Linking carbon and nitrogen mineralization with microbial responses to substrate availability — the DECONIT model

Abstract

Simulation of decomposition and inorganic nitrogen release in complex biogeochemical models can be based on different principles. A major problem is the link between carbon and nitrogen mineralization and a description of microbial growth dynamics in dependence of a suite of possible substrates. This contribution considers a first order decomposition model with several carbon pools and one nitrogen pool to investigate how the decomposition of plant types and mineralization of nitrogen is related to carbon quality. The model structure assumes that nitrogen is mobilised with the rate at which the lignin compounds decompose. The decomposition module is coupled with microbial dynamics by adjusted Michaelis Menten equations that relate microbial growth to the availability of various substrates. The model was calibrated using Markov Chain Monte Carlo (MCMC) applied to measured litter remnants, concentrations of lignin, cellulose and nitrogen from 30 in situ incubations of foliage litters. Additionally, data from a laboratory incubation experiment were used to analyse the formation of microbial biomass, dissolved organic nitrogen, ammonium (NH ⁺₄ ) and microbial respiration. Parameter sensitivity was analysed according to the rate of acceptance of various settings in the MCMC calibration chain. The most important parameters for the decomposition process were the decomposition rate of lignin, and the temperature response parameter Q₁₀. The most important parameters for the formation of microbial biomass, dissolved organic nitrogen, ammonium and microbial respiration, were the potential growth rate of the microbial population and the rate of microbial decay. Estimated optimal decomposition rates for field experiments are 0.003 ± 0.002 d⁻¹ for lignin like compounds, 0.006 ± 0.004 d⁻¹ for cellulose like compounds and 0.0286 ± 0.052 d⁻¹ for solutes. The temperature response parameter Q₁₀ is 3.2 ± 0.6 and the optimum decomposition temperature is 28.1 ± 4.3°C. Important model parameters on microbial biomass and nitrification are the maximum microbial growth rate μ_MAX = 0.13 ± 0.82 gC_mic gC ⁻¹_mic d⁻¹ or the rate of microbial decay D = 0.006 ± 0.014 gC_mic gC ⁻¹_mic d⁻¹. The model performance was tested for independent datasets. Generally, correlations between modelled and measured values, expressed in R², were high for the remaining tissue dry weight, or concentrations of lignin, cellulose and solutes or organic nitrogen (R ² > 0.84). Due to uncertainties in measurements of DON and NH ⁺₄ concentrations, microbial biomass or basal respiration and significant site variability in these parameters, the model performance for these parameters as expressed as R² was somewhat lower, but statistically highly significant, and in the range of 0.1-0.96.