Abstract

BARRERA-GUZMAN, Luis Ángel et al. POTENTIAL DISTRIBUTION OF DOMESTICATED Sechium edule (CUCURBITACEAE) IN MEXICO. Acta biol.Colomb. [online]. 2022, vol.27, n.3, pp.326-335.  Epub June 11, 2024. ISSN 0120-548X.  https://doi.org/10.15446/abc.v27n3.93485.

Mexico is the centre of origin of the chayote (Sechium edule Jacq. Sw), an important plant in human consumption and in pharmaceuticals. The objective of this study was to determine the potential distribution of domesticated S. edule in Mexico using seven species distribution algorithms, to efficiently manage S. edule resources and help its conservation by identifying patterns of geographic distribution. Otherwise, areas of high suitability can be used to produce improved seed at a lower cost. 162 GBIF occurrence points and nine layers in raster format were used to evaluate seven algorithms of species distribution models. To evaluate the reliability and performance of the models, the statistics Area Under the Curve (AUC) and true skill statistic was used. Predominant climate types were Cwb (33.3 %) and Aw (17.9 %); predominant soil types were leptosol (33.3 %) and phaozem (16.7 %). The seven models showed areas of high suitability (> 0.75) in Chiapas, Guerrero, Oaxaca, Veracruz, Tabasco, Puebla and Hidalgo states. AUC values for the seven models were > 0.8 and their performance was adequate (0.4 > TSS < 0.7). Classification tree analysis was found to be the best algorithm measured by AUC (0.90); however, the seven models were adequate to explain S. edule distribution in Mexico. S. edule climatic adaptability also allows to be distributed towards the Yucatan Peninsula and western Mexico. The distribution of S. edule in Mexico, according to the studied algorithms, is limited to total annual precipitation and temperature seasonality.

Keywords : algorithms; geographical distribution; performance; temperature; weather.

        · abstract in Spanish     · text in English     · English ( pdf )