Mesh data of air temperature in a geographical information system are useful for the precise management of farmland with unique meteorology, especially in hilly and mountainous areas. However, there have been few cases in which the application of mesh data has increased agricultural productivity because the mesh data have been too incomplete to be handled as practical data for agriculture. Thus, a practical method of developing mesh data of air temperature is reviewed; this paper is a review of author studies. Air temperature data are estimated from the variation in air temperature between an estimation site and an existing observation site. There are two factors behind the differences in air temperature: the radiative cooling intensity at the estimation site and the radiative cooling intensity at the existing observation site. Furthermore, a method of compiling high-resolution mesh data from coupled general circulation model data to make quantitative assessments of the effects of climate change is presented. A correction model of downscaling data employing a regional climate model was developed through stepwise multiple regression analysis using the variation in monthly mean air temperatures between the observation and regional climate model output as a dependent variable and geographical factors as independent variables.