Watersheds are crucial for reveal microbial distribution patterns
Our results showed that the heterogeneous distribution of microbial diversity could be better explained by using watersheds as the units compared with grids. More endemic species were observed using watershed units (19 species by using watersheds vs 11 species by using grids). Additionally, 84.21% rare species were clustered in the same or adjacent watersheds, while only 63.16% were clustered in the same or adjacent grids.
At the genetic level, the 5 clades of A. oligospora were consistent with the natural watershed divisions and these divisions were further supported by machine learning results. The 64.70%,60.71%,80.00%,63.89% and 61.11% of the strains from clade 1 to clade 5 were correspondingly distributed in the Yangtze River, Red River, Pearl River, Mekong, and Salween-Irrawaddy watersheds respectively. These findings suggest that watersheds are the most crucial factor in explaining the observed spatial distribution pattern of A. oligospora , more so than other environmental factors. Historical events leading to the formation of Yunnan’s six large watersheds, including the uplift of Qinghai-Tibet Plateau and the spatial compression of the Hengduan Mountains, therefore likely shaped the spatial distribution of NTF genetic diversity. Because environments in mountainous region were geographically isolated by mountains, rivers, and other natural barriers, many new species evolved here (Ding et al ., 2020). As such, patterns of microbial diversity are easier to observe in this region given its clear watershed boundaries, complex terrains, high environmental heterogeneity, and abundant biodiversity.
Watersheds with clear boundaries and ecogeographical significance make them natural division units for biogeographic research. The dividing ridges between adjacent watersheds are critical dispersal barriers, the relatively isolated materials and energy flow were shaped. Previous studies have considered watersheds as ecosystem boundaries, and discussed the possibility of setting watersheds as the appropriate units in landscape ecology (Berkes et al., 1998). For example, Paula et al. (2018) found that watersheds were better units when predicting diversity based on environmental heterogeneity. The hierarchical structure of watersheds also makes them appropriate for understanding how historical and contemporary factors contribute to diversity patterns. Moving from large scales (secondary or tertiary watersheds) to smaller scales (sub-watersheds), the influence of ridge barriers hindering distribution gradually decrease while environmental influences increase. The effects of historical events and contemporary environmental background are therefore perfectly coupled by watershed units. As such, using watersheds research units should help solve most of the problems in studying the spatial distribution of microbial diversity.
Meanwhile, from the perspective of watershed, the possibility of integrating plants, animals and microorganisms biogeographic research could be provided. As with microorganisms, biogeographic research on plants and animals is also plagued by confusing or deficient research units, methods, and analyses. For instance, the biogeographic provinces between the plants and animals were not coincident. Watershed units can provide an alternative that is applicable to multiple biological groups, allowing for research that consider the interactive effects of plants, animals, and microorganisms in shaping biogeographic patterns. For the core issues of biogeography, including the historical and environmental influences on the spatial distribution, the scale effects on the biological distribution patterns could be also solved. Finally, we believe that watersheds could support explanations on the origin of biodiversity, which is a fundamental scientific problem.