Correspondence between pre-fire and biomass losses in boreal peatlands and upland forests using lidar and optical remotely sensed data

Abstract:

In Western Canada, climate change associated with warmer and drier air temperatures are reducing water availability in boreal peatlands and forests, increasing their susceptibility to wildland fire by exposing deep carbon-rich peat layers. Boreal peatland and upland ecosystems are difficult to monitor due to their broad and remote area coverage. Yet, there are significant uncertainties with regard to the impacts of climate change and disturbance on these ecosystems. To better understand and monitor changes, long-term optical remotely sensed data can be used as an indicator of consumption for mapping burn severity in peatlands and upland ecosystems. There is some uncertainty on how optical burn severity indices such as the difference Normalized Burn Ratio (dNBR) correlate with the loss of biomass during combustion and the remaining vegetation structures after a wildfire. The uncertainties includes the application of lower spatial resolution pixels to small landscape features such as wetlands, atmospheric constituents and canopy occlusion. Improved understanding of interactions between the burned surface and reflected/absorbed electromagnetic radiation would reduce uncertainties associated with burn severity and biomass losses, and how these might affect post-fire regeneration and long-term ecosystem services in boreal peatlands and forests.

In this study, we assessed the correspondence between pre-fire vegetation structures and post-fire biomass losses in boreal peatlands and forested uplands using airborne multi-spectral lidar remote sensing and optical burn severity indices. The objective was to determine how lidar structural vegetation metrics representing vegetation losses compare with the operational Canadian Forest Service (CFS) Landsat dNBR product. The results of this research will provide an understanding of the correspondence between vegetation structures and post-fire biomass losses due to combustion, and how the active spectral characteristics compare with optical burn severity metrics in boreal peatlands and uplands.

Keywords— Biomass, Wildfire, Lidar, Peatlands