Abstract
The recognition of spatial distribution of climatic variables is essential for planning land use and occupation. This is especially relevant in regions where the agricultural frontier is expanding, as it is the case of Tocantins State, Brazil. One of the tools widely used for spatialization of environmental variables is geostatistics, which allows the identification of the spatial dependence of these variables. In this context, this study evaluates the performance of geostatistical interpolators ordinary kriging (OK) and cokriging (CK) by adjusting different semivariogram models, and the subsequent spatialization of the variables mean air temperature, insolation, air relative humidity, and potential evapotranspiration for Tocantins State. The main results and conclusions were: i) variogram analysis is essential to improve the mapping results of each variable; ii) cross-validation showed acceptable errors, indicating reliability of results; (iii) the OK outperformed CK, which can be explained by the good spatial dependence structure presented by the primary variable; and iv) the maps produced can corroborate the management of natural resources and land use planning in Tocantins State.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have