1 Introduction
Large-scale spatial distribution patterns of species richness and their formation mechanisms are central to ecology and biogeography. It is also the basic scale for measuring regional diversity and the basis for constructing evolutionary and ecological models and conservation strategies (Gotelli & Colwell, 2001; Jenkins, Pimm & Joppa, 2013; D’Antraccoli et al., 2019). The prediction of species ranges can usually be achieved in three ways: collection or surveyed species distribution record points, expert mapping of species distributions, and the ranges inferred from species distribution models (Guisan & Thuiller, 2005). At present, species distribution models (SDM) are frequently used in studies on species distribution prediction because of their relative flexibility and better discriminatory and predictive power. SDM can use the relationship between species distribution points and local environmental variables to predict the potential distribution areas of species (Zhang et al., 2019; Abdulwahab, Hammill & Hawkins, 2022; Sanczuk et al., 2022). China is a vast territory, and covering all biological surveys is challenging. Therefore, SDM can guide future field surveys to a certain extent, provide references for further exploration and discovery of species distribution, and provide a scientific basis for the formulation of species protection measures (Nguyen & Leung, 2022). Among the simulation methods of various distribution models, the maximum entropy model uses environmental variables and species distribution sites to calculate constraints in the case of a small sample size. It explores the possible distribution of maximum entropy under this constraint to predict the habitat suitability of species in the study area, resulting in better simulation results than other models (Wang et al., 2021).
How the interactions of the modern environment, evolutionary history, and ecological processes shape the patterns of species richness remains an interesting but controversial issue in biogeography. Ecologists have been trying to determine the effects of various environmental variables on the distribution and diversity of organisms in different ecological regions. The factors determining richness patterns are critical for understanding the structure and dynamics of organisms in an area (Holt et al., 2018). Species are not randomly distributed over the land surface; rather, their distribution patterns are based on climate, topography, and anthropogenic influences in recent decades (Li et al., 2015; Xu et al., 2019). Consequently, various theories and hypotheses have been developed to explain how geographical patterns of species richness are formed.
The energy-water hypothesis is the most common and discussed hypothesis for explaining species richness patterns (Hawkins et al., 2003; Pandey et al., 2020). This hypothesis states that the availability of energy and water determines the total plant resources that control biological activity and that total plant resources, in turn, determine changes in biodiversity (Jimenez-Alfaro et al., 2016). Second, habitat heterogeneity, another form of environmental variation that affects the production and maintenance of diversity, is considered one of the most important factors controlling species richness gradients. Increased space and shelter and opportunities for isolation and adaptation enhance species coexistence, persistence, and diversification (Stein, Gerstner & Kreft, 2014; Stein et al., 2015). Third, seasonal changes in climate and unsystematic changes in daily maximum and minimum temperatures increase organisms’ tolerance levels by altering their thermal environments, enabling them to become geographically widespread (Mi et al., 2022). Finally, human-induced environmental changes, such as habitat fragmentation, land-use changes, and disturbances, can lead to habitat loss for species (Li et al., 2015; Xu et al., 2019). These hypotheses explore the main factors influencing species richness formation based on different influencing factors.
Some studies have tested a single hypothesis (Sun et al., 2020), whereas others have tested multiple hypotheses (Gebauer et al., 2018; Ding et al., 2019; Pandey et al., 2020). A single variable or hypothesis is limited in its interpretation of species richness distribution patterns because it is a multiple-complex phenomenon that determines species richness distribution patterns. Thus, multiple modeling approaches are best suited for quantifying the contribution of various hypotheses to spatial richness distribution patterns. In the context of global biodiversity loss and concomitant climate change, attempts have been made to determine the relationships between species populations and their determinants (Xu et al., 2019; Pandey et al., 2020; Sun et al., 2020). Some studies have explained the distribution patterns of the regional richness of rodents in China (Zhou, Ma & Ye, 2002; Xing, Zhou & Ma, 2008) but have not considered the mechanisms that determine richness patterns. Chi et al. (2020, 2021) studied the distribution pattern of terrestrial mammal abundance in China and its relationship with environmental factors. Such studies include rodents in mammals, which inevitably do not fully consider their distribution pattern, resulting in studies that do not fully reflect the distribution pattern of rodents. Moreover, few studies have been on the richness distribution patterns of endemic and non-endemic rodent groups in China. Endemic species are those found only in specific locations or regions, not anywhere else in the world. They are usually restricted to a limited geographic range, with small ranges and population sizes, and sometimes with low genetic diversity and specific habitat requirements (Myers et al., 2000; Isik, 2011). Multiscale drivers and geographic distribution patterns of endemic species are also important topics in conservation biogeography because these species are particularly vulnerable to climate change and habitat degradation (Wu et al., 2016). It has been shown that there is a lack of consistency between all species richness or non-endemic species richness and endemic species richness (Orme et al., 2005; Lamoreux et al., 2006). Areas with high species richness may have many endemic species but not necessarily coherent patterns (Vetaas & Grytnes, 2002).
Therefore, we attempted to explore the geographic distribution pattern of rodents (Rodentia and Lagomorpha) in China. We divided rodent species into endemic and non-endemic species and assumed that the factors affecting endemic and non-endemic species distribution are different. We investigated the relative importance of energy-water, climatic seasonality, habitat heterogeneity, and human factors that may contribute to the distribution patterns of rodents in China. The main objectives of this research were to (1) explore the distribution pattern of rodents and their endemic and non-endemic species and (2) assess the explanatory power of energy-water, habitat heterogeneity, climate seasonality, and human factors for rodent distribution patterns in China.