Analysis of Factors Controlling Soil Organic Matter Dynamics As Affected By Management Practices: A Model Inter-Comparison Study.

A wide range of long-term experiments (LTEs) exist within Europe, and these LTEs provide unique and valuable information relevant for the farming systems in which they occur. These LTEs can be used to derive factors of soil carbon change under different management and climate so that the models can be used to confidently upscale to regional, national and continental scales. We have tested the DNDC and DayCent models at four long term experiments covering 9 treatments and 4 management practices. The simulated grain yields matched the measurement in the majority of treatments and management practices. Overall, both DNDC and DayCent simulated grain yields satisfactorily. Both DNDC and DayCent simulated top soil (20 cm depth) soil organic carbon (SOC) in crops under different tillage management well, but both models overestimated SOC in plots where crop residues were incorporated, suggesting that further improvements are required in the crop residue and decomposition modules of these models. The performance of both DNDC and DayCent are comparable with other studies in the literature. Several limitations of DNDC and DayCent models were identified. These limitations can be overcome by incorporating new process and increasing quality and quantity of input data. DNDC and DayCent are useful tools for land managers and policymakers when recommending long-term practices to enhance SOC gains in agriculture. However, it is recommended that both models should be further developed to better simulate the impact of management practices such as residue management, since both models showed poor fit for observed yields and soil carbon dynamics for these practices..