Estimation of Potential Evapotranspiration in the Yellow River Basin Using Machine Learning Models
Potential evapotranspiration (<i>PET</i>) is an important input variable of many ecohydrological models, but commonly used empirical models usually input numerous meteorological factors. In consideration of machine learning for complex nonlinear learning, we evaluated the applicability o...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-09-01
|
| Series: | Atmosphere |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2073-4433/13/9/1467 |
| _version_ | 1827663611836760064 |
|---|---|
| author | Jie Liu Kunxia Yu Peng Li Lu Jia Xiaoming Zhang Zhi Yang Yang Zhao |
| author_facet | Jie Liu Kunxia Yu Peng Li Lu Jia Xiaoming Zhang Zhi Yang Yang Zhao |
| author_sort | Jie Liu |
| collection | DOAJ |
| description | Potential evapotranspiration (<i>PET</i>) is an important input variable of many ecohydrological models, but commonly used empirical models usually input numerous meteorological factors. In consideration of machine learning for complex nonlinear learning, we evaluated the applicability of three machine learning algorithms in <i>PET</i> estimation in the Yellow River basin (YRB), in addition to determining significant factors affecting the accuracy of machine learning. Furthermore, the importance of meteorological factors at varying altitudes and drought index grades for <i>PET</i> simulation were evaluated. The results show that the accuracy of <i>PET</i> simulation in the YRB depends on the input of various meteorological factors; however, machine learning models including average temperature (<i>T<sub>mean</sub></i>) and sunshine hours (<i>n</i>) as input achieved satisfactory accuracy in the absence of complete meteorological data. Random forest generally performed best among all investigated models, followed by extreme learning machine, whereas empirical models overestimated or underestimated <i>PET</i>. The importance index shows that <i>T<sub>mean</sub></i> is the most influential factor with respect to <i>PET</i>, followed by <i>n</i>, and the influence of <i>T<sub>mean</sub></i> on <i>PET</i> gradually decreased with increased altitude and drier climate, whereas the influence of n shows the opposite trend. |
| first_indexed | 2024-03-10T00:46:12Z |
| format | Article |
| id | doaj.art-7a9afec6072742f9b062f0ea36d6805f |
| institution | Directory Open Access Journal |
| issn | 2073-4433 |
| language | English |
| last_indexed | 2024-03-10T00:46:12Z |
| publishDate | 2022-09-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Atmosphere |
| spelling | doaj.art-7a9afec6072742f9b062f0ea36d6805f2023-11-23T15:00:05ZengMDPI AGAtmosphere2073-44332022-09-01139146710.3390/atmos13091467Estimation of Potential Evapotranspiration in the Yellow River Basin Using Machine Learning ModelsJie Liu0Kunxia Yu1Peng Li2Lu Jia3Xiaoming Zhang4Zhi Yang5Yang Zhao6State Key Laboratory of Eco-Hydraulics in Northwest Arid Region of China, Xi’an University of Technology, Xi’an 710048, ChinaState Key Laboratory of Eco-Hydraulics in Northwest Arid Region of China, Xi’an University of Technology, Xi’an 710048, ChinaState Key Laboratory of Eco-Hydraulics in Northwest Arid Region of China, Xi’an University of Technology, Xi’an 710048, ChinaState Key Laboratory of Eco-Hydraulics in Northwest Arid Region of China, Xi’an University of Technology, Xi’an 710048, ChinaState Key Laboratory of Simulation and Regulation of Water Cycle in River Basin, Institute of Water Resources and Hydropower Research, Beijing 100048, ChinaNingxia Soil and Water Conservation Monitoring Station, Yinchuan 750002, ChinaState Key Laboratory of Simulation and Regulation of Water Cycle in River Basin, Institute of Water Resources and Hydropower Research, Beijing 100048, ChinaPotential evapotranspiration (<i>PET</i>) is an important input variable of many ecohydrological models, but commonly used empirical models usually input numerous meteorological factors. In consideration of machine learning for complex nonlinear learning, we evaluated the applicability of three machine learning algorithms in <i>PET</i> estimation in the Yellow River basin (YRB), in addition to determining significant factors affecting the accuracy of machine learning. Furthermore, the importance of meteorological factors at varying altitudes and drought index grades for <i>PET</i> simulation were evaluated. The results show that the accuracy of <i>PET</i> simulation in the YRB depends on the input of various meteorological factors; however, machine learning models including average temperature (<i>T<sub>mean</sub></i>) and sunshine hours (<i>n</i>) as input achieved satisfactory accuracy in the absence of complete meteorological data. Random forest generally performed best among all investigated models, followed by extreme learning machine, whereas empirical models overestimated or underestimated <i>PET</i>. The importance index shows that <i>T<sub>mean</sub></i> is the most influential factor with respect to <i>PET</i>, followed by <i>n</i>, and the influence of <i>T<sub>mean</sub></i> on <i>PET</i> gradually decreased with increased altitude and drier climate, whereas the influence of n shows the opposite trend.https://www.mdpi.com/2073-4433/13/9/1467potential evapotranspirationmachine learningempirical modelimportance indexYellow River basin |
| spellingShingle | Jie Liu Kunxia Yu Peng Li Lu Jia Xiaoming Zhang Zhi Yang Yang Zhao Estimation of Potential Evapotranspiration in the Yellow River Basin Using Machine Learning Models Atmosphere potential evapotranspiration machine learning empirical model importance index Yellow River basin |
| title | Estimation of Potential Evapotranspiration in the Yellow River Basin Using Machine Learning Models |
| title_full | Estimation of Potential Evapotranspiration in the Yellow River Basin Using Machine Learning Models |
| title_fullStr | Estimation of Potential Evapotranspiration in the Yellow River Basin Using Machine Learning Models |
| title_full_unstemmed | Estimation of Potential Evapotranspiration in the Yellow River Basin Using Machine Learning Models |
| title_short | Estimation of Potential Evapotranspiration in the Yellow River Basin Using Machine Learning Models |
| title_sort | estimation of potential evapotranspiration in the yellow river basin using machine learning models |
| topic | potential evapotranspiration machine learning empirical model importance index Yellow River basin |
| url | https://www.mdpi.com/2073-4433/13/9/1467 |
| work_keys_str_mv | AT jieliu estimationofpotentialevapotranspirationintheyellowriverbasinusingmachinelearningmodels AT kunxiayu estimationofpotentialevapotranspirationintheyellowriverbasinusingmachinelearningmodels AT pengli estimationofpotentialevapotranspirationintheyellowriverbasinusingmachinelearningmodels AT lujia estimationofpotentialevapotranspirationintheyellowriverbasinusingmachinelearningmodels AT xiaomingzhang estimationofpotentialevapotranspirationintheyellowriverbasinusingmachinelearningmodels AT zhiyang estimationofpotentialevapotranspirationintheyellowriverbasinusingmachinelearningmodels AT yangzhao estimationofpotentialevapotranspirationintheyellowriverbasinusingmachinelearningmodels |