Nonlinear control of climate, hydrology, and topography on streamflow response through the use of interpretable machine learning across the contiguous United States
Runoff has been greatly affected by climate change and human activities. Studying nonlinear controls on runoff response is of great significance for water resource management decision-making and ecological protection. However, there is limited understanding of what physical mechanisms dominate the r...
Main Authors: | , |
---|---|
Format: | Article |
Language: | English |
Published: |
IWA Publishing
2023-11-01
|
Series: | Journal of Water and Climate Change |
Subjects: | |
Online Access: | http://jwcc.iwaponline.com/content/14/11/4084 |
_version_ | 1797202758846119936 |
---|---|
author | Yu Wu Na Li |
author_facet | Yu Wu Na Li |
author_sort | Yu Wu |
collection | DOAJ |
description | Runoff has been greatly affected by climate change and human activities. Studying nonlinear controls on runoff response is of great significance for water resource management decision-making and ecological protection. However, there is limited understanding of what physical mechanisms dominate the runoff response and of their predictability over space. This study analyzed the spatial patterns of runoff response including runoff changes and its sensitivity to climate–landscape variations in 1,003 catchments of the contiguous United States (CONUS). Then, an interpretable machine learning method was used to investigate the nonlinear relationship between watershed attributes and runoff response, which enables the importance of influencing factors. Finally, the random forest model was employed to predict runoff response according to the predictors of catchment attributes. The results show that alteration of runoff is up to 56%/10 years due to climate change and human activities. Catchment attributes substantially altered runoff over CONUS (−60% to 56%/10 years). Climate, topography, and hydrology are the top three key factors which nonlinearly control runoff response patterns which cannot be captured by the linear correlation method. The random forest can predict runoff response well with the highest R2 of 0.96 over CONUS.
HIGHLIGHTS
Spatial patterns of runoff response were explored over 1,000+ catchments of the contiguous United States.;
Interpretable machine learning method was used to investigate the nonlinear relationship between runoff response and catchment attributes.;
Climate, hydrology, and topography are the top three dominant factors that nonlinearly control the runoff response, which can be well predicted through a random forest based on 56 indices.; |
first_indexed | 2024-03-09T09:05:18Z |
format | Article |
id | doaj.art-9d4274be539c422d8785e6e940d2298d |
institution | Directory Open Access Journal |
issn | 2040-2244 2408-9354 |
language | English |
last_indexed | 2024-04-24T08:08:32Z |
publishDate | 2023-11-01 |
publisher | IWA Publishing |
record_format | Article |
series | Journal of Water and Climate Change |
spelling | doaj.art-9d4274be539c422d8785e6e940d2298d2024-04-17T08:37:53ZengIWA PublishingJournal of Water and Climate Change2040-22442408-93542023-11-0114114084409810.2166/wcc.2023.279279Nonlinear control of climate, hydrology, and topography on streamflow response through the use of interpretable machine learning across the contiguous United StatesYu Wu0Na Li1 Heilongjiang Province Hydraulic Research Institute, Institute of Soil and Water Conservation, Harbin 150070, China Heilongjiang Agricultural Reclamation Survey and Research Institute, Harbin 150090, China Runoff has been greatly affected by climate change and human activities. Studying nonlinear controls on runoff response is of great significance for water resource management decision-making and ecological protection. However, there is limited understanding of what physical mechanisms dominate the runoff response and of their predictability over space. This study analyzed the spatial patterns of runoff response including runoff changes and its sensitivity to climate–landscape variations in 1,003 catchments of the contiguous United States (CONUS). Then, an interpretable machine learning method was used to investigate the nonlinear relationship between watershed attributes and runoff response, which enables the importance of influencing factors. Finally, the random forest model was employed to predict runoff response according to the predictors of catchment attributes. The results show that alteration of runoff is up to 56%/10 years due to climate change and human activities. Catchment attributes substantially altered runoff over CONUS (−60% to 56%/10 years). Climate, topography, and hydrology are the top three key factors which nonlinearly control runoff response patterns which cannot be captured by the linear correlation method. The random forest can predict runoff response well with the highest R2 of 0.96 over CONUS. HIGHLIGHTS Spatial patterns of runoff response were explored over 1,000+ catchments of the contiguous United States.; Interpretable machine learning method was used to investigate the nonlinear relationship between runoff response and catchment attributes.; Climate, hydrology, and topography are the top three dominant factors that nonlinearly control the runoff response, which can be well predicted through a random forest based on 56 indices.;http://jwcc.iwaponline.com/content/14/11/4084budyko frameworkclimate changehydrological responseinterpretable machine learningmachine learning |
spellingShingle | Yu Wu Na Li Nonlinear control of climate, hydrology, and topography on streamflow response through the use of interpretable machine learning across the contiguous United States Journal of Water and Climate Change budyko framework climate change hydrological response interpretable machine learning machine learning |
title | Nonlinear control of climate, hydrology, and topography on streamflow response through the use of interpretable machine learning across the contiguous United States |
title_full | Nonlinear control of climate, hydrology, and topography on streamflow response through the use of interpretable machine learning across the contiguous United States |
title_fullStr | Nonlinear control of climate, hydrology, and topography on streamflow response through the use of interpretable machine learning across the contiguous United States |
title_full_unstemmed | Nonlinear control of climate, hydrology, and topography on streamflow response through the use of interpretable machine learning across the contiguous United States |
title_short | Nonlinear control of climate, hydrology, and topography on streamflow response through the use of interpretable machine learning across the contiguous United States |
title_sort | nonlinear control of climate hydrology and topography on streamflow response through the use of interpretable machine learning across the contiguous united states |
topic | budyko framework climate change hydrological response interpretable machine learning machine learning |
url | http://jwcc.iwaponline.com/content/14/11/4084 |
work_keys_str_mv | AT yuwu nonlinearcontrolofclimatehydrologyandtopographyonstreamflowresponsethroughtheuseofinterpretablemachinelearningacrossthecontiguousunitedstates AT nali nonlinearcontrolofclimatehydrologyandtopographyonstreamflowresponsethroughtheuseofinterpretablemachinelearningacrossthecontiguousunitedstates |