Location Invariant Flood Forecasting using Fourier Neural Operator

Location Invariant Flood Forecasting using Fourier Neural Operator | 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 Search Search ADVANCED SEARCH SCROLL This is a preprint and has not been peer reviewed.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. 9 January 2025 V1 Latest version Share on Location Invariant Flood Forecasting using Fourier Neural Operator Authors : Chetan Kumar 0000-0003-2755-7338 [email protected] , Diana Mcspadden , Steven Goldenberg , Malachi Schram , Heather Richter , Yidi Wang , Binata Roy , and Jonathan L Goodall Authors Info & Affiliations https://doi.org/10.22541/au.173645444.46575894/v1 313 views 110 downloads Contents Abstract Supplementary Material Information & Authors Metrics & Citations View Options References Figures Tables Media Share Abstract Real-time flood prediction is crucial for coastal urban cities prone to flooding; enabling communities, emergency services, and transportation authorities to prepare and take necessary precautions. Traditional approaches for flood prediction involve high-fidelity physics-based hydrodynamic models, which are computationally expensive, particularly in terms of time. To address these limitations, data-driven machine learning models have been developed for real-time predictions, focusing primarily on specific spatial dimensions such as street segments, floodplain areas, or geographic locations. In this study, we propose a generalized machine learning model based on Fourier Neural Operators (FNO), capable of operating across geographic locations not seen during training. For this research, we analyze eleven storm events in Norfolk, VA, spanning from 2017 to 2022, each lasting approximately four days with a 15-minute time interval. Our FNO model utilizes water depth maps generated by the TUFLOW hydrodynamic model, with a spatial resolution of 2.5m x 2.5m, along with rainfall data from seven observation sites maintained by the Hampton Roads Sanitation District. We employ inverse-distance weighted interpolation to account for geographic variations of rainfall across the study area. For model evaluation and generalization, we implement a k-fold cross-validation approach, randomly dividing the study area into five folds. Our findings show that training for 24 time steps (360 minutes) result in a model capable of accurate predictions in the next 6 time steps (90 minutes). Additionally, sequential experiments using k-fold with eleven storm event folds demonstrate that a look-back period of 360 minutes yields similarly low error rate predictions in the 15 minutes look-ahead interval. These findings underscore the efficiency and generalizability of our FNO-based model, demonstrating its capability to effectively handle predictions over extended periods for unseen geographic locations. Additionally, our machine learning model achieves an order of magnitude speed up compared to traditional physics-based hydrodynamic models. Supplementary Material File (agu_abstract_archive.pdf) Download 111.30 KB Information & Authors Information Version history V1 Version 1 09 January 2025 Copyright This work is licensed under a Non Exclusive No Reuse License. Keywords environmental sciences hydrology Authors Affiliations Chetan Kumar 0000-0003-2755-7338 [email protected] Old Dominion University, Thomas Jefferson National Accelerator Facility Joint Institute on Advanced Computing for Environmental Studies View all articles by this author Diana Mcspadden Old Dominion University, Thomas Jefferson National Accelerator Facility Joint Institute on Advanced Computing for Environmental Studies Thomas Jefferson National Accelerator Facility View all articles by this author Steven Goldenberg Thomas Jefferson National Accelerator Facility View all articles by this author Malachi Schram Old Dominion University, Thomas Jefferson National Accelerator Facility Joint Institute on Advanced Computing for Environmental Studies Thomas Jefferson National Accelerator Facility View all articles by this author Heather Richter Old Dominion University, Thomas Jefferson National Accelerator Facility Joint Institute on Advanced Computing for Environmental Studies View all articles by this author Yidi Wang Department of Civil and Environmental Engineering, University of Virginia View all articles by this author Binata Roy Department of Civil and Environmental Engineering, University of Virginia View all articles by this author Jonathan L Goodall Department of Civil and Environmental Engineering, University of Virginia View all articles by this author Metrics & Citations Metrics Article Usage 313 views 110 downloads .FvxKWukQNSOunydq8rnd { width: 100px; } Citations Download citation Chetan Kumar, Diana Mcspadden, Steven Goldenberg, et al. Location Invariant Flood Forecasting using Fourier Neural Operator. Authorea . 09 January 2025. DOI: https://doi.org/10.22541/au.173645444.46575894/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.173645444.46575894/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:'a00e9a3e1ce8df94',t:'MTc3OTY0OTg0OA=='};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())}}}})();