Inductive predictions of hydrologic events using a Long Short-Term Memory network and the Soil and Water Assessment Tool. (June 2022)

• Record Type: Journal Article
• Authors: Majeske, Nicholas; Zhang, Xuesong; Sabaj, McKailey; Gong, Lei; Zhu, Chen; Azad, Ariful
• Is Part Of: Environmental modelling & software. Volume 152(2022)
• Publication Date: 2022-06
• Subjects: Hydrology -- Machine learning -- Long short-term memory -- Wabash river -- Little river -- Watershed; Environmental monitoring -- Computer programs -- Periodicals; Ecology -- Computer simulation -- Periodicals; Digital computer simulation -- Periodicals; Computer software -- Periodicals; Environmental Monitoring -- Periodicals; Computer Simulation -- Periodicals; Environnement -- Surveillance -- Logiciels -- Périodiques; Écologie -- Simulation, Méthodes de -- Périodiques; Simulation par ordinateur -- Périodiques; Logiciels -- Périodiques; Computer software; Digital computer simulation; Ecology -- Computer simulation; Environmental monitoring -- Computer programs; Periodicals; Electronic journals; 363.70015118
• DOI: 10.1016/j.envsoft.2022.105400
• Languages: English
• ISSNs: 1364-8152

Abstract

Abstract: We present machine learning methods to predict hydrologic features such as streamflow and soil moisture from spatially and temporally varying hydrological and meteorological data. We used a temporal reduction technique to reduce computation and memory requirements and trained a Long Short-Term Memory (LSTM) network to predict soil moisture and streamflow over multiple watersheds. We show LSTM networks can be trained in a fraction of the time required by complex process-based and attention-based models such as Soil and Water Assessment Tool (SWAT) and GeoMAN without sacrificing accuracy. We also demonstrate that outside data - sourced from a watershed other than the target - can be used to train LSTM to comparable or even superior prediction accuracy. The success of LSTM in such spatially-inductive settings shows hydrologic features can be predicted with minimal prior knowledge of the watershed in question. Finally, we make all methodologies of this work publicly available as an end-to-end software pipeline that facilitates rapid prototyping of hydrologic learners. Graphical abstract: Image 1 Highlights: Achieved highly accurate spatiotemporal inductions for hydrological features within and across watersheds. LSTM delivers predictive performance comparable to more sophisticated process-based and ML models at a fraction of the time. Demonstrated the efficacy of LSTM as a complimentary model of a process-based model (SWAT). Developed a modular software pipeline for end-to-end analysis of spatiotemporal hydrologic data.