High resolution three dimensional spatial patterns of soil organic carbon storage in eroding agricultural landscapes

Soil Organic Carbon (SOC) is the largest terrestrial pool of carbon (1500Gt), and is an important factor controlling overall soil quality. The quantification of SOC and its change over time is crucial in a context of climate change and concerns about soil fertility. Current methods to assess SOC changes, however, pose some important limitations to our ability to accurately assess SOC dynamics in a spatial and temporal context. The overall aim of this thesis was therefore to improve our capability to assess 3D SOC storage patterns at spatial high resolution and to improve our understanding of the impact of soil movement on SOC storage in an agro-ecosystem. In this thesis, we developed and evaluated a method to predict the vertical SOC profile at a high resolution in a landscape affected by erosion. We show that the surface SOC content is related to soil redistribution processes whereas the shape of the vertical distribution is influenced by the soil moisture and temperature. In a second phase, we developed and evaluated a procedure to measure the surface SOC content at very high resolution using low altitude aerial sensing systems (UAV’s). The method developed permits to map the surface SOC content of arable fields at a resolution of c. 10cm with an accuracy that is similar to those obtained under laboratory conditions. Furthermore, we assessed the potential of combining the UAV-based sensors with the soil profile depth assessments for quantifying the three dimensional SOC distribution. We showed that the remote sensing technique proposed here has a strong potential for 3D mapping due to the strong correlation between the topsoil and subsoil SOC content. However, further research using more performing sensors are needed to control for soil variability (i.e. clay content) that may interfere with the SOC content. Using these new methodologies, we then quantified and evaluated the impact of soil redistribution on three dimensional SOC patterns and its impact on the net C balance for the period 1997-2014. We showed that the texture was a key factor controlling the SOC pattern, but due to the high variability of the SOC and its slow evolution over the time, detecting significant changes was challenging. Nevertheless, our methodology allowed detecting a significant decrease in SOC stock for the topsoil in the eroding areas and an increase in SOC stock for the subsoil layers of stable areas. This analysis was consistent with model-based estimates of changes in SOC stocks in response to erosion and deposition. Finally, this work showed that it is important to further improve methodologies and tools to assess 3D SOC storage in order to obtain a solid SOC baseline when detecting changes in SOC storage. The improvements presented in this thesis provided a significant step in this direction. We conclude that this thesis has improved our understanding of the consequences of erosion on SOC dynamics, and this may provide a basis to manage these impacts to enhance soil fertility but also mitigate climate change.