Developing machine learning models for air temperature estimation using MODIS data

Air temperature is a key variable in a wide range of environmental applications, including land–atmosphere interaction, climate change research and hydrology and crop growth models, among others. The objective of this study was to estimate daily maximum (Tmax) and minimum (Tmin) temperatures, based...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autores principales: Ovando, Gustavo, Sayago, Silvina, Bocco, Mónica
Formato: Artículo revista
Lenguaje:Inglés
Publicado: Facultad de Ciencias Agropecuarias. 2022
Materias:
Random Forest
Artificial Neural Networks
maximum/minimum air temperature
land surface temperature
AQUA/TERRA satellite
bosques aleatorios
redes neuronales artificiales
temperatura máxima/mínima
temperatura de la superficie terrestre
satélites AQUA/TERRA
Acceso en línea:https://revistas.unc.edu.ar/index.php/agris/article/view/33225
Aporte de:
Revistas de la UNC de Universidad Nacional de Córdoba
id I10-R10-article-33225
record_format ojs
spelling I10-R10-article-332252023-04-27T15:07:43Z Developing machine learning models for air temperature estimation using MODIS data Desarrollo de modelos de aprendizaje automático para estimar temperatura del aire utilizando datos MODIS Ovando, Gustavo Sayago, Silvina Bocco, Mónica Random Forest Artificial Neural Networks maximum/minimum air temperature land surface temperature AQUA/TERRA satellite bosques aleatorios redes neuronales artificiales temperatura máxima/mínima temperatura de la superficie terrestre satélites AQUA/TERRA Air temperature is a key variable in a wide range of environmental applications, including land–atmosphere interaction, climate change research and hydrology and crop growth models, among others. The objective of this study was to estimate daily maximum (Tmax) and minimum (Tmin) temperatures, based on MODIS AQUA/TERRA land surface temperature (LST), NDVI, extraterrestrial solar radiation and precipitation data. Artificial neural networks (ANN) and random forests (RF) models were developed to predict these temperatures covering weather stations in Córdoba (Argentina) for 2018-2020. The results show that RF and ANN machine learning algorithms are capable of modeling non-linear relationships between registered temperatures and LST MODIS data, in a very robust way. The validation of the models confirms that Tmax and Tmin can be accurately estimated using, jointly or separately, AQUA and TERRA LST. The best models present determination coefficients equal to 0.81/0.91 and root mean square error of 2.7/2.1 ºC for Tmax/Tmin, when using AQUA LST day/night satellite overpass time data, respectively. The robustness and confidence of the models developed, and the ease and free accessibility of input data at a global scale, suggest that these methodologies have the potential to be applied to other regions. La temperatura del aire es una variable clave en una amplia gama de aplicaciones ambientales, que incluyen interacción tierra-atmósfera, cambio climático, modelos de cultivos e hidrológicos, entre otros. El objetivo de este estudio fue estimar las temperaturas máxima y mínima diaria, con datos de temperatura de la superficie terrestre (LST) de MODIS AQUA/TERRA, NDVI, radiación solar extraterrestre y precipitación. Se desarrollaron modelos de redes neuronales artificiales (ANN) y bosques aleatorios (RF) para predecir estas temperaturas considerando estaciones meteorológicas de Córdoba (Argentina) para el período 2018-2020. Los resultados muestran que las metodologías de RF y ANN fueron capaces de modelar relaciones no lineales entre la temperatura registrada y los datos de LST de MODIS, de manera muy robusta. La validación de los modelos confirma que Tmax y Tmin se pueden estimar con precisión utilizando, en conjunto o por separado, AQUA y TERRA LST. Los mejores modelos presentaron coeficientes de determinación iguales a 0,81/0,91 y error cuadrático medio de 2,7/2,1 ºC para Tmax/Tmin, cuando se utilizaron datos de AQUA correspondientes a día/noche, respectivamente. La solidez y el ajuste de los modelos desarrollados, sumado a la libre accesibilidad de datos a escala global, sugieren que esta metodología puede ser aplicada a otras regiones. Facultad de Ciencias Agropecuarias. 2022-06-30 info:eu-repo/semantics/article info:eu-repo/semantics/publishedVersion application/pdf application/pdf https://revistas.unc.edu.ar/index.php/agris/article/view/33225 AgriScientia; Vol. 39 No. 1 (2022); 15-28 AgriScientia; Vol. 39 Núm. 1 (2022); 15-28 1668-298X 10.31047/1668.298x.v39.n1 eng https://revistas.unc.edu.ar/index.php/agris/article/view/33225/38043 https://revistas.unc.edu.ar/index.php/agris/article/view/33225/41049 Derechos de autor 2022 Gustavo Ovando, Silvina Sayago, Mónica Bocco https://creativecommons.org/licenses/by-sa/4.0
institution Universidad Nacional de Córdoba
institution_str I-10
repository_str R-10
container_title_str Revistas de la UNC
language Inglés
format Artículo revista
topic Random Forest
Artificial Neural Networks
maximum/minimum air temperature
land surface temperature
AQUA/TERRA satellite
bosques aleatorios
redes neuronales artificiales
temperatura máxima/mínima
temperatura de la superficie terrestre
satélites AQUA/TERRA
spellingShingle Random Forest
Artificial Neural Networks
maximum/minimum air temperature
land surface temperature
AQUA/TERRA satellite
bosques aleatorios
redes neuronales artificiales
temperatura máxima/mínima
temperatura de la superficie terrestre
satélites AQUA/TERRA
Ovando, Gustavo
Sayago, Silvina
Bocco, Mónica
Developing machine learning models for air temperature estimation using MODIS data
topic_facet Random Forest
Artificial Neural Networks
maximum/minimum air temperature
land surface temperature
AQUA/TERRA satellite
bosques aleatorios
redes neuronales artificiales
temperatura máxima/mínima
temperatura de la superficie terrestre
satélites AQUA/TERRA
author Ovando, Gustavo
Sayago, Silvina
Bocco, Mónica
author_facet Ovando, Gustavo
Sayago, Silvina
Bocco, Mónica
author_sort Ovando, Gustavo
title Developing machine learning models for air temperature estimation using MODIS data
title_short Developing machine learning models for air temperature estimation using MODIS data
title_full Developing machine learning models for air temperature estimation using MODIS data
title_fullStr Developing machine learning models for air temperature estimation using MODIS data
title_full_unstemmed Developing machine learning models for air temperature estimation using MODIS data
title_sort developing machine learning models for air temperature estimation using modis data
description Air temperature is a key variable in a wide range of environmental applications, including land–atmosphere interaction, climate change research and hydrology and crop growth models, among others. The objective of this study was to estimate daily maximum (Tmax) and minimum (Tmin) temperatures, based on MODIS AQUA/TERRA land surface temperature (LST), NDVI, extraterrestrial solar radiation and precipitation data. Artificial neural networks (ANN) and random forests (RF) models were developed to predict these temperatures covering weather stations in Córdoba (Argentina) for 2018-2020. The results show that RF and ANN machine learning algorithms are capable of modeling non-linear relationships between registered temperatures and LST MODIS data, in a very robust way. The validation of the models confirms that Tmax and Tmin can be accurately estimated using, jointly or separately, AQUA and TERRA LST. The best models present determination coefficients equal to 0.81/0.91 and root mean square error of 2.7/2.1 ºC for Tmax/Tmin, when using AQUA LST day/night satellite overpass time data, respectively. The robustness and confidence of the models developed, and the ease and free accessibility of input data at a global scale, suggest that these methodologies have the potential to be applied to other regions.
publisher Facultad de Ciencias Agropecuarias.
publishDate 2022
url https://revistas.unc.edu.ar/index.php/agris/article/view/33225
work_keys_str_mv AT ovandogustavo developingmachinelearningmodelsforairtemperatureestimationusingmodisdata
AT sayagosilvina developingmachinelearningmodelsforairtemperatureestimationusingmodisdata
AT boccomonica developingmachinelearningmodelsforairtemperatureestimationusingmodisdata
AT ovandogustavo desarrollodemodelosdeaprendizajeautomaticoparaestimartemperaturadelaireutilizandodatosmodis
AT sayagosilvina desarrollodemodelosdeaprendizajeautomaticoparaestimartemperaturadelaireutilizandodatosmodis
AT boccomonica desarrollodemodelosdeaprendizajeautomaticoparaestimartemperaturadelaireutilizandodatosmodis
first_indexed 2022-08-20T01:01:49Z
last_indexed 2023-05-25T23:38:42Z
_version_ 1770717062500450304

Ejemplares similares