Estimating actual crop evapotranspiration by using satellite images coupled with hybrid deep learning-based models in potato fields

Estimating actual crop evapotranspiration (ETc act) is a fundamental requirement for effective crop water management, aiming to achieve precision irrigation amidst changing environmental conditions. Despite the introduction of various methods for estimating ETc act values, these methods are still as...

Full description

Saved in:
Bibliographic Details
Main Authors: Larona Keabetswe, Yiyin He, Chao Li, Zhenjiang Zhou
Format: Article
Language:English
Published: Elsevier 2024-12-01
Series:Agricultural Water Management
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S0378377424005274
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1846123502696923136
author Larona Keabetswe
Yiyin He
Chao Li
Zhenjiang Zhou
author_facet Larona Keabetswe
Yiyin He
Chao Li
Zhenjiang Zhou
author_sort Larona Keabetswe
collection DOAJ
description Estimating actual crop evapotranspiration (ETc act) is a fundamental requirement for effective crop water management, aiming to achieve precision irrigation amidst changing environmental conditions. Despite the introduction of various methods for estimating ETc act values, these methods are still associated with several challenges and limitations. Motivated by the robustness of deep learning models, this study employed two hybrid models that integrate Convolution Neural Network with either Random Forests (CNN-RF) or Support Vector Machine (CNN-SVR) to estimate potato ETc act using a limited set of input features. Three models were configured and compared for each CNN-RF (CNN-RF1, CNNRF2, CNNRF3) and CNN-SVM (CNN-SVM1, CNN-SVM2, CNN-SVM3), by using different combinations of variable input features derived from meteorological data (air temperature (Ta), vapour pressure deficit (VPD), net radiation (Rn)) and MODIS satellite data (land surface temperature (LST), fraction of photosynthetically active radiation (Fpar), leaf area index (LAI)). The results demonstrated the outstanding performance of the CNN-RF models, showcasing their superiority over the CNN-SVM models. Notably CNN-RF1 model yielded the best results, achieving a normalized root mean squared error (nRMSE) of 11.7 and 31.5 %; Nash–Sutcliffe Efficiency (NSE) of 0.97 and 0.80; Correlation coefficient (CC) of 0.99 and 0.91 at training and testing phases respectively. While, both models exhibited lower performance when using remote sensing satellite-based inputs, CNN-RF2, managed to produce satisfactory estimates of nRMSE = 16.7, 40.9 %; NSE = 0.95, 0.66; CC = 0.98, 0.813; Bias = −0.41, −4.377 W/m2 during training and testing respectively. The ETc act of potato showed to have high correlation with LST (0.74–0.84) and using LST as proxy for Ta, resulted in improved model estimates compared to using satellite data exclusively. These findings suggest that, in situations where climatic data is limited, remote sensing can serve as a viable alternative for modelling potato ETc act when coupled with CNN-RF, and further improvement can be achieved by incorporating meteorological data fusion.
format Article
id doaj-art-a800d14df6da436e8ba91c2923c987c6
institution Kabale University
issn 1873-2283
language English
publishDate 2024-12-01
publisher Elsevier
record_format Article
series Agricultural Water Management
spelling doaj-art-a800d14df6da436e8ba91c2923c987c62024-12-14T06:29:47ZengElsevierAgricultural Water Management1873-22832024-12-01306109191Estimating actual crop evapotranspiration by using satellite images coupled with hybrid deep learning-based models in potato fieldsLarona Keabetswe0Yiyin He1Chao Li2Zhenjiang Zhou3College of Biosystems Engineering and Food Science, Zhejiang University, Hangzhou 310058, PR ChinaCollege of Biosystems Engineering and Food Science, Zhejiang University, Hangzhou 310058, PR ChinaCollege of Biosystems Engineering and Food Science, Zhejiang University, Hangzhou 310058, PR ChinaCorresponding author.; College of Biosystems Engineering and Food Science, Zhejiang University, Hangzhou 310058, PR ChinaEstimating actual crop evapotranspiration (ETc act) is a fundamental requirement for effective crop water management, aiming to achieve precision irrigation amidst changing environmental conditions. Despite the introduction of various methods for estimating ETc act values, these methods are still associated with several challenges and limitations. Motivated by the robustness of deep learning models, this study employed two hybrid models that integrate Convolution Neural Network with either Random Forests (CNN-RF) or Support Vector Machine (CNN-SVR) to estimate potato ETc act using a limited set of input features. Three models were configured and compared for each CNN-RF (CNN-RF1, CNNRF2, CNNRF3) and CNN-SVM (CNN-SVM1, CNN-SVM2, CNN-SVM3), by using different combinations of variable input features derived from meteorological data (air temperature (Ta), vapour pressure deficit (VPD), net radiation (Rn)) and MODIS satellite data (land surface temperature (LST), fraction of photosynthetically active radiation (Fpar), leaf area index (LAI)). The results demonstrated the outstanding performance of the CNN-RF models, showcasing their superiority over the CNN-SVM models. Notably CNN-RF1 model yielded the best results, achieving a normalized root mean squared error (nRMSE) of 11.7 and 31.5 %; Nash–Sutcliffe Efficiency (NSE) of 0.97 and 0.80; Correlation coefficient (CC) of 0.99 and 0.91 at training and testing phases respectively. While, both models exhibited lower performance when using remote sensing satellite-based inputs, CNN-RF2, managed to produce satisfactory estimates of nRMSE = 16.7, 40.9 %; NSE = 0.95, 0.66; CC = 0.98, 0.813; Bias = −0.41, −4.377 W/m2 during training and testing respectively. The ETc act of potato showed to have high correlation with LST (0.74–0.84) and using LST as proxy for Ta, resulted in improved model estimates compared to using satellite data exclusively. These findings suggest that, in situations where climatic data is limited, remote sensing can serve as a viable alternative for modelling potato ETc act when coupled with CNN-RF, and further improvement can be achieved by incorporating meteorological data fusion.http://www.sciencedirect.com/science/article/pii/S0378377424005274Remote sensingCrop water useRandom ForestsLimited inputsAmeriFlux
spellingShingle Larona Keabetswe
Yiyin He
Chao Li
Zhenjiang Zhou
Estimating actual crop evapotranspiration by using satellite images coupled with hybrid deep learning-based models in potato fields
Agricultural Water Management
Remote sensing
Crop water use
Random Forests
Limited inputs
AmeriFlux
title Estimating actual crop evapotranspiration by using satellite images coupled with hybrid deep learning-based models in potato fields
title_full Estimating actual crop evapotranspiration by using satellite images coupled with hybrid deep learning-based models in potato fields
title_fullStr Estimating actual crop evapotranspiration by using satellite images coupled with hybrid deep learning-based models in potato fields
title_full_unstemmed Estimating actual crop evapotranspiration by using satellite images coupled with hybrid deep learning-based models in potato fields
title_short Estimating actual crop evapotranspiration by using satellite images coupled with hybrid deep learning-based models in potato fields
title_sort estimating actual crop evapotranspiration by using satellite images coupled with hybrid deep learning based models in potato fields
topic Remote sensing
Crop water use
Random Forests
Limited inputs
AmeriFlux
url http://www.sciencedirect.com/science/article/pii/S0378377424005274
work_keys_str_mv AT laronakeabetswe estimatingactualcropevapotranspirationbyusingsatelliteimagescoupledwithhybriddeeplearningbasedmodelsinpotatofields
AT yiyinhe estimatingactualcropevapotranspirationbyusingsatelliteimagescoupledwithhybriddeeplearningbasedmodelsinpotatofields
AT chaoli estimatingactualcropevapotranspirationbyusingsatelliteimagescoupledwithhybriddeeplearningbasedmodelsinpotatofields
AT zhenjiangzhou estimatingactualcropevapotranspirationbyusingsatelliteimagescoupledwithhybriddeeplearningbasedmodelsinpotatofields