Using modelling and tools for advice on improved agricultural management to achieve climate neutrality in croplands

Abstract

Cropland soils show large potential to sequester carbon to achieve climate neutrality. Changes in management can affect an increase of carbon sequestration or reducing carbon losses in form of emissions or leaching. However, the impact of management changes on the sequestration and other processes needs to be quantified to provide advice to farmers. Experiments to analyse impacts of management changes are costly and labour intensive. Additionally, these experiments take time and cover only a small range of environmental conditions. Therefore, modelling is widely used to over-come these limitations. Model results allow the estimation of all relevant fluxes for the overall greenhouse gas emission balance or, depending on the model, for some parts. This is a fast and efficient method to quantify soil organic carbon (SOC) changes due to modifications in agricultural management. Even though, models proved their quality of simulating SOC changes, there are some restrictions in the use of models for actual advice based on model results. In the here presented study, three key points will be analysed: First, the additional impacts beside the SOC changes. Carbon sequestration can be offset by emission of other greenhouse gases or management changes affect yield, which needs to be included in the analysis. While these two variables are well covered by usual model approaches, other aspects like food quality are more difficult to include. Second, how does the complexity of the model affect the result. The simple assumption that more complex models are potentially more accurate, but also require more input data is in most cases realistic (this is a generic assumption which is not always true). More input data and more complexity are also associated with potentially increased uncertainty. Third, who is running the model. While research-based advice using more complex models might be potentially more accurate, models used by farmers might be more specific and direct in providing key information. Additionally, the impact of the increased data demand and required data can affect an increased error. These points are analysed on examples and case studies. This includes an analysis beyond the carbon sequestration and how to include these aspects in the analysis. Further, results of a tool developed for stakeholders/farmers is compared with results of a biogeochemical model for selected sites. Finally, an analysis of the limitations of the models due to data demand and data availability. The analysis of wheat yields shows mainly positive impacts on the SOC change, but mainly reduced yield. The comparison of the two models indicates the impracticability of the more complex option, as the data demand is not orientated on the data availability. The decision based on model results requires a careful use of models and a good understanding of the results.