Systematic modeling of soil functions
Soils are complex systems in which a large number of physical, chemical and biological processes interact at different spatial scales. The various soil functions result from these interactions. Models for describing the dynamics of soil functions in response to management measures or climatic changes need to represent the relevant soil processes and their interactions. In BonaRes, we are developing a systemic model approach in which biotic and abiotic processes in soils are linked with each other via mathematical process descriptions in accordance with our current state of knowledge.
Thus, a network of interactions is created that reproduces the complex behavior of the soil as realistically as possible. This includes the non-linear dynamics of soil properties, their convergence towards stable states but also their instability in response to external disturbances.
This modeling approach is a useful tool in several aspects:
It allows the prediction of the effects of agricultural practices on soil functions.
The need for current research to improve our understanding of the system can be
identified more effectively.The relevance of individual processes for the overall system can be assessed.
More information on the model
BODIUM is designed to simulate the complex relationships between soil properties, climatic conditions, soil management and their interactions in the soil. The concept and claim of BODIUM is to represent the soil as an overall system with all relevant physical, chemical and biological processes and their interrelationships. On this basis, it will be possible to simulate changes in soil functions in response to changing boundary conditions (management and climate). The BODIUM model is based on proven model concepts for plant growth as well as water and material flows in soils and supplements these with additional aspects that are essential for soil functions but have hardly been taken into account so far. One example is the temporal dynamics of the soil structure due to soil cultivation or biological activity, which are of decisive importance for the water balance, the carbon balance and root growth. Another example is the differentiation of various groups of soil organisms, their specific demands on food quality and their influence on turnover processes, structure formation and mixing in soils.
The BODIUM model runs without complex calibration of the individual model parameters for the respective site. The parameterization of the soil processes is based on our current understanding of systemic processes. This means that the model can be used for different locations and can also simulate management or climate scenarios for which there is still little or no empirical data. The discrepancy between measured and modelled variables (e.g. yields, Nmin values, water contents) are not minimized by calibration, but rather indicate missing or insufficiently modelled processes and thus help to gradually improve the model. The primary aim of BODIUM is not to predict the coming harvest as accurately as possible, which could be achieved by calibration using current state variables, but to predict the medium to long-term changes in soil functions, which can only be achieved using a process-based model.
In order to obtain reliable results, we first need to know the processes in the soil - because we can't convert into program code what we don't understand. To do so, we use new findings from the BonaRes joint projects. On this scientific basis, we can improve and expand known model approaches and process descriptions. For example, we integrated the influence of tillage on the pore structure and the subsequent settlement of the soil into the BODIUM model as a "dynamic soil structure". In addition to the influence of cultivation on the soil pores, the various processes depending on the pore structure can also be modeled. The description of complicated biological processes is also improved with BODIUM. This includes, among other things, how microbial communities control the nutrient cycle in the soil. Using new methods, for example from microbiology, the process knowledge in this area is constantly growing and we are integrating it into BODIUM. The model can thus be constantly adapted and improved according to scientific progress.
A critical question is: how do we know that the model delivers reliable results? These results are based on the mathematical description of many physical, chemical and biological processes. Some of the relevant parameters are very difficult to measure directly or can only be accessed via highly simplified laboratory experiments. Nevertheless, in order to be able to assess the quality of the model predictions, we use existing data from long-term experiments. Here, the essential boundary conditions of soil management, climate and weather are known in detail and important state variables such as nutrient content, carbon stocks or harvest quantities are continuously measured. These state variables are the result of the processes depicted in the model and their interactions. The robustness of the results and the plausibility of the model assumptions are then determined by how well the dynamics of these state variables are represented by the model (without calibration).
Application
Science
The BODIUM model simulates physical, chemical and biological soil processes according to the current state of knowledge. It takes into account local soil properties and site conditions. This makes it possible to predict changes in soil functions depending on climate and land use. The source code of the first version is freely available.
Agriculture
BODIUM4Farmers is based on the systemic soil model BODIUM. The open access model tool with an intuitive user interface is designed to support farms in balancing their long-term farm management with good yields and healthy soils.
Contributing Projects
Publications
BODIUM Approach
Contact
Sara König (UFZ)
sara.koenig@ufz.de
+49(0)341 6025 2347
Ulrich Weller (UFZ)
ulrich.weller@ufz.de
+49(0)341 6025 4410