Enhancing the accuracy and generalizability of reference evapotranspiration forecasting in California using deep global learning

Study region: This research focuses on the Central Valley of California, a climatically homogeneous region known for its significant agricultural productivity and reliance on extensive irrigation. Our study utilizes monthly reference evapotranspiration (ETO) time series data from 55 standardized wea...

Full description

Saved in:
Bibliographic Details
Main Authors: Arman Ahmadi, Andre Daccache, Minxue He, Peyman Namadi, Alireza Ghaderi Bafti, Prabhjot Sandhu, Zhaojun Bai, Richard L. Snyder, Tariq Kadir
Format: Article
Language:English
Published: Elsevier 2025-06-01
Series:Journal of Hydrology: Regional Studies
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S2214581825001648
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849716180846116864
author Arman Ahmadi
Andre Daccache
Minxue He
Peyman Namadi
Alireza Ghaderi Bafti
Prabhjot Sandhu
Zhaojun Bai
Richard L. Snyder
Tariq Kadir
author_facet Arman Ahmadi
Andre Daccache
Minxue He
Peyman Namadi
Alireza Ghaderi Bafti
Prabhjot Sandhu
Zhaojun Bai
Richard L. Snyder
Tariq Kadir
author_sort Arman Ahmadi
collection DOAJ
description Study region: This research focuses on the Central Valley of California, a climatically homogeneous region known for its significant agricultural productivity and reliance on extensive irrigation. Our study utilizes monthly reference evapotranspiration (ETO) time series data from 55 standardized weather stations as part of the California Irrigation Management Information System (CIMIS). Study focus: ETO is a critical component of regional water cycles, indicating atmospheric water demand. This study evaluates the potential of deep learning (DL) models for ETO forecasting, particularly emphasizing the efficacy of a global learning scheme compared to traditional local learning. Global learning involves training forecasting models on pooled data from multiple time series, tested over new instances. We compared the performance of statistical models and advanced DL models, demonstrating significant accuracy enhancements in global learning schemes. We also explored automatic hyperparameter optimization for these models to achieve state-of-the-art forecasting accuracy, yielding RMSE values below 10 mm/month for one-year-ahead forecasts on new, unseen stations. New hydrological insight for the region: Applying global learning methodologies to DL models markedly improved forecasting performance, showcasing an ability to generalize findings to ungauged regions and even newly established weather stations. This suggests a promising avenue for enhancing water resource management efficiency in data-scarce areas. Our findings argue that such data-centric methodological shifts could play a critical role in better managing the irrigation demands of the Central Valley, thereby supporting sustainable water usage and agricultural productivity in the region.
format Article
id doaj-art-0bcdec491782455381cac4d3bb48706c
institution DOAJ
issn 2214-5818
language English
publishDate 2025-06-01
publisher Elsevier
record_format Article
series Journal of Hydrology: Regional Studies
spelling doaj-art-0bcdec491782455381cac4d3bb48706c2025-08-20T03:13:07ZengElsevierJournal of Hydrology: Regional Studies2214-58182025-06-015910233910.1016/j.ejrh.2025.102339Enhancing the accuracy and generalizability of reference evapotranspiration forecasting in California using deep global learningArman Ahmadi0Andre Daccache1Minxue He2Peyman Namadi3Alireza Ghaderi Bafti4Prabhjot Sandhu5Zhaojun Bai6Richard L. Snyder7Tariq Kadir8Department of Environmental Science, Policy and Management, University of California, Berkeley, Berkeley, CA, USA; Corresponding author.Department of Biological and Agricultural Engineering, University of California, Davis, Davis, CA, USACalifornia Department of Water Resources, Sacramento, CA, USACalifornia Department of Water Resources, Sacramento, CA, USADepartment of Ocean Engineering, University of Rhode Island, Kingston, RI, USACalifornia Department of Water Resources, Sacramento, CA, USADepartment of Computer Science, University of California, Davis, Davis, CA, USADepartment of Land, Air and Water Resources, University of California, Davis, CA 95616, USACalifornia Department of Water Resources, Sacramento, CA, USAStudy region: This research focuses on the Central Valley of California, a climatically homogeneous region known for its significant agricultural productivity and reliance on extensive irrigation. Our study utilizes monthly reference evapotranspiration (ETO) time series data from 55 standardized weather stations as part of the California Irrigation Management Information System (CIMIS). Study focus: ETO is a critical component of regional water cycles, indicating atmospheric water demand. This study evaluates the potential of deep learning (DL) models for ETO forecasting, particularly emphasizing the efficacy of a global learning scheme compared to traditional local learning. Global learning involves training forecasting models on pooled data from multiple time series, tested over new instances. We compared the performance of statistical models and advanced DL models, demonstrating significant accuracy enhancements in global learning schemes. We also explored automatic hyperparameter optimization for these models to achieve state-of-the-art forecasting accuracy, yielding RMSE values below 10 mm/month for one-year-ahead forecasts on new, unseen stations. New hydrological insight for the region: Applying global learning methodologies to DL models markedly improved forecasting performance, showcasing an ability to generalize findings to ungauged regions and even newly established weather stations. This suggests a promising avenue for enhancing water resource management efficiency in data-scarce areas. Our findings argue that such data-centric methodological shifts could play a critical role in better managing the irrigation demands of the Central Valley, thereby supporting sustainable water usage and agricultural productivity in the region.http://www.sciencedirect.com/science/article/pii/S2214581825001648Time series forecastingGlobal learningDeep learningReference evapotranspirationWater management
spellingShingle Arman Ahmadi
Andre Daccache
Minxue He
Peyman Namadi
Alireza Ghaderi Bafti
Prabhjot Sandhu
Zhaojun Bai
Richard L. Snyder
Tariq Kadir
Enhancing the accuracy and generalizability of reference evapotranspiration forecasting in California using deep global learning
Journal of Hydrology: Regional Studies
Time series forecasting
Global learning
Deep learning
Reference evapotranspiration
Water management
title Enhancing the accuracy and generalizability of reference evapotranspiration forecasting in California using deep global learning
title_full Enhancing the accuracy and generalizability of reference evapotranspiration forecasting in California using deep global learning
title_fullStr Enhancing the accuracy and generalizability of reference evapotranspiration forecasting in California using deep global learning
title_full_unstemmed Enhancing the accuracy and generalizability of reference evapotranspiration forecasting in California using deep global learning
title_short Enhancing the accuracy and generalizability of reference evapotranspiration forecasting in California using deep global learning
title_sort enhancing the accuracy and generalizability of reference evapotranspiration forecasting in california using deep global learning
topic Time series forecasting
Global learning
Deep learning
Reference evapotranspiration
Water management
url http://www.sciencedirect.com/science/article/pii/S2214581825001648
work_keys_str_mv AT armanahmadi enhancingtheaccuracyandgeneralizabilityofreferenceevapotranspirationforecastingincaliforniausingdeepgloballearning
AT andredaccache enhancingtheaccuracyandgeneralizabilityofreferenceevapotranspirationforecastingincaliforniausingdeepgloballearning
AT minxuehe enhancingtheaccuracyandgeneralizabilityofreferenceevapotranspirationforecastingincaliforniausingdeepgloballearning
AT peymannamadi enhancingtheaccuracyandgeneralizabilityofreferenceevapotranspirationforecastingincaliforniausingdeepgloballearning
AT alirezaghaderibafti enhancingtheaccuracyandgeneralizabilityofreferenceevapotranspirationforecastingincaliforniausingdeepgloballearning
AT prabhjotsandhu enhancingtheaccuracyandgeneralizabilityofreferenceevapotranspirationforecastingincaliforniausingdeepgloballearning
AT zhaojunbai enhancingtheaccuracyandgeneralizabilityofreferenceevapotranspirationforecastingincaliforniausingdeepgloballearning
AT richardlsnyder enhancingtheaccuracyandgeneralizabilityofreferenceevapotranspirationforecastingincaliforniausingdeepgloballearning
AT tariqkadir enhancingtheaccuracyandgeneralizabilityofreferenceevapotranspirationforecastingincaliforniausingdeepgloballearning