Abstract

OROZCO-RAMIREZ, Q.; LORENZEN, M.; FERNANDEZ DE CASTRO MARTINEZ, G.  and  CRUZ RAMIREZ, M. A.. Social and Biophysical Factors of the Forest Transition in the UNESCO Mixteca Alta Global Geopark. Invest. Geog [online]. 2022, n.108, e60465.  Epub Sep 12, 2022. ISSN 2448-7279.  https://doi.org/10.14350/rig.60465.

The analysis of land-use/cover changes is a crucial issue for Applied Geography. In Mexico, although deforestation prevails at the national level, it occurs concurrently with regional forest recovery processes - the so-called forest transition. Despite the extensive literature on forest transition in Mexico and other developing countries, few studies have simultaneously addressed the social and biophysical factors of this phenomenon. The Mixteca Alta region of the Mexican state of Oaxaca has increased in forest cover in the past decades because of the combination of these factors. This paper analyzes land-use and land-cover changes from 1967 to 2020 in the UNESCO Mixteca Alta Global Geopark (GMA, for its acronym in Spanish) based on aerial photographs (1967) and satellite imagery (2020). During the study period, oak forests increased by 59% and pine-oak forests by 349%, while grasslands declined by 39%, agricultural areas by 17%, and areas without vegetation by 68%. Overall, 5992 hectares of forest were recovered (14.5% of GMA total area). We also determined the main social and biophysical factors associated with this forest transition using three categories of change: no change, reforestation, and deforestation. This analysis was conducted with 15-meter-resolution raster data. We studied the importance of 21 variables based on Random Forest classification. The most important variables associated with forest transition were distance to the forest, geomorphological landscape, geology, slope, elevation, number of tractors, and mean temperature. The article discusses why biophysical variables were more important than social variables.

Keywords : forest transition; spatial models; random forest; raster analysis; deagrarianization.