Search

Aim: Species–area relationships (SAR) are widely utilized for estimating the species richness and its spatial turnover across various scales. Despite the prevalent characterization of SAR using the power law in many microbial community studies, its efficacy remains unvalidated. This study aims to characterize the microbial SAR and its mechanisms in alpine grassland soils on the Qinghai‐Tibet Plateau (QTP). Location: Qinghai‐Tibet Plateau, China. Time Period: August 2014. Major Taxa Studied: Soil bacteria. Methods: Soil samples were collected from five alpine grassland sites on the QTP. Employing a nested sampling strategy at each site, soil samples were collected in plot sizes ranging from 0.5 × 0.5 m2 to 2048 × 2048 m2. Soil bacterial communities were analysed by sequencing 16S ribosomal RNA gene amplicons using an Illumina MiSeq. Results: The bacterial SAR exhibited a logarithmic power law (R2: 0.952–0.999), outperforming the power law (R2: 0.701–0.852). Consequently, the most widely adopted power law led to an overestimation of species richness by up to 15.07% in areas >256 × 256 m2, and the regional maximum theoretical richness based on Chao1 by up to 9.88%. Mechanistically, the passive sampling hypothesis was refuted through the rarefied species richness analysis, and the disproportionate effect hypothesis was rejected based on analyses of the effective numbers of species number conversions for the probability of interspecific encounters (SPIE). Notably, Pearson and multiple linear regression analyses indicated that the spatial turnover of bacterial richness was determined by the environmental heterogeneity (R2: 0.855–0.999), rather or better than environmental variables themselves, supporting the 'environment heterogeneity hypothesis'. Main Conclusions: Soil bacterial SAR in alpine grasslands exhibited a logarithmic power relationship. Spatial turnover was primarily governed by the environmental heterogeneity. In contrast, the traditional power law leads to an overestimation of soil bacterial diversity at the regional scale. [ABSTRACT FROM AUTHOR]

• Grassland degradation is predictable on a large scale by Random Forest. • Specific microbial groups, mainly Solirubrobacter, were screened as credible biomarkers. • Grassland degradation exhibited a longitudinal pattern and was correlated with current Net primary productivity. Extensive grassland degradation under climate change and intensified human activities has threatened ecological security and caused a variety of environmental problems. However, it is still challenging to predict the grassland degradation status on a large scale because it is a multi-factorial phenomenon with complex changes in ecosystem structure and function, which is hard to be fully characterized through mechanism models. The emergence of machine learning algorithms provides a potential to model complex systems and mine information from multi-source data without elucidating underlying mechanisms. Here, we utilized random forest and neural network algorithms to predict the grassland degradation represented by the net primary productivity (NPP) changing rate based on multi-source data including geographic, meteorological, plant traits, land use type and microbial variables in the Chinese Northern grassland. Particularly, the microbial roles in determining the degradation status were concerned. Results show that a high prediction precision was achieved by random forest model, rather than by neural network model, with a mean relative error of 16.9% and a mean square error of 9.273e-05. Besides identified longitude, arid index and current NPP state, specific soil microbial groups, mainly Solirubrobacter, were screened as credible biomarkers. Regarding model fitting, geographic, meteorological and plant variables explained 61.8% of the total variance, which was enhanced up to 72.8% by the rest microbial markers. These findings provide a theoretical basis to establish a pre-warning system for grassland management and policy-making. [ABSTRACT FROM AUTHOR]

• Non-linear patterns of microbial N genes occurred along precipitation gradient. • Breaking points were clearly related to certain soil types. • Attribution of different environmental factors varied by ecosystem. Despite the importance of microorganisms in soil nitrogen (N) cycling, studies on spatial patterns of microbial N-cycling gene abundances in temperate grasslands are still lacking, whose productivity is limited by N. Here, we investigated N functional genes in soil microorganisms from 60 sites in temperate grasslands across 1661 km in Inner Mongolia, China. Abundances of all N functional genes and 16S rDNA tended to decrease from northeast to southwest, consistent with the changing tend of precipitation but contrary to that of temperature. Non-linear saturation curves dominated patterns for abundances of most N functional genes and 16S rDNA along precipitation as they increased with the rising precipitation when <288–343 mm, but remained stable after these breaking points. Interestingly, these breaking points were clearly related to certain soil types, e.g. nifH did not change with precipitation in Chernozems and Gleysols, but increased with it in other soil types. Non-linear saturation patterns were also observed for most N functional gene and 16S rDNA abundances with temperature and soil total organic carbon. In contrast, no consistent spatial pattern for ratios of N functional genes: 16S rDNA was observed, which varied greatly by different N functional genes. Moreover, decay relationships were discovered for the abundance-based matrix of N functional genes over geographic distance. From the temperate meadows to typical steppes and to temperate desert steppes, the influence of relatively long-term environmental variables increased, but that of short-term ones decreased. Overall, we revealed non-linear patterns of N functional gene abundances along precipitation with a clear relationship with soil type, and discovered that attributions of geographic distance, plant community, historical-contingency and contemporary-disturbance to N functional gene community similarity decay were ecosystem–dependent. [ABSTRACT FROM AUTHOR]