Modeling of air temperatures by multiple linear regressions in the Rhône-Alpes region (France): enhancement of topographic and meteorological variables by remote sensing data

摘要

With the phenomenon of urban heat island and thermal discomfort felt in urban areas, exacerbated by climate change, it isnecessary to best estimate the air temperature in every part of a territory, especially in the context of the on-goingrationalization of the Météo-France network. This study proposes to estimate the temperature of the air from 35 explanatoryvariables, notably from remote sensing using multiple linear regressions. The collinearity of the explanatory variables isanalyzed by the Pearson correlation matrix and the Variance Inflation Factor. In fine, for each day of study in each studyarea, the part of the variance explained is very high (greater than 73%). On the other hand, this estimate of the airtemperature can never be a substitute for measurements on the ground.