Advancing Regional Climate Predictability Through ML-enabled Dynamical System Approach

Jian Lu,Zhangshuan Hou,Yi Deng

DOI: https://doi.org/10.2172/1769654

2021-01-01

Abstract:than 80% of the global burned area, and emit ~1.4 PgC y-1 into the atmosphere together with other dust and aerosols that strongly affect regional climate.

What problem does this paper attempt to address?

Current Progress in Subseasonal-to-decadal Prediction Based on Machine Learning

Zixiong Shen,Qiming Sun,Xinyu Lu,Fenghua Ling,Yue Li,Jiye Wu,Jing-Jia Luo,Chaoxia Yuan

2024-01-01

Applied Computing and Geosciences

Abstract:The application of machine learning (ML) techniques to climate science has received significant attention, particularly in the field of climate predictions, ranging from sub-seasonal to decadal time scales. This paper reviews recent progress of ML techniques employed in climate phenomena prediction and the enhancement of dynamic forecast models, which provide valuable insights into the great potentials of ML techniques to improve climate prediction capabilities with reduced computational time and resource consumption. This paper also discusses several major challenges in the application of ML to climate prediction, including the scarcity of datasets, physical inconsistency, and lack of model transparency and interpretability. Additionally, this paper sheds light on how climate change impacts ML model training and prediction, and explores three key areas with potential breakthroughs: large-scale climate models, knowledge discovery driven by ML, and hybrid dynamical-statistical models, underscoring the important role of the integration of “ML and dynamical models” in building a bridge between the artificial intelligence and climate science.
Global Fire Modelling and Control Attributions Based on the Ensemble Machine Learning and Satellite Observations

Yulong Zhang,Jiafu Mao,Daniel M. Ricciuto,Mingzhou Jin,Yan Yu,Xiaoying Shi,Stan Wullschleger,Rongyun Tang,Jicheng Liu

DOI: https://doi.org/10.1016/j.srs.2023.100088

2023-01-01

Science of Remote Sensing

Abstract:Contemporary fire dynamics is one of the most complex and least understood land surface phenomena. Global fire controls related to climate, vegetation, and anthropogenic activity are usually intertwined, and difficult to disentangle in a quantitative way. Here, we leveraged an ensemble of five machine learning (ML) models and multiple satellite-based observations to conduct global fire modeling for three fire metrics (burned area, fire number, and fire size), and quantified driving mechanisms underlying annual fire changes in a spatially resolved manner for the period 2003-2019. Ensemble learning is a meta-approach that combines multiple ML predictions to improve accuracy, robustness, and generalization performance. We found that the optimized ensemble ML well reproduced annual dynamics of global burned area (R2 = 0.90, P < 0.001), total fire numbers (R2 = 0.86, P < 0.001), and averaged fire size (R2 = 0.70, P < 0.001). Additionally, the ensemble ML captured key spatial patterns of multi-year mean magnitudes, annual variabilities, anomalies, and trends for different fire metrics. Our ML-based fire attributions further highlighted the dominant role of enhanced anthropogenic activity in reducing global burned area (-1.9 Mha/yr, P < 0.01), followed by climate control (-1.3 Mha/yr, P < 0.01) and insignificant positive vegetation control (0.4 Mha/yr, P = 0.60). Spatially, climate dominated a much larger burned area (53.7%) than human (23.4%) or vegetation control (22.9%); however, the counteracting effects from regional wetting and drying trends weakened the net climate impacts on global burned area. The fire number and fire size exhibited similar spatial control patterns with burned area; globally, however, fire number tended to be more affected by climate while fire size more influenced by human activities. Overall, our study confirmed the feasibility and efficiency of ensemble ML in global fire modeling and subsequent control attributions, providing a better understanding of contemporary fire regimes and contributing to robust fire projections in a changing environment.
Recent Advances in China on the Predictability of Weather and Climate

Wansuo Duan,Lichao Yang,Mu,Bin Wang,Xueshun Shen,Zhiyong Meng,Ruiqiang Ding

DOI: https://doi.org/10.1007/s00376-023-2334-0

IF: 5.8

2023-01-01

Advances in Atmospheric Sciences

Abstract:This article summarizes the progress made in predictability studies of weather and climate in recent years in China, with a main focus on advances in methods to study error growth dynamics and reduce uncertainties in the forecasting of weather and climate. Specifically, it covers (a) advances in methods to study weather and climate predictability dynamics, especially those in nonlinear optimal perturbation methods associated with initial errors and model errors and their applications to ensemble forecasting and target observations, (b) new data assimilation algorithms for initialization of predictions and novel assimilation approaches to neutralize the combined effects of initial and model errors for weather and climate, (c) applications of new statistical approaches to climate predictions, and (d) studies on meso- to small-scale weather system predictability dynamics. Some of the major frontiers and challenges remaining in predictability studies are addressed in this context.
A Dynamical Pathway Bridging African Biomass Burning and Asian Summer Monsoon

Cao Dianbin,Park Tae-Won,Song Yongjia,Wang Yuhang,You Zhenyu,Lin Yanluan,Deng Yi

DOI: https://doi.org/10.1007/s00382-021-05788-8

IF: 4.901

2021-01-01

Climate Dynamics

Abstract:The Asian summer monsoon (ASM) affects more than one-third of the world’s population due to its close connection with floods, droughts thus water resources in densely populated Asian countries. The effects of aerosols emitted in remote regions on the ASM, in contrast to local emissions, remain largely unclear. Here we demonstrate through a hierarchy of climate models that aerosol emissions from the central African wildfires could strengthen the circulation of the ASM (South Asian Monsoon in particular), increase precipitation over South Asia and reduce precipitation immediately north and south of it. The enhanced latent heating over South Asia provides a critical positive feedback to the initial strengthening of monsoon westerlies associated with wildfire-driven anomalous Rossby wave source. The atmospheric dynamical bridge discovered here effectively connects African biomass burning with hydroclimate variability over East Asia in boreal summer and offers a new source of monsoon predictability across a range of timescales.
AttentionFire_v1.0: interpretable machine learning fire model for burned-area predictions over tropics

Zhu, Qing

DOI: https://doi.org/10.5194/gmd-16-869-2023

IF: 5.1

2023-02-03

Geoscientific Model Development

Abstract:African and South American (ASA) wildfires account for more than 70 % of global burned areas and have strong connection to local climate for sub-seasonal to seasonal wildfire dynamics. However, representation of the wildfire–climate relationship remains challenging due to spatiotemporally heterogenous responses of wildfires to climate variability and human influences. Here, we developed an interpretable machine learning (ML) fire model (AttentionFire_v1.0) to resolve the complex controls of climate and human activities on burned areas and to better predict burned areas over ASA regions. Our ML fire model substantially improved predictability of burned areas for both spatial and temporal dynamics compared with five commonly used machine learning models. More importantly, the model revealed strong time-lagged control from climate wetness on the burned areas. The model also predicted that, under a high-emission future climate scenario, the recently observed declines in burned area will reverse in South America in the near future due to climate changes. Our study provides a reliable and interpretable fire model and highlights the importance of lagged wildfire–climate relationships in historical and future predictions.

geosciences, multidisciplinary
Quantifying The Drivers And Predictability Of Seasonal Changes In African Fire

Yan Yu,Jiafu Mao,Peter E. Thornton,Michael Notaro,Stan D. Wullschleger,Xiaoying Shi,Forrest M. Hoffman,Yaoping Wang

DOI: https://doi.org/10.1038/s41467-020-16692-w

IF: 16.6

2020-01-01

Nature Communications

Abstract:Africa contains some of the most vulnerable ecosystems to fires. Successful seasonal prediction of fire activity over these fire-prone regions remains a challenge and relies heavily on in-depth understanding of various driving mechanisms underlying fire evolution. Here, we assess the seasonal environmental drivers and predictability of African fire using the analytical framework of Stepwise Generalized Equilibrium Feedback Assessment (SGEFA) and machine learning techniques (MLTs). The impacts of sea-surface temperature, soil moisture, and leaf area index are quantified and found to dominate the fire seasonal variability by regulating regional burning condition and fuel supply. Compared with previously-identified atmospheric and socioeconomic predictors, these slowly evolving oceanic and terrestrial predictors are further identified to determine the seasonal predictability of fire activity in Africa. Our combined SGEFA-MLT approach achieves skillful prediction of African fire one month in advance and can be generalized to provide seasonal estimates of regional and global fire risk. Fire is an important component of many African ecosystems, but prediction of fire activity is challenging. Here, the authors use a statistical framework to assess the seasonal environmental drivers of African fire, which allow for a better prediction of fire activity.
Improving predictions of land-atmosphere interactions based on a hybrid data assimilation and machine learning method

Xinlei He,Yanping Li,Shaomin Liu,Tongren Xu,Fei Chen,Zhenhua Li,Zhe Zhang,Rui Liu,Lisheng Song,Ziwei Xu,Zhixing Peng,Chen Zheng

DOI: https://doi.org/10.5194/hess-2022-379

IF: 6.3

2022-12-15

Hydrology and Earth System Sciences

Abstract:The energy and moisture exchange between the land surface and atmospheric boundary layer plays a critical role in regional climate simulations. This paper implemented a hybrid data assimilation and machine learning framework (DA-ML method) into the Weather Research and Forecasting (WRF) model to optimize surface soil and vegetation conditions. The hybrid method can integrate remotely sensed leaf area index (LAI), multi-source soil moisture (SM) observations, and land surface models (LSMs) to accurately describe land surface states and fluxes. The performance of the hybrid method on the regional climate was evaluated in the Heihe River Basin (HRB), the second largest endorheic river basin in Northwest China. The findings indicate that the DA-ML method improved the estimation of evapotranspiration (ET) and generated a spatial distribution consistent with the ML-based watershed ET (ETMap). The WRF simulations overestimated (underestimated) the air temperature (specific humidity) in the vegetated areas of the HRB. In contrast, the estimated air temperature and specific humidity from WRF (DA-ML) agree well with the observations, especially in the midstream oasis. The DA-ML framework enhanced oasis-desert interactions by improving the soil and vegetation characteristics. The wetting and cooling effects and wind shield effects of the oasis were enhanced by the DA-ML. The wetting and cooling effect of the oasis can transfer water vapor to the surrounding desert, which benefits the oasis-desert ecosystem. The results show that the wetting and cooling effects only negligibly changed the local precipitation in the midstream oasis. However, upstream of the HRB, the integration of LAI and SM will induce water vapor intensification and promote precipitation, particularly on windward slopes.

geosciences, multidisciplinary,water resources
Recent Advances in Regional Climate Modeling and Applications

LIU Hong-Bo,ZHANG Da-Lin,WANG Bin

DOI: https://doi.org/10.3969/j.issn.1006-9585.2006.05.010

2006-01-01

Abstract:Significant advances have been made in regional climate modeling research during the past decades.Numerous regional climate models(RCMs)have now become an important tool for climate research and operational prediction after extensive model developments and verifications.Many encouraging results have been published in simulating historic extreme climate events and predicting our future climate,particularly in the monthly to seasonal climate variables.By analyzing the high-resolution,dynamically consistent model datasets,remarkable improvements have been made in understanding various physics processes,such as land-surface and hydrological processes,the planetary boundary layer,clouds and precipitation,cloud-radiation interaction at the weekly to seasonal scales.However,RCMs still face many challenging issues due to the complicated scale interaction over a large-range temporal and spatial scales(e.g.,from mesoscale to synoptic and planetary scale disturbances),the interaction of individual climate sub-systems(e.g.,atmosphere,biosphere,water,ice-snow,land-surface),and the lack of understanding of some complex physical processes,especially clouds and soil moisture.This article summarizes the recent progress in the present and future climate simulation,the extreme weather and climate studies,and in understanding various physics processes as well as the short-term climate prediction using RCMs.The prospects of RCMs related to the above subjects,especially the weekly to subseasonal regional climate simulation and prediction,will also be discussed.
Challenges and opportunities for improved understanding of regional climate dynamics

Matthew Collins,Shoshiro Minobe,Marcelo Barreiro,Simona Bordoni,Yohai Kaspi,Akira Kuwano-Yoshida,Noel Keenlyside,Elisa Manzini,Christopher H. O’Reilly,Rowan Sutton,Shang-Ping Xie,Olga Zolina

DOI: https://doi.org/10.1038/s41558-017-0059-8

IF: 30.7

2018-01-29

Nature Climate Change

Abstract:Dynamical processes in the atmosphere and ocean are central to determining the large-scale drivers of regional climate change, yet their predictive understanding is poor. Here, we identify three frontline challenges in climate dynamics where significant progress can be made to inform adaptation: response of storms, blocks and jet streams to external forcing; basin-to-basin and tropical–extratropical teleconnections; and the development of non-linear predictive theory. We highlight opportunities and techniques for making immediate progress in these areas, which critically involve the development of high-resolution coupled model simulations, partial coupling or pacemaker experiments, as well as the development and use of dynamical metrics and exploitation of hierarchies of models.

environmental sciences,environmental studies,meteorology & atmospheric sciences
Development of a regional feature selection-based machine learning system (RFSML v1.0) for air pollution forecasting over China

Li Fang,Jianbing Jin,Arjo Segers,Hai Xiang Lin,Mijie Pang,Cong Xiao,Tuo Deng,Hong Liao

DOI: https://doi.org/10.5194/gmd-15-7791-2022

IF: 5.1

2022-10-24

Geoscientific Model Development

Abstract:Abstract. With the explosive growth of atmospheric data, machine learning models have achieved great success in air pollution forecasting because of their higher computational efficiency than the traditional chemical transport models. However, in previous studies, new prediction algorithms have only been tested at stations or in a small region; a large-scale air quality forecasting model remains lacking to date. Huge dimensionality also means that redundant input data may lead to increased complexity and therefore the over-fitting of machine learning models. Feature selection is a key topic in machine learning development, but it has not yet been explored in atmosphere-related applications. In this work, a regional feature selection-based machine learning (RFSML) system was developed, which is capable of predicting air quality in the short term with high accuracy at the national scale. Ensemble-Shapley additive global importance analysis is combined with the RFSML system to extract significant regional features and eliminate redundant variables at an affordable computational expense. The significance of the regional features is also explained physically. Compared with a standard machine learning system fed with relative features, the RFSML system driven by the selected key features results in superior interpretability, less training time, and more accurate predictions. This study also provides insights into the difference in interpretability among machine learning models (i.e., random forest, gradient boosting, and multi-layer perceptron models).

geosciences, multidisciplinary
Improving regional climate simulations based on a hybrid data assimilation and machine learning method

DOI: https://doi.org/10.5194/hess-27-1583-2023

IF: 6.3

2023-04-18

Hydrology and Earth System Sciences

Abstract:The energy and water vapor exchange between the land surface and atmospheric boundary layer plays a critical role in regional climate simulations. This paper implemented a hybrid data assimilation and machine learning framework (DA-ML method) into the Weather Research and Forecasting (WRF) model to optimize surface soil and vegetation conditions. The hybrid method can integrate remotely sensed leaf area index (LAI), multi-source soil moisture (SM) observations, and land surface models (LSMs) to accurately describe regional climate and land–atmosphere interactions. The performance of the hybrid method on the regional climate was evaluated in the Heihe River basin (HRB), the second-largest endorheic river basin in Northwest China. The results show that the estimated sensible ( H ) and latent heat ( L E ) fluxes from the WRF (DA-ML) model agree well with the large aperture scintillometer (LAS) observations. Compared to the WRF (open loop – OL), the WRF (DA-ML) model improved the estimation of evapotranspiration (ET) and generated a spatial distribution consistent with the ML-based watershed ET (ETMap). The proposed WRF (DA-ML) method effectively reduces air warming and drying biases in simulations, particularly in the oasis region. The estimated air temperature and specific humidity from WRF (DA-ML) agree well with the observations. In addition, this method can simulate more realistic oasis–desert boundaries, including wetting and cooling effects and wind shield effects within the oasis. The oasis–desert interactions can transfer water vapor to the surrounding desert in the lower atmosphere. In contrast, the dry and hot air over the desert is transferred to the oasis from the upper atmosphere. The results show that the integration of LAI and SM will induce water vapor intensification and promote precipitation in the upstream of the HRB, particularly on windward slopes. In general, the proposed WRF (DA-ML) model can improve climate modeling by implementing detailed land characterization information in basins with complex underlying surfaces.

geosciences, multidisciplinary,water resources
Climate predicts wildland fire extent across China

Ali Hassan Shabbir,Jie Ji,John W. Groninger,Ghislain N. Gueye,Jason H. Knouft,Eddie J. B. van Etten,Jiquan Zhang

DOI: https://doi.org/10.1016/j.scitotenv.2023.164987

IF: 9.8

2023-01-01

The Science of The Total Environment

Abstract:Wildland fire extent varies seasonally and interannually in response to climatic and landscape-level drivers, yet predicting wildfires remains a challenge. Existing linear models that characterize climate and wildland fire relationships fail to account for non-stationary and non-linear associations, thus limiting prediction accuracy. To account for non-stationary and non-linear effects, we use time-series climate and wildfire extent data from across China with unit root methods, thus providing an approach for improved wildfire prediction. Results from this approach suggest that wildland area burned is sensitive to vapor pressure deficit (VPD) and maximum temperature changes over short and long-term scenarios. Moreover, repeated fires constrain system variability resulting in non-stationarity responses. We conclude that an autoregressive distributed lag (ARDL) approach to dynamic simulation models better elucidates interactions between climate and wildfire compared to more commonly used linear models. We suggest that this approach will provide insights into a better understanding of complex ecological relationships and represents a significant step toward the development of guidance for regional planners hoping to address climate-driven increases in wildfire incidence and impacts.
Global Fire Forecasts Using Both Large‐Scale Climate Indices and Local Meteorological Parameters

Huizhong Shen,Shu Tao,Yilin Chen,Mehmet Talat Odman,Yufei Zou,Ye Huang,Han Chen,Qirui Zhong,Yanyan Zhang,Yuanchen Chen,Shu Su,Nan Lin,Shaojie Zhuo,Bengang Li,Xilong Wang,Wenxin Liu,Junfeng Liu,Gertrude K. Pavur,Armistead G. Russell

DOI: https://doi.org/10.1029/2019gb006180

IF: 6.5

2019-01-01

Global Biogeochemical Cycles

Abstract:Fire forecasts that predict dry‐season fire activities several months in advance are beneficial for fire management. On a global scale, however, the predictability of fires is limited because fires depend on multiple factors and lack a single dominant predictor to describe diverse fire characteristics across regions. Here, based on 33 local meteorological parameters (MPs) and 37 large‐scale climate indices (CIs), we establish four empirical model clusters to predict global interannual fire variability. We show that across various geographic locations, the models provide reliable fire forecasts at least three months prior to the peak fire months. Compared to MPs, CIs such as the Oceanic Niño Index are comparable or even superior predictors. Globally, as well as in most continents, the El Niño–Southern Oscillation is the major driving force, explaining 17% of interannual fire variability, with strong implications for fire carbon emissions and the global carbon cycle. Other important predictors include the Northern Atlantic sea surface temperature (9%), the Southern Atlantic sea surface temperature (5%), and the Pacific/North American Pattern (3%). The predictive models reveal a strong interaction between MPs and CIs, indicating potential climate‐induced modification of fire responses to meteorological conditions. We show that the newly developed predictive models can benefit future fire management in response to climate change.
Regional Integrated Environmental Modeling System: Development and Application

Shuyu Wang,Congbin Fu,Helin Wei,Yun Qian,Zhe Xiong,Jinming Feng,Deming Zhao,Li Dan,Zhiwei Han,Bingkai Su,Ming Zhao,Yaocun Zhang,Jianping Tang,Hongnian Liu,Jian Wu,Xinmin Zeng,Min Chen,Lizhi Wang

DOI: https://doi.org/10.1007/s10584-013-0973-3

2013-01-01

Climatic Change

Abstract:The demand for high-confidence regional climate change scenarios is increasing. It is therefore vitally important to better understand the behavior of Earth’s climate system on regional scale and advance the knowledge of regional responses to global climate. With their ability to represent meso-scale forcings, such as coastline, complex topography, anthropogenic aerosols and land cover/use changes, Regional Climate Models (RCMs) are developed and used worldwide to investigate the effects of the above-mentioned meso-scale forcings on the local circulations that regulate the regional distribution of climatic variables. Considering the complexity of Asian Monsoon system, which is not only a physical process but also modulated by the interaction among physical, biological, chemical and social processes, a modeling framework Regional Integrated Environmental Modeling System (RIEMS) was proposed, developed and well tested before it was widely used in regional climate studies in the East Asia monsoon region.
Regional carbon emission evolution mechanism and its prediction approach: a case study of Hebei, China

Jingmin Wang,Fan Yang,Keke Chen

DOI: https://doi.org/10.1007/s11356-019-06021-0

IF: 5.8

2019-08-05

Environmental Science and Pollution Research

Abstract:Industrialization and urbanization process has brought both economic development and a series of environmental problems. Hebei Province is a high haze area and one of the key areas of China's "Blue Sky Defense" policy. As carbon emission is one of the most important factors causing haze, it is necessary to screen the influencing factors effectively to make quantitative analysis and predict carbon emissions in different degrees of coordination. Faced with the new situation of Beijing-Tianjin-Hebei coordinated development and more environmental constraints, this paper designs a system for predicting carbon emissions in Hebei Province innovatively using the data during 1990–2016. In the first step, 7 influence factors are determined by "graded screening." Then, this paper improved the traditional support vector machine (SVM) by improved gray wolf optimizer (IGWO), and established IGWO-SVM model. Finally, the carbon emissions in Hebei Province from 2017 to 2025 are predicted under three stages of different cooperative development speed of 7 influence factors by the improved intelligent algorithm model. Based on the forecasting results, this paper put forward some pertinent opinions to provide theoretical basis for formulating relevant policies on carbon emissions in Hebei Province to effectively control carbon emissions from the source.

environmental sciences
Advances in regional climate modeling since 1990

Zhao Zongci,Luo Yong

1997-01-01

Journal of Meteorological Research

Abstract:Due to the close relationship between regional climate anomalies and social-economy and society development,climatologists worldwide paid great attention to the regional climate anomalies over a long period of time and the corresponding investigation of regional climat emodeling has made great progresses.Since 1990 the regional climate simulations have made a more substantial achievement.This paper will focus on the reliability and uncertainties of regional climate modeling by global climate models,the advances on regional climate modeling in the world and the outlook of regional climate modeling.
Statistical Downscaling and Dynamical Downscaling of Regional Climate in China: Present Climate Evaluations and Future Climate Projections

Jianping Tang,Xiaorui Niu,Shuyu Wang,Hongxia Gao,Xueyuan Wang,Jian Wu

DOI: https://doi.org/10.1002/2015jd023977

2016-01-01

Abstract:Statistical downscaling and dynamical downscaling are two approaches to generate high‐resolution regional climate models based on the large‐scale information from either reanalysis data or global climate models. In this study, these two downscaling methods are used to simulate the surface climate of China and compared. The Statistical Downscaling Model (SDSM) is cross validated and used to downscale the regional climate of China. Then, the downscaled historical climate of 1981–2000 and future climate of 2041–2060 are compared with that from the Weather Research and Forecasting (WRF) model driven by the European Center‐Hamburg atmosphere model and the Max Planck Institute Ocean Model (ECHAM5/MPI‐OM) and the L'Institut Pierre‐Simon Laplace Coupled Model, version 5, coupled with the Nucleus for European Modelling of the ocean, low resolution (IPSL‐CM5A‐LR). The SDSM can reproduce the surface temperature characteristics of the present climate in China, whereas the WRF tends to underestimate the surface temperature over most of China. Both the SDSM and WRF require further work to improve their ability to downscale precipitation. Both statistical and dynamical downscaling methods produce future surface temperatures for 2041–2060 that are markedly different from the historical climatology. However, the changes in projected precipitation differ between the two downscaling methods. Indeed, large uncertainties remain in terms of the direction and magnitude of future precipitation changes over China.
Predicting global patterns of long-term climate change from short-term simulations using machine learning

L. A. Mansfield,P. J. Nowack,M. Kasoar,R. G. Everitt,W. J. Collins,A. Voulgarakis

DOI: https://doi.org/10.1038/s41612-020-00148-5

IF: 9.4475

2020-11-19

npj Climate and Atmospheric Science

Abstract:Abstract Understanding and estimating regional climate change under different anthropogenic emission scenarios is pivotal for informing societal adaptation and mitigation measures. However, the high computational complexity of state-of-the-art climate models remains a central bottleneck in this endeavour. Here we introduce a machine learning approach, which utilises a unique dataset of existing climate model simulations to learn relationships between short-term and long-term temperature responses to different climate forcing scenarios. This approach not only has the potential to accelerate climate change projections by reducing the costs of scenario computations, but also helps uncover early indicators of modelled long-term climate responses, which is of relevance to climate change detection, predictability, and attribution. Our results highlight challenges and opportunities for data-driven climate modelling, especially concerning the incorporation of even larger model datasets in the future. We therefore encourage extensive data sharing among research institutes to build ever more powerful climate response emulators, and thus to enable faster climate change projections.

meteorology & atmospheric sciences
Global Prediction of Flash Drought Using Machine Learning

Lei Xu,Xihao Zhang,Tingtao Wu,Hongchu Yu,Wenying Du,Chong Zhang,Nengcheng Chen

DOI: https://doi.org/10.1029/2024gl111134

IF: 5.2

2024-11-06

Geophysical Research Letters

Abstract:Flash droughts are rapidly developing extreme weather events with sudden onset and quick intensification. Global prediction of flash droughts at sub‐seasonal time scales remains a great challenge. Current state‐of‐the‐art dynamic models subject to large errors and demonstrate low skills in global flash drought prediction. Here, we develop a machine learning‐based framework that uses meteorological forecasts as inputs to predict global root‐zone soil moisture and flash droughts from 1 day to 2 week lead times. The results indicate that 33% and 24% global flash drought onset and termination events can be correctly predicted by machine learning at 7 day lead time, versus 19% and 11% fractions by state‐of‐the‐art dynamic model. The developed machine learning model demonstrates substantial improvements over dynamic model in global soil moisture prediction, and thus enhances global flash drought forecasting skills in space and time. The presented framework may benefit global flash drought prediction and early warning at sub‐seasonal scales.

geosciences, multidisciplinary
Dynamical-generative downscaling of climate model ensembles

Ignacio Lopez-Gomez,Zhong Yi Wan,Leonardo Zepeda-Núñez,Tapio Schneider,John Anderson,Fei Sha

2024-10-03

Abstract:Regional high-resolution climate projections are crucial for many applications, such as agriculture, hydrology, and natural hazard risk assessment. Dynamical downscaling, the state-of-the-art method to produce localized future climate information, involves running a regional climate model (RCM) driven by an Earth System Model (ESM), but it is too computationally expensive to apply to large climate projection ensembles. We propose a novel approach combining dynamical downscaling with generative artificial intelligence to reduce the cost and improve the uncertainty estimates of downscaled climate projections. In our framework, an RCM dynamically downscales ESM output to an intermediate resolution, followed by a generative diffusion model that further refines the resolution to the target scale. This approach leverages the generalizability of physics-based models and the sampling efficiency of diffusion models, enabling the downscaling of large multi-model ensembles. We evaluate our method against dynamically-downscaled climate projections from the CMIP6 ensemble. Our results demonstrate its ability to provide more accurate uncertainty bounds on future regional climate than alternatives such as dynamical downscaling of smaller ensembles, or traditional empirical statistical downscaling methods. We also show that dynamical-generative downscaling results in significantly lower errors than bias correction and spatial disaggregation (BCSD), and captures more accurately the spectra and multivariate correlations of meteorological fields. These characteristics make the dynamical-generative framework a flexible, accurate, and efficient way to downscale large ensembles of climate projections, currently out of reach for pure dynamical downscaling.

Atmospheric and Oceanic Physics,Machine Learning

Advancing Regional Climate Predictability Through ML-enabled Dynamical System Approach

Current Progress in Subseasonal-to-decadal Prediction Based on Machine Learning

Global Fire Modelling and Control Attributions Based on the Ensemble Machine Learning and Satellite Observations

Recent Advances in China on the Predictability of Weather and Climate

A Dynamical Pathway Bridging African Biomass Burning and Asian Summer Monsoon

AttentionFire_v1.0: interpretable machine learning fire model for burned-area predictions over tropics

Quantifying The Drivers And Predictability Of Seasonal Changes In African Fire

Improving predictions of land-atmosphere interactions based on a hybrid data assimilation and machine learning method

Recent Advances in Regional Climate Modeling and Applications

Challenges and opportunities for improved understanding of regional climate dynamics

Development of a regional feature selection-based machine learning system (RFSML v1.0) for air pollution forecasting over China

Improving regional climate simulations based on a hybrid data assimilation and machine learning method

Climate predicts wildland fire extent across China

Global Fire Forecasts Using Both Large‐Scale Climate Indices and Local Meteorological Parameters

Regional Integrated Environmental Modeling System: Development and Application

Regional carbon emission evolution mechanism and its prediction approach: a case study of Hebei, China

Advances in regional climate modeling since 1990

Statistical Downscaling and Dynamical Downscaling of Regional Climate in China: Present Climate Evaluations and Future Climate Projections

Predicting global patterns of long-term climate change from short-term simulations using machine learning

Global Prediction of Flash Drought Using Machine Learning

Dynamical-generative downscaling of climate model ensembles