The relationship between simulated and remotely sensed forest parameters

Howland research forest covers an area of 550 acres and has been studied extensively over the past two decades. The forest is also home to old-growth spruce and hemlock providing habitat for many species. In this project we parameterized a spatially explicit version of SIBBORK individual-based gap model for the old-growth section of the Howland forest. SIBBORK simulations were conducted over a 300x300m domain with sub-plot sizes of 10x10m within which each tree was positioned to a specific location. The parameterization was calibrated using field data for the study site and results show that the simulated forest reproduces the structural complexity of Howland old growth forest, based on comparisons of key variables including, aboveground biomass, forest height and basal area. The model was initialized using a stem map data from 1989. Afterwards the model was advanced 26 years and results were compared to the 2015 stem map.

We will present the relationship between the simulated forest parameters (e.g. tree height) and their comparison to available remote sensing data (e.g. lidar and stereogrammetry-based tree heights). Data from G-LiHT, a portable airborne imaging system that combines hi-resolution imagery, lidar, imaging spectroscopy and thermal mapping, will be used; as well as imagery from commercially available high-resolution satellite imagery. Furthermore, we will provide an outlook to the observational capabilities that will be available in the near future, like the two LiDAR (NASA's GEDI and ICESAT 2) and two SAR (NASA's ISRO NiSAR and ESA's Biomass) systems to be launch within the next 4 years.