View in EDS

Large‐Scale Drought Forecasting in the U.S. Southern Plains Through a Hybrid Cluster‐Based Wavelet‐Machine Learning Approach.

Saved in:
Bibliographic Details
Title: Large‐Scale Drought Forecasting in the U.S. Southern Plains Through a Hybrid Cluster‐Based Wavelet‐Machine Learning Approach.
Authors: Lee, SangHyun, Danandeh Mehr, Ali, Moriasi, Daniel, Mirchi, Ali
Source: Water Resources Research; Nov2025, Vol. 61 Issue 11, p1-23, 23p
Subject Terms: DROUGHT forecasting, MACHINE learning, WAVELET transforms, CLIMATE change adaptation, HYDROLOGIC models, PLAINS, SPATIAL variation, EVAPOTRANSPIRATION
Abstract: High‐resolution gridded data sets provide valuable opportunities to enhance drought forecasting, but applying complex machine learning algorithms across large spatial domains is computationally challenging. This study presents a novel hybrid approach for forecasting the gridded Standardized Precipitation‐Evapotranspiration Index (SPEI) across the U.S. Southern Plains (SP), with lead times of 1 and 3 months. We developed a clustering‐based method using 21 centroid grid cells, each representing a unique cluster of similar grid cells based on various hydrologic characteristics, to train and evaluate multilayer perceptrons (MLPs), long short‐term memory (LSTM), and genetic programming (GP). Based on the superior performance of the trained MLPs in terms of Nash‐Sutcliffe efficiency and root‐mean‐square error, they were extended to corresponding grid cells for each cluster, enabling spatially adaptive drought prediction at a high resolution. The use of discrete wavelet transform (DWT) further enhanced model accuracy by capturing key temporal patterns in the SPEI series. Notably, our results showed that physical and hydrologic attributes strongly influenced input selections. While a 12‐month lag period worked well in regions with weaker seasonality, areas with strong seasonality benefited from selection of effective lags by using mutual information. For 3‐month‐ahead forecasts, including decomposed potential evapotranspiration in addition to precipitation as inputs improved accuracy in drier regions but decreased accuracy in humid areas. The forecast maps based on the hybrid DWT‐MLP models effectively captured the spatial variability of drought, with high correlations to observed values, demonstrating their effectiveness for regional drought early warning systems to inform water resources management adaptations. Plain Language Summary: This study presents a new method to improve drought forecasting for the U.S. Southern Plains, aiming to provide early warnings and support water planning efforts. Drought forecasts are challenging due to the complexity of processing large environmental data sets, especially when computing power is limited. We resolved this issue by focusing on key regions that share similar drought characteristics across neighboring areas and trained machine learning models only for these areas. We applied these trained models to the areas showing similar drought patterns so that the machine learning models can predict drought conditions for the entire homogeneous area. Our approach captures essential large‐scale environmental patterns by combining two techniques: a method that helps the model recognize important weather patterns, and a second method for selecting the most relevant input variables based on regional conditions. The results show this model accurately predicts drought conditions across the Southern Plains region one to 3 months in advance. Our findings suggest that the presented modeling framework can serve as a valuable tool for creating drought forecast maps to help communities prepare for droughts in a timely manner. Key Points: A computationally efficient cluster‐based hybrid machine learning (ML) approach is presented for large‐scale drought predictionThe forecast maps capture drought patterns across the Southern Plains region with high accuracy at 1‐ and 3‐month lead timesEffective lag and feature selection for ML models are strongly influenced by local weather factors such as aridity and seasonality [ABSTRACT FROM AUTHOR]
Copyright of Water Resources Research is the property of Wiley-Blackwell and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
Database: Biomedical Index
Description
Abstract:High‐resolution gridded data sets provide valuable opportunities to enhance drought forecasting, but applying complex machine learning algorithms across large spatial domains is computationally challenging. This study presents a novel hybrid approach for forecasting the gridded Standardized Precipitation‐Evapotranspiration Index (SPEI) across the U.S. Southern Plains (SP), with lead times of 1 and 3 months. We developed a clustering‐based method using 21 centroid grid cells, each representing a unique cluster of similar grid cells based on various hydrologic characteristics, to train and evaluate multilayer perceptrons (MLPs), long short‐term memory (LSTM), and genetic programming (GP). Based on the superior performance of the trained MLPs in terms of Nash‐Sutcliffe efficiency and root‐mean‐square error, they were extended to corresponding grid cells for each cluster, enabling spatially adaptive drought prediction at a high resolution. The use of discrete wavelet transform (DWT) further enhanced model accuracy by capturing key temporal patterns in the SPEI series. Notably, our results showed that physical and hydrologic attributes strongly influenced input selections. While a 12‐month lag period worked well in regions with weaker seasonality, areas with strong seasonality benefited from selection of effective lags by using mutual information. For 3‐month‐ahead forecasts, including decomposed potential evapotranspiration in addition to precipitation as inputs improved accuracy in drier regions but decreased accuracy in humid areas. The forecast maps based on the hybrid DWT‐MLP models effectively captured the spatial variability of drought, with high correlations to observed values, demonstrating their effectiveness for regional drought early warning systems to inform water resources management adaptations. Plain Language Summary: This study presents a new method to improve drought forecasting for the U.S. Southern Plains, aiming to provide early warnings and support water planning efforts. Drought forecasts are challenging due to the complexity of processing large environmental data sets, especially when computing power is limited. We resolved this issue by focusing on key regions that share similar drought characteristics across neighboring areas and trained machine learning models only for these areas. We applied these trained models to the areas showing similar drought patterns so that the machine learning models can predict drought conditions for the entire homogeneous area. Our approach captures essential large‐scale environmental patterns by combining two techniques: a method that helps the model recognize important weather patterns, and a second method for selecting the most relevant input variables based on regional conditions. The results show this model accurately predicts drought conditions across the Southern Plains region one to 3 months in advance. Our findings suggest that the presented modeling framework can serve as a valuable tool for creating drought forecast maps to help communities prepare for droughts in a timely manner. Key Points: A computationally efficient cluster‐based hybrid machine learning (ML) approach is presented for large‐scale drought predictionThe forecast maps capture drought patterns across the Southern Plains region with high accuracy at 1‐ and 3‐month lead timesEffective lag and feature selection for ML models are strongly influenced by local weather factors such as aridity and seasonality [ABSTRACT FROM AUTHOR]
ISSN:00431397
DOI:10.1029/2024WR039744