Learning to Infer Weather States Using Partial Observations

Abstract Accurate state estimation of the high‐dimensional, chaotic Earth's atmosphere marks a Sisyphean task, yet is indispensable for initiating weather forecasts and gauging climate variability. While much effort is devoted to assimilating observations and forecasts to infer weather states,...

Full description

Saved in:
Bibliographic Details
Main Authors: Jie Chao, Baoxiang Pan, Quanliang Chen, Shangshang Yang, Jingnan Wang, Congyi Nai, Yue Zheng, Xichen Li, Huiling Yuan, Xi Chen, Bo Lu, Ziniu Xiao
Format: Article
Language:English
Published: Wiley 2025-03-01
Series:Journal of Geophysical Research: Machine Learning and Computation
Online Access:https://doi.org/10.1029/2024JH000260
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850063183871475712
author Jie Chao
Baoxiang Pan
Quanliang Chen
Shangshang Yang
Jingnan Wang
Congyi Nai
Yue Zheng
Xichen Li
Huiling Yuan
Xi Chen
Bo Lu
Ziniu Xiao
author_facet Jie Chao
Baoxiang Pan
Quanliang Chen
Shangshang Yang
Jingnan Wang
Congyi Nai
Yue Zheng
Xichen Li
Huiling Yuan
Xi Chen
Bo Lu
Ziniu Xiao
author_sort Jie Chao
collection DOAJ
description Abstract Accurate state estimation of the high‐dimensional, chaotic Earth's atmosphere marks a Sisyphean task, yet is indispensable for initiating weather forecasts and gauging climate variability. While much effort is devoted to assimilating observations and forecasts to infer weather states, the inherent low‐dimensional statistical structure in atmospheric circulation, shaped by geophysical laws and geographic boundaries, is underutilized as an informative prior for state inference, or as reference for assessing representative of existing observations and planning new ones. We realize these potential by learning climatological distribution from climate reanalysis/simulation, using a deep generative model. For a case study of estimating 2 m temperature spatial patterns, the learned distribution faithfully reproduces climatology statistics. A combination of the learned climatological prior with few station observations yields strong posterior of spatial pattern estimates, which are spatially coherent, faithful and adaptive to observational constraints, and uncertainty‐aware. This allows us to evaluate each observation's value in reducing pattern estimation uncertainty, and guide optimal observation network design by pinpointing the most informative sites. Our study showcases how generative models can extract and utilize information produced in the chaotic evolution of climate system.
format Article
id doaj-art-7ec668e479e949529ddade6bf02b9c2c
institution DOAJ
issn 2993-5210
language English
publishDate 2025-03-01
publisher Wiley
record_format Article
series Journal of Geophysical Research: Machine Learning and Computation
spelling doaj-art-7ec668e479e949529ddade6bf02b9c2c2025-08-20T02:49:43ZengWileyJournal of Geophysical Research: Machine Learning and Computation2993-52102025-03-0121n/an/a10.1029/2024JH000260Learning to Infer Weather States Using Partial ObservationsJie Chao0Baoxiang Pan1Quanliang Chen2Shangshang Yang3Jingnan Wang4Congyi Nai5Yue Zheng6Xichen Li7Huiling Yuan8Xi Chen9Bo Lu10Ziniu Xiao11National Key Laboratory of Earth System Numerical Modeling and Application Institute of Atmospheric Physics Chinese Academy of Science Beijing ChinaNational Key Laboratory of Earth System Numerical Modeling and Application Institute of Atmospheric Physics Chinese Academy of Science Beijing ChinaSchool of Atmospheric Sciences Chengdu University of Information Technology Chengdu ChinaNational Key Laboratory of Earth System Numerical Modeling and Application Institute of Atmospheric Physics Chinese Academy of Science Beijing ChinaNational Key Laboratory of Earth System Numerical Modeling and Application Institute of Atmospheric Physics Chinese Academy of Science Beijing ChinaNational Key Laboratory of Earth System Numerical Modeling and Application Institute of Atmospheric Physics Chinese Academy of Science Beijing ChinaClustertech LTD Hong Kong ChinaNational Key Laboratory of Earth System Numerical Modeling and Application Institute of Atmospheric Physics Chinese Academy of Science Beijing ChinaKey Laboratory of Mesoscale Severe Weather Ministry of Education, and School of Atmospheric Sciences Nanjing University Nanjing ChinaNational Key Laboratory of Earth System Numerical Modeling and Application Institute of Atmospheric Physics Chinese Academy of Science Beijing ChinaNational Climate Center China Meteorological Administration Beijing ChinaNational Key Laboratory of Earth System Numerical Modeling and Application Institute of Atmospheric Physics Chinese Academy of Science Beijing ChinaAbstract Accurate state estimation of the high‐dimensional, chaotic Earth's atmosphere marks a Sisyphean task, yet is indispensable for initiating weather forecasts and gauging climate variability. While much effort is devoted to assimilating observations and forecasts to infer weather states, the inherent low‐dimensional statistical structure in atmospheric circulation, shaped by geophysical laws and geographic boundaries, is underutilized as an informative prior for state inference, or as reference for assessing representative of existing observations and planning new ones. We realize these potential by learning climatological distribution from climate reanalysis/simulation, using a deep generative model. For a case study of estimating 2 m temperature spatial patterns, the learned distribution faithfully reproduces climatology statistics. A combination of the learned climatological prior with few station observations yields strong posterior of spatial pattern estimates, which are spatially coherent, faithful and adaptive to observational constraints, and uncertainty‐aware. This allows us to evaluate each observation's value in reducing pattern estimation uncertainty, and guide optimal observation network design by pinpointing the most informative sites. Our study showcases how generative models can extract and utilize information produced in the chaotic evolution of climate system.https://doi.org/10.1029/2024JH000260
spellingShingle Jie Chao
Baoxiang Pan
Quanliang Chen
Shangshang Yang
Jingnan Wang
Congyi Nai
Yue Zheng
Xichen Li
Huiling Yuan
Xi Chen
Bo Lu
Ziniu Xiao
Learning to Infer Weather States Using Partial Observations
Journal of Geophysical Research: Machine Learning and Computation
title Learning to Infer Weather States Using Partial Observations
title_full Learning to Infer Weather States Using Partial Observations
title_fullStr Learning to Infer Weather States Using Partial Observations
title_full_unstemmed Learning to Infer Weather States Using Partial Observations
title_short Learning to Infer Weather States Using Partial Observations
title_sort learning to infer weather states using partial observations
url https://doi.org/10.1029/2024JH000260
work_keys_str_mv AT jiechao learningtoinferweatherstatesusingpartialobservations
AT baoxiangpan learningtoinferweatherstatesusingpartialobservations
AT quanliangchen learningtoinferweatherstatesusingpartialobservations
AT shangshangyang learningtoinferweatherstatesusingpartialobservations
AT jingnanwang learningtoinferweatherstatesusingpartialobservations
AT congyinai learningtoinferweatherstatesusingpartialobservations
AT yuezheng learningtoinferweatherstatesusingpartialobservations
AT xichenli learningtoinferweatherstatesusingpartialobservations
AT huilingyuan learningtoinferweatherstatesusingpartialobservations
AT xichen learningtoinferweatherstatesusingpartialobservations
AT bolu learningtoinferweatherstatesusingpartialobservations
AT ziniuxiao learningtoinferweatherstatesusingpartialobservations