Discrimination ofPicea chihuahuanaMartinez populations on the basis of climatic, edaphic, dendrometric, genetic and population traits

Abstract

BackgroundPicea chihuahuana, which is endemic to Mexico, is currently listed as “Endangered” on the Red List. Chihuahua spruce is only found in the Sierra Madre Occidental (SMO), Mexico. About 42,600 individuals are distributed in forty populations. These populations are fragmented and can be classified into three geographically distinct clusters in the SMO. The total area covered byP. chihuahuanapopulations is less than 300 ha. A recent study suggested assisted migration as an alternative to theex situconservation ofP. chihuahuana, taking into consideration the genetic structure and diversity of the populations and the predictions regarding the future climate of the habitat. However, detailed background information is required to enable development of plans for protecting and conserving species and for successful assisted migration. Thus, it is important to identify differences between populations in relation to environmental conditions. The genetic diversity of populations, which affect vigor, evolution and adaptability of the species, must also be considered. In this study, we examined 14 populations ofP.chihuahuana, with the overall aim of discriminating the populations and form clusters of this species.MethodsEach population was represented by one 50 × 50 m plot established in the center of its respective location. Climate, soil, dasometric, density variables and genetic and species diversities were assessed in these plots for further analyses. The putatively neutral and adaptive AFLP markers were used to calculate genetic diversity. Affinity Propagation (AP) clustering technique andk-means clustering algorithm were used to classify the populations in the optimal number of clusters. Later stepwise binomial logistic regression was applied to test for significant differences in variables of the southern and northernP. chihuahuanapopulations. Spearman’s correlation test was used to analyze the relationships among all variables studied.ResultsThe binomial logistic regression analysis revealed that seven climate variables, the geographical longitude and sand proportion in the soil separated the southern from northern populations. The northern populations grow in more arid and continental conditions and on soils with lower sand proportion. The mean genetic diversity using all AFLP studied ofP. chihuahuanawas significantly correlated with the mean temperature in the warmest month, where warmer temperatures are associated to larger genetic diversity. Genetic diversity ofP. chihuahuanacalculated with putatively adaptive AFLP was not statistically significantly correlated with any environmental factor.DiscussionFuture reforestation programs should take into account that at least two different groups (the northern and southern cluster) ofP. chihuahuanaexist, as local adaptation takes place because of different environmental conditions.