Application of Lidar Terrain Surfaces for Soil Moisture Modeling

作者：Southee Florence Margaret^*; Treitz Paul M; Scott Neal A

来源：Photogrammetric Engineering and Remote Sensing, 2012, 78(12): 1241-1251.

DOI：10.14358/PERS.78.11.1241

Soil moisture gradients and nutrient fertility are used to classify forest types in Ontario, Canada based on ecological land classification (ELC). An existing lidar dataset for the Romeo Malette Forest near Timmins, Ontario was used to derive three terrain indices (topographic wetness index Owl), percent elevation index (PEI), and canopy height model (CHM)) at varying resolutions (2 in, 5 in, 10 in and 20 in) to determine the resolution that best characterizes soil moisture patterns in a boreal forest landscape. Depression removal algorithms were examined to determine how they affect the TWI, and thus, soil moisture estimation. This paper stresses the importance of gathering data at a resolution that is sufficient for mapping fine-scale basin features to accurately model soil moisture in forested environments. The results of this research indicate that 5 in resolution data provided the best overall relationship with measured seasonal soil moisture. More generally, the results indicate that high spatial resolution variables (i.e., 2 in, 5 in) may be more suited to modeling soil moisture trends at shallow depths (0 to 15 cm), while coarser resolutions (i.e., 10 in, 20 m) may be more adept at resolving trends over greater depths (0 to 40 cm).

• 出版日期2012-12