Effect of the net radiation proxies on maize and soya evapotranspiration estimation using machine learning methods.
Accurate evapotranspiration (ET) estimation is essential for water management in crops, but it is not an easy task. Empirical ET methodologies require precise net radiation (Rn) measurements to obtain accurate results. Nevertheless, Rn measurements are not easy to obtain from meteorological stations...
| Main Authors: | , , , |
|---|---|
| Format: | Online |
| Language: | eng |
| Published: |
Facultad de Ciencias Agropecuarias
2022
|
| Subjects: | |
| Online Access: | https://revistas.unc.edu.ar/index.php/agris/article/view/37104 |
| _version_ | 1834859335741603840 |
|---|---|
| author | Venturini, Virginia Walker, Elisabet Fonnegra Mora, Diana Carolina Fagioli, Gianfranco Walker, Elisabet |
| author_facet | Venturini, Virginia Walker, Elisabet Fonnegra Mora, Diana Carolina Fagioli, Gianfranco Walker, Elisabet |
| author_sort | Venturini, Virginia |
| collection | Portal de Revistas |
| description | Accurate evapotranspiration (ET) estimation is essential for water management in crops, but it is not an easy task. Empirical ET methodologies require precise net radiation (Rn) measurements to obtain accurate results. Nevertheless, Rn measurements are not easy to obtain from meteorological stations. Thus, this study explored the use of machine learning algorithms with two Rn substitutes, to estimate daily ET: the extraterrestrial solar radiation (Ra) and a modelled Rn (RnM). Support Vector Machine (SVM), Kernel Ridge (KR), Decision Tree (DT), Adaptive Boosting (AB), and Multilayer Perceptron (MLP) were applied to model FLUXNET Rn and ET observations. Adaptive Boosting produced the best field Rn measurements (RnO), yielding a Root Mean Square Error of about 16 % of the mean observed Rn. The resulting Rn (AB RnM) was used to model daily crops ET employing the above-mentioned machine learning methods with RnO, AB RnM, and Ra, in conjunction with meteorological variables and the NDVI index. The evaluated methods were suitable to estimate ET, yielding similar errors to those obtained with RnO, when contrasted with ET observations. These results demonstrate that AB and KR are applicable with rutinary meteorological and satellite data to estimate ET. |
| format | Online |
| id | oai:ojs.revistas.unc.edu.ar:article-37104 |
| institution | Universidad Nacional de Cordoba |
| language | eng |
| publishDate | 2022 |
| publisher | Facultad de Ciencias Agropecuarias |
| record_format | ojs |
| spelling | oai:ojs.revistas.unc.edu.ar:article-371042023-11-30T18:44:58Z Effect of the net radiation proxies on maize and soya evapotranspiration estimation using machine learning methods. Efecto de los sustitutos de radiación neta en la estimación de la evapotranspiración del maíz y la soja mediante métodos de aprendizaje automático Venturini, Virginia Walker, Elisabet Fonnegra Mora, Diana Carolina Fagioli, Gianfranco Walker, Elisabet water stress net radiation crops Adaptive Boosting estrés hídrico radiación neta cultivos aprendizaje automático acelerador adaptativo Accurate evapotranspiration (ET) estimation is essential for water management in crops, but it is not an easy task. Empirical ET methodologies require precise net radiation (Rn) measurements to obtain accurate results. Nevertheless, Rn measurements are not easy to obtain from meteorological stations. Thus, this study explored the use of machine learning algorithms with two Rn substitutes, to estimate daily ET: the extraterrestrial solar radiation (Ra) and a modelled Rn (RnM). Support Vector Machine (SVM), Kernel Ridge (KR), Decision Tree (DT), Adaptive Boosting (AB), and Multilayer Perceptron (MLP) were applied to model FLUXNET Rn and ET observations. Adaptive Boosting produced the best field Rn measurements (RnO), yielding a Root Mean Square Error of about 16 % of the mean observed Rn. The resulting Rn (AB RnM) was used to model daily crops ET employing the above-mentioned machine learning methods with RnO, AB RnM, and Ra, in conjunction with meteorological variables and the NDVI index. The evaluated methods were suitable to estimate ET, yielding similar errors to those obtained with RnO, when contrasted with ET observations. These results demonstrate that AB and KR are applicable with rutinary meteorological and satellite data to estimate ET. La estimación precisa de la evapotranspiración (ET) es esencial para gestionar agua en cultivos, pero no es una tarea fácil. Las metodologías empíricas de ET requieren mediciones precisas de la radiación neta (Rn) para obtener resultados confiables. Sin embargo, estas mediciones no son rutinarias en las estaciones meteorológicas. Este trabajo exploró el uso de aprendizaje automático para estimar la ET diaria con dos sustitutos de Rn: la radiación solar extraterrestre (Ra) y la Rn modelada (RnM). Se utilizó Support Vector Machine (SVM), Kernel Ridge (KR), Decision Tree (DT), Adaptive Boosting (AB) y Multilayer Perceptron (MLP) para modelar observaciones de FLUXNET. Adaptive Boosting brindó los mejores resultados con observaciones de Rn (RnO), con un valor para la raíz del error cuadrático medio de aproximadamente el 16 % de Rn medio observado. La Rn resultante (AB RnM) se utilizó para modelar la ET, usando RnO, AB RnM y Ra, junto a variables meteorológicas y el índice NDVI. Los métodos evaluados estimaron adecuadamente la ET, arrojando errores similares a los obtenidos con RnO, cuando se contrastan con las observaciones de ET. Estos resultados demuestran que AB y KR son aplicables con datos rutinarios meteorológicos y de satélite para estimar la ET. Facultad de Ciencias Agropecuarias 2022-12-30 info:eu-repo/semantics/article info:eu-repo/semantics/publishedVersion application/pdf text/html https://revistas.unc.edu.ar/index.php/agris/article/view/37104 10.31047/1668.298x.v39.n2.37104 AgriScientia; Vol. 39 No. 2 (2022); 1-17 AgriScientia; Vol. 39 Núm. 2 (2022); 1-17 1668-298X 10.31047/1668.298x.v39.n2 eng https://revistas.unc.edu.ar/index.php/agris/article/view/37104/40237 https://revistas.unc.edu.ar/index.php/agris/article/view/37104/41184 Derechos de autor 2022 Virginia Venturini, Elisabet Walker, Diana Carolina Fonnegra Mora, Gianfranco Fagioli https://creativecommons.org/licenses/by-sa/4.0 |
| spellingShingle | water stress net radiation crops Adaptive Boosting estrés hídrico radiación neta cultivos aprendizaje automático acelerador adaptativo Venturini, Virginia Walker, Elisabet Fonnegra Mora, Diana Carolina Fagioli, Gianfranco Walker, Elisabet Effect of the net radiation proxies on maize and soya evapotranspiration estimation using machine learning methods. |
| title | Effect of the net radiation proxies on maize and soya evapotranspiration estimation using machine learning methods. |
| title_alt | Efecto de los sustitutos de radiación neta en la estimación de la evapotranspiración del maíz y la soja mediante métodos de aprendizaje automático |
| title_full | Effect of the net radiation proxies on maize and soya evapotranspiration estimation using machine learning methods. |
| title_fullStr | Effect of the net radiation proxies on maize and soya evapotranspiration estimation using machine learning methods. |
| title_full_unstemmed | Effect of the net radiation proxies on maize and soya evapotranspiration estimation using machine learning methods. |
| title_short | Effect of the net radiation proxies on maize and soya evapotranspiration estimation using machine learning methods. |
| title_sort | effect of the net radiation proxies on maize and soya evapotranspiration estimation using machine learning methods |
| topic | water stress net radiation crops Adaptive Boosting estrés hídrico radiación neta cultivos aprendizaje automático acelerador adaptativo |
| topic_facet | water stress net radiation crops Adaptive Boosting estrés hídrico radiación neta cultivos aprendizaje automático acelerador adaptativo |
| url | https://revistas.unc.edu.ar/index.php/agris/article/view/37104 |
| work_keys_str_mv | AT venturinivirginia effectofthenetradiationproxiesonmaizeandsoyaevapotranspirationestimationusingmachinelearningmethods AT walkerelisabet effectofthenetradiationproxiesonmaizeandsoyaevapotranspirationestimationusingmachinelearningmethods AT fonnegramoradianacarolina effectofthenetradiationproxiesonmaizeandsoyaevapotranspirationestimationusingmachinelearningmethods AT fagioligianfranco effectofthenetradiationproxiesonmaizeandsoyaevapotranspirationestimationusingmachinelearningmethods AT walkerelisabet effectofthenetradiationproxiesonmaizeandsoyaevapotranspirationestimationusingmachinelearningmethods AT venturinivirginia efectodelossustitutosderadiacionnetaenlaestimaciondelaevapotranspiraciondelmaizylasojamediantemetodosdeaprendizajeautomatico AT walkerelisabet efectodelossustitutosderadiacionnetaenlaestimaciondelaevapotranspiraciondelmaizylasojamediantemetodosdeaprendizajeautomatico AT fonnegramoradianacarolina efectodelossustitutosderadiacionnetaenlaestimaciondelaevapotranspiraciondelmaizylasojamediantemetodosdeaprendizajeautomatico AT fagioligianfranco efectodelossustitutosderadiacionnetaenlaestimaciondelaevapotranspiraciondelmaizylasojamediantemetodosdeaprendizajeautomatico AT walkerelisabet efectodelossustitutosderadiacionnetaenlaestimaciondelaevapotranspiraciondelmaizylasojamediantemetodosdeaprendizajeautomatico |