Ecological constraints on elevational gradients of bird species richness in Tajikistan

The avifauna in Tajikistan has been widely studied for the last century, but specific work on species richness pattern along elevation gradients in Tajikistan is rarely investigated. Here, we reported the first study of bird species richness (BSR) in the high-altitude mountain systems (Tien Shan and Pamir-Alay) of Tajikistan which are very sensitive to the recent climate changes. We aim to explore the relationship of BSR pattern with elevation gradient and to determine the potential drivers underlying the patterns. We collected occurrence data from field surveys, published articles, and open access websites to compile a list of bird species along elevational gradients across the whole country. The BSR was counted by 100 ​m elevational bands ranging from 294 ​m to 5146 ​m. The patterns of BSR were calculated separately for five groups: all breeding birds, Passeriformes, Non-Passeriformes, large elevational range species, and small elevational range species. We calculated ecological and climatic factors of planimetric area, mid-domain effect (MDE), habitat heterogeneity (HH), mean annual temperature (MAT), temperature annual range (TAR), annual precipitation (AP), normalized difference vegetation index (NDVI), human influence index (HII), and human disturbance (HD) in each elevational band. A combination of polynomial regression, Pearson's correlation, and general least squares model analyses were used to test the effects of these factors on the BSR. A unimodal distribution pattern with a peak at 750–1950 m was observed for all breeding birds. The similar pattern was explored for Passeriformes and Non-Passeriformes, while species with different elevational range sizes had different shapes and peak elevations. For all the breeding birds and Passeriformes, BSR was significantly related to spatial, climate and human influence factors, while BSR of Non-Passeriformes positively correlated with all the given factors. First, second and fourth range classes of birds were significantly correlated with human influence factors. Moreover, large-ranged species had positive correlations with the mid-domain effect and weakly with habitat heterogeneity. We found that area, MAT and AP were the main factors to explain the richness pattern of birds, and the species richness increases with these three factors increasing. Multiple factors such as area and climate explain 84% of the variation in richness. Bivariate and multiple regression analyses revealed a consistent influence of spatial and climate factors in shaping the richness pattern for nearly all bird groups.