Multivariate analysis and vegetation mapping of a biodiversity hotspot in the Hindu Kush Mountains

To explore and understand the floristic composition of one of the most important biodiversity hotspots in the Hindu Kush region of Pakistan, a multivariate analysis of species was carried out using the DCA and CCA ordination methods. Ordination plots were developed to evaluate interaction between tree-tree, tree-herbaceous plants and plants-environmental variables, using Canoco 4.6 for Windows. Results obtained for the DCA analysis of the tree species only indicate that Taxus baccata has the lowest correlation response with the other species and thus forms no association with the other native trees like Picea smithiana, Cedrus deodara and Abies pindrow. The DCA results for the locations-only data indicate that location 18 (Sulatanr and Rodingar) provides distinct microclimatic conditions for the flora and can be termed as the superior biodiversity hotspots of the area. Results also indicate that all the lowland species have positive correlation response to rock, water bodies, field (Farms) and unpaved paths and are closely associated with each other. Different tree species or group of species were found to support specific sub-flora while species i.e. Pinus roxburghii was found to be ecologically associated with no immediate sub-flora. It was observed that this species of Pine is the favourite plant used for re and afforestation purposes in the area. It is recommended that the species selection for the reforestation efforts should be revisited, and the impact of the climate change should be evaluated on these species if the ultimate aim of these projects is to improve the forest services and biodiversity of the area.