–2022 (43 out of 44), suggesting an operational potential in the seasonal forecasting of TCs landfall on Taiwan island.

This study utilizes the HighResMIP-CMIP6 data to investigate future changes in tropical cyclone (TC) activity, including TC genesis frequency (TCGF) and track density (TCTD), in response to El Niño-Southern Oscillation (ENSO). The TC activity changes from uncoupled (atmosphere-only) and coupled runs are very diverse, indicating large uncertainty for future projections. Full-coupled models projected an El Niño-like Sea surface temperature (SST) warming over the equatorial Pacific, while uncoupled (atmosphere-only) models reveal a La Niña-like warming pattern. This SST difference further induced diverse atmospheric circulation anomalies, eventually contributing to distinct TCGF and TCTD changes during ENSO phases for future projections. Further detailed analysis reveals that the El Niño-like or La Niña-like warming patterns are two extreme cases. It shows a strong variance of simulated SST across different models. The accurate projection and simulation of central Pacific SSTs and Eastern Pacific SSTs is the key to reducing the diverse behaviors of TC activity among these models. We detected three crucial regions, that is the central Pacific, the subtropical eastern North Pacific, and the equatorial eastern Pacific, that could modulate TC activity in response to ENSO in numerical models. These results help to further improve the accuracy of models to simulate future TC changes.

We evaluated the ability of the fvGFS with a 13-km resolution in simulating tropical cyclone genesis (TCG) by conducting hindcast experiments for 42 TCG events over 2018–19 in the western North Pacific (WNP). We observed an improved hit rate with a lead time of between 5 and 4 days; however, from 4- to 3-day lead time, no consistent improvement in the temporal and spatial errors of TCG was obtained. More “Fail” cases occurred when and where a low-level easterly background flow prevailed: from mid-August to September 2018 and after October 2019 and mainly in the eastern WNP. In “Hit” cases, 850-hPa streamfunction and divergence, 200-hPa divergence, and genesis potential index (GPI) provided favorable TCG conditions. However, the Hit–Fail case differences in other suggested factors (vertical wind shear, 700-hPa moisture, and SST) were nonsignificant. By contrast, the reanalysis used for validation showed only significant difference in 850-hPa streamfunction. We stratified the background flow of TCG into four types. The monsoon trough type (82%) provided the most favorable environmental conditions for successful hindcasts, followed by the subtropical high (45%), easterly (17%), and others (0%) types. These results indicated that fvGFS is more capable of enhancing monsoon trough circulation and provides a much better environment for TCG development but is less skillful in other types of background flow that provides weaker large-scale forcing. The results suggest that the most advanced high-resolution weather forecast models such as the fvGFS warrant further improvement to properly simulate the subtle circulation features (e.g., mesoscale convection system) that might provide seeds for TCG.

Typhoons affecting the Korean Peninsula are getting stronger and dying down (WMO, 2021; IPCC, 2021; Na and Jung, 2021). The National Oceanic and Atmospheric Administration (NOAA) and the National Weather Service (NWS) have established disaster prevention and changed the national management policy stance by providing information on risks in advance about the effects of extreme weather disasters, responding to them, and taking action in advance. A paradigm shift was attempted by declaring the Weather-Ready Nation (WRN), a national strategy to minimize damage, as a strategic plan for the next 10 years in 2011. Starting with the declaration of paradigm shift of WRN, various countries around the world (Britain, France, Japan, China, etc.) have conceptualized the words READY and PREPARE in relation to tropical cyclones such as typhoons and hurricanes, establishing disaster prevention. In addition, with the help of the World Meteorological Organization, several developing countries in East Asia (Philippines, Vietnam, Thailand, etc.), which are heavily affected by typhoons, are also actively introducing impact forecasts (Katie et al., 2016; Nguten et al., 2021; Nofal et al., 2021; Thuc et al., 2022; Do and Kuleshov, 2022).This study aims to evaluate the possibility of typhoon impact prediction in Korea by utilizing the risk index calculated by the Typhoon-Ready System, which was developed to establish the basis for typhoon impact prediction in Korea. The possibility and validity of the risk index were analyzed through the analysis of representative typhoon cases applying the complex weather disaster risk index based on impact prediction, and in particular, the possibility of typhoon impact prediction in Korea was evaluated through the risk level proposal for regional-specific risk indices, which are the core of impact prediction. This work was supported by the National Research Foundation of Korea(NRF) grant funded by the Korea government(MSIT) (No.RS-2023-00212688).

Genesis Potential Indices (GPIs) indicate the potential of the Tropical cyclone (TC) occurrence in response to large-scale environmental conditions. Recent studies use simulated GPIs from the Coupled Model Intercomparison Project Phase 6 (CMIP6) to investigate the physical explanation of projected global TC changes. In this work, we first evaluate the ability of CMIP6 historical simulations (1979-2003) to capture long-term variations in GPIs (dynamic GPI and Emanuel GPI) compared to the GPIs from observational data sets. The presence of inherent biases in the models prevents both the GPI simulations and the individual terms of GPIs from adequately reproducing the observed variation. Thus, bias correction was conducted via direct approach using univariate (quantile delta mapping; QDM) and multivariate (N-dimensional probability density function transform; MBCn) methods and component-wise with QDM on CMIP6 simulated GPIs in historical and future projections (2075-2099) under high emission (SSP5-8.5) scenario. Overall model evaluation on all BC methods in historical GPI simulations under the direct bias correction exhibits a slightly better performance than the component-wise approach. Even though we observed no significant difference between the future changes from GPIs from bias-corrected CMIP6 and the raw models, we observed substantial model uncertainty reduction in the corrected models, emphasizing the improved reliability of the models for future projections. This study provides a comprehensive assessment of bias correction techniques used in CMIP6 models for GPI simulation, along with the projected potential for TC formation.

This study investigates the large-scale circulation anomalies induced by straight-moving tropical cyclones (TCs) over the western North Pacific (WNP) during winter. Corresponding to the straight-moving TCs, a quasi-stationary wave train is excited as alternative geopotential height anomalies in the upper troposphere stretching from East Asia to the North Pacific. Specifically, the anomalous anticyclones are initially formed over East Asia to the north of TCs and then lead to the subsequent anomalies in the downstream areas. Further analysis reveals that the upper-level anticyclonic anomalies are excited by negative Rossby wave sources, which are mainly attributed to the poleward vorticity advection by anomalous divergence relevant to TCs. In addition, the diagnosis indicates that the generation of wave source is caused by the product of the TC-induced divergent flows and the prominent meridional vorticity gradient in association with East Asian upper-tropospheric westerly jet. The above processes differ from the recurving TCs in summer and autumn, which undergo extratropical transition when they move northward into the mid latitude. These findings imply that the tropical disturbances over the WNP, such as straight-moving TCs, can remotely affect weather over the extratropics, and thus have implications for improving the weather forecast over the extratropics through improving tropical disturbance forecast.

Moderate tropical cyclone precipitation (TCP) is of great significance to regional water resource supply, while extreme TCP could bring significant adverse impacts to ecosystems and society, especially when tropical cyclones intensify rapidly, leaving no time to take prevention actions. Whether rapid intensification (RI) of tropical cyclones (TCs) affect TCP in both land and ocean remains unknown. Here we classified TCs which have undergone increases in the maximum sustained wind speed (MSW) by at least 30 knots within 24-h into RI category. We analyzed TCP totals provided by daily precipitation from Precipitation Estimation from Remotely Sensed Information using Artificial Neural Networks-Climate Data Record (PERSIANN-CDR) and spatial extent from 1983 to 2019 in the four categories based on regions (land and ocean) and RI-experiencing characteristics (with- and without-RI). TCP totals and spatial extent was identified by the restricted moving neighborhood method and semi-variogram framework. The results show that TCP totals on the ocean are larger than those on the land, since RI-experiencing TCP are higher than TCP without RI experiencing, although RI processes tend to increase TCP totals in the extremely high percentiles more significantly on land than ocean. The four regions of the Northeast Pacific Ocean (EP), South Pacific Ocean (SP), Northwest Pacific Ocean (WP), and North Atlantic Ocean (NA) show increases in regional mean and extreme TCP totals. The highest increase in the extreme TCP totals (0.37 mm day−1 year−1) over the NA region occurs in the RI_ocean category, which is 2.6 times the average positive enhancement trend across all basins. The increasing rate of the extreme TCP totals over the WP region is higher in track points with RI-experiencing than without RI-experiencing. The category of RI_land over the regions of NA, EP and SP shows a significant increase in the regional mean TCP spatial extent.

The South China Sea (SCS) is the largest marginal sea in the Northwest Pacific Ocean, and it encounters frequent typhoons. The atmosphere and ocean will create significant thermal and dynamic responses during the intense disturbance caused by typhoons. However, these responses have not been thoroughly investigated owing to the complicated marine environment. According to the satellite data, the SCS Basin was observed to have a strong SST response to Typhoon Mangkhut, resulting in widespread SST cooling. A coupled model was used to investigate the atmospheric and oceanic responses to Typhoon Mangkhut. Best-track data, satellite SST, and ARGO measurements show that the coupled WRF-CROCO simulation displays better track, intensity, SST, temperature, and salinity profiles than those of the WRF-only simulation. Results show that the typhoon induced rightward intensifications in wind speed, ocean current, and SST. The following are some remarkable atmosphere and ocean responses: (1) the SST below the inner-core region is cooled by 1°C, resulting in a 37%–44% decrease in wet enthalpy, and the central pressure is increased by ~9 hPa. Therefore, the changes in SST below the inner-core region of the SCS Basin have a significant impact on air-sea fluxes under high-wind conditions. (2) the ocean boundary layer analysis shows that near-inertial oscillations on the right side of the typhoon track and a strong inertial current up to ~2.28 m/s in the upper ocean were observed, which resonated with the local wind and flow field on the right side and induced strong SST cooling. (3) a decrease in SST decreased the moist static energy of the typhoon boundary layer, thereby weakening the typhoon’s intensity. The difference in equivalent potential temperature and sea surface pressure have a good correlation, indicating that the influence of moist static energy on typhoon intensity cannot be overlooked.

This study examines the characteristics of PBL for landfalling TC cases using data from the Boseong Tall Tower. The Boseong Tall Tower, the second highest multi-level meteorological tower in Asia, provides vertical profiles of the lower planetary boundary layer (PBL), enabling investigation of the characteristics of the lower PBL over land. It is particularly useful when a tropical cyclone (TC) makes landfall, as observations with sonde over land can be difficult. Previous studies have suggested that the track, intensity and structure of a TC are influenced by turbulent processes in the Planetary Boundary Layer (PBL), and the parameterization of the TC boundary layer used in the ocean may not match well with land. Therefore, it is important to understand the characteristics of the PBL over land, especially when TCs make landfall, for better TC prediction and prevention. Based on our analysis, it appears that the log layer may not accurately represent the depth of the surface layer when assuming a near-neutral surface layer. This suggests that alternative assumptions may be necessary. In contrast to previous studies, our findings indicate that the fitted surface frictional speed (u_*s) slightly increases with wind speed, while the fitted roughness length (z₀) decreases with mean wind speed. These suggest that stronger surface wind speeds result in a wider range of wind speeds, which may be influenced by ocean surface features due to the proximity of the Boseong Tall Tower to the coastline. Additionally, we compare the vertical eddy diffusivity (K_m) as measured by observation and the parameterization used in the numerical model. [Funding. This study was supported by the Korea Meteorological Administration Research and Development Program (KMI2022-01312)].

A prediction system for wind and precipitation during typhoons affecting the Korean Peninsula was developed based on the characteristics of wind and precipitation areas along the track of past typhoons, and to improve the accuracy, sensitivity experiments were conducted based on the selection criteria for similar past typhoons. As a result of Typhoon KHANUN by selecting similar typhoon cases in the past, the wind prediction was good result, but precipitation prediction was relatively inaccurate. To improve the observation-based typhoon prediction system, we conducted a prediction experiment using an ensemble numerical model. Among the predicted ensemble members, we selected those whose typhoon centers were within 50 km of the official typhoon information and then averaged the precipitation/wind forecast fields. The wind prediction results showed over estimated, while precipitation prediction showed significant improvement in accuracy compared to observations.

The decrease in sea surface temperature (SST) with latitude is one of the factors that weakens the intensity of typhoons moving towards mid-latitude. Atmosphere-only models (i.e., WRF) with high horizontal resolution which use prescribed SST for low boundary conditions sometimes overestimate the intensity of decaying typhoons in mid-latitude. The air-sea interaction reduces this overestimate of typhoon intensity by producing realistic SST cooling below the typhoon. However, excessive SST cooling in coupled models often occurs and this can decrease the typhoon intensity. In this case, ocean vertical structure may be associated. We investigate the effect of air-sea interaction and the vertical structure of ocean below typhoon BAVI in 2020 by comparing 3-km high-resolution WRF and WRF-ROMS coupled model. For the WRF-ROMS experiment, 3D ocean data is necessary, and the HYCOM analysis data is usually used. However, the salinity of HYCOM in East China Sea has a large uncertainty because HYCOM does not fully reflect the Changjiang river freshwater. So we conducted WRF-ROMS simulation with Changjiang river discharge to create realistic salinity input for typhoon BAVI simulation. As a result, WRF with prescribed HYCOM SST overestimates the intensity of typhoon BAVI, but WRF-ROMS with HYCOM salinity underestimates the typhoon intensity. WRF-ROMS with realistic salinity shows the salinity in East China Sea is lower than normal due to high Changjiang river discharge, and the freshwater inflow makes ocean stratification in East China Sea. Ocean stratification makes a barrier layer and reduces the vertical mixing and SST cooling under typhoon BAVI, increasing typhoon intensity.

Tropical cyclones (TCs) in the western North Pacific (WNP) follow two distinct tracks: some move in a west-northwest direction, while others recurve in a north-northeast direction. Recurving TCs possess a major disaster risk for the densely populated coastline of Korean Peninsula and Japan. While there have been studies on recurving TCs in the WNP, the statistical characteristics of the recurving TCs and whether their features are changing due to climate change have not been thoroughly investigated to date. Apart from that, there are discrepancies in extracting the recurving TCs in this basin. Therefore, this study proposes a new methodology comprising a set of criteria to realistically extract recurving TCs in this basin. TCs extracted based on proposed new criteria, we investigate statistical characteristics such as the seasonal distribution and trend, recurving location (latitude and longitude), the number of occurrences, lifetime maximum intensity (LMI) and recurving time nexus, translation speed, and the recurving angle of recurving TCs during the period 1981–2021 among others. Moreover, we further examine the reasons of varying frequencies of recurving TCs in different years. Results suggest that position and strength of WNP subtropical high (WNPSH) plays a pivotal role in recurving TCs varying frequencies in WNP. Finally, we investigate whether recurving TCs characteristics change over time. These findings will contribute to understanding the characteristics of recurving TCs in the WNP, especially in the face of climate change.

This study investigates the water vapor transport to the Tibetan Plateau (TP) induced by Bay of Bengal (BOB) tropical cyclones (TCs) in May during 1979-2019. The water vapor around the BOB TCs can be transported to the south of the TP by the southerlies at the upper troposphere east of the TC center. Whether the water vapor can cross the southern boundary of the Tibetan Plateau (SBTP) may rely on the different configurations of BOB TCs and long-wave trough and ridge. We found that the water vapor may (may not) cross SBTP if the TC’s wind field converges with the westerlies in front of the trough (ridge). The physical mechanism is explored according to the momentum equation. When the TC moves northward around the SBTP, its wind field could interact with the trough (Type-T TC) and ridge (Type-R TC) in the westerlies. In the area close to the trough line (ridge line), the meridional acceleration is positive (negative), hence meridional component of velocity increases (decreases). Therefore, it is more favorable for the water vapor transported northward if a Type-T TC enters the westerlies from the area close to the trough line than the ridge line. The simulation of the BOB TC cases configuration using WRF model confirms the importance of the relative position between the Type-T TC and the SBTP to the water vapor transport over the SBTP. For the Type-T TC, the meridional moisture budget over the SBTP is positively correlated with the TC intensity and negatively correlated with the distance between the TC center and the SBTP. However, for the Type-R TC, the meridional moisture budget is not related to the TC intensity and the distance. The configuration of TC and long-wave trough or ridge may be predictable through the intraseasonal oscillation activity in the eastern BOB and the southern Arabian Sea. May and October to December are the bimodal phases of BOB TC frequency, which decreases month by month from October to December and is relatively low in May. However, the contribution rate of MMBa is the highest in May. Seasonal variation in the meridional position of the westerlies is the key factor affecting the contribution rate. The relatively south position of the westerlies in November and December results in the lower contribution rate of MMBa.

This study evaluates potential sites for wind power generation in the Taiwan Strait and inland areas of Taiwan, employing data from multiple satellites, reanalysis, and meteorological stations. The calculation of wind energy is based on transforming these wind speed data to the elevation of wind turbines, considering blade surface area and other coefficients for potential electrical energy estimation. In offshore areas, water depth determines the suitability of fixed or floating wind turbines, with depths exceeding 60 meters being suitable for floating turbines, and shallower waters being appropriate for fixed ones. Inland Taiwan's wind patterns are influenced by topography and buildings. Due to the stability of the wind, coastal areas emerge as prime locations for wind turbine installation. The spatiotemporal characteristics of wind at each potential site are the primary considerations in this assessment. Wind speeds in the Taiwan Strait range from 4 m/s to 15m/s, typically higher in winter. According to the Meteorological Information Based Green Energy Operations Center in Taiwan, the effective wind speed range for turbines is 3 m/s to 25 m/s, which means that wind turbines can operate throughout the year in the Taiwan Strait. Additionally, climatic-scale variations in wind speed are also considered. From 2012 to 2023, an annual increase of approximately 0.15m/s was observed in February, April, May, October, and December in the Taiwan Strait. Comprehensive assessments indicate that the central offshore areas of the Taiwan Strait exhibit higher wind energy potential, with water depths conducive to fixed wind turbines.

The demand for long-term climate analysis is increasing with the flow of global climate change. Accordingly, the need for Level 3(L3) data from satellite observation has also increased in climate studies. In particular, the Geostationary Environmental Monitoring Spectrometer (GEMS) mounted on geostationary satellites is highly evaluated in East Asia, and it is time to develop a L3 algorithm tailored to the model grid to effectively utilize GEMS for climate analysis. Therefore, this study developed a long-term climate analysis data production algorithm based on GEMS data. Data for January 1, 2022, were estimated using GEMS AOD, NO₂, and total O₃ data. To consider the homogeneity of data, all data that produce L3 data was collected by using only Full Mode Scan. Data quality correction was performed by applying Cloud Fraction and Final Algorithm Flag. In addition, we used the area weighted average calculation method to achieve stable production. In this data production, the most used spatial resolution, 0.25° X 0.25°, was adopted, and using this data, the combination with other satellites for long-term data production will be conducted.

The Tibetan Plateau Vortexes (TPV) can trigger mesoscale convective systems (MCSs) under appropriate conditions. These MCSs not only contribute most of the precipitation in the plateau area, but also move eastward along with TPV under appropriate conditions, triggering a series of mesoscale convective activities along the way. These precipitation events are often intense and long-lasting, often causing serious flooding disasters. However, due to the lack of observation data and complex terrain over plateau areas, the study of weather systems in plateau areas has always been a challenge. This work utilizes Fengyun-4 meteorological satellite and reanalysis data to study the cloud top structure characteristics of plateau vortices with different movement paths. A comparative analysis was conducted on the differences in cloud top height, cloud top temperature, cloud top brightness temperature, and brightness temperature difference of different satellite channels among three paths of plateau vortices: Colonel in situ, moving eastward but not descending to the plateau, and moving eastward beyond the plateau. The influence of these variables on the movement path of plateau vortices was also discussed.

The conventional emission estimation for air quality simulation was based on the bottom-up method, which relies on the accuracy of the emission inventory. During Taiwan's winter and spring seasons, the northeasterly monsoonal wind prevails, which transports the air pollutants from East Asia to Taiwan. However, the updates of emissions in East Asia is slow, making it challenging to promptly capture changes in emission sources. This leads to forecasting bias during transboundary pollution events. Recently, more real-time and high-resolution satellite data have been used to constrain the bottom-up estimated emissions. This study aims to use TROPOMI satellite data to update the bottom-up estimated emissions in East Asia. This correction intends to address the issue of underestimated transboundary air pollutants in Taiwan. Through a mass balance adjustment of emissions derived from satellite and model-retrieved NO2 column densities, the adjusted NOx emissions are higher than the bottom-up estimated emissions, which furthermore enhances the model performance of the transboundary pollution event.

As extreme weather events such as heatwaves and wildfires increasingly occur worldwide due to climate change, efforts are underway to reduce greenhouse gas emissions. Among various greenhouse gases, methane significantly impacts climate change, with a global warming potential approximately 82.5 times that of CO₂ over a 20-year period. Reducing methane gas emissions, which have a relatively potent warming effect, is essential for mitigating climate change, and continuous monitoring is crucial for identifying and managing sources and amounts of emissions. Satellites can periodically observe the same areas, enabling continuous monitoring of methane emissions, and their global coverage allows for the detection of methane gas in extensive or inaccessible areas. The Sentinel-2 satellite, with its short revisit cycles and high-resolution imagery across various spectral bands, is useful for methane gas detection. Previous research has conducted methane detection studies on well-known methane emission sources using the shortwave infrared (SWIR) bands of various satellites. In this study, we utilized the Google Earth Engine (GEE), which allows for rapid analysis of satellite data in a web browser, to leverage the short revisit cycles and broad spectral band coverage of the Sentinel-2 satellite. We detected and monitored methane plumes from the 2.8 Jikdong Coal Mine located in Suncheon, South Pyongan Province, North Korea. The results of this study are expected to provide important data for the establishment of environmental policies and strategies for climate change mitigation. Furthermore, it demonstrates the practicality of remote sensing technology in methane emission monitoring and highlights the potential use of Sentinel-2 satellite data in climate change studies.

Total Precipitable Water(TPW), a column of water vapor content in the atmosphere, is an important meteorological factor and play a critical role in the occurrence of precipitation and atmospheric river. The current TPW algorithm operational at NMSC calculates TPW based on AMI Aerosol Profile(AAP) in for clear-sky areas, and uses values calculated from numerical weather prediction (NWP) for cloudy-sky regions. The validation results for this algorithm with radiosonde show an RMSE 6.9mm and a bias of 0.377mm. To fuse the distribution of TPW in all sky conditions, this study proposes a retrieval algorithm based on the light gradient boosting machine (LGBM) model using GK2A/AMI data, Radiosonde, ERA5, and three NWP models data. Through statistical verification, with a root mean square error (RMSE) of 6.697~7.056mm, a bias of -0.929~0.377, and a Pearson's R of 0.942~0.946. We utilize the previously generated NWP prediction field to retrieve TPW in real-time. Detailed introduction of new TPW will be presented in the conference. This research was supported by the “Technical development on high-impact weather detection and prediction using meteorological satellite data”(KMA2020-00121) of “Technical Development on Weather Forecast Support and Convergence Service using Meteorological Satellites” project funded by the National Meteorological Satellite Center, Korea Meteorological Administration.

The theoretically calculated radiance (B) in the process of utilizing satellite data into numerical weather prediction models plays an important role in various processes. For example, the difference between observed data O and calculated B is used to understand the error characteristics of the observed data or to determine whether clouds are present in the observed data. Therefore, accurately calculating B is one of the most basic processes for the use of satellite data. To achieve this, various efforts are being made, such as improving the accuracy of radiative transfer models as well as accurately inputting surface characteristics. In this study, we focus on using temperature and humidity profiles along the slant path considering the direction in which satellite data is obtained, rather than the vertical path for the input of the radiative transfer model. Previous studies on clear-sky conditions have shown that applying a slant path result in a noticeable deviation when the viewing angle is far from nadir and in regions with high variability in atmospheric temperature and humidity. This study extends this to investigate the effect of the slant path radiative transfer model in the presence of clouds. When the horizontal range of clouds is not extensive, the presence of clouds varies depending on the chosen path. Furthermore, application of the slant path explicitly considers the parallax effect of clouds in the simulation process, allowing for a more accurate calculation of B values. Taking all these factors into account, we aim to analyze the impact of slant path simulations on vertical temperature and humidity retrieval by analyzing the results of 1DVAR models for all-sky cases.

Heavy precipitation events are of great importance as large amounts of rainfall concentrated in a short duration can generate severe natural disasters, which then cause widespread damage to human societies. Many previous studies have shown an increasing trend in the atmospheric water vapor due to global warming, readily indicated by the Clausius-Clapeyron relation. However, most of studies on the occurrence frequency of heavy precipitation are generally conducted on regional or national scales.In this study, we use three satellite-based precipitation estimation products—Integrated Multi-satellitE Retrievals for GPM (IMERG), Global Satellite Mapping of Precipitation (GSMaP), and Global Precipitation Climatology Project (GPCP)—and one rain gauge-based precipitation product, NOAA Climate Precipitation Center (CPC), and examine the global trend in occurrence frequency of heavy precipitation from 2001 to 2020. Using the number of days when the daily precipitation amount exceeds the 95th percentile of precipitation amount (FP95) and utilizing the Mann-Kendall test, Sen’s slope estimator, linear regression, and innovative trend analysis, we find that the trends of FP95 from the various analysis methods are generally comparable but highly vary depending on the precipitation products. The trends of FP95 in the lands from IMERG and GPCP are generally similar with those from CPC. However, the trends in the oceans are noticeably inconsistent across all the products. The FP95 from GSMaP increases significantly both in the lands and in the oceans except for some regions, but those from GPCP are decreasing on most of the oceans. Because satellite-based precipitation estimation products derive the amount of precipitation indirectly from atmospheric elements such as water vapor, liquid water path, and cloud top temperature, trends in the atmospheric parameters provided by a polar orbiting satellite are also analyzed to explain the discrepancies in trends of FP95 across the precipitation products.

Heavy rains occurring in a short time can cause significant damage, but it is difficult to accurately predict or observe its occurrence and dissipation. In this study, we investigated the possibility of understanding the development of convective precipitation clouds using thermodynamic variables (such as lifting convective level (LCL), level of free convection (LFC), stability index, etc.) with a 10-minute interval and 2 km spatial resolution produced by GK-2A. The accuracy of these variables produced by GK-2A is estimated by comparing with the collocated radiosonde data. In case of stability indices, the weighted linear regression coefficients for KI, LI, SSI, and TTI all showed values above 0.7, and the errors were randomly distributed without bias. The error for SSI and LI were larger at night than during the day, and the overall pattern of differences between satellite and radiosonde products was similar across stations. Furthermore, an analysis of 21 significant outlier cases that occurred during the comparison revealed that all outlier cases corresponded to one or more of the following: nighttime near-surface PBL, near-surface inversion layer, or instability in temperature and humidity profiles in the radiosonde data. LCL deviation ranged widely from -120 to 70, and RMSE varied between 30 and 130 based on initial ascent altitudes. Using 962.25 hPa as a reference, the deviation was -3.24, with an RMSE of 31.77, indicating the highest accuracy. Derived temperature and humidity profiles from GK-2A align closely with radiosonde data under suitable ascent altitudes. In selected convective cases, the LCL decreased by approximately 70.94 hPa (0.23 hPa/min) in the 6 hours before convective cloud formation, confirming a link between higher vertical convection altitudes and a heightened potential for strong convection. We also plan to present additional analysis on the relation of LFC and CAPE to the convective precipitation.

Convective clouds, accompanied by heavy rainfall and thunderstorms, can lead to substantial damages. Hence, the detection of Convective Initiation (CI) plays a crucial role in weather monitoring and prediction. While radar observations mainly detect precipitation particles in the mature stage of convective clouds, satellite data can identify convection cells before the formation of precipitation particles. Satellite data also offers the advantage of covering a wide area with high temporal resolution. In this study, we aimed to predict convective initiation by adopting the U-Net-based LightningCast model, originally developed for lightning prediction by NOAA/CIMSS, as our foundational model. Since the GEO-KOMPSAT-2A (GK2A) satellite lacks a lightning detection sensor like Geostationary Lightning Mapper (GLM), we employed ground-based radar reflectivity as ground truth. After aligning radar echoes with the GK2A Korean Peninsula area, we selected radar reflectivity that exceeded 35dBZ at least once within 60 minutes as our target data. We utilized observation data from GK2A Advanced Meteorological Imager (AMI) during the summer as input and conducted a study to identify the optimal channel combination through sensitivity testing. This research was supported by the “Technical development on high-impact weather detection and prediction using meteorological satellite data”(KMA2020-00121) of “Technical Development on Weather Forecast Support and Convergence Service using Meteorological Satellites” project funded by the National Meteorological Satellite Center, Korea Meteorological Administration.

Prediction of wind vectors in the low troposphere is beneficial in many application fields. Conventionally, Supervisory Control and Data Acquisition (SCADA) has been used for the low-level (1000 hPa) wind estimation. On the other hand, a recent study has developed the algorithm that can predict clear-sky high-level (500-750 hPa) wind vectors a few minutes to hours ahead using the Korean geostationary satellite (GK2A) water vapor channels only. Here we find the correlation between the two level wind fields from the GK2A and SCADA by using Convolutional Long Short-Term Memory (ConvLSTM) model. Then this model is applied to predict low-level wind fields for the entire Korean Peninsula. Our results show that our low-level wind fields have higher accuracy than those from the numerical weather prediction. Also shown is a potential that GK2A high-level wind can be used to predict low-level wind where the observations are unavailable. We expect that this study can serve to improve weather and wind power prediction.

The Rapid-Scan Atmospheric Motion Vector (RS-AMV) provides satellite-based three-dimensional wind information, and it can be useful information for initializing numerical weather prediction(NWP) model through data assimilation. In addition, It can be useful for a forecaster to be tracking the convection cell and analyzing wind information around the typhoon. Advanced Meteorological Imagery (AMI) on board the GEO-KOMSAT-2A(GK-2A) can capture the imagery of East Asia every two minutes. With this advantage, the RS-AMV can be generated with better spatio-temporal resolution than AMV. RS-AMV is derived as following steps: 1) target selection via optimal method using largest standard deviation in specific area, 2) height assignment based on cross-correlation with Cloud Top Pressure (CTP) product, 3) calculate displacement between the similar pixels in earlier or later image, 4) assign quality information on each vectors in RS-AMV via comparison with surrounding pixels. In this study, the quantitative and qualitative evaluations were performed using in-situ observation and analysis fields from NWP model. As the results of the quality assessment, we found that the major issues of RS-AMV are height assignment and wind speed. Additionally, the cause of the problem and which parts of the algorithm caused the problem were presented. This study was carried out with the support of the Korea Meteorological Administration R&D program[Meteorological satellite forecast support and convergence service technology development] (KMA2020-00121).

Water vapor imagery is a suitable satellite data to track upper atmospheric flow. It is useful for analyzing synoptic meteorological phenomena as tracking cyclonic flows, vorticities, dry areas and wet areas that appear in water vapor images, and identifying the location of troughs/ridges by utilizing the strengthening and weakening tendencies of dry areas and anticyclonic air flows. Synoptic meteorological phenomena can be analyzed extensively using water vapor image. Infrared and visible images are useful for analyzing clouds distributed in the upper, middle, and lower layers of the atmosphere. In water vapor images, upper-level waves and darkening-related troughs, which are precursor stages of low pressure system that cannot be seen in clouds that appear in visible and infrared images, can be identified through air flow analysis. In this study, we specifically showed some methods to analyze troughs through the tendency of dry areas to strengthen or weaken using water vapor imagery. Through this study, we found that the troughs are located right in front of the maximum brightness temperature area of the dry zone and the ridges are located in the area where the wet region begins to gradually dry in the water vapor image, and where the clouds begin to dissipate in the infrared and visible images.

Hyperspectral infrared (IR) sounder measures radiances at CO₂ and H₂O bands with high spectral resolution, and these measurements can be used to retrieve atmospheric temperature/ humidity profiles which are important in short-term weather forecasts. GIIRS (Geostationary Interferometric Infrared Sounder) of FY-4B is the first hyperspectral IR sounder in the geostationary orbit, and measures wide area with high temporal resolution. GIIRS-measured radiances contain vertical information on atmospheric CO₂ and H₂O, allowing to estimate atmospheric temperature/humidity from space. Of course, development of relevant retrieval algorithm is necessary. Improving weather forecasting capabilities, the Korean National Meteorological Satellite Center developed a physical algorithm so-called AAP (Advanced Meteorological Imager (AMI) Atmospheric Profile) for the retrieval of atmospheric thermodynamical properties. The AAP algorithm is based on the 1D-Var scheme and was originally developed for GEO-KOMPSAT-2A AMI sensor. In this research, sensitivity tests were conducted to find the best set of GIIRS channels and observation/background error covariances. Then obtained results were used for adapting the original AAP algorithm for GIIRS sensor. It was found that the modified AAP algorithm performs well with 36 GIIRS channels. Results indicate that the rms error of the modified AAP algorithm compared with ERA5 atmospheric profile is reduced when the observation error is prescribed as larger value. In addition, artificial intelligence algorithm was also developed for temperature/humidity retrievals by training GIIRS-measured radiances with collocated ERA5 atmospheric temperature/humidity profiles. ANN, CNN, ResNet, and U-Net methods were used, and results were compared for that purpose. GIIRS-measured TBs were directly related to ERA5 profiles without any first guess profile. The test results show that the algorithm based on ResNet yields the highest performance compared to the other algorithms. However, it is noted that these neural network algorithms perform not better than the physical algorithm in terms of rms errors.

Geostationary Ocean Color Imager – II (GOCI-II) onboard GEO-KOMPSAT-2B has operated since 19 February 2020 and has monitored aerosol optical depth (AOD) over East Asia. Although GOCI-II has provided invaluable AOD datasets such as hourly measurements, biases of GOCI-II AOD are not sufficiently analyzed to exploit GOCI-II AOD data in a data assimilation system for air quality forecast. This study aims to analyze the characteristics of GOCI-II AOD data in terms of bias and root mean square (RMS) error from December 2021 to January 2023, by comparing GOCI-II data with aerosol optical properties derived from the AErosol RObotic NETwork (AERONET) data. To make comparable datasets between GOCI-II and AERONET, GOCI-II AODs were averaged within a square grid box centered on each AERONET site having size of x^o. The sensitivity test was conducted by increasing the grid box size x from 0.1˚ to 1.0˚ with the increment of 0.1˚. The bias of GOCI-II AOD to AERONET AOD depends on the selected grid box size in most AERONET sites, showing smaller biases as grid box size decreases, possibly due to data representativeness. It is expected that there will be certain threshold at which bias will not decrease further. In addition, bias of GOCI-II AOD according to the aerosol type was examined. Here, aerosol types were classified into 6 categories: dust, non-absorbing coarse, mixture, high-absorbing fine, moderating-observing fine, and non-absorbing fine. AOD over the eastern China showed higher value compared to neighboring countries during the study period. The characteristics of GOCI-II AOD identified by this study will be utilized to improve a design of pre-processing process of satellite-derived AOD for data assimilation.

Atmospheric Motion Vector (AMV) is a satellite product that tracks atmospheric current by monitoring the movement of clouds and water vapor in three sequential satellite images. It is valuable for enhancing the precision of forecast and analysis fields in numerical model data assimilation. Additionally, it's crucial for understanding wind structures in typhoons. The National Meteorological Satellite Center (NMSC) from Korea Meteorological Administration (KMA) is currently producing Rapid-Scan (RS)-AMV, which has an improved spatiotemporal resolution than AMV, using the rapid-scan data from Geo-Kompsat-2A (GK2A). This output features a height assignment technique that utilizes Cloud Top Pressure (CTP) to obtain local wind information with less reliance on numerical weather prediction (NWP) models. In this study, we focused on increasing the vector count by computing vectors through optical flow. We analyzed atmospheric current using these vectors and validated the accuracy of the optical flow AMV by comparing them with RS-AMV. Optical flow is a technique that calculates the movement of pixels between two image frames as vectors. we used the Farnebäck algorithm, which calculates this movement for all pixel values in the image. Height assignment using GK2A Cloud Top Pressure (CTP) was applied to the optical flow AMV. The optical flow AMV calculated in this way was verified qualitatively and quantitatively through case analysis and observation data such as radiosonde and wind profiler.

The GK-2A AAP (Geo-Kompsat-2A AMI Atmospheric Profile) and other 1D-Var atmospheric profile retrieval algorithms have common limitations related to ill-posed problems due to a lack of observed information, leading to unstable retrieved profiles or reliance on initial guess profiles. In our study, we addressed these challenges by employing Empirical Orthogonal Functions (EOFs) and Principal Components (PCs) for temperature and humidity profiles to mitigate the ill-posed equation problem through variable reduction. The number of components of vertical profile was reduced using main EOFs and PCs. Furthermore, we modified the GK-2A AAP based on Principal Component Analysis (PCA-AAP) to achieve optimized atmospheric vertical information. The temperature and humidity profiles were calculated using PCA-AAP, with the Unified Model (UM) and GK-2A radiance serving as inputs, mirroring the approach employed in GK-2A AAP. Validation of the PCA-AAP retrieved profiles was conducted using radiosonde data in August 2022. The accuracy of PCA-AAP was found to be nearly comparable to that of both the initial guess and GK-2A AAP. Notably, PCA-AAP successfully avoided retrieving inaccurate profiles observed in GK-2A AAP, attributed to the stabilizing effect of main EOFs. The results demonstrate that the PCA technique complements the GK-2A AAP algorithm, providing stable and continuous vertical profiles. Consequently, we plan to enhance the accuracy of EOFs by incorporating global radiosonde data and model analysis data. This improvement is anticipated to contribute significantly to supporting very short-range forecasting and nowcasting more effectively. This research was supported by the “Technical development for utilizing meteorological satellite data for numerical weather predication”(KMA2020-00122) of “Technical Development on Weather Forecast Support and Convergence Service using Meteorological Satellites” project funded by the National Meteorological Satellite Center, Korea Meteorological Administration.

Extreme weather often causes major disasters, such as the flash flood brought by intense rainfall. To assess rainfall position and intensity, we usually use satellite images and radar reflectivity as a reference. Radar reflectivity provides high-resolution information of rainfall structure, but the range that radar can detect is limited based on the location and range of each radar site. In remote areas or places such as the Pacific Ocean, the radar data cannot cover. In contrast, satellite images offer a broader range of coverage, providing information in a global scale. Therefore, we propose a machine learning framework to train models that combines radar reflectivity and satellite image data. Firstly, the visible light and infrared data are obtained from the Himawari-9 satellite in Taiwan area from July to December, 2023. Then, the radar reflectivity is provided by the Taiwan Central Weather Administration (CWA) in Taiwan area as labels. In this study, the U-net model captures the features of satellite images to predict rainfall intensity and generate radar reflectivity. The reason we choose U-net is because of its excellent performance in image segmentation. Hence, it is an optimal choice to identify the edges of cloud layers and predict rainfall intensity. This research not only aims to extend the detection range of radars, providing more comprehensive weather information, but also helps observe the formation and causes of typhoons and other severe rainfall events. This efficient framework can contribute to monitor various extreme weather conditions and provide real-time warnings, thereby reducing the losses caused by disasters.

In recent times, global temperatures have been rising due to global warming, with East Asia being one of the regions experiencing the highest rate of increase. Furthermore, under high CO2 emission scenarios, the increasing evaporative demand in the East Asia region is projected to rise more rapidly than land surface evaporation. This land-atmosphere moisture difference exhibits high vapor pressure deficits, which can lead to extreme temperatures and droughts. The estimation of potential evapotranspiration, a key indicator of evaporative demand, remains unclear in the available reanalysis or model data, as they do not account for plant responses to CO2. We only focused on stomatal changes in response to CO2 changes, and the results showed an increase in surface resistance with increased CO2, the effect of which was largely offset by vapor pressure deficits, which affects the increase in potential evapotranspiration. In particular, in the case of drought, when plant responses are not considered, potential evapotranspiration tends to overestimate drought in East Asia by approximately 17% compared to scenarios that take vegetation into account. We examined the characteristics of land-atmosphere interactions in East Asia based on land types using a land-atmosphere coupling matrix. Specifically focusing on the characteristics of land-atmosphere coupling and its changes with and without considering plant responses, we analyzed the resulting changes in extreme climate events such as heatwaves and droughts.

We have successfully incorporated a 3-dimensional sub-grid terrain solar radiative effect (3D STSRE) parameterization scheme into a convection-permitting Weather Research and Forecasting model (WRF_CPM) in this study. Impacts of 3D STSRE scheme on the ability of WRF_CPM in forecasting the precipitation in summer over the Tibetan Plateau (TP) and nearby regions with complex terrain have been systematically addressed by conducting experiments without and with the 3D STSRE scheme. Results show that the application of 3D STSRE scheme can obviously mitigate the overestimation of surface solar radiation (SSR) and rainfall over TP and nearby regions, especially over the areas with much more rugged terrain (i.e., southern TP) in the WRF_CPM without 3D STSRE scheme. Further mechanism analyses indicate that the decreased surface heating induced by the reduction of SSR reduces the intensity of the thermal-low pressure over the TP, which leads to the diminished strength of southwesterly winds and thereafter the weaker convergence of moisture flux over the southern TP. Moreover, the weakened surface thermal forcing makes the local atmosphere more stable, suppressing the vertical water vapor transport and local convection. These effects greatly alleviate the overestimation of precipitation over the southern TP produced by the WRF_CPM without the 3D STSRE scheme.

Changes in the soil freeze–thaw status will inevitably affect the thermal conditions and properties of the Tibetan Plateau (TP), thereby affecting its upper atmosphere, and further afield in East Asia and even globally. In this study, using the soil temperature simulated by the Community Land Model Version 5.0, the timing and duration of the soil freeze–thaw status were divided into freeze start-date, freeze end-date, and freeze duration. Results showed: (1) The model effectively reproduces seasonal multi-layer soil temperature changes, showing strong correlation with observations. (2) There is a clear trend of delayed freezing, advanced thawing, and shortened freeze duration from northwest to southeast over the TP. From 1979 to 2018, the freeze start-date was delayed by 7.3 days (1.9 days/decade), the freeze end-date advanced by 6.4 days (1.7 days/decade), and the freeze duration shortened by 13.7 days (3.6 days/decade). The timing and duration of the soil freeze–thaw status vary across different regions of the TP. The freeze start-date in all areas of the TP has been delayed in the past 39 years. Except for the sub-cold zone and arid regions of the TP, the freeze end-date has occurred earlier and the freeze duration has shortened, with the most significant changes in the sub-cold zone and humid regions. (3) The timing and duration of the soil freeze–thaw status are significantly correlated with surface air temperature, elevation, and latitude. The strongest correlation is with surface air temperature, followed by altitude and latitude. The western TP shows a stronger correlation than the eastern TP. The rate of change in the soil freeze–thaw status increases with altitude to 3000 m above sea level, while this rate decreases with elevation above 3000 m. The greatest rate of change is observed at 29°N.

Satellite data shows increasing leaf area of vegetation since 2000, which may exert biophysical effects on near-surface air temperature (SAT). However, such effects remain largely unknown because prior studies either focus on land surface temperature, which differs from SAT, or rely on simulations, which are limited by model uncertainties. As a widely used metric in climate and extremes research, SAT is more relevant to human health and terrestrial ecosystem functions. Therefore, it is necessary to explore impacts of greening on SAT and extremes based on observations. Here, we investigate the greening effects on SAT and subsequent extremes over 2001–2022 based on an enhanced global dataset for the land component of the fifth generation of European ReAnalysis (ERA5-land). We find that greening can cause cooling effects on the mean SAT and more pronounced cooling effects on SAT extremes in Europe, East Asia and Central North America, where evident greening was observed since 21^st century. An attribution analysis suggests that the main driving factor for the additional cooling effect of greening in hot extremes is the enhanced evapotranspiration. Moreover, the biophysical effects of vegetation on high temperatures vary with climate conditions and vegetation types. Our study reveals a considerable climate benefit of greening on SAT, which may have implications for climate mitigation in different regions around world.

This study used radiation observation data and remote sensing data to explore the different factors affecting radiation estimation and prediction based on machine learning algorithms. The results indicate that: (1) The estimation of radiation is influenced by weather condition levels and pollution levels, where the estimation of global solar radiation is linearly related to the weather condition level, while the diffuse solar radiation shows a non-linear relationship. In heavily polluted weather, the use of machine learning for solar radiation estimation performs poorly. (2) In all prediction scenarios, the feature selection (FS) of the Random Forest (RF) model is higher than that of the Support Vector Regression (SVR) model, and this performance advantage becomes more pronounced when the lead time exceeds 90 minutes. (3) Remote sensing data can assist in improving radiation prediction. Full-channel input of FY-4A remote sensing data can provide better prediction results, but considering time and computational cost, the optimal three-channel model is a better choice.

The snow cover on the Tibetan Plateau (TP) undergoes sub-seasonal changes, which exert a substantial influence on weather and climate in the surrounding and downstream regions. However, current climate models struggle to accurately capture the rapid changes of TP snow cover. Our research indicates that, most global climate models in phase 6 of the Coupled Model Intercomparison Project (CMIP6) largely overestimate the snow amount over the TP, along with notable biases in the rapid fluctuations of TP snow cover. Similarly, these biases are also present in regional climate simulations. Through in-depth analysis, we identify snowfall and snowmelt as primary contributors to the observed rapid changes in TP snow cover. The simulated biases in rapid snow changes over the TP result from the excessive snowfall and albedo bias in both global and regional climate simulations. Our findings reveal the potential sources of simulated biases in rapid snow changes over the TP and offer a promising perspective for improving the accuracy of TP regional and global climate simulations.

Based on reanalysis data from ERA5_Land, MEERA2, HadISST, NCEP-NCAR reanalysisⅠ and other datasets, the response characteristics between interdecadal Eurasia snow cover in winter and spring and sea surface temperature in Northern Hemisphere (NHSST) in winter are investigated. Barnett and Preisendorfer canonical correlation analysis (BP-CCA) and regression analysis are used in this paper. Main conclusions are as follows: In the past 40 years, except for Equatorial East Pacific region, the NHSST has increased. And except for eastern Siberia, the snow cover basically decreased. The interdecadal influence of winter NHSST on Eurasia snow cover lasts until spring after removing the linear trend. In the first mode of BP-CCA, winter and spring snow cover increase in Europe and decrease in central Eurasia when the North Atlantic and Northwest Pacific Sea temperatures rise (AMO+, PDO-) in winter. At this time, the height field and surface temperature decrease, water vapor and precipitation increase, trough deepens in Europe, and then snow cover increases in winter and spring. In central Eurasia, the variations are revised. In the second mode of BP-CCA, there is a significant reverse change between snow cover over Eurasia and Northeast Pacific SST with a quasi-16a period, and the North Atlantic SST is a tripole type. When the SST increases in the Northeast Pacific Ocean and the mid-latitude Atlantic Ocean in winter, the trough and ridge in Eurasia all weaken, surface temperature increase, and then winter snow cover decrease. In spring, the water vapor flux and precipitation decreased significantly, and snow cover decreased.

The influence of soil moisture on atmospheric precipitation has been extensively studied, but few of these studies have considered the role of land-atmosphere coupling in afternoon precipitation events at a sub-daily timescale. Here, using in-situ observations and reanalysis datasets, we investigated the effect of the soil moisture anomaly (SMA) on warm seasons’ afternoon precipitation in the North China Plain (NCP), which is identified as a strong land-atmosphere coupling region. Afternoon precipitation events (APE) were separated from all precipitation events in the NCP during the warm seasons of 2010 to 2019. It follows from a comparative analysis that APE is more likely to be initiated on drier soil, which has little dependence on the thresholds used for identifying APE. However, no affirmative relationship is found between precipitation amount in the first hour of APE (APE_1hour) and the soil moisture. Further analyses indicate that larger amounts of APE_1hour result from higher convective available potential energy (CAPE), higher moist static energy (MSE), or weaker vertical shear of horizontal wind. When considering the joint effects of SMA and atmospheric variables, APE tend to occur on drier (wetter) soil with lower (higher) lower-tropospheric stability, CAPE, or MSE. This study highlights the significant roles of land-atmosphere interactions on local atmospheric precipitation, especially the joint roles of soil moisture and atmospheric variables on precipitation.

Evapotranspiration (ET) is a key parameter regulating land–atmosphere interaction processes and the water cycle. The seasonal and interannual variability of ET and its environmental controls over an alpine meadow in a subfrigid humid zone of the Tibetan Plateau (TP) are reported on. Direct measurements were made using the eddy covariance method over a 10-year period with a significant increase in growing season length (GSL). The results showed that annual ET for the alpine meadow site was 492 ± 66 mm in comparison with 635 ± 88 mm of precipitation (P), with a ratio of ET/P ranging from 0.63 (2017) to 1.04 (2010). The path analysis and Priestley–Taylor coefficient (ET/ET_eq) revealed that the daily ET experienced energy-limited and water-limited seasons within the year. During the water-limited non-growing season, the daily surface conductance (Gs) and ET/ET_eq increased positively with soil water content (SWC). During the energy-limited growing season, the 16-day average ET, ET/ET_eq, and Gs scaled positively with the normalized difference vegetation index (NDVI). ET/ET_eq increased nonlinearly with an increase in Gs, but was insensitive to increases in Gs greater than the threshold Gs_*. The mean Gs and Gs_* were strongly regulated by the maximum NDVI. The maximum NDVI, Gs_*, and annual mean Gs explained 46%, 80%, and 68%, respectively, of interannual variation in annual ET. Thus, we concluded that biophysical factors, rather than P and GSL, mainly controlled the interannual variability in annual ET. These findings are critical for understanding the response mechanism of ET to the changing biotic and abiotic conditions in the TP.

Soil moisture is one of the critical variables in the hydro-climate system; however, the gauge-based measurement poses a great challenge in its spatial coverage. To obtain more comprehensive soil moisture data over Taiwan, this study aims to leverage the Weather Research Forecasting model-Hydrological modeling system (WRF-Hydro). To ensure a more reliable regional simulation, the default soil texture, soil hydraulic parameters, and land use/cover data in WRF-Hydro are replaced with locally curated data to better reflect the actual surface characteristics of Taiwan. The Taiwan Hydraulic Digital Elevation Model (HyDEM) is used to configure catchment characteristics in the pre-processing of WRF-Hydro. Gauge- or radar-based precipitation is used as the meteorological forcing to drive WRF-Hydro. Simulated soil moisture will be compared and validated with gauge- and satellite-based data in Taiwan. We believe that the outcome of this work will provide more effective soil moisture information for ensuing hydro-climatic modeling and disaster monitoring.

This study examines the role played by aerosol in torrential rain that occurred in the Seoul area, which is a conurbation area where urbanization has been rapid in the last few decades, using cloud-system-resolving model (CSRM) simulations. The model results show that the spatial variability in aerosol concentrations causes the inhomogeneity of the spatial distribution of evaporative cooling and the intensity of associated outflow around the surface. This inhomogeneity generates a strong convergence field in which torrential rain forms. With the increases in the variability in aerosol concentrations, the occurrence of torrential rain increases. This study finds that the effects of the increases in the variability play a much more important role in the increases in torrential rain than the much-studied effects of the increases in aerosol loading. Results in this study demonstrate that for a better understanding of extreme weather events such as torrential rain in urban areas, not only changing aerosol loading but also changing aerosol spatial distribution since industrialization should be considered in aerosol–precipitation interactions.

Effects of an aerosol layer on warm cumulus clouds in the Korean Peninsula when the layer is above or around the cloud tops in the free atmosphere are compared to those effects when the layer is around or below the cloud bases in the planetary boundary layer (PBL). For this, simulations are performed using the large-eddy simulation framework. When the aerosol layer is in the PBL, aerosols absorb solar radiation and radiatively heat up air enough to induce greater instability, stronger updrafts and more cloud mass than when the layer is in the free atmosphere. Interestingly, even though the aerosol layer in the PBL intercepts more solar radiation reaching the surface, resulting in reduced surface heat fluxes, instability, updraft intensity, and cloud mass, the overall effect still leads to more cloud mass. Hence, there is a variation of cloud mass with the location of the aerosol layer, which arises from an interplay between aerosol-induced changes in the surface fluxes and those in radiative heating of air. Even without impacts of aerosols on radiation, there is still more cloud mass when the aerosol layer is in the PBL, because aerosol-induced changes in droplet nucleation induces aerosol-induced invigoration of updrafts. However, the variation of cloud mass is much smaller when there is only aerosol impacts on droplet nucleation than when aerosol impacts on both of radiation and nucleation are present. Hence, aerosol impacts on nucleation and those on radiation work together to amplify the variation of cloud mass with the altitude of the aerosol layer. This study reveals that this variation of cloud mass also reduces as aerosol concentrations in the layer decrease. Furthermore, the transportation of aerosols by updrafts reduces aerosol concentrations in the PBL. 

In middle and high latitudes, marine stratocumulus clouds often form after cold fronts pass, with cold air flowing from the continents during winter. Because these stratocumulus clouds persist over a relatively broad region and a relatively long duration compared to clouds associated with typical low-pressure systems, they can produce significant cloud radiative forcing, which is negative owing to their low altitudes and large cloud water contents. As they move downwind, they usually experience transition to cumulus clouds with lower cloud fraction. While increasing sea surface temperature has been recognized as the main factor to control the transition, some recent studies have pointed out that precipitation, which can be modulated by aerosols, is also important to the transition. In this study, we present the physical properties of boundary-layer clouds embedded in the cold-air outbreaks over the Yellow Sea occurred from 2013 to 2023. We identify the entire cases seen on the MODIS observations during the period, a total of 119 cases. The cloud fraction peaks at the middle of the cloud trajectories and then decreases as the clouds move downwind. The cloud top temperature generally increases along the trajectories although the cloud top height is also increasing, while the cloud water path remains relatively unchanged as the clouds move downwind. Beyond analyzing the mean properties of the entire cold-air outbreak stratocumulus clouds, we select some cases and categorize them as either ‘clean’ or ‘polluted’ based on the average carbon monoxide concentration over the upwind region of the trajectories. We then illustrate the effects of pollution on the cloud evolution by comparing the properties of the two cloud groups. Additionally, we conduct a series of numerical simulations and examine in detail how aerosols influence the cloud evolution and the transition to cumulus clouds.

Precipitation is an important process that contributes to the energy budget of the earth by modulating the water cycle and also assists in controlling the concentration of atmospheric pollutants by scavenging particulate matter (PM) (i.e., wet removal). Previous studies attempted to quantify the impact of wet removal mechanisms on PM concentrations, however, they performed the research based on short-term measurements, limited sites, or a few precipitation event cases. Thus, this study showed the spatiotemporal characteristics of precipitation and the concentration of PM (i.e., PM2.5 and PM10) in six regions (i.e., Seoul, Busan, Daejeon, Daegu, Gwangju, and Jeju) individually, and quantified the impact of the wet removal on the concentration of PM by precipitation events based on long-term measurements in six regions in Korea. Precipitation and PM10 surface measurements in six regions were used for 22 years (2001-2022), while PM2.5 surface measurements were only available for 8 years (2015-2022). All measurements were made with a one-hour resolution. The spatiotemporal variabilities (i.e., yearly, seasonal, and monthly averages) of precipitation and the concentration of PM were calculated in each region. The precipitation events were identified when the threshold time interval (i.e., 2 hours) was presented within the hourly precipitation data. Based on the duration, average intensity, and accumulated amount of precipitation events, the characteristics of precipitation events were analyzed. The variation in the concentration of PM during the precipitation events and the rate of wet removal were quantified as functions of the characteristics of precipitation events (i.e. duration, average intensity, and accumulated amount of precipitation events) for each of the six regions. The spatiotemporal characteristics of the regional wet removal effects that occurred by precipitation events were calculated, and the influence of the precipitation properties on the wet removal effects was evaluated.

Meteorological phenomena are highly nonlinear, and small differences in initial values can significantly affect the results. Therefore, ensemble calculations are a commonly used computational technique in weather/climate simulations to reduce the inevitable uncertainty inherent in individual simulations and to quantitatively evaluate the degree of uncertainty. However, obtaining an average value with sufficient accuracy requires a large number of ensembles, which in turn requires significant computational resources. For example, in the case of the atmospheric model NICAM with a resolution of 3.5 km and 1024 ensembles, 131072 nodes were used, but with this number of nodes, it was not possible to calculate 1024 ensembles at once, and a method of repeating 4 cycles of 256-member ensemble calculations per cycle was employed. Therefore, in our study, we coupled a low-resolution ensemble with single high-resolution computation. This method can replace high-resolution large-scale ensembles with fewer computational resources. Furthermore, we can expect to obtain more accurate results by coupling low-resolution calculations, which excel in reproducing large-scale fields, with high-resolution calculations, which excel in reproducing detailed fields. A general-purpose coupler, h3-Open-UTIL/MP, was applied to this study. Two execution modes are possible on this coupler: many-to-one coupling, where multiple ensemble runs are coupled to a single model, and many-to-many coupling, where multiple coupled models are executed as an ensemble. In this study, the later mode was applied. We coupled low resolution NICAM ensemble and high resolution NICAM single model, on the Wisteria/BDEC-01 of the University of Tokyo. For the calculations, optimization of the single model and variable resource allocation were applied to minimize computational resources (node-time) as much as possible. In this presentation, the details of the ensemble coupling will be explained and the computational performance will be discussed.

Speeding up the post-processes in addition to the weather and climate simulation itself is important for speeding up the overall workflow. There are a wide variety of post-processes, but one typical post-process required for many models is remapping from model native grids to latitude-longitude grids. Here, we developed a series of parallelized remapping algorithms for NICAM, a global weather and climate model with an icosahedral grid system, and demonstrated their performance with global 14-0.87-km mesh model data on the supercomputer Fugaku. The original remapping tool in NICAM supports parallelization only in reading and interpolating data. In our proposed algorithms, the process of data writing is parallelized by separating output files or using the MPI-IO library, both of which enable us to remap 0.87-km mesh data with 670 million horizontal grid points and 94 vertical levels. The benchmark with 14-km mesh data shows that the developed algorithms significantly outperform the original algorithm in terms of elapsed time (by 7.4-8.7 times) and memory usage (by 2.8-5.0 times). Among the proposed algorithms, the separation of output files, along with reduced MPI communication size, leads to a better performance in the elapsed time and its scalability, and the use of the MPI-IO library leads to a better performance in memory usage. The remapping year per wall-clock day, assuming a six-hourly output interval, is up to 0.56 with 3.5-km mesh data, demonstrating the feasibility of handling global cloud-resolving climate simulation data in a practical time. This study demonstrates the importance of IO performance, including MPI-IO, in accelerating weather and climate research on future supercomputers.

The Korean Integrated Model (KIM) is a global weather forecast model independently developed by the Korea Institute of Atmospheric Prediction Systems (KIAPS). Comprising a spectral element-based dynamical core and a scale-aware physics package on cubed-sphere grids, KIM is currently deployed by the Korea Meteorological Administration (KMA) in a 12 km resolution, with data assimilation for weather forecasting. However, acknowledging the challenges in accurately simulating precipitation systems, particularly sudden and intense heavy rains around the Korean Peninsula, there is a growing need for a higher resolution model. In this study, we attempted to simulate KIM with a 3-4 km horizontal resolution, strategically avoiding cumulus parameterization specifically for cases of extreme heavy rainfall. The primary aim was to assess the possibility of positioning KIM as a Global Cloud-Resolving Model (GCRMs). As a result, KIM demonstrating a stable 10-day simulation of extreme rainfall case and more realistic compared to the 8 km simulation.

Himawari-8, as a geostationary satellite, carries Advanced Himawari Imager which can perform full-disk observations every 10 minutes. Due to coastal water being turbid, shallow, and mixed with the land, the aerosol retrieval product cannot accurately retrieve the Aerosol Optical Depth (AOD) over shallow and turbid coastal waters. To fill the gap in this area, we developed a coastal water retrieval algorithm with a spatial resolution of 2 km. The land mask is first performed by normalized difference water index (NDWI), and then the cloud mask is performed by a spatial variation test based on 2.3 μm and some traditional reflectance and bright temperature threshold tests. After that, the water pixels are classified into open ocean pixels and coastal pixels by power-law fitting algorithm and Elevation and Topography at 1 arc minute (ETOPO1) bathymetry. After performing gas correction, the AOD and aerosol properties were retrieved using the Dark-Target algorithm for the open ocean pixels. For the coastal pixels, the AOD was retrieved using the 2.3 μm top of atmosphere reflectance and the aerosol properties of the nearest open ocean pixel. This algorithm can improve the accuracy and coverage of the coastal AOD.

Since the industrial revolution, human beings have produced a large amount of air pollutants. Air pollution results in changes in the atmospheric composition, which in turn alters the energy budget of the Earth system and, consequently, causes the present-day climate crisis. The main climate active substances in the air pollutants include the greenhouse gases and atmospheric aerosols. The greenhouse gases absorb IR radiation and cause warming of the atmosphere, whereas aerosols could cause warming or cooling effects, depending on the microphysical and optical properties. The eastern Asia is among the most polluted regions in the world. Along the course of energy transition and air pollution mitigation, monitoring data provide the crucial evidence to evaluate the status of environmental changes. This study has been conducting observation of atmospheric aerosols (PM₁₀) and major greenhouse gases (CO₂, CH₄, N₂O) at three remote sites (Penghu, Green Island, Cape Fuguei) around Taiwan. The preliminary results of observation for 2018 – 2019 showed that, in terms of annual mean level, the background mixing ratios of CO₂, CH₄, and N₂O were 418 ppmv, 1.95ppmv, and 334.3 ppbv, respectively. Moreover, according to the observation data from the Penghu, Green Island and Cape Fuguei research stations, we estimated the background level of atmospheric aerosols (annual mean of PM₁₀) was 30.6 ugm^-3.

At the Meteorological Service of Singapore, Himawari geostationary data are used for transboundary haze monitoring and retrieval of Aerosol Layer Height with the application of different Random Forest architectures and Sentinel-5p/TROPOMI training data. The Numerical Atmospheric-dispersion Modelling Environment (NAME) is used for dispersal modelling and forecasting of forest fire emissions, and for further derivation of AOD. We evaluate performance of the system in recent haze incidents using the GEMS AOD product as reference. GEMS is the first satellite sensor dedicated to atmospheric monitoring in geostationary orbit, providing hourly AOD and Aerosol Layer Height information. The aim of this study is to explore the potential of GEMS to support operational, high frequency haze monitoring and validation of dispersion model outputs, specifically in the challenging environment of Southeast Asia.

Aerosols are closely associated with air quality, climate change, and public health. Moreover, aerosol concentrations exhibit significant regional variations and considerable temporal fluctuations. Therefore, studying the long-term changes in aerosol concentration and analyzing the trends are essential for air quality improvement and climate change research. In this study, we present the long-term trends of Aerosol Optical Depths (AODs) across Asia and compare the trends of Fine-mode Aerosol Optical Depths (FAODs) and Coarse-mode Aerosol Optical Depths (CAODs). AOD represents the attenuation of solar radiation due to aerosols in the entire atmosphere, serving as an indicator of aerosol quantity. The Aerosol Robotic Network (AERONET) is a ground-based observational network providing information on aerosol optical properties worldwide. AERONET holds long-term data and offers continuous observations through daily measurements. For this reason, we analyzed the long-term trends of AOD using AERONET data. To examine changes in aerosol quantity attributed to anthropogenic sources and yellow dust, we presented trends in Coarse-mode Aerosol Optical Depths (CAODs) and Fine-mode Aerosol Optical Depths (FAODs), categorizing aerosols based on size. By comparing CAODs and FAODs, common trends were identified in trend analysis across countries or sub-regions. Through this analysis, it becomes evident that aerosol concentrations vary significantly regionally. However, it is apparent that factors such as each country's local environment and air quality policies exert substantial influence on aerosol trends.

Since its launch in 2020, the GOCI-II (Geostationary Ocean Color Imager-II) onboard the GEO-KOMPSAT-2B (GK-2B) satellite has provided aerosol products using the Yonsei aerosol retrieval (YAER) algorithm (Lee et al., 2023). The GOCI-II YAER algorithm retrieves aerosol optical depth (AOD) at 550 nm using an inversion algorithm with a precalculated look-up table (LUT) over UV to near-IR wavelengths. The surface reflectance database is collected using the Cox and Munk method (Cox and Munk, 1954) and the minimum reflectance technique (Hsu et al., 2004) over ocean and land, respectively. When calculating the minimum reflectance, the minimum value of Lambertian Equivalent Reflectance (LER) of each wavelength is designated as the surface reflectance at each pixel. The 550 nm AOD is the weighted average of AOD of two aerosol types that minimize the standard deviation among the six pre-assumed types. In this study, we improved the performance of the GOCI-II YAER algorithm by optimizing the surface reflectance calculation method and tuning the aerosol type selection phase. First, the spectral AOD of the YAER algorithm was validated to the AOD from the AErosol RObotic NETwork (AERONET) to test the stability of the minimum reflectance. The wavelength on which its AOD showed the highest consistency with AERONET was used as the standard of the minimum reflectance to fix the surface reflectance of all other wavelengths. Second, aerosol type selection phase was tuned to take more aerosol optical information during the selection. As a result, the updated product showed improved validation statistics when compared to AERONET AOD in terms of % within expected error (EE), the correlation coefficient, and the root mean squared error (RMSE). The improved GOCI-II AOD can help mitigate the air quality issues and expand our knowledge of diurnal variations of aerosols over Northeast Asia.

There are remaining uncertainties from satellite remote sensing retrieval of nitrogen dioxide (NO₂), one of important pollutants for understanding urban air quality study. In this research, we analyze the impact for spatial resolution of observation pixels on retrieved differential slant column density (dSCD), regarded as one of uncertainty factor. Airborne measurement data which offers a higher resolution remote sensing data compared to satellites and useful for evaluating spatial variability were collected from KORUS-AQ and SIJAQ campaign. Comparing dSCD maps among different spatial resolutions, we confirm that some point sources and structures of NO₂ density distribution were diluted and smoothed at a low resolution. Also, we confirm increase of standard deviation values that shows deviation of information on pixels over 2 km spatial resolution. In future, we can study the impact of spatial resolution on other trace gases that has different spatiotemporal variability. Furthermore, we can advance the understanding of satellite data according to spatial resolution.

This study aims to quantify NO_x lifetimes and emissions in Seoul, South Korea, using data from Geostationary Environment Monitoring Spectrometer (GEMS) tropospheric NO₂ observations in 2022 along with ECMWF wind field data. NO_x lifetimes are determined by analyzing systematic difference obtained through fitting NO₂ line density to a model function, considering wind directions and wind speed. The results indicate that, with a fitting interval of 100×100km², the calculated lifetimes ranged from 14.14 to 50.19 minute in spring, 16.23 to 53.61 minute in summer, 19.32 to 62.89 minute in autumn and 24.2 to 62.02 minute in winter (R²>0.8). When the fitting interval is 100×200km², lifetime ranged from 34.05 to 82.61 minute in spring, 33.81 to 80.28 minute in summer, 30.13 to 94.07 minute in autumn and 61.11 to 94.89 minute in winter (R²>0.8). The quantified NO_x emissions for 2022 were determined based on the lifetime fitting results, ranging from 210.66 Gg to 714.93 Gg with a 100×100km² lifetime fitting interval and from 174.33 Gg to 390.88 Gg with 100×200km² lifetime fitting interval. To verify these results, we used Clean Air Policy Support System (CAPSS) emissions inventory data. The mean of NO_x emissions for 2016-2020 within the fitting interval area was 203.7±21.22 Gg. In conclusion, adopting a 100×200km² fitting interval provided better visibility of the movement of the NO₂ plume, resulting in longer calculated lifetimes. The study acknowledges the high uncertainty in lifetime fitting results during winter and emphasizes the careful interpretation of winter emission results. Future studies are proposed to compare NO_x lifetimes and emissions in other East Asian regions for verification. This work was supported by Korea Environment Industry & Technology Institute (KEITI) through "Climate Change R&D Project for New Climate Regime", funded by Korea Ministry of Environment (MOE) (2022003560006).

Trans-Pacific transport is important because it not only affects air quality in East Asia but also in North America. In April, trans-Pacific transport is associated with the Western Pacific pattern (WP). When the WP is positive phase, favorable atmospheric circulation for trans-Pacific transport is formed over the North pacific, leading to an increase of the Aerosol Index (AI) over the North America. Climate change leads to extreme weather phenomena. WP is no exception. Changes in the WP due to climate change are expected to impact trans-Pacific transport as well. Coupled Model Intercomparison Project phase 6 (CMIP6) makes possible to examine the changes associated with climate change and thus predicting the changes in air quality. In this analysis, we investigated whether CMIP6 simulated the spatiotemporal pattern of Aerosol Optical Depth (AOD) according to WP and examined the tendencies of the WP phase under different Shared Socioeconomic Pathways (SSPs). CMIP6 effectively simulated the changes in atmospheric circulation and the AOD associated with the variations in the WP phase. When comparing the four SSPs (SSP1-2.6, SSP2-4.5, SSP3-7.0, SSP5-8.5), the frequency and intensity of positive WP phase were highest in the highest CO₂ emission scenario, SSP5-8.5. This suggests that the higher carbon emission, the more frequent and intense trans-Pacific transport. Further investigation is needed to understand the underlying reasons for increased frequency and intensity of positive WP phase in high carbon emission. Additionally, the analysis should be conducted using more model data. This work was supported by Korea Environment Industry & Technology Institute (KEITI) through "Climate Change R&D Project for New Climate Regime", funded by Korea Ministry of Environment (MOE) (2022003560006).

PM_2.5 denotes aerosol particles in the air with aerodynamic diameter of less than 2.5μm. PM_2.5 contributes to social and economic challenges and negatively affects human health. Research into the temporal and spatial variability of PM_2.5 is crucial for its effective management. The Ministry of Environment has operated a ground-based observation network, AirKorea, since the early 2000s to continuously monitor air pollutants, including PM_2.5 and its precursors (NO₂, SO₂, etc.). This study analyzed diurnal variations in PM_2.5 concentrations in South Korea using data from AirKorea observation network from 2016 to 2021. In South Korea, PM_2.5 levels typically begin to rise at 8 a.m. during rush hour, peak around 10 a.m., and then decrease after noon as radiation levels increase. A subsequent increase is observed starting around 6 p.m. during the evening rush hour, with another peak occurring at night. Many previous studies have indicated that the morning peak in PM_2.5 is influenced by increased anthropogenic emissions, and the afternoon decrease is affected by an increase in Planetary Boundary Layer (PBL) height. However, research specifically investigating the nighttime peak is limited. In this study, to determine the cause of the PM_2.5 nighttime peak in South Korea, we analyzed the differences in meteorological conditions (PBL height, wind speed, relative humidity) and air pollutants (PM₁₀, NO₂, SO₂) on days with the PM_2.5 nighttime peak compared to days without it. Additionally, we used PM_2.5 compositional data to compare the concentrations of its constituents on days with the nighttime peak to those on days without such peaks in Seoul, Ulsan, Jeju, Gwangju, and Daejeon. This work was supported by Korea Environment Industry & Technology Institute (KEITI) through "Climate Change R&D Project for New Climate Regime", funded by Korea Ministry of Environment (MOE) (2022003560006).

This study explores the key energy source between the Earth and the atmosphere, solar radiation, which directly and indirectly influences the weather. Changes in the Earth's system impact radiative equilibrium, and the concept of radiative forcing (RF) was introduced to quantify these effects. RF is influenced by specific atmospheric components such as trace gases, with significant impacts primarily attributed to changes in clouds and aerosols. To understand direct aerosol radiative forcing (DARF) caused by aerosols, we developed an algorithm considering aerosol optical depth (AOD) and aerosol types using the Geostationary Environment Monitoring Spectrometer (GEMS) on the GK-2B (Geostationary Korea Multi-Purpose Satellite-2B) satellite. GEMS AOD algorithm version 2.0 and the libRadtran version 2.0.4 radiative transfer model were employed as input data. Aerosol types were classified as Black Carbon, dust, and non-absorbing aerosol, and the OPAC model was used for DARF calculations based on aerosol types. Surface (SFC), top of atmosphere (TOA), and atmospheric (ATM) DARF spatial distributions were examined in major Korean cities. To obtain city-specific DARF time series from November 2021 to April 2023, city boundaries were defined, and the average AOD values of all GEMS pixels within each boundary were used as input data. For validation purposes, a comparison with Aerosol Robotic Network (AERONET) inversion data for RF was also conducted. This study aims to enhance understanding of air pollution issues in South Korea and the variation characteristics of DARF in different regions throughout the day. This work was supported by Korea Environment Industry & Technology Institute (KEITI) through "Climate Change R&D Project for New Climate Regime", funded by Korea Ministry of Environment (MOE) (2022003560006).

The hydroxyl radical (OH) is a pivotal oxidant in the troposphere, playing a crucial role in determining the lifetime of critical pollutants such as carbon monoxide (CO) and methane (CH4). The primary generation of tropospheric OH occurs through the photolysis of ozone (O3) in the presence of water vapor (H2O), making tropospheric OH's abundance and oxidative capability heavily dependent on diurnal patterns. Yet, due to its ephemeral lifetime of approximately one second and its scant concentration in the troposphere, direct measurement of tropospheric column OH (TCOH) and its diurnal fluctuation remains unattainable through global satellite monitoring. Efforts have been made to use satellite observations as a proxy to study OH variability on temporal scales longer than a day, often in conjunction with chemistry transport or machine learning models. With the launch of the Geostationary Environment Monitoring Spectrometer (GEMS) in February 2020, which provides hourly measurements of atmospheric pollutants like O3, nitrogen dioxide (NO2), and formaldehyde (HCHO) over Asia, our understanding of these pollutants' diurnal variations has significantly advanced. This study introduces a method for approximating the diurnal variation of TCOH by employing satellite data on OH precursors, including the hourly-changing observations from GEMS. Utilizing training data from satellite observations and TCOH values from chemical composition reanalysis fields, the supervised machine learning model adeptly predicts the diurnal variation of TCOH when provided with inputs of OH precursors. The findings also indicate a promising avenue for integrating this approach into high-resolution CH4 modeling for East Asia.

Volatile organic compounds (VOCs) are precursors to surface ozone and secondary organic aerosols. In East Asia, simulations from chemical transport models often miss significant VOCs due to uncertain emission sources. Formaldehyde (HCHO), a byproduct of VOC oxidation, can act as an indicator to provide observational constraints on VOC emissions. The Geostationary Environment Monitoring Spectrometer (GEMS) commenced hourly monitoring of HCHO vertical column densities (VCDs) over East Asia starting in August 2020. This study assesses the updated operational GEMS HCHO retrieval algorithm (version 3) by comparing it with TROPOMI and ground-based observations (FTIR, MAX-DOAS) throughout its three years of operation. We introduce several enhancements to the retrieval algorithm, particularly regarding background corrections and the input parameters that influence air mass factor (AMF) calculations. We observe that utilizing hourly background surface reflectance from GEMS, rather than OMI Lambertian equivalent reflectivity, results in 20–30% lower AMFs over land areas in East Asia. The revised GEMS HCHO data generally align with the spatial distributions observed by TROPOMI HCHO (r=0.67–0.88), despite slightly underestimating VCDs (NMB=-28% to -17%) across the scanning domain. Additionally, we explore factors affecting AMF sensitivity using HCHO vertical profiles obtained from airborne measurements during the KORUS-AQ campaign. We note that certain geometric conditions, particularly in the early morning (7–8 KST) and afternoon (15–16 KST), significantly impact the AMF, potentially increasing reliance on model accuracy in VCD calculations. Our findings underscore the necessity of precise model simulations in calculating HCHO AMF.

Nitrogen dioxide (NO₂) is a gaseous atmospheric pollutant that contributes as catalysts to the formation of ozone in the troposphere and plays a significant role in atmospheric chemistry. The ground-based Multi-Axis Differential Optical Absorption Spectroscopy (MAX-DOAS) have been employed to investigate tropospheric trace gases (including NO₂) and aerosols at a fixed location with high temporal resolution and accuracy. The continuous measurements of ground-based MAX-DOAS are advantageous in effectively observing the high temporal variations of NO₂ Tropospheric Vertical Column Density (NO₂ TropoVCD). Aiming to obtain the reliable NO₂ TropoVCD in Seoul Metropolitan Area (SMA), two MAX-DOAS have been operated in Seoul (37.59°N, 127.03°E, 80 m) and Yongin (37.34°N, 127.27°E, 167 m), downwind of Seoul, since October 2023 and October 2021, respectively. During Airborne and Satellite Investigation of Asian Air Quality (ASIA-AQ) campaign, the purposes of this study are summarized as follows: (1) validation of NO₂ TropoVCD from Geostationary Environmental Monitoring Spectrometer (GEMS) using the MAX-DOAS, (2) comparison of the differences in MAX-DOAS NO₂ TropoVCD between two sites, (3) assessment of NO₂ advection from Seoul to Yongin using the NO₂ vertical profile. We believe these results will contribute to improving the accuracy of GEMS NO₂ products and the understanding of NO₂ behavior in SMA.

In this study, we developed a ground-based sulfur dioxide (SO2) Slant Column Density (SCD) retrieval algorithm based on Differential Optical Absorption Spectrometer (DOAS) from hyperspectral measurements to monitor industrial stack plumes. Our algorithm has an advantage of a capability to continuously and remotely monitor industrial SO2 plumes rather than in-situ SO2 measurements. Also, it has a valuable advantage that we know uncertainty of instruments we use and experimental situation we have approximately. And we calculated the optical path length using the observation geometry for mixing ratio retrieval. In a industrial stack plume observation situation, the optical path length that considers the distance between the instrument and the stack and the plume thickness together can increase the accuracy of the mixing ratio retrieval. The experimental data to test this algorithm of SCD and mixing ratio was obtained in Haman, South Korea in August, 2022 and Gimhae, South Korea in November 2023. Since most in-situ measurements has a product in the unit of mixing ratio, it has a valuable advantage of being able to confirm the accuracy of observation through comparison of the retrieved mixing ratio and in-situ.

Accurate knowledge of mesospheric winds and waves is essential for studying the dynamics and climate in the mesosphere and lower thermosphere (MLT) region. In this study, we conducted a comparative analysis of the mesosphere tidal results obtained from these two adjacent meteor radars at low latitudes in Kunming， China from November 2013 to December 2014 were analyzed and compared. These two radars operate at different frequencies of 37.5 MHz and 53.1 MHz, respectively. However, an overall good agreement was observed between the two radars in terms of horizontal winds and tide observations. The results showed that the dominant tidal waves of the zonal and meridional winds are diurnal and semidiurnal tides. Moreover, we conducted a statistical comparison of tide amplitudes and phases observed by both radars, revealing a high degree of correlation between the tidal parameters. Additionally, the variances and covariances of tidal amplitudes and phases were found to be quite similar for the two radars. Thus, mesospheric tides at low latitudes observed by two collocated meteor radars with different frequents exhibit strong consistency.

Various mean-flow oscillations in the equatorial middle atmosphere can be generated by interaction between the zonally averaged mean flow and numerous atmospheric wave modes. Quasi-biennial oscillation (QBO) is one of the most well-known examples in the wave-mean interaction in the tropical lower stratosphere. However, the QBO can be modulated by momentum forcing due not only to equatorial waves but also to horizontally propagating Rossby waves that originate from the extratropical region. In the northern winters of 2015/2016 and 2019/2020, the downward progression of westerly QBO phase was interrupted by localized and intensified easterlies, mainly due to exceptionally large wave momentum forcing due to extratropical Rossby waves. In this study, global wave modes represented by Hough and vertical structure functions are obtained from MERRA-2 reanalysis data by decomposing global winds and temperature into components associated with Kelvin, Rossby-gravity, inertia-gravity, and Rossby waves. Then, resolved wave momentum forcing (Eliassen-Palm (EP) flux divergence) due to each wave mode is estimated. To compute EP flux and its divergence, methods of obtaining temperature perturbations and vertical wind velocity induced by each wave mode have been developed. Comparison of EP flux divergence for climatological QBO indicates that dominant eastward (westward) wave momentum forcing is due to Kelvin waves (Rossby waves). The QBO momentum budget in terms of zonal wind tendency due to advection, the resolved wave momentum forcing due to each wave mode, analysis increments, the parameterized gravity wave drag, and the residual tendency will be discussed together with wave momentum budget in the disrupted QBO periods.

Many researches over the past decades have been shown that OH airglow known as the peak emission altitude of about 87km is very useful to study not only the propagation and dissipation of atmospheric gravity waves but also the mesospheric temperature variation with time. Recently, KOPRI (Korea Polar Research Institute) has started to operate Mesospheric Temperature Mapper (MTM) at King Sejong Station, Antarctica (62.2S, 58.8W) since its installation on Jan. 2023, in order to study the small-scale gravity wave activities and temperature variation near mesopause region. MTM instrument provides two dimensional OH airglow intensity and rotational temperature every 30s with a high spatial resolution of ~0.5km.In this study, we introduce the KSS MTM instrument and present some preliminary results which compare the temperatures obtained from MTM for 26 clear nights during 2023 austral winter season with the temperatures from co-located KSS Meteor Radar and Aura/MLS satellite.

This study presents regional climate simulations for East Asia and China using the Model for Prediction Across Scale-Atmosphere (MPAS-Atmosphere) driven by ERA-Interim reanalysis. In a 29-year simulation (1988-2016), MPAS-A with a global variable resolution (VR) configuration (92-25km mesh refinement over East Asia) is evaluated for precipitation, near-surface air temperature, and circulation against observed climate using combined observational datasets. Large-scale deviations, such as the northward displacement of rain belts, excessively triggered precipitation on the ocean, and stationary surface air temperature biases, are identified. These deviations are attributed to simulated circulation, moisture transports. Focusing on the summer monsoon with the highest uncertainty in East Asian climate, comparative experiments are conducted using a similar 92-25km grid under global variable resolution settings (MPAS-VR) and limited-area settings (MPAS-RCM) for simulations from April to August during 1998-2017. Constrained by the “reanalysis-based” lateral boundary conditions (LBCs), MPAS-RCM could well capture the evolution of large-scale fields and becomes a proper reference for evaluating MPAS-VR. MPAS-VR reproduced the conclusions of the previous 29-year simulation. MPAS-RCM, benefiting from a more realistic water vapor transport, alleviates the wet bias and presents better daily variations and rain belts’ evolution. However, MPAS-RCM did not exhibit a clear advantage in the distribution of climate indices related to extreme events over land. Additionally, its performance in terms of climatology and interannual variability of the surface temperature are insufficient compared to MPAS-VR. With the same dynamic core and model physics, those climate features are similarly resolved by the two simulations. the study underscores the robustness and potential of the VR approach, particularly regarding extreme rainfall and heat wave.

This study presents an in-depth evaluation of the Weather Research and Forecasting (WRF) model's performance in the context of gray-zone (9 km) dynamical downscaling over China, emphasizing the added value it brings to regional climate modeling. The term "gray-zone" refers to scales where traditional parameterization schemes are neither fully effective nor entirely redundant, a common challenge in atmospheric modeling. Our work bridges this gap, focusing on the transitional scale phenomena. The research utilized the advanced WRF model, incorporating recent updates in its physics and dynamics schemes, to conduct high-resolution simulations over China. The study period spanned multiple years to capture a comprehensive range of meteorological conditions, including extreme climate events. Our evaluation criteria involved a comparative analysis of the WRF model outputs against observational datasets and ERA5 reanalysis to assess its performance and added value. Key findings demonstrate that the WRF model, when applied in the gray-zone, significantly enhances the representation of mesoscale meteorological features compared to ERA5. This improvement was particularly notable in complex terrain areas, where the model accurately captured local near-surface air temperature patterns and precipitation distributions. Furthermore, the model showed a marked improvement in simulating extreme climate events, such as heavy rainfall and heatwaves, which are critical for regional climate impact assessments. The added value of the WRF gray-zone downscaling is evident in its ability to provide more accurate and detailed climate information at a regional scale. This is particularly beneficial for China, where diverse topography and climatic conditions pose a challenge for conventional models. The significant advancements and potential of gray-zone dynamical downscaling using the WRF model over China facilitate better decision-making in sectors like agriculture, water resource management, and urban planning, where precise climate data is crucial.

Future dry-wet changes of Northwest China and their mechanisms remain controversial. Therefore, this work projected seasonal and annual dry-wet conditions in the arid and semi-arid regions by analyzing the variation of water availability, which is defined as precipitation minus evaporation, based on the downscaled future regional climate change under 1.5/2.0°C stable warming scenarios (1.5s/2.0s) using Weather Research and Forecasting (WRF) model. The results showed that, the water availability in arid increased by 1.09 and 1.24 mm/month under future warming scenarios of 1.5s and 2.0s, respectively, while in semi-arid, the increase was lower than in arid and even decreased in summer. The results of changes in moisture transport suggested that the stronger increase in water availability in arid may be related to the inflow of moisture from arid and the outflow from semi-arid in summer. The moisture budget analysis further demonstrated that, the increase of water availability in arid was mainly due to the enhanced contribution of the thermodynamic term caused by warming, which further increased by 8% with an additional warming of 0.5°C. Whereas the decrease in summer semi-arid was from a negative contribution of the non-linear term, which may be related to reduced moisture transport due to changes in the East Asian summer monsoon.

In this study, Weather Research and Forecasting (WRF) based dynamic downscaling simulation was performed over China at a horizontal resolution of 25 km. To reduce the systematic large-scale biases from lateral boundary conditions, a dynamic blending (DB) technique was introduced in downscaling, and its performance was compared with downscaling without DB (NO_DB) as well as the driving global climate model (GCM, i.e., HadGEM3). In the present-day simulation for verification, added values were found in DB, relative to the GCM and NO_DB, in simulating the precipitation extremes, especially over southeastern China. Possible causes responsible for this improvement were further analyzed. In the GCM, excessive moisture and atmospheric heating in the upper troposphere in conjunction with abnormally strong deep convection resulted in excessive extreme precipitation over southeastern China, while in NO_DB, insufficient moisture and atmospheric heating in the whole troposphere in conjunction with abnormally weak convection suppressed extreme precipitation there. In comparison, DB showed a closer representation of observations in terms of vertical velocity and vertical profiles of atmospheric moisture and heating, accounting for the improved simulation of extreme precipitation. In the future projection under the SSP5-8.5 scenario, both the GCM and DB predicted that extreme precipitation would increase in most parts of China; however, DB indicated a larger increase over southeastern China relative to the GCM. Stronger moisture flux convergence over southeastern China in DB accounted for this larger increase, and in addition, the thermodynamic effect associated with more precipitable water dominated the stronger moisture flux convergence.

High-resolution climate simulations has gained importance in recent climate change impact studies. This approach allows for the investigation of the "gray zone," denoting resolution higher than 10 km where model exhibit uncertainties. Therefore, our objective is to determine the added value high-resolution model simulations over the Peninsular Malaysia based on high resolution simulations using RegCM4 driven by the Fifth Generation of the European Centre for Medium-Range Weather Forecasts (ECMWF) Reanalysis (ERA5). Two one-way nested domains were employed. The first domain covers the entire Southeast Asia and was configured at 25km grid resolution. The second domain covers the entire Peninsular Malaysia and was configured at 5km grid resolution. Given the complex resolution of Peninsular Malaysia there is potential for significant advantages in adopting high-resolution simulations. Climate Hazards Group InfraRed Precipitation with Station (CHIRPS) and Climate Hazards Group Infrared Temperature with Stations (CHIRTS) data were used as references data for precipitation and temperature respectively. The analysis covers the period 1990 to 2014 and extreme climate indices were calculated using Expert Team on Climate Change Detection and Indices (ETCCDI). Extreme precipitation used were RX1day, RX5day, CWD, CDD, R95Pptot, R99ptot and SDII. Fo temperature, it is TNN, TXN, TXX, TXN, TN90P, TN10P, TX10P and TX90P. The 5km resolution model shows remarkable improvements in simulating rainfall distribution and characteristics during the DJF season in the Northeastern part of Peninsular Malaysia. High-resolution 5km models demonstrate better capabilities in simulating extreme wet indices. Additionally, the high resolution simulations also improve the rainfall diurnal cycles over Peninsular Malaysia. The 5km resolution model also shows enhancements in simulating extreme temperatures, especially over the high-altitude regions. These findings show the relevancy of the high resolution model in simulating climate extremes. Moreover, it highlights the need of high resolution simulation in studying the climate over a complex topography in Peninsular Malaysia.

Tibetan Plateau (TP, with the height > 3000 m) is a region with complex topographical features and a large diversity of climate both in space and time. Future climate change over TP and the surrounding areas is investigated based on the ensemble of a set of the 21st century climate change projections using a regional climate model, RegCM4. The model is driven by five different GCMs at a grid spacing of 25 km. Results show the RegCM4 greatly improves the temperature and precipitation simulations by providing finer scale spatial details of them over the region. The topographic effects are well reproduced by RegCM4 but not the GCMs. General warming and increase in precipitation are found in both GCM and RegCM4 simulation, but with substantial differences in both the spatial distribution and magnitude of the changes. For temperature, RegCM4 projected a more pronounced warming in DJF over TP compared to its surrounding areas. The increase of precipitation is more pronounced and over the basins in DJF for RegCM4. For the extreme indices of snowfall, RegCM4 generally reproduces the spatial distributions although with overestimation in the amount. General decreases in SNOWTOT and S1mm, with greater magnitude over the eastern part are projected. Both S10mm and Sx5day show decrease over the eastern part but increase over the central and western parts. Notably, S10mm shows a marked increase (more than double) with high cross-simulation agreement over the central TP. Significant increases in all four indices are found over the Tarim and Qaidam basins, and northwestern China north of the TP. The projected changes show topographic dependence over the TP in the latitudinal direction, and tend to decrease/increase in low-/high-altitude areas.

This study presents future projection of extreme temperature over the Korean Peninsula (KP) under the global warming levels (GWLs) of 2.0℃ and 1.5℃. For projection, bias-corrected large ensemble of the Regional Climate Model in CORDEX-East Asia Phase 2 is used. Under GWL of 2.0℃ (1.5℃), extreme minimum and maximum temperature intensity indices (TNn and TXx) would increase by 2.61℃ (1.91℃) and 2.38℃ (1.58℃) compared to present period. Extreme temperature frequency indices (TN10p and TX90p) would decrease 17.7 days (14.0 days) and increase 30.9 days (20.1 days). In addition, extreme cold and warm events occurring once every 20 years during present period are expected to change a reoccurrence of 28.9 years (22.6 years) and 5.9 years (7.2 years). Meanwhile, there are several noteworthy points in the projection. The extreme temperature over the KP would increase over the northern KP than the southern KP, which is more distinct in TNn than TXx. The spatial pattern in TX90p is similar from that of TXx but the pattern in TN10p is different from that of TNn. In other words, a smaller increase in TN10p is expected over the inland KP.

This study comprehensively examines the evolving feedback mechanisms shaping water availability in Africa, particularly emphasizing the critical need for adaptive measures in response to severe drought events. Drawing from an extensive dataset that incorporates satellite-based information, observational data, CORDEX AF44 RCP8.5, and CMIP6 SSP585 scenario, the research employs advanced indices, including the Standardized Precipitation-Evapotranspiration Index (SPEI). Methodologically, the investigation analyzes drought trends spanning 1983-2005 and projects into the future (2051-2100). Noteworthy findings reveal a conspicuous reduction in drought severity anticipated from the 2070s-2080s afterward, underscoring the necessity of addressing evolving drought conditions. The study highlights the intricate nexus between water availability and soil moisture; it even recorded a positive sensitivity coefficient, emphasizing the pivotal role of considering soil moisture feedback and moisture convergence in estimating water availability. Mean flow convergence is a substantial contributor, constituting approximately 55% of the overall moisture convergence across the African continent. These nuanced insights advocate for a holistic understanding of evolving feedback mechanisms, emphasizing the necessity for proactive measures to mitigate climate-induced water scarcity. This research has significant implications for informing effective water resource management and adaptive strategies, especially in the context of prolonged drought episodes in Africa.

In East Asia, about half of the annual precipitation occurs during the summer, with the East Asian Summer Monsoon (EASM) region being affected by various temporal scale variability factors, making its prediction challenging. Therefore, understanding the characteristics of the East Asian summer precipitation and improving simulation performance through optimal experimental design are essential meteorological tasks. In particular, summer precipitation in East Asia, mainly convective, requires consideration of heat, moisture, and momentum exchanges between the land, ocean, and atmosphere, and mixing processes in the boundary layer triggered by the complex topography of East Asia. Hence, elements such as the Cumulus Parameterization Scheme (CPS) and the Planetary Boundary Layer (PBL) scheme significantly impact the simulation of these processes. The 2022 East Asian summer precipitation was similar to the normal year, but the Korean Peninsula recorded the highest hourly rainfall since observation by the Korea Meteorological Administration (KMA) began, showing a different pattern from the normal year. This study examined the differences in the simulation of summer precipitation in the East Asian region in 2022 using the Weather Research and Forecasting (WRF) model based on PBL schemes and CPS combinations. The PBL scheme is vital in describing the interactions between convection and horizontal flow within the boundary layer. Among various PBL schemes, this study employed the nonlocal Yonsei University (YSU) scheme and the combined local and nonlocal closure Asymmetric Convective Model version 2 (ACM2) to compare the simulation performance and precipitation characteristics according to different physical schemes. We also investigated the impacts on weather prediction through differences in the vertical structure and large-scale environmental field.

In this study, the performance of five RCMs and their ensemble mean for present-day climate simulations are evaluated. RCMs have relatively high performance for climatology patterns for Korea peninsular, East China and Japan, while relatively low performance is observed for the Tibetan plateau and India. Future climate changes under four SSP scenarios are analyzed for late 21st century (2081 ~ 2100) compared to present day (1995 ~ 2014). East Asia is expected to experience temperature increases of 2.4℃ to 6.2℃ and precipitation increases of 6.7% to 12.6%, with stronger changes in higher-emission scenarios. Among the five RCMs, HadGEM3-RA projects the largest increase in temperature while GRIMs is characterized by the strongest increase in precipitation. In line with mean warming rates, warm extreme days (TX90p) are projected to increase by 35.7 ~ 93.3 days and cold extreme days (TN10p) are projected to decrease by 23.4~35.2 days. The results of this study can be used as a reference for future detailed analyses of East Asian climate changes and its impacts as well as for assessing the importance of carbon neutrality.

In the climate scientific community, there has been a recent emergence of deep learning as a new approach for downscaling large-scale atmospheric fields to a regional scale. Our study focuses on the added value of integrating a coupled-monsoon-trained deep learning architecture in downscaling the drivers of Compound heatwave and drought (temperature and precipitation) during the passage of the West African monsoon flow. In our study, we design three deep learning frameworks (CNN, UNET, and UNET++) following the perfect prognosis approach with a primary target of exploring each model's transferability ability in warm and cold conditions. Our study's findings highlight the significant contribution in terms of the added value of deep learning coupled-monsoon architecture to high-elevation regions, particularly in the central Sahel region. Conclusively, by implementing this approach, we intend to improve the accuracy of climatic information for present and future compound events risk over the West Africa region.

In forthcoming scenarios, the African continent is expected to face significant challenges in water availability due to soil dryness and low air aridity index, with higher vapor pressure deficits (VPD). This study examines the complex interaction relationship between soil moisture (SM), precipitation (P), and evapotranspiration (E), and how they relate to drought. Specifically, the focus is on understanding changes in SM, VPD seasonal dynamics, and drought conditions using the Standardized Precipitation-Evapotranspiration Index (SPEI). Using various data sources such as observations, satellite-based data, reanalysis data, CORDEX RCMs under the Africa 44 domain, and CMIP6 GCMs, we analyze historical 1981-2005 and future climate scenarios 2051-2100. Positive feedbacks in the SM-(P-E) relationship show a positive varying coefficient, except for an unusually high level observed in satellite data. According to the SPEI with a 1-month interval, a shift from wet to dry conditions is projected around 2085. Future conditions under RCP8.5 suggest less severe "Extreme" and "Very" dry periods compared to the past, while the study recorded GCMs indicate a higher intensity and frequency of drought than RCMs. Additionally, VPD is notably active in the Sahel and Southern Africa, with future projections anticipating a stronger relationship between SM and VPD in summer compared to winter. These findings underscore the pressing need for proactive water resource management strategies in the face of anticipated challenges, emphasizing the importance of understanding the intricate dynamics between land-atmosphere interaction for effective drought mitigation and adaptation efforts.

As temperatures rise, businesses and households will need more energy to keep cool, increasing the cooling demand. Several studies used climate model simulations to show that climate change will likely increase cooling demand in many parts of the world. Climate models are the primary source of long-term climate information and are known to have biases. However, these biases are not explicitly discussed while investigating the cooling demand. The present study addresses how climate model biases and bias adjustment of heat stress may play a role in assessing cooling demand. Cooling degree days are calculated as a proxy for temperature and human discomfort to excess heat to investigate the cooling demand. Estimation is performed using climate information from an ensemble of general circulation models (MRI-CGCM3, MIROC5, CCSM4), reanalysis data (JRA25), and regional climate models (NHRCM, NRAMS, WRF) forced by general circulation models and reanalysis before and after the bias adjustment. Results are validated using cooling degree days based on observed temperature data from NARO. Reduction in temporal deviations for cooling degree days based on forcing data and corresponding regional climate models is noticed after the bias adjustment, indicating its effectiveness. Detrended cooling degree days suggest that changes in fluctuations after bias adjustment may be attributed to reduced temperature biases. The change is more prominent in cooling degree days determined from forcing data, indicating that the biases in coarser resolution data may significantly impact estimating cooling demand more than in regional climate simulations. These findings show how model biases and bias adjustments may affect the representation of cooling demand, highlighting the need to consider such uncertainties in future studies.

Compound hot extremes (CHEs)—the concurrence of daytime and nighttime heat—have been increasing under anthropogenic warming, causing serious damage to human society and ecosystems. However, the anthropogenic fingerprint in past and future changes in daily CHEs and the corresponding fingerprint-constrained population exposure remain unclear. Here, using a fingerprint-based detection and attribution method, we quantify contributions of different external forcings to the historical increase in CHEs by defining three daily-scale metrics: the probability ratio (PR) of CHEs, and the proportion of CHEs in the number of extreme hot days/nights (PTday/PTnight). All three metrics have increased significantly by 98.04%, 139.67%, and 141.91% globally from 1950 to 2014, especially in Europe and North America. We find it is very likely (>90%) for human to leave fingerprints in increasing daily CHEs. The increases in PR, PTday, and PTnight that are attributable to greenhouse gas emissions are 113.62%, 144.88%, and 145.26%, respectively. By the end of the 21st century under a high-emission scenario, our fingerprint-constrained projections indicate that these metrics would increase to approximately 131.23, 0.41, and 0.43, respectively, which is a significant reduction in magnitude and uncertainty, relative to the raw projections. In some regions such as the mid and high-latitudes, almost all daytime or nighttime extreme-heat events would become CHEs. We further find that CHEs disproportionately affect densely populated areas in fingerprint-constrained projections of population exposure. Our results indicate that adaptive measures are required to alleviate the increasing proportion of CHEs and the disproportionate population exposure in densely populated areas.

The Madden-Julian oscillation (MJO) and the quasi-biweekly oscillation (QBWO) are prominent components of the intraseasonal oscillations over the tropical Indo-Pacific Ocean. This study examines the tropical cyclone (TC) genesis over the Bay of Bengal (BOB) and the South China Sea (SCS) on an intraseasonal scale in May−June during 1979−2021. Results show that the convection associated with the two types of intraseasonal oscillations simultaneously modulates TC genesis in both ocean basins. As the MJO/QBWO convection propagated, TCs form alternately over the two basins, with a significant increase (decrease) in TC genesis frequency in the convective (non-convective) MJO/QBWO phase. Based on the anomalous genesis potential index associated with the MJO/QBWO, an assessment of the influence of various factors on TC genesis is further assessed. Middle-level relative humidity and lower-level relative vorticity play key roles in the MJO/QBWO modulation on TC genesis. The MJO primarily enhances large-scale cross-equatorial moisture transport, resulting in significant moisture convergence, while the QBWO generally strengthens the monsoon trough and induces the retreat of the western North Pacific subtropical high, increasing the regional lower-level relative vorticity. The potential intensity and vertical wind shear make small or negative contributions. This study provides insights into the neighboring basin TC relationship at intraseasonal scales, which has a potential to improve the short-term prediction of regional TC activity.

Under global climate warming and local anthropogenic influence, urban areas are threatened by increasingly serious heat extremes across the globe, posing a great threat to human health and the survival of city residents. Using global climate reanalysis and local daily air temperature datasets since the 1960s, we disentangle the synergies between urban heat island (UHI) and compound heat extremes in Chinese megacities of different climate zones. We find that the heat extremes in China have nearly tripled in frequency, intensity, and duration since this century. The nighttime- and compound-heat extremes remarkably amplified in cities, and account for 76.3% of extreme heat events in the past decade. After classifying the weather stations into urban and non-urban types with LULC maps and quantifying the urban impacts of amplified heat extremes in four metropolitan areas of China, we find the BTH region seems to be the most vulnerable region suffering from urbanization-accelerated heat extremes, with a relative contribution of the frequency, intensity, and duration by nearly 55.3%, 51.7%, 55.4%, respectively. Overall, urbanization has a positive contribution to the frequency, duration, and intensity of high temperatures in the BTH, YRD, and SCB regions. The PRD does not clearly match, possibly due to the strong influence of land-sea circulation and the geographical location near the equator. Urbanization-accelerated heat extremes are more pronounced in temperature climate zones, which may be attributed to the stronger UHI intensity resulting from lower wind speed and less evapotranspiration.

Sub-seasonal prediction of regional compound heatwaves and their predictability sources remain unclear. In this study, the underlying mechanisms associated with a long-lasting compound heatwave occurred from 1-18 July 2010 over southern China, and the major sources of its sub-seasonal prediction skill are identified. The result shows that both the development and decay of this compound heatwave are mainly dominated by atmospheric processes (i.e., adiabatic heating associated with anticyclonic circulation), whereas the land-atmosphere coupling processes play an important role in sustaining the heatwave. Further analysis indicates that the tropical intra-seasonal oscillation with periods of 30-90 days and 10-30 days produce anticyclonic circulation over southern China and facilitate the occurrence and maintenance of the heatwave during its entire and second half periods, respectively. The National Centers for Environmental Prediction (NCEP) Climate Forecast System version 2 (CFSv2) shows a low skill in predicting the 2010 compound heatwave over southern China when the lead time is longer than 2-pentad leads, which is mainly attributed to the model’s bias in representing the intensity and phase of intra-seasonal oscillations.

The frequency and intensity of summer heatwaves have been increasing in China over the past few decades. Long-term records of such extreme events can be used to put the recent changes in a long-term context. That is essential to understanding the drivers of the long-term variations and changes of the high-temperature extremes. However, the observational records of the heatwave events are short, and cannot cover the long time period, especially before the industrial era. The tree-ring proxy has the potential to infer the variability of summer heatwaves over the past time, and has been successfully extend the variation of the frequency of the summer heatwaves over inner East Asia. Here we quantify the relationships between the variability of the radial growth of trees and the variabilities of the frequency, intensity and duration of the heatwaves over China on various timescales based on the publicly-available tree-ring width data in China, and the heatwave features calculated based on the ERA5 dataset. The results show overall increasing trends in the frequency, intensity and duration of the summer heatwaves over all the tree-ring sites in China for the past few decades. In contrast, the trees over the different sites exhibit distinct trends in their radical growth, with some showing increasing trends and some showing radical declines in the growths over the recent two decades. The statistical analysis shows that the correlation between the tree-ring growth and the heatwave frequency in summer changes from negative to positive with the increase in the elevations of the tree-ring sites. Despite an overall positive response, tree-ring growths show negative responses to the heatwave frequency in some high-elevation sites. Further results will show the spatial pattern and reasons of the response diversity of the tree-ring growth to the summer heatwaves in China over the past decades.

Meteorological data assimilation (DA) combined with the online coupled chemistry meteorology model (CCMM) can improve the air quality forecasts, as it helps reduce the uncertainties of meteorological initial conditions while considering the interaction between meteorology and air quality. In this study, the effects of meteorological DA on weather and particulate matter (PM) forecasts of the online CCMM are examined. Weather Research and Forecasting model coupled with Chemistry (WRF-Chem) and WRFDA three-dimensional variational (3DVAR) system are used. The nesting domains of the model represent East Asia (outer domain) and the Korean Peninsula (inner domain), respectively. The two high PM concentration cases were selected, and four experiments with different DA domains were conducted for each case. When the meteorological DA was performed in the outer domain, the forecast errors (root-mean-squared error (RMSE) and bias) of meteorological variables were the smallest, while the errors were the largest in the experiment without DA. In addition, when the meteorological observations were assimilated in the outer domain, the forecast errors of PM concentration were the smallest. The effect of the meteorological DA on the PM forecast error reduction lasted approximately 58-66 hours. The synoptic meteorological forecasts were improved by performing meteorological DA in the outer domain. The improved forecasts in the outer domain enhanced the meteorological initial and boundary conditions of the inner domain, which consequently led to the improvement of the meteorological and PM forecasts in the inner domain. Acknowledgements: This work was supported by a grant from a National Research Foundation of Korea (NRF) grant funded by the South Korean government (Ministry of Science and ICT) (Grant 2021R1A2C1012572) and Yonsei Signature Research Cluster Program of 2023 (2023-22-0009).

China has experienced increasingly serious ozone (O₃) pollution in recent years, posing a great threat to agricultural ecosystem and human health. Here, we analyzed ozone trends across China using multi-source observations combined with multi-model calculations. Ozone increases steadily in China between 2013−2022, with a fast increase rate of 4.4 μg m⁻³ yr⁻¹ in Phase I of Action Plan and a much smaller 0.6 μg m⁻³ yr⁻¹ in Phase II. Results highlight that the deteriorative O₃ pollution in Phase I and early Phase II is dominated by the nonlinear O₃-emission response. Persistent decline in O₃ precursors has shifted its chemical regime in urban areas and began to show a positive influence on ozone mitigation in recent years. Meteorological influence on O₃ variations is minor until 2019 (~10%), but it greatly accelerates or relieves the O₃ pollution after then, showing comparable contribution to emissions. Strong connections between surface O₃ levels and SSC synoptic weather patterns are identified, with the dry tropical (DT) weather pattern featured by dry and hot air masses as the main contributor to 35.3% of the high ozone occurrences and partly explaining the observed ozone trends in recent decade (e.g., +9.7% ozone increase per tripling of the DT frequency). Using the agricultural statistical data and empirical algorithms, it is estimated that the annual yields of winter wheat, single-cropping rice, double-cropping early rice and late rice are reduced by 9.0−17.0%, 6.0−8.2%, 4.8−9.4% and 3.9−7.0%, respectively, in China during 2014–2022 due to the ozone exposure. Epidemiological model predicts totally 0.8−3.0 thousand yr⁻¹ more deaths across China with altered anthropogenic emissions since clean air actions, and additional health burdens by −1.5~0.3 thousand yr⁻¹ from perturbated meteorology. This study calls for stringent emission control and climate adaptation strategies to attain the ozone pollution mitigation in China.

The rapid development of Chinese cities is accompanied by air pollution. Although the implementation of air pollution control strategies in recent years has alleviated PM_2.5 pollution, O₃ pollution and the synergistic pollution of PM_2.5 and O₃ have become more serious. To understand the underlying chemical interaction mechanisms between PM_2.5 and O₃, we applied the modified Weather Research and Forecasting model with Chemistry (WRF-Chem) to study the effects of aerosol-photolysis feedback and heterogeneous reactions on the two pollutants and revealed the contribution of different mechanisms in different seasons and regions in Yangtze River Delta (YRD) in eastern China. We found that, through the aerosol-photolysis feedback, PM_2.5 decreased the surface photolysis rates J_NO2 and J_O1D, resulting in a decrease in O₃ concentration in the VOC-sensitive area and a slight increase in the NO_x-sensitive area. The heterogeneous reactions reduced O₃ concentration in the YRD in spring, autumn and winter by consuming H_xO_y. While in summer, the heterogeneous absorption of NO_x decreased O₃ in the NO_x-sensitive areas and increased O₃ in the VOC-sensitive areas. Heterogeneous reactions also promoted the secondary formation of fine sulfate and nitrate aerosols, especially in winter. Through the combined effect of two chemical processes, PM_2.5 can lead to a decrease in O₃ concentration of -3.3 ppb (-7.6%), -2.2 ppb (-4.0%), -2.9 ppb (-6.3%), and -5.9 ppb (-18.7%), in spring, summer, autumn and winter in YRD. Therefore, if the PM_2.5 concentration decreases, the weakening effect of PM_2.5 on the ozone concentration will be reduced, resulting in the aggravation of ozone pollution. This study is important for understanding the synergistic pollution mechanism and provides a scientific basis for the coordinated control of urban air pollution.

Land-sea atmosphere interactions (LSAIs) are important processes affecting ozone (O₃) pollution in coastal areas. While the effects of small-scale LSAIs, such as sea-land breezes, have been widely studied, the impact of large-scale LSAIs on O₃ pollution is not yet fully understood. Here we investigated an O₃ episode to illuminate the role of large-scale LSAIs in O₃ pollution in the southeastern coastal, Bohai-Yellow Seas and adjacent regions through observations and model simulations. The results are presented below. 1) O₃ pollution in southeastern coastal cities and the northern Bohai Sea’s coastal regions was mainly caused by transport from over the eastern China seas (ECS), the Bohai-Yellow Seas, respectively. 2) Affected by the Mongolian High, the study regions initially experienced a typical unimodal O₃ diurnal variation. Precursors emitted from Beijing-Tianjin-Hebei, Northeast China (NEC) and Japan-Korea were transported to ECS, and precursors from Beijing-Tianjin-Hebei, Shandong, and NEC were transported to the Bohai-Yellow Seas. Photochemical reactions produced O₃ in marine air masses, causing higher O₃ levels over the sea than in coastal regions. As the Mongolian High shifted eastward and expanded, northeasterly winds transported O₃-rich marine air masses from over ECS to the southeastern coastal regions, and southerly winds transported them from over the Bohai-Yellow Seas to their coastal region, which prolonged pollution and weakened diurnal variations. 3) In the southeastern coastal, Japan-Korea contributed the most, with an average of about 5 ppb and a peak of up to 30 ppb. The contributions of Beijing-Tianjin-Hebei and NEC were comparable, with an average of about 2 ppb and hourly peaks of 19 and 10 ppb, respectively. In the northern Bohai Sea, Shandong’s emission contributed significantly in both phases (27.5% and 26.1%, respectively), and emissions from the Korean Peninsula and marine shipping had a notable impact on O₃ during the second phase (10.7% and 13.7%, respectively).

The urban landscape pattern, which can affect the air pollutants distribution in urban areas, was significantly changed due to the expansion of built-up land. Deeply understanding of the mechanisms by which urban landscape patterns influence the PM2.5 concentration distribution is fundamental for urban pollution control. However, the relationship between the built-up land landscape pattern and PM2.5 concentration distribution is still unclear at different grid scales in different seasons. In this study, four landscape metrics were calculated to quantify the distribution characteristics of built-up land in Nanjing, and the impact of built-up land landscape patterns on PM2.5 concentration distribution at different grid scales in different seasons was further analyzed. Results showed that the dominant landscape pattern metrics affecting PM2.5 concentration in different areas of Nanjing was closely related to the topography: in the main urban areas with flat terrain, the aggregation, shape, and proportion of built-up land significantly affected PM2.5 concentration, while the influence of the shape and proportion of built-up land was more noticeable in the hilly southwest suburbs. The effect of built-up land landscape pattern on PM2.5 concentration in winter was more significant than that in summer in general, and the relatively greater seasonal differences in this effect can be seen in the aggregation and proportion of built-up land. With the grid increasing, the correlation between the built-up land landscape pattern and the PM2.5 concentration distribution increased in the main urban areas but mostly decreased in the southwest suburbs, indicating that the PM2.5 concentration distribution in the main urban areas was mostly contributed by the spatial transmission of pollutants, while the PM2.5 concentration distribution in the southwest suburbs was mostly dominated by local emissions. We suggest that the planning of flexible urban strategies with respect of PM2.5 pollution control in different areas should be implemented considering the influence of topography.

Monitoring and predicting fine particulate matter (PM2.5) concentrations is crucial for public health, especially in urban areas where air pollution is a significant risk. Understanding factors influencing PM2.5 variabilities – such as synoptic-field changes, interannual and seasonal variability, and climatic indices – is vital for effective air quality management and public health protection in densely populated and industrially advanced regions like East Asia. The Arctic Oscillation (AO) significantly influences regional weather patterns, potentially affecting air quality in urban areas like Seoul, Korea. This study investigates the relationship between PM2.5 concentrations in Seoul and the AO using ECMWF Reanalysis v5 (ERA5) and the Weather Research and Forecasting (WRF) model. From 2015 to 2023, we analyzed daily average PM2.5 concentrations from 48 air quality stations in Seoul, focusing on November to April, which typically experience numerous high concentration events. A regression analysis was conducted to understand the impact of AO phases on Seoul's air quality. This analysis identified that PM2.5 concentrations correlate with various weather variables, showing different patterns during positive and negative AO phases. These correlations uncovered the influence of the AO on air quality. Moreover, accurate forecasting of high PM2.5 concentration events is vital for public health and environmental management. Thus, we also evaluated the WRF model's performance in forecasting weather variables during high pollution events to enhance predictive accuracy and preparedness. This research found the impact of the AO on Seoul's air quality, verifying the interaction between synoptic-scale weather variables and urban-scale air pollution. Consequently, forecasting local-scale PM2.5 concentrations in East Asia must consider dealing with climate indices, particularly the AO phase.

Particulate matter (PM_2.5) is generated not only by primary emission sources but also by secondary processes from precursors in the atmosphere. For this reason, the Ministry of Environment has recently implemented measures to reduce precursor emissions from power plants and on-road mobile sources nationwide. Despite the intensive efforts, the PM_2.5 concentration remains steady both in the east and west regions of Gangwon province. To establish effective PM_2.5 reduction measures in the Gangwon province, it is necessary to identify not only its emission sources in the regions but also the PM_2.5 contribution of precursors. In addition, the Yeongseo (west) and Yeongdong (east) regions have different emission source characteristics in Gangwon province. This study aims to evaluate the PM_2.5 contribution to air pollutants emitted from the Yeongseo and Yeongdong regions, separately in February 2023. We utilized the CMAQ-BFM(Brute Force Method) to calculate the PM_2.5 contribution of air pollutants. PM₁₀, PM_2.5, SO₂, NO_X, NH₃, and VOCs are chosen for emitted air pollutants in the contribution analysis. The simulation results of PM_2.5 concentration were verified using data from urban air quality monitoring stations in the study area. Results showed that emission contribution from the Yeongseo region was approximately 20%, while that from Yeongdong region was approximately 10%. We estimate the contribution of each air pollutant for each region individually. Acknowledgment" "This research was supported by Particulate Matter Management Specialized Graduate Program through the Korea Environmental Industry & Technology Institute(KEITI) funded by the Ministry of Environment(MOE)."

Fine particulate matter (PM2.5) episodes in Korea were primarily known to be originated from the inflow of pollutants from upwind countries and regions or occur due to atmospheric stagnation. To effectively manage such high PM2.5 concentrations, this study classified the occurrence types of high-concentration PM2.5 cases by utilizing an atmospheric recirculation factor (RF) and the concentrations of PM2.5 components (SO42-, NO3-, NH4+). Additionally, we quantitatively analyzed the dominance of inter-regional influence through time-lag correlation analysis. The target regions included Seoul, Ansan, Seosan, Daejeon, Gwangju, Ulsan, and Jeju, where the atmospheric environment research institutes operated by the national institute of environmental research are located. The study period is from January to December in 2021. Hourly concentration data of PM2.5 and its composition components, along with local meteorological forecasting model (LDAPS) data from the Korea Meteorological Administration, were analyzed. Four types of high-concentration cases were identified: inflow from abroad, stagnation after inflow, inflow from other regions, and atmospheric stagnation. As a result, the frequencies of inflow from abroad and atmospheric stagnation after inflow were high in the western coastal and metropolitan regions. Additionally, it was confirmed that sulfate and nitrate components are suitable indicators for foreign inflow and atmospheric stagnation, respectively. This work was supported by “the FRIEND (Fine Particle Research Initiative in East Asia Considering National Differences) Project through the National Research Foundation of Korea (NRF) funded by the Ministry of Science and ICT (2023M3G1A1090663).” and "This research was supported by Particulate Matter Management Specialized Graduate Program through the Korea Environmental Industry & Technology Institute(KEITI) funded by the Ministry of Environment(MOE) (C10168180002)."

As one of the precursors of severe storms, convective clouds are a special type of clouds that play a critical role in the global hydrological cycle and radiative energy budget at a local or global scale. Due to complex physical and dynamic processes and large spatiotemporal variability, convective clouds especially for the small spatial scales contribute to significant uncertainties in existing weather and climate models' simulations. Convective cloud detection and tracking is one of the crucial tasks to understand their development and evolution characteristics. Based on the Himawari-8 data with higher spatial and temporal resolutions, a convective cloud detection and tracking algorithm was constructed by combining machine learning, region-growing, area-overlapping, and Kalman filter algorithms. First, a complicated and strict spatiotemporal matching strategy between the Himawari-8 AHI data and CloudSat cloud profile radar was established, such as parallax correction model was constructed to minimize the impact of parallax. In addition, to expand the sample volume of convective clouds, region-growing algorithm is further used. Then, a convective cloud detection model suitable for both day and night time was constructed based on the XGBoost machine learning model. The convective cloud detection results are extensively evaluated by comparisons with cloud property products of JAXA AHI, traditional threshold algorithm, and CloudSat data. In addition, to achieve tracking of large-scale, small-scale, and fast-moving convective clouds, the automatic tracking of convective clouds algorithm is achieved by comprehensively using the area overlapping and Kalman filter algorithms. Validation results indicate that the algorithm developed can detect convective clouds of different scales with high accuracy, and the results show good continuity during day-to-night transitions.

Effective Decadal/near‐term climate prediction can provide key climate information one or several decades in advance for policymakers and stakeholders, it has recently developed a high profile in the scientific community and beyond. However, current decadal climate prediction skill is far from successful. Our studies provide a new decadal increment method, which can skillfully predict the decadal variabilities of east Asian summer monsoon，summer precipitation over North China, extreme precipitation over South China and Pacific Decadal Oscillation. There are three steps in the increment method. First, the decadal variability (DV) of predictand is obtained by applying a 5‐year running mean. Second, the 3‐year decadal increment (DI) of the DV (DV at the current year minus the DV at the 3 years before, DI_DV) is predicted by a statistical forecast model with the leading several years of predictors in DI form. Third, the predicted DI_DV is added to the observed DV at 3 years ago to obtain the final DV prediction. This decadal increment method provides a promising and valuable approach to decadal climate prediction.

Winter precipitation anomalies in South China (SC) frequently result in severe disasters. However, the evaluation of prediction performance and distinctions between positive precipitation anomaly events (PPA, wet condition) and negative precipitation anomaly events (NPA, dry condition) in current operational models remains incomplete. This study employed the Climate Forecast System version 2 (CFSv2) to assess winter precipitation prediction accuracy in SC from 1983 to 2021. Differences in predicting PPA and NPA events and the underlying physical mechanisms were explored. The results indicate that CFSv2 can effectively predict interannual variations in winter precipitation in SC, as there is a significant time correlation coefficient of 0.68 (0.62) between observations and predictions, with a lead time of 0 (3) months. The model revealed an intriguing asymmetry in prediction skills: PPA outperformed NPA in both deterministic and probabilistic prediction. The higher predictability of PPA, as indicated by the perfect model correlation and signal-to-noise ratio, contributed to its superior prediction performance when compared to NPA. Physically, tropical signals from the ENSO and extratropical signals from the Arctic Sea ice anomaly, were found to play pivotal roles in this asymmetric feature. ENSO significantly impacts PPA events, whereas NPA events are influenced by a complex interplay of factors involving ENSO and Arctic Sea ice, leading to low NPA predictability. The capability of the model to replicate Arctic Sea ice signals is limited, but it successfully predicts ENSO signals and reproduces their related circulation responses. This study highlights the asymmetrical features of precipitation prediction, aiding in prediction models improvement.

In order to improve the ability of the CMA global ensemble prediction system (CMA-GEPS) to forecast extreme weather, aiming at the difficulty of reasonably calculating model climate due to small samples of historical forecasts from CMA-GEPS together with the lack of the re-forecast dataset, a method to build the model climate required by extreme forecast index (EFI) was investigated via using the insufficient samples of deterministic forecasts through extending the forecast samples in both time and space. By employing the operational forecast data of CMA-GEPS and the ERA5 reanalysis dataset, the forecast ability of CMA-GEPS for extreme high temperature in four representative regions at home and abroad for the summer of 2022 was evaluated. Results from the relative operating characteristic curve showed that CMA-GEPS EFI had the ability to discriminate extreme high temperature within the short- and medium-range forecast lead times of 1-10 days. Taking the maximum TS score as the criterion, the critical thresholds of EFI for issuing warning signals of extreme high temperature was determined. The forecast ability of EFI was decreased with the increase of forecast lead times, and different performance was exhibited in different regions: the forecast ability of extreme high temperature in the middle and lower reaches of the Yangtze River in China was higher than that in North China for all lead times; the forecast ability of EFI in western Europe was better than that in central Europe for the 1-7-d lead times, meanwhile the EFI forecast ability in central Europe for the 8-10-d lead times was better. Evaluation results from the economic value model revealed that risk decisions based on the EFI forecast information owned certain economic value. A case study further demonstrated that the CMA-GEPS EFI could provide early warnings for extreme high temperature in the medium forecast range.

The Korea Institute for Atmospheric Prediction Systems (KIAPS) was established in 2011 with a mission to develop a global atmosphere-only numerical weather prediction system for operational use at the Korea Meteorological Administration (KMA). This system was completed on schedule, and made operational at KMA in April 2020, immediately giving a world-class performance. The system is based on a new atmospheric model called the "Korean Integrated Model" (KIM), which is based on a cubed-sphere grid and uses the spectral element method within its dynamical core. Deterministic data assimilation (DA) is based on a hybrid-4DEnVar algorithm, and forecast uncertainties are modelled by a 50-member Ensemble Prediction System (EPS). The EPS is based on a local ensemble transform Kalman filter (LETKF) DA algorithm, and further schemes are included to account for deficiencies and uncertainties in the DA process and the forecast model. In this presentation, we examine the sensitivity to stochastically perturbed physical tendencies (SPPT), which aims to account for uncertainties in the forecast model and test a three-scale pattern, which consists of a linear combination of three independent random patterns, each describing a different correlation scale. To reduce computational costs, the sensitivity tests were carried out within a low-resolution framework, and we analyzed the effect of the schemes on the ensemble mean error and spread in temperature, geopotential height, and specific humidity in extended medium-range forecasts. In regions where the control experiment, including only initial perturbations, did not adequately reflect the ensemble spread, it is anticipated that the SPPT approach will result in increased forecast skill.

The Afro-Asian monsoon is a major component of the global monsoon system, with a large-scale rainbelt covering Africa, South Asia, and East Asia. Previous studies have demonstrated that the Afro-Asian summer monsoon (AfroASM) rainfall exhibits prominent decadal variability. By investigating observations and ensemble simulations, this study reveals that the decadal variation of the rainfall in the Sahel and eastern China is mainly caused by the interdecadal Pacific variability (IPV) and the Atlantic multidecadal variability (AMV), with positive (negative) phase of the AMV contributing to intensified (weakened) AfroASM rainfall, and the IPV corresponds to opposite impacts. The AMV and IPV influence AfroASM rainfall by modifying the tropical easterly jet (TEJ). Positive AMV favors a stronger TEJ, but positive IPV leads to a weaker TEJ, vice versa for their negative phases. The AMV elicits more contribution in the exit region of the TEJ, while the IPV contributes more in its entrance region, which concurs that the rainfall in the Sahel receives a larger impact from the AMV, and the IPV influences eastern China rainfall more strongly.

Extensive observational and modeling studies demonstrate the influence of urbanization on the intensity of convective storms and extreme precipitation. Nevertheless, the impact of urbanization on the tornado triggering environment is relatively lacking at present. Here we examine the impact of anthropogenic aerosol emission and land surface changes on the triggering environment of an EF3 supercell tornado on 14th May, 2021 in Yangtze River Delta, a rapidly urbanized region in China. The event is simulated by Weather Research and Forecasting Model coupled with Chemistry (WRF-Chem) with an updated land use category. Significant Tornado Parameter (STP) is used to assess the tornado-triggering environment based on evaluation of convective available potential energy, wind shear, storm relative helicity, and lifting condensation level. Sensitivity tests show that anthropogenic aerosols due to urbanization enhances tornado-triggering environment, with the increase of storm relative helicity (SRH1) dominates the enhancement. Meanwhile, condensation of aerosol releases more latent heat, changes the airflow, and leads to the increase of SRH1. Moreover, change of land surface alters the sensible and latent heat fluxes, thus affecting the dynamical and thermodynamical conditions. Together, changes of aerosol and land surface due to urbanization influence the environment that are conducive to tornado formation.

Urban areas worldwide are increasingly threatened by climate-driven risks. In response to the need for urban climate adaptation, an explicit urban representation in global climate models contributes to realistically simulating urban heat and water fluxes across regions. This study attempts to enhance representation of urban surface heterogeneity within the Community Earth System Model (CESM), by developing a new scheme to represent urban areas based on local climate zones (LCZs). In CESM's land component, the Community Land Model (CLM), we incorporated the urban LCZ scheme as an option alongside its default scheme. The CESM-LCZ scheme increases the subgrid-level urban land use types to 10 LCZs, with each LCZ represented by a unique set of urban canopy parameters (UCPs). The modified model is verified by present-day simulations and applied for future urban climate projections. We envision that the global-scale simulations of urban climates will provide scientific support for broader communities, enabling them to conduct assessments of LCZ-related impacts and inform urban planning decisions. 

Hydrometeorological disasters caused by extreme precipitation pose significant hazards in the Philippines, underscoring the urgent need for a robust monitoring system to safeguard human lives and assets in the country. This study proposes a tool for identifying potential extreme rainfall events using average recurrence interval (ARI). To enhance the reliability of ARI statistics beyond historical records, we employed Regional Frequency Analysis (RFA) across the Philippines, utilizing TRMM 3B42 rainfall data. Following the methodology outlined by Hosking and Wallis (1997), rainfall data were first pooled into homogeneous regions before estimating precipitation quantiles for different ARIs. Our findings reveal that RFA produced a lower Root Mean Square Error (RMSE) compared to traditional frequency analysis, signifying the reliable implementation of resulting ARI statistics for monitoring applications. To illustrate, three past cases of extreme rainfall events that vary in intensity and impact were examined. The equivalent ARI maps of these events captured areas that experienced flooding and landslides associated with the extreme rainfall event. Expanding our analysis, we applied ARI thresholds for extreme rainfall detection to our experimental WRF forecasts, comparing the outcomes with results based on monthly climatological rainfall thresholds. This comparative evaluation demonstrated the added value of ARI thresholds in detecting rainfall amounts surpassing monthly climatological levels, providing crucial insights into extreme weather monitoring. Lastly, recognizing the importance of medium-term rainfall events lasting up to a week, initial assessments exploring the physical and statistical properties of medium-term events revealed a seasonal pattern in different regions, particularly demonstrating improved homogeneity in areas affected by tropical cyclone disturbances, while variability remains over Mindanao and the Eastern Philippines during the northwest monsoon season. Despite these regional nuances, there is potential in using ARI for effective extreme weather monitoring applications, emphasizing its relevance across diverse temporal scales and geographical contexts.

Climate change leads to more frequent and intense extreme temperature events, causing a significant number of excess deaths in cities in China. Using an epidemiological approach, we analyze all-cause deaths related to heatwaves and cold spells in 2,852 Chinese counties from 1960 to 2020. Economic losses associated with these events are determined through the value of statistical life. Findings reveal that cold-related cumulative excess deaths (1,133 thousand) are approximately 2.5 times higher than heat-related deaths, despite an increase in heat-related fatalities in recent decades. Monetized mortality due to heat-related events is estimated at 1,284 billion CNY, while cold-related economic loss is 1,510 billion CNY. Notably, cities located in colder regions experience more heat-related excess deaths, and vice versa. Economic development does not significantly reduce mortality risks to heatwaves across China. This study provides insights into the spatial-temporal heterogeneity of heatwaves and cold spells mortality, essential for policymakers ensuring long-term climate adaptation and sustainability.

Studying on the impacts of block-scale urban morphology on strong wind holds significant implications for predicting local severe winds, which have the potential to trigger safety and economic hazards. In this study, the strong wind induced by typhoon "Muifa" (2212) in Shanghai, China, was simulated using the Weather Research and Forecasting (WRF) model coupled with Parallelized Large-eddy simulation Model for Urban applications (PALM-4U). The findings were as follows: (1) Frontal area fraction (λf), building area fraction (λp), building surface to plan area ratio (λb) and height-to-width ratio (hw) showed a significant correlation with the strong wind indices (gust factor (GF) and relative wind speed (RS)). (2) The impacts of block-scale urban morphological indices on the average strong wind indices can be mainly categorized into two stages: acceleration and stabilization, and the transitional zone correspond to λf (0.07-0.09), λp (0.35-0.45), λb (1.3-1.6) and hw (2.2-2.4), respectively. (3) The accelerating effects of the urban morphological indices on the extreme features of the strong wind indices were manifested as the shifting and broadening of the probability density functions. In extreme cases, the average values would be significantly surpassed by the extreme values during the stabilization phase. The strong wind indices with 1% probability of occurrence can result in differences reaching λf (GF: 0.83, RS: 1.12), λp (GF: 1.03, RS: 1.34), λb (GF: 0.81, RS: 1.17) and hw (GF: 0.92, RS: 1.16), respectively.

In this study, four extreme precipitation indices with the highest impact on Ulaanbaatar's flood risk were chosen, and their changes over the last 60 years of climate data have been investigated. In addition, Future trends in maximum daily precipitation indices( RX1day), which are used as parameters for flood risk estimation, were estimated as well using regional ECAM5 and HadGEM2 model data. According to the findings, the number of consecutive rainy days increased from 1975 to 1994 and then decreased beginning in 1995. Throughout 1966 and 2022, observed daily maximum precipitation (RX1day) and extreme precipitation (R95p) indices rose by 2 mm and 7 mm, respectively. However, there was not a clear pattern in the number of days with precipitation greater than 20 mm (R20) across the years. However, there was no discernible trend in the number of days with precipitation of more than 20 mm (R20) over the years.

The interactions between urban heat islands (UHIs) and heat waves have been studied in many cities around the world because of their scientific interest as well as their influences on urban heat stress. However, few studies examined how these interactions vary with the characteristics of heat waves. Using observation data in Daegu, South Korea for 2001–2022, this study reveals that the interactions between UHIs and heat waves are contrasting under two distinct heat wave types (dry heat waves and moist heat waves). Dry (Moist) heat waves are defined as heat waves with daily mean relative humidity lower (higher) than its 10th (90th) percentile. Dry heat waves exhibit significantly higher temperature, lower cloud fraction, and lower soil moisture than those under moist heat waves. The mean UHI intensity, which is defined as the mean urban-rural difference in daily minimum 2-m temperature, under dry heat waves is significantly stronger than that under non-heat waves by 1.07 °C, but that under moist heat waves is weaker by 0.24 °C. This indicates that the interactions between UHIs and heat waves are primarily synergistic (negative) under dry (moist) heat waves. Interestingly, the temperature-humidity index, humidex index, and discomfort index in the urban area are not significantly different between dry and moist heat waves during the nighttime. This implies that heat waves that synergistically interact with UHIs do not necessarily lead to higher urban heat stress than those that negatively interact with UHIs. Therefore, monitoring changes in humidity under heat waves is indispensable for assessing extreme urban heat stress as well as interactions between UHIs and heat waves.

As one of the acyclic sesquiterpenes (SQTs), (E)-β-farnesene plays an important role in SOA formation through atmospheric oxidation to form highly oxidized multifunctional molecules (HOMs). In this study, we comprehensively explore the mechanism of oxidation of (E)-β-farnesene with OH radical by quantum chemical calculations. The OH addition reactions with lower free barriers are more dominant than H-atom abstraction reactions. The first-generation products involving acetone, (E)-4-methyl-8-methylenedeca-4,9-dienal, 4-methylenehex-5-enal, 6-methylhept-5-en-2-one, formaldehyde, (E)-7,11-dimethyldodeca-1,6,10-trien-3-one and (E)-6,10-dimethyl-2-methyleneundeca-5,9-dienal are produced from the subsequent reactions of the OH-(E)-β-farnesene adducts with NO and HO₂. The second-generation products, including 4-oxopentanal, 4-methylenehex-5-enal, (E)-4-methyl-8-oxodeca-4,9-dienal, formaldehyde and (E)-4-methyl-8-methylenenon-4-enedial. HOMs formation via the autoxidation mechanism is limited by the H-shift. Our proposed mechanism, involving multiple H abstractions by OH radical and subsequent reactions with O₂ and HO₂, is a more thermodynamically favorable pathway for HOMs formation. The complicated oxidation mechanisms and products of sesquiterpenes (including (E)-β-farnesene) make it difficult to evaluate their contributions to SOA formation in the atmospheric chemical models. We hope the mechanism of (E)-β-farnesene oxidation in the atmosphere from a theoretical perspective would be conducive to clarifying the atmospheric fate of (E)-β-farnesene and even sesquiterpenes.

In April and May 2022, Shanghai implemented city-wide static management measures to control the Omicron variant, resembling a large-scale experiment. Despite a 30% reduction in VOCs and 50% in NO₂, the 23% increase in average ozone levels surprised researchers. Diurnal patterns showed four distinct ozone variations, with Clusters 3 and 4 experiencing significant increases during static management. Observation-Based Model analysis indicated over 30% increases in OH, HO₂, and RO₂ concentrations in 2022 compared to 2020 and 2021. Despite HONO photolysis being the main ROx radical primary source, radical cycling dominated overall production. With a substantial NO2 decrease relative to VOCs, the VOCs/NO₂ ratio rose from 1.6 in 2020 to 3.0 in 2022, affecting radical cycling. The OH radical propagation/termination ratio increased to 2.10, indicating the different precursor reduction proportions strengthened radical cycling. The unequal reduction in VOCs and NO₂ during static management caused the rise in ozone concentration in Shanghai. This work was supported by National Natural Science Foundation of China (22176037, 42075097, 22376030, 42375089, 21976031) and National Key Research and Development Program of China (2022YFC3700101).

Tangshan City, a major industrial city in the Beijing-Tianjin-Hebei region in the North China Plain, faces severe air pollution. Air pollutants and CO₂ emissions share highly common homology, process characteristics and spatio-temporal consistency of emissions. In alignment with commitment to achieving carbon peaking and carbon neutrality, China has proposed plans to promote efficient coordination in pollutants and carbon reduction. However, there is a lack of research on understanding the benefits of co-emission reduction of CO₂ and air pollutants. In this study, we use the Community Multiscale Air Quality (CMAQ) model with Weather Research and Forecasting Model (WRF) to simulate the synergistic scenarios of carbon and pollution reduction in 2030 and 2060, respectively. The scenarios include the baseline scenario, reflecting carbon emission reduction consistent with China's 2020 Nationally Determined Contribution (NDC) Target, the reinforcing scenario, outlining a mitigation strategy in line with the global 1.5-degree climate target and China's enhanced NDC commitments, and the moderate scenario, representing an intensity level between the baseline and reinforcing ones. Biogenic emissions are estimated using the Model of Emissions of Gases and Aerosols from Nature (MEGAN), while anthropogenic emission scenario are based on the Multi-resolution Emission Inventory for China (MEIC). This study would help elaborate on the synergies between environmental governance and carbon mitigation and provide a basis for enforcing integrated coordinated environmental pollution and CO₂ emission control measures in Tangshan and across China.

Surpassing other regions in China, the Fenwei Plain (FWP) became the most polluted areas area in China in recent years. Despite the pronounced severity of air pollution in this region, there is a notable scarcity of research on the subject, resulting in an unclear understanding of the factors contributing to changes of air pollution during clean air actions. This study seeks to elucidate the variations in air pollution within the FWP from 2013 to 2020, employing simulation results from the Community Multiscale Air Quality (CMAQ) v5.0.2 model. Meteorological fields are generated using the Weather Research and Forecasting (WRF) model, anthropogenic emissions are sourced from the Multi-resolution Emission Inventory for China (MEIC), and biogenic emissions are modeled using the Model of Emissions of Gases and Aerosols from Nature (MEGAN). Initially, the model's performance will be validated against observations, followed by an investigation into the changes in PM2.5 and O3 from 2013 to 2020. Ultimately, the study aims to assess the contributions of variations in meteorological conditions and emissions to the changes in PM2.5 and O3. This research provides a crucial reference for the prevention and control of air pollution in the FWP.

Rapid urbanization and industrialization have resulted in diverse anthropogenic activities and emissions between urban, peri-urban and rural regions, leading to varying levels of exposure to air pollutants and associated health risks. However, endeavors to mitigate air pollution and health benefits displayed considerable heterogeneity across different regions. Therefore, comprehending the changes in air pollutant concentrations and health impacts within the urban expansion context is imperative for promoting environmental equity. This paper uses GDP- and population-weighted methods to distinguish anthropogenic emissions from urban, peri-urban and rural regions in the world, and quantified their contributions to air pollutants and fast atmospheric response using the Community Earth System Model 2 (CESM2) from 1975 to 2015. The study reveals the impact of urbanization on environmental justice between countries at different levels of development in the world.

Lightning-induced nitrogen oxides (LNO_x) can affect ozone (O₃) through photochemical processes. However, the impact of LNO_x on the Tibetan Plateau (TP) O₃ valley, an area with a lower O₃ column compared to surrounding areas, remains controversial. Prior to this study, the impact of monsoonal transport on LNO_x and related effects on the TP O₃ valley has been overlooked. The objectives of this study are to differentiate between locally generated LNO_x and monsoon-transported LNO_x over the TP and to evaluate their impacts on tropospheric NO_x vertical column density, tropospheric column O₃, and O₃ at different altitudes. The Weather Research and Forecasting model coupled with Chemistry model was used to replicate LNO_x, NO_x, and O₃ over the TP during July 2010, for which nesting techniques and sensitivity experiments were implemented. The findings emphasize the importance of LNO_x as a source of NO_x in this region, with nonlocal lightning activity being the primary source. Conversely, the influence of locally generated LNO_x on NO_x levels over the TP is comparatively minor. Overall, LNO_x contributes to an increase in O₃ over the TP, but one order lower than the O₃ depletion. The impact of LNO_x on O₃ varies across different altitude ranges; it is broadly “C” shaped, reflecting increase in O₃ at near-surface and in upper troposphere–lower stratosphere (UTLS) regions and decrease in the middle troposphere. The increase in surface-level O₃ over the TP is mainly caused by monsoon-transported LNO_x, while locally generated LNOx also plays a crucial role in O₃ formation in the UTLS region.

China is confronting severe ozone (O₃) pollution although particulate matter reduced significantly, causing damage to public health and ecological systems. O₃ was influenced significantly by the atmospheric oxidation capacity (AOC), which was evaluated by the sum of oxidation rates of VOCs via reactions with HO_x (OH and HO₂) and NO₃ radicals. Here, we updated the source-oriented CMAQ model to track the sources of HO_x and NO₃ radicals in the North China Plain (NCP) and Yangtze River Delta (YRD) in China. Specifically, we evaluated the contributions of different anthropogenic and natural sources to AOC by tracking the chemical reactions of HO_x and NO₃ radicals. Consequently, it’s significant to find that anthropogenic emissions made an important contribution to AOC in China. In the future, AOC contributed by different sources should be considered when designing O₃ control policies.

Airborne PM_2.5 pollution is a leading risk factor for human health, which has been associated with multiple cardiovascular and respiratory diseases. Nitrate is one of the important toxic component and its fraction in PM_2.5 is growing in China. The Community Multiscale Air Quality (CMAQ) model is widely used to estimate PM_2.5 and its components. The substantial computational cost of CMAQ, including solving a complex system of partial differential equations (PDEs) for transport and aerosol dynamics, and ordinary differential equations (ODEs) for chemistry, poses a bottleneck to its widespread application. Machine learning-based methods have the potential to capture complex physicochemical processes in the atmosphere, and hence replace CMAQ to accelerate pollutant concentration estimation. This study aims to use a neural network emulator based on graph neural networks (GNNs) in an “encoder-processor-decoder” configuration to simulate the surface nitrate concentration of CMAQ. The model will be applied in winter and summer months of 2019 and evaluated against CMAQ results. Computational efficiency will be compared and possible uncertainties will be identified.

This study aims to demonstrate the benefits of using novel high spatiotemporal retrieval products from newer satellites for top-down emission estimates of nitrogen oxides (NO_x) and non-methane volatile organic compounds (NMVOCs) for the summer of 2019 over the contiguous United States. Recent satellite retrievals have not only advanced spatiotemporal resolution but also greatly reduced error and uncertainty due to reduced noise in the retrievals compared to spaceborne sensors launched in the past. We applied inverse modeling techniques using tropospheric nitrogen dioxide (NO₂) and formaldehyde (HCHO) column retrieval products from the Ozone Monitoring Instrument (OMI) and TROPOspheric Monitoring Instrument (TROPOMI) in conjunction with the Weather Research Forecast and Community Multiscale Air Quality Modeling system (WRF-CMAQ). In order to provide a better representation of background chemical composition and avoid misalignment of emission adjustment, we applied monthly scaling factors for ozone (O₃) and CO boundary concentrations in addition to the inclusion of lightning and aviation emissions. Satellite-constrained NO_x and NMVOCs posterior emissions showed a mitigated discrepancy between observed and modeled columns. The improvement in the model performance was greater when using TROPOMI, primarily benefiting from reduced errors/biases of the satellite retrievals that enabled us to explore corresponding changes in O₃ concentrations and production sensitivity regimes using the ratio of HCHO and NO₂.

Coastal cities not only in China suffer from multiple natural hazards, such as strong winds, storm surges, extreme precipitation and flooding, from tropical cyclones (TCs). Due to a limited reliable observation period e.g., only ca. 50 typhoon seasons, robust hazard assessments of TCs are often challenging. Here we investigate the time development of the TC frequency-intensity distribution by use of a centennial multi-member seasonal hindcasts ensemble, the ECMWF CSF-20C data set. Based on this we build time-varying event-sets of physical consistent, but non-realised TCs with damage potential for coastal megacities following the approach from Osinski et al. (2016) and Ng and Leckebusch (2021). In this presentation, we show preliminary results for the long-term TC hazard variability resulting from those physically consistent event sets with special attention to the impact-relevant footprints of pure ensemble events. Additionally, the study will focus on several coastal megacities in China in different decades and thus contributing to the general understanding of the variability of the hazards and their time-varying uncertainty. Further implications and applications, e.g. for storm surge assessments are also discussed.

The relationship between the spring North Atlantic Oscillation (NAO) and the tropical cyclone frequency over the western North Pacific (WNPTCF) in summer and autumn is investigated by use of observation data. It is found that their linkage appears to have an interdecadal change from weak connection to strong connection and then weak connection. During the period of 1950–1971 and 2011-2021, the NAO was insignificantly correlated to the WNPTCF. However, during the period of 1972-2010, they were significantly correlated with stronger (weaker) NAO corresponding to more (fewer) tropical cyclones in the western North Pacific. The relationship between the NAO and the WNPTCF is modulated by the Pacific Decadal Oscillation (PDO). When PDO is in its positive phase, anomalous upper-level westerlies occur over the tropical North Atlantic, allowing the energy associated with the positive phase of NAO propagates towards lower latitudes, leading to cold sea surface temperature anomalies in the tropical North Atlantic during spring. Subsequently, these cold sea surface temperature anomalies adjust the circulation patterns in the tropical North Atlantic, causing positive sea surface temperature anomalies in the Central Pacific, triggering anomalous low-level cyclonic patterns in the Northwest Pacific region, along with anomalous upper-level anticyclonic patterns, ultimately resulting in an increase in the frequency of typhoons. However, when the PDO is in its negative phase, the connection between the NAO and cold sea surface temperature anomalies weakens, breaking the relationship between the NAO and WNPTCF.

In this research, we examine the analysis of a decade's worth of lightning and rainfall data in Mindanao, Philippines, employing a geospatial aggregation methodology. Our approach hinged on the processing of extensive datasets from the Global Precipitation Measurement (GPM) and the World Wide Lightning Location Network (WWLLN). The core of this method involved segmenting Mindanao into a detailed grid system, where each pixel represented a specific geographic area. For each pixel, we calculated a statistical measure that captures the typical intensity of rainfall and frequency of lightning, meticulously avoiding the influence of extreme weather anomalies. This allowed us to construct a highly accurate geospatial climate profile across different regions of Mindanao. The results shows that for annual, seasonal and diurnal climate maps, there has been substantial temporal variations in weather patterns. This illustrates that while certain months consistently show heightened climatic activity, the same periods in subsequent years may not exhibit the same intensity or distribution of rainfall and lightning. The analysis revealed four distinct climatic zones, each characterized by unique patterns of rainfall and lightning. The first zone was identified by its simultaneous high intensity of rainfall and frequent lightning occurrences, suggesting a climate prone to regular and intense thunderstorms. The second zone, marked by high rainfall but less frequent lightning, indicated a wet climate with fewer thunderstorms. In contrast, the third zone showed an inverse pattern with less frequent rainfall but higher occurrences of lightning, pointing to a drier climate where lightning storms are more common. Lastly, the fourth zone, characterized by both infrequent rainfall and lightning, represented the most stable and mild weather conditions. Geospatial mapping has yielded detailed insights into Mindanao's varied climate, vital for improving weather forecasts and developing disaster strategies for each zone's specific climatic conditions.

Tropical cyclones (TCs) cause severe economic losses and casualties, which has led to extensive studies on the response of TCs to large-scale environmental changes under global warming and atmospheric CO₂ concentration increase. However, changes in TCs due to CO₂ concentration decrease, which is necessary for climate mitigation, have not been established. As such, the present study revisits the TC response to CO₂ increase and then investigates the TC response to CO₂ decrease using the Community Earth System Model 1.2 (CESM1.2). When the atmospheric CO₂ concentration is increased until quadrupled (ramp-up, RU), TC genesis increases over the eastern North Pacific and South Pacific, while it decreases over the southern Indian Ocean. Such changes are not completely reversed by symmetrically reducing the atmospheric CO₂ concentration (ramp-down, RD). In the RD, TC genesis decreases over the eastern North Pacific, but increases over the Southern Hemisphere basins. Thus, TC genesis over the South Pacific becomes higher in the late RD than in the early RU, which indicates the asymmetric TC response to symmetric atmospheric CO₂ change. To gain insight into this asymmetry in TC genesis response, the large-scale environmental changes associated with TC genesis are further analyzed. It is found that the South Pacific experiences an increase in relative humidity and vertical updraft in both RU and RD, thus becoming more favorable to TC genesis. These atmospheric changes appear to be linked with the intensification and southward shift of the Intertropical Convergence Zone during RU and RD, respectively. Nearly reversible changes in these atmospheric variables, and hence the TC genesis, are observed in the southern Indian Ocean. Meanwhile, low-level vorticity change, due to sea surface temperature change centered in the eastern Pacific, is likely responsible for TC genesis change in the eastern North Pacific.

Carbohydrates are the main source of energy for living organisms. Carbohydrates are released into the atmosphere through natural and man-made emissions and are regarded as organic molecular tracers. Carbohydrates on atmospheric particles may come from biomass burning, dust, fungal spores and pollen, etc. In this study, we collected atmospheric particulate samples from April 2019 to September 2020 on the Matsu Islands, which are located at estuary of Minjiang River. During the sampling period, the concentrations of total mass and the water-soluble organic carbon (WS-OC) were 40.03 ± 7.00 μg m^-3 and 69.21 ± 72.44 nmol m^-3. respectively. For carbohydrate species, the concentrations of total water-soluble carbohydrate and monosaccharide were 0.15 ± 0.11 nmol m^-3 and 0.08 ± 0.10 nmol m^-3., respectively. Monosaccharides account for nearly 50% of total sugars, which means that monosaccharides in the atmosphere are more available to organisms. In terms of particle size composition, monosaccharide was mainly distributed in fine particles (particle size < 095 μm), while polysaccharide was mainly distributed in particle size between 3 μm ~7.2 μm and < 0.49 μm. In order to further explore the sources of carbohydrates, Principal Component Analysis (PCA) and Positive Matrix Factorization (PMF) were used in this study. The results show four main aerosol sources in the region: biomass burning, anthropogenic sources, marine sources, and mixed sources, indicating that PCA and PMF provide precise insights into aerosols sources. The main source of carbohydrates is biomass burning, exhibiting a seasonal trend. During the biomass burning frequency in spring, it provides a carbon source for microorganisms, promoting nutrient cycling and plant growth, thereby exerting various effects on biogeochemistry.

The ambiguity surrounding the correlation between sea spray aerosol (SSA) formation and sea surface temperature (SST) hinders the accurate estimation of the impact of SSA on global climate. Here, we developed a temperature-controlled plunging SSA simulation tank to comprehensively investigate the impact of SST on SSA formation from two perspectives of SSA particle size distribution and organic matter enrichment. Our findings showed that the production of accumulation modal and coarse modal particles decreases as SST decreases from 30 ℃ to 0 ℃, while the production of Aitken modal particles decreases from 30 ℃ to 12 ℃, followed by a significant increase from 12 ℃ to 0 ℃. Enrichment factor (EF) results revealed a tenfold in the surfactant’s EF and a fourfold increase in the algae-derived dissolved organic carbon’s EF at 30 ℃ compared to 0 ℃, indicating that the temperature-dependent nature of organic matter in SSA particles significantly enhances their enrichment ability. Using the experimental results, we predict that SST variations may lead to a disparity of at least 191 ± 46 % regarding the contribution of Aitken-mode submicron sea salt particles to cloud condensation nuclei in polar and equatorial waters. Finally, considering the importance of SST for organic enrichment, our estimation indicates the global flux of DOC emitted via SSA is 25.2 Tg C yr^-1.

Local Convection in Dabie Mountains (LCDM) occurs more frequently over the Dabie Mountains and brings severe weather to adjacent areas. In order to understand the characteristics of LCDM, their spatial distribution, the monthly and diurnal variations, and possible mechanisms are investigated. Based on radar composite reflectivity data over the 5-y period of 2014–2018 during warm seasons (April–September), a total of 195 cases of LCDM are identified. The LCDM exhibits maximum frequency on the windward slopes of the Dabie Mountains with a secondary maximum on lee slopes. It is demonstrated that LCDM peaks in July and August, while their diurnal variation exhibits a major peak in the afternoon during 12:00–16:00 local solar time (LST). Most LCDM does not leave the Dabie Mountains (NoOut-Type), accounting for 89.7% overall, and has an average 3.5 h lifespan. In contrast, the lifespans of Out-Types (i.e., LCDMs that move away from the Dabie Mountains) are longer (5.8 h on average), while most Out-Type LCDMs develop on southern slopes (‘South-Type’) and a few are also reinforced on northern slopes (‘North-Type’). The South-Type mainly produces short-duration heavy precipitation, while the ‘North-Type’ predominately generates thunderstorms high winds. It is suggested that LCDM is thermally induced, and that both the ‘South-Type’ and ‘North-Type’ are controlled by southerly wind perturbation. Lifting by upslope wind and heat sources over windward slopes has led to ‘South-Type’ development, while ascent induced by wave-like perturbations on lee slopes has led to ‘North-Type’. These mechanisms should be further investigated in future work by using field experiments and numerical simulations.

The Uncertainty in the initial conditions of the numerical model affected the errors in the numerical weather prediction system. Aircraft are considered one of the best platforms to obtain atmospheric spatial information in observational gaps, especially over the sea. National Institute of Meteorological Sciences (NIMS) has operated an atmospheric research aircraft to fill observational gaps. These Aircraft observational data provide a continuous distribution of meteorological variables and contribute significantly to improving the performance of the numerical predictions. In this study, the effect of data assimilation (DA) on the prediction of typhoons affecting the Korean Peninsula was evaluated using high-resolution numerical modeling with atmospheric research aircraft observational data. This analysis was performed using two sets of simulation experiments: (1) WRF (V.4.1.2) run with DA (i.e., WRF-3DVAR), including atmospheric research aircraft observational data (Dropsonde, AIMMS-20) to improve the prediction of meteorological variables (e.g., temperature, wind components, and relative humidity), and (2) WRF run without DA (i.e., CTL). To reflect the data assimilation technique that improved the initial field more effectively, we used the GDAPS, ERA5, and KIM reanalysis fields provided by UM, ECMWF, and KMA as the initial/boundary conditions of the model. Overall, the results simulated by WRF-3DVAR for typhoon events in the study area are in good agreement with those observed in the CTL run. In particular, the air temperature showed improved results in all experiments with data assimilation (Dropsonde, AIMMS-20, and Dropsonde+AIMMS-20), and the wind speed showed improved simulation results in Dropsonde and Dropsonde+AIMMS-20, with the exception of AIMMS-20.

The National Institute of Atmospheric Sciences (NIMS) employs its KingAir 350H research aircraft for various missions, including the "Severe Weather" (SW) mission. This mission aims to enhance understanding of severe weather events like heavy rain, snowfall, and typhoons through advanced observations, ultimately improving numerical model accuracy. A key instrument aboard the aircraft is the Stepped Frequency Microwave Radiometer (SFMR). Mounted underneath, this C-band radiometer measures ocean brightness temperature to calculate sea surface wind speed and rain rate. Our study compared wind speeds derived from the SFMR with those obtained from dropsondes deployed along the aircraft's flight path. We also investigated the sensitivity of SFMR wind speed estimates to sea surface temperature and salinity, both serving as initial input data. Between June 2022 and October 2023, 21 SW-mission flights covered the Yellow Sea. SFMR data revealed an average wind speed of 8.7 m/s and a maximum of 22.1 m/s. Sea surface temperature and salinity data from buoys and research stations were interpolated to 0.1˚ intervals for sensitivity analysis. Calibration coefficients derived from a dedicated calibration flight on October 26, 2022, were applied to re-calculate wind speeds, which were then compared with pre-calibration estimates and wind speeds obtained from dropsondes at 30m, 150m, and 500m above the sea surface. Notably, the strongest correlation was observed between SFMR-derived wind speeds and those at 500m. This study demonstrates the potential of NIMS aircraft SFMR data for accurate sea surface wind speed measurements, particularly when employing calibration and considering the influence of lower-level wind patterns. Further research can refine these techniques and enhance our understanding of air-sea interaction.

South Korea experienced an unprecedented heatwave in October 2021. Such autumn heatwaves have adverse effects on crop production and lead to increased energy consumption. To prepare for heatwaves, analyzing the impact of meteorological conditions is crucial. Therefore, we conducted numerical experiments to analyze the impact of Typhoon MINDULLE (2116) on the heatwave that occurred in October 2021. Using the Weather Research and Forecasting (WRF) model, we performed the experiment replicating reality (TC) and under similar conditions but eliminating typhoon winds (TC-removed). Each experiment consisted of four ensemble members and ran for 41 days, beginning on September 23. During the heatwave, TC simulated 2-m temperatures up to 1.35°C higher than those of TC-removed. This temperature difference was linked to the typhoon's influence, which enhanced warm air advection from the northwest Pacific. These findings indicate that typhoon variability should be considered when forecasting extreme temperatures on the Korean Peninsula.

※ This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT) (No.2022R1A2C1008858), and by the Korea Environment Industry & Technology Institute (KEITI) through the 'Climate Change R&D Project for New Climate Regime,' funded by the Korea Ministry of Environment (MOE) (2022003560001).

The ocean heat fluxes over the sea are critical for simulating typhoon track and intensity because ocean heat fluxes are the energy source for typhoon maintenance and development. Over sea, where there is less observational data than on land, airborne observation can provide high-quality initial conditions for numerical models and help predict typhoon intensities and tracks. The National Institute of Meteorological Sciences (NIMS) initiated the annual operation of NIMS Atmospheric Research Aircraft (NARA) to investigate the atmospheric horizontal and vertical structure preceding severe weather phenomena such as typhoon, and heavy rainfall with the aim of reducing the uncertainty of atmospheric observations as of January 2018.This study investigates the impact of Dropsonde data assimilation for prediction of the typhoon OMAIS over the sea around Korean peninsula using the Weather Research and Forecasting (WRF) Advanced Research WRF model and its 3-dimensional data assimilation (3DVar) technique. Typhoon OMAIS is the first typhoon to land on the Korean Peninsula in 2021. Six experiments were conducted, three with Dropsonde data assimilation at three different initial conditions (UM GDPS, ECMWF ERA5, KIM) and three without Dropsonde data assimilation at three different initial conditions.

Extreme and persistent marine heatwaves (MHWs) occur frequently in the Northeast Pacific, with huge impacts on climate, ecosystem and socio-economic. This study investigates the atmospheric circulations associated with the 33 MHWs since 1951 in observations. The composite results reveal that the MHWs in the Northeast Pacific can be triggered by a couple of anticyclonic and cyclonic anomalies, i.e., the anticyclonic anomaly to the northeast of the MHW region and cyclonic anomaly to the southwest. This atmospheric circulation pattern can be detected as the dominant mode through EOF analysis on 500-hPa geopotential height anomalies over the Northeast Pacific-North America region, following the Pacific–North American teleconnection. These observational results are verified by using the outputs of 34 models in the historical simulation from phase 6 of the Coupled Model Intercomparison Project (CMIP6). Further diagnosis of the heat budget is performed, in attempt to illustrate the processes of MHW formation and maintenance.

Central India is affected by Indian Summer Monsoon (ISM) droughts. On a smaller scale, the Bundelkhand sub-region has experienced a series of droughts that have not coincided with ISM droughts. Analysis of daily rainfall data from the past century reveals two distinct drought types in Bundelkhand: Type-1 and Type-2. Type-1 droughts coincide with large-scale ISM droughts. Their evolution and drivers are well studied. In contrast, Type-2 droughts exhibit a different sub-seasonal evolution compared to Type-1. Their onset is in July, and the seasonal rainfall deficit is confined to Bundelkhand. Interestingly, parts of western India receive excess rainfall during Type-2 droughts. This study focuses on the spatiotemporal evolution and the potential drivers of Type-2 droughts. We observe that changes in large-scale circulation moisture convergence are consistent with the rainfall deficits in Bundelkhand. We also analyse the impacts of Type-2 droughts on the regional soil moisture (SM) and evapotranspiration (ET) patterns. SM deficit in Type-2 droughts prevails throughout the cropping season (June-October), leading to lesser vegetation cover and ET. We discuss the implications of our results in the context of a potential "negative feedback loop": decreasing ET leads to decreased atmospheric moisture, further reducing rainfall.

Heat waves are extreme weather events that cause socioeconomic damage over extensive areas. Heat waves are consistently being observed in many regions of the world. However, in terms of short-range predictions, researches to reduce forecast errors for heat waves are insufficient. In this study, the effects of observations used for data assimilation (DA) on forecast errors for variables associated with heat wave were evaluated using observing system experiments with the Weather Research and Forecasting model and a three-dimensional variational DA. All observations used in DA contributed to reducing forecast errors for the meteorological variables (e.g., geopotential height, temperature, and wind) associated with heat wave. As the forecast time increased, upper atmospheric geopotential heights in East Asia and 2-m temperatures around Korea, Japan, and eastern China were underestimated. Upper atmospheric observations reduced forecast errors more than near-surface observations. Advanced Microwave Sounding Unit-A (AMSU-A) had the greatest impact in reducing underestimation of the 200 and 500 hPa geopotential heights and the second greatest impact in reducing underestimation of the 2-m temperature. Radiosonde observations had the greatest impact in reducing underestimation of the 2-m temperature. When specific observations were not used in DA, the Tibetan high and Northern Pacific high were contracted compared to the analysis from the experiment using all observations for DA, causing the clockwise upper atmospheric wind bias through the geostrophic relationship. Therefore, upper atmospheric observations are important for reducing heat wave simulation errors in East Asia. The results of this study could contribute to design an optimal observing system for heat wave forecasts in East Asia. Acknowledgments: This work was supported by a National Research Foundation of Korea (NRF) grant funded by the South Korean government (Ministry of Science and ICT) (Grant 2021R1A2C1012572) and the Yonsei Signature Research Cluster Program of 2023 (2023-22-0009).

In the context of global warming, the phenomenon of warming and wetting occurred in northwest China. By comparing the three indexes of total precipitation, precipitation days and precipitation intensity between CN05.1 data, meteorology station from Meteorology Observation Network (CMA) and CMIP6 data, the results show that the precipitation trend in northwest China is significant, and both GHG forcing and AA forcing have some contribution. The results of water vapor decomposition of different external forcing show that the precipitation increasing in northwest China is mainly caused by the dynamic effects of greenhouse gases and anthropogenic aerosols. Further analysis shows that greenhouse gas emissions lead to the northward shift and westward extension of the Western Pacific subtropical high and the Mongolian high anticyclone anomaly, resulting in the easterlies on the north side of the Western Pacific subtropical high and the easterlies on the south side of the Mongolian High anticyclone, which together enter the northwest of China, thus weakening the westerlies and bringing more water vapor to the northwest of China. The anthropogenic aerosol forcing led to the weakening of the subtropical jet axis, resulting in the formation of anomalous anticyclone and cyclone in Central Asia, which affected the precipitation in northwest China.

The Seoul metropolitan area in South Korea experienced a record-breaking rainfall on August 8^th, 2022. The rainfall intensity reached up to 141.5 mm hr^-1, causing significant property damages and life losses. The observations show that most clouds that entered the rainfall region passed over the Yellow Sea where sea surface temperature (SST) was over 1.5 ℃ warmer than climatology. In particular, on the night of August 8^th when the heaviest rainfall occurred, small clouds formed over the Gyeonggi Bay and were merged with pre-existing rain clouds over the rainfall region. The present study investigates the possible effect of warm SST in the Yellow Sea and Gyeonggi Bay on the heavy rainfall event by performing a series of the Weather Research and Forecasting (WRF) model experiments. Previous studies have suggested that warm SST around the Korean Peninsula cloud lead to heavy rainfall event in land. However, the model experiments do not support them. When SST is replaced with climatology, colder than the observation, rainfall intensity becomes even stronger inland of the Korean Peninsula. When SST is increased by 1 ℃ and 2 ℃, the similar results are obtained. This results indicate a complicated pathway of SST impact on inland precipitation.

In South Korea, about 90% of the heavy rainfall events (HREs) are concentrated between June and September. Given the significant socio-economic impacts, understanding their climatological and dynamic/thermodynamic characteristics is of great importance. Although the developing mechanisms of HREs in South Korea are reasonably well known, the factors that determine the intensity of HREs need further research. The present study separates HREs into two groups: i.e., advisory HREs (aHREs), with 12-h accumulated rainfall between 110 mm and 180 mm and warning HREs (wHREs), with 12-h accumulated rainfall over 180 mm. These two groups are then compared from synoptic and thermodynamic perspectives. The synoptic features show no distinct difference between the two group, such as upper-level jet and baroclinic structure. The vertical motion diagnosed by solving the quasi-geostrophic omega equation is also similar. While dynamic omega is similar or even smaller in wHREs, diabatic omega has a slightly larger fraction. The two HREs differ in moisture transport. The atmospheric river is significantly stronger and better organized in wHREs, indicating that the thermodynamic conditions play a more crucial role in determining the intensity of HREs in South Korea.

With the growing impact of global warming, changes in the frequency and intensity of concentrated heavy rainfall events have been observed. To gain a detailed understanding of the heavy rainfall events occurring on the Korean Peninsula, cluster analysis was conducted. This study utilizes data collected over the past decade (2013-2022) from approximately 700 Automated Weather Stations(AWS) and Automated Surface Observation System(ASOS) stations to analyze precipitation patterns on the Korean Peninsula. Approximately 400 stations were chosen through a simple QC process, considering the missing rate. A comprehensive dataset was compiled, including variables such as rainfall duration (1/3/12 hours), frequency and intensity, and monsoon months (May-September). These variables were normalized to a median of 0 using the Robust transformation based on their deviation from the median. Three common clustering methods K-Means, Self Organizing Map, and Hierarchical Clustering were utilized, and the number of clusters was determined as six through Explained Cluster Variance (ECV) analysis, with K-Means showing the highest ECV. Cluster-1 displayed lower overall rainfall frequency, peaking in August, and was primarily located inland, excluding Gyeonggi-do areas. Cluster-2 exhibited higher rainfall frequencies in July and August, particularly in the western part of the inland region. Cluster-3 covered the eastern and southern coastal areas, including parts of Jeju, experiencing increasing rainfall frequencies from May/June to September. Cluster-4, located inland, demonstrated concentrated heavy rainfall, especially in August. Cluster-5, covering South and East Jeju along with parts of the southern coast, showed moderate and evenly distributed rainfall frequencies, with a peak in July. Finally, Cluster-6, encompassing Jeju and Geoje, consistently displayed high rainfall frequencies, especially in August and September, recording the highest number of heavy rainfall warnings. In this presentation, we will discuss more detailed characteristics of concentrated heavy rainfall occurrences for each cluster, including daily variations.

Drought can develop rapidly over a short period, the so-called “flash drought”. The widely used standardized precipitation evapotranspiration index (SPEI), with traditional time scales longer than 1 month, cannot easily capture flash drought signals. Here, the SPEI with a 5-day (pentad) time scale was proposed to investigate flash droughts in China from 1980 to 2017. New criteria for flash droughts based on the high-resolution index were also developed. Flash droughts were stronger and longer in western and northern China. The intensity and duration of mild, moderate, and severe flash droughts increased during the study period, while severe flash droughts showed the largest increase. The high-resolution SPEI permits us to examine the process of a drought event, that is, how a mild flash drought develops into a severe flash drought. The contribution of precipitation deficit to the duration of mild flash droughts was 50%. As the intensity of flash droughts increased, the contribution of high temperature and increased net radiation to flash drought duration increased from 50% (mild flash droughts) to 74% (severe flash droughts). When flash droughts occurred, Northwestern China and Southwestern China were dominated by unusually weak water vapor transport, while Northeastern China, Northern China, Southern China, and Middle-Lower Yangtze were mainly dominated by abnormal downward motion. The downward motion was generally accompanied by clear sky conditions with fewer clouds, higher air temperature, and higher surface solar radiation, which accelerated and amplified flash droughts triggered by initial precipitation deficits.

The Korea Meteorological Administration(KMA) is producing an impact-based forecast data based on both the deterministic forecast and ensemble forecast for heat waves (HW) and cold waves (CW). Ensemble prediction system for impact-based forecast is Multi-Model ensemble system which integrates Unified Model(global, global ensemble, local, and local ensemble models), ECMWF(global and global ensemble models) and KIM(Korean Integrated Model) global model(Hereafter, impact-based forecast based on the deterministic forecast and Multi-Model ensemble are called `DIMF` and ‘MEPS’, respectively.) MEPS determines the risk level by using the probability of occurrence of abnormal temperatures in Korea. Once maximum feels-like temperature(HW) or lowest temperature(CW) from all 93 Multi-Model Ensemble members were extracted, their probability distribution was determined by using a Generalized Extreme Value (GEV) distribution. The performance of MEPS for HW was compared with DIMF for July and August 2022. Verification was conducted by evaluating how well impact-based forecast level(safe, concern, caution, warning, alarm) was matched to the observed risk level in 175 regions. As a result, the ETS score of MEPS is better than DIMF at ‘caution’ and ‘warning’ level in HW but tend to overestimate HW, and the CW MEPS has slightly lower performance than DIMF. If the GEV distribution is used, definition such as a 2-day-lasting high/low temperature or temperature decrease cannot be reflected. Instead of the GEV, a method of obtaining the probability by the actual number of members satisfying the specific condition among the total number of ensembles was developed, and in this way, the tendency of overestimation of the HW in MEPS was slightly improved. In addition, two MEPS that using GEV and the new method about 2023 July to August are compared and the effect of the addition of the KIM ensemble on the performance is investigated. 

Urban solar radiation measurements are crucial for evaluating photovoltaic potential and human thermal sensation. Standard pyranometers are expensive and deploying them in multiple locations across the city is challenging. Urban climatological research demands inexpensive and compact solar radiation sensor. There are commercially available pyranometers using silicon photodiodes which are cheaper and downsized compared with standard ones. However, the wavelength covered by silicon photodiodes is 0.4 to 1.1 μm, which is limited to the narrow band for the entire wavelength range of shortwave radiation (0.29 to 3.0 μm); thus, applying pyranometers with silicon photodiodes to urban area might produce inaccurate data due to the varying spectral reflectance of building materials. Here, we scrutinized several types of pin photodiodes which are even less expensive and smaller than photodiode pyranometers. We assessed the viability of assembling pyranometers by ourselves using pin photodiodes. Pin photodiodes emit an electrical current corresponding to the light intensity. The sensitivity-wavelength range of a standard photodiodes is 0.7 – 1.1 μm, although this can vary among different products. Moreover, the size of the photosensitive areas also differs between various models. In this study, 12-day fixed-point outdoor observations were carried out using four varieties of pin photodiodes with different wavelength ranges and photosensitive areas to select the appropriate sensor type for solar radiation measurement. The results showed that the solar radiation measurement accuracy was superior in pin photodiodes with a larger photosensitive area and a wavelength range extending from ultraviolet to near-infrared. In addition, a mobile observation was performed to evaluate whether PIN photodiode pyranometers can be used in the complex radiant environment with different reflective properties of various building materials. The results showed that reflected solar radiation from vegetation and asphalt was not measured accurately with PIN photodiode pyranometers.

Despite the fatal impact of heavy precipitation on people’s lives and the social economy, its accurate estimating remains challenging, especially in the Northeast China with the complex terrain and distribution of land and sea. In this study, we show how this issue can be addressed by kilometer-scale simulations as well as how to reduce computational costs. Three typical heavy precipitation events are simulated at 3 km horizontal resolution, and each event is simulated with 24 combinations of parameterization schemes and the simulations are evaluated with gauge observations and compared to satellite products (IMERG, CMORPH, and GSMaP). For the hourly maximum precipitation, the ensemble mean of simulations outperforms all satellite products in the cold vortex case and the heavy snowfall case, and is second only to the best-performing IMERG in the typhoon case. For average accumulated precipitation, the ensemble mean of simulations outperforms all satellite products in the cold vortex case and the heavy snowfall case, and is second only to the best-performing CMOPRH in the typhoon case. The physical configuration of the Morrison scheme and Mellor-Yamada-Janjic scheme without scale-aware cumulus scheme demonstrates the best overall performance. The microphysics scheme significantly impacts the maximum hourly precipitation, whereas the planetary boundary layer (PBL) scheme has a greater influence on the average accumulated precipitation. Employing a scale-aware cumulus scheme only has a pronounced impact on the average accumulated precipitation of the typhoon case. Additionally, the difference in precipitation changes caused by employing the scale-aware cumulus scheme on the microphysics scheme caused by different PBL schemes can exceed 10%. Conducting 24 simulations can be unnecessary because of similarities among simulations (e.g., patterns of accumulated precipitation). Referring to the estimation of data reconstruction to supplement additional simulations based on perturbing one physics scheme at a time resulting in 7 simulations is a recommended new strategy.

In this study, we conducted sensitivity tests using the Weather Research and Forecasting (WRF) model for summer heavy rainfall cases in Korea in terms of horizontal resolution and physical parameterization schemes. The model employs one-way nesting with horizontal resolutions of 27 km, 9 km, and 3 km, respectively, over the Korean Peninsula (KP), using the NCEP FNL Operational Model Global Tropospheric Analyses with horizontal resolution of 1.0°x1.0° for the initial and boundary conditions. Two heavy rainfall events, occurred in July 2011 and August 2022 over central South Korea, have been considered. The selected cases occurred when blocking highs were located in the northeast of the KP. Over the KP, warm humid air was supplied to the low-level atmosphere from the North Pacific high (NPH), whereas cold and dry air flew into the mid-level of the troposphere along the trough located in the north of KP — causing strong instability. Low-level jet stream was combined with upper-level jet stream; the southwesterly, which supplied heat and water vapor, was lifted and formed tall convective clouds, resulting in heavy rainfalls. The dataset observed from the Automated Weather Stations (AWS) over KP was used to evaluate the model performance. The outcome presented shows importance of finding the best-performing scheme set and horizontal resolution for predicting extreme precipitation.

An inverse relationship exists between temperature and relative humidity (RH) generally, to be more precise with assumption that total water vapor in the atmosphere doesn’t change. In 2023 summer, however, both temperature and RH were elevated over Korea, deviating from the expected physical relationship. The combination of high temperature and high relative humidity induces many heat-related patients. Thus, this study aims to elucidate the variation in compound heat-humid extremes and to comprehend aspects that temperature alone cannot capture. Analyzing station data from the Korean Meteorological Administration (KMA), the time series of daily maximum temperature (Tmax) in Korea during the summer exhibits a consistent increase from 1973 to 2023. In contrast, the RH initially decreased until 2001 and subsequently began to rise. The period is divided into two periods: P1 (1973-2001) and P2 (2002-2023), identifying Tmax and RH. In P1, shows insignificant increase, while RH significantly decreases. In P2, both Tmax and RH exhibit significant increases, and the increase in RH is predominant in southern part of Korea. Extending this analysis to East Asia using ERA5, Japan, Manchuria region and southern China, and the maritime continent demonstrate similar trend to Korea. Since 2001, the rise in RH has brought the atmosphere close to saturation, leading to increased discomfort for humans and a higher incidence of heat-related diseases.

In July–August 2022, Pakistan suffered historic flooding while record-breaking heatwaves swept southern China, causing severe socio-economic impacts. Similar extreme events have frequently coincided between two regions during the past 44 years, but the underlying mechanisms remain unclear. Here we use observations and a suite of model experiments to show that the upper-tropospheric divergent wind induced by convective heating over Pakistan excites a stationary anomalous anti-cyclone over eastern China, which further leads to persistent heatwaves. Atmospheric model ensemble simulation further indicates that this dynamic pathway linking Pakistan flooding and East Asia heatwaves is intrinsic to the climate system, largely independent of global sea surface temperature forcing. This dynamic connection via the upper troposphere westerly waveguide is robust, offering hopes to improve the sub-seasonal prediction of extreme events in East Asia.

Extreme precipitation refers to a bipolar climate phenomenon in which a high amount of precipitation occurs in a short period or a drought persists for a long period. In a future climate with increased CO2 concentrations, the characteristics of extreme precipitation can undergo significant variations. This study focuses on East Asia (110°-150°E, 20°-50°N) and employs six indices from the Expert Team on Climate Change Detection and Indices (ETCCDI) to assess the reversibility of extreme precipitation events. The Carbon Dioxide Removal (CDR) experiment, simulated by the National Institute of Meteorological Sciences and the Korea Meteorological Administration (NIMS-KMA) climate model, involves increasing the CO2 concentration by 1% per year from the Pre-Industrial (PI) level and decreasing it from four different carbon-neutral points: A (44 years), B (51 years), C (70 years), and D (140 years) from the initial year. The NIMS-KMA simulation proves most effective among eight models from the Coupled Model Intercomparison Project Phase 6 (CMIP6). Results indicate that extreme precipitation indices respond nonlinearly to CO2 concentration, with intensity and frequency indices showing hysteresis. Reversibility is limited, and delayed carbon neutrality leads to increased irreversibility. Notably, the R99 frequency index exhibits the highest irreversibility, ranging from 10.61% to 29.50% from point A to point D. This suggests that postponing carbon neutrality may strengthen the central Pacific warming pattern, intensify subtropical high pressure in the northwest Pacific, and increase water vapor flow into East Asia.

In 2021, Ministry of Ecology and Environment of the People’s Republic of China issued the “Pilot Work Plan for Carbon Monitoring and Evaluation” to implement carbon monitoring and assessment pilots for regions, cities, and key industries. The plan specified the need for inversion research on greenhouse gas (GHG) emissions, with a requirement for a prior inventory. Past studies commonly utilized the Emission Database for Global Atmospheric Research (EDGAR) as the prior inventory. However, numerous studies highlighted location mismatches in EDGAR, particularly for high-emission facilities, leading to significant errors in inversion results at the city scale. Therefore, ensuring an accurate and reliable gridded prior inventory at the city level is crucial for improving GHG inversion. In this study, a gridded emission inventory was developed for the predominant greenhouse gas in the atmosphere, namely carbon dioxide (CO₂). The approach involved establishing a relationship between tabular GHG emission inventories and spatial proxy data to compile a gridded emission inventory at the city scale. The key challenge of the study lay in acquiring tabular emission inventories and spatial proxy data, especially the spatial location information and activity data for high-emission point sources, which are mostly confidential in the study area. The China City CO₂ Emission Dataset (2020) served as the tabular emission inventory, while proxy data for non-point source emissions were derived from publicly available population data, road networks, land use, etc. Obtaining proxy data for point source emissions involved using web crawler technology. Points of interest extracted from online map APIs served as spatial proxy data, and the associated business registration information obtained through crawling was utilized as numerical weight data. Based on the proxy data, emissions from each sector were spatially allocated to the target grid to compile the gridded CO₂ emission inventory.

Globally, the power sector is significant for air quality and climate. The past three decades have witnessed the dramatic expansion of global biomass- and fossil fuel-fired power plants, while the tremendously diverse power infrastructure shapes different air pollutant and CO₂ emission characteristics. Here, a new version of the Global Power Emissions Database (GPEDv2) compiles unit-level air pollutant emissions (e.g. SO₂, NOx, and PM_2.5) during 1990-2020 based on bottom-up compilation approaches. By combining GPEDv2 constructed in this study and the previously developed China coal-fired power Plant Emissions Database (CPED), we find that global CO₂ emissions from the power sector increased from 7.1 Gt in 1990 to 13.0 Gt in 2020, driven by the growth of power demand. In contrast to the 87% increase in CO₂ emissions, global SO₂ emissions of power plants decreased by 56% (from 50.4 Mt to 22.2 Mt), NOx only increased by 5% (from 19.6 Mt to 20.6 Mt), and PM_2.5 decreased by 38% (from 3.8 Mt to 2.4 Mt). The disproportionately changes in air pollutant emissions were mainly due to the role of emission controls. With stringent emission control measures deployed in Europe, the United States and China, the predominant air pollutant emission contributor has shifted from Europe and the United States in 1990 (e.g. 65% for SO₂) to India, the Middle East and North Africa and the Rest of Asia (excluding China) in 2020 (e.g. 61% for SO₂). Our results suggest that developing countries that currently lack effective emission control should strengthen emission standards to substantially reduce air pollutant emissions from power plants.

To mitigate climate change, China is striving to reduce its carbon dioxide (CO2) emissions before 2030 and has derived a series of national environmental policies to advocate green development. Analysing the changes in regional atmospheric CO2 concentrations could be more convincing in evaluating the impact of the associated strategies. In this study, a three-year continuous observation campaign of the atmospheric CO2 mole fraction was first conducted at the summit of the Nanling Mountains, southern China in 2015–2017 to probe the timely feedback of the implementation of the green policies of the government. The effects of regional emissions and removals on the CO2 mole fraction were subsequently analysed to explore the primary processes and regional contributions driving the background CO2 concentration. Approximately 86.4% of the observed values were filtered as background events, with average CO2 concentrations of 395.9 ± 6.8 ppm. The curve-fitting CO2 mole fraction showed a slight short-term descending trend (0.6–2.2 p.m. y− 1) that was rarely seen against the backdrop of a continuous rise of the global average over the years. In addition to the possible impacts of the changes in regional-scale CO2 sources and sinks, we inferred that the vigorously enforced sustainable development policies of the Chinese government should be accentuated to reduce CO2 emissions. Although the conclusion relies heavily on a relatively short dataset, this study may globally help the people to reduce carbon emissions.

Methane (CH4) is the second most important greenhouse gas, following carbon dioxide. The analysis of the distribution of atmospheric methane concentrations on a regional scale has become a focal point of international research. Antarctica is the most vulnerable region of the global ecological environment. Monitoring of methane concentrations above Antarctica will provide a basis for assessing the impact of the greenhouse effect on the global climate. The Atmospheric Infrared Ultra-Spectral Sounder (AIUS) carried on the Gaofen 5 satellite is China’s first infrared occultation detector and is capable of observing ozone and various trace gases over Antarctica. This paper assesses the retrieval algorithm (based on the Optimal Estimation Method) for deriving atmospheric methane profiles over Antarctica in 2019 from the AIUS spectra. The results indicate that the deviation between the April, May, and December 2019 data and the ACE-FTS methane profiles is generally within ± 5% in the 12-60 km altitude range. In August, when the altitude is below 20 km, the retrieval accuracy is within ± 3%, and above 20 km, the accuracy gradually decreases with increasing altitude. The retrieval error of the November data is within ± 3% in the 10-60 km range. The methane concentration in Antarctica shows an obvious seasonal variation. The methane concentration in November and December is higher than in other months, which may be related to the significant release of methane from the melting of the Antarctic ice core during the summer.

This study investigates the impacts of near-term climate forcers (NTCFs) mitigation on the near-term (2045-2054) global surface air temperature (SAT) and Arctic sea ice concentration (SIC) changes. Specifically, the impacts of methane and non-methane NTCFs are analyzed by examining five climate models from three Shared Socioeconomic Pathways (SSPs) scenarios: control scenario (SSP3-7.0), low non-methane NTCF emission scenario (SSP3-7.0-lowNTCF), and low all NTCF emission scenario (SSP3-7.0-lowNTCFCH4). The impacts of all NTCF and non-methane NTCF mitigations are defined as the changes in SAT and SIC between the SSP3-7.0-lowNTCFCH4 and SSP3-7.0 scenarios and between the SSP3-7.0-lowNTCF and SSP3-7.0 scenarios, respectively. The methane mitigation is also isolated by the difference between the SSP3-7.0-lowNTCFCH4 and SSP3-7.0-lowNTCF scenarios. Global SAT is projected to increase under the control scenario, which is based on regional rivalry over carbon emission policies, with the warming being the largest in the Arctic. The methane mitigation reduces global warming, while the non-methane NTCF mitigation strengthens global warming. The former is qualitatively stronger than the latter, especially in the Arctic from late fall to early winter (October to January). Consistent with the change in Arctic SAT, Arctic SIC is projected to increase with methane mitigation, but decrease with non-methane NTCF mitigation. These results suggest that air quality control policies, accompanied by methane reduction, are necessary to slow down global warming and preserve Arctic sea ice.

Current and expected future aerosol emission changes are particularly strong in East and South Asia, where high population densities imply high potential climate risk. Hence, there is an urgent need for improved knowledge about the near-term influences of changes in aerosol emissions. Here we have developed a set of Systematic Regional Aerosol Perturbations (SyRAP) using the reduced complexity climate model FORTE 2.0 to explore the effects of aerosol-driven climate change. Results show that the increased Black Carbon(BC) concentrations over China and India lead to decreased local surface Temperature (Ts) and precipitation, with seasonal differences in the spatial distribution. Chinese (Indian) BC emissions also impact on Indian (Chinese) climate in specific seasons. The changes of shortwave radiation (SW) dominate the surface cooling and the lower tropospheric warming due to the absorption of BC. The reductions of column-integrated diabatic cooling lead to the decreased local precipitation, while the changes in atmospheric circulation play an opposite role (weakened EAWM, enhanced EASM and ISM). The horizontal/vertical distributions of air temperature anomalies can induce the changes in cloud cover and atmospheric circulation, which further impact on the radiation flux and precipitation. Additionally, the increased surface albedo in winter is helpful to decrease Ts and precipitation.

Recently, there has been significant global interest in carbon neutrality. The Intergovernmental Panel on Climate Change (IPCC) Special Report approved in 2018 set a target to limit the temperature increase to 1.5°C or less by 2100, emphasizing the need to achieve carbon neutrality by 2050. However, the World Meteorological Organization (WMO) announced in 2023 that the global average temperature had already increased by 1.4°C compared to pre-industrial levels. During the 28th Conference of the Parties (COP) to the United Nations Framework Convention on Climate Change (UNFCCC) in 2023, the first Global Stocktake (GST) was conducted to assess progress towards the goals of the Paris Agreement, outline future directions, and it concluded that the implementation of the Paris Agreement is insufficient across all sectors and has not reached the specified levels. As a result, there should be increased efforts in climate change mitigation and carbon neutrality. The Republic of Korea (Korea) has announced two carbon neutrality scenarios with a goal of achieving net-zero by 2050 in 2022. The transition from fossil fuel-dominated energy supply to renewable and environmentally friendly energy sources is expected to lead to a reduction in fossil fuel consumption, consequently influencing air quality. Notably, among the factors affecting air quality, such as Total Energy Supply, Carbon Intensity, Control Status, and Local Air Pollutant Emission Effects, a substantial change is anticipated in Carbon Intensity. This study examines the predicted changes in each factor affecting air quality under carbon neutrality scenario in Korea and comprehensively estimates the expected air quality changes, particularly in the concentrations of primary air pollutants such as carbon monoxide and sulfur dioxide.

The morphology and phase separation of aerosol particles play a crucial role in scattering and absorbing solar radiation, significantly impacting atmospheric energy balance and climate processes. Therefore, understanding these characteristics is important for predicting aerosol behavior in atmospheric chemistry and climate change. Although interest in this area is growing, significant gaps remain in studies examining the morphology and phase separation of ambient aerosols. To bridge this gap, fine particulate matter (PM_2.5) was collected from four Northeast Asian sites (Seoul, Seosan, Beijing, and Ulaanbaatar) in the autumn of 2023 using a high-volume air sampler. These PM_2.5 particles were dissolved in a water and methanol solution, nebulized onto a hydrophobic substrate, and placed in a flow-cell for observation. The particles were analyzed by optical microscopy as the relative humidity (RH) decreased. Interestingly, as the RH decreased, a variety of morphologies and phase separation were observed in the ambient aerosols. It is anticipated that these findings will greatly enhance knowledge about how aerosols behave in the context of atmospheric chemistry and climate change, highlighting the importance of further research. The results will be presented.

In order to analyze the physicochemical mechanisms affecting the variation of PM_2.5 in urban air, we established the UAPRS (Urban Air Pollution Research Site) in Taichung City, the second largest city in Taiwan. The WRF/CMAQ model was used to simulate the impacts of various emissions on the central region of Taiwan. Taking the high pollution event from November 3 to 6, 2021, which was caused by the prevailing easterly or southeasterly winds being blocked by the Central Mountain Range, which resulted in the formation of a weak wind zone in the central region of Taiwan. The wind speed is often weak near the ground at night and the height of the boundary layer is low, one of the reasons for the accumulation of PM_2.5 near the ground in the Dadu Mountain on the west side of Taichung City, and the increase in PM_2.5 concentration. The model evaluation shows that the sources of pollution in Taichung City are not only the boundary condition, but also the point, line and surface effects. SO₄^2- is mainly generated from point sources in Taichung City, mainly from Emission, from medium of H₂O₂, Fe, Mn, and O₃; NO₃^- is also mainly generated from point sources in Taichung City, and HNO₃ formed at noontime and ANO₃ at others; NH₄⁺ from Taichung area source; OM is more complex, mainly from line emission of Taichung City, and other sources such as area and point emissions of Taichung City, and emissions from Changhua County. The most important mechanism is low volatility/semivolatile oxidized combustion OC, which produced most OC in the midday hours, followed by low volatility/semivolatile POA, which may produce OC in the morning or evening; and EC, which mainly comes from line sources in Taichung City and Changhua County.

Aerosol acidity is a key parameter in aerosol chemistry. It can alter the phase partitioning of semi-volatile species, chemical reaction rate, and trace metal solubility, which further impact human health, climate, and terrestrial and oceanic ecosystems. Aerosol acidity is usually characterized by aerosol pH, which is defined as the negative logarithm of the hydrogen ion activity within aerosol liquid water with a base of 10. It is notable that existing studies typically employ the arithmetic mean of pH to represent the average trend of aerosol pH, in fact, this is inappropriate. In this work, the probability distributions of the aerosol pH and aerosol water content (AWC) in the North China Plain (NCP) for winter 2018, as well as their joint probability distribution, were mapped based on GEOS-Chem simulations. Considering the interdependence between aerosol pH and aerosol water content, we proposed an AWC-based pH averaging approach. The results showed that the pH_{weighted mean} and pH_mean during the study period differed by nearly 2 units. Moreover, due to the buffering effect of NH₃/NH₄⁺, this difference become significantly stabilized when the relative humidity (RH) surpasses 50%. In addition, comparing the variations between different averaging methods at diverse temporal resolutions (3-hour, daily, weekly, monthly, and 3-month) reveals that insufficient temporal resolution may result in the transient fluctuations in AWC to be smoothed. This, in turn, can lead to an overestimation of aerosol pH and an underestimation of AWC. Consequently, it is advisable to use hourly-resolution datasets as the inputs for thermodynamic modeling, and employing the AWC-weighted method when reporting the average pH throughout the study is recommended.

Lightning is a phenomenon caused by the discharge of positive and negative charges within clouds, and cloud-to-ground lightning poses significant risks to human life and property. Therefore, understanding the characteristics of clouds where lightning occurs and detecting the areas of lightning occurrence are important. In this study, we investigated the possibility for region-based lightning detection using data from the high spatiotemporal resolution geostationary meteorological satellite GK2A (GeoKompsat-2A) and ground lightning observation equipment LINET (LIghtning NETwork). To use datas with different spatial and temporal resolutions together, lightning events within 5 minutes before and after the satellite observation time were defined as occurring at the satellite observation time. Spatially, the satellite pixel with the shortest distance was found based on the lightning occurrence point, and the pixel with the lowest brightness temperature of the infrared channel in the 15X15 area based on the pixel was defined as the lightning pixel to match time and space. The study selected 12 cases based on the frequency of lightning occurrences during the summer of 2020-2021 (June to August). Among these, 10 cases were designated as training cases, and 2 as validation cases. Lightning primarily occurs in clouds rising near the tropopause due to deep convection. Therefore, to examine the relationship between lightning and brightness temperature, the study analyzed the brightness temperature and lightning occurrence probability graph for the training cases. The results indicated a higher probability of lightning occurrence when the infrared channel brightness temperature was lower, and the brightness temperature difference was close to 0K. Currently, the plan is to derive a regression equation for calculating the probability of lightning occurrence using the training cases. Subsequently, a region-based approach for lightning detection will be designed based on the characteristics of these lightning pixels, and the performance of the regression equation will be evaluated.

Turbulence in thunderstorms affects the charge distribution, which in turn affects the lightning channel morphology. Thus, the lightning channel morphology can reflect the characteristics of turbulence. The current understanding of the correlation between the two is still limited to the relationship between macroscopic thunderstorm dynamic characteristics and lightning activity. In this paper, based on three-dimensional radiation source localization data from the Lightning Mapping Array and radar-based data, our analysis shows that the overall morphology and detailed morphology of the lightning channel correspond to different eddy dissipation rate (EDR) characteristics. Lightning with complex channel morphology occurs in regions with large EDRs. In single lightning events, channels that extend directly within a certain height range without significant bifurcation and turning tend to propagate in the direction of decreasing EDRs, while channel bifurcations and turns usually occur in regions with large radial velocity gradients and large EDRs. This study shows the relationship between channel morphology and thunderstorm dynamics and provides a new method for the direct application of channel-level localization data to understand thunderstorm dynamics characteristics.

This research aims to investigate using radar-retrieved variables to improve short-term rainfall forecasts over mountainous regions. Two advanced and fully developed retrieval algorithms named WInd Synthesis System using Doppler Measurement (WISSDOM) and Terrain Permitting Thermal Retrieval Scheme (TPTRS) are used to retrieve the three-dimensional wind, pressure and temperature fields over complex terrain. In addition, the moisture field is estimated based on the statistical relationship between sounding-observed relative humidity and radar reflectivity. The cases selected for analysis are from IOP#5 of TAHOPE 2022, representing an afternoon thunderstorm case over northern Taiwan, and SoWMEX IOP#8, representing a squall line case over southern Taiwan. Results of retrievals in two cases indicate that strong upward motion is consistent with the convergence. In addition, the regions with positive/negative temperature perturbations also coincide with strong updraft/downdraft in the IOP5 case due to the latent heat release. On the other hand, near the surface, the negative temperature perturbation represents the cold pool and high-pressure structure, respectively. Compared with the observations, the model can well capture significant areas of accumulated rainfall for up to three hours in the IOP5 case and six hours in the IOP8 case. The results also imply that the model spin-up time can be effectively shortened after directly assimilating the retrieved kinematic, thermodynamic, and moisture fields.

EI Nino-Southern Oscillation (ENSO) is a complex ocean-atmosphere interaction phenomenon leading to extreme weather events worldwide and water vapor plays a vital role in the evolution of ENSO. Recently, the Global Navigation Satellite Systems (GNSSs) have been regarded as a relatively new mechanism for quiring high spatial and temporal resolution water vapor due to its unique features of high accuracy, global coverage and all-weather operation capability. However, GNSS derived precipitable water vapor (PWV) has not been well researched for its potential in the study of ENSO. This research is hence to investigate the correlation between ENSO and GNSS-derived PWV in a global context, in particular the variations of PWV time series in response to ENSO before and during floods and droughts. In this study, composite analysis and singular spectrum analysis (SSA) were used to obtain the non-linear trends of PWV anomalies at 16 GNSS stations close to the sea for Nino and Nina years with moderate and higher intensities. Then, the relationships between these non-linear trends of PWV anomalies and ONI show that the correlations on the western Pacific and the eastern Indian Ocean are significantly negative, on the contrary, the correlations on the eastern Pacific and the western Indian Ocean are positive. These results show well consistent with the variations of precipitation in response to ENSO, which to some extent reflect the drought and flooding in these areas. A case study conducted at the COCO station near Indonesia shows that the non-linear trend of the PWV anomalies depicts the evolution of one severe drought event and one severe flood event occurring in Indonesia. These results suggest that GNSS-derived PWV together with other climatic variables can be used as an indication of the evolution of ENSO events and as a possible indicator of drought and flood occurrence.

Ground-based global navigation satellite system (GNSS) water vapor monitoring technology is characterized by low cost and high temporal and spatial resolution compared to traditional water vapor detection methods. In GNSS meteorology, real-time water vapor retrieval is required for weather forecasting applications such as extreme precipitation and typhoon tracks. In this study, real-time orbit and clock error products provided by various analysis centers, such as IGS, CNES, etc., are used to obtain tropospheric zenith delay (ZTD) values using the real-time precision point positioning (PPP) technique. The accuracy of the real-time ZTD estimation is analyzed by comparing it with the high-precision ZTDs calculated using the final products. Then we extracted the tropospheric wet delay (ZWD) values from the ZTD to further calculate the atmospheric precipitable water vapor (PWV) values, and compared them with the PWV calculated from the radiosonde data to analyze the accuracy of the water vapor calculated from the different real-time products, as well as their spatial and temporal properties.

Water vapor, an important greenhouse gas in the troposphere, is closely related to weather and climate phenomena that affects daily life of human beings. Accurately determining the water vapor content helps us deepen the understanding of atmospheric dynamic processes. Precipitable water vapor (PWV) is an essential parameter for measuring the content of water vapor, and some researchers have focused on real-time PWV retrieval for time-critical meteorological applications. This contribution focuses on the algorithm, software development, and accuracy evaluation of real-time PWV retrievals based on the Global Navigation Satellite Systems (GNSS) precise point positioning (PPP) technique. First, an enhanced multi-GNSS real-time PPP software was developed to estimate the zenith tropospheric delay (ZTD) over the GNSS station. Second, an improved ZHD lapse-rate model was developed to enhance the accuracy of VMF1/VMF3-predicted ZHD. Third, an investigation into the modeling of the weighted mean temperature was introduced. Finally, the accuracy of real-time GNSS PWV was analyzed.

Bioaerosols, are also known as primary particles (PBA), are a subset of atmospheric particles. Bioaerosols are either living or dead airborne particles originating in plants, animals, and microorganisms. Early warning and management are crucial for controlling the spread of bioaerosols. Herein, we developed aptamer (SA)-based electrochemical biosensors for rapid and sensitive detection of bioaerosols. The platform features two components: DNA aptamers for their ability to bind and undergo target-induced assembly on the chip surface and functional materials for their ability to provide a high surface density of aptamer-binding sites and facilitate the electron transfer at the biointerface. The sensors were capable of detecting targets with a detection limit to picomole level. Furthermore, this design allowed the detection of ATP in cultured microorganisms and collected real bioaerosols. Overall, this strategy of interfacing DNA aptamers with composite materials represents a versatile approach for the ubiquitous detection of biochemical targets in bioaerosols.

West Antarctic and the Antarctic Peninsula have experienced dramatic warming in austral spring since the 1970s. Using observations and the Community Atmosphere Model version 4 (CAM4), this study explores the physical mechanism by which the tropical Pacific and Indian Ocean temperature anomaly mode (PIM) affects the dipolar surface air temperature (SAT) anomalies across the West Antarctic in austral spring. The positive phase of the PIM, characterized by positive sea surface temperature anomalies (SSTAs) in the tropical central-eastern Pacific and western Indian Ocean and negative SSTAs in the Maritime Continent, can generate two branches of stationary Rossby wave trains propagating from the tropical central Pacific and southeastern Indian Ocean to the West Antarctic, with an anticyclonic anomaly appearing over the Amundsen Sea. The northerlies advect warmer air to the Ross–Amundsen Seas, but southerlies advect colder air to the Antarctic Peninsula–Weddell Sea, resulting in the dipole of SAT anomalies over the West Antarctic. In this process, the role of tropical central-eastern Pacific SSTAs dominate, and it is amplified by the SSTAs around the Maritime Continent. The SSTAs in the western Indian Ocean combined with the SSTAs over the Maritime Continent further contribute to the western pole of the SAT. Only simulation that includes a prescribed PIM forcing can exactly reproduce the observations of the dipolar SAT response across the West Antarctic, indicating the need to treat the tropical Pacific and Indian Oceans as a unified whole.

Atmospheric Rivers (AR) accounts for more than 90% of poleward moisture transport across midlatitudes of both hemispheres and has a crucial impact on hydrological cycle and surface mass balance over the Antarctic ice sheets. Here we investigate the interannual variations of atmospheric river and associated poleward moisture transport in austral summer and winter using 43-year MERRA-2 reanalysis data. 3-hourly integrated water vapor transport (IVT) is used to detect AR with latitude dependent thresholds of IVT to better detect AR-like features in the polar regions. In austral summer, first two modes from eigen analysis of sea ice are highly correlated with AR activities and poleward moisture transport in the Weddell Sea and Ross Sea, respectively. In austral winter, AR impact is limited to the sea ice in the Weddell Sea. In addition, ZW3 circulation is highly correlated with sea ices in both regions in austral winter. To better understand the effect of AR on sea ice in different parts of the Antarctic, AR genesis regions are identified and AR propagation from the genesis regions are analyzed. Based on 3-hourly AR statistics, four main AR genesis regions in the southern hemisphere (i.e., southern Africa and southwestern Indian Ocean, southern Pacific Ocean, South America monsoon region, and Australia) are identified. The variations of AR frequency and intensity at each genesis regions and associated patterns of atmospheric circulations and its impact on the poleward moisture transports as well as surface heat and radiation flexes will be discussed.

Antarctica is experiencing unprecedented regional warming and accelerated loss of mass. Planetary boundary layer processes play a key role in these changes. The Antarctic boundary layer often features a strong surface-based inversion (SBI) due to the semi-permanent net radiation deficit at the surface. The stably stratified atmosphere can affect not only mixing process but also interact with clouds and blowing snow. This paper presents the work on Antarctic boundary layer analysis using co-located data from dropsonde, radiosonde, and the CALIPSO satellite, where the in-situ observations provide high resolution vertical atmosphere profile and CALIPSO provides the cloud and blowing snow information. Results show that 1) the Antarctic boundary layer demonstrates clear regional signatures. Factors such as wind shear, radiative forcing, terrain and air mass influences, play a role in the boundary layer mixing; 2) signals of interaction between cloud and the boundary layer are clearly observed. Increased downwelling longwave radiation during cloudy cases evidently abate radiative cooling losses during all seasons, contributing to surface and boundary layer warming, especially during spring, fall, and winter; 3) blowing snow, which is a common phenomenon over the Antarctic continent, can serve as an indicator of boundary layer mixing. SBI strength plays a role in the blowing snow layer development. 

Atmospheric Rivers (AR) are globally occurring phenomena in which water vapor generated in tropical oceans flows towards continents through jet streams, spanning hundreds to thousands of kilometers. When AR events occur, they transport substantial amounts of water vapor, leading to heavy rainfall and snowfall in inland. In the Antarctic, the occurrence of atmospheric rivers results in the melting of ice shelves, the formation of polynya in the Antarctic sea ice, and localized accumulation of snow on glaciers. Previous studies have indicated that atmospheric rivers frequently occur due to tropical cyclones and are associated with climate modes on various timescales, such as the Southern Annular Mode and Pacific-South American pattern 2 on interannual timescales, and the Pacific Decadal Oscillation on decadal timescales. However, the analysis on the phenomenon related to intraseasonal timescale is rare. Therefore, this study diagnoses the relationship between the Madden-Julian Oscillation (MJO) characteristics and the frequency of atmospheric rivers in the Antarctic. The MJO is the most prominent atmospheric oscillation, accounting for 20-30% of the intraseasonal variability in tropical regions. It has a temporal cycle of 30-70 days and exhibits a slow eastward propagation with a speed of approximately 5 m/s. In this study, MJO phases are categorized into four groups (phases 2-3, phases 4-5, phases 6-7, and phases 8-1). In phases 4-5, the vertically integrated water vapor transport (vIVT) flowing into the Antarctic Peninsula (AP) is large, and the surface temperature is increasing in the eastern part of AP, which can accelerate the melting of glaciers.

During the Arctic summer expeditions of the Korea Polar Research Institute IBRV Araon, synthetic ship-borne surface and upper-air observations were carried out by the radiosonde balloons, micro-pulse lidar, and surface automatic weather station. Valuable in-situ observation data were obtained over a variety of surface and synoptic conditions, including upper-air met variables (temperature, humidity, pressure and horizontal winds), backscattered signals from clouds, and surface met variables and downward radiative fluxes. These data with reanalysis are used to study the surface radiative effect of clouds over the Arctic Ocean in late summer. Cloud base heights and cloud base temperatures are estimated from the combination of micro-pulse lidar backscatter signals and radiosonde temperature profiles. Results show the cloud base heights are close to the surface during the cold period, but become higher as the temperatures warm. A positive linear relationship dominates between cloud-controlled downward longwave radiative flux and surface air temperature regardless of the sea ice concentration ranges (high: ≥70%, intermediate: 10-70%, low: <10%), but the presence of the warm and clear period obscures the linear relationship.

We developed a seasonal climate forecasting system by utilizing the National Center for Atmospheric Research's (NCAR) Community Atmosphere Model version 6 (CAM6) and incorporating sea surface temperature and sea ice concentration boundary data from the National Centers for Environmental Prediction's (NCEP) Climate Forecast System (CFS) reforecast dataset. In the 4-month winter hindcast spanning from 2000 to 2019, initialized on November 1st each year with 11 ensemble members, it was found an anomaly correlation coefficient (ACC) of 0.2 for the detrended winter (December-January-February) mean surface air temperature (SAT) when compared to reanalysis data in the Arctic region (north of 67°N). The climate feedback response analysis method (CFRAM) indicated that cloud-radiation feedback had the most significant impact on SAT bias in the Arctic. Given the similarity in the relationship between Arctic and East Asia SAT within the hindcast results and observations, this implies that improving the accuracy of Arctic cloud simulation could enhance the SAT prediction skill in both the Arctic and East Asia regions.

Shandong Province in China, with three sides facing the sea, is the province in the northern region of our country that is most affected by typhoons (TC). The number of typhoons affecting Shandong Province is not as high as those along the southeast coast. However, once they occur, they are often interact with mid-latitude weather systems, causing widespread torrential rainfall. Against the background of global warming, TCs can form at higher latitudes, and there is a trend for the location of their maximum intensity to shift towards the poles. Northern China is more susceptible to the impact of typhoons than before, and the typhoon precipitation in this region needs to be widely concerned. This study selected 43 typhoons that have affected Shandong Province from those generated in the Northwest Pacific since 2000 for statistical analysis. The results indicate that on average, there are 2 typhoons that affect Shandong per year, with a concentration between July and September. Most of the typhoons that affect Shandong Province are transitioning typhoons, and their movement speed during the impact period is relatively fast. The intensity of these typhoons is weakened compared to their maximum intensity during their lifetime. The intensity of precipitation is mainly affected by the typhoon's path, movement speed, and intensity，the transitioning status also plays a role, specifically: the shorter the distance between the typhoon center and Shandong, the slower the movement speed, and the stronger the intensity, the greater the precipitation in Shandong during the impact period. Based on these factors, this study further derived a quantitative estimation formula for extreme precipitation in Shandong Province. Due to the influence of environmental vertical wind shear, heavy rainfall in Shandong caused by typhoons with smaller (larger) path curvature mainly occurs on the left (right) side of the typhoon's movement path.

This study revisits the issue of why tropical cyclones (TCs) develop more rapidly with lower environmental inertial stabilities, using ensemble axisymmetric numerical simulations and energy diagnostics based on the isentropic analysis, with the focus on the relative importance of the outflow-layer and boundary layer inertial stabilities to TC intensification and energy cycle. Results show that although lowering the outflow-layer Coriolis parameter and thus inertial stability can slightly strengthen the outflow, it does not affect the simulated TC development, whereas lowering the boundary layer Coriolis parameter largely enhances the secondary circulation and TC intensification as in the experiment with a reduced Coriolis parameter throughout the model atmosphere. This suggests that TC outflow is more likely a passive result of the convergent inflow in the boundary layer and convective updraft in the eyewall. The boundary layer inertial stability is found to control the convergent inflow in the boundary layer and depth of convection in the eyewall and thus the temperature of the energy sink in the TC heat engine, which determines the efficiency and overall mechanical output of the heat engine and thus TC intensification. It is also shown that the hypothesized isothermal and adiabatic compression legs at the downstream end of the outflow in the classical Carnot cycle are not supported in the thermodynamic cycle of the simulated TCs, implying that the hypothesized classical TC Carnot cycle is not closed. It is the theoretical maximum work of the heat engine, not the energy expenditure following the outflow downstream, that determines the mechanical work used to intensify a TC.

Challenges persist in accurately predicting sharp changes in tropical cyclone (TC) motion over a short period of time, even with the employment of state-of-art forecasting technologies. The precise connection between these sudden changes and specific TC structure remains unclear. Here, we delve into the relationship between TC asymmetry (TCA) of outer-core size and anomalous motion, using best-track data spanning the period from 2001 to 2022. Results indicate that TCs characterized by lower TCA tend to display more pronounced deflections than their normal motion. Furthermore, fast-moving TCs exhibit heightened asymmetry and a propensity to accelerate, whereas slow-moving ones lean towards greater symmetry. In addition, TCs demonstrating substantial angular deviations are more prevalent at lower speeds, while fast-moving ones rarely generate anomalous deflections. These findings provide valuable insights into the potential impact of TCA on anomalous TC motion, which can ultimately be used to enhance the accuracy of TC track forecasting.

In 2023, Storm Freddy emerged as the most long-lived tropical cyclone (TC) in record, lasting 35 days over the southern tropical Indian Ocean and spanning both weather and sub-seasonal to seasonal time ranges. The primary objective of this study is to understand the factors contributing to the poor predictability of Freddy in forecasts spanning over 2 weeks. This holds significant importance as our understanding about the ability of the numerical models to predict long-lived TCs remains limited. Using over 7,000 global ensemble forecasts from five global Numerical Weather Prediction (NWP) centers and a high-resolution regional model, we identified three key factors contributing to the limited predictability of Freddy: the strength of the Mascarene High, the position of Storm Dingani (2023), and the size of Freddy. In large track-error results of the global forecasts and regional simulations for Freddy, the strength of the Mascarene High was underestimated, Dingani was located further northeast, and Freddy was either too large or too small. These findings were further validated through a high-resolution regional model. Specifically, Freddy's track and intensity most closely matched the observations when these three factors were most closely represented. Our study underscores the pivotal role played by the interaction between TCs and multi-scale systems for long-lived TCs.

Urban population and greenhouse gases emissions have grown in recent years. Although Tokyo, Japan, is one of the largest cities in the world with a population of over 37 million, a quantitative evaluation of carbon dioxide (CO₂) emissions from Tokyo is insufficient. We estimated net CO₂ fluxes from Tokyo for two years from 2019 to 2020 in combination with a global high-resolution model simulation by the Nonhydrostatic ICosahedral Atmospheric Model (NICAM) and the observation at around 250 m height of Tokyo Skytree (35.71°N, 139.81°E), a freestanding broadcasting tower. Observed atmospheric CO₂ variations were well simulated at remote sites around Japan owing to use of flux data from a global inverse analysis. Tagged tracers simulations estimated Tokyo-originated CO₂ concentration and showed that variations of CO₂ concentration in Tokyo area are strongly influenced by wind; southerly winds caused by the sea breeze transport airmass with high CO₂ concentrations from power plants and industrial areas distributed along the bay. By removing CO₂ data with low wind condition under southern-wind condition, the correlation coefficient of Tokyo-originated CO₂ concentrations between the observation and the model is increased reaches the maximum when CO₂ data with wind speed lower than 5.5 m/s are removed. Meanwhile, Radon, which has a constant flux distribution in the model, does not show drastic change of the correlation coefficient under southern wind condition with removing low wind speed, suggesting that the changes of correlation of CO₂ with removing low wind is caused by the unrealistic flux distribution in the model. This study corrected a prescribed CO₂ flux from Tokyo by observation with the removal of the low wind speed data to 10.5 ± 2.2 kg C m^-2 year^-1 (82.5 ± 17.3 Tg C year^-1 in net emissions from the target area in Tokyo).

Long-term and high-frequency atmospheric CO₂ measurements at multiple sites in Salt Lake Valley (SLV), Utah, have demonstrated that the annual and monthly patterns of CO₂ variability align with a priori estimates of CO₂ emissions from various anthropogenic and biological sources. In this study, we investigate whether short-term changes in anthropogenic sources, as captured in the Vulcan emissions dataset for the US, can be detected through atmospheric CO₂ observations. Specifically, we focus on the Thanksgiving holiday period, during which traffic and energy usage patterns are expected to differ from other periods in November. We observed that on-road CO₂ emissions peak during weekday morning and evening rush hours, but are lower during the Thanksgiving holidays and weekends. The CO₂ concentrations also show the double peak, suggesting the dominance of traffic emissions in the CO₂ diurnal cycle. Interestingly, atmospheric CO₂ concentrations during the Thanksgiving holidays were higher than the other periods of November at all SLV monitoring sites between 2008 and 2013, predominantly due to increased non-onroad emissions. While the temporal patterns of emissions and CO₂ mole fractions at the monitoring sites were similar, there were differences in the spatial patterns. The suburban Rose Park and downtown Murray sites exhibited the highest CO₂ mole fractions among all observed locations. Vulcan anthropogenic emissions data, however, indicated that the emissions at Murray were relatively lower compared to those at Rose Park. The Murray site is located next to a major Interstate freeway, which runs north-south, whereas the Rose Park site is away from the freeway. The CO₂ mole fractions were elevated primarily when southerly winds advected emissions along the freeway. This study highlights the role of local processes in understanding CO₂ variations in an urban monitoring network. 

The quantification of methane emissions from fugitive natural gas leakages presents a challenge, however, it is essential for the effective mitigation of methane emissions. The Korea Carbon Project (KCP) endeavored to quantify the fugitive methane emissions from the Seoul Combined Cycle Power Plant (SCCPP) using mobile measurements over approximately 6 months, from January 12 to July 12, 2023. This approach employed an electric vehicle-based atmospheric GHG monitoring platform, which had instruments for measuring CH₄, C₂H₆, and CO₂ mounted on it. Three major emission sources from the SCCPP were identified through the mobile measurement strategies and their sources were determined by ethane to methane ratios and methane to carbon dioxide ratios. The maximum methane enhancements from the three major sources were quantified at 56,039, 3,795, and 1,051 ppb, respectively. To quantify the methane emissions from these sources, the Gaussian Plume Dispersion Model (GPDM) and OTM-33a method were used. The emission rates from three major sources at the SCCPP were 7.66±0.8, 7.97±1.03, and 0.31±0.03 GgCO₂eq/yr, respectively. The emission rates obtained from the mobile measurements were compared with the bottom-up approach, based on the 2023 fuel consumption data at the SCCPP. These results revealed that the emissions from the three major sources identified at the SCCPP through mobile measurements constituted 35.41%, 21.16%, and 0.73% of the respective bottom-up based inventory estimations. This study suggests that the discrepancy between actual and known emissions could be reduced through measurement-based estimation of methane emissions.

Efforts to enhance greenhouse gas (GHG) emission reduction in East Asia play a pivotal role on both global and regional scales in advancing climate mitigation strategies. This study aimed to better constrain anthropogenic CO2 emission estimates by expanding the network of near-surface in-situ stations for CO2 observations across South Korea. To achieve an optimal CO2 network design, we conducted an Observing System Simulation Experiment (OSSE) coupled with STILT, utilizing meteorological data from the Korean Integrated Model (KIM). Our inversion setup incorporated two CO2 emission datasets with a 0.1o resolution: EDGAR v6 for prior emissions and GRACED for truth emissions. A uniform model-mismatch error of 3 ppm was introduced across sites. The effectiveness of the existing five in-situ stations, termed the base network, in South Korea was evaluated to gauge their ability to constrain CO2 surface flux estimates. However, the findings revealed a reduction in flux uncertainty of only 29.2%, which fell short of the desired uncertainty reduction goal. In this base network, the Lotte World Tower (LWT: 37.5126˚E, 127.1025˚E) in Seoul and the Anmyeondo (AMY: 36.538576˚ N, 126.330071˚ E) site in Taean county stood as major contributors, with estimated reductions of 17.48% and 6.35%, respectively. Consequently, we proposed and developed an extended network, identifying seven candidate sites based on consideration of logistical factors, existing infrastructures, and proximity to the emission source regions. An incremental optimization scheme ranked their contributions, resulting in an additional 25% reduction, bringing the total to 54.13%. However, it is noteworthy that diminishing returns (ranging from 13% to less than 0.1%) were observed with an increase in station count mainly due to the possibility that adding a station earlier in the sequence might render subsequent stations redundant. Despite this, the proposed CO2 network successfully reduced uncertainty in emissions.

The majority of people residing in urban areas is leading to the current rapid urbanization. This urbanization process has resulted in the complex microclimate phenomena unique to cities attributed to the reduction in green areas and the construction of high-rise buildings. Buildings, asphalt, and concrete in cities absorb and release radiation, inducing to heat-related environmental issues. Additionally, atmospheric stagnation between densely packed high-rise buildings, coupled with enhanced social activities due to population growth lead to elevated CO₂ concentrations. Urban environments are changed based on land surface conditions such as adjacent building or green spaces, and more. This study simulates urban environmental changes in Gwangjin-gu district in Seoul using the ENVI-met model, which takes into accounts factors such as buildings, vegetation, land cover, and meteorological conditions. Through four-season simulations, it was confirmed that enlarging green spaces is the most effective method to mitigate the urban heat island and reduce CO₂ concentrations. Our findings offer crucial guidelines for sustainable urban development and climate change mitigation, aiming to contribute to urban planning and environmental policymaking.

To attain Net Zero by 2050, it is crucial to accurately estimate the emissions of CO and long-lived greenhouse gases (GHGs), such as CO₂ and CH₄. Traditionally, GHGs monitoring has been conducted by long-term measurements at fixed locations and on global scale using satellite observations to capture trends. Recently, aircrafts have been employed to measure the vertical and spatial distribution of GHGs concentrations from the surface to the upper troposphere. Using GLA331-MCEA1 (OA-ICOS; ABB-LGR) which was equipped on a research aircraft (KingAir-C90GT), we measured CO, CO₂, and CH₄ concentrations over the Yellow Sea of Korea and nearby large coal-fired power plants in 2022 and 2023. To investigate the characteristics of regional emission sources, we calculated the ratios of ΔCH₄/ΔCO₂, ΔCH₄/ΔCO, where delta (Δ) indicates the subtraction of background concentrations (the 1st percentile of each flight), from the measured values. Subsequently, based on back-trajectories from the HYSPLIT model, we analyzed the regional and altitude differences in ΔCH₄/ΔCO₂ and ΔCH₄/ΔCO, which was divided into administrative districts. The estimated ΔCH₄/ΔCO₂ and ΔCH₄/ΔCO ratios obtained through airborne measurement will be compared with those from bottom-up GHGs emission inventories such as Emissions Database for Global Atmospheric Research (EDGAR v6.0) to validate the accuracy of emission inventories.

In 2020, greenhouse gas (GHGs) emissions from the transportation sector in South Korea accounted for 14.7% of the total. Among various sectors encompassed within the transportation sector, road transport contributed about 96% to the emissions. Therefore, reducing carbon dioxide (CO₂) emissions from the road transport sector is essential for achieving national carbon neutrality. However, the current calculation of CO₂ emissions in the road transport sector is conducted only at the national level. To achieve more efficient emission reduction, it is necessary to calculate emissions on a regional and road-specific basis, considering the characteristics of each area and road. Therefore, as the initial phase of this study, a model for estimating CO₂ emissions from roads in Seoul was developed. This study aims to estimate CO₂ emissions from roads in Incheon, Busan, and Ulsan, using the Seoul model as a reference. To achieve this, the first step involved validating and improving the Seoul model. This process included utilizing feature importance analysis for data classification, addressing sample bias through clustering, and conducting model validation using independent datasets. In the next step, the estimation of road CO₂ emissions in Incheon, Busan, and Ulsan was conducted using the Seoul model. Subsequently, model construction specific to each of these cities was carried out. The reason for choosing this approach is to assess the general applicability of the model. In the future, independent data will be acquired and model improvement processes will be undertaken to validate the models for these cities. The result of this study is considered a crucial step in estimating regional and road-specific CO₂ emissions on a nationwide scale. Through this, it is anticipated that the findings will contribute to policymaking in the road transport sector and enable efficient reduction of CO₂ emissions.

Industrial complex areas can be large sources of air pollutants and greenhouse gases. It is very important to understand in detail the distribution of pollutants in areas with large emissions. In order to understand the detailed emission characteristics of an industrial complex, it is necessary to identify the contribution of emission sources at a detailed scale. Therefore, this study seeks to estimate the contribution of CO2 and BC emission sources based on machine learning techniques. To calculate contribution, five tree-based algorithmic models were used to select the model with the highest predictive performance. And the contribution of greenhouse gases was calculated using the SHAP (Shapley Additive exPlanations) technique. The input variables for machine learning are CO2, BC, U, V, road separation distance, land cover ratio, latitude, and longitude. The spatial distribution of CO2 (GMP252, 2-second intervals; H41-H6, 1-second intervals) and BC (MA200, 1-second intervals) were measured through mobile measurements based on bicycles and electric vehicles. Measurements were conducted 13 times over 4 days from August 21st to August 24th, 2023, and 20 times over 3 days from December 19th to 21st, 2023. CO2 and BC showed high concentrations mainly on the road, and seasonal differences were founded. The analyzed results showed the distribution of emission sources to CO2 and BC concentrations in industrial complexes and the priority of the emission sources which should be reduced first. Acknowledgements: This work was supported by “Korea Environment Industry & Technology Institute(KEITI) through Project for developing an observation-based GHG emissions geospatial information map, funded by Korea Ministry of Environment(MOE) (RS-2023-00232066).” and "This research was supported by Particulate Matter Management Specialized Graduate Program through the Korea Environmental Industry & Technology Institute(KEITI) funded by the Ministry of Environment(MOE)."

The increase in CO₂ in the atmosphere is the major concern worldwide due to its impacts on global warming. The government has announced the goal of net zero greenhouse gas emissions until the year of 2050. It is necessary to quantitatively understand the spatiotemporal and seasonal variations of CO₂ fluxes from the emission sources. In Korea, relevant researches have been utilized the eddy covariance technique by implementing gas analyzer at the top of towers to calculate upward and downward fluxes at a single point. To complement the limited applicability of eddy covariance method, in this study, we used drones to calculate and compare the seasonal CO₂ fluxes quantitatively in industrial and non-industrial complexes. The drone observations were conducted in the Sihwa industrial complex in Siheung, Korea, and the Kangwon National University athletic field in Chuncheon, Korea. The drone flights were taken for 3 hours after sunrise and 3 hours before sunset, in winter and summer, from 0m to 250m above ground. The estimation results showed the net CO₂ fluxes at the two different sites and seasons, revealing the emission characteristics monitored by drone observations. Acknowledgements: This work was supported by “Korea Environment Industry & Technology Institute(KEITI) through Project for developing an observation-based GHG emissions geospatial information map, funded by Korea Ministry of Environment(MOE) (RS-2023-00232066).” And "This research was supported by Particulate Matter Management Specialized Graduate Program through the Korea Environmental Industry & Technology Institute(KEITI) funded by the Ministry of Environment(MOE) (C10168180002)."

Urban CO₂ monitoring is crucial for climate change mitigation, as CO₂ is a major contributor to climate change and is primarily emitted in densely populated urban areas. Recently with the increase in ground-based Fourier transform spectrometer measurements and satellite data in urban area, the importance of understanding XCO₂ (Total Column-averaged dry-air mole fraction of carbon dioxide) data is also growing. However, in South Korea, there is little research on XCO₂ using ground-based Fourier transform spectrometer in urban area. In this research, we show the results of XCO₂ measurements from EM27/SUN ground-based Fourier transform spectrometer in the period of May 2020 to July 2023 over Seoul, South Korea. XCO₂ ­data exhibited an increasing trend of 2.16 ppm yr^-1, and a seasonal amplitude of 7.79 ppm. These results are then compared with the surface in-situ CO₂ data (SNUCO2M) measured near the EM27/SUN site. When compared with the trend and variability of surface in-situ CO₂, total column CO₂ showed smaller increases and variability, implying lower concentrations and changes in the upper atmosphere. This study also calculated and compared the footprints of column CO₂ and surface CO₂ using X-STILT (X-Stochastic Time-Inverted Lagrangian Transport model) and STILT model. We have discovered that the footprints of column CO₂ are significantly larger than those of surface in-situ CO₂, indicating that column CO₂ is influenced by broader area within urban regions. This work was supported by Korea Environmental Industry & Technology Institute (KEITI) through “Project for developing an observation-based GHG emissions geospatial information map”, funded by Korea Ministry of Environment (MOE)(RS-2023-00232066).

An atmospheric river (AR) is a strong filamentary water vapor transport that plays a critical role in regional hydroclimate systems. While climate conditions can affect wildfire activities, the process by which ARs are associated with wildfire patterns remains unclear. Here, we characterize ARs in 2016 and 2020, and associate them with fire spread and burned areas along with other climate conditions in the western U.S. We found the record-high wildfire activity in 2020 was associated with hotter, drier, and windier conditions, with its peak shifted from July to August, unlike the climatological fire seasonality in the western U.S. It was also linked to satellite-observed low soil moisture during pre- and on fire season but high vegetation greenness, a proxy of fuel load, during the pre-fire season. ARs were more frequent but weaker in the summer, while ARs were less frequent and short-lived in the fall of 2020 than those of 2016. The year 2016 experienced a 'coupled' precipitation-wind pattern (i.e. higher wind accompanying high precipitation). In contrast, precipitation was much lower in 2020 than in 2016, showing a 'decoupled' precipitation-wind pattern, particularly in the spring and fall. Under ARs, the contrasting precipitation-wind patterns in 2020 (dry-windy) and 2016 (wet-windy) were more evident. For example, the surface wind (precipitation) in the AR cases was higher by 9% (34%) than in the non-AR cases in 2020 (both years) (p < 0.01) over land. The daily fire activity records demonstrate that long-lived, successive, and coastal ocean-originated (centered) ARs with high precipitation help suppress fire activity (e.g., September-November 2016), while short-lived or no ARs with strong wind and little precipitation rather yield fire activity (e.g., August and September 2020). This result highlights how ARs can be associated with wildfire activity patterns during the pre-fire and fire seasons in the western U.S.

East Asian atmospheric rivers (ARs) play a critical role in the hydroclimate as they are closely associated with extreme precipitation. Although it is well known that ARs will become more frequent in a warming climate, the future changes of East Asian ARs and their impacts on precipitation in high-resolution climate models remain unclear. The present study investigates the East Asian ARs in the CMIP6 High Resolution Model Intercomparison Project (HighResMIP) models. The results show a robust increase in AR frequency and associated precipitation in the near future (2040-2050) under the SSP5-8.5 scenario. These increases are dominated by thermodynamic processes with a minor role of dynamic processes. Comparing low- and high-resolution model simulations, it is further found that AR-related precipitation is better captured in the latter. This result suggests that the high-resolution modeling would be beneficial for quantifying the hydrological impacts of East Asian ARs in a warming climate.

Atmospheric rivers (ARs) are closely related to local precipitation which can be both beneficial and destructive. Although several studies have evaluated their predictability, there is a lack of studies on East Asia ARs. This study evaluates the subseasonal prediction skill of East Asian ARs in the European Centre for Medium-Range Weather Forecasts (ECMWF) subseasonal-to-seasonal prediction model for 2001-2020 summer. The model successfully predicts the spatial distribution of AR frequency in East Asia, but underestimates it along the boundary of the western North Pacific subtropical high at lead times of one or two weeks. When the prediction skill is quantified by computing the anomaly correlation coefficient and mean square skill score, weekly AR prediction skill saturates around 2 weeks. Such prediction limit is primarily set by wind field errors with a minor contribution of moisture distribution errors. This result suggests that the improved prediction of atmospheric circulation field can improve the predictability of East Asian summer ARs and the associated precipitation.

Atmospheric rivers (ARs) are narrow, elongated, synoptic jets of water vapor that play important roles in the global water cycle and regional weather/hydrology. The need for characterizing and understanding AR life cycles (i.e., from genesis to termination) and their variations between different regions motivated the continual development of an AR tracking algorithm suitable for global studies, namely, Tracking Atmospheric Rivers Globally as Elongated Targets (tARget). The algorithm identifies AR objects at individual time steps based on thresholding integrated water vapor transport (IVT) and other requirements, and tracks each AR object in time and space. Building on three previous versions of the algorithm, this paper discusses further refinements to the algorithm since the last release in 2019, to better handle ARs in tropical and polar areas where the background climate is different from the midlatitudes, as well as “zonal” ARs (those with weak poleward IVT or even equatorward IVT) which the previous versions of the algorithm are not designed to capture. The further refined algorithm is applied to the ERA5 reanalysis over 1940–current at 6-hour intervals and a 0.25° × 0.25°horizontal resolution. The AR detection result is evaluated based on four analyses: (1) comparison of AR defining characteristics to those of Zhu and Newell (1998); (2) comparison of AR landfall dates in western North America, Britain, and East Antarctica with AR detection methods independently developed in other studies for those regions; (3) comparison of AR durations with ground observations of 91 AR events during 2004–2010 from an AR observatory in northern California; and (4) comparison of AR width and total IVT across the AR width with dropsonde observations of 21 ARs over the northeastern Pacific during 2005–2016.

Atmospheric rivers (ARs) are closely associated with extreme precipitation and hydrological events in East Asia to cause significant damages to human society. Therefore, predicting the impact of climate change on ARs is crucial for preventing damages caused by extreme precipitation and ensuring effective operation of water management facilities. Projections of future (2080~2099) changes in ARs and related hydrology under the SSP5-8.5 scenario utilizing the CMIP6 multi-model ensemble indicate that the annual-average integrated vapor transport (IVT) in East Asia in the period 2080~2099 will increase by approximately 32.5% compared to the past period (1995~2014) while the annual-average AR frequency will increase by about 111% of the past period. Examination of the water vapor and moist wind components of IVT reveals that the future increase in IVT is primarily due to increases in water vapor. It is inferred that the IVT increase is largely responsible for the increase the AR occurrence frequencies. The changes in AR due to climate change also affect precipitation to increase the total precipitation in East Asia. Examination of the changes in AR characteristics shows that the frequency of intense AR events will also increase due to the climate change. Specifically, significant increases in the frequency of strong AR events during the East Asian summer monsoon season are projected to occur in the future. Projections regarding the frequency and intensity of AR events vary substantially following regions, highlighting the necessity for more detailed regional analyses to further understand the impacts of the climate change on ARs and related hydrology in East Asia.

Length of day (LOD) is one of the Earth rotation parameters whose precise estimation and prediction are critical in applications such as interplanetary spacecraft tracking and navigation. LOD fluctuates on interannual or shorter time scales due to the solid Earth’s interaction with the atmospheric non-tidal phenomena such as El Niño–Southern Oscillation (ENSO) and Quasi-Biennial Oscillation (QBO). A recent study (Scaife et al. 2022) reported that LOD can be predicted more than a year ahead, though its result is based on only a single model. In this study, we examine LOD prediction skills in nine state-of-the-art decadal prediction systems that participated in the Decadal Climate Prediction Project (DCPP). LOD prediction skill remains significant for about a year in multiple prediction systems and multi-model mean (MMM), with a notable rebound at 12–15 lead months. This is consistent with the previous research and suggests that models in general can predict LOD a year ahead. LOD predicted by MMM is highly correlated with a linear combination of ENSO and QBO (r = 0.89) and with ENSO alone (r = 0.88), which is qualitatively consistent with observation (r = 0.55 and 0.46 for ENSO and QBO and ENSO alone, respectively). This indicates that LOD variation is well explained by ENSO and QBO, with ENSO playing a major role and QBO being a minor factor. Prediction skill of LOD at the time of rebound is also highly correlated with that of ENSO and QBO combined (r > 0.8), and with that of ENSO (r = 0.78), which further confirms a stronger link between LOD and ENSO. These results suggest that a skillful ENSO prediction may lead to an accurate prediction of LOD and related extra-tropical climate components up to a year ahead.

Recently, the international community has committed to achieving carbon neutrality by 2050 under the Paris Agreement, which aims to keep global average temperatures from rising more than 1.5°C/2°C above pre-industrial levels. By using Community Earth System Model Version 2 (CESM2), we examined the responses of Northern Hemisphere surface temperature under a net-zero emission scenario among ensemble members. Under a net-zero emission scenario, anthropogenic CO₂ emissions are increases linearly based on the SSP5-8.5 scenario, and then gradually reduced at the same rate until anthropogenic CO₂ emission reach a net-zero. After a net-zero emission was achieved, two pathways of Northern Hemisphere surface temperature change are observed after a certain period. By comparing with ensemble members with different pathways, we found that there exists a significant difference in North Atlantic Ocean, which could be associated with the changes of Atlantic Meridional Overturning Circulation (AMOC) intensity. Acknowledgements: This study was supported by the NRF grant funded by Korean government (NRF-2018R1A5A1024958). 

In this study, we use the Community Earth System Model version 2 (CESM2) ocean model hierarchy, in which the identical atmosphere, land, and sea-ice components are coupled with ocean models of varying complexity. In particular, we have developed and evaluated the Pencil Ocean Model (PenOM), a one-dimensional (1-D) ocean columns model with the lateral ocean processes disabled. The PenOM is implemented within the Parallel Ocean Program version 2 (POP2), which is the same model framework used for the full 3-D ocean component of CESM2. The PenOM provides a more realistic 1-D thermodynamic ocean compared to a slab ocean model by including multiple ocean layers and allowing a prognostic mixed layer depth evolution and vertical mixing, while is simpler than the fully 3-D ocean general circulation model. In a set of multi-century preindustrial control simulations using the CESM2 ocean model hierarchy, the internal variability of sea surface temperature (SST) in the absence of externally forced signals is analyzed. The results from the PenOM coupled simulation and the 3-D ocean case are compared to examine the respective roles of 1-D ocean processes (e.g., mixing, interannual mixed layer depth variations, and entrainment) and 3-D ocean dynamics (e.g., wind and buoyancy-driven circulation). The objective of this research is to discern the individual contributions of atmospheric forcing, 1-D ocean processes, and 3-D ocean dynamics to the variability and predictability of SST on interannual-to-decadal time scales. In addition, we aim to assess the impact of ocean dynamics on the atmospheric circulation.

The Atlantic Multidecadal Variability (AMV) with dominant timescales of 55 to 70 years serves as a crucial source of predictability for climate variability and changes in many regions. This influence extends not only to the continents surrounding the North Atlantic but also remotely through global atmospheric teleconnection. Although considerable relevant studies have been conducted, challenges remain in better identifying the origins of its variability and predictability and improving predictions of AMV and its teleconnection at longer leads. This study aims to determine the origins of AMV variability and to assess the predictability and long-lead forecast skills of AMV and its teleconnection by using a multi-year Earth System Prediction system based on Community Earth System Model Version 2 (CESM2). The prediction system consists of 50-member uninitialized simulations, 30-member ocean assimilations, and 30-member 5-year hindcasts initiated from every January 1^st from 1960 to 2021. Comparison between the mean of large ensemble simulations and observation indicates that external forcings explain about 30% of the observed nonfiltered AMV variability for the last 62 years. We estimate that the ocean internal processes contribute up to 25% (10%) to AMV predictability at a 1-year (5-year) forecast lead. We also investigate prediction skills for regional temperature and precipitation anomalies associated with AMV variability.

The demand for decision-relevant and evidence-based near-term climate information is increasing. This includes understanding and explaining the variability and changes in ecosystems to support disaster management and adaptation choices. As climate prediction from seasonal to decadal (S2D) expands to encompass Earth system dimensions, including terrestrial and marine ecosystems, it is crucial to deepen our scientific understanding of the long-term predictability sources for ecosystem variability and change. Here we explore to what extent terrestrial ecosystem variables are driven by large-scale - potentially predictable -climate modes of variability and external forcings or whether regional random environmental factors are dominant. To address these issues, we utilize a multi-year prediction system based on Community Earth System Model version 2 (CESM2). The system consists of 50-member uninitialized historical simulations, 20-member ocean assimilations, and 20-member hindcast initiated from every January 1^st integrating for 5 years from 1961 to 2021. The key variables assessed are surface temperature, precipitation, soil moisture, wildfire occurrence, and Gross Primary Productivity. Our results suggest that land surface processes and ecosystem variables over many parts of the globe can be potentially predictable 1 to 3 years ahead originating from anthropogenic forced signals and modes of climate variability, particularly El Nino and Southern Oscillation and Atlantic Multi-decadal variability. These global modes of climate variability shift regional temperature and precipitation patterns, leading to changes in soil moisture, wildfire occurrence, and terrestrial productivity.

Over the past few decades, the Indian Ocean has exhibited the most significant warming trends globally, imparting profound regional and global consequences. However, the specific ramifications of this warming, particularly concerning terrestrial and agricultural productivity and their impact on terrestrial carbon sinks, have remained unclear. In this study, we unveil a compelling link between Tropical Indian Ocean (TIO) warming and the reduction of terrestrial and agricultural productivity in North America, with the potential to contribute to the decline of carbon sinks. TIO warming initiates an atmospheric teleconnection process, characterized by the propagation of quasi-stationary waves into the extratropics. This process culminates in a widespread deficit of precipitation and soil moisture over North America. Subsequently, agricultural crop yields and terrestrial productivity, accounting for 10-20% of Gross Primary Productivity (GPP), are significantly impacted, thereby contributing to a positive carbon-climate feedback loop. Through simulations replicating TIO warming, we validate the suppression of terrestrial activities in North America under both historical and projected CO2 forcings. This study underscores the pivotal role played by TIO warming in shaping the global terrestrial ecosystem.

Recent deep learning-based models have shown great potential in predicting and interpreting weather and climate phenomena. For example, the data-driven weather forecast models (e.g., GraphCast and Pangu-Weather) showed better forecast skills of the global atmosphere within a week than the operational numerical weather prediction. However, with their short history, understanding of trained physical processes in the deep learning-based model is not satisfactory. Motivated by the above, this study aims to improve the accuracy and capability of data-driven global climate prediction. This study uses a deep learning-based model trained with daily-mean atmospheric variables in the European Centre for Medium-Range Weather Forecasts (ECMWF) reanalysis (ERA5) at 250-km horizontal resolution for 1979-2015. A method to better structuralize the horizontal structure of the global atmosphere into the deep learning model shows great potential to improve data-driven prediction skills, which exhibit reliable deterministic forecast skills within a week. The inference results also exhibit realistic subseasonal climate variability, such as eastward propagation of the Madden-Julian Oscillation (MJO). Additionally, we propose an idealized model to understand the remote influence of atmospheric perturbation from the pre-trained model. The idealized model experiments forced by El Nino- and MJO-like perturbation exhibit realistic atmospheric teleconnection, indicating that the realistic physical processes are adequately trained in the data-driven method. The results of this study shed light on the necessary processes in the model architecture for state-of-the-art global climate prediction.

Climate models often overestimate the precipitation in western Tibetan Plateau (TP) during winter, due to their poor ability in representing the orographic drag of unresolved complex terrain. In this study, the parameterization scheme of orographic gravity wave drag (OGWD) in the Model for Prediction Across Scales (MPAS) is revised to account for the nonhydrostatic effects (NHE) on the surface momentum flux of vertically-propagating orographic gravity waves. The effect of revised OGWD scheme on the simulation of winter precipitation over the TP is examined using parallel numerical experiments with the original and revised schemes, respectively. The results show that the revised scheme can effectively alleviate the precipitation biases in western TP by improving the atmospheric circulation and water vapor transport. The NHE reduces the surface wave momentum flux of orographic gravity waves which results in weaker zonal OGWD in the mid-low troposphere and thus stronger westerlies over the TP, reducing the easterly biases in the experiment with the original OGWD scheme. The weakened zonal OGWD promotes a plateau-scale cyclonic circulation to the north of the TP, which suppresses the northern branch of the bifurcated westerlies detouring the TP. According to quantitative analysis, the water vapor transport into the western TP and surrounding areas by upstream westerlies is notably reduced by the northeasterlies of the cyclonic circulation, which eventually leads to the decrease of precipitation in this region.

Terrestrial carbon fluxes are crucial to the global carbon cycle. Quantification of terrestrial carbon fluxes over the Tibetan Plateau (TP) has considerable uncertainties due to the unique ecosystem and climate and scarce flux observations. This study evaluated our recent improvement of terrestrial flux parameterization in the weather research and forecasting model coupled with the vegetation photosynthesis and respiration model (WRF-VPRM) in terms of reproducing observed net ecosystem exchange (NEE), gross ecosystem exchange (GEE), and ecosystem respiration (ER) over the TP. The improvement of VPRM relative to the officially released version considers the impact of water stress on terrestrial fluxes, making it superior to the officially released model due to its reductions in bias, root mean square error (RMSE), and ratio of standard deviation (RSD) of NEE to 0.850 µmol·m⁻² ·s ⁻¹, 0.315 µmol·m⁻² ·s ⁻¹, and 0.001, respectively. The improved VPRM also affects GEE simulation, increasing its RSD to 0.467 and decreasing its bias and RMSE by 1.175 and 0.324 µmol·m⁻² ·s ⁻¹, respectively. Furthermore, bias and RMSE for ER were lowered to −0.417 and 0.954 µmol·m⁻² ·s ⁻¹, with a corresponding increase in RSD by 0.6. The improved WRF-VPRM simulation indicates that eastward winds drive the transfer of lower CO2 concentrations from the eastern to the central and western TP and the influx of low-concentration CO2 inhibits biospheric CO2 uptake. The use of an improved WRF-VPRM in this study helps to reduce errors, improve our understanding of the role of carbon flux cycle over the TP, and ultimately reduce uncertainty in the carbon flux budget.

The top-hat approximation, which is widely used with bulk entraining plume in the mass-flux type convective parameterization schemes, is verified at different horizontal scales by using large eddy model (LES) and cloud resolving model (CRM) simulations, especially those at the limit of vanishing cloud fraction. Five convection cases consisting of three shallow convection cases: RICO, BOMEX, and ATEX from the Global Energy and Water Cycle Experiment (GEWEX) Cloud System Study (GCSS) programs and two deep convection cases including GATE and KWAJEX are conducted. Due to the fact that convective cloudiness fraction increases with increasing horizontal resolution, the cloudy part of turbulent flux increases, leading to considerable errors manifested in top-hat approximation. These errors, however, are largely compensated by the decrease of error in the environmental part, resulting in the relative error in top-hat approximation not increasing significantly with increasing horizontal resolution. At first glance, this brings encouraging results to the modeling community that top-hat approximation is still valid at high resolutions. Nevertheless, increase of error in the cloudy part poses additional difficulties for turbulence schemes, which often struggle to represent turbulence in cloudy part. A multi-plume based scheme is needed instead of a bulk entraining model in order to partially account for the inhomogeneity in cloudy part. Preliminary tests show that three plumes are sufficient to well parameterize the turbulent flux in the cloudy part.

Marine stratocumulus plays a crucial role in regulating Earth’s radiation budget by exerting a strong negative radiative cooling effect. Many state-of-the-art climate models have struggled to accurately simulate the marine stratocumulus features, leading to substantial shortwave radiation biases at both the surface and the top of the atmosphere. These biases may be attributed to deficiencies in physical parameterizations, though the exact causes need to be identified and addressed by individual models. This study aims to address the marine stratocumulus biases in Taiwan Earth System Model Version 1 (TaiESM1) through a two-fold approach: (1) identifying the sources of these biases, and (2) modifying relevant physical parameterization schemes, such as boundary layer and cloud schemes, to resolve these biases. Specifically, this study will utilize DYCOMS-II and VOCALS-REx field observation data, ERA5 reanalysis, and a hierarchy of TaiESM1 simulations including single-column model simulations, hindcast simulations, and AMIP-type and fully coupled simulations. This hierarchy of simulations enables a comprehensive examination of the physical parameterizations in TaiESM1, facilitating the identification and resolution of marine stratocumulus biases. Furthermore, the proposed methodology can be a valuable framework for other modeling centers seeking to address biases in their models. 

The parameterization of orographic gravity wave drag (OGWD) is essential for the simulation of the atmospheric circulation, especially in regions of complex terrain. Current OGWD schemes assume hydrostatic OGWs, but the parameterized OGWs in fine-resolution models with narrow subgrid-scale orography can be significantly affected by nonhydrostatic effects (NHE). This study revises the OGWD scheme in the Weather Research and Forecasting model by accounting for NHE on the surface wave momentum flux of upward-propagating OGWs. To evaluate the performance of the revised OGWD scheme in short-range forecast, two sets of simulations are conducted for nine Northeast China cold vortices (NECVs) in the warm season of 2011 using the original and revised OGWD schemes, respectively. Results show that the model tends to underestimate the NECV intensity, producing too high geopotential height and too weak horizontal wind. The NECV intensity biases are alleviated when accounting for NHE in the OGWD scheme, especially in the mid-to-lower troposphere. Process analyses reveal that NHE act to reduce the lower-tropospheric OGWD by reducing the surface wave momentum flux of OGWs, which strengthens the NECV in the lower troposphere. The NECV intensifies in the mid-upper troposphere as well, owing to the enhancement of post-trough cold advection caused by the strengthened low-level cyclonic circulation. NHE increase as the model resolution changes from 27 km to 3 km, suggesting greater importance of accounting for NHE in OGWD parameterization in high-resolution models.

On May 3, 2017, an intense rainfall event that resulted in heavy flooding occurred in several areas over Davao City, Philippines. According to Lagare et al. (2023), mesoscale convective systems (MCS), due to weak large-scale forcing and different mechanisms, were observed over Davao City during this date. This paper evaluates the impact of urban canopy models (UCM) on a heavy rainfall event over Davao City using the coupled Weather Research and Forecasting (WRF)-Urban modeling system. The WRF’s default urban surface model (BULK Urban Parameterization), Single-Layer Urban Canopy Model (SLUM), and Multi-layer Building Energy Parameterization (BEP) were used to investigate the impact of urbanization to the formation or movement of the mesoscale convective systems. BULK treats the urban canopy as a single rough surface, SLUM takes into account different surface types such as roofs, walls, and streets, while BEP consider individual building features and urban surfaces as it generates the vertical flow of heat, moisture, and momentum. WRF was initially run with 9-, 3- and 1-km horizontal resolution using the 100-m Copernicus Global Land Service Land Cover that has more Urban Land Surface categories than the 1-km MODIS land cover data. UCM sensitivity runs were performed by downscaling the 1-km WRF run to 500-m grid. The multilayer urban canopy parameterization simulated lower biases than the BULK scheme. Significant differences in rainfall and MCS formation were found between the different urban canopy models. Specifically, MCS1 and MCS2 formations and their merging into MCS3 over downtown Davao were more evident in BEP compared to Bulk. MCS3 over the urban area was not observed in SLUM simulation. The multilayer scheme also weakened wind speed and delays moisture transport over the downtown area. Further analysis will be done to tune the model with different stages of Davao City’s urbanization.

Some extreme meteorological events, such as heat wave, atmospheric stagnation and temperature inversion, may have an important impact on air quality through physical and chemical processes, and have a great impact on regional air quality(Han et al., 2014; Lei et al., 2018; Liao et al., 2018; Steiner et al., 2010). When the temperature inversion occurs, the atmospheric boundary layer is stable, which inhibits the transfer of matter and energy in the boundary layer and is not conducive to the diffusion of pollutants. Atmospheric stagnation tends to have a weak wind speed in the boundary layer and a small precipitation, which are not conducive to the diffusion and washout of pollutants. Heat waves are characterized by persistent high temperatures, which have a significant impact on ozone production. Studies show that heat wave days have a greater impact on ozone and PM2.5 concentrations; atmospheric stagnation has a greater impact on PM2.5, and temperature inversion will slightly increase ozone concentrations but have a greater impact on PM2.5.（Yang, J., & Shao, M., 2021).In general, when heatwaves, atmospheric stagnation or temperature inversions are present alone or in combination, ozone and particulate matter concentrations are usually higher in most areas than when there is no extreme weather. Studies have shown that the frequency of extreme meteorological events in China has been on the rise in the past decade(Ding et al.,2016). Therefore, continuing to study the relationship between extreme meteorological events and the pollution of ozone and particulate matter in different regions of China will be of great help to the prevention and control of air pollution in different regions.

Severe PM₁₀ events in South Korea are caused by stable atmospheric circulation conditions related to high-pressure anomalies in the upper troposphere. However, research on why these atmospheric circulation patterns occur is unknown. In this study, we propose new large-scale teleconnection pathways that cause severe PM₁₀ events during the mid-winter in South Korea. This study investigated extremely high (EH)-PM₁₀ instances in South Korea during mid-winter and examined the corresponding atmospheric teleconnection patterns to identify the factors contributing to EH-PM₁₀ events. K-means clustering analysis revealed that EH-PM₁₀ cases were associated with two large-scale teleconnection patterns originating from the North Atlantic Ocean and the Barents-Kara Sea. Atmospheric variability in two key regions trigger the Rossby wave propagation, causing favorable atmospheric conditions for EH-PM₁₀ events in Korea. These large-scale teleconnections led to a high-pressure anomaly over the Korean Peninsula, reducing atmospheric ventilation and weakening the surface pressure system, causing a rapid increase in PM₁₀ concentration within a few days. Understanding this teleconnection phenomenon may assist in implementing emission reduction measures based on the results of short-term forecasts of severe PM₁₀ events. [This work was supported by a grant from the National Institute of Environment Research (NIER), funded by the Ministry of Environment (MOE) of the Republic of Korea (NIER-2021-03-03-007) and the Korea Polar Research Institute (KOPRI) grant titled “Development and Application of the Earth System Model-based Korea Polar Prediction System (KPOPS-Earth) for the Arctic and Midlatitude High-impact Weather Events” funded by the Ministry of Oceans and Fisheries (KOPRI PE23010)].

Previous studies have noticed that the Coupled Model Intercomparison Project Phase 6 (CMIP6) models with a stronger cooling from aerosol-cloud interactions (ACI) also have an enhanced warming from positive cloud feedback, and these two opposing effects are counter-balanced in simulations of the historical period. However, reasons for this anti-correlation are less explored. In this study, we perturb the cloud ice microphysical processes to obtain cloud liquid of varying amounts in two Earth System Models (ESMs). We find that the model simulations with a larger liquid water path (LWP) tend to have a stronger cooling from ACI and a stronger positive cloud feedback. More liquid clouds in the mean-state present more opportunities for anthropogenic aerosol perturbations and also weaken the negative cloud feedback at middle to high latitudes. This work, from a cloud state perspective, emphasizes the influence of the mean-state LWP on effective radiative forcing due to ACI (ERFACI).

The dust direct radiative effect (DRE) depends sensitively on the particle size distribution (PSD) and complex refractive index (CRI). Although recent studies constrained the PSD in the models, CRI uncertainties are still large. Therefore, whether dust warms or cools the climate system remains unclear. Here, we jointly analyze the dust PSD and CRI to constrain the dust DRE by leveraging measurements of dust CRI globally. We find that employing regionally dependent CRI significantly enhances the scattering of dust in shortwave and reduces its absorption in longwave, which is opposite to that caused by increasing the coarse dust fraction via constraining the PSD. Constraining both PSD and CRI ultimately leads to a net DRE of -0.71 W m^-2, a cooling stronger than current model estimates. It is crucial to include PSD and regionally varying CRI more realistically to accurately represent the impact of dust on the climate system.

Wildfires emit large amounts of black carbon and light-absorbing organic carbon, known as brown carbon, into the atmosphere. These particles perturb Earth’s radiation budget through the absorption of incoming shortwave radiation. Given that wildfires are predicted to increase globally in the coming decades, it is important to quantify these radiative impacts. Compared to black carbon, the atmospheric warming effect exerted by brown carbon remains highly variable and poorly represented in climate models. Recently, observational studies found a type of dark brown carbon (i.e., tar balls) generated from wildfires can contribute more than black carbon (nearly three-quarters) to the short visible light absorption. Those indicate that the climate effects of brown carbon may be larger than ever imagined in the future, while the parameterizations of tar balls from wildfires have not been considered in the climate models. And the climate effects of brown carbon under the scenario of future wildfires have not been explored. In this study, we first added parameterizations of tar balls' light absorption to the Modal Aerosol Module (MAM4) in CAM5.4 to represent the light absorption of brown carbon from wildfires. We then collected multi-type observational data (vertical and surface) from several wildfire events to validate the model simulation. Based on this, we finally estimated the direct radiative effects of tar balls and analyzed their effects on clouds and precipitation under the scenario of future wildfires.

Subsidence often occurs on clear nights, which affects the development of the nocturnal boundary layer (NBL) and the processes of material and energy transfer. Subsidence in the boundary layer appears to be rare and short-lived, so it is difficult to be observed. In this paper, data from a large tethered balloon (1900 m³) with instruments for measuring meteorological parameters and air pollutants, a ground-based aerosol lidar, a Doppler wind lidar and some other ground observations are combined to analyze the influence of a subsidence process on the boundary layer structure and downward transport of various pollutants. Pollutants were horizontally transported to the observation site driven by southwest low-level jet. Under the downdraft and turbulent kinetic energy transported towards the surface, the vertical subsidence of the pollutants lasted for nearly 5 hours. During this period, the observation data of the tethered balloon hovering at 500 m shows that the concentration of PM2.5, NO³_-, SO₄^2- and NH₄⁺ increased by approximately 17 μg m^-3, 11 μg m^-3, 2.3 μg m^-3 and 5 μg m^-3, respectively. The vertical subsidence of pollutants had significant effects on the secondary inorganic aerosol concentrations, but the concentration of organic aerosols was not affected. In addition, the O3 concentration in the lower atmosphere increased by 8 ppb. The downward transport of pollutants obviously lagged behind the downward vertical motions by about one hour. The subsidence process also enhanced the inversion intensity of the lower atmosphere. The inversion layer effectively hindered the diffusion of pollutants, and the concentration of ground PM2.5 increased by 42 μg m^-3. Our results show that under the effects of descending movements and the downward vertical flux of turbulent kinetic energy, pollutants could also subsided into the lower layer of the NBL.

In stable atmospheric boundary layers, turbulence can be categorized into two regimes: a moderate turbulence regime where turbulence is relatively prompted and a weak turbulence regime where turbulence is suppressed. Monin-Obukhov similarity theory, which is the foundation of the modern micrometeorology, is only valid under the moderate turbulence regime. Therefore, understanding the transition between the two turbulence regimes is crucial for the development of stable atmospheric boundary layer theories and parameterizations. Currently, however, there is no physically based, non-dimensional index for characterizing the transition between the two turbulence regimes. This study aims to bridge this research gap by proposing a new non-dimensional index and determining the transition threshold value theoretically. Using CASES-99 dataset, this study reveals that the transition between the moderate and the weak turbulence regimes is related to the establishment of pancake vortices, the vertical length scale of which is comparable to the buoyancy length scale. In order to obtain a non-dimensional index characterizing this process, a simplified spectral model is established. By incorporating horizontal and vertical length scales of turbulence into the spectral model, this study finds that the transition occurs at a critical horizontal Froude number of 0.28. Furthermore, the critical horizontal Froude number can recover previously used height- and site-dependent mean wind speed thresholds by invoking the stability-corrected logarithmic wind profile. These findings hold significant scientific and practical implications for the representation of stably stratified turbulence in numerical simulations for weather and pollutants dispersion. The identified critical horizontal Froude number can be used to constrain the application range of Monin-Obukhov similarity theory and further improve turbulence parameterizations in stable atmospheric boundary layers.

This study presents a novel approach in measuring the atmospheric boundary layer height (ABLH) by utilizing data from a very high frequency (VHF) radar installed at the Advanced Centre for Atmospheric Radar Research in Cochin University of Science and Technology (CUSAT) in India. To ensure the result accuracy, clear-sky days were identified through a dedicated year-long experimental campaign in 2021 involving simultaneous radar and radiosonde operations. The mixing ratio and associated gradient profiles from radiosonde are employed to estimate the reference ABLH values. The new method combines three parameters, viz., signal-to-noise ratio, wind speed and wind direction - in estimating the boundary layer height from radar datasets. The ABLH is determined as the altitude at which the sum of normalized standard deviations of the three parameters peak and then markedly decrease. The results demonstrate a significant correlation (r=0.91 at a 99% confidence level) between the ABLH derived from this methodology and that estimated from the radiosonde data. In contrast to sparsely available radiosonde profiles, this research provides extensive opportunities to investigate the high temporal variability such as the diurnal cycle of ABLH using radar data. Nevertheless, there are certain hurdles in the estimation of ABLH during rainy/cloudy conditions which is yet to be overcome.

In this study, the Extreme Hourly Precipitation Areas (EHPAs) of three extreme levels (i.e., between the 95^th and 99^th percentiles, between the 99^th and 99.9^th percentiles, beyond the 99.9^th percentile) in the Pearl River Delta over South China are identified; then the Extreme Hourly Precipitation Event (EHPE) and associated Convective Cores (CCs) are tracked, and their macro-and-microphysical characteristics are analyzed using multi-year dual-polarization radar observations. Results show that > 90% of EHPAs are smaller than 10 km², and 65 – 75% of EHPEs last only one hour. They tend to be more localized and persist longer with increasing hourly-precipitation extremity. The EHPEs overlap with the CCs during 50 – 64% of the EHPEs’ life span. Their occurrence frequencies are nearly quadrupled after the monsoon onset over South China Sea, with a major (secondary) peak at about 1400 LST (0600 LST) in the diurnal variations. The CCs are non-linear shaped with about 65% being meso-γ-scale and embedded within mostly meso-β or α-scale 20dBZ regions. The CCs generally contain active warm-rain processes and about 70% possess moderate-to-intense mixed-phase microphysical processes. The ratios of ice water path to liquid water path are about 0.37, and coalescence dominates (about 68%) the liquid-phase processes. The average size of raindrop is slightly larger than the “maritime-like” regime and the average concentration is much higher than the “continental-like” regime. These CCs’ characteristics mostly resemble those of the convection producing extreme instantaneous precipitation, except for a larger horizontal scale and less evident changes with the hourly-precipitation extremity.

How anthropogenic aerosols influence extreme precipitation is an open question. On May 7, 2017, a record-breaking rainfall event hit the capital city Guangzhou in South China with a maximum hourly rainfall of 184.4 mm and a maximum 3-hour rainfall accumulation of 382.6 mm, causing serious societal impacts. The observations of PM2.5 indicate increases of aerosol concentration on May 6, 2017 resulting in polluted conditions under which the rainstorms initiated and developed. To investigate the possible contribution of anthropogenic air pollution to this high-impact rainfall event, this study utilizes the Chemistry version of the Weather Research and Forecasting model (WRF-Chem), coupled with spectral-bin microphysics (SBM) and a multilayer urban model embedded with a building energy model (BEM-BEP). The control experiment (CTRL) with realistic anthropogenic emissions can reproduce the diurnal variations of PM2.5 and its surge to peak amount before the convection initiation as found in the observations. The locations, structures and evolutions of convective storms and associated surface rainfall accumulation are also simulated reasonably well compared with the observations. In contrast, the “clean scenario” experiment (CLEAN) with anthropogenic emissions reduced to 10% of CTRL doesn’t capture the observed spatiotemporal distributions of PM2.5, e.g., the near-ground aerosol number concentration at the moment of convection initiation is only about 1/5 of that in the CTRL. A much smaller amount of maximum rainfall accumulation in CLEAN than CTRL (235 vs. 339 mm) is associated with the weaker convective intensity and shorter duration of the rainstorm during the later stage of the rainfall event. We are making ensemble simulations and in-depth analysis to obtain more robust results about urbanization-induced aerosol impacts on the rainfall.

This study analyzes the effect of the Australian High (AH) on the seasonal phase locking of Indian Ocean Dipole (IOD) events. The anomalous strong AH associated with the positive phase of the Antarctic Oscillation can cause significant easterly wind anomalies and northward cross-equatorial flow over the western Maritime Continent (MC) by strengthening the Australian winter monsoon during May–August. The AH-associated easterly anomalies and northward cross-equatorial flow can create thermodynamic air-sea feedback and contribute to a significant cooling anomaly in the western MC and the tropical eastern Indian Ocean. Without considering the effect of ENSO, these processes contribute to the occurrence of positive IOD events, which begin in early summer, peak in late summer, and decay rapidly thereafter. The effect of ENSO can extend the peak period of IOD into the boreal autumn of that year. An anomalous weak AH corresponds to the occurrence and seasonal phase locking of negative IOD events. Through combined empirical orthogonal function analysis, we find that the effect of the AH can well explain the seasonal phase locking of 34 IOD events (40 in total), which provides an important theoretical basis for the prediction of IOD events.

Although the trend of sea-ice extent under global warming has been studied extensively in recent years, most climate models have failed to capture the recent rapid change in the Arctic environment, which has brought into question the reliability of climate model projections of sea ice and suggested a potential shift in Arctic climate dynamics. Here, based on the results of a time-variant emergent constraint method with a weighting scheme, we show that an ice-free Arctic might occur earlier (by at least 5 ~ 10 years) than previously estimated. In other words, Arctic ice will likely disappear before the 2050 s. The observationally constrained date for an ice-free Arctic in September under fossil-fuel-based development (i.e., Shared Socioeconomic Pathway (SSP) 5–8.5) scenarios yields a central estimate of 2050–2054 with a 66% confidence range (equivalent to the IPCC’s ‘likely’ range) of 2037–2066, while an ice-free Arctic will likely occur for another 20 years and 11 years under ambitious mitigation scenarios (i.e., SSP2-4.5) and SSP3-7.0. An ice-free Arctic is unlikely to occur under the sustainable development scenario (i.e., SSP1-2.6). Looking forward, this time-variant emergent constraint may also help detect tipping points in the climate system. Our findings provide useful information to help policy makers cope with climate change.

To study the natural differences between precipitation events with various intensities, it is necessary to focus on the features of their corresponding rain cells, where a rain cell is usually regarded as a continuous precipitation entity in which the rain rate (or radar reflectivity) is higher than surrounding. Thus, the two-dimensional data, Near Surface precipitation Rate (NSR), provided by the Dual-frequency Precipitation Radar (DPR) onboard the GPM satellite with a horizontal resolution of 5 km, can be used for identifying these rain cells. However, the NSR distribution is usually complex with rain cells often adhesive to each other in a larger precipitation system, bringing difficulties for distinguishing them. Also, the large number of rain cells makes manual identification impractical. Hence, an automatic rain-cell-identification algorithm using NSR is proposed in this study. Firstly, identify the NSR peak pixels, which are the pixels with the local maximum NSR value in an area of continuous precipitation with a minimum NSR value set up. Then, assuming each peak corresponds to a rain cell, the scope of the rain cell is determined by expanding from the peak when following two requirements are met: the NSR value of a pixel is larger than some percentage (e.g., 50%) of the corresponding peak NSR, and the NSR value tends to decrease approximately as the range from the peak increases. Results show that the proposed algorithm is capable of distinguishing adjacent rain cells by reconciling the completeness and separation of these rain cells, which are more reasonable than those from the algorithms simply relying on NSR value thresholds and pixel continuity. Therefore, this proposed method can help further studies on more accurate and reliable precipitation features of rain cells and their connection with larger systems as well as their precipitation mechanisms.

Weather plays a vital role in human society. Having advanced access to meteorological information thus can enhance the ability for decision-making and disaster preparedness. The Numerical Weather Prediction (NWP) model was developed due to the increasing demand for accurately predicting weather changes for the next few hours or days. However, the traditional physical-based model requires extensive computational time and costs, which is not an optimal solution in rapidly changing weather conditions. Recently, there has been an increasing interest in using machine learning models as an alternative to NWP models. Current research mainly focuses on two categories based on the types of datasets used: (1) “Reanalysis data” for predicting various meteorological parameters at different altitudes globally; and (2) “Surface observation data” for predicting specific meteorological parameters at individual stations. Research by NVIDIA, Huawei, and Google has shown that the application of reanalysis data with machine learning techniques in weather forecasting is becoming more efficient and accurate than traditional physical-based models. However, reanalysis data still rely on NWP models for preliminary data processing, resulting in high computational costs. As for surface observation data with machine learning techniques, common research focuses on the prediction of temperature and precipitation. These have shown commendable performance in many regions, but heavy rainfall predictions are still challenging due to the imbalanced non-Gaussian data distribution. Therefore, this study aims to utilize surface observation data to enhance the accuracy of machine learning predictions, especially in forecasting extreme precipitation with highly uneven rainfall distribution in Taiwan.

Southeast Asia experiences vigorous convection patterns due to solar heating, leading to extreme weather events. Thunderstorms occurring in these areas exhibit a diurnal variation, emphasizing the importance of considering diurnal variation to detect thunderstorms. This study suggests a novel method to detect thunderstorms considering the diurnal variation of convection using the geostationary satellite GEO-KOMPSAT 2A (GK2A). We utilize 10-minute GK2A infrared (IR) brightness temperature (BT) and cloud microphysical parameters (e.g. cloud optical thickness, cloud effective radius) to detect thunderstorms effectively. The storm detection results were validated with Global Precipitation Measurement (GPM) Integrated Multi-Satellite Retrievals (IMERG) IR precipitation variables. As result, the thunderstorm detection was further improved when cloud microphysical parameters were applied with BT. This study proposes an optimized threshold of cloud microphysical parameters over time to minimize the false alarm ratio value also maximize the probability of detection and critical success index value. This study will contribute to reducing damage due to thunderstorms in Southeast Asia. 

Recent studies have shown an increase in frequency and intensity of heavy rainfall (Hev_Ran) events on the Korean Peninsula, located East Asia, due to global warming. Additionally, several studies have shown that the atmospheric environment and MCS (Mesoscale Convective Systems) causing Hev_Ran in East Asia differs from those in North America and Europe. Therefore, this study aims to reevaluate the possibility of detecting MCS causing Hev_Ran in Korea using instability indices (Inst_Ind) derived from rawinsonde data. Considering the regional, seasonal, and temporal variations of Hev_Ran events in Korea, this study also conducts investigation on the detection capability by Inst_Ind, region, and time as well as threshold optimization. For the recent ten years during the rainy season, hourly accumulated precipitation data from the AWS and rawinsonde data from 8 stations in Korea were used for this purpose. Due to the differing observation frequencies of the two datasets, this study defines the collocated data as those AWS data within -2h~+2h temporally and 100km spatially based on rawinsonde observations. Comparing the Inst_Ind during climate average (Cli_Ave) and Hev_Ran, significant differences were noted for KI, SWEAT, and TPW with more than 20% differences. However, CAPE, LI, SSI, SRH, and TTI did not show significant differences. POD and FAR were used to reevaluate various Hev_Ran intensity (30, 40, and 50mm/h) detection level of the Inst_Ind. Analysis has indicated usefulness in detecting Hev_Ran using SSI, KI, and TPW showing high POD (0.92~0.98) and FAR (0.91~0.99). However, CAPE, LI, SWEAT, SRH indicated less effectiveness regardless of Hev_Ran intensity, showing low POD (0.32~0.48) and high FAR (0.90~0.98).This presentation will further detail the optimization of thresholds for Inst_Ind and provide detailed presentation of the detection capability of Hev_Ran systems and MCS, segmented by station and time.

This research focuses on the crucial role of sea ice balance in the Earth's climate system, highlighting its significance in global climate change. Among the factors affecting this balance, ocean heat flux is key due to its impact on sea ice dynamics. Ocean heat flux, the transfer of thermal energy between ocean and ice, is essential in sea ice's freezing and melting processes. Our research reveals that increases in ocean heat flux are directly correlated with heightened rates of sea ice melting. This correlation is particularly pronounced in polar regions, where even minor rises in ocean temperatures can significantly disrupt the fragile equilibrium necessary for sea ice formation. Inversely, the study also identifies periods when a decrease in heat flux plays a vital role in the stabilization and augmentation of sea ice. Previous studies have indicated that direct measurement of ocean heat flux is not feasible, and approximations derived from sea ice buoy data are often imprecise due to the presence of noise. Addressing this challenge, our study employs observational data to determine boundary temperatures (at the top and bottom positions of ice). These temperatures are then analyzed using a Fourier series to predict ocean heat flux, effectively eliminating noise components and emphasizing principal wave terms. The methodology encompasses a detailed examination of both temperature variations and changes in sea ice, aiming to establish a comprehensive understanding of the heat transfer mechanisms at play. In conclusion, our findings assert the criticality of sea ice in forecasting and managing the future state of the Earth's climate. The research emphasizes the need for specific and empirically derived values of ocean heat flux, moving away from arbitrary assumptions, to enhance the accuracy of climate predictions and inform effective environmental strategies.

This study suggests a possible mechanism of how the Arctic sea-ice loss can influence the mid-latitude climate in the Eurasian continent. It is shown that the low sea-ice concentration over the Barents-Kara-Laptev Seas in autumn typically leads to cold Eurasian in winters. It is demonstrated that the Arctic-to-midlatitude connection depends on the state of late autumn atmospheric circulation. When the autumn sea-ice reduction is accompanied by anticyclonic circulation over northern Eurasia, Eurasia becomes anomalously cold in the early winter. However, when cyclonic circulation is dominant, Eurasian cold anomalies appear in the late winter. This seasonally-delayed response is further found to be related to the wind-driven sea-ice drift that causes warm anomalies over the Barents-Kara Seas in the following winter. These observational results are confirmed by model simulations, indicating that the recent Eurasian cold winters could be linked to their forced response to the Arctic sea-ice loss.

This study presents an analysis of the future of the Warm Arctic-Cold Continent (WACC) phenomenon in East Asia and North America. Utilizing large-ensemble datas, we project the WACC events for the upcoming decades, focusing on the evolving geographic boundaries of extreme cold temperatures. Our findings indicate a notable northward shift in these boundaries, underlining global warming's influence on the distribution of cold air. There are two key insights: First, our results align with previous studies, confirming an increase in WACC occurrences up to the 2020s. Second, a significant decrease in WACC events is projected from the 2030s onwards, signaling a major shift in extreme winter weather patterns. These findings necessitate a re-examination of current models and theories regarding extreme winter weather. Importantly, this research highlights the need for updated climate models to enhance future preparedness and response strategies.

Asian Dust Storms (ADSs) have been one of the high-impact atmospheric hazards in South Korea (SK). They originate in dry regions, such as the Gobi Desert and Inner Mongolia, primarily during spring, and are transported to SK along the westerlies, significantly impacting air quality in SK. To deal with the air pollution problem, it is vital to use a numerical model to enhance air quality forecasting skills. In particular, the performance of numerical air quality prediction is highly related to the accurate forecast of the dust storm occurrence at the source and the dust concentration. In this study, we built an intelligent optimization system by coupling the micro-genetic algorithm (μGA) and the WRF-Chem model—the WRF-Chem-μGA system. This system can find optimal parameters related to dust emission in WRF-Chem to improve air quality forecasting. For optimization, we selected the case of the recent ADS, which significantly impacted SK. Overall, the WRF-Chem with the optimized set of parameter values outperforms the non-optimized ones in forecasting the ADS events in SK.

The Korea Institute of Atmospheric Prediction Systems (KIAPS) has embarked on enhancing its predictive capabilities from very short range (~6 hours) to extended medium range (~30 days), including coupling to various earth system components like the land surface, oceans, and sea ice. The first phase of the KIAPS project delivered the global atmosphere-only NWP system that was made operational at the Korea Meteorological Administration (KMA) in April 2020. The NWP model - named the Korean Integrated Model (KIM) - is a non-hydrostatic model on a cubed-sphere grid that employs the spectral element method for its dynamical core. The global data assimilation (DA) system is based on a hybrid-4DEnVar system for the deterministic analysis, and an LETKF for ensemble perturbation updates, and is already giving good performance. However, some systematic model error is found in the KIM forecast. Notably, a positive temperature bias exists below 500 hPa across most regions, except for high latitudes. Furthermore, while an overall wet bias prevails in this region, a dry bias was identified at 850 hPa. Data assimilation uses various observations to produce analysis fields with reduced errors, but persistent systematic errors remain. There is a question as to how much analysis field error can be corrected at most if data assimilation is perfect. To find an answer to this, assuming the analysis field of ERA5 as a truth, a virtual Sonde observation with a dense global grid was created based on ERA5 and used for data assimilation. Results indicated a substantial reduction in analysis errors across most regions when using virtual Sonde observations, in spite of regional variability. Our presentation will focus on quantifying the maximum error correction achievable through data assimilation. Furthermore, the results of virtual Sonde data assimilation will be compared with the current data assimilation results.

The ensemble data assimilation (EDA) is widely applied nowadays for its ability to estimate the flow-dependent error covariance. For EDA, localization methods are commonly applied to mitigate the false correction induced by the sampling error. However, localization shrinks covariance over a specified distance much and could seriously sacrifice the benefits from the flow-dependent error covariance. Consequently, several localization schemes, dual localization (DL), successive covariance localization (SCL), and scale-dependent localization (SDL), are proposed to deal with the sampling error in more elaborate manners with consideration of scale. While DL and SDL perform scale separation to the background ensemble perturbation, SCL distributes observations to different scales. Meanwhile, DL and SCL perform localization to observation error covariance (R localization), and SDL performs localization to background error covariance (B localization). In this study, we compare the three schemes with a main concern on the ability to improve covariance estimation. With a convective-scale 1000-member ensemble forecast during the Tokyo Olympics, we evaluate how accurate these schemes can be with a smaller ensemble and measure the skill by comparing with the 1000-member-one without localization. Not only the Kalman gain and increment are inspected, but we also estimate the equivalent B localization matrix for DL and SCL to get a more concrete picture of how the three schemes refine the noisy ensemble-estimated covariance. An OSSE based on the QG model is further conducted to examine whether the findings result from the offline diagnosis persist, and how they evolve in an online cycling experiment.

Regional atmospheric models require lateral boundary conditions obtained from global circulation models. The regional analysis sometimes suffers from deterioration of the large-scale structure than that of the global analysis due to limitations in the domain size and observations. The large-scale error may cause the displacement error for disturbances such as typhoons or synoptic-scale fronts and degrade the performance of convective-scale DA. Although several scale-dependent blending methods of global and regional analyses have been proposed to alleviate those large-scale errors, these blending methods may hinder the optimality of individual DA. Guidard and Fischer (2008) and Dahlgren and Gustafsson (2012) introduced the augmented information vector with the global analysis into the regional variational assimilation and reported promising results. However, their formulations require several assumptions for the error correlations and ignore the covariance between the global and the regional forecast errors. In this study, we extend their augmented variational formulation to an ensemble variational method to relax those assumptions and take the flow-dependency of the forecast error into account. We test the proposed method in the one-way coupling system using ideal one-dimensional models proposed by Lorenz (2005) and compare the results with those of the separate assimilations and of the previous studies. The effect of the covariance between the two domains will also be discussed.

A dense observation network of surface weather stations has been established in Taiwan, which can provide surface pressure, wind, temperature, and humidity measurements in real-time. These observations carry crucial information about spatial and temporal variations in convective-scale weather systems. However, significant biases exist between near-surface observation and model simulation outputs, posing a major challenge in assimilating surface data effectively. The complex terrain of Taiwan even deteriorates the biases between the observations and the model. Therefore, implementing a bias correction scheme in surface data assimilation in Taiwan becomes a critical issue to consider. This study was inspired by the well-established variational bias correction (VarBC) method commonly used in satellite data assimilation. The VarBC method is extended to the surface data assimilation and implemented in a WRF/WRFDA-based convective-scale data assimilation system at the Central Weather Administration (named CWA-RWRF) to assimilate dense surface observations in Taiwan. Utilizing the statistical relationship between innovations and multiple predictor variables in the bias model of VarBC, adaptive bias correction is achieved during assimilation. The impact of the VarBC on surface data assimilation is investigated with a Meiyu front-related heavy rainfall events characterized by coastal convection over western Taiwan from 6 to 8 June 2022. A set of VarBC models is introduced for assimilating 10-meter wind, 2-meter temperature, and 2-meter humidity observations, while radar observation data are also assimilated as in the usual operation. Results indicate that the use of the VarBC reduces the mean errors in the analysis and forecast fields, with particularly significant improvements in 10-meter wind speed and daytime temperature. The influence of the bias correction on the dynamic and thermodynamic conditions for the rainfall events will be further discussed in the presentation.

In air quality prediction, initial conditions are essential for a coupled atmosphere-chemistry model based on atmospheric and aerosol/chemistry observations. Typically, a greater quantity and higher quality of observations result in more accurate model outcomes. However, forecast errors within a specific region of interest can magnify from observational deficiencies in a particular upstream area, primarily originating from initial errors in that region. Therefore, the primary task is to identify these upstream areas where minor initial errors could escalate into significant forecast errors within the area of interest. Conditional Nonlinear Optimal Perturbation for Initial Conditions (CNOP-I) can serve as a crucial tool for adaptive (targeted) observations. Computing CNOP-I involves multiple energy equations defining various variables such as kinetic energy, dry energy, moist energy, etc. This study comprises of two primary objectives: categorizing synoptic weather patterns through which Asian Dust Storms (ADSs) intrude the Korean Peninsula and identifying sensitive areas for adaptive observations concerning each distinct synoptic pattern. A thorough investigation was conducted on ADS events occurring in South Korea over the past 32 years (1990 to 2021). Through principal component analysis, it was confirmed that key variables like temperature, vertical velocity, divergence, specific humidity, ozone mass mixing ratio, and eastward wind play pivotal roles in the occurrence and movement of ADSs, with robust downdrafts and divergence emerging as crucial factors ultimately leading ADSs to converge on the Korean Peninsula. Utilizing these factors as primary variables, dust influx synoptic patterns are classified using the K-means clustering method to identify sensitive areas for adaptive observations in air quality prediction through CNOP-I. These proposed targeted observation areas related to dust patterns aim to improve air quality prediction by categorizing weather conditions triggering severe ADS occurrences in South Korea and identifying upstream regions for intensified observations through international collaborations.

Climate extremes, such as hot temperature and heavy precipitation events, have devastating effects on human societies. As the planet warms, they have become more intense and more frequent. To avoid irreversible damage from climate extremes, many countries have committed to achieving net-zero anthropogenic carbon emissions, or carbon neutrality, by 2050s. Here, we quantify the impact of carbon neutrality on population exposure to climate extremes using multi-model projections from the Coupled Model Intercomparison Project Phase 6 (CMIP6) based on the Shared Socioeconomic Pathway (SSP)1-1.9 and SSP3-7.0 scenarios. It is found that the increasing exposure of the population to hot-temperature and heavy-precipitation extremes can be substantially reduced by 87–98% in the late 21st century by achieving carbon neutrality. The benefits of carbon neutrality are particularly pronounced in Africa and Asia. The potential benefits of carbon neutrality are also significant in North America, Europe, and Oceania, where a reduction in climate extremes is more than twice as important as population decline in reducing population exposure to climate extremes. These results provide important scientific support for ongoing efforts to achieve net-zero carbon emissions by 2050s to reduce potential climate risk and its inequity across continents.

In recent years, the strengthened subtropical pacific high-pressure system has frequently led to prolonged high temperatures. Due to the influence of heatwave events, certain urban areas consistently exhibit higher temperatures than their rural counterparts—a phenomenon known as the urban heat island effect. Previous studies have focused less on the interaction between large-scale circulation patterns and urban heat island effects. The purpose of this study is to examine impact of the Western North Pacific Subtropical High (WNPSH), the predominant synoptic weather system during summer on heat waves and its subsequent influence on the urban heat island. Utilizing EOF analysis and anomalous analysis, the results indicate that the intensity of the WNPSH system has increased in recent years, particularly for the leading EOF (EOF1). This phenomenon may be attributed to the global warming effect. Thus, we focus on EOF1 and its impact on temperature, humidity, and precipitation in Taiwan. A heatwave event was defined as the maximum daily temperature exceeding the 95th percentile for three consecutive days in the plain area (altitude < 500 m). After identifying the heatwave days, the WRF model coupled with the urban canopy model was used to investigate the impact of heat wave events on the urban heat island problem in Taiwan. EOF1 is regressed with humidity, precipitation, and specific humidity to examine the heat wave effects. The preliminary analysis indicates that the warming phase may be more significant in urban sites than rural ones, especially in coastal sites where higher humidity may cause uncomfortable conditions. Detailed discussion will be illustrated during the meeting.

This study aims to predict the future climate of South Korea by utilizing climate change scenario. The focus is on forecasting the country’s future climate under warming targets of 1.5°C, 2.0°C, and 3.0°C. The timing of the warming target attainment points was determined using the global climate model (UKESM1), calculating the years when the global temperature rose by 1.5°C, 2.0°C, and 3.0°C relative to the pre-industrial era(1850-1900). In this study, we utilized high-resolution (1km) gridded observation data(2000-2019) and climate change scenario (2021-2100) using PRIDE model in South Korea. We generated 20 ensemble members based on 5 regional climate models (RCM) and 4 Shared Socioeconomic Pathway (SSP) scenarios for analysis. Using the present period of 2000-2019, the annual mean temperature is projected to increase by +0.7°C, +1.4°C, and +2.6°C during the 1.5°C, 2.0°C, and 3.0°C warming periods, respectively. Precipitation is projected to slightly decrease during the 1.5°C warming target, it is anticipated to significantly increase with further warming. The analysis of future extreme climate conditions due to warming reveals an increase in summer heat extremes and a decrease in winter cold extremes. Extreme precipitation and extreme wind/humidity indices are generally expected to increase with warming. According to the research findings, if additional warming beyond 1.5°C occurs, South Korea could face more severe extreme climate events. Therefore, the results of this study are expected to provide essential information for formulating climate change adaptation and mitigation policies different regions in South Korea.

One of the various thermal indices, apparent temperature (AT) is widely used by meteorological agencies around the world in a wide range of fields, such as heat health warning systems and worker productivity assessments. In this study, we investigated the characteristics of AT changes in South Korea using a national standard climate change scenarios with a high resolution of 1km based on the Coupled Model Intercomparison Project Phase 6(CMIP6). We analyzed thermal stress and its associated contributions during the climate change period in South Korea and five major cities. In the present-day(PD), high AT occurs in major cities due to high temperatures (TAS) and relative humidity (RH). We find that changes in AT are mainly attributed to changes in TAS, whereas changes in RH play a relatively minor role. Similar annual trends between AT and TAS show that the long-term increase in AT is characterized by an increase in TAS, with RH changes playing a relatively minor role. Our research results demonstrate that significant AT can occur even when TAS is relatively low, primarily due to high humidity. Particularly under future warmer climate conditions, high AT may occur first in the five major cities and then expand to surrounding areas. Under future warmer climate conditions, increased TAS and RH from March to June, traditionally not considered hot seasons, may lead to earlier onset of heat risk and more frequent occurrence of high heat stress events.

The 3 July 2019 Kaiyuan City, Liaoning Province, tornadic supercell is simulated with the advanced research version of the Weather Research and Forecasting (WRF-ARW) model using three nested grids with 4.05-km, 1.35-km, and 450-m grid spacings. The radar radial velocity and reflectivity data from eight operational Doppler weather radars in different wavebands together with automatic weather station (AWS) data are assimilated through the ARPS three-dimensional variational data assimilation (3DVAR) and cloud analysis systems on the 450-m grid to generate a more accurate initial condition that includes a well-analyzed supercell and associated low-level mesocyclone. Sensitivity experiments show that the analyses and forecasts of the tornadic supercell of the control experiment assimilating both radar and AWS data are significantly improved relative to the one without special data assimilation. The near-surface vortex intensification is not produced when the radar radial velocity and reflectivity data are assimilated alone without AWS data. One assimilation cycle experiment performs better than multi-cycle ones, due to the additional latent heating added by the cloud analysis in each cycle. The use of a divergence constraint in the 3DVAR plays an important role in establishing the low-level mesocyclone during the assimilation and forecast. A tornado-resolving simulation with further nested grids 150 m, 50 m, and 16 m grid spacings starts from the 450-m control initial condition is conducted. The inner-most grid covers the entire period of the observed tornado outbreak and successfully captures the development of tornadic vortices. The intensity of the simulated tornado on the 16-m grid reaches the enhanced Fujita scale 2 (EF2) intensity. The simulated tornado vortices evolved through one-cell, two-cell, and multi-vortex stages. This study represents the first time a real tornado is successfully simulated through the assimilation of multi-radar and high-density surface station data.

It has long been known that coupling between the various Earth system components (the ocean, atmosphere, sea ice, and land) produces improved forecasts on seasonal and longer time scales. KIAPS (Korea Institute of Atmospheric Prediction Systems) is now developing a coupled atmosphere-land-ocean-sea ice model aimed at extended-range forecasts, using NEMO (Nucleus for European Modelling of the Ocean) as the ocean model. Also, we are developing a weakly-coupled atmosphere-ocean data assimilation (WCDA) system to provide more balanced initial conditions for this new coupled model. The atmospheric model is based on the KIM (Korean Integrated Model) v4.0 and the atmospheric data assimilation (DA) system uses the “KVAR” variational DA software to run 3DVAR-FGAT analyses within a 6-hour DA cycle. The ocean system is based on KMA’s “GODAPS” system (Global Ocean Data Assimilation and Prediction System), which uses the NEMO ocean model. The ocean DA system uses the NEMOVAR software to run 3DVAR-FGAT analyses, with KIM surface forcing. To match the atmospheric DA cycles, we changed the ocean DA cycles from 24-hours to 6-hours. And we upgraded the NEMO version used by the coupled model. We developed the WCDA system which coupled the ocean DA system to the KIM atmospheric DA system. And we also produced the uncoupled DA system for control experiments. Here, we will introduce our progress for the new WCDA system for the KIM coupled model and present preliminary results.

The Korea Institute for Atmospheric Prediction Systems (KIAPS) is developing a coupled earth system NWP system aimed at improving the performance of extended-medium-range forecasts. The system will be based on a new coupled model which combines the existing “Korean Integrated Model” (KIM) atmosphere model with models for other earth system components, including the NEMO (Nucleus for European Modelling of the Ocean) ocean model and the SI3 (Sea Ice Modelling Integrated Initiative) sea ice model. To provide the ocean data assimilation component, we are adapting the KMA GODAPS (Global Ocean Data Assimilation and Prediction System), which consists of the NEMO ocean model, the CICE (Community Ice CodE) sea ice model, and the NEMOVAR (NEMO VARiational) data assimilation (DA) system, and obtains surface forcing from KMA UM (Unified Model) forecasts. As part of this work, we have added support for use of KIM rather than KMA UM surface forcing, and we have reduced the cycling period and window length from 24 hours to 6 hours, to match the atmospheric DA cycling system. In this study, we will summarize the configuration of the ocean DA system, and then discuss the impact of changing surface forcing and shortening the cycling period on key analysis and forecast metrics.

KIAPS (Korea Institute of Atmospheric Prediction Systems) has developed a global ocean data assimilation system base on KMA (Korea Meteological Administration) GODAPS (Global Ocean Data Assimilation and Prediction System). The ocean data assimilation (DA) system is based on NEMOVAR, and uses a 3-Dimension Variational data assimilation – First Guess at Appropriate Time DA method. The assimilated observations include SST observations from satellite and moored buoys, temperature profiles collected by Argo floats, and satellite observations of sea-level anomaly and sea ice concentration. Our actual first goal is to develop a weakly-coupled DA system that combines Korean Integrated Model (KIM)’s existing atmosphere/land DA system with a separate DA system for the NEMO (Nucleus for European Modelling of the Ocean) model that has been chosen for use within the KIM coupled model. In order to make the system suitable for weakly-coupled DA with KIM, we tried to change ocean and sea-ice model from NEMO-CICE (Community Ice CodE) to NEMO-SI3 (Sea Ice Modelling Integrated Initiative) based on the ocean DA system with 6-horly DA window. We will present studies of updated ocean-sea ice model aimed at assessing the impact of the changes on ocean analysis performance.

KMA (Korea Meteorological Administration) operational global NWP system is based on a 6-hour analysis cycle, with 6-hour data assimilation windows. Early-observation-cutoff analyses are used to initialise the main forecasts, and late-observation-cutoff analyses used to generate background forecasts for the next analysis update. In this system, the most up-to-date forecasts can be sometimes be based on observational data that is at least 6 hours old. By producing analyses and forecasts more frequently, we can reduce this time gap, and potentially improve the quality of the most up-to-date forecast products at certain times of day. In this study, we evaluate two methods of producing 3-hourly rather than 6-hourly analyses: 1. Simply adding extra analyses and forecasts to the current system, without changing the underlying data assimilation cycle. 2. Switching to a 3-hour cycle based only on early-observation-cutoff analyses, but keeping the analysis window lengths at 6 hours. The latter method requires us to overlap the data assimilation windows, but observations are still assimilated only once. Apart from more frequent forecast delivery, this new cycling strategy can benefit from producing smaller analysis increments, and therefore reducing linearization errors in the analysis steps. On the other hand, removing the late-observation-cutoff analyses reduces the number of observations available for assimilation. After introducing our experiment configurations, and the impacts on observation usage, we present results from extended NWP experiments designed to answer the following questions: 1. What is the impact of moving to the 3-hourly cycling system on forecast performance. 2. What is the impact of using fewer observations. 3. What is the impact on gravity-wave noise.

One of most advantageous thing of OSSE is that researcher can verify the result of NWP with respect to the truth (called as nature run (NR)). Furthermore, observation can be simulated for any proposed (for example, making new type of sensor, applying realistic forward operator and method, specifying the locations, and error characteristics of the observation). To use such advantage of OSSE, we developed the prototype of global OSSE system used Korea Integrated Model (KIM). In this study, we will introduce the theoretical background of OSSE, forecast system of KIM, and the contents considered for development of the OSSE system used KIM, briefly. To verify the OSSE system used KIM, we performed the denial experiments for synoptic observation (aircraft) and satellite observation (MHS), not only in OSSE but also in OSE. Because of the absence of wind and temperature observation of aircraft in aircraft denial experiment, the performance decrease of KIM analysis against IFS reanalysis were mainly shown at 250 hPa (which is a cruising altitude of aircraft) of the u, v, and T variable. In MHS denial experiment, the performance decrease of T and q variable were appear in KIM analysis. The reaction of analysis due to the observation denial in both OSSE and OSE weren't in the same perfectly, but the result from OSE and OSSE system was similar, we evaluated the OSSE system used KIM is well made enough to analyze the effect of observation. The reaction of analysis due to the observation denial in both OSSE and OSE weren't in the same perfectly, but the result from OSE and OSSE system was similar, we evaluated the OSSE system used KIM is well made enough to analyze the effect of observation.

The ATMS data have, until now, only been assimilated over sea in Korea Integrate Model(KIM). To extend the coverage of the ATMS over land and sea-ice, we firstly assimilated only the high peaking, non-sensitive surface, channels (ch8~15 over land and ch22~23 over sea-ice). To extend the use of ATMS high peaking channel over land, bias correction and observation error covariance were adjusted. As a result, the analysis field error of stratosphere was reduced especially and the forecast skill improved (GPH NH: 0.48~2.65%, Asia: 1.35~ 2.35%). In order to assimilate more surface-sensitive microwave observation, the contribution of the atmosphere must be separated by removing the effect of surface radiance from the observed brightness temperature. Compared to the ocean, the land skin temperature has higher uncertainties, and surface emissivity is larger at around 0.8-0.95 with higher heterogeneities due to the complex surface condition, making it harder to model. So, estimating the land emissivity/surface temperature directly from satellite is tested. Using the radiation transfer equation and observation, the observation-dependent surface emissivity estimation method was introduced. Preprocessing was performed to remove the edge area and scattered data that cause errors in emissivity retrieval, and remove the data that differed significantly from the emissivity climate value. It was diagnosed that the model surface temperature is about 3K higher than the value calculated based on ATMS observation, which cause 20% difference in the observation increment of channel 6. It is necessary to use a more appropriate surface temperature for microwave surface radiance and correct the resulting bias. The number of observation available on land increased by 200% when an observation-dependent emissivity is applied compared to the default emissivity model. It will be tested to support a more reliable assimilation of the surface-sensitive channels over land.

The renewed KIAPS (Korea Institute of Atmospheric Prediction Systems) aims at developing a unified framework for seamless prediction from very short range (~6 hours) to extended medium range (~30 days), including coupling to various Earth system components, such as the land surface, oceans, and sea ice. For very short range forecasts, KIAPS is developing at the high-resolution (1-5km) model targeting storm-scale high-impact weather over East Asia, centered on the Korean peninsula. We have been developing a convective-scale DA system based on Local Ensemble Transform Kalman Filter (LETKF) that can respond to high-density observations such as radar, with increased horizontal resolution of the model. In order to assimilate from meso-scale observations to synoptic-scale observations, we have completed the development of LETKF-based radar reflectivity DA system for a “testbed” system based on a high-resolution, limited-area version of WRF and a testbed background ingestion system in KIM Package for Observation Processing (KPOP). The LETKF-based convective-scale DA system has been extended to assimilate conventional observations (Sonde, Surface, Aircraft, etc.) provided by KPOP in addition to radar reflectivity observations. Also, we have established the basic framework called the regional ensemble analysis system that is consist of testbed, KPOP, LETKF-based convective DA system. In order to get reasonable ensemble analysis fields through LETKF-based convective-scale DA system and tune radar DA techniques, we need to investigate the analysis performance of radar reflectivity DA added conventional observations. So, we will perform cycle experiments for the assimilation of radar reflectivity adding Sonde, Aircraft, and Surface observations using the regional ensemble analysis system. In this presentation, we will introduce results for the evaluation of LETKF-based radar reflectivity DA on observation-space statistics and the characteristics of ensemble analysis fields. And then, we will proceed with tuning the LETKF-based radar reflectivity method depending on the evaluated analysis performance.

The aircraft-based observation bias exists in temperature data due to various factors such as the aircraft type, flight phases, and equipment sensors. Nonetheless, the observation used in the numerical weather prediction (NWP) models is one of the important conventional data. Although the static bias correction (BC) based on the linear regression was applied to the Korean Integrated Model (KIM) Package for Observation Processing (KPOP) system which is being operated in practice, there were limitations of a spatial discontinuity and a dependency on the calculation period of BC coefficients. To reduce these constraints and consider the spatial and temporal characteristics and correlations of each observation at the same time, a bias estimation model was devised based on a Multi-layer Perceptron (MLP) applied an attention method in this study. The attention method makes an estimated bias concentrate on the correlation of more related observation variables. After the predicted bias was removed in the pre-processed procedure, the statistics value improved about 40% and the number of observations available in the data assimilation (DA) system increased about 3.5% in a given test period. As results of DA cycle tests, the error of analysis fields against IFS data was reduced over the North Pacific and North Atlantic Ocean at 250hPa, where most observations exist. Meanwhile, the error increased over North America and Europe area. It results from the different characteristics of model background error according to those regions. In future research, we will take into account way to further reflect regional error characteristics into this BC model.

This study presents a machine learning (ML) based calibration process and instrument error correction experiment for the novel low-cost mini-radiosonde system, Storm Tracker (ST). To address temperature and moisture errors caused by solar radiation heating. We adopted the Generalized Linear Model (GLM) and Gradient Boosting Model (GBM) as the kernel for ML model development. We used the co-launch radiosonde data from ST and Vaisala-RS41 (VS) between 2018 and 2022 as training inputs to develop a two-stage calibration approach. The data calibration system can correct the observed temperature with the ML model and then calibrate the moisture with the corrected temperature as input. In the data calibration workflow, we also designed a suitable experimental platform based on ST's characteristics to test the instrument's measurement error. It can calibrate the instrument's error as the standard reference for each ST instrument. The results show that the measurement error of ST is a normal distribution. About 70% of the ST instrumental errors can be reduced to within the limitation of instrument accuracy after statistical error correction. The remaining 30% of the instruments still retain the characteristics of random errors, which can be corrected through our ML models. The linear calibration model (GLM) and the non-linear calibration model (GBM) have been compared. Results show that GLM exhibits good calibrating capability for observing the planetary boundary layer (1000-700 hPa). On the other hand, GBM, with a more complex structure, maintains even better calibrating performance statistically. It can reduce the overestimation of convective available potential energy (CAPE) and column water vapor (CWV) amount for PBL observations. Furthermore, this study's systematic data correction process has been made accessible through a dedicated website. This user-friendly platform allows for the seamless application of the correction process, providing a convenient tool for real-time ground-based data correction.

This study analyzes remote rainfall events in northeastern Taiwan from September to February over the period of 1980-2020. These events are triggered by the interaction between typhoons and the northeasterly monsoon flow. When background northeasterly winds exceed 7 m/s, a “remote rainfall-prone area” can be defined north of the Philippines, spanning the northern South China Sea to the north Philippine Sea. The confluence of typhoon outer circulations and strong monsoon winds creates a convergence zone, enhancing rainfall in northeast Taiwan by an average of 80–220 mm per day. With typhoons positioned over the rainfall-prone area, the probability of daily maximum rainfall exceeding 200 mm rises above 20%, particularly in the region between 20°N–22°N and 116°E–124°E, northward to Luzon Island. Here, the probability can top 45%. The zone of highest risk for extreme remote rainfall events lies between 20°N–22°N and 118°E–120°E, with over 90% probability when the key convergence zone aligns. Although this convergence zone does not necessarily coincide with significant baroclinic forcing, the abundant moisture supplied from the typhoons remains conducive to heavy rainfall events.

The Yeongdong region experiences various mesoscale phenomena due to the complex topography of the Taebaek Mountains, particularly during winter. Predominant mesoscale features such as Cold-Air Damming (CAD) and barrier winds significantly impact snowfall characteristics, posing challenges for accurate snowfall forecasting. We aim to improve snowfall forecast by classifying Strong Snow (SS) and Strong CAD (SC) cases under similar weather conditions (850hPa temperature below -10℃) using rawinsonde data. Through analysis of observations and LDAPS data, the main factors contributing to snowfall differences between SS and SC cases are investigated. We introduce the Froude number (Fr) to quantify snowfall variability induced by Barrier wind and CAD. In SS1 and SS2 cases, strong northerly to northeasterly flow up to 2km was observed in rawinsonde data. LDAPS revealed a dome-shaped structure over the East Sea with a temperature gradient of 0.5-1K. Snowfall occurred when northeasterly flow penetrated inland deeply. In contrast, SC1 and SC2 cases exhibited a strong inversion layer, with lower-level northwesterly flow thicker up to 1km than the northeasterly layer. LDAPS indicated a strong CAD with a temperature gradient exceeding 2K, inhibiting snowfall by preventing snow clouds from penetrating inland. During the SC3 case on February 14, 2023, the northwesterly barrier wind was poorly represented in the LDAPS simulation. Strong barrier wind (approximately 10 ms-1) inhibited the inland penetration of snow clouds according to rawinsonde, Wind Profiler, and Buoy data. Subsequently, heavy snowfall occurred on February 15 with the inflow of strong northeasterly winds. The investigation of the Froude number confirms the significant influence of barrier wind and CAD on snowfall location and amount. Fr criteria for quantitative differentiation between SS and SC cases at each locations. The investigation of the Froude number would help to quantify the influence of barrier wind and CAD on snowfall location and amount.

Characteristics of downslope windstorm (DW) has been used for the criteria of DW selection mainly based on 1-min average wind and the other meteorological conditions in the Yeongdong region for 2000 - 2020. The criteria is as follows; First, a classification procedure for the downslope windstorm is proposed using surface wind speed (greater than 99 percentile), 1-hour longevity of strong wind (SW), westerly wind direction, low humidity (less than 20 percentile), and leeside warming. The number of DW days satisfying the proposed criteria is 221 (2.9% of total days) for 2000 – 2020. The DW occurrences show distinctive annual variation with its peak in April. Mean wind speed of DW days is 8.2 m s^-1 with its duration of 2 hr 30 min and relative humidity of 28 % at Gangneung. Two DW episodes are chosen satisfying the criteria for the intensive analysis. The first episode was 7 May 2021. The sounding shows that the layer of wind speed greater than 25 m s^-1 was lowered down to 925 hPa at Gangneung (leeside) relative to 850 hPa at Hoengseong (windward), in the afternoon along with significant warming and drying. The second episode (11 April 2023) was more than 2 times stronger than the first in terms of its intensity and duration except for much more humid condition. The preliminary analysis shows that this kind of strong downslope windstorm was attributable to the partial reflection of mountain wave in the lower troposphere. Froude numbers of Wonju (windward) and Gangneung (downwind) for the both DW events were increased 4 and 5 times greater than those of normal days, respectively. We need to investigate DW mechanisms in terms of gravity wave dynamics and further prerequisite conditions of favorable downslope windstorms to improve DW forecast in the leeside (Yeongdong) of the mountains.

Efforts to understand extreme events often require reliable climate metrics information. Several studies have focused on correcting the climate model simulation to study the effects of climate change. However, these conventional bias correction methods often fail to distinguish the inter-variable dependence of climate variables output over the climate scenario, even though it is considered relevant for the various impact studies. To counter this, this study evaluated the effectiveness of post-processing bias correction of extreme climate indices rather than the climate models data at a global scale. The Shared Socioeconomic Pathways (SSP3-70) in the ICCP CESM2 large ensemble climate model and ERA5 reanalysis datasets were used in this study. The multi-year drought indices of Standardized Precipitation Evapotranspiration (SPEI) and Quantile Delta Mapping Bias Correction (QDM)-BC method were used. The period between 1979-2014 was used as the calibration period, while 2015-2022 as the evaluation period of the outputs of the SPEI indices over the comparable period (1979–2022). The results show that the QDM-BC post-processing is effective in preserving the distribution of the SPEI as compared to applying the BC to raw climate data. Therefore, this study highly recommends post-processing bias-correction methods on climate extremes in order to preserve the climate change signal and the internal variability in the changing climate of each ensemble member.

Greenland occasionally experiences extreme temperature conditions in terms of temperature and precipitation, which exert a significant impact on the ice sheet covering major parts of the land mass. By employing stochastical methods, we aim to disentangle the role of natural variability and anthropogenic forcings on the development and maintenance of prevailing temperature anomalies over Greenland. Based on current reanalysis datasets and the CESM2 Large Ensemble, we analyze and attribute anomalous warm years to natural variability and anthropogenic forcings. The preliminary results indicate the dominance of blocking-like conditions over Greenland, which coincided with an anomalous anti-cyclonic motion advecting moisture and warm air into the interior of Greenland, destabilizing the atmosphere there. This results in radiative radiative feedback enhancing local warming. Further, analysis of the role of climate variability modes gives implications for the predictability of summer temperature extremes.

The 2021 IPCC report found that most studies show declining trends for the global diurnal temperature range (DTR) since the 1950s, decreasing mainly during 1960–1980. This issue is revisited here using an up-to-date in-situ data set, Hadley Center Integrated Surface Database, constrained by rigorous station selection conditions. The global observed DTR trend was found to reverse during 1980–2021, increasing significantly at a rate of 0.091 ± 0.008°C decade⁻¹. The trend was dominated by a faster rate of increasing daily maximum air temperature. This increasing observed trend in the past four decades was not fully captured in raw CMIP6 models, as models only partially capture the spatial patterns. With global CMIP6 outputs and regionally-available observations, the global land DTR was then estimated, through emergent constraints, to be 0.063 ± 0.012°C decade⁻¹. The study raises concern for risks of increasing DTR globally and provides new insights into global DTR assessment.

The reasons for large discrepancies between observations and simulations, as well as for uncertainties in projections of the equatorial Pacific zonal sea surface temperature (SST) gradient, are controversial. We used CMIP6 models and large ensemble simulations to show that model bias and internal variabilities affected, i.e., strengthened, the SST gradient between 1981 and 2010. The underestimation of strengthened trends in the southeast trade wind belt, the insufficient cooling effect of eastern Pacific upwelling, and the excessive westward extension of the climatological cold tongue in models jointly caused a weaker SST gradient than the recent observations. The phase transformation of the Interdecadal Pacific Oscillation (IPO) could explain ~51% of the observed SST gradient strengthening. After adjusting the random IPO phase to the observed IPO change, the adjusted SST gradient trends were closer to observations. We further constrained the projection of SST gradient change by using climate models’ ability to reproduce the historical SST gradient intensification or the phase of the IPO. These models suggest a weakened SST gradient in the middle of the 21^st century.

The faster warming for Arctic Ocean surface air temperature (SAT) relative to that at lower latitude is connected with various processes, including local radiation feedback, poleward oceanic and atmospheric heat transport. However, we do not fully understand how combinations of different low-frequency internal climate modes influence Arctic amplification on the decadal timescale. Here, the decadal Arctic SAT variation, its connection with the Atlantic Meridional Overturning Circulation (AMOC) and possible underlying mechanisms, are investigated based on several independent observational proxies, pre-industrial experiments, and historical large ensembles of two CMIP6 models. Our study suggests that AMOC and Arctic SAT vary in phase on the decadal timescale, whereas this relationship is weak at the interannual timescale. Further analysis shows that the AMOC accompanied with cross-basin oceanic water/heat transport between Atlantic and Arctic would alter air–sea interface exchange over the melting ice regions, and then amplified poleward atmospheric heat and moisture transports which originated from Pacific Ocean in budget analysis. The resulting enhanced downward longwave radiation finally warms the Arctic SAT. Additionally, the North Pacific Oscillation (NPO) can modulate the relationship between AMOC and Arctic SAT on decadal scale. Specifically, a phase shift in the NPO can contribute 10%–40% of the correlation between AMOC and Arctic SAT. Our study provides potential sources for predicting the Arctic climate and constraining its uncertainty in future projections.

The phase state and size distribution of particulate matter (PM) are important factors affecting their characteristics. The phase state of PM can affect particle size distribution, however there is limited information about how the phase state of fine PM (PM_2.5) _­in the atmosphere affects its size distribution. In this study, conducted from 2020 to 2022 in Seoul, Seosan, and Beijing, we investigated the impact of the phase state of PM_2.5 on its size distribution. PM_2.5 filter samples were collected and analyzed using optical observation techniques and the poke and flow method to investigate the phase state of PM_2.5. Additionally, during the same periods, the particle number size distributions (PNSD) of PM_2.5 and the chemical constituents of the PM_2.5 were analyzed. By focusing on the impact of the phase state on the size distribution of PM_2.5, this study contributes to our understanding of how phase state and size distribution influence atmospheric pollution. The results will be presented.

PM2.5 is composed of various chemical substances, and these substances can alter the optical properties of the air. To precisely understand this, we simultaneously analyzed in-situ observations through SPARTAN (Surface Particulate Matter Network) and remote-sensing data through AERONET (Aerosol Robotic Network). SPARTAN sampler provides information on PM mass concentration and chemical composition for each aerosol sample. This study compared SPARTAN and AERONET data from 2016 to 2022 in 11 global observation sites. Focusing on three dominant components—Ammoniated Sulfate, BC, and Fine Soil—found in SPARTAN sampling data, we observed that during periods when Fine Soil mass exceeded BC mass, the wavelength difference of single scattering albedo (SSA) between 870 and 440 nm from AERONET was positive. Conversely, during periods with higher BC mass, this difference was negative. This relationship suggests that the chemical characteristics of aerosols have optically influenced their absorption properties. Therefore, we propose some improvements in existing aerosol type classification algorithms, regarding aerosol chemical states. Furthermore, considering substances with optical absorption and non-absorption characteristics, we calculated the mass ratio of Ammoniated Sulfate to the sum of BC and Fine Soil masses and then analyzed its correlation with SSA at 440 nm. The results, presented for each site, showed relatively strong correlations (R > 0.7).

Atmospheric aerosols are enriched with surfactants that reduce the surface tension at the solution/air interface. The reduction of aerosol surface tension leads to enhanced cloud condensation nuclei (CCN) activity, exerting potential impacts on the climate. However, research on surface tension mostly focused on laboratory experiments or modeling work due to limitations in instrumentation and measurement methods. In this study, we alternatively derive the surface tension of aerosols at activation by using hygroscopicity and CCN activity field measurements at two megacities of China based on the κ-Köhler theory. The surface tension calculations for both cities are mostly lower than that of pure water, confirming the reduction of surface tension by the surfactants. Furthermore, the dependence of surface tension to organic mass concentration are calculated in respect to the Szyszkowski-Langmuir equation for various organic species. We find that the dependence in southern city is similar to that of levoglucose which is primarily produced by biomass burning. In contrast, in the northern city the dependence is similar to that of dicarboxylic acids originated from oxidative degradation of anthropogenic (e.g., primary traffic and cooking emissions) and biogenic volatile organic compounds. This work provides the observational constrains of surface tension for urban aerosols, which helps to improve the CCN parameterization in climate models.

Ammonia (NH₃), a prominent alkaline gas, contributes a vital role in the production of atmospheric particles by neutralizing acid gases. This process presents challenges to global air quality, human health, and the overall climate. Top of Form The exploration of NH₃ emissions and their atmospheric impacts in South Korea's livestock regions is currently insufficient. Therefore, this study seeks to fill research voids by generating high-resolution data on atmospheric NH₃, acid gases, PM_2.5, and other gas concentrations, along with meteorological variables. The investigation takes place in Gimje, South Korea, spanning from September 22 to October 10, 2023. Real-time NH₃ measurements were conducted at both a downwind site (near livestock sites) and an upwind site (farther away) using Los Gatos Research and Cavity Ringdown Spectrometers NH₃ analyzers, respectively. Data on PM_2.5, various gases (O₃, SO₂, NO, NO₂), and meteorological variables such as wind direction, wind speed, temperature, and relative humidity were measured. An online atmospheric Aerosols and Gases Monitor was utilized to measure acid gases, including HNO₃, HONO, and HCl. A strategic 24-hour passive NH₃ sampling was implemented at diverse sites, including pig farms, manure treatment plants, cow farms, farmlands, and upwind locations. NH₃ concentrations at upwind and downwind sites were statistically similar, indicating widespread livestock emissions in Gimje. Notably, passive sampling at pig farms and manure treatment plants revealed higher NH₃ levels due to emissions and volatilization. Diurnal patterns showed increased NH₃ during morning livestock activities and nighttime accumulation influenced by lower mixing height, temperature, high humidity, and low wind speed. Afternoons witnessed decreased NH₃ concentrations attributed to higher wind speed and temperature, along with reduced humidity. This study emphasizes understanding daily and local NH₃ variations for effective pollution control, offering insights crucial for environmental protection and public health in Gimje and beyond.

The photolysis of nitrous acid (HONO) and formaldehyde (HCHO) is a critical source of hydroxyl (OH) radicals in the atmosphere, strongly affecting ozone generation. In order to explore the pollution characteristics of nitrous acid (HONO) and formaldehyde (HCHO), long-term observations of HONO and HCHO were conducted in Shenzhen, China from January 2022 to September 2023. The results showed that the average concentration of HONO in Shenzhen during the observation period was highest in winter (0.71 ppbv), followed by autumn (0.64 ppbv), spring (0.59 ppbv) and summer (0.49 ppbv) and the average concentration of HCHO was highest in summer (3.79 ppbv), followed by autumn (2.74 ppbv), winter (2.44 ppbv) and spring (2.13 ppbv). The diurnal variation of HONO showed a similar trend throughout the four seasons, with daytime valleys and nighttime peaks. It was also affected by traffic emissions during the morning and evening peak periods. HCHO showed a more obvious diurnal unimodal characteristics and prominent photochemical generation characteristics. OH radical budget analysis showed that HONO photolysis was the main contributor(60%) to OH radical production in Shenzhen, and it had a significant promoting effect on the early morning photochemical reaction, especially in winter, while HCHO played a more important role in summer and noon. Therefore, by effectively controlling HONO and HCHO, atmospheric oxidation capacity can be reduced, which helps to control O₃ pollution.

Atmospheric nitrogen deposition has attracted more and more attention as a part of acid deposition. Excessive nitrogen deposition not only has a bad influence on ecosystem, but also affect human health. The temporal and spatial distribution of nitrogen deposition flux is influenced by anthropogenic emissions and meteorological conditions. There is a great significance to understand how nitrogen deposition results response to the change of emission conditions and meteorological conditions. We conducted numerical simulations basing on WRF-Chem in China from 2013 to 2019 to study the contribution of anthropogenic emissions and meteorological conditions on nitrogen deposition results. It can be concluded that anthropogenic emission contributes more to nitrogen deposition, while meteorological conditions contribute less, which precipitation is the main meteorological factor affecting nitrogen deposition. Reduced nitrogen deposition has a good consistency with NH₃ emission, while oxidized nitrogen deposition has a non-linear relationship with oxidized nitrogen emission. In addition, the contribution of dry and wet deposition to the total nitrogen deposition change is also different in different regions. In the Beijing-Tianjin-Hebei(BTH) region and the Yangtze River Delta(YRD) region, the contribution of dry and wet deposition is similar, while the contribution of wet deposition to the settlement change is larger in the Pearl River Delta region(YRD).

Water surface microlayer (SML) is ubiquitous in the environment and provides a unique medium for interfacial processing. To investigate the formation of organic compounds including the N-containing organics released in the gas phase, under more realistic conditions, in this study, real-time measurements of volatile organic compounds (VOCs) produced by interfacial oxidation chemistry of gaseous O₃ (100 ppb) with an authentic SML by using a novel secondary electrospray ionization ultra-high-resolution quadrupole Orbitrap mass spectrometer (SESI-UHR-MS). We show that ozone oxidation chemistry at the surface microlayers can lead to a large suite of unsaturated and saturated CHO organic compounds in the ambient air of the urban environment. In addition, a large fraction of N-containing organic compounds is produced during this chemistry which can have an impact on human health and the environment. In particular the compounds containing C=N bond are known by their toxicity. We also used Fourier transform ion cyclotron resonance mass spectrometry (FTI-CR MS) for molecular characterization of the organic compounds produced by interfacial oxidation chemistry of gaseous O₃ with authentic SML. We show that ozone oxidation chemistry can lead to a large suite of unsaturated and saturated CHO organic compounds including nitrogen (N) containing organic compounds and sulfur(S)-containing organics. The results indicate that the ozone oxidation chemistry at the authentic SML leads to the formation of N-containing compounds ten times more than the N-containing compounds formed by this chemistry on the SML sampled at the upper part of the river at the periphery of the city. We also show that an important number of aromatic compounds with light-absorbing- and toxic- properties are formed which could be discharged into the ocean or atmosphere via gas-water interchange, imposing a great concern on the urban area, in term of human health impact and environmental issues. 

In recent years, fractional contributions of semi-volatile ammonium and nitrate in urban PM_2.5 in China have been increasing due to the successful reduction of sulfate. Their contribution to PM_2.5 is strongly affected by the gas-particle partitioning ratio, which, in turn, is influenced by physical conditions (i.e., temperature and relative humidity) and chemical composition of PM_2.5. Therefore, it is essential to investigate the sensitivity of PM to ammonia and nitrate availability under different atmospheric conditions. We analyzed the concentrations of water-soluble inorganic and gaseous compounds in Hong Kong from 2013 to 2017 and in Shanghai from 2020 to 2022. The ionic PM_2.5 compositional data and meteorological parameters were input in ISORROPIA-II to calculate aerosol water content (AWC) and aerosol pH. By examining the datasets by season, we characterized the sensitivity of PM_2.5 mass to HNO₃ and NH₃ availability under different aerosol pH and AWC, following the four-regime plot proposed by Nenes. We further quantified the sensitivity of PM_2.5 during winter when PM_2.5 levels, as well as the fractional contributions of ammonium and nitrate, were at their highest. Results show that the reduction of water-soluble inorganic compounds was proportional to the reduction of TNO₃ (NO₃^-+HNO₃) and sulfate. However, the reduction of TNH₃ (NH₄⁺+NH₃) becomes increasingly efficient as more TNH₃ is reduced. During pollution episodes, reducing TNO₃ and TNH₃ would contribute to lowering PM concentration in Hong Kong. However, in Shanghai, reducing TNH₃ by even 30% still would have little effect on PM concentration, so reducing TNO₃ would be a more effective approach to improving air quality in Shanghai in the near future. The findings of our study show the importance of controlling TNO₃ and TNH₃ in Shanghai and Hong Kong and provide valuable insights for other urban aeras to design appropriate policies to improve air quality.

Nitro-aromatic compounds (NACs) in the atmosphere are receiving increasing attention due to their light absorption and biological toxicity. However, there is a scarcity of comprehensive understanding on their distribution characteristics, sources, and formation pathways. In this study, particulate, gaseous, and cloud water samples were simultaneously collected during cloud events at the summit of Mount Tai in northern China in spring, summer, and winter and the contents of 11 major NACs were determined. During cloud events, most NACs were mainly distributed in the particle phase, except dinitrophenols which were mainly distributed in the gas phase in winter. The field-derived effective Henry's law coefficients were several orders of magnitude higher than their theoretical values in pure water. Moreover, the measured concentrations of particulate NACs were substantially greater than the theoretical predictions, especially in spring. The above results indicate that NACs were partly formed via aqueous-phase reactions inside the cloud droplets or on the wet particle surfaces, which changed their distribution patterns. Furthermore, an observationally constrained multiphase chemical box model was developed and employed to elucidate the NACs formation mechanisms in the fine particles. The simulation results indicate the crucial roles of gas-particle partitioning and heterogenous uptake on aerosols. The results highlight the importance of aqueous-phase reactions on the formations of NACs and provides modelling insights on the formation of particulate NACs and the contributions from different sources, addressing the underestimation in traditional models regarding gas-particle partitioning.

The interannual variability of evapotranspiration (ET) in the Maritime Continent (MC) is generally low, despite significant fluctuations in precipitation influenced by the El Niño-Southern Oscillation (ENSO). This study explores the underlying dynamics of this phenomenon, using observations, reanalysis data, and climate model. We decomposed ET into canopy evaporation (CE), canopy transpiration (CT), and soil evaporation (SE) to understand their interactions. Our findings reveal that during El Niño years, reduced precipitation leads to a significant decrease in CE and SE, while CT increases notably due to higher incoming solar radiation. This self-compensating mechanism between CE and CT results in only a slight decrease in ET during El Niño years, thus dampening ET's interannual variation. Conversely, La Niña years show opposite patterns. Furthermore, the MC rainforests have experienced severe deforestation in recent decades. Idealized deforestation simulations in CESM indicate that deforestation leads to a decrease in ET and an increase in precipitation. Notably, deforestation removes the dampening effect of CT, resulting in an amplified interannual variation of ET. This change in ET’s interannual variability due to deforestation also leads to a decreased interannual variation of precipitation. Our detailed investigation reveals that the mean states can modulate deforestation's impacts on precipitation, consequently influencing its interannual variation. Overall, this study underscores the crucial role of forests' CT in regulating the interannual variability of ET and highlights the complex interplay between deforestation, ET, and precipitation in the MC. It emphasizes the significant impacts of deforestation and the critical importance of tropical rainforests in the hydroclimatological cycle.

Deforestation can impact surface temperature via biophysical processes. Earth system models (ESMs) are commonly used tools to examine biophysical effects of deforestation, but the model capacity to represent deforestation effects remains unclear. In this study, we comprehensively evaluate the performance of four ESMs of the Coupled Model Intercomparison Project Phase 6 (CMIP6) in representing deforestation effects with a satellite-based benchmark. The results show that the ESMs can basically capture the sign of the temperature response but over- or underestimate the magnitude. Such biases are the consequence of biases in the simulated responses of albedo and sensible and latent heat fluxes. Specifically, the ESMs consistently overestimate the albedo response under snow-covered conditions, for example, in the northern latitudes and in the cold season. The ESMs fail to fully reproduce the observed responses of sensible and latent heat fluxes, and the model bias depends on the model, region and season. The ESMs and observations even disagree on the sign of responses of sensible and latent heat fluxes in some cases. An attribution analysis further shows that biases in the simulated surface temperature response mainly result from biases related to the response of the surface energy partitioning. Biases related to the albedo response only play an important role under snow-covered conditions. Given these model biases, we highlight that when the CMIP6 models are used to investigate deforestation effects, the simulated result should be interpreted with caution. Moreover, the identified model deficiency shown here also has implications for model improvement.

Human activities have notably affected the Earth’s climate through greenhouse gases (GHG), aerosol, and land use/land cover change (LULCC). To investigate the impact of forest changes on regional climate under different shared socioeconomic pathways (SSPs), changes in surface air temperature and precipitation over China under low and medium/high radiative forcing scenarios from 2021 to 2099 are analyzed using multimodel climate simulations from the Coupled Model Intercomparison Project Phase 6 (CMIP6). Results show that the climate responses to forest changes are more significant under the low radiative forcing scenario. Deforestation would increase the mean, interannual variability, and the trend of surface air temperature under the low radiative forcing scenario, but it would decrease those indices under the medium/high radiative forcing scenario. The changes in temperature show significant spatial heterogeneity. For precipitation, under the low radiative forcing scenario, deforestation would lead to a significant increase in northern China and a significant decrease in southern China, and the effects are persistent in the near term (2021–40), middle term (2041–70), and long term (2071–99). In contrast, under the medium/high radiative forcing scenario, precipitation increases in the near term and long term over most parts of China, but it decreases in the middle term, especially in southern, northern, and northeast China. The magnitude of precipitation response to deforestation remains comparatively small.

Land surface processes and their coupling to the atmosphere are crucial in simulating the regional and global climate. Consequently, it is critical to identify and quantify uncertainty in these land surface model (LSM) processes in order to comprehend the land-atmosphere interactions. The heterogeneous characteristics, such as land use/land cover and soil texture, of the land surface contribute to the uncertainty associated with vegetation and soil parameters in LSM. They can alter the exchange of water and energy fluxes between land and atmosphere and thus impact the meteorological conditions at local and regional scales. In this study, we use the University of Torino land surface Process model for Interaction in the Atmosphere (UTOPIA), a one-dimensional land surface model that represents the interactions at the interface of the atmosphere, vegetation, and soil layers, to examine the sensitivity of land cover and soil texture in LSMs. We will investigate the impact of energy flux partition, soil moisture redistribution, and runoff generation due to land use/land cover and soil texture changes.

The accurate estimation of soil texture is crucial as it significantly impacts soil moisture and other hydrological variables. While the Weather Research and Forecasting Hydrological Extension (WRF-Hydro) model is a useful tool for investigating various aspects of hydrological processes and their interactions with the atmosphere, the default soil map provided by USGS and MODIS exhibits potential issues associated with coarse resolution and limited accuracy. To address these limitations, this study conducts a series of sensitivity experiments that consider additional data sources or alternative soil mapping approaches within WRF-Hydro model framework. A comparative analysis is performed by focusing on hydrological variables such as soil moisture and runoff. The study will enhance our understanding of how changes in soil properties influence key hydrological processes and shed light on the impact of diverse soil conditions on the robustness of simulation results. Acknowledgment This work was supported by Korea Environment Industry & Technology Institute(KEITI) through Water Management Program for Drought, funded by Korea Ministry of Environment(MOE)(2022003610003).

Land-atmosphere coupling serves as an important process that is associated with the variability of hydroclimate from local to regional scales given perturbed land surface conditions. This research was inspired by a potential geo-engineering project, commonly referred to as the Bradfield Scheme, proposed decades ago aiming at introducing a large water expanse into central Australia for the hydroclimatic changes that are favorable toward increased agricultural productivity. Based on idealized simulations by the community earth system model (CESM), we investigate how an inland lake perturbs the land-atmosphere interactions in central Australia and how the local hydroclimate responds to those perturbations. Moreover, the CESM employed in our research enables water isotopes and activates water tracers. We find that the idealized lake would strengthen the precipitation recycling process but fail to cause significant changes in total precipitation (e.g., the precipitation of water vapor originating from the lake region is shown to trivially contribute to total precipitation). The negligible impact on precipitation might be explained by the perturbed land-atmosphere interactions via the changes in the local thermodynamics and dynamics: the lake increases the latent heat flux through changing the surface energy budget, which corresponds to a significantly enhanced moisture flux into the overlying atmosphere; however, it also leads to significant evaporative cooling, creating strong divergence in the lower atmosphere and suppressing precipitation formation. Our research indicates that it is necessary to consider the interactions between land and atmosphere to better understand the underlying mechanisms of (either natural or anthropogenic) land use/land cover changes.

The intensifying concentration of the population in the metropolitan area of South Korea in recent times has led to the urban heat island (UHI) effect, promoting temperature increases in city centers and deteriorating the quality of life. In this study, we examined the impact of changing land use from urban areas to cropland in Seoul for 24 hours from 6 AM on July 26th, 2022, on the weather elements in downstream Namyangju using the WRF model. The results showed a shift in the convergence of winds from the northern part of Seoul towards Namyangju, particularly displaying strong convergence during the night. The temperature decrease in Namyangju was over 1.4℃ higher at night compared to daytime. During the day, temperatures dropped significantly in all directions from Seoul, except for the western part. However, at night, temperatures fell in a limited area to the east. Both Seoul and Namyangju experienced an increase in humidity, attributed to the evaporation of moisture from the cropland in Seoul. This humid atmospheric layer was generated around 1:30 PM and disappeared around 2 AM the following day. This study analyzed temperature changes, wind convergence, and humidity increase in downstream cities when alleviating Seoul's UHI effect. It can serve as crucial research material for predicting downstream city temperatures considering UHI and for planning satellite city constructions to mitigate the concentration of Seoul. 

The Maritime Continent (MC) has experienced severe deforestation in recent decades, leading to atmospheric destabilization. Here, we focus on the changes in extreme precipitation under the land use changes (LUC) and the global warming (GW) in MC. Our results suggested a “wet-get-wetter, dry-get-drier” paradigm for precipitation responses to MC LUC and GW, leading to an increased frequency of both extreme heavy and light precipitation events. However, the mechanisms show different mechanisms between LUC and GW in MC. LUC results in a warmer land surface, inducing an unstable atmospheric environment that affects local convection, manifesting as a negative effect of forest coverage on precipitation changes. As the GW, leading to the capacity of water vapor increases with rising temperatures in the atmosphere, resulting in the enhancement of precipitation. Our research found that LUC and GW cause more potent local effects, resulting in increased (decreased) precipitation in areas that already receive more (less) precipitation. Our study further explores how these two scenarios affect the intensity of extremely heavy and light precipitation in the MC. Atmospheric moisture budget analysis reveals that extreme precipitation events are associated with the dynamic component of vertical moisture advection after LUC or GW.

Marine low clouds are effective at cooling the Earth's surface by reflecting solar radiation. Existing research suggests that the uncertainty surrounding the cloud feedback in subtropical and extratropical regions is crucial to understanding the model spread in Equilibrium Climate Sensitivity (ECS). A recent study by Furtado et al. (2023) discovered that the seasonal patterns of clouds in the mid-latitude oceans of the Northern Hemisphere can provide insights into how clouds may respond to long-term surface warming. Jiang et al. (2023) introduced an additional constraint on ECS based on the seasonal fluctuations of extratropical marine low cloud fraction. Both Jiang et al. (2023) and Furtado et al. (2023) showed a correlation between the seasonal and long-term changes in the extratropical marine low cloud fraction per degree of surface warming. However, the underlying physical mechanisms behind the cloud fraction changes on these two time scales are yet to be fully established. In this study, we examine the contributions of the boundary layer inversion strength versus the extratropical cyclones to the extratropical marine low cloud fraction change on the seasonal and centennial time scales. By investigating these factors, this study aims to enhance our understanding of how the climate system responds to increasing greenhouse gases and improve the accuracy of climate change predictions.

High cloud cover is a crucial factor that regulates the energy balance and plays a significant role in temperature, precipitation, and the overall climate of the tropical region. However, the simulation of high cloud cover in tropical regions is subject to significant uncertainties, which can affect the accuracy of climate models and the prediction of future climate change. The study focuses on the simulation of high cloud cover and high cloud feedback in tropical regions using the Coupled Model Intercomparison Project Phase 6 (CMIP6) models. It is found that from 1984 to 2014, cirrus cloud cover decreases while cirrostratus and deep convection cloud cover increase, with no significant trend in total high cloud cover according to the International Satellite Cloud Climatology Project (ISCCP) data. The study also reveals that the CMIP6 models shows significant discrepancies in the simulating the spatial distribution and temporal variation of high cloud cover. The trend of high cloud cover in tropical regions mainly varies from east to west, with an increase in high cloud cover in the western Pacific and a decrease in the eastern Pacific, and vice versa. The time variation of simulated high clouds is considerably correlated with the sea surface temperature changes from 1980 to 2014. Future research will explore the interactions among large scale circulation, convection, radiation, and high/low cloud feedbacks and SST warming patterns in the models, which can improve our understanding of the complex high cloud feedback in tropical regions and enhance the accuracy of climate models.

Surface emissivity is essential for many remote-sensing applications including the retrieval of surface skin temperature from satellite-based infrared measurements, the determination of cloud detection thresholds, and the estimation of the surface longwave radiation emission, an important component of the energy budget of the surface-atmosphere interface. The CERES (Clouds and the Earth’s Radiant Energy System) Project is measuring broadband shortwave and longwave radiances and deriving cloud properties from the MODIS on Terra and Aqua and from the VIIRS on NOAA-19 and NOAA-20 orbiters to produce combined global radiation and cloud property data sets. Zhou et al. (IEEE Trans. Geosci. Remote Sens., 49, 2011) used Infrared Atmospheric Sounding Interferometer (IASI) data to create a high spectral resolution surface emissivity atlas for remote sensing and modeling applications. The IASI measures spectral radiances between 3.62 and 15.5 mm. The VIIRS I4 channel width is from 3.55 to 3.93 mm, while MODIS Band 20 is from 3.66 to 3.84 mm. Comparisons of top-of-atmosphere (TOA) radiance calculations with MODIS and VIIRS observations for these bands suggest that the IASI emissivity atlas near 3.7 µm may not be suitable for CERES cloud retrievals. In this paper, the IASI emissivities for the VIIRS and MODIS bands centered near 11mm are used to derive surface skin temperature from nighttime MODIS/VIIRS data. The Goddard Earth Observing System for Instrument Teams (GEOS-IT) numerical weather analyses provide temperature and water vapor profiles fused to correct the observed radiances for atmospheric absorption and emission. Global seasonal emissivity maps are then derived for the VIIRS and MODIS 3.7mm bands that are consistent with the derived skin temperatures and the observed TOA radiances. These seasonal climatology maps will be validated and used in CERES Edition 5 and other CERES-related cloud retrieval algorithms to provide improved clear-sky radiances and derived cloud properties.

The optical properties and bidirectional reflectance of the snowpack mixed with different concentrations of soot and dust are discussed in this study. Two mixing rules, Bruggeman and Maxwell-Garnett mixing rules, were considered and compared. It indicates that the extinction coefficient is less changed with the changing of soot and dust concentrations. The single-scattering albedo (asymmetry factor) decreases (increases) with increased concentrations of soot and dust. The optical properties of the snowpack are more sensitive to the change of soot concentration when the soot concentration is over 0.1 ppmw. From the change of optical properties, there is slightly stronger absorption using the Bruggeman mixing rule than using the Maxwell-Garnett mixing rule, while the relative differences between them are small. For semi-infinite snowpack, a reduction of the bidirectional reflection distribution function (BRDF) occurs with increasing concentrations of soot and dust, and the snowpack with heavy dust pollution can cause a larger BRDF reduction than that with heavy soot pollution. This phenomenon is more significant for large snow grain radii. For infinite snowpack, three cases were set to refer to observation. The reduction of BRDF is sharp in the case with a large concentration of dust at the snowpack surface.

We looked into the long-term variability of aerosols and clouds over Eastern China (EC) using MICAPS ground-based data from 1995 to 2019 and MODIS aerosol optical depth (AOD) retrieval from 2003 to 2019. The ground-based observed visibility (VIS) was used to characterize aerosol concentrations. Our study indicate that VIS shows a significant shift over EC in recent decades, as evidenced by falling (P1: 1995–2006)-flat (P2: 2007–2013)-rising (P3: 2014–2019) periods. Correspondingly, a change was observed in cloud fraction (CF) from a strong increase(P1:0.84%/year) to a slow fluctuation(P3:-0.02%/year) during the three periods over the Beijing-Tianjin-Hebei (BTH), but no significant variation over the Yangtze River Delta (YRD) region. In addition, the CF-VIS shows a negative correlation during both P1(BTH:-0.69%/km; YRD:-0.52%/km) and P3(BTH:-0.45%/km; YRD:-0.29%/km) periods while this relationship weakened during P3 (35% weaker in BTH and 44% weaker in YRD). However, it is important to note that this weakening may be influenced by a few factors, such as data processing methods, the location of aerosols relative to the cloud, relative humidity (RH), and large-scale circulation etc. Further studies, including numerical simulations and analysis, are required to determine how much change of cloudiness is attributed to aerosols.

Multiple parallel rainband (MPRB) is a type of organization of mesoscale convective systems. MPRBs are often accompanied by two different scales of train effects and thus often result in heavy precipitation. Our previous work showed that MPRBs in China have the highest frequency in the Beibu Gulf and its coastal areas and with more than half of them occurring in the mountainous areas along the western coast of the Beibu Gulf. This study tries to understand the formation mechanism of MPRBs in this highest frequency area of coastal Beibu Gulf based on a case study. Result shows that horizontal convective rolls (HCRs) played an important role in the formation of the MPRBs. The HCRs were generated by large vertical wind shear and inflection point instability over coastal area with a low-height mountain orientated almost perpendicular to the orientation of HCRs. Coastal mountains block water vapor from the southern sea and favors the formation of the HCRs and convection initiation. The interaction between the HCRs with the mountainous terrain led to the continuous initiation of convection in the upstream part of HCRs, and formed MPRBs through back-building process. Sensitivity experiments revealed that the height of the coastal terrain apparently affected the structure of MPRBs, but had little effect on the structure of the HCRs. As the height of the mountain range decreases, the MPRB structure keeps weakening or even disappears.

The results indicated that the BLCL in the Hetao area was formed by the combined influence of surface differences, boundary layer atmospheric circulation and complex terrain. And the BLCL mainly occurred at the border between the Hetao irrigation area and the Kubuqi Desert, and in a smaller arid area along the southeast bank of the Yellow River, with BLCL lengths ranging from 100 to 200 km. BLCL was a shallow system with convection heights of approximately 1000 to 1300 m. The daily variation of BLCL was significant, with a high incidence period from 12:00 to 17:00, accounting for 80%. The probability of BLCL formation in July and August of midsummer was as high as 60%, and 39% of BLCL could trigger convection. The prediction indicators about the formation of BLCL were as follows: (1) The temperature in Hangjinhouqi (53420) in Hetao Irrigation District is 2.5 ℃ lower than that in Habailaigeng (C3183) in Kubuqi Desert,and the relative humidity difference between them is up to 20% .(2) Continuous strong southerly winds was in the Kubuqi Desert and Mu Us Sandy Land areas in Hetao area, with a maximum wind speed of 10 meters above the ground greater than 4 ms ^{− 1}.(3) The sea level pressure field is high in the east and low in the west, and a dense isobaric zone was in the Jiziwan of Yellow River with 3-4 isobars. The formation of BLCL is of great significance to the convection triggering in this region, which is manifested in convection triggering, convection strengthening, and convection organization. The generation and convective triggering of BLCL in this specific region are closely related to the distribution characteristics of rapid increase in precipitation from west to east in the Hetao region.

Based on the radar data, this study explored the statistical characteristics of the convection initiations (CIs) during summers from 2012 to 2016 over the Hetao area in North China, where there is a sharp vegetation contrast between irrigation and desert. The durations of the most convections that triggered in the Hetao area are short and the development are weak. The high-incidence areas of CIs are deserts, mountains, and the edge of the Ordos Plateau. The convections over the desert are triggered around noon and last longer time, while the convections over mountains or at the edge of plateaus are mostly initiated at night and last for a shorter time. Some CIs are associated with boundary layer convergence lines (boundaries) caused by non-uniform underlying surfaces. Under different synoptic patterns, the initiation time of boundary-related convections are later than that of nonboundary-related convections, and the durations all longer. When the Hetao area is controlled by a weak high-pressure ridge at 500 hPa and there are southerly winds at low level, the long-duration boundary-related CIs are most likely to be triggered over the Kubuqi Desert.

Global warming is rapidly altering the dynamics of terrestrial vegetation, with consequences for regional climate dynamics, biodiversity, agricultural systems, and socio-economic challenges associated with land-use changes. Under the global warming, corresponding global greening trends have been observed in recent decades. Particularly, since the 2000s, vegetation growth has significantly increased in East Asia compared to other regions. To determine whether the recent strong greening trend in East Asia is a result of anthropogenic climate change, or a part of a natural oscillation on longer time scales than the observation period, a combination of satellite observations, tree-ring proxies, and earth system model simulations was used. Tree-ring based reconstruction reveals that the recent significant increase in vegetation in East Asia is a highly unprecedented phenomenon, with levels not observed in the past 200 years. In a warming climate, the enhanced Kuroshio Current and Western North Pacific Subtropical High in winter preceded by the growing season appear to amplify the greening trends in East Asia by inducing favorable warm and wet conditions for vegetation growth. Multiple model simulations from the Coupled Model Intercomparison Project Phase 6 (CMIP6) also depict the recent strong greening trend in East Asia as a manifestation of anthropogenic climate change and project further intensification in the 21st century.

This study aimed to predict on the rice harvest date in South Korea using one-month temperature forecasts. A global prediction data from the NOAA Climate Forecast System was downscaling using the Weather Research and Forecasting (WRF) model to obtain detailed predictive information. The WRF model utilized a double-nested modeling system, generating 5km gridded information centered on South Korea. Temperature predictive data over 11 years (2012-2022) were employed as input for a phenological model to predict the rice harvest date in advance. The seeding date was assumed to be January 1, and the reference temperature for harvesting was set at 1400°C + 55 days. The maximum and minimum temperature hindcasts tends to underestimate compared to observation. Due to the cold bias of temperature forecast data, the rice harvest date derived from hindcasts tend to be delayed compared to observation. This study highlights the potential of the one-month prediction system in obtaining high-resolution agricultural information. Funding Source: This work was supported by a grant (no. RS-2020-RD009438) from the Rural Development Administration, Republic of Korea. Reference: Hur, J, E-S Im, S Ha, Y-S Kim, E-S Kim, J Lee, S Jo, K-M Shim, M-G Kang, 2023: 1-month Prediction on Rice Harvest Date in South Korea Based on Dynamically Downscaled Temperature, Korean Journal of Agricultural and Forest Meteorology, 25(4), 267~275.

In this study, we analyzed the characteristics of the error in one-month temperature prediction data generated through joint development between the Rural Development Administration and the Hong Kong University of Science and Technology and performed corrections. For this purpose, we colleted hindcast data and weather observation data from 2013 to 2021, and environmental information on latitude, altitude, slope direction, slope, and Land cover index, and analyzed the characteristics of errors. In the case of maximum and minimum temperatures, the higher the elevation, the larger the forecast error. Various machine learning methods were used to correct errors in temperature prediction data, and on average, the RMSE of predicted data corrected with machin learning decreased by 0.475 (maximum temperature) and 0.400 (minimum temperature), respectively, compared to uncorrected predicted data. Through this study, it was found that errors in prediction data are affected by topographical conditions, and that machine learning methods can effectively improve errors by considering various environmental factors. This work was supported by a grant (no. RS-2020-RD009438) from the Rural Development Administration, Republic of Korea. Reference

The changes in rice climatic yield potential (CYP) across the Korean Peninsula are evaluated based on the new climate change scenario produced by the National Institute of Agricultural Sciences with 18 ensemble members at 1 km resolution under a Shared Socioeconomic Pathway (SSP) and Representative Concentration Pathways (RCP) emission scenarios. To overcome the data availability, we utilize solar radiation for CYP instead of sunshine duration which is relatively uncommon in the climate prediction field. The result show that maximum CYP(CYPmax) decreased, and the optimal heading date is progressively delayed under warmer temperature conditions compared to the current climate. This trend is particularly pronounced in the SSP5-85 scenario, indicating faster warming, except for the northeastern mountainous regions of North Korea. This shows the benefits of lower emission scenarios and pursuing more efforts to limit greenhouse gas emissions. On the other hand, the CYPmax shows a wide range of feasible futures, which shows inherent uncertainties in future climate projections and the risks when analyzing a single model or a small number of model results, highlighting the importance of the ensemble approach. This work was supported by a grant (no. RS-2021-RD009055) from the Rural Development Administration, Republic of Korea.

In April 2020, the 2m temperature over Siberia reached a historically unprecedented heatwave, while the Korean Peninsula experienced a cold spell. This study examines whether the operational model (GloSea6) for KMA accurately predicted the Siberian heatwave and Korean Peninsula cold spell in April 2020. Simultaneously, the impact of snow initialization on the sub-seasonal prediction skill of the model was investigated by prescribing different snow conditions for the Eurasian region. The snow experiments included three conditions: climate-prescribed snow state, snow state produced by the operational model's land surface component (JULES), and assimilated snow state using satellite data. In 2020, similar to the substantial decrease observed in snow amounts, the results generated by the land surface model and assimilation portrayed a dry condition compared to climatic values. In particular, the assimilated snow state exhibited the closest resemblance to the observations. Based on these findings, this study confirms that changes in snow state significantly influence the upper-level wave patterns and temperature variations in the Siberian region. The reduction of snow in the Eurasian region due to future climate change is expected to enhance land-atmosphere interaction in spring, playing a crucial role in the maintenance and strengthening of upper-level stationary waves in the Siberian region.※ This work was funded by the Korea Meteorological Administration Operational System Operation and Development for Climate Prediction Program under Grant KMA2018-00322.

The frequency and intensity of abnormal meteorological and climatic phenomena are increasing due to climate change. According to the abnormal climate report from the Korea Meteorological Administration, the number of high-temperature days in South Korea is showing an increasing trend along with the rise in global mean temperature. This trend especially showing a rapid increase since the 2010s. The increase in temperature and the number of high-temperature days can affect the growth and yield of crops, and can also affect the health of farmers engaged in agriculture. The National Institute of Agricultural Sciences of the Rural Development Administration has established the Pusan National University/Rural Development Administration (PNU/RDA) CGCM-WRF Chain by migrating the PNU CGCM-WRF Chain. The RDA produces 5km-resolution long-term prediction data using PNU/RDA-WRF Chain system to produce and provide weather and climate prediction for use in use agricultural part. Therefore, in this study, we evaluated the predictability of the number of summer high-temperature days in the South Korea over the past 30 years using 5km-resolution long-term prediction data. The correlation between predicted and observed of number of high-temperature days showed a statistically significant at the 99% confidence level, and an accuracy close to 80%.

In 2014, a Land-Atmosphere Modeling Package for supporting agricultural and forest management was developed at NCAM in South Korea. This modeling package is comprised of two components; one is the Weather Research and Forecasting modeling system (WRF) coupled with Noah-MultiParameterization options (Noah-MP) Land Surface Model (LSM), and the other is offline one-dimensional LSMs including JULES. Recently, the LAMP has been reinforced by adding FDDA, WRF-LES, and chemistry modules. The LAMP was initially started for weather and climate research in the agricultural and forestry sector within the university with government support, but has since expanded its support to local governments, private companies, and air force weather units. It can provide a dynamically and physically scientific tool to be easily applied for high-resolution NWP and making it easy to perform biophysical application and mechanism explanation on small-scale complex terrains. The results are visualized in one, two, and three dimensions using NCL and VAPOR and verified on a regular basis. It has artificial intelligence-based postprocessing methods to increase medium-range predictability. Some example applications using the LAMP are as follows: 1) to understand local wind circulations around observation sites; 2) to characterize agricultural drought in small-scale farmlands based on soil moisture (SM) deficit; 3) to predict snow depth and wetness to reduce the damage to agricultural facilities; 4) to analyze the effects of air temperature reduction by green and cool roofs in Seoul; 5) understand spatiotemporal variations in the atmospheric ventilation index; 6) to simulate rainfall and SM for landslide risk assessment; 7) to compute livestock heat stress in a mechanically ventilated broiler house; and 8) to analyze structure of low level jets near the origin of migratory insect pests. We plan to introduce U. S. National Water Model schemes for the process-level enhancement of terrestrial energy, water, carbon circulation analysis and prediction.

We study the IOP2 during Northern Coast Observation, Verification of Dynamics Experiment 2021 (NoCOVID21), and we focused on the details about the interaction between the typhoon (Choi-wan) and the mei-yu front, and the relationship between heavy rainfall over Taiwan and the southwesterly flow. On 3 Jun 2021, the period we defined as the pre-merging period 1, the Choi-wan typhoon and the mei-yu front were farther away with a relative dry region between them, and the two systems kept their characteristics. The atmosphere was warm and moist around the typhoon center, but it was baroclinic across the mei-yu frontal system with large temperature gradient. On 4 Jun, the pre-merging period 2, as the two systems were getting close gradually, the relative dry region was replaced by the moist airmass. The tropical cyclone weakened when its circulation was modified by Taiwan terrain, and the temperature gradient in the frontal region also weakened. On 5 Jun, the merging stage, the tropic cyclone moved to northeast of Taiwan, merging with the mei-yu front and transforming into a strong extratropical cyclone. The extratropical cyclone is characterized by stronger vorticity, circulation, ascent, total precipitable water and rainfall. After 2000 UTC 5 Jun, the extratropical cyclone weakened rapidly. During 5 Jun to 6 Jun, the prefrontal synoptic-related low-level jet (SLLJ) and the marine boundary layer jet (MBLJ) coexisted over southwest of Taiwan, and the MBLJ played an important role on moist transport. The warm and moist LLJs were blocked and lifted by Taiwan terrain, and the upper level divergence was present during 5 Jun to 6 Jun. As a result, the heavy rainfall event occurred over southwestern Taiwan.

Climate change has led to more frequent extreme weather events, posing challenges worldwide and in Taiwan, where advanced weather forecasting is crucial for disaster management and resource planning. For example, during drought, it is vital for authorities to assess the potential for mid-term drought relief within a 10-day period. This necessitates accurate and reliable mid-term precipitation forecasts. And these are urgently needed to address various challenges posed by climate changes, whether in agricultural water resources, water management, urban planning, or disaster risk and insurance planning. Focusing on 1-9-day precipitation forecast, this study developed a post-processing technique to calibrate and downscale the European Centre for Medium-Range Weather Forecasts (ECMWF) single deterministic forecast data (HRES), and generate calibrated probabilistic and single deterministic precipitation forecast. Forecast evaluation shows that generated probabilistic forecasts have superior performance in reliability, discrimination ability, and forecast skill, even for extreme precipitation events. Besides, the post-processed single deterministic forecast displays smaller-scale precipitation characteristic. In other words, more valuable meteorological forecast information can be obtained through the developed post-processing technique.

In recent years, the increasing frequency of extreme weather events has emphasized the critical need for accurate warnings of heat and cold events, substantially impacting agricultural practices in Taiwan. These events not only disrupt crop growth and development but also markedly influence crop quality and yield. Addressing this challenge, we have developed a statistical post-processing technique that integrates the Bayesian Processor of Output (BPO) and Logistic Regression methods. This approach aims to enhance the quality of extended-range probabilistic forecasts for consecutive days of heat and cold events in the agricultural region. Our evaluation results demonstrate the effectiveness of this methodology in bias correction and downscaling. Calibrated probabilistic forecasts using BPO exhibit better reliability, discrimination, and forecast skill for single-day heat and cold events compared to raw probabilistic forecasts. However, recognizing the interdependence of consecutive heat and cold events, direct multiplication of calibrated forecast probabilities for each consecutive day results in unreliable forecasts. To address this issue, Logistic Regression is applied in conjunction with BPO to adjust forecast probabilities for consecutive days of heat and cold events, leading to a substantial improvement in reliability, particularly for consecutive heat events. This innovative statistical post-processing technique provides a more reliable and skillful prediction for consecutive days of heat and cold events, enabling more effective preparation and mitigation strategies for the agricultural sector in response to changing environmental conditions in Taiwan.

The raindrop size distribution (DSD) information is useful to delineate the microphysical characteristics of precipitation and their underlying processes. Furthermore, the RSDs aid in improving the quantitative precipitation estimation (QPE) algorithms and cloud modeling simulations. The RSDs were found to exhibit disparities with geography, season and precipitation type. Here, we used long-terms observations of ground-based disdrometers over north Taiwan to investigate the seasonal and spatial disparities in RSD characteristics. The results demonstrated dominant seasonal variations in the RSDs over the regional differences. Seasonal disparities and underlying microphysical attributes are delineated using remote sensing and re-analysis data sets. The radar reflectivity and rainfall rate (Z-R) relations, which are essential in QPE are established for each season are established. Furthermore, the empirical relations between any two parameters of the Gamma distribution (N(D) =N₀D^μ e^-ΛD; where D is the raindrop diameter in mm, N(D) is the number distribution function in m⁻³ mm⁻¹, N₀ is the intercept parameter in m⁻³ mm^−1−μ, μ is the shape parameter, Λ is the slope parameter (mm⁻¹) ) are also appraised; which helps in improving the microphysics parameterizations. Apart from the traditional way of estimating the RSD relations, here we machine learning approach to establish the RSD relations. 

This study investigates the environmental conditions associated with the frequency of tropical cyclone (TC) occurrences (TCOF) and their impacts on the TC’s rain microphysics over the North Indian Ocean (NIO) throughout the major monsoon stages (pre-monsoon, monsoon, and post-monsoon) from 2014 to 2021. The Arabian Sea (AS) and the Bay of Bengal (BOB) areas are given emphases within the NIO domain. High TCOF is attributed to the combined presence of warm SST and cyclonic wind or positive 850-hPa relative vorticity. During the monsoon season, TCs exhibit limited intensification as a result of a pronounced deep layer wind shear. Using case studies, results show notable differences in the precipitation pattern across the major monsoon stages. Although heightened reflectivity and substantial precipitation are consistently identified within 50km radius from the storm’s center in all monsoon stages, TC Vayu (2019) during monsoon stage stands out with the most intense rainfall and highest reflectivity within the inner band region. This is followed by TC Fani (2019) during pre-monsoon and lastly by TC Kyaar (2019) during post-monsoon. The heightened reflectivity and substantial precipitation observed during the monsoon season are apparently attributed to the abundant presence of relative humidity and cloud liquid water content throughout the monsoon months from June to September.

Heavy to extremely heavy rainfall events are the major source of flash floods, landslides and agricultural damage. An increase in heavy rainfall events, more particularly in between May to September months, over Taiwan necessitate for the detailed investigation. The present study is aimed to investigate the spatial and temporal variations in the heavy rainfall events over Taiwan. Long-term data sets from the ground-based rain gauges, disdrometers, airborne radars (TRMM/GPM DPR) are used to investigate the rainfall and microphysical attributes of heavy rainfall events. The results showed higher occurrence frequency of heavy rainfall events over central Taiwan than the rest of the island. The contour frequency by altitude diagram of rainfall and raindrop size distribution estimates from the GPM DPR data products revealed contrasts in the microphysical features of heavy rainfall events across Taiwan. Apart from this, Modern-Era Retrospective analysis for Research and Applications, Versions2 (MERRA-2), re-analysis. Moderate Resolution imaging spectroradiometer (MODIS) and re-analysis data sets are also used to explore the influence of aerosol-cloud interactions on heavy rainfall events over Taiwan. 

The variability of the drop size distribution (DSD) caused by microphysical processes is one of the reasons for poor quantitative precipitation forecasts (QPF) under bulk microphysics schemes. Therefore, it is necessary to understand the actual microphysical process variations in different weather systems. After analyzing the long-term characteristics of DSD, the data in Taiwan indicated that the microphysics processes of afternoon convection in summer exhibit higher variability compared to other weather systems, particularly under warm rain processes. On the other hand, this study attempts to quantitatively analyze the collision effects during the warm rain process using the DSD function. The collision rate (CR) and probability of collision (Pr_collision) are effective in categorizing DSD features and predicting collision process fingerprints. In contrast, after comparing with the radar variables, the current bulk microphysics schemes make it difficult to effectively understand the warm rain collision process because of the limitation of the DSD shape. Furthermore, this study performs a synthetic analysis of the observed afternoon convection data during the 2022 TAHOPE/PRECIP experiment to understand the variation of the warm rain collision process under different convection scenarios. The results show that in addition to the collisional aggregation process, the importance of collisional fragmentation on convective evolution has been underestimated.

High temporal resolution aerosol optical depth (AOD) observations derived from new-generation geostationary (GEO) satellites possess unique advantages. Unfortunately, the AOD products obtained by GEO satellites are limited to specific region. In this study we use polar orbit satellite data to expand the spatial range of GEO-AOD and propose a multi-source AOD data fusion method based on machine learning. We fully explore the complementary information from multi-source satellite observations and reanalysis data, and generate hourly full-coverage AOD maps in China for an entire day. The verification results indicate fused AOD dataset matches well with the ground AOD measurement. Also, the temporal and spatial AOD variation trends can be reconstructed based on this method, which consistent with the ground air quality monitoring stations observations. The fused dataset substantially enhances the availability of GEO-AOD data, and it has far-reaching significance for the evaluation of air pollution in the near surface regions of China.

Canada imposed new methane regulations on upstream oil and gas sector in 2020 aiming to achieve a 40-45% reduction in methane emissions by 2025, and a 75% reduction by 2030. In 2021 and 2022, Environmental Defense Fund (EDF) conducted multiscale (facility level and regional level) aerial mass balance measurement campaigns in Alberta to assess the actual methane emissions from upstream oil gas sector. In partnership with Scientific Aviation, we completed aerial mass-balance flights around eleven geographically distinct oil and gas production regions within Alberta and compared to the local emission inventories estimated from a model developed by EDF following the same methodologies used by Environment and Climate Change Canada for national inventory reporting. Our analysis revealed that for most regions, measured emissions were ~1.5 to 2 times higher than the inventory. We conducted measurements at 16 upstream O&G facilities in Alberta in 2021, and the measurements revealed that emissions were, on average, 1.7 (SD: 0.6) times higher than the reported emissions for the same year. On a subsequent campaign in 2022, we focused on understudied O&G sectors covering 24 midstream and end-use facilities. These sites were found to be emitting, on average, 4.0 (SD: 1.1) times more CH4 than reported. This work highlights the need for direct measurements to verify modelled inventories and regulation-driven reductions. 

Considering the higher global warming potential of Methane (CH₄) and low residence time in comparison with the other greenhouse gases (GHGs) such as carbon dioxide, the mitigation efforts for minimizing the adverse effect of climate change has recently given increased focus on reducing methane emissions. Accurate estimation of the regional GHG emissions is crucial towards addressing the climate change scenario. In the Indian context, major contributions of methane emissions come from domestic ruminants, fossil fuels, waste, rice agriculture, wetland emissions, etc. In the present study, we use column averaged dry-air mixing ratio of methane (XCH₄) from TROPOMI (Tropospheric Monitoring Instrument) on the ESA Copernicus Sentinel-5 Precursor satellite to infer India’s CH₄ emissions. We attempt to assess the potential of these retrievals to quantify methane emission hotspots and to improve the emission estimates over the Indian sub-continent. Along with the satellite observations, the Eulerian atmospheric transport model Weather Research Forecast (WRF) coupled with Chemistry (WRF-Chem-GHG) has been used for the forward transport simulations of atmospheric methane. Also, the results from an inversion study by using these satellite observations will be presented.

The coal energy sector in China, including Shanxi and the Golden Energy Triangle, is marked by numerous coal mines and industries focusing on coal use. Both due to co-emissions as well as over time in the atmosphere in-situ, these particles interact with various low-volatility atmospheric components and form coated structures in which the absorbing aerosol (black carbon and coal dust) forms core, while refractory species (sulfate, nitrate, water, etc.) form shell. This study conducted field observations of surface BC concentration, size distribution, and AOD in coal-mining areas. A Mie model based on a “core-shell” configuration is used with all individual waveband observations, to form a solution space and uncertainty bounds of the per-particle optical information (SSA, ABS, EXT). These values are then used to respectively invert the column number and mass loading. Firstly, fine-mode black carbon aerosols are shown to exhibit a trimodal lognormal fine-mode size distribution, differing from previous studies. These results align with the fact that local sources include a very diverse set of sources including high temperature combustion, low-temperature combustion, coal to chemicals, coal dust, and coal transportation. Secondly, the absorbing aerosol demonstrates strong absorption of solar radiation in the near ultraviolet and blue bands, leading to a situation in which the typical use of aethalometer observations in the typical 880nm band used to obtain BC, yields a significantly low SSA bias. Thirdly, the SSA distribution characteristics of the absorbing aerosols do not entirely conform to the power-law relationship of AAE, resulting in deviations when estimating SSA across all size bins. This finding is consistent with additional bias with respect to many models, which tend to underestimate aerosol mixing and aging. Finally, analyzing the radiative effects and properties from the different radiative bands and size distributions provides detailed information on aerosol physical and optical properties.

At a global scale, biogenic volatile organic compounds (BVOCs) comprise a major fraction of total non-methane VOCs, significant impacting ecosystem functioning and climate change. Literature has consistently demonstrated that the presence of anthropogenic pollutants, such as NO_x, can perturb the formation of biogenic secondary organic aerosol (BSOA). During nighttime, BVOCs primarily undergo oxidation by O₃ and NO₃. The latter pathway leads to the formation of nitrogen-containing organic aerosol (NOA), which plays a vital role in air quality and human health. Current research mainly focuses on NOA formation from individual BVOC precursor. However, in the real atmosphere, multi-reactant oxidation process occurs simultaneously, leading to complex interactions and non-linear effects on NOA formation. The underlying mechanisms of the non-linear formation of NOA from BVOC mixtures remain largely unexplored. This study aims to investigate the formation of NOA resulting from the oxidation of monoterpenes (e.g., limonene) and sesquiterpenes (e.g., longifolene), and elucidate the non-linear formation mechanism, properties, and characteristics of NOA in the multi-precursor system. Three key aspects will be examined: (1) the yields and physicochemical characteristics (e.g., functional groups, volatility, light-absorbing properties) of NOA produced from both individual BVOC system and BVOC mixtures; (2) the interactions between oxidation products and intermediates derived from the oxidation of various BVOCs; (3) the non-linear formation mechanisms of NOA in the presence of BVOC mixtures. This study will provide a comprehensive understanding of NOA formation in the multi-precursor system and their implications for atmospheric chemistry.

Extreme weather becomes more frequent with global warming. The combined extreme weather of high temperature and drought has an impact on air quality and human health, etc. The vegetation-atmosphere feedback during the combined heat wave and drought is also an important natural process affecting the generation of O3. Understanding the trend of heat-drought combined extreme weather and its impact on ozone and BVOC can help mitigate its negative impacts on the environment and human health. Using ERA5 data and major heat wave and drought indicators, the trend of heat wave and heat-drought combined extreme weather in summer in China during 1960-2022 was analyzed. The effects of temperature and soil moisture on isoprene emissions were investigated using Station for Observing Regional Processes of the Earth System (SORPES) observation data. Since the 1960s, the average number of heat wave days, frequency, cumulative intensity and the max duration of heat wave and heat-drought combined extreme weather have been on the rise. The cumulative intensity of heat-drought combined extreme weather is about 1℃ higher than that of heat wave per year. Isoprene emissions increased with increasing temperature, but the relationship between isoprene emissions and soil moisture was complicated. Isoprene emissions increase during mild and moderate droughts; During severe droughts, isoprene emissions decrease. Under the influence of high-temperature drought combined weather, isoprene emissions may increase, which will exacerbate ozone pollution.

Not only anthropogenic emissions but also biogenic VOC emissions contribute to the concentration of ambient air pollutants. According to the latest “National Forest Resource Survey” from the Forestry and Nature Conservation Agency, Ministry of Agriculture of Taiwan, the total forest area of Taiwan is 2,197,090 hectares, with a forest coverage rate of approximately 60.71%. Therefore, biogenic VOC emissions might play an essential role in air pollutant production in Taiwan. The Model of Emissions of Gases and Aerosols from Nature (MEGAN) provides biogenic emissions by considering the plant functional type (PFT), leaf area index (LAI), and meteorological condition. Since the initial PFT data was from a relatively distant year with a different spatial distribution compared with the observation, this study utilized the land use index to reproduce the PFT dataset and investigated its impact on air quality modeling. WRF version 3.8.1 and CMAQ version 5.2 were conducted to simulate the meteorological field and concentration of air pollutants, respectively, from September 27 to October 10, 2021. The discussion focused on October 5 and 6, when the mountain blocked the synoptic easterly wind, causing the weak wind and severe air pollution in western Taiwan. The simulation with updated PFT data shows lower and higher VOC concentrations (about 20-30 ppbC) compared with the original simulation in western Taiwan's plain region and the central mountain area, respectively. The update of the PFT data is able to improve the overestimation of daytime ozone concentrations in mountainous areas.

Compound drought and heatwaves (DHW) events have much attention due to their notable impacts on socio-ecological systems. However, studies on the mechanisms of DHW related to land-atmosphere interaction are not still fully understood in regional aspects. Here, we investigate drastic increases in DHW from 1980 to 2019 over northern East Asia, one of the strong land-atmosphere interaction regions. Heatwaves occurring in severely dry conditions have increased after the late 1990s, suggesting that the heatwaves in northern East Asia are highly likely to be compound heatwaves closely related to drought. Moreover, the soil moisture–temperature coupling strength increased in regions with strong increases in DHW through phase transitions of both temperature and heat anomalies that determine the coupling strength. As the soil moisture decreases, the probability density of low evapotranspiration increases through evaporative heat absorption. This leads to increase evaporative stress and eventually amplify DHW since the late 1990s. In particular, seasonal changes in soil moisture and evapotranspiration between spring and summer contributed to the amplification of DHW by enhancing land-atmosphere interaction.

This study analyzed the long-term change in the interannual variation of the East Asian summer (July-August) precipitation pattern represented by the Meiyu-Baiu-Changma. The July and August GPCP monthly precipitation data from 1979-2022 were divided into six clusters by using Self-Organizing Map (SOM). Two features were noteworthy: First, the temporal distribution of the clusters showed a shift between clusters, indicating a decadal change in precipitation patterns over time. The influence of the Okhotsk Sea air mass and tropical monsoon air mass in the former period, and the North Pacific air mass in the latter period, on the formation of the East Asian summer precipitation pattern was evident. In the latter period, the westward extension and intensification of the WNPSH breaks down the original triple precipitation pattern and affects precipitation in Northeast Asia.

Climate change significantly affects the frequency, intensity, and timing of both mean and extreme precipitation. Extreme precipitation has caused tremendous socio-economic losses and severe impacts on human life, livelihoods, and ecosystems. In South Korea, heavy rainfall events during the boreal summer (June-to-August) have become more frequent and sporadic in recent years. Given the severity of these events, there is an urgent need for investigation into summer extreme rainfall. While previous studies have addressed daily extreme precipitation, there is still a gap in understanding hourly extreme rainfall. Therefore, this study aims to fill that gap by examining trends, spatio-temporal variability, and long-term variations in mean and extreme precipitation over South Korea during the boreal summertime. We utilized daily and hourly observational data and employed various analysis methods. Over the past 50 years (1973–2022), there has been a noticeable increase in maximum hourly precipitation, even though the boreal summer mean precipitation has seen only marginal growth. Regionally, an increase in mean and extreme rainfall occurred in the northern part of the central region and the southern coast of the Korean peninsula. Furthermore, the rise in the intensity and frequency of extreme precipitation, along with an increase in dry days, has contributed significantly to the total summer precipitation in recent years. Our findings provide scientific insights into the progression of extreme summer precipitation events in South Korea.

Recently, extreme precipitation events in the Asian region have been anticipated to occur more frequently and intensely in the future due to the increase in greenhouse gas concentrations. However, extreme precipitation exhibits different features depending on surface characteristics, and different dynamics. Because most studies on extreme precipitation in the Asian region have been centered around the broad monsoon region, understanding of dynamic-thermodynamic characteristics of extreme precipitation in specific areas is lacking. In this study, we identified specific regions in terms of extreme precipitation quantification to investigate the regional driving factors of extreme precipitation in the Asian monsoon region during the period from 1979 to 2022 (June, July, and August). Through the moisture budget equation, we examined the dynamics, thermodynamic characteristics, and contributions of extreme precipitation. Based on frequency, quantity, and intensity of extreme precipitation, the EASM region was divided into the Yangtze River, Korean Peninsula, and Japan Southern regions, while the SASM region was divided into Indian Central-North, Himalayan Mountains, Bay of Bengal, Indian Northwest, and West Ghat regions. In the dynamic and thermodynamic analysis, extreme precipitation in all specific regions was found to be predominantly influenced by the dynamic convergence term (DY CON). However, unlike the South Asia region where extreme precipitation was mainly dominated by the DY CON, the East Asia region showed high contributions from the thermodynamic term and the convergence of the nonlinear term. Among them, the KP region had the highest contribution of 21% from the thermodynamic advection term, emphasizing regional differences in the characteristics of extreme precipitation. The results imply that understanding the precipitation characteristics of specific regions can contribute to the comprehension of extreme precipitation prediction.

In previous studies, it has been demonstrated that the remote forcing associated with climate teleconnections, such as El Niño-Southern Oscillation (ENSO) and the Indian Ocean Dipole (IOD), have a significant impact on wildfires in Indonesia. However, there still exists considerable diversity in wildfire activity among El Niño events, especially with the highest values observed in the major season from August to October. To identify the role of inter-basin coupling, the accelerated termination of IOD and the dominant influence of Indian Basin Wide Mode (IOBM) during the El Niño winter, we categorized cases into four types (i.e., IOD-ENSO-IOBM, (IOD)-ENSO-IOBM, IOD-ENSO-(IOBM), (IOD)-ENSO-(IOBM). The results show that the first (third) type exhibits more vigorous carbon emissions compared to the second (fourth) type. We found a notable difference between the first (second) type, associated with the strong phase of IOBM, and the third (fourth) type, associated with the weak or negative phase of IOBM.

The East Asian monsoon system (EAM) shows strong annual cycle with the wet season in summer and dry in winter. Apart from the annual cycle, the monsoon system exhibits distinct climatological intraseasonal oscillations (CISOs) (Wang and Xu 1997), also known as the fast annual cycle (LinHo and Wang 2002). This study focuses on investigating the relationship between the CISOs, South China Sea (SCS) summer monsoon onset, and extreme rainfall events during the spring to summer transition period (March-June) in the monsoon region of 5°S-35°N, 60°E-150°E. The CISOs are identified based on the harmonic analysis of the OLR data in the 44-year period of 1979-2022. A significant CISO is identified over the SCS as an oscillation with the positive (dry) peak in early May and negative (wet) valley in late May. The timing of the valley point coincides with the dates of the commencement of the SCS summer monsoon. The spring to summer transition of the EAM is further analyzed using the ERA5 daily 850-hPa wind data. We found that the March-July daily wind maps can be classified to five types (weather type, WT). The WT occurrence frequency can well represent the sub-seasonal characteristics of the transition. The ENSO impacts on the EAM can be seen in the interannual variations of the WT frequency. The time windows with high occurrence probability of extreme rainfall events will be presented in the context of multiple-scale climate features such as the CISO, transient ISOs, WTs and ENSO. The implication for the S2S forecasts of the extreme events will be discussed.