Location and differentiation of chestnut tree (Castanea sativa Mill.) individuals grafted with traditional cultivars using spatio-temporal analysis of remote-sensed imagery, dendro-chronological analysis, and SSR genotyping

The objectives of this work are to apply species distribution model (SDM) based on maximum entropy (MaxEnt) to predict locations of old chestnut (Castanea sativa Mill.) cultivars. To this end, we used geographic information systems (GIS) to represent spatial locations of individuals of chestnut cultivars non-catalogued in previous studies, and that were geolocated in preliminary fieldwork. Different environmental, climatic and anthropogenic variables were used as explanatory factors for the distribution model. The robustness demonstrated by our model, in comparison with those developed for well-known cultivars, establishes it as a reliable tool for in situ identification of new individuals among these non-catalogued cultivars, with the help of GIS to process historic and current remote sensed data. The model is expected to guide fieldwork for the verification of previously analysed data and sample collection for laboratory analyses including dendrochronological surveys and genotyping using SSR markers to genetically identify the localized cultivars. We expect that these results will allow the identification and location of traditional cultivars yielding a high value from genetic, ecological, and cultural perspectives.