4.1 Maxent model and environmental variables
Maxent model can connect the distribution data of species with geographic and ecological information, and calculate the maximum entropy value of species distribution, so as to predict the potential distribution area of species(Giannakopoulos et al., 2019). Maxent model can achieve a good prediction effect only by using limited species distribution data. Therefore, it is often used to simulate the potential distribution area of species, drift of potential distribution area, suitability evaluation and other studies on global climate change involving species distribution and genetic diversity in conservation biology and invasion biology(Hannah et al., 2007; Mays, Hung, Shaner, Denvir, & Primerano, 2018; Prieto-Torres, Navarro-Sigüenza, Santiago-Alarcon, & Rojas-Soto, 2016).
Environmental variables commonly used for Maxent model analysis include climate variables, landscape variables such as topography and landform, and the impact of human factors on species distribution is rarely considered (Pearson, Raxworthy, Nakamura, & Townsend, 2010; Penman, Pike, Webb, & Shine, 2010). However, the influence of human factors on ecological environment is extensive and continuous, and its role cannot be ignored(Li et al., 2014). From the contribution of the four environmental factors selected in this study, the climate factor is still the main factor affecting distribution of B. gargarizans in China. Among them, the contribution of Bio9 and Bio14 is relatively large, the contribution was 38.4% and 16.2%, and the optimal distribution range of B. gargarizans was -1.85℃ ~ 13.01℃ and 3.59mm ~ 145.20mm, respectively. Secondly, human influence also has a certain degree of contribution, and the overall influence trend shows that, within a certain range, the distribution probability of B. gargarizans increases with the enhancement of human influence (Table S5, Fig.3), this is caused by the overlap of the ecological niche of the B. gargarizans and the area of human activities and some human infrastructure is more conducive to the distribution of amphibians(Tytar, Sobolenko, Nekrasova, & Mezhzherin, 2015). Compared with other environmental factors, the overall contribution of topographic and geomorphic factors is relatively low, and see Fig.7 for details.