A new regional reference evapotranspiration model based on quantile approximation of meteorological variables
Reference evapotranspiration (ETo) is a variable that can assist in estimating agricultural water use in water-scarce regions. Estimating ETo with limited data is an important alternative to overcome the current shortage of meteorological data in many areas around the world. For this purpose, this s...
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-03-01
|
| Series: | Agricultural Water Management |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S0378377425000137 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832586788370120704 |
|---|---|
| author | Guomin Huang Jianhua Dong Lifeng Wu Jingwei Luo Rangjian Qiu Yaokui Cui Yicheng Wang |
| author_facet | Guomin Huang Jianhua Dong Lifeng Wu Jingwei Luo Rangjian Qiu Yaokui Cui Yicheng Wang |
| author_sort | Guomin Huang |
| collection | DOAJ |
| description | Reference evapotranspiration (ETo) is a variable that can assist in estimating agricultural water use in water-scarce regions. Estimating ETo with limited data is an important alternative to overcome the current shortage of meteorological data in many areas around the world. For this purpose, this study introduces a new method for establishing a simplified regional ETo model. The method, which creating ETo models based on temperature at meteorological stations that have the highest quantile matching with the target station's meteorological variables based on the closest meteorological data characteristics. To test the performance of the new method, we used data from 120 meteorological stations in Northwest China from 2000 to 2021 to develop XGBoost models to establish the new regional ETo model. We compared the proposed method with local models and two conventional regional ETo models to evaluate its performance. While the new method increased the Root Mean Square Error (RMSE) by an average of 13.4 % compared to local models, it demonstrated significant advantages over conventional regional models. Specifically, the RMSE decreased by 6.4–7.1 %, the Normalized RMSE (NRMSE) decreased by 5.5–7.3 %, computation time was reduced by 18.4–21.8 times, and spatial memory usage was reduced by 147–211 %. These improvements make the proposed method more efficient and scalable, particularly for regional applications in data-scarce areas. |
| format | Article |
| id | doaj-art-55b5c981c0144f08ae032440581559ea |
| institution | Kabale University |
| issn | 1873-2283 |
| language | English |
| publishDate | 2025-03-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Agricultural Water Management |
| spelling | doaj-art-55b5c981c0144f08ae032440581559ea2025-01-25T04:10:52ZengElsevierAgricultural Water Management1873-22832025-03-01308109299A new regional reference evapotranspiration model based on quantile approximation of meteorological variablesGuomin Huang0Jianhua Dong1Lifeng Wu2Jingwei Luo3Rangjian Qiu4Yaokui Cui5Yicheng Wang6School of Soil and Water Conservation, Nanchang Institute of Technology, Nanchang 330099, ChinaSchool of Water Resources and Hydropower Engineering, Wuhan University, Wuhan 430072, China; Corresponding author.School of Soil and Water Conservation, Nanchang Institute of Technology, Nanchang 330099, China; State Key Laboratory of Simulation and Regulation of Water Cycle in River Basin, China Institute of Water Resources and Hydropower Research, Beijing 100038, China; Corresponding author at: School of Soil and Water Conservation, Nanchang Institute of Technology, Nanchang 330099, China.School of Soil and Water Conservation, Nanchang Institute of Technology, Nanchang 330099, ChinaSchool of Water Resources and Hydropower Engineering, Wuhan University, Wuhan 430072, ChinaInstitute of RS and GIS, School of Earth and Space Sciences, Peking University, Beijing 100871, ChinaState Key Laboratory of Simulation and Regulation of Water Cycle in River Basin, China Institute of Water Resources and Hydropower Research, Beijing 100038, ChinaReference evapotranspiration (ETo) is a variable that can assist in estimating agricultural water use in water-scarce regions. Estimating ETo with limited data is an important alternative to overcome the current shortage of meteorological data in many areas around the world. For this purpose, this study introduces a new method for establishing a simplified regional ETo model. The method, which creating ETo models based on temperature at meteorological stations that have the highest quantile matching with the target station's meteorological variables based on the closest meteorological data characteristics. To test the performance of the new method, we used data from 120 meteorological stations in Northwest China from 2000 to 2021 to develop XGBoost models to establish the new regional ETo model. We compared the proposed method with local models and two conventional regional ETo models to evaluate its performance. While the new method increased the Root Mean Square Error (RMSE) by an average of 13.4 % compared to local models, it demonstrated significant advantages over conventional regional models. Specifically, the RMSE decreased by 6.4–7.1 %, the Normalized RMSE (NRMSE) decreased by 5.5–7.3 %, computation time was reduced by 18.4–21.8 times, and spatial memory usage was reduced by 147–211 %. These improvements make the proposed method more efficient and scalable, particularly for regional applications in data-scarce areas.http://www.sciencedirect.com/science/article/pii/S0378377425000137XGBoostGeneralized modelNorthwest ChinaArid and semi arid regionTemperature based modelCross station |
| spellingShingle | Guomin Huang Jianhua Dong Lifeng Wu Jingwei Luo Rangjian Qiu Yaokui Cui Yicheng Wang A new regional reference evapotranspiration model based on quantile approximation of meteorological variables Agricultural Water Management XGBoost Generalized model Northwest China Arid and semi arid region Temperature based model Cross station |
| title | A new regional reference evapotranspiration model based on quantile approximation of meteorological variables |
| title_full | A new regional reference evapotranspiration model based on quantile approximation of meteorological variables |
| title_fullStr | A new regional reference evapotranspiration model based on quantile approximation of meteorological variables |
| title_full_unstemmed | A new regional reference evapotranspiration model based on quantile approximation of meteorological variables |
| title_short | A new regional reference evapotranspiration model based on quantile approximation of meteorological variables |
| title_sort | new regional reference evapotranspiration model based on quantile approximation of meteorological variables |
| topic | XGBoost Generalized model Northwest China Arid and semi arid region Temperature based model Cross station |
| url | http://www.sciencedirect.com/science/article/pii/S0378377425000137 |
| work_keys_str_mv | AT guominhuang anewregionalreferenceevapotranspirationmodelbasedonquantileapproximationofmeteorologicalvariables AT jianhuadong anewregionalreferenceevapotranspirationmodelbasedonquantileapproximationofmeteorologicalvariables AT lifengwu anewregionalreferenceevapotranspirationmodelbasedonquantileapproximationofmeteorologicalvariables AT jingweiluo anewregionalreferenceevapotranspirationmodelbasedonquantileapproximationofmeteorologicalvariables AT rangjianqiu anewregionalreferenceevapotranspirationmodelbasedonquantileapproximationofmeteorologicalvariables AT yaokuicui anewregionalreferenceevapotranspirationmodelbasedonquantileapproximationofmeteorologicalvariables AT yichengwang anewregionalreferenceevapotranspirationmodelbasedonquantileapproximationofmeteorologicalvariables AT guominhuang newregionalreferenceevapotranspirationmodelbasedonquantileapproximationofmeteorologicalvariables AT jianhuadong newregionalreferenceevapotranspirationmodelbasedonquantileapproximationofmeteorologicalvariables AT lifengwu newregionalreferenceevapotranspirationmodelbasedonquantileapproximationofmeteorologicalvariables AT jingweiluo newregionalreferenceevapotranspirationmodelbasedonquantileapproximationofmeteorologicalvariables AT rangjianqiu newregionalreferenceevapotranspirationmodelbasedonquantileapproximationofmeteorologicalvariables AT yaokuicui newregionalreferenceevapotranspirationmodelbasedonquantileapproximationofmeteorologicalvariables AT yichengwang newregionalreferenceevapotranspirationmodelbasedonquantileapproximationofmeteorologicalvariables |