Decomposition of plant residues of different quality in soil—DAISY model calibration and simulation based on experimental data

Abstract

A parameter setup for the DAISY model calibrated on data measured in the field was evaluated on data obtained from a lab-incubation experiment with residues of leguminous green manure plants, blue grass, rape straw and barley straw. The aim of this study was to test and further develop the principles and parameters for the turnover and the initial characterisation of these plant materials in the DAISY model. The field-calibrated parameter set led to considerable problems when applied to the lab-incubation experiment. For mineral N, soil microbial biomass N and added organic matter, none of the model simulations was fully satisfactory. The only exception was the treatment without addition of plant material. As a consequence, the parameters controlling the turnover of added organic matter and soil microbial biomass have been modified. Further conceptual changes have been suggested. It was not possible to simulate the initial decay and N release from added organic matter (AOM) by simply subdividing it into a water-insoluble part (AOM1) and a water-soluble part (AOM2). However, the C/N ratio and the cellulose content of the added plant residues may be useful indicators for the partitioning of plant materials into a slowly decomposable (AOM1) and a rapidly decomposable (AOM2) part. The general concept of two AOM-pools with predefined constant turnover rates and C/N ratios is questioned for plant residues with very different properties. After addition of easily decomposable green plant materials, death and maintenance respiration rates of modelled soil microbial biomass pools had to be reduced considerably in order to fit simulated mineral N to measured values. This is in contrast to the assumption of two SMB-pools with different but constant properties. After these modifications it was possible to achieve reliable simulations of mineral N, cumulative soil respiration and added organic matter. However, a major problem remained after recalibration. It was not possible to simulate SMB-N satisfactorily in the treatments with red clover, white clover or white melilot. It is concluded that the DAISY model does not fully reflect the flow of N through SMB after addition of easily decomposable leguminous plant materials and the following turnover into soil microbial residual N (SMR-N). The introduction of a separate SMR-pool is proposed.