Necromass-derived soil organic carbon and its drivers at the global scale

Abstract

Microbial residual carbon (MRC) plays an important role in maintaining soil carbon stability and is therefore a focus of soil organic carbon (SOC) estimation. However, the global pattern of MRC and its underlying determinants remains uncertain, mainly due to the lack of spatially explicit data on the ratio of gram-positive to gram-negative bacteria which are required to make the conversion from muramic acid to MRC. In the absence of such data, a constant value (65% gram-positive and 35% gram-negative) is commonly used, despite the fact that the ratio varies with environmental conditions. Here, we synthesized a comprehensive data set composed of field data on gram-positive and gram-negative bacteria from 487 sites across the globe, obtained from published literature. Our results show that the ratio is affected by a suite of environmental factors, such as soil moisture and pH. We developed a multiple linear regression model between the ratio and the influencing variables to recalculate the MRC based on amino sugar data from 654 sites. The global distribution pattern of the newly-obtained values of MRC and its driving factors were analyzed. We found striking differences in the estimation of the MRC contribution to SOC when using the newly generated coefficient compared with the currently widely used coefficient, notably in temperate grassland. MRC was greater in topsoil than in deep soil, while the converse was true for MRC/SOC. The MRC and the MRC to SOC ratio varied in different ecosystems, with the highest values of MRC and MRC/SOC occurring in temperate forest and tundra, and the lowest in desert and wetland. The key factors determining microbial residual C were microbial characteristics (microbial biomass C and N) and soil properties (SOC and total nitrogen). We provide a globally gridded map of MRC produced by the boosted regression-tree model, which further reveals that MRC in mid-high latitude areas is higher than in other regions of the northern hemisphere. These results emphasize that assessment of microbial necromass-derived SOC should consider the heterogeneity of the ratio of gram-positive to gram-negative bacteria, which would provide a huge boost to our depth of understanding of SOC composition and stability.