Background and Objective

Modelling of soil organic carbon dynamics on reginal to national scale is challenging due to the high spatial variability of drivers such as soil properties, climate and land management. Many different models have been developed during the last decades that estimate soil organic carbon dynamics. In a systematic assessment models will be selected and validated that best fit to pedo-climatic conditions of German agricultural soils. In order to run the models the C input with biomass from crop residues and roots have to be estimated as input variable for the model. There are different allocation functions available which are also tested and selected. The aim of the project is to develop a model ensemble that is able to estimate current and future soil carbon dynamics at national scale und to predict soil carbon under different management scenarios.

Approach

In order to select and validate soil carbon models, data from the German long-term monitoring network are used as well as data from long-term field experiments. In total six commonly used soil carbon models and five different allocation functions to estimate the C input to the soil are used in a comparative study. This will result in a model ensemble which will be used to estimate for all 3100 sampling points of the Soil inventory Agriculture if soils are source or sink for CO₂.