Quantitative aspects of heterogeneity in soil organic matter dynamics in a cool‐temperate Japanese beech forest: a radiocarbon‐based approach

Soil is the largest carbon reservoir in terrestrial ecosystems; it stores twice as much carbon as the atmosphere. It is well documented that global warming can lead to accelerated microbial decomposition of soil organic carbon (SOC) and enhance the release of CO 2 from the soil to the atmosphere; however, the magnitude and timing of this effect remain highly uncertain due to a lack of quantitative data concerning the heterogeneity of SOC biodegradability. Therefore, we sought to identify SOC pools with respect to their specific mean residence times (MRTs), to use those SOC pools to partition soil respiration sources, and to estimate the potential response of the pools to warming. We collected surface soil and litter samples from a cool‐temperate deciduous forest in Japan, chemically separated the samples into SOC fractions, estimated their MRTs based on radiocarbon ( 14 C) isotope measurements, and used the data to construct a model representing the soil as a complex of six SOC pools with different MRT ranges. We estimate that a minor, fast‐cycling SOC pool with an MRT of less than 10 years (corresponding to the O horizon and recognizable plant leaf fragments in the A1 horizon) is responsible for 73% of annual heterotrophic respiration and 44% of total soil respiration. However, the predicted response of these pools to warming demonstrates that the rate of SOC loss from the fast‐cycling SOC pool diminishes quickly (within several decades) because of limited substrate availability. In contrast, warming will continue to accelerate SOC loss from slow‐cycling pools with MRTs of 20–200 years over the next century. Although using a 14 C‐based approach has drawbacks, these estimates provide quantitative insights into the potential importance of slow‐cycling SOC dynamics for the prediction of positive feedback to climate change.