Palaeoclimate has a major effect on the diversity of endemic species in the hotspot of mountain biodiversity in Tajikistan.

In a period of ongoing climate changes, identifying drivers of overall and endemic species diversity is a key element in constructing new ecological patterns and determining the main goals of conservation. Such studies are especially crucial if they concern biodiversity hotspot areas. In this study, we explore patterns and drivers of plant endemism (the proportion of endemic plant species to overall plant species richness; PE) in Tajikistan. We used three groups of climatic measures featuring the contemporary and glacial climates as well as climatic changes since the Last Glacial Maximum in the Pleistocene (LGM). To explore relationships between PE and climatic groups, and the most important climatic variables, we applied the Generalised Additive Model and regression trees method respectively. Glacial climate predicted PE variation the most (74.3%), followed by climate stability (55.4%) and current climate (62.4%). The most important variables represented change in precipitation of driest quarter, glacial mean annual temperature and current annual precipitation. LGM climate and its change to date have the greatest influence on contemporary PE patterns in Tajikistan, revealing the evolutionary dependencies between limited-range plants and past climate. Accordingly, annual temperature and precipitation regimes have been the most crucial drivers of PE since the LGM until today. The study revealed the dependence of the PE on a stabilized water-energy supply. The changing temperature and precipitations regimes during the ongoing climate warming may, therefore, increases the threat to geographically isolated cryophilous plants of Tajikistan, while their escape potential to suitable cold habitats is highly topographically limited.