LAUSR

Grassland ecosystems are among the largest terrestrial ecosystems in China, and grazing, as an important grassland management method, has direct and indirect impacts on grassland ecosystems. Meta-analyses can be used to systematically evaluate and summarize multiple findings from existing studies, but there have been few comparisons of meta-analysis methods. In this review, we summarize the effects of grazing on grassland plants and soil in the existing meta-analysis studies in China from 38 meta-analysis papers. The results show that they have consistent conclusions, such as grazing reduces the aboveground biomass by approximately half, increases the soil pH, decreases the C:N:P ratio, and reduces the number of topsoil microorganisms, but the conclusions of light and moderate grazing index changes vary greatly from study to study. The belowground biomass was generally found to increase, but it slightly decreased in some cases, and the total biomass generally decreased, but it slightly increased in other cases. Vegetation coverage increased during moderate grazing; the soil moisture content was highest for light grazing, and microbial diversity increased at light to moderate levels of grazing. There are also very inconsistent conclusions due to the different datasets and quantities of samples used in meta-analysis studies, as well as variations in the types and scales of grassland areas. The ranges of changes in other indicators were large, especially for the root-shoot ratio and soil carbon. However, changes in the aboveground biomass were generally stable. We suggest subsequent meta-analyses of grazing should further clarify the classification of grassland types and compare conclusions at different scales. Additionally, standardized network analyses are recommended for field manipulation experiments to further improve the accuracy of meta-analysis and reduce the temporal and spatial limitations of existing data.