UC Irvine

Most of the carbon (C) that enters the soil is broken down by the microbial community and either respired or stored in soil depending on the microbial allocation strategy. Changes in how the microbial community uses C can significantly affect soil C pool sizes, so new models have begun to explicitly represent microbial allocation. Most models use a parameter called carbon use efficiency (CUE) to represent microbial allocation, which partitions consumed C between respiration and growth. Here we compare a “Typical Microbial Model” with this representation of microbial allocation to two other models. One is the “Microbial Allocation Model” that represents CUE as an emergent property of the microbial community, explicitly modeling multiple processes involved in CUE. The second is the “Two-Pool Biomass Model” that similarly accounts for CUE as an emergent property but also represents the biomass using two C pools with different turnover times. We assessed the models’ relative ability to track a14C-glucose tracer pulse over three weeks through the extractable C pool, the microbial biomass, and respiration. We also used the 14C data to test how estimates of microbial CUE change during the incubation. Our results suggest that CUE estimates in soil are sensitive to incubation timing and are at no point stable. Isotopic data can best parameterize models when a time course of measurements is used. Our model comparison showed that the Two-Pool Biomass Model best fit our data. Using the Two-Pool Biomass model to represent microbial allocation is more biologically realistic and better matches the dynamics observed in our microbial C partitioning data. Mechanistic biogeochemical models should consider using a dynamic allocation scheme to represent CUE, rather than assuming a static value.