{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,20]],"date-time":"2026-04-20T18:41:03Z","timestamp":1776710463662,"version":"3.51.2"},"reference-count":52,"publisher":"Public Library of Science (PLoS)","issue":"4","license":[{"start":{"date-parts":[[2026,4,20]],"date-time":"2026-04-20T00:00:00Z","timestamp":1776643200000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"the 2025 High-Level Talent Cultivation Program of Zhaoqing University","award":["gcc202512"],"award-info":[{"award-number":["gcc202512"]}]}],"content-domain":{"domain":["www.plosone.org"],"crossmark-restriction":false},"short-container-title":["PLoS One"],"abstract":"<jats:p>Typhoon-induced extreme floods pose severe threats to subtropical watersheds, yet systematic integration of typhoon physics into machine learning flood prediction remains limited. This study developed a physics-informed machine learning framework for the Boluo watershed, South China, emphasizing typhoon feature engineering. Four models (Linear Regression (LR), Artificial Neural Network (ANN), Random Forest (RF) and XGBoost (XGB)) were systematically evaluated across three feature engineering scenarios: Baseline (conventional hydrometeorological variables), With Typhoon (original typhoon observations), and Enhanced Typhoon (19 physics-informed derived features). Physics-driven design included sigmoid-transformed distance decay functions representing saturating near-field typhoon influence, multi-day cumulative impact indices integrating antecedent storm effects, and trajectory-based kinematic features characterizing translation speed and directional evolution. ANN-EnTY achieved superior performance with Kling-Gupta Efficiency (KGE) of 0.946 and Root Mean Square Error (RMSE) of 174 m\u00b3\/s, representing a 3.1% improvement in KGE and 16.7 m\u00b3\/s reduction in RMSE compared to the Baseline scenario. During a representative extreme flood event with peak flow of 7670 m\u00b3\/s, ANN-EnTY reduced peak prediction error by approximately 4% relative to the best-performing baseline model. SHAP analysis revealed upstream flow dominance (72.6%), while typhoon features, contributing only 2% overall, played critical synergistic roles during extremes. This dual-mode pattern of routine memory-driven versus extreme event-driven responses provides mechanistic insights for operational flood warning systems. The framework offers replicable methodology for typhoon-prone watersheds with direct implications for disaster preparedness and water management.<\/jats:p>","DOI":"10.1371\/journal.pone.0346237","type":"journal-article","created":{"date-parts":[[2026,4,20]],"date-time":"2026-04-20T17:44:38Z","timestamp":1776707078000},"page":"e0346237","update-policy":"https:\/\/doi.org\/10.1371\/journal.pone.corrections_policy","source":"Crossref","is-referenced-by-count":0,"title":["Enhancing flood prediction through physics-driven typhoon feature engineering and machine learning"],"prefix":"10.1371","volume":"21","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-2030-076X","authenticated-orcid":true,"given":"Zhi","family":"Zhang","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yusha","family":"Xiao","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Biqing","family":"Chen","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-7578-3378","authenticated-orcid":true,"given":"Kaihao","family":"Long","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Feng","family":"Liang","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jiwu","family":"Liao","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"340","published-online":{"date-parts":[[2026,4,20]]},"reference":[{"issue":"3","key":"pone.0346237.ref001","doi-asserted-by":"crossref","DOI":"10.1029\/2024WR039054","article-title":"Predicting the Evolution of Extreme Water Levels With Long Short\u2010Term Memory Station\u2010Based Approximated Models and Transfer Learning Techniques","volume":"61","author":"S Daramola","year":"2025","journal-title":"Water Resources Research"},{"issue":"2302","key":"pone.0346237.ref002","first-page":"20240287","article-title":"Challenges and opportunities of ML and explainable AI in large-sample hydrology","volume":"383","author":"L Slater","year":"2025","journal-title":"Philos Trans A Math Phys Eng Sci"},{"issue":"2","key":"pone.0346237.ref003","article-title":"Improving Typhoon Predictions by Integrating Data\u2010Driven Machine Learning Model With Physics Model Based on the Spectral Nudging and Data Assimilation","volume":"12","author":"Z Niu","year":"2025","journal-title":"Earth and Space Science"},{"issue":"4","key":"pone.0346237.ref004","doi-asserted-by":"crossref","DOI":"10.1002\/met.70083","article-title":"AI\u2010Based Tropical Cyclone Rainfall Forecasting in the Philippines Using Machine Learning","volume":"32","author":"CG Mesias","year":"2025","journal-title":"Meteorological Applications"},{"issue":"4","key":"pone.0346237.ref005","doi-asserted-by":"crossref","first-page":"498","DOI":"10.2166\/nh.2024.016","article-title":"A hybrid deep learning approach for streamflow prediction utilizing watershed memory and process-based modeling","volume":"55","author":"BA Yifru","year":"2024","journal-title":"Hydrology Research"},{"issue":"1","key":"pone.0346237.ref006","doi-asserted-by":"crossref","DOI":"10.1029\/2024WR038192","article-title":"Improving Streamflow Prediction Using Multiple Hydrological Models and Machine Learning Methods","volume":"61","author":"H Solanki","year":"2025","journal-title":"Water Resources Research"},{"issue":"3","key":"pone.0346237.ref007","doi-asserted-by":"crossref","DOI":"10.1002\/hyp.14847","article-title":"On strictly enforced mass conservation constraints for modelling the Rainfall\u2010Runoff process","volume":"37","author":"JM Frame","year":"2023","journal-title":"Hydrological Processes"},{"issue":"1","key":"pone.0346237.ref008","doi-asserted-by":"crossref","first-page":"1211","DOI":"10.1080\/19942060.2022.2073565","article-title":"A comparison of machine learning models for suspended sediment load classification","volume":"16","author":"N AlDahoul","year":"2022","journal-title":"Engineering Applications of Computational Fluid Mechanics"},{"key":"pone.0346237.ref009","doi-asserted-by":"crossref","first-page":"885","DOI":"10.1016\/j.scitotenv.2019.06.097","article-title":"Impact of rainfall variability and anthropogenic activities on streamflow changes and water stress conditions across Tapi Basin in India","volume":"687","author":"PJ Sharma","year":"2019","journal-title":"Sci Total Environ"},{"key":"pone.0346237.ref010","doi-asserted-by":"crossref","first-page":"128618","DOI":"10.1016\/j.jhydrol.2022.128618","article-title":"Enhancing predictive skills in physically-consistent way: Physics Informed Machine Learning for hydrological processes","volume":"615","author":"P Bhasme","year":"2022","journal-title":"Journal of Hydrology"},{"issue":"7","key":"pone.0346237.ref011","doi-asserted-by":"crossref","DOI":"10.1029\/2023WR035790","article-title":"Streamflow Prediction at the Intersection of Physics and Machine Learning: A Case Study of Two Mediterranean\u2010Climate Watersheds","volume":"60","author":"S Adera","year":"2024","journal-title":"Water Resources Research"},{"key":"pone.0346237.ref012","doi-asserted-by":"crossref","first-page":"171684","DOI":"10.1016\/j.scitotenv.2024.171684","article-title":"Improving the simulations of the hydrological model in the karst catchment by integrating the conceptual model with machine learning models","volume":"926","author":"C Sezen","year":"2024","journal-title":"Sci Total Environ"},{"key":"pone.0346237.ref013","doi-asserted-by":"crossref","first-page":"106091","DOI":"10.1016\/j.envsoft.2024.106091","article-title":"Spatio-temporal deep learning model for accurate streamflow prediction with multi-source data fusion","volume":"178","author":"Z Wang","year":"2024","journal-title":"Environmental Modelling & Software"},{"issue":"1","key":"pone.0346237.ref014","doi-asserted-by":"crossref","first-page":"2762","DOI":"10.1038\/s41598-025-87187-1","article-title":"Enhanced streamflow forecasting using hybrid modelling integrating glacio-hydrological outputs, deep learning and wavelet transformation","volume":"15","author":"JH Ougahi","year":"2025","journal-title":"Sci Rep"},{"issue":"11","key":"pone.0346237.ref015","doi-asserted-by":"crossref","first-page":"1536","DOI":"10.3390\/w10111536","article-title":"Flood Prediction Using Machine Learning Models: Literature Review","volume":"10","author":"A Mosavi","year":"2018","journal-title":"Water"},{"issue":"1","key":"pone.0346237.ref016","article-title":"Distributed Typhoon Track Prediction Based on Complex Features and Multitask Learning","volume":"2021","author":"Y Sun","year":"2021","journal-title":"Complexity"},{"issue":"7","key":"pone.0346237.ref017","doi-asserted-by":"crossref","first-page":"676","DOI":"10.3390\/atmos11070676","article-title":"Machine Learning in Tropical Cyclone Forecast Modeling: A Review","volume":"11","author":"R Chen","year":"2020","journal-title":"Atmosphere"},{"key":"pone.0346237.ref018","doi-asserted-by":"crossref","first-page":"302","DOI":"10.1016\/j.jhydrol.2014.12.009","article-title":"A real-time forecasting model for the spatial distribution of typhoon rainfall","volume":"521","author":"G-F Lin","year":"2015","journal-title":"Journal of Hydrology"},{"key":"pone.0346237.ref019","doi-asserted-by":"crossref","first-page":"623","DOI":"10.1016\/j.jhydrol.2016.06.026","article-title":"A hybrid approach to monthly streamflow forecasting: Integrating hydrological model outputs into a Bayesian artificial neural network","volume":"540","author":"GB Humphrey","year":"2016","journal-title":"Journal of Hydrology"},{"issue":"12","key":"pone.0346237.ref020","doi-asserted-by":"crossref","first-page":"5089","DOI":"10.5194\/hess-23-5089-2019","article-title":"Towards learning universal, regional, and local hydrological behaviors via machine learning applied to large-sample datasets","volume":"23","author":"F Kratzert","year":"2019","journal-title":"Hydrol Earth Syst Sci"},{"issue":"15","key":"pone.0346237.ref021","doi-asserted-by":"crossref","first-page":"4013","DOI":"10.5194\/hess-26-4013-2022","article-title":"Flood forecasting with machine learning models in an operational framework","volume":"26","author":"S Nevo","year":"2022","journal-title":"Hydrol Earth Syst Sci"},{"issue":"4","key":"pone.0346237.ref022","doi-asserted-by":"crossref","first-page":"2045","DOI":"10.5194\/hess-25-2045-2021","article-title":"Rainfall\u2013runoff prediction at multiple timescales with a single Long Short-Term Memory network","volume":"25","author":"M Gauch","year":"2021","journal-title":"Hydrol Earth Syst Sci"},{"issue":"1","key":"pone.0346237.ref023","doi-asserted-by":"crossref","first-page":"21446","DOI":"10.1038\/s41598-024-72205-5","article-title":"Study on long short-term memory based on vector direction of flood process for flood forecasting","volume":"14","author":"T Xie","year":"2024","journal-title":"Sci Rep"},{"issue":"1","key":"pone.0346237.ref024","article-title":"A fast physics-based perturbation generator of machine learning weather model for efficient ensemble forecasts of tropical cyclone track","volume":"8","author":"J Pu","year":"2025","journal-title":"npj Clim Atmos Sci"},{"issue":"1","key":"pone.0346237.ref025","doi-asserted-by":"crossref","first-page":"6870","DOI":"10.1038\/s41598-024-80100-2","article-title":"Research on a hybrid model for flood probability prediction based on time convolutional network and particle swarm optimization algorithm","volume":"15","author":"Q Yu","year":"2025","journal-title":"Sci Rep"},{"issue":"10","key":"pone.0346237.ref026","doi-asserted-by":"crossref","first-page":"105008","DOI":"10.1088\/2515-7620\/ae0c5c","article-title":"Probabilistic flood susceptibility mapping using explainable AI for the Western United States","volume":"7","author":"CC Ibebuchi","year":"2025","journal-title":"Environ Res Commun"},{"key":"pone.0346237.ref027","doi-asserted-by":"crossref","first-page":"102904","DOI":"10.1016\/j.ecoinf.2024.102904","article-title":"Improving the explainability of CNN-LSTM-based flood prediction with integrating SHAP technique","volume":"84","author":"H Huang","year":"2024","journal-title":"Ecological Informatics"},{"issue":"10","key":"pone.0346237.ref028","article-title":"Uncovering the Dynamic Drivers of Floods Through Interpretable Deep Learning","volume":"12","author":"Y Xu","year":"2024","journal-title":"Earth\u2019s Future"},{"issue":"1","key":"pone.0346237.ref029","doi-asserted-by":"crossref","first-page":"91","DOI":"10.1016\/j.ijsrc.2024.10.002","article-title":"Interpreting machine learning models based on SHAP values in predicting suspended sediment concentration","volume":"40","author":"H Lamane","year":"2025","journal-title":"International Journal of Sediment Research"},{"key":"pone.0346237.ref030","doi-asserted-by":"crossref","first-page":"144352","DOI":"10.1109\/ACCESS.2021.3119110","article-title":"Shapley Values for Feature Selection: The Good, the Bad, and the Axioms","volume":"9","author":"D Fryer","year":"2021","journal-title":"IEEE Access"},{"key":"pone.0346237.ref031","first-page":"102202","article-title":"Explainable artificial intelligence (XAI) for interpreting predictive models and key variables in flood susceptibility","volume":"57","author":"S Kumar","year":"2024","journal-title":"Journal of Hydrology: Regional Studies"},{"key":"pone.0346237.ref032","doi-asserted-by":"crossref","DOI":"10.1007\/978-0-387-84858-7","volume-title":"The elements of statistical learning: Data mining, inference, and prediction","author":"T Hastie","year":"2009","edition":"2"},{"key":"pone.0346237.ref033","volume-title":"Deep Learning","author":"I Goodfellow","year":"2016"},{"issue":"1","key":"pone.0346237.ref034","doi-asserted-by":"crossref","first-page":"5","DOI":"10.1023\/A:1010933404324","article-title":"Random Forests","volume":"45","author":"L Breiman","year":"2001","journal-title":"Machine Learning"},{"key":"pone.0346237.ref035","first-page":"785","article-title":"XGBoost: A scalable tree boosting system.","author":"T Chen","year":"2016","journal-title":"Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining"},{"key":"pone.0346237.ref036","doi-asserted-by":"crossref","DOI":"10.3389\/frwa.2025.1562842","article-title":"Developing an explainable deep learning module based on the LSTM framework for flood prediction","volume":"7","author":"Z Zhang","year":"2025","journal-title":"Front Water"},{"key":"pone.0346237.ref037","doi-asserted-by":"crossref","first-page":"130650","DOI":"10.1016\/j.jhydrol.2024.130650","article-title":"Multiple spatio-temporal scale runoff forecasting and driving mechanism exploration by K-means optimized XGBoost and SHAP","volume":"630","author":"S Wang","year":"2024","journal-title":"Journal of Hydrology"},{"issue":"1","key":"pone.0346237.ref038","doi-asserted-by":"crossref","first-page":"5451","DOI":"10.1038\/s41598-025-89837-w","article-title":"Multi-step ahead forecasting of daily streamflow based on the transform-based deep learning model under different scenarios","volume":"15","author":"M He","year":"2025","journal-title":"Sci Rep"},{"issue":"6","key":"pone.0346237.ref039","doi-asserted-by":"crossref","first-page":"613","DOI":"10.2166\/nh.2024.003","article-title":"Linking explainable artificial intelligence and soil moisture dynamics in a machine learning streamflow model","volume":"55","author":"A Ley","year":"2024","journal-title":"Hydrology Research"},{"issue":"1","key":"pone.0346237.ref040","doi-asserted-by":"crossref","first-page":"4824","DOI":"10.1038\/s41467-024-49115-1","article-title":"Global expansion of tropical cyclone precipitation footprint","volume":"15","author":"L Qin","year":"2024","journal-title":"Nat Commun"},{"issue":"12","key":"pone.0346237.ref041","doi-asserted-by":"crossref","first-page":"2357","DOI":"10.5194\/hess-27-2357-2023","article-title":"The suitability of differentiable, physics-informed machine learning hydrologic models for ungauged regions and climate change impact assessment","volume":"27","author":"D Feng","year":"2023","journal-title":"Hydrol Earth Syst Sci"},{"issue":"6","key":"pone.0346237.ref042","doi-asserted-by":"crossref","DOI":"10.1029\/2022WR034118","article-title":"Developing a Physics\u2010Informed Deep Learning Model to Simulate Runoff Response to Climate Change in Alpine Catchments","volume":"59","author":"L Zhong","year":"2023","journal-title":"Water Resources Research"},{"key":"pone.0346237.ref043","article-title":"Investigation of hydrometeorological influences on reservoir releases using explainable machine learning methods","volume":"5","author":"M Fan","year":"2023","journal-title":"Front Water"},{"key":"pone.0346237.ref044","doi-asserted-by":"crossref","first-page":"18","DOI":"10.1016\/j.atmosres.2013.12.015","article-title":"Rainfall asymmetries of tropical cyclones prior to, during, and after making landfall in South China and Southeast United States","volume":"139","author":"W Xu","year":"2014","journal-title":"Atmospheric Research"},{"issue":"1","key":"pone.0346237.ref045","doi-asserted-by":"crossref","first-page":"117","DOI":"10.1175\/JAMC-D-13-0359.1","article-title":"Observed Rainfall Asymmetry in Tropical Cyclones Making Landfall over China","volume":"54","author":"Z Yu","year":"2015","journal-title":"Journal of Applied Meteorology and Climatology"},{"key":"pone.0346237.ref046","doi-asserted-by":"crossref","first-page":"105160","DOI":"10.1016\/j.atmosres.2020.105160","article-title":"Generation and enhancement mechanisms for extreme orographic rainfall associated with Typhoon Morakot (2009) over the Central Mountain Range of Taiwan","volume":"247","author":"W Agyakwah","year":"2021","journal-title":"Atmospheric Research"},{"issue":"8004","key":"pone.0346237.ref047","doi-asserted-by":"crossref","first-page":"559","DOI":"10.1038\/s41586-024-07145-1","article-title":"Global prediction of extreme floods in ungauged watersheds","volume":"627","author":"G Nearing","year":"2024","journal-title":"Nature"},{"issue":"4","key":"pone.0346237.ref048","doi-asserted-by":"crossref","DOI":"10.1029\/2023WR036170","article-title":"Distributed Hydrological Modeling With Physics\u2010Encoded Deep Learning: A General Framework and Its Application in the Amazon","volume":"60","author":"C Wang","year":"2024","journal-title":"Water Resources Research"},{"key":"pone.0346237.ref049","doi-asserted-by":"crossref","first-page":"101172","DOI":"10.1016\/j.gsd.2024.101172","article-title":"Physics-informed neural networks in groundwater flow modeling: Advantages and future directions","volume":"25","author":"ASA Ali","year":"2024","journal-title":"Groundwater for Sustainable Development"},{"issue":"4","key":"pone.0346237.ref050","doi-asserted-by":"crossref","first-page":"917","DOI":"10.5194\/hess-28-917-2024","article-title":"A comprehensive study of deep learning for soil moisture prediction","volume":"28","author":"Y Wang","year":"2024","journal-title":"Hydrol Earth Syst Sci"},{"issue":"2","key":"pone.0346237.ref051","doi-asserted-by":"crossref","first-page":"1961","DOI":"10.1007\/s41748-025-00726-8","article-title":"Improving Real-Time Flood Forecasting: Probabilistic Validation of Assimilated Remotely-Sensed Soil Moisture Data","volume":"10","author":"SS Soltani","year":"2025","journal-title":"Earth Syst Environ"},{"issue":"4","key":"pone.0346237.ref052","doi-asserted-by":"crossref","first-page":"779","DOI":"10.2166\/hydro.2024.208","article-title":"Machine learning techniques for flood forecasting","volume":"26","author":"FAA Hadi","year":"2024","journal-title":"Journal of Hydroinformatics"}],"container-title":["PLOS One"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0346237","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,4,20]],"date-time":"2026-04-20T17:44:48Z","timestamp":1776707088000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0346237"}},"subtitle":[],"editor":[{"given":"Babak","family":"Mohammadi","sequence":"first","affiliation":[],"role":[{"role":"editor","vocabulary":"crossref"}]}],"short-title":[],"issued":{"date-parts":[[2026,4,20]]},"references-count":52,"journal-issue":{"issue":"4","published-online":{"date-parts":[[2026,4,20]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pone.0346237","relation":{},"ISSN":["1932-6203"],"issn-type":[{"value":"1932-6203","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,4,20]]}}}