Artificial intelligence and remote sensing for spatial prediction of daily air temperature:case study of Souss watershed of Morocco

作     者：Modeste Meliho Abdellatif Khattabi Driss Zejli Collins Ashianga Orlando Caleb E.Dansou Modeste Meliho;Abdellatif Khattabi;Driss Zejli;Collins Ashianga Orlando;Caleb E.Dansou

作者机构：Ecole Nationale des Sciences Appliquées Kénitra-Laboratoire Ingénierie des Systèmes Avancés(ISA)UniversitéIbn TofailKenitraMorocco Ecole Nationale Forestière d’IngénieursSaléMorocco Ecole Nationale des Sciences Appliquées Kénitra-Laboratoire Ingénierie des Systèmes Avancés(ISA)SaléMorocco Sciences de donnéesÉcole des sciences de l’information(ESI)RabatMorocco 

出 版 物：《Geo-Spatial Information Science》 (地球空间信息科学学报（英文）)

年 卷 期：2022年第25卷第2期

页      面：244-258页

核心收录：

学科分类：07[理学] 0707[理学-海洋科学] 

主　　题：Air temperature remote sensing machine learning Morocco 

摘      要：Air temperature(Tair)is a fundamental variable in climate research and climate impact *** field observations do not accurately capture its spatial distribution due to the sparse and uneven distribution of weather stations,especially in remote areas where the local variability is *** circumvent this problem,in this study,remote sensing and weather station data were used to estimate Tair in the Souss watershed in *** statistical methods,including linear regression and partial least squares(PLS),and four machine learning algorithms,namely k-nearest neighbors,random forest(RF),extreme gradient boost,and Cubist,were used for modeling and predicting Tair and its performance were evaluated using random subsets and *** resolution imaging spectroradiometer predictors,including Terra band 32 emissivity,Terra nighttime land surface temperature,Terra local time of night observation,Aqua band 31 emissivity,Aqua daytime land surface tempera-ture,and Aqua nighttime land surface temperature(ALSTN),and auxiliary inputs,including sky-view,elevation,slope,and hillshade,were used as inputs for *** results showed that the Cubist and RF were the most accurate models(RMSE=2.09°C and 2.13°C,R2=0.91 and 0.90,respectively),while PLS had the lowest predictive power(RMSE=2.71°C;R2=0.83).The overall performance of the models for estimating Tair in the study area was generally satisfac-tory,with RMSE limited to less than 3°C for all ***,the station data reliability was still an issue,with only four of the seven stations marked by complete meteorological data.