Regresar

Pbkp_redicting non-inventoried forest elements using forest inventory data: The case of winter forage for woodland caribou

Abstract:

Growing development pressures and expectations that forest managers provide future wildlife habitat require better understanding of species' habitat needs, particularly food, cover, and space requirements, and an ability to spatially depict these needs. In forest management in Canada, the primary data used to identify and quantify wildlife habitat reside in remotely sensed forest resource inventories (FRI) that were originally developed to assess timber values for merchantable tree species. Although FRI- and field-based sampling do not always show strong agreement, research has shown that FRI can be informative for wildlife habitat assessments. However, much uncertainty remains when investigating forest characteristics that are not visible to the interpreters, such as sub-canopy features. Here, we used 152 plots in northwestern Ontario to compare the ability of field-based and remotely sensed forest inventories to pbkp_redict Cladonia lichen cover, a primary winter food source for woodland caribou. The best model for field-based data, which included percentage of jack pine and black spruce in the tree canopy, tree height, stand age, soil moisture, and stem density, correctly pbkp_redicted 92% of cases where Cladonia spp. were absent (n = 107 plots) and 62% of cases where they were present (i.e., cover >1%; n = 45 plots). FRI performed poorly by contrast, with corresponding percentages of 96 and 19%. FRI provide weak data support for differentiating winter forage availability for woodland caribou, an important habitat factor at the stand level. These findings have important implications for pbkp_redictions of herd productivity, and suggest that improved remote-sensing capabilities are required in order to assess woodland caribou winter habitat.