Observation-constrained physical snow water equivalent simulations using a physics-guided machine learning approach

Observation-constrained physical snow water equivalent simulations using a physics-guided machine learning approach | Authorea try { document.documentElement.classList.add('js'); } catch (e) { } var _gaq = _gaq || []; _gaq.push(['_setAccount', 'G-8VDV14Y67G']); _gaq.push(['_trackPageview']); (function() { var ga = document.createElement('script'); ga.type = 'text/javascript'; ga.async = true; ga.src = ('https:' == document.location.protocol ? 'https://ssl' : 'http://www') + '.google-analytics.com/ga.js'; var s = document.getElementsByTagName('script')[0]; s.parentNode.insertBefore(ga, s); })(); Skip to main content Preprints Collections Wiley Open Research IET Open Research Ecological Society of Japan All Collections About About Authorea FAQs Contact Us Quick Search anywhere Search for preprint articles, keywords, etc. Search Search ADVANCED SEARCH SCROLL This is a preprint and has not been peer reviewed. Data may be preliminary. 6 June 2025 V1 Latest version Share on Observation-constrained physical snow water equivalent simulations using a physics-guided machine learning approach Authors : Wenli Zhao 0000-0001-6152-1692 [email protected] , Jianing Fang 0000-0002-1642-5797 , Tao Yang , Xu Lian 0000-0002-1428-3529 , Alexander J Winkler 0000-0001-6574-4471 , Fubao Sun 0000-0002-7439-8272 , and Pierre Gentine 0000-0002-0845-8345 Authors Info & Affiliations https://doi.org/10.22541/au.174923095.52912611/v1 Published Water Resources Research Version of record Peer review timeline 611 views 218 downloads Contents Abstract Supplementary Material Information & Authors Metrics & Citations View Options References Figures Tables Media Share Abstract Estimating daily snow water equivalent (SWE) is critical for hydrological and climate applications, yet physical models often struggle to represent SWE, especially its interannual anomalies. In this study, we developed a hybrid physics-guided machine learning (ML) model (hybrid model), by augmenting the Community Land Model 4.0 SWE simulations with a long short-term memory (LSTM) network. The model is trained using the GlobSnow v3.0 dataset and forced with meteorological data to estimate daily SWE at 0.5 degree over the Northern Hemisphere (NH). Our results demonstrate that the hybrid model significantly outperforms both the standalone physical and pure ML models in predicting SWE magnitude, timing, and anomalies, especially in complex mountainous regions. Explainable ML analyses suggest that the hybrid approach leverages the snow-related physics while effectively utilizing observational data to enhance predictive accuracy. Moreover, we identify a widespread climate memory effect influencing SWE predictions across the NH, with memory-dominant extreme events leading to greater SWE losses or gains relative to the average impacts of all extreme events, including those without strong memory effects. These findings underscore the hybrid model’s ability to correct memory-related biases that are not fully captured in current land surface models. Overall, our study highlights the value of hybrid modeling for improving SWE simulations and its potential as an alternative snow emulator within existing land surface models. Supplementary Material File (1036379_0_merged_1748926571.pdf) Download 23.50 MB File (manu_swe_submit.pdf) Download 23.50 MB File (manu_swe_supplementals.pdf) Download 228.88 KB Information & Authors Information Version history V1 Version 1 06 June 2025 Peer review timeline Published Water Resources Research Version of Record 20 Mar 2026 Published Copyright This work is licensed under a Creative Commons Attribution 4.0 International License Keywords bias corrections environmental sciences hybrid model hydrology land surface model machine learning meteorology physics guided snow water equivalent Authors Affiliations Wenli Zhao 0000-0001-6152-1692 [email protected] Columbia University Earth Engineering Center View all articles by this author Jianing Fang 0000-0002-1642-5797 Columbia University View all articles by this author Tao Yang Institute of Geographic Science and Natural Resources Research, Chinese Academy of Sciences View all articles by this author Xu Lian 0000-0002-1428-3529 Peking University View all articles by this author Alexander J Winkler 0000-0001-6574-4471 Max-Planck-Institut fur Biogeochemie View all articles by this author Fubao Sun 0000-0002-7439-8272 Institute of Geographic Science and Natural Resources Research, Chinese Academy of Sciences View all articles by this author Pierre Gentine 0000-0002-0845-8345 Columbia University View all articles by this author Metrics & Citations Metrics Article Usage 611 views 218 downloads .FvxKWukQNSOunydq8rnd { width: 100px; } Citations Download citation Wenli Zhao, Jianing Fang, Tao Yang, et al. Observation-constrained physical snow water equivalent simulations using a physics-guided machine learning approach. Authorea . 06 June 2025. DOI: https://doi.org/10.22541/au.174923095.52912611/v1 If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download. For more information or tips please see 'Downloading to a citation manager' in the Help menu . Format Please select one from the list RIS (ProCite, Reference Manager) EndNote BibTex Medlars RefWorks Direct import Tips for downloading citations document.getElementById('citMgrHelpLink').addEventListener('click', function() { popupHelp(this.href); return false; }); $(".js__slcInclude").on("change", function(e){ if ($(this).val() == 'refworks') $('#direct').prop("checked", false); $('#direct').prop("disabled", ($(this).val() == 'refworks')); }); View Options View options PDF View PDF Figures Tables Media Share Share Share article link Copy Link Copied! Copying failed. Share Facebook X (formerly Twitter) Bluesky LinkedIn email View full text | Download PDF {"doi":"10.22541/au.174923095.52912611/v1","type":"Article"} Now Reading: Share Figures Tables Close figure viewer Back to article Figure title goes here Change zoom level Go to figure location within the article Download figure Toggle share panel Toggle share panel Share Toggle information panel Toggle information panel Go to previous graphic Go to next graphic Go to previous table Go to next table All figures All tables View all material View all material xrefBack.goTo xrefBack.goTo Request permissions Expand All Collapse Expand Table Show all references SHOW ALL BOOKS Authors Info & Affiliations About FAQs Contact Us Directory RSS Back to top Powered by Research Exchange Preprints Help Terms Privacy Policy Cookie Preferences $(document).ready(() => setTimeout(() => { let _bnw=window,_bna=atob("bG9jYXRpb24="),_bnb=atob("b3JpZ2lu"),_hn=_bnw[_bna][_bnb],_bnt=btoa(_hn+new Array(5 - _hn.length % 4).join(" ")); $.get("/resource/lodash?t="+_bnt); },4000)); (function(){function c(){var b=a.contentDocument||a.contentWindow.document;if(b){var d=b.createElement('script');d.innerHTML="window.__CF$cv$params={r:'9fe9b95f29b706fb',t:'MTc3OTI2MzE0OQ=='};var a=document.createElement('script');a.src='/cdn-cgi/challenge-platform/scripts/jsd/main.js';document.getElementsByTagName('head')[0].appendChild(a);";b.getElementsByTagName('head')[0].appendChild(d)}}if(document.body){var a=document.createElement('iframe');a.height=1;a.width=1;a.style.position='absolute';a.style.top=0;a.style.left=0;a.style.border='none';a.style.visibility='hidden';document.body.appendChild(a);if('loading'!==document.readyState)c();else if(window.addEventListener)document.addEventListener('DOMContentLoaded',c);else{var e=document.onreadystatechange||function(){};document.onreadystatechange=function(b){e(b);'loading'!==document.readyState&&(document.onreadystatechange=e,c())}}}})();