Sensitivity analysis of six soil organic matter models applied to the decomposition of animal manures and crop residues

Submitted: 1 February 2016
Accepted: 14 May 2016
Published: 14 September 2016
Abstract Views: 1689
PDF: 788
HTML: 493
Publisher's note
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.

Authors

  • Daniele Cavalli daniele.cavalli@unimi.it Department of Agricultural and Environmental Sciences - Production, Landscape, Agroenergy, University of Milano, Italy.
  • Pietro Marino Gallina Department of Agricultural and Environmental Sciences - Production, Landscape, Agroenergy, University of Milano, Italy.
  • Luca Bechini Department of Agricultural and Environmental Sciences - Production, Landscape, Agroenergy, University of Milano, Italy.
Two features distinguishing soil organic matter simulation models are the type of kinetics used to calculate pool decomposition rates, and the algorithm used to handle the effects of nitrogen (N) shortage on carbon (C) decomposition. Compared to widely used first-order kinetics, Monod kinetics more realistically represent organic matter decomposition, because they relate decomposition to both substrate and decomposer size. Most models impose a fixed C to N ratio for microbial biomass. When N required by microbial biomass to decompose a given amount of substrate-C is larger than soil available N, carbon decomposition rates are limited proportionally to N deficit (N inhibition hypothesis). Alternatively, C-overflow was proposed as a way of getting rid of excess C, by allocating it to a storage pool of polysaccharides. We built six models to compare the combinations of three decomposition kinetics (first-order, Monod, and reverse Monod), and two ways to simulate the effect of N shortage on C decomposition (N inhibition and C-overflow). We conducted sensitivity analysis to identify model parameters that mostly affected CO2 emissions and soil mineral N during a simulated 189-day laboratory incubation assuming constant water content and temperature. We evaluated model outputs sensitivity at different stages of organic matter decomposition in a soil amended with three inputs of increasing C to N ratio: liquid manure, solid manure, and low-N crop residue. Only few model parameters and their interactions were responsible for consistent variations of CO2 and soil mineral N. These parameters were mostly related to microbial biomass and to the partitioning of applied C among input pools, as well as their decomposition constants. In addition, in models with Monod kinetics, CO2 was also sensitive to a variation of the half-saturation constants. C-overflow enhanced pool decomposition compared to N inhibition hypothesis when N shortage occurred. Accumulated C in the polysaccharides pool decomposed slowly; therefore model outputs were not sensitive to a variation of its decay constant. Six-month organic matter decomposition was generally higher for models implementing classical Monod kinetics, followed by models with first-order and reverse Monod kinetics, due to the effect of soil microbial biomass growth on decomposition rates. Moreover, models implementing Monod kinetics predicted positive priming effects of native organic matter after soil amendment, according to co-metabolism theory. Thus, priming was proportional to the increase of the microbial biomass and in turn to the decomposability of applied organic matter. We conclude that model calibration should focus only on the few important parameters.

Dimensions

Altmetric

PlumX Metrics

Downloads

Download data is not yet available.

Citations

How to Cite

Cavalli, D., Marino Gallina, P., & Bechini, L. (2016). Sensitivity analysis of six soil organic matter models applied to the decomposition of animal manures and crop residues. Italian Journal of Agronomy, 11(4), 217–236. https://doi.org/10.4081/ija.2016.757