GC11B-1041
High-Resolution Modeling Disturbance-Induced Forest Carbon Dynamics with Lidar and Landsat Observations
Monday, 14 December 2015
Poster Hall (Moscone South)
Maosheng Zhao1, Chengquan Huang2, George C Hurtt2, Ralph Dubayah2, Justin Fisk2, Ritvik Sahajpal2, Steve Flanagan2, Anu Swatantran2, Wenli Huang2, Hao Tang2, Jarlath ONeil-Dunne3 and Kristofer D Johnson4, (1)University of Maryland College Park, College Park, MD, United States, (2)University of Maryland, Department of Geographical Sciences, College Park, MD, United States, (3)University of Vermont, Rubenstein School of Environment & Natural Resources, Burlington, VT, United States, (4)USDA Forest Service Northern Research Statiuon, Newtown Square, PA, United States
Abstract:
Forest stands are dynamic in a status from severely, partially disturbed, or undisturbed to different stages of recovery towards maturity and equilibrium. Forest ecosystem models generally use potential biomass (an assumption of equilibrium status) as initial biomass, which is unrealistic and could result in unreliable estimates of disturbance-induced carbon changes. To accurately estimate spatiotemporal changes of forest carbon stock and fluxes, it requires accurate information on initial biomass, the extent and severity of disturbance, and following land use. We demonstrate a prototype system to achieve this goal by integrating 1-m small footprint Lidar acquired in year 2004, 30-m Landsat disturbances from 1984 to 2011, and an individual-based structure height Ecosystem Demography (ED) model. Lidar provides critical information on forest canopy height, improving the accuracy of initial forest biomass estimates; impervious surfaces data and yearly disturbance data from Landsat provide information on wall-to-wall yearly natural and anthropogenic disturbances and their severity (on average 0.32% for the natural and 0.19% for the anthropogenic for below test area); ED model plays a central role by linking both Lidar canopy height and Landsat disturbances with ecosystem processes. We tested the system at 90-m spatial resolution in Charles County, Maryland, by running ED model for six experiments, the combinations of three initial biomass (potential, moderate and low initial biomass constrained by Lidar canopy height) with two disturbance scenarios (with and without anthropogenic disturbances). Our experiments show that estimated changes of carbon stock and flux are sensitive to initial biomass status and human-induced land cover change. Our prototype system can assess regional carbon dynamics at local scale under changing climate and disturbance regimes, and provide useful information for forest management and land use policies.
