Abstract:Extreme weather and climate events are exacerbated by global warming,leading to increasingly severe losses from weather-related disasters.Despite progress,theories and methods of climate system prediction still face numerous challenges,and the precision of climate predictions remains inadequate for societal needs.To address these issues and enhance the scientific rigor and accuracy of climate prediction,the “Center for Climate System Prediction Research,” jointly led by Nanjing University of Information Science and Technology and Sun Yat-sen University,receive support from the National Natural Science Foundation of China (2021-01—2025-12).Over the initial three years of the project,the research team conducted extensive and systematic studies,resulting in several significant advances:(1) unveiling key changes,driving forces,and physical mechanisms of the climate system;(2) investigating the impacts of air-sea-land-ice interactions on extreme weather and climate events in China;(3) making substantial progress in developing numerical models for the climate system and integrating prediction systems;(4) advancing prediction theories and methods for the climate system across extended-range,subseasonal to seasonal (S2S),and decadal time scales.This study offers a brief overview of these advancements and identifies key scientific questions for future exploration,including climate and environmental change attribution,bridging paleoclimate with present climate studies,understanding climate system variability and extremes across time and space,the role of artificial intelligence in climate science,decadal prediction,and risk response systems.

Abstract:The East Asian winter monsoon (EAWM) system is one of the most active circulation systems during winter in the Northern Hemisphere.It exerts a major impact on the near-surface temperature and precipitation in East Asia through interactions with the Siberian high,Aleutian low (AL),East Asian trough,and East Asian westerly jet stream.Despite the importance of the relationship between the AL and East Asian winter monsoon,there have been relatively few studies on it and its variations.In this study,an EOF analysis is performed to analyze the intensity change,north-south and east-west movement,and Northwest-Southeast propagation modes of the AL,and to depict the relationships among these four modes and the northern and southern patterns of the EAWM.The EOF analysis is also used to extract the main modes of near-surface temperature.The explanatory variance of the first mode is 48.9％,which shows that the near-surface temperature is uniformly distributed throughout the country,and that the temperature decreases rapidly to the north along the meridional gradient.This pattern is called the northern EAWM mode.The second mode accounts for 16.2％ of the total variance,showing a reverse distribution from north to south,which is called the southern EAWM monsoon.The results show that,although the intensity variation of the AL and the two East Asian winter monsoon modes are insignificant,the meridional movement of the AL is strongly correlated with the southern EAWM mode,with a feature of decadal variation.Considering that the interdecadal variation of AL and EAWM will affect their correlative features,the time series of the four modes and two types of winter monsoon are studied by 21-point sliding correlation.From 1964 to 1994,there was a significant positive correlation between the north-south movement of the AL and the southern EAWM mode.After 1994,the axis position of the north-south oscillation over the North Pacific moved eastward under the influence of ENSO,and the configurations of the AL and Siberian high have not been conducive to the formation of the pressure gradient in southeastern China.In this study,we take 1978 as a typical case of southern EAWM mode,and obtain the distribution patterns of the north-south movement AL and southern winter monsoon.The Northwest-Southeast propagation mode is strongly correlated with the northern EAWM pattern,and the decadal change of this relationship is mainly affected by the North Atlantic Oscillation.There is no significant correlation between the propagation mode and northern EAWM monsoon during the period of 1961—1974.After 1974,the North Atlantic Oscillation stimulates anomalous positive potential height anomalies in the central Siberian region,expands the influence extent of the Northwest-Southeast propagation mode,and brings northeasterly anomalies to the middle and higher latitudes.We also take 1976 as a case study of northern EAWM monsoon,and the study results are consistent with our predictions.In this paper,the relationship between the intensity change and east-west movement of AL pressure and their winter winds is not very significant,thus we have not carried out further analysis on this issue.In future research,we will continue to focus on the influence of four-mode interannual changes in the AL on winter monsoon,so as to supplement the results of this paper.

Abstract:Both the Siberian high and the East Asian trough play crucial roles in the East Asian winter climate system.Changes in the strength and position of the Siberian high can lead to the movement of cold or warm air masses over the East Asian continent,while variations in the East Asian trough can influence circulation patterns in the western Pacific.These combined effects can alter temperature and precipitation distributions in East Asia during the winter,significantly impacting the winter climate.This study defines the Siberian High (SH) index,the East Asian Trough (ET) index,and a Meridional Wind (V index) index between the high and low-pressure systems based on climatological characteristics.The relationships between these indices are explored using power spectrum analysis and linear regressions.The study investigates the significant periods of variability and the impact mechanisms of the Siberian high and East Asian trough on winter temperatures in East Asia.Additionally,a simple linear regression model is constructed using a non-filtering method and cross-validation for extended-range forecasts of intra-seasonal temperatures in southern China during the winter.The main conclusions are as follows:The most significant periods of the SH and ET indices occur on intra-seasonal timescales,accounting for 60％ of the total standard deviation.Power spectrum analysis results show that the energy peaks of the SH and ET indices are concentrated in the 10—50-day period,while the V index is more concentrated in the 10—40-day period.On the intra-seasonal timescales,the correlation coefficients between the SH/ET indices and the V index are 0.69 and -0.73,respectively (through a 99％ confidence test).The reconstructed meridional wind fields related to the SH/ET indices are used to calculate the ratio of the standard deviation compared to the actual meridional wind field.Quantitative results show that the contributions of the intra-seasonal SH and ET indices to the V index are 82.6％ and 42.2％,respectively.Lead-lag regression analysis for the SH,ET,and V indices shows similar regression patterns,with slow eastward propagation of northwest-southeast distributing wave trains in the 500 hPa geopotential height field,consistent configurations of the low-level 850 hPa moisture,near-surface 925 hPa circulations,precipitation,and 2 m temperature fields.The three indices all correspond to a northwest-southeast-oriented low-frequency Rossby wave train in the middle and higher troposphere,which propagates eastward or southeastward.The upper-level circulation fields are well linked to low-level water vapor,precipitation,and surface air temperature fields.When the Siberian high deepens or the East Asian trough develops,northerly winds are enhanced over the key East Asian region,facilitating the transport of dry and cold air from high latitudes in Siberia to East Asia,resulting in dry and cold weather conditions.Based on comprehensive regression results,the V index and the intra-seasonal component of 2 m temperature during winter in southern China are selected as the predictors and predictands.A linear regression model is built for extended-range forecasting,and the forecast performance is evaluated by calculating the time correlation coefficient and standardized root mean square error.Cross-validation and independent forecast experiments indicate that reliable forecasts of the intra-seasonal 2 m temperature in the region with a lead time of 25 days can be achieved.

Abstract:Northeast China experienced an unprecedented sequence of continuous heavy snowfall days in November 2013,the most significant event since 1982—2020.This event was characterized by two intense snowfall processes occurring from the 17th to the 20th and the 25th.The first process had a longer duration,while the second exhibited greater snowfall intensity.This study investigates the causes and predictability of these events from the perspective of the anomalous climatic background of November 2013 and the detailed dynamics of the two intense snowfall processes.The results show that the positive phase of the Arctic Oscillation (AO),the negative phase of the North Pacific Oscillation-like (NPO-like) pattern,increased sea ice growth north of the Barents Sea (November compared to September),and anomalously warm sea surface temperatures in the tropical-southern Indian Ocean during November were responsible for the persistent heavy snowfall event in Northeast China.The increased sea ice growth suggested a heightened release of latent heat flux into the atmosphere,resulting in higher temperatures that favored the strengthening of the positive phase of the AO and the excitation of Rossby wave trains,subsequently weakening the Aleutian low.Moreover,anomalously warm sea surface temperatures facilitated enhanced convection over the tropical Indian Ocean and weakened convection over the tropical western Pacific,leading to a 'cyclone-anticyclone’ circulation anomaly in the Northwest Pacific-Aleutian region,presenting a negative phase of the NPO-like pattern.These atmospheric circulations favored the transport of water vapor from the North Pacific.Additionally,the causes of the two intense snowfalls were analyzed.From the 12th to the 16th,in the five days preceding the first intense snowfall (17th to 20th),the North Atlantic Oscillation (NAO) maintained a continuous positive phase,triggering the persistent eastward propagation of Rossby wave trains.This condition corresponded to a meridional 'cyclone-anticyclone’ anomaly over the Northwest Pacific-Aleutian region,facilitating continuous moisture transport from the North Pacific to Northeast China.During the second intense snowfall on November 25,2013,the Ural blocking high intensified significantly,and the deepening of the Northeast low vortex promoted the transport of warmer and moister air from the tropical western Pacific to the Northeast region,augmenting daily snowfall intensity.Finally,the prediction skill of the anomalous climatic background in November 2013 was evaluated using CFSv2.While CFSv2 effectively predicted anomalously warm sea surface temperatures in the tropical-southern Indian Ocean one month in advance,its skill for predicting anomalous convection in the tropical Indian Ocean and western Pacific,the NAO,and tropical-midlatitude teleconnections was relatively limited.On the subseasonal scale,ECMWF (CMA) reasonably predicted spatial distributions of snowfall for two processes 29 (12) and 13 (16) days in advance,respectively.This prediction skill could be attributed to better anticipation of the daily variations of key circulation systems such as the NAO and the Ural blocking high.Therefore,future efforts should concentrate on enhancing the subseasonal-to-seasonal predictive skills of tropical-midlatitude teleconnections,moisture transport,and the stratospheric polar vortex.

Abstract:An accurate understanding of Earth system change mechanisms and predicting their impacts relies on the development of Earth system models.Compared to global models,Regional Earth System Models (RESMs) concentrate more on medium-to small-scale processes and their regional impacts within specific areas,featuring higher spatial resolution and more detailed physical processes.RESMs enable coupled simulations of multi-layered Earth interactions,thereby enhancing the ability to reproduce,analyze,and forecast extreme climate events.Consequently,the development and application of RESMs hold significant scientific and practical importance for addressing various climate change-related challenges and assisting in prediction and decision-making across multiple fields,including disaster prevention and mitigation,water resources management,agriculture,energy,environmental conservation,and resource exploitation.The concept of RESMs was initially proposed by Giorgi around 1995.Over the past three decades,their development has primarily followed two approaches.The first approach,known as independent development,involves coupling regional weather/climate models with specialized models tailored to specific application goals.This method aims to broaden the application scope of regional models in specific fields based on reliable atmospheric,land surface,and oceanic descriptions.Typically,specialized models are directly coupled with regional weather/climate models,resulting in relatively simple model structures and limited functions.Representative models include WRF-Hydro,PFWRF,WRF-HMS,RegCM-FVCOM,CWRF-FVCOM,WRF-Crop,WRF-CMAQ,and WRF-Chem.The second approach,holistic integration,seeks to construct a comprehensive model of coupled multi-sphere processes for digital twin regional Earth systems.This approach aims to create a unified and coordinated framework that emphasizes deep integration among various models.Such an approach not only requires technical compatibility among the models but also demands theoretical and methodological innovations to better simulate and understand the complex dynamics and interactions within Earth systems.Representative models of this approach include RegCM-ES,TerrSysMP,ROM,and R-CESM.Irrespective of the development approach,RESMs exhibit the following common characteristics:(i) Multi-layer coupling:RESMs provide a more detailed representation and online coupling of land surface and ocean processes compared to regional weather/climate models.They integrate biogeochemical,hydrological,human activity,and atmospheric chemistry processes,enabling a comprehensive understanding and simulation of the dynamic relationships among various Earth system components.Through the use of couplers,RESMs achieve flux coupling and interactions across different spatiotemporal scales,thereby enhancing the precision of simulations of natural cycles and the impacts of human activities on these cycles.(ii) Higher spatial resolution:RESMs can simulate small-scale processes,explicitly representing atmospheric convection,boundary layer processes,oceanic mesoscale eddies,complex vegetation structures on land surfaces,and changes in land use.These capabilities lead to more accurate simulations and predictions of extreme weather and climate events and their effects on local environments.(iii) Integration of data assimilation:The initial state in RESMs involves multiple processes across different layers,and the initial value of any variable can influence the entire model.Assimilating observational data from multiple sources and layers into the model’s initial state not only reduces the initial errors of related processes but also minimizes error propagation throughout the entire system,thereby shortening the model’s spin-up time and enhancing simulation accuracy.In light of the overview provided,this study advocates for the integration of interdisciplinary research efforts through open-source collaboration to expedite the development of RESMs in China.There is an urgent need to conduct interdisciplinary research utilizing the newly established model,with a particular focus on interactions among multi-layer and multi-scale processes.Additionally,efforts should be directed towards establishing a regional digital twin platform for monitoring and early warning based on high-resolution RESMs.Such platform could play a crucial role in disaster prevention and mitigation in critical regions and support vital decision-making processes.

Abstract:Numerical prediction models often encounter difficulties in predicting the intensity changes of tropical cyclones (TCs),particularly in the case of super typhoons and the progression of TC intensification.This study aims to investigate the large-scale environmental factors influencing the generation and development of super typhoons in the western North Pacific (WNP).We analyze the effects of thermodynamic and dynamic environmental factors on the intensity of TCs with tracks similar to Super Typhoon Haiyan (2013),the strongest TC in the past century in the WNP.Typhoons with tracks resembling Haiyan (2013) are categorized into two groups:super TCs and regular TCs.The Lanczos method is utilized to filter out synoptic-scale disturbances and typhoon disturbances,and mean composite differences are used to compare the composite environmental factors of the two groups.Observational analysis results indicate that,compared to dynamic factors,thermodynamic factors play a more significant role in strengthening TCs,such as higher temperatures and more abundant vapor in the lower troposphere,along with higher relative humidity in the middle troposphere.This configuration facilitates the release of a large amount of latent heat,which is beneficial for ascending motion.Meanwhile,low-level Ekman pumping enhances TC strength through positive feedback.To quantitatively assess the relative importance of environmental conditions in regulating TC intensity,a box difference index (BDI) analysis method is used to rank the environmental factors.Accordingly,925 hPa moist static energy (MSE),950 hPa specific humidity,and 900 hPa temperature are selected as the top-3 predictors for estimating TC intensity.Furthermore,we calculate the values of ocean heat content (OHC) and TC moving speed for each case,which are found to be larger for super TCs compared to regular TCs and may contribute to an increase in TC intensity.As the typhoon moves faster,the cold water upwelling caused by the typhoon leads to a smaller cooling effect,allowing the OHC to increase and facilitating typhoon intensity strengthening.Additionally,OHC and TC moving speed can also serve as additional predictors.Lastly,idealized numerical model experiments with the Weather Research and Forecasting (WRF) model are conducted to reveal the relative importance of environmental temperature and moisture vertical profiles in affecting TC intensity.The control experiment,identified as CTRL,and a set of sensitivity experiments,identified as Axel_SH,Axel_T,and Axel_SH+T,respectively,are designed.The sensitivity experiments Axel_SH,Axel_T,and Axel_SH+T indicate that the vertical profile of the area-averaged specific humidity,temperature,and specific humidity plus temperature of Typhoon Alex is replaced by that of Tyhpoon Haiyan.The sensitivity experiment results indicate that the relative contribution of environmental moisture and temperature profiles is 1∶4,suggesting that the environmental static stability parameter is the most important factor regulating super TC formation.Overall,the study confirms the significant impact of thermodynamic factors on typhoon intensity with tracks similar to Typhoon Haiyan (2013),providing quantitative insights into the contributions of different environmental factors.Given the challenges in predicting TC intensification,numerical modeling efforts should focus on understanding the interaction with the entire upper-ocean column and improving physical parameterization.

Abstract:With population growth and socio-economic development, the use of fossil fuels not only impacts the environmental but also highlights its finite nature. Consequently, the quest for environmentally friendly and sustainable alternative energy solutions has become urgent. Offshore wind, as an emerging energy source, offers a continuous power supply for China. However, the instability of wind energy’s interannual variations can lead to insufficient energy supply for the wind power industry, emphasizing the importance of examining and predicting these variations.In this study, we employed the Time Series K-means clustering method to categorize winter interannual variations of wind energy along the Chinese coast into four regions: the North China Sea (NCS), East China Sea (ECS), Northern South China Sea (NSCS), and Southern South China Sea (SSCS). Subsequently, regression analysis was used to explore the relationship between interannual variations in regional wind energy and large-scale circulation anomalies. We found that interannual variations in NCS are related to the Arctic Oscillation (AO)-related cyclones (Anticyclones) in Northeast China, while those in ECS are associated with the central-type El Niño-Southern Oscillation (ENSO) and the Siberian High. Wind power in both SSCS and NSCS is influenced by the eastern-type ENSO-related Philippines cyclones (Anticyclones), with the north-south position of the continental high-pressure system also affecting their interannual variations; when the continental high-pressure system shifts northward (southward), NSCS (SSCS) is mainly affected.Considering the relatively high predictability of climate modes, we evaluated the predictive skill of wind energy along the Chinese coast using five climate models. Regarding climatology predictions, CMCC and JMA overestimate wind energy in the southern sea, contrasting with underestimations from NUIST and DWD. SEAS5 aligns closely with ERA5. Conversely, in the northern sea, all models except SEAS5 tend to overestimate wind energy. In terms of root mean square error (RMSE) in predictive skill, significant deviations are observed among various models for regions abundant in wind energy resources such as the Taiwan Strait, the Luzon Strait, and areas west of the Nansha Islands. CMCC exhibits the largest prediction error of wind energy resources in the Chinese Sea, while the SEAS5 model demonstrates the smallest prediction error. Concerning predictions of interannual variations, climate models show higher predictive skill for the wind energy index in the South China Sea, reflecting the models' strong predictive skill for ENSO. However, for northern regions, current climate models face challenges in predicting the influences of climate modes on wind power.This paper elucidates the relationship between the interannual variations of winter wind energy along the Chinese coast and large-scale circulation anomalies caused by climate factors. However, it does not delve into the underlying mechanism of this association from the perspective of atmospheric dynamics. Further investigation is needed to explore this intrinsic mechanism. Additionally, the interannual variations of wind energy in other seasons along the Chinese coast and their influencing factors also merit further exploration.

Abstract:The newly released ERA5 reanalysis dataset by the ECMWF provides hourly precipitation reanalysis values,adding another reference for global hourly precipitation research.However,current evaluations of ERA5 hourly precipitation are limited,focusing more on monthly and daily precipitation assessments or hourly precipitation within restricted areas.Few studies address global-scale assessments of hourly precipitation.Motivated by this gap,we assess the global hourly precipitation frequency of ERA5.Sparse station-observed hourly precipitation data,particularly over open oceanic and remote inland areas,limits its usefulness for global-scale assessment of ERA5 reanalysis precipitation data,despite its presumed accuracy over satellite estimates or precipitation reanalysis data.Consequently,we employ four satellite datasets—GPM,GSMaP-M,GSMaP-G,and CMORPH—to evaluate hourly precipitation frequency of ERA5 in this study.We analyze the spatial distribution of climatological annual total precipitation and precipitation frequency in ERA5 and the four satellite datasets.Additionally,we examine the error sources of hourly precipitation frequency of ERA5 by evenly dividing precipitation events from the 5 datasets into ten intensity bins based on the hourly precipitation intensity percentiles.We then compare precipitation frequency differences among different regions,including tropical and extra-tropical areas,as well as marine and land areas.The main findings are as follows:1) the magnitude and spatial pattern of climatological annual total precipitation of ERA5 close align with the four satellite datasets,indicating ERA5 effectively represents total precipitation.2) The climatological hourly precipitation frequency of ERA5 is 2-3 times higher than that of satellite data,suggesting a systematic error in the hourly precipitation frequency of ERA5 compared to satellite estimates,or an overestimation of average precipitation frequency of ERA5.3) The overall overestimation of hourly precipitation frequency of ERA5 mainly stems from an abnormal overestimation of moderate and light precipitation frequencies.Notably,ERA5's global average for the lightest hourly precipitation (0—10％) is approximately 6 times higher than that observed in satellite estimations.4) ERA5 underestimates the frequency of extreme hourly precipitation compared to satellite estimations.5) In addition,oceanic overestimation is generally greater than that on land,with slightly stronger overestimation in the tropical region compared to the extratropical region.Although ERA5 systematically overestimates hourly precipitation frequency,its spatial pattern remains close to satellite estimates.The potential impact of this systematic overestimation in ERA5 on related research warrants further evaluation.

Abstract:Exploring the potential predictability of seasonal precipitation in Eastern China beyond the preceding winter El Nino-Southern Oscillation (ENSO) has long been a significant challenge.This study investigates the influences of the non-canonical Atlantic Niño (NCA) on precipitation patterns in Eastern China during the early and late rainy seasons using observational and reanalysis data from 1979 to 2020.Results indicate that the NCA leads to increased precipitation in southern China during the early rainy season,while it decreases precipitation along the southern coast but increases precipitation in northern China during the late rainy season.This is because,in the early rainy season,warm sea surface temperature (SST) anomalies associated with the NCA in the North Tropical Atlantic trigger La Niña through the “wind-evaporation-SST” feedback,inducing an anomalous anticyclone over the western North Pacific.The southwesterly flow along the northwest flank of this anomalous anticyclone transports warm and moist air into southern China,resulting in increased precipitation.Conversely,during the late rainy season,NCA-related warm SST anomalies migrate southward to the equatorial Atlantic,modifying the Walker circulation,reinforcing La Niña,and intensifying the anomalous anticyclone over the western North Pacific.This enhanced anticyclone covers the southern coastal region of China,reducing precipitation there,while further northward moisture transport along the enhanced anticyclone increases precipitation in northern China.Additionally,NCA influences Eastern China’s precipitation by stimulating different mid-high-latitude Rossby waves in the early and late rainy seasons.In the early rainy season,the Rossby wave is primarily triggered by anomalous warm SST in the North Tropical Atlantic,propagating from northern Africa to the Qinghai-Tibet Plateau and eventually reaching the Yangtze River basin.Although the SST anomalies in the Tropical North Atlantic weaken during the late rainy seasons,the North Atlantic Oscillation can still strengthen through local air-sea coupling,thereby developing SST in the North Atlantic.The Rossby wave,induced by an anomalous SST in the North Atlantic,across the Eastern European Plain,the Siberian Plain,and Lake Baikal,propagates southeastward to northern China,forming an anticyclonic anomaly in the upper troposphere and facilitating precipitation in northern China.This study reveals distinct seasonal variations in NCA-related precipitation anomalies in Eastern China,highlighting the complexity of large-scale circulation responses to the NCA.It provides a vital scientific basis for improving precipitation forecasting and early warning systems in China.Further investigation is needed to determine the extent to which numerical models can simulate NCA and its teleconnections.

Abstract:The frequency of extreme high temperature events has increased against the backdrop of global warming,posing serious risks to natural ecosystems,socio-economic development,and human safety.The Eurasian mid-high latitudes,or core regions of the Belt and Road area,feature fragile ecological environments highly susceptible to climate change,with limited adaptive capacities to extreme weather events.In recent decades,the frequent occurrence of extreme high-temperature events in these latitudes has resulted in tens of thousands of fatalities and billions of dollars in economic losses.Accurate prediction of extreme high temperatures in this region,especially on a decadal scale,is urgently needed by governmental decision-makers to effectively address climate change and promote sustainable development.This paper assesses the decadal predictive skill of current state-of-the-art dynamical models (CMIP6 DCPP) for summer extreme high temperatures in the Eurasian mid-high latitude region.We utilize the anomaly correlation coefficient (ACC) to assess the model's skill in capturing the observed variability phase and the mean-square skill score (MSSS) as a deterministic verification metric sensitive to amplitude errors.By comparing DCPP hindcasts (initialization) with historical simulations (external forcing),we examine the sources of predictive skill.The evaluation results show that multi-model ensemble average (MME) exhibits high predictive skill for the region south of 60°N (South Eurasia,SEA),accurately forecasting its linear growth trend and prominent decadal variability during 1968—2008.However,MME shows almost no predictive skill for the decadal variability of extreme high temperatures in the North Eurasia (NEA) region,only forecasting a linear growth trend lower than observed.To improve decadal predictive skills,we developed a three-layer recurrent neural network (RNN).This model utilizes the large-sample model predictions of 86 initial fields as input,with training and testing periods of 1968—2007 and 2008—2022,respectively.Significant improvements in extreme high temperature skills in NEA and SEA during test period of 2008—2020 were observed in the RNN model.The ACC skills of NEA and SEA in RNN are 0.86 and 0.83,respectively,compared to-0.61 and-0.03 in MME.Meanwhile,the MSSSs of NEA and SEA in RNN are 0.37 and 0.52,whereas they are-1.1 and-0.94 in MME,respectively.Real-time forecasts from RNN indicate that extreme high temperatures in the SEA region will continue to rise from 2021 to 2026,with a record-breaking event in 2026.Meanwhile,the NEA region is predicted to experience anomalously fewer events in 2022,followed by fluctuating increases.A comparison of the performance of various input sizes in RNN reveals that large sample sizes are necessary for the RNN model.Additionally,incorporating additional predictors with significant physical mechanisms for extreme high-temperature events may further enhance decadal prediction skills,warranting further investigation.Nevertheless,this study provides new insights into current decadal prediction of extreme climate,offering promising scientific support for governmental decision-makers in addressing climate change.

Abstract:Beyond conventional weather forecasts and seasonal climate predictions,a continuum of prediction across all timescales,known as weather-climate seamless prediction,has garnered considerable attention and progressed over the last decades.However,among predictions on various timescales,subseasonal predictions,bridging weather forecasts and short-term climate predictions still face challenges.This timescale typically extends beyond two weeks but falls short of a season,where the influence of initial conditions has faded away while the forcing from boundary conditions remains insignificant.Consequently,understanding the sources of subseasonal predictability and achieving effective subseasonal predictions encounter significant challenges.As reported in numerous previous studies,tropical intraseasonal oscillations (ISO),comprising the Madden-Julian Oscillation (MJO) and Boreal Summer Intraseasonal Oscillation (BSISO),can provide dominant sources of global subseasonal-to-seasonal (S2S) predictability.Based on the subseasonal climate forecast system of Nanjing University of Information Science and Technology (NUIST CFS1.1),the atmospheric initialization of individual members and ensemble strategy are slightly modified toward an upgraded version named NUIST CFS1.1 Pro,which consists of nine members and saves computational costs.Furthermore,using the real-time multivariate MJO index and two BSISO indices,BSISO1 and BSISO2,the prediction skills of tropical ISOs during different seasons are evaluated.The results show that skillful prediction (ACC>0.5) for MJO,BSISO1,and BSISO2 can extend to 26,17,and 12 lead days,respectively,and for strong events (amplitude> 1),it can be extended to 30,21,and 13 lead days,respectively.In predicting these tropical ISOs,the NUIST CFS1.1 Pro outperforms two newly-developed subseasonal forecast systems in China (i.e.,BCC_CSM2 and FGOALS-f2).Moreover,it achieves competitive performances compared to eight major operational prediction systems participating in the international S2S project,with a relatively leading level in predicting winter MJO and summer BSISO1,as well as a medium level in predicting BSISO2.As for the winter MJO and BSISO1 predictions,the target phases 2,3,6,and 7 display higher skills than the other four phases.Further analysis indicates that the NUIST CFS1.1 Pro can accurately capture the eastward propagation of the winter MJO at lead times of 21—25 days.Additionally,it partly predicts the MJO-related 2 m temperature anomalies in China,particularly the cold anomalies during phases 2 and 3.In summer,the NUIST CFS1.1 Pro well predicts the northward and northwestward propagation of BSISO1 at lead times of 16—20 days,especially anomalous convection and low-level circulation over the northwestern Pacific.This leads to successful predictions of the spatial pattern of precipitation anomalies in East China associated with BSISO1.However,the predictions of NUIST CFS1.1 Pro on these time scales severely underestimate tropical ISO signals and their impacts on air temperature and precipitation over China,warranting further efforts for improvement.In particular,there is ample room for the NUIST CFS1.1 to improve in the prediction of MJO teleconnections over China during the winter.For instance,the warm anomalies over most of China in phases 6 and 7 of the MJO cannot be successfully predicted,whereas the prediction of tropical convection and circulation displays good skill.

Abstract:In recent years,the occurrence of compound humid-heat weather has been increasing worldwide due to global warming.These events predominantly affect subtropical coastal areas,posing significant threats to the environment,economy,and various other aspects.Eastern China,in particular,is highly susceptible to such extreme,compound humid-heat events.Neglecting the influence of humidity on high-temperature weather could lead to a serious underestimation of the associated hazard level.Given China's vast territory,the distribution and variation of humid-heat events can vary across different regions.In eastern China,a substantial population is exposed to perilous humid-heat conditions.Therefore,it is crucial to study the changing characteristics of humid-heat events in China.This study aims to explore the spatial-temporal changing patterns of compound humid-heat events across China from 1961 to 2020 using daily maximum wet-bulb temperature (WBT).WBT was calculated from daily relative humidity,daily maximum temperature,and daily mean pressure.Results indicated that:(1) Both the mean and maximal daily maximum WBT in China exhibit a similar spatial pattern,with warmer temperatures in the south and cooler temperatures in the north.The regions of southern China and the Sichuan Basin emerge as hotspots for maximal daily WBT.While the mean daily maximum WBT shows an upward trend,the maximal daily maximum WBT does not exhibit a significant trend.The mean daily maximum WBT generally exhibits a 2—6 year period,whereas the maximal daily maximum WBT has relatively shorter periods.Meanwhile,changes in mean daily maximal WBT and maximal daily maximum WBT vary across different regions of China.Both the mean daily maximum WBT and the maximal daily maximum WBT show a decreasing trend in eastern Xinjiang,while they both increase rapidly in eastern Northwest China.The mean daily maximum WBT increases slightly in southern China and eastern Northwest China,while the maximal daily maximum WBT increases in northeastern China and decreases in southern China.(2) The distribution of extreme humid-heat thresholds in China is similar to that of the daily maximum WBT.The intensity of extreme humid-heat events shows an increasing trend,while the frequency of such events increases at a rate of 0.098 d/a.The intensity of extreme humid-heat events has increased the most in eastern Northwest China,but it has shown a decreasing trend in southern China.However,the frequency of extreme humid-heat events has increased in most areas of China over the past 60 years.(3) The threshold distribution of the extreme events with return periods of 10 (20) years in China is also similar to that of the maximum daily maximum WBT.Such events mainly occur in the Sichuan Basin,with a significant increase in the northwest and eastern regions of China and a decreasing trend in southern China.The measurement of relative humidity was changed to automatic methods in the early 21st century,and relative humidity decreased abnormally during the same period,which may be related to a sudden drop in eastern Xinjiang.Therefore,the homogenizing process for relative humidity data is necessary in future research.The findings of this study have significant implications for climate change policies.Improving our understanding of changes in humid-heat events in China can help us better respond to these climate-related disasters.

Abstract:This study elucidates the spatiotemporal characteristic of June—July mean Meiyu rainfall over the middle and lower reaches of the Yangtze River basin(27°—33°N,108°—123°E) using Chinese monthly gauge precipitation data and global atmospheric reanalysis datasets from 1961 to 2000.Three physically meaningful precursors play pivotal roles in enhancing Meiyu rainfall during June and July.First,positive sea level pressure anomalies over the subtropical western Pacific (SWP) during April—May strengthened the western North Pacific subtropical high by exciting Kelvin wave responses and enhancing Walker circulation.This phenomenon facilitates moisture transport from the tropics to the Yangtze River via southerly winds.The mechanism underlying SWP’s impact on Meiyu highlights the persistent influence of atmosphere-ocean interaction over the Indo-Pacific basin from spring to summer.Second,the negative tendency of sea level pressure over the North Atlantic from March to May (NAP) reflects the influence of North Atlantic Oscillation (NAO)-related mid-latitude wave trains on Meiyu.From spring to early summer,the evolution of NAO-related wave trains across Eurasia strengthens the Northeast Asian cyclone and enhances Meiyu rainfall.Third,the cooling tendency of surface temperature over East Siberian from January to April (EST) is closely associated with the extratropical westerly jet by amplifying the temperature gradient between the tropics and polar regions.This condition favors the maintenance of meridional circulation over East Asia and enhances Meiyu rainfall.The aforementioned mechanisms have been verified in corresponding numerical experiments based on a linear baroclinic model.Consequently,a physically-based empirical (PE) model based on these three predictors exhibited significant prediction skills,with a temporal correlation coefficient (TCC) of 0.79 and 0.77 and a mean square skill score (RMSE) of 0.59 and 0.68 during the training period (1961—2000) and independent forecast period (2001—2022),respectively.For comparison,the partial least squares (PLS) regression method and five machine learning methods (Random Forest,LightGBM,Adaboost,Catboost,and XGboost) are employed to conduct seasonal predication of Meiyu based on the same potential precursors.Although the PLS model and five machine learning models exhibit prefect hindcast skills (TCCs of LightGBM,Catboost,and XGboost all being 1.00) during the training period,their skills diminish dramatically in the independent forecast period of 2001—2022 (with the maximum TCC being 0.43 and the minimum RMSE being 0.94),indicating a significant overfitting problem.Hence,the PE model based on physically meaningful precursors demonstrates superior and stable independent prediction skills in Meiyu rainfall forecasts.The findings of this study underscore the advantages of the PE model and emphasize caution in the use of machine learning methods in climate prediction.Additionally,the comparison of multiple methods for seasonal prediction of Meiyu in this study provides practical scientific references for operational departments engaged in seasonal climate prediction.

Abstract:The Tibetan Plateau,often referred to as the “Roof of the World” and the “Third Pole”,is of considerable importance due to its high altitude,vast scale,and complex terrain,rendering it a pivotal element in global climate dynamics.In the last five decades,the plateau has witnessed a pronounced warming trend,with temperatures increasing at a rate twice that of the global average.Precise forecasting of future climate change in this region is paramount for various sectors,including agriculture,ecosystems,and socio-economic development.This study employs data from an experiment involving 18 models in the CMIP6 model,wherein the CO₂ concentration suddenly quadruples (abrupt-4×CO₂),to investigate the response of the Tibetan Plateau to greenhouse gas forcing.Specifically,the study focuses on feedback processes using the climate feedback response analysis method (CFRAM).The findings reveal that surface warming on the plateau is primarily driven by greenhouse gas forcing and positive water vapor feedback,further amplified by albedo and cloud feedback.Processes such as surface heat storage,sensible heat,and latent heat play roles in moderating temperature increases.Cloud feedback emerges as a significant source of uncertainty in plateau warming response,followed by albedo and water vapor feedbacks,while sensible and latent heat processes aid in mitigating this uncertainty.Variations in projected warming,particularly in central-eastern and southern regions of the plateau,stem from inter-model differences in surface heat storage and atmospheric dynamics.Enhanced parameterization to surface albedo and cloud cover is identified as an effective strategy to alleviate spatial uncertainty in model predictions of regional warming across the Tibetan Plateau.The spatial distribution of uncertainty in feedback processes varies,with maximum standard deviations observed in different regions for each process,corresponding to areas projected to experience significant warming.In summary,although greenhouse gas forcing models generally exhibit consistent trends across the Tibetan Plateau,variations in feedback processes and regional dynamics highlight the necessity for enhanced parameterization and resolution in climate models to improve predictions in this pivotal region.

Abstract:A systematic diagnostic study was conducted to examine the characteristics and causes of the hollow typhoon “Mulan”,which formed in the South China Sea in August 2022.The study applied on-site observation precipitation data,global analysis data,and merged precipitation data from multiple sources.The results indicated that typhoon “Mulan”,originating from a monsoon depression in the South China Sea,exhibited typical features of depressions in the region.Satellite observations revealed that “Mulan” lacked typical typhoon characteristics,with no deep convection development near its center.Strong convection and rainstorms were primarily distributed in the typhoon’s periphery,with precipitation significantly higher there than near the center.In terms of atmospheric circulation,“Mulan” featured multiple smaller vortices within its early-stage cyclonic circulation,with strong winds mainly concentrated in the periphery.Therefore,typhoon “Mulan” exhibited characteristics of a hollow typhoon.Despite being weak,“Mulan” brought strong winds and heavy rainfall to regions of South China,including Guangdong,Guangxi,and southern Yunnan,due to a combination of factors including a northeastern low-level jet,ample water vaper supply from the South China Sea,and intense convective activity over land.The strong winds on the northeast side of the typhoon resulted from the convergence of the southwest monsoon from the South China Sea and the southeast monsoon from the northwest Pacific Ocean.Blocked by the zonal subtropical high over the mid-latitudes of the Asian continent,the typhoon’s direction of movement changed from northwest to westward as it approached land,following an inverted parabolic path shape.Using the Weather Research and Forecasting (WRF) model,a retrospective simulation of 84 hours was conducted,employing the nesting of two domains with horizontal grid spacings of 9 and 3 km.The model reasonably reproduced the circulation structure and evolution process of "Mulan",although with some discrepancies,particularly in the simulated monsoon trough and typhoon track.Comparison with Final Operational Global Analysis (FNL) data showed significant improvement in precipitation simulation with the WRF model,particularly at higher resolution.This study contributes to understanding the formation and characteristics of hollow typhoons in the South China Sea and highlight the potential of mesoscale models for enhancing typhoon simulation and forecasting.Future research will focus on the cloud microphysical characteristics of “Mulan” and evaluate different parameterization schemes of the WRF model to enhance the forecasting ability for weak typhoons with heavy precipitation.