Identifying Tilled Soils in Northumberland County using High Resolution Digital Elevation Models derived from LiDAR Data
The global agricultural community is becoming increasingly aware of the detrimental effects that tilling has on the health and conservation of agricultural top soils. Within Canada, the Ontario Ministry of Agriculture, Food and Rural Affairs (OMAFRA) aims to combat increasing soil erosion, and has expressed interest in utilizing remote sensing technologies to monitor soil erosion at a provincial level. With a better understanding of where soil is likely to be eroded, better management of agricultural soils can begin so that they are preserved for the future. Identifying farms that are actively tilling their fields is the first step in combatting soil erosion within Ontario. This project aims to identify tillage using Light Detection and Ranging (LIDAR) due its very fine spatial resolution. The Ontario Ministry of Natural Resources and Forestry (OMNRF) is in the process of releasing a very large LiDAR dataset for selected areas within the province, and this project received early access to some of that data. This project creates and implements a model using WhiteboxTools in an attempt to identify tillage based upon its rough surface texture and anisotropic geometry in Northumberland County. The tillage identification model was designed so that the tilled soils could be identified based upon their anisotropic geometry. However, the model proved to be ineffective as the minute scale of the features and role of image noise were not considered thoroughly prior to design. Furthermore, shortcomings in model validation were also exposed. Nonetheless, portions of the model functioned as intended, and areas of future research could be explored due to the findings of this project.