Species and environment variables
In the study, we used data on the species catalog, which was obtained from 14 literatures listed in Table S1 (Huang et al., 2008; Zhang et al., 2007; Han et al., 2015; Xu et al., 2017; Xu et al., 2020; He, 2018; Yi, 2005; Li et al., 2004; Yang et al., 1998; Xiong, 2005; Kwon et al., 2016; Nakamura et al., 1970; Yoshifumi et al., 1994; Chuan et al., 2003). The 14 sampling sites differed in established time, region, and researchers (Table 1 and 2). The soil fauna identification may vary among plots at family levels. Thus, the order names in the samples were checked based on the Pictorial Keys to Soil Animals of China (Yin et al., 1998) and Catalogue of Life China (2013). Concurrently, all species were divided into four functional types based on their feeding guilds, as follows, saprozoic, omnivores, phytophage, and predacity (Table S2) (Zhang et al., 2001; Xu et al., 2017).
In each sampling site, soil animals were sampled from litter layer and/or soil cores in multiple seasons (Table 2). The sampling superficial area was 0.73 ± 0.58 m2 and the sampling volume was 0.071 ± 0.061 m3 (Table 2), and the sampling quantity was 339 ± 153. Furthermore, the minimum number of soil fauna individuals was 9684 and the maximum was 52673. This way, these samplings can well reflect the profile of the species composition in that particular region, because the total sampling areas were designed to adequately cover both microhabitats and plant species (Basset et al., 2012). Therefore, the effects of sampling difference on large-scale species composition among plots were negligible. Not only that, the 14 scientific studies selected in this paper declareed that the sampling completeness of each study could fully reflect the composition of local soil fauna community.
The longitude and latitude of the samples, climate factors [mean annual precipitation (MAP, mm), mean annual temperature (MAT, °C), mean temperature of the coldest months (MTCM, °C), extreme minimum temperature (EMT, °C)], and soil factors including soil organic carbon (SOC, g/kg), soil bulk density (SBD, kg/m3), and pH value were analyzed and compared among the samples. Longitude and latitude, climate factors, and soil factors were obtained from National Earth System Science Data Center (http://www.geodata.cn/), Google Earth Engine (earthengine.google.com/), or literature (Table 1 and 3).