Factors driving microbial biomass and necromass relationships display ecosystem‐dependent responses

Authors:
Publisher website
Google Scholar
Abstract
Microorganisms help govern soil organic carbon (SOC) turnover and accumulation. Whilst it is increasingly clear that microbial necromass is a precursor of SOC formation, the relationship between living microorganisms, necromass turnover and SOC persistence remains elusive. In this study, we used phospholipid fatty acids and amino sugars to quantify living versus dead microbial carbon concentrations and evaluated the utility of each pool as an indicator of SOC persistence across a range of climates (low‐, mid‐ and high‐latitude sites) and ecotypes (old‐growth forests vs. managed croplands). We found that microbial necromass was higher in forest than in cropland soils and was positively correlated with soil moisture, SOC and total nitrogen (TN). However, the flow of microbial biomass into necromass and SOC was decoupled in forest sites, likely because the high soil SOC/TN ratio accelerated necromass turnover and recycling by living microorganisms. In contrast, microbial biomass and necromass pools were tightly coupled in croplands and influenced by multiple environmental and biological factors (e.g., necromass concentrations exhibited greater variability in soils with more bacteria than fungi, and those with more gram‐positive than gram‐negative taxa). Contrasting our expectations, the proportion of microbially‐derived necromass in SOC was decoupled from soil properties and microbial biomass in both ecotypes. Whilst SOC and pH appear to be universal drivers of necromass cycling, feedbacks between living biomass, necromass and SOC are shaped by local factors. Our results contribute to ecological theory by highlighting the environmental and biological factors underpinning SOC formation and turnover that can be used to inform land‐management practices that optimize below‐ground carbon sequestration.
Funding Information
  • National Natural Science Foundation of China (31930070, 32241037, 41977051)

This publication has 87 references indexed in Scilit: