Expanding carbon stock and flux modeling capabilities through the improvement of LiDAR-derived forest age estimates

Understanding the terrestrial carbon cycle is integral to modeling and mitigating global climatic change as well as generating accurate carbon inventory maps. Understanding the relationships between forest function, structure and carbon sequestration is therefore of the utmost importance. Ecosystem models have been developed that estimate future carbon fluxes in forested ecosystems based on forest structural and functional properties. The Ecosystem Demography (ED) model, in particular, has capitalized on the recent availability of LiDAR data over large areas. LiDAR can be used to estimate forest height, and ED is initialized with the distribution of LiDAR heights to estimate carbon flux, assuming a relationship between forest age and height. However, forest height alone is not always an appropriate proxy for age, as forest height can be stunted by environmental constraints. This research attempts to explore how forest age is related to other forest structural attributes, including stand density, canopy gap frequency and vertical structure. LVIS data are used to model and map forest structure along a 2 km wide transect running from Maryland to southern Mississippi. A validated forest disturbance product derived from ~25 years of LANDSAT imagery is used as forest age data. Empirical relationships are developed between forest age and structural metrics. These relationships are also constrained with known allometric scaling relationships that aim to improve model efficacy. If LiDAR data can be used as a suitable proxy for forest age, global LiDAR datasets may be able to provide unprecedented capabilities for modeling global forest carbon stocks and fluxes at present, and into the future.