Comparing stem volume predictions of coastal Douglas-fir stands in British Columbia using a simple physiological model and LiDAR remote sensing

Abstract

Coastal Douglas-fir (Pseudotsuga menziesii spp. menziesii (Mirb.) Franco) is highly desirable for timber production and considered a critically important species in the Pacific Northwest of North America, occurring in some of the most globally productive forest. With changes in predicted climate within the region likely to result in warmer and longer growing seasons and potentially drier summer conditions, Douglas-fir is a key candidate for ongoing research. A number of new tools are available to estimate the productivity of Douglas-fir forests, including process-based models that allow predictions of growth based on underlying physiological principles rather than statistical models of past growth and advanced remote sensing tools that allow estimation and extrapolation of forest inventory variables over the landscape. In this paper, we use a suite of tools to predict standing volume in a Douglas-fir-dominated stand in coastal British Columbia. The approaches were applied over Malcolm Knapp Research Forest (MKRF) in British Columbia, Canada. A simple physiological model, Physiological Principles Predicting Growth (3-PG), was used to predict both gross and merchantable volume across the region, using spatial estimates of climatic variables at 25 m cell resolution. Using contemporary airborne light detection and ranging data (LiDAR), we also predicted standing volume using an area-based approach at 25 m over the same stands. Results indicated accurate predictions of stem volume from 3-PG (r2=0.91, P < 0.01) and LiDAR (r2=0.70, P < 0.01) despite a long time lag between plot measured volume and LiDAR acquisition. Interestingly, comparison of stem volume between the two approaches was also close (r2=0.65, P < 0.01), indicating reasonable agreement between the fundamentally different approaches. At this fine, 25 m, spatial resolution, both sets of predictions have the potential to augment regional forest productivity assessments and offer insight into the future changes in productivity of the species, especially under a continuously changing climate.