Silvilaser 2019 - Poster Presentations »
DEFINING OLD-GROWTH FORESTS WITH AIRBORNE LiDAR
Forests in their later stages of development attain attributes that support biodiversity and provide a variety of benefits to human populations. Despite their irreplaceable values, old-growth forests are declining worldwide due to anthropogenic pressures. Conservation approaches to maintain old-growth forests do exist, such as Old-growth management areas (OGMAs) in British Columbia, CA. Currently, OGMAs are mainly defined by estimates of forest age. However, it is uncertain if this selection strategy suitably identifies forests with characteristics/attributes expected in old-growth forests. In this work, we developed a range of old-growth indexes that capture multiple forest attributes associated with old-growth forests. We then developed LiDAR-derived metrics that we utilized with a random forest (RF) modelling framework to model old-growth forest attributes across the landscape. We designed three RF models in classification mode using field estimated classes as response variables: (1) age classes (54% accuracy), unsupervised classification of old-growth attributes (2) with age (69% accuracy) and (3) without age (68% accuracy). In addition, we designed three other RF models in regression mode with field estimated (4) stand age (R2: 35%), an old-growth index (5) with age (R2: 71%) and (6) without (R2: 71%). We found that for both classification and regression models, age was not a crucial attribute, as age alone were not well predicted in neither classification nor regression models. As well, we found that excluding age did not significantly reduce the overall model’s performances. Based on a combination of model performance and simplicity of the models themselves, we selected a LiDAR-derived model (model 6) to scale up the old-growth index from plot to landscape level. Using this model we found that 14.7% of the study area is covered by forests with high old-growth values. Inside the currently designated OGMAs’ areas, the “old-growth” cover is 24.9%, indicating that OGMAs are retaining forests with some high old-growth value. However, only 2.5% of the OGMAs have more than 50% of its area covered by forests with high old-growth value. This research brings light to old-growth and OGMAs’ definition and their assessment through the use of fine scale remotely sensed data, LiDAR. More importantly, the identification of the amount and location of old-growth forests over the landscape can aid to the conservation of this rare resource and its services.