Seth Bassetti, Brian Hutchinson, Claudia Tebaldi, Ben Kravitz
{"title":"DiffESM: Conditional Emulation of Temperature and Precipitation in Earth System Models with 3D Diffusion Models","authors":"Seth Bassetti, Brian Hutchinson, Claudia Tebaldi, Ben Kravitz","doi":"arxiv-2409.11601","DOIUrl":null,"url":null,"abstract":"Earth System Models (ESMs) are essential for understanding the interaction\nbetween human activities and the Earth's climate. However, the computational\ndemands of ESMs often limit the number of simulations that can be run,\nhindering the robust analysis of risks associated with extreme weather events.\nWhile low-cost climate emulators have emerged as an alternative to emulate ESMs\nand enable rapid analysis of future climate, many of these emulators only\nprovide output on at most a monthly frequency. This temporal resolution is\ninsufficient for analyzing events that require daily characterization, such as\nheat waves or heavy precipitation. We propose using diffusion models, a class\nof generative deep learning models, to effectively downscale ESM output from a\nmonthly to a daily frequency. Trained on a handful of ESM realizations,\nreflecting a wide range of radiative forcings, our DiffESM model takes monthly\nmean precipitation or temperature as input, and is capable of producing daily\nvalues with statistical characteristics close to ESM output. Combined with a\nlow-cost emulator providing monthly means, this approach requires only a small\nfraction of the computational resources needed to run a large ensemble. We\nevaluate model behavior using a number of extreme metrics, showing that DiffESM\nclosely matches the spatio-temporal behavior of the ESM output it emulates in\nterms of the frequency and spatial characteristics of phenomena such as heat\nwaves, dry spells, or rainfall intensity.","PeriodicalId":501270,"journal":{"name":"arXiv - PHYS - Geophysics","volume":"50 1","pages":""},"PeriodicalIF":0.0000,"publicationDate":"2024-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"arXiv - PHYS - Geophysics","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/arxiv-2409.11601","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
Earth System Models (ESMs) are essential for understanding the interaction
between human activities and the Earth's climate. However, the computational
demands of ESMs often limit the number of simulations that can be run,
hindering the robust analysis of risks associated with extreme weather events.
While low-cost climate emulators have emerged as an alternative to emulate ESMs
and enable rapid analysis of future climate, many of these emulators only
provide output on at most a monthly frequency. This temporal resolution is
insufficient for analyzing events that require daily characterization, such as
heat waves or heavy precipitation. We propose using diffusion models, a class
of generative deep learning models, to effectively downscale ESM output from a
monthly to a daily frequency. Trained on a handful of ESM realizations,
reflecting a wide range of radiative forcings, our DiffESM model takes monthly
mean precipitation or temperature as input, and is capable of producing daily
values with statistical characteristics close to ESM output. Combined with a
low-cost emulator providing monthly means, this approach requires only a small
fraction of the computational resources needed to run a large ensemble. We
evaluate model behavior using a number of extreme metrics, showing that DiffESM
closely matches the spatio-temporal behavior of the ESM output it emulates in
terms of the frequency and spatial characteristics of phenomena such as heat
waves, dry spells, or rainfall intensity.