Integrating UAV data and soil-crop modelling for Enhanced Soil Health Monitoring

The necessity of soil health monitoring is paramount in reversing soil degradation and promoting sustainable farming. Including cover crops in the crop rotation is one of the sustainable soil management practices contributing to soil health. Cover crops contribute to soil health by nutrient retention and carbon accumulation during their growth and return of organic matter to the soil upon their incorporation. During monitoring, the sampling frequency can change from annual in the case of SOC to weekly or daily for fertilization and irrigation. Remote sensing techniques offer a solution, enabling the monitoring of vegetation over time and space, thereby enhancing our understanding of the impact of cover crops on the main crop. However, this technology makes it possible to see the surface of the field which can assist with the above-ground changes of the system. Process-based modelling and data assimilation can subsequently link the above-ground component with soil functions. In-situ data collection that includes crop characteristics such as biomass and N-uptake is essential both for transforming remote sensing data into crop characteristics and for calibrating models. Using Unmanned Aerial Vehicles (UAVs) can potentially collect data at high frequency, which can be used to enhance soil process modelling. The development of this UAV-based method has the potential to be scaled up to a satellite level in the future.In our research, we have combined the study of nutrient cycling and the effect of cover crops on soil health. To achieve this, we have used the WOFOST-SWAP-ANIMO model to simulate the varying influence of cover crop monocultures and mixtures on Soil Organic Carbon (SOC) and Nitrogen cycling in a 7-year crop rotation on sandy soil. The model simulates vegetation characteristics such as biomass, leaf area index, and yield, as well as soil moisture and mineral Nitrogen concentrations. This will give us a good estimation of the vegetation input into the soil as well as the nutrient uptake from both cover crops and main crops. Soil sampling is also important to model calibration/validation to be able to simulate the N dynamics of biological activity under the surface. Our findings suggest that the model, in conjunction with UAV data and field sensors, can effectively monitor soil health indicators crucial for field management practice selection, such as the Carbon cycle and Nitrogen use efficiency.