Soil organic matter balances in organic versus conventional farming-modelling in field experiments and regional upscaling for cropland in Germany

The question of whether organic farming leads to higher soil organic matter (SOM) levels in arable soils compared with conventional farming is an ongoing debate. Building on several studies reported in the literature, we hypothesize that the impact on SOM levels is not an intrinsic characteristic of any farming system but is the result of the actual structure of the farming system, in particular, the composition and management of crop rotations, and the availability and utilization of organic manure. The SOM balances for organic versus conventional farming in Germany are compared by considering data on the structure of organic and conventional farming systems from agricultural census reports and then applying the SOM balance model HU-MOD. Preliminary testing confirmed the applicability of the model using a survey on soil organic carbon (SOC) change and SOM balances in four long-term field experiments in Germany and Switzerland and found that more positive SOM balances coincided with higher SOC levels. We therefore conclude that, where the SOM supply level of organic farming systems is higher than in conventional management, a shift from conventional to organic agriculture would increase SOM levels. Upscaling using agricultural census data in Germany, we found that SOM balances of organic farming were more positive than for conventional farming in the scenarios without consideration of animal manure application, but SOM balances for the two systems were not different where animal manure application rates were assumed to be at the current average rate for all cropped land. However, in fact, animal manure availability and application shows strong regional variations, and it is likely that this would affect the mean cropland SOM balance if it were possible to calculate it based on such spatially disaggregated data. We confirm the applicability of simple SOM balance models to compare the impact of farming systems and cropland structures on SOC levels. More work is needed to develop data inputs at a sufficient spatial and structural resolution to support more detailed evaluation.