Estimation of monthly reference evapotranspiration with scarce information using machine learning in southwestern Colombia

This research aimed to identify an alternative method to estimate reference evapotranspiration (ETo) with scarce climatological information in southwestern Colombia between 1983-2017 by evaluating and comparing different machine learning techniques. The FAO Penman-Monteith (FAO-PM56) was used as the...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autores principales: Triana Madrid, Juan Camilo, Ocampo Marulanda, Camilo, Carvajal Escobar, Yesid, Torres Lopez, Wilmar Alexander, Triana, Joshua, Canchala, Teresita
Formato: Articulo
Lenguaje:Español
Publicado: 2023
Materias:
Ciencias Astronómicas
Artificial neural network
FAO-56 Penman-Monteith
Performance metrics
Southwestern Colombia
Evapotranspiration
Redes neuronales artificiales
Métricas de desempeño
Suroccidente Colombiano
Evapotranspiración
Acceso en línea:http://sedici.unlp.edu.ar/handle/10915/164620
Aporte de:
SEDICI (UNLP) de Universidad Nacional de La Plata
id I19-R120-10915-164620
record_format dspace
spelling I19-R120-10915-1646202024-04-09T20:03:20Z http://sedici.unlp.edu.ar/handle/10915/164620 Estimation of monthly reference evapotranspiration with scarce information using machine learning in southwestern Colombia Estimación de evapotranspiración de referencia con informacion escasa utilizando machine learning en el suroccidente colombiano Triana Madrid, Juan Camilo Ocampo Marulanda, Camilo Carvajal Escobar, Yesid Torres Lopez, Wilmar Alexander Triana, Joshua Canchala, Teresita 2023-10 2024-04-09T18:03:49Z es Ciencias Astronómicas Artificial neural network FAO-56 Penman-Monteith Performance metrics Southwestern Colombia Evapotranspiration Redes neuronales artificiales Métricas de desempeño Suroccidente Colombiano Evapotranspiración This research aimed to identify an alternative method to estimate reference evapotranspiration (ETo) with scarce climatological information in southwestern Colombia between 1983-2017 by evaluating and comparing different machine learning techniques. The FAO Penman-Monteith (FAO-PM56) was used as the reference method and four empirical methods (Hargreaves, Thornthwaite, Cenicafe, and Turc) were assessed with five metrics to evaluate the method of best fit to FAO-PM56, root mean square error (RMSE), mean absolute error (MAE), mean bias error (MBE), Nash-Sutcliffe model efficiency coefficient (NSE), and Pearson correlation coefficient (R). Three models were designed using machine learning techniques to estimate ETo, multiple linear regression (MLR), artificial neural networks (ANN), and autoregressive integrated moving average model (ARIMA). The results showed that the ARIMA-M3 model reported the best performance metrics (RMSE = 4.13 mm month-1, MAE = 3.15 mm month-1, MBE = -0.08 mm month-1, NSE = 0.96 and r = 0.98). However, it restricts in that it can only be used locally and cannot be extrapolated to other climatological stations, because it was calibrated with specific conditions (exogenous variables) and stations, unlike the ANN-M1 model, which only requires training the network for its application. This method will allow estimating ETo in places with scarce information, as vital for water management in places with much uncertainty regarding accessibility and availability. Esta investigación tuvo como objetivo identificar un método alternativo para estimar la evapotranspiración de referencia (ETo) con escasa información climatológica en el suroeste de Colombia entre 1983-2017, evaluando y comparando diferentes técnicas de machine learning. Se utilizó el método de FAO Penman-Monteith (FAO-PM56) como método de referencia y se evaluaron 4 métodos de empíricos (Hargreaves, Thornthwaite, Cenicafé y Turc) con cinco métricas para evaluar el método de mejor ajuste al FAO-PM56, error cuadrático medio (RMSE), error medio absoluto (MAE), error medio de sesgo (MBE), coeficiente de eficiencia del modelo de Nash-Sutcliffe (NSE) y coeficiente de correlación de Pearson (R). Se diseñaron tres modelos utilizando técnicas de machine learning para estimar la ETo, regresión lineal múltiple (MLR), redes neuronales artificiales (ANN) y modelo de media móvil integrada autorregresiva (ARIMA). Los resultados mostraron que el modelo ARIMA-M3 presentó la mejor métrica de rendimiento (RMSE = 4,13 mm mes-1, MAE = 3,15 mm mes-1, MBE = -0,08 mm mes-1, NSE = 0,96 y R = 0,98). Sin embargo, tiene la restricción de que sólo se puede utilizar localmente y no se puede extrapolar a otras estaciones climatológicas, porque se calibró con estaciones y condiciones especificas (variables exógenas), a diferencia del modelo RNA-M1, que sólo requiere entrenar la red para su aplicación. Este método permitiría estimar la ETo en lugares con escasa información, lo que es vital para la gestión del agua en lugares con mucha incertidumbre en cuanto a accesibilidad y disponibilidad. Centro Argentino de Meteorólogos Articulo Articulo http://creativecommons.org/licenses/by-nc-sa/4.0/ Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0) application/pdf
institution Universidad Nacional de La Plata
institution_str I-19
repository_str R-120
collection SEDICI (UNLP)
language Español
topic Ciencias Astronómicas
Artificial neural network
FAO-56 Penman-Monteith
Performance metrics
Southwestern Colombia
Evapotranspiration
Redes neuronales artificiales
Métricas de desempeño
Suroccidente Colombiano
Evapotranspiración
spellingShingle Ciencias Astronómicas
Artificial neural network
FAO-56 Penman-Monteith
Performance metrics
Southwestern Colombia
Evapotranspiration
Redes neuronales artificiales
Métricas de desempeño
Suroccidente Colombiano
Evapotranspiración
Triana Madrid, Juan Camilo
Ocampo Marulanda, Camilo
Carvajal Escobar, Yesid
Torres Lopez, Wilmar Alexander
Triana, Joshua
Canchala, Teresita
Estimation of monthly reference evapotranspiration with scarce information using machine learning in southwestern Colombia
topic_facet Ciencias Astronómicas
Artificial neural network
FAO-56 Penman-Monteith
Performance metrics
Southwestern Colombia
Evapotranspiration
Redes neuronales artificiales
Métricas de desempeño
Suroccidente Colombiano
Evapotranspiración
description This research aimed to identify an alternative method to estimate reference evapotranspiration (ETo) with scarce climatological information in southwestern Colombia between 1983-2017 by evaluating and comparing different machine learning techniques. The FAO Penman-Monteith (FAO-PM56) was used as the reference method and four empirical methods (Hargreaves, Thornthwaite, Cenicafe, and Turc) were assessed with five metrics to evaluate the method of best fit to FAO-PM56, root mean square error (RMSE), mean absolute error (MAE), mean bias error (MBE), Nash-Sutcliffe model efficiency coefficient (NSE), and Pearson correlation coefficient (R). Three models were designed using machine learning techniques to estimate ETo, multiple linear regression (MLR), artificial neural networks (ANN), and autoregressive integrated moving average model (ARIMA). The results showed that the ARIMA-M3 model reported the best performance metrics (RMSE = 4.13 mm month-1, MAE = 3.15 mm month-1, MBE = -0.08 mm month-1, NSE = 0.96 and r = 0.98). However, it restricts in that it can only be used locally and cannot be extrapolated to other climatological stations, because it was calibrated with specific conditions (exogenous variables) and stations, unlike the ANN-M1 model, which only requires training the network for its application. This method will allow estimating ETo in places with scarce information, as vital for water management in places with much uncertainty regarding accessibility and availability.
format Articulo
Articulo
author Triana Madrid, Juan Camilo
Ocampo Marulanda, Camilo
Carvajal Escobar, Yesid
Torres Lopez, Wilmar Alexander
Triana, Joshua
Canchala, Teresita
author_facet Triana Madrid, Juan Camilo
Ocampo Marulanda, Camilo
Carvajal Escobar, Yesid
Torres Lopez, Wilmar Alexander
Triana, Joshua
Canchala, Teresita
author_sort Triana Madrid, Juan Camilo
title Estimation of monthly reference evapotranspiration with scarce information using machine learning in southwestern Colombia
title_short Estimation of monthly reference evapotranspiration with scarce information using machine learning in southwestern Colombia
title_full Estimation of monthly reference evapotranspiration with scarce information using machine learning in southwestern Colombia
title_fullStr Estimation of monthly reference evapotranspiration with scarce information using machine learning in southwestern Colombia
title_full_unstemmed Estimation of monthly reference evapotranspiration with scarce information using machine learning in southwestern Colombia
title_sort estimation of monthly reference evapotranspiration with scarce information using machine learning in southwestern colombia
publishDate 2023
url http://sedici.unlp.edu.ar/handle/10915/164620
work_keys_str_mv AT trianamadridjuancamilo estimationofmonthlyreferenceevapotranspirationwithscarceinformationusingmachinelearninginsouthwesterncolombia
AT ocampomarulandacamilo estimationofmonthlyreferenceevapotranspirationwithscarceinformationusingmachinelearninginsouthwesterncolombia
AT carvajalescobaryesid estimationofmonthlyreferenceevapotranspirationwithscarceinformationusingmachinelearninginsouthwesterncolombia
AT torreslopezwilmaralexander estimationofmonthlyreferenceevapotranspirationwithscarceinformationusingmachinelearninginsouthwesterncolombia
AT trianajoshua estimationofmonthlyreferenceevapotranspirationwithscarceinformationusingmachinelearninginsouthwesterncolombia
AT canchalateresita estimationofmonthlyreferenceevapotranspirationwithscarceinformationusingmachinelearninginsouthwesterncolombia
AT trianamadridjuancamilo estimaciondeevapotranspiraciondereferenciaconinformacionescasautilizandomachinelearningenelsuroccidentecolombiano
AT ocampomarulandacamilo estimaciondeevapotranspiraciondereferenciaconinformacionescasautilizandomachinelearningenelsuroccidentecolombiano
AT carvajalescobaryesid estimaciondeevapotranspiraciondereferenciaconinformacionescasautilizandomachinelearningenelsuroccidentecolombiano
AT torreslopezwilmaralexander estimaciondeevapotranspiraciondereferenciaconinformacionescasautilizandomachinelearningenelsuroccidentecolombiano
AT trianajoshua estimaciondeevapotranspiraciondereferenciaconinformacionescasautilizandomachinelearningenelsuroccidentecolombiano
AT canchalateresita estimaciondeevapotranspiraciondereferenciaconinformacionescasautilizandomachinelearningenelsuroccidentecolombiano
_version_ 1807222895213019136

Ejemplares similares