Measuring post-disturbance forest regrowth in North and South Carolina using 28-years of Landsat data.

Khaldoun Rishmawi, University of Maryland/College Park, rishmawi@umd.edu (Presenter)
Jeffrey Masek, NASA GSFC, jeffrey.g.masek@nasa.gov
Feng Aron Zhao, University of Maryland, zhao26@umd.edu
Christopher Neigh, NASA GSFC, christopher.s.neigh@nasa.gov
Chengquan Huang, University of Maryland, cqhuang@umd.edu

The carbon balance of a forest system is fundamentally linked to its cycle of disturbance and recovery. Forest disturbance events were obtained from the North America Forest Dynamics (NAFD) study products. Post-disturbance changes in forest biomass, canopy cover and LAI were modeled from annual Landsat reflectance trajectories (1984-2011) by inversion of the GEOSAIL radiative transfer model using the look-up table approach. Training and validation data were provided by the US Forest Service Forest Inventory and Analysis (FIA) program. The number of FIA plots in the study region is approximately 9,600 plots of which 2,300 plots had complete records of tree species, diameter at breast height (DBH), and tree height measurements. A subset of FIA and co-located Landsat data were used to parameterize GEOSAIL inputs (e.g. optical and angular properties of components) and to constrain the often highly uncertain parameters of the allometric equations that relate FIA field measurements (forest type, height, and basal area) to forest structural elements that influence top of canopy reflectances (e.g. canopy cover, LAI, tree height to width ratio). The parameterized GEOSAIL was then used to populate Look-Up Tables (LUTs) that relate aboveground biomass, canopy LAI, and fraction cover to Top of Canopy (TOC) reflectance values. The LUTs were then used to estimate biomass, LAI and canopy cover from the Landsat reflectance trajectories. An estimate of the error budgets were obtained by comparison with the remaining FIA field measurements. The combination of annual Landsat reflectance trajectories with ground-based FIA measurements provided robust estimates of biomass accumulation and canopy closure of recently disturbed forests in the Carolinas. Results also illustrate regional variability in post-disturbance forest recovery rates. The results can be used to study the relation of post-disturbance forest dynamics to variations in soil, climate, topography, disturbance regimes, and post-disturbance land management.

Presentation Type:  Poster

Session:  General Contributions   (Tue 4:35 PM)

Associated Project(s): 

• Goward, Samuel: US Forest Disturbance History from Landsat ...details

Poster Location ID: 218