ECOLOGICAL NICHE OF SEMIDOMESTICATED POPULATIONS OF Capsicum pubescens RUIZ & PAV. BASED ON ACCESSIONS FROM VERACRUZ, MEXICO

To be able to cultivate a wild species it is necessary to modify the genetic scheme resulting from natural selection processes to one adapted to human-managed conditions, it implies to detect geographic areas similar to those where the species originated. This study analyzes a model of potential geographic areas for Capsicum pubescens Ruiz & Pav adaptation, aiming to detect the appropriate ecological niche conditions in Mexico, and to describe the relationships between the environment and the morphological characteristics of the fruit. The recent maximum entropy modeling algorithm (MaxEnt) was used to model the niche of C. pubescens within an important region in Central Veracruz, Mexico. A total of 44 sites of presence and four bioclimatic variables were used to detect adequate niches for the species; also, a partial least squares regression analysis was performed by combining presence sites, bioclimatic variables and morphological characteristics of the fruit. A final suitability map was built identifying the areas suitable for C. pubescens growth. The contributions of the predictor variables to the model were annual rainfall (Bio12) 43.9 %, potassium layer (K) 23 %, altitude (DEM) 22.3 % and mean annual temperature (Bio1) 10.7 %, with a value of area under the curve of 99.7 %. The partial minimum squares corroborated the importance of covariables, which intervene in the expression of morphological characteristics of the fruit, helping to better understand the relationships between species and the environment. Areas not yet explored had occurrence values over 90 %, mainly in the mountainous regions of Chihuahua, Tamaulipas, Nuevo Leon, and the Santa Martha Sierra in southern Veracruz. A group of outstanding accessions was identified that could serve as the basis to initiate a breeding program for this species.