The contributions of soil mesofauna to leaf and root litter decomposition of dominant plant species in grassland

Abstract

Litter decomposition of perennial dominant species primarily regulates carbon and nutrient cycling in grassland, which is driven by litter quality, soil micro-environment, and the decomposer organisms. However, how does soil fauna contribute to litter decomposition remains poorly understood, especially for below-ground root at different decomposition stages in grassland ecosystem. Here in upland grassland, we simultaneously examined leaf and root litter decomposition rates of relatively higher-quality (Trifolium repens L.) and lower-quality (Lolium perenne L.) perennial dominant species over time in control and fauna-excluded treatments, using litterbags with different mesh sizes to limit the accessibility of soil mesofauna. We found that the presence of soil mesofauna significantly increased annual litter mass loss by 20.01%, and could amplify the effects of litter quality on decomposition. The larger contribution of mesofauna to root litter decomposition and higher inputs of root in grassland, suggests that a stronger role of soil mesofauna in accelerating grassland ecosystem processes than previously thought, and the leaves and roots need to be considered simultaneously when evaluating the role of soil fauna in litter decomposition. Furthermore, for lower-quality species, the disparities in the contribution of soil fauna to leaf and root litter only existed in the first 7 months, but disappeared in the subsequent 5 months. However, for higher-quality species, the stronger contribution of soil mesofauna on the root litter decomposition persisted at all decomposition stages. This suggests that the decomposition stage could affect the roles of mesofauna in leaf and root litter decomposition, and the interactive effects are largely dependent on litter species quality in grassland. Overall, our findings highlight the importance of incorporating the influence of these factors in global decomposition models, which may improve the accuracy of predicting biogeochemical process.