Abstract

FLORES GARNICA, José Germán  and  REYES CARDENAS, Oscar. Spatial distribution of Pinus oocarpa Schiede ex Schltdl. through estimating Kernel density. Rev. mex. de cienc. forestales [online]. 2019, vol.10, n.53, pp.21-40. ISSN 2007-1132.  https://doi.org/10.29298/rmcf.v10i53.406.

In the state of Chiapas, the species Pinus oocarpa is widely distributed throughout the Sierra Madre de Chiapas, specifically in the subprovinces of the central depression, the central highlands and the eastern mountains. In the present work the spatial distribution of this species was analyzed through Kernel density analysis, using information generated by the National Inventory of Forests and Soils of Mexico (INFyS), the proposed alternative process estimates the spatial distribution of this species through Kernel density analysis. For this reason, the presence of the species can allegedly be represented as a series of georeferenced events, which occur differentially along a given region, determining spatial variations in their density. To make the comparison between the different maps that present the distribution of P. oocarpa, a confusion matrix was defined for each one of the maps, which were based on a binary classification. The distribution defined through the Kernel density estimation confirms that P. oocarpa is located basically along the Sierra Madre of Chiapas and the Central High Plateau, generally agreeing with most of the illustrations of the geographical distribution of this species. The methodology proposed herein can be used in other regions where information from forest inventories (at national, state, region, or management level) is available, by limiting the definition of the distribution of the species of interest to the area covered by the sampling design.

Keywords : Bandwidth; Kappa coefficient; interpolation; National Forest and Soil Inventory; confusion matrix; continuous surfaces.