The growth and yield of important agricultural crops are severely diminished under heat stress, as well as being eminently affected by subtle changes in wind. Thus, the main objective of this study is to develop a geostatistical interpolator according to the spatial distribution of mean air temperature and wind speed in the state of Pernambuco. Annual mean air temperature and wind speed data were used to evaluate the applicability of two geostatistical interpolators to different LULC in three periods (2010, 2015, and 2020). The datasets adopted include TerraClimate and the MapBiomas Project (Collection 7). This study detected that OK's R2 proved to be a measure sensitive to the variability of the data since it differed by more than 40% among the geostatistical methods, but OK was not affected by this effect. So, this highlighted the geostatistical power of the OK regardless of the presence of outliers, as well as revealing that COK can be successfully used as an alternative interpolator of the Tair using Tmin as a covariate, since it showed only 1% difference in the R2 and CCC compared to OK, and is also able to obtain better fit of the model. Another finding was the sensitivity of the R2 to inter-annual climate variability, where the years with above-average Ws and severe drought associated with El Niño produced higher coefficients (≥ 0.73). This manuscript made it possible to propose appropriate management techniques to mitigate the effects of climate change and identify strategic development regions with high farming potential. [ABSTRACT FROM AUTHOR]