Regional Applicability of Forest Height and Aboveground Biomass Models for the Geoscience Laser Altimeter System

Abstract

Abstract Accurate estimates of forest aboveground biomass are needed to reduce uncertainties in global and regional terrestrial carbon fluxes. In this study we investigated the utility of the Geoscience Laser Altimeter System (GLAS) onboard the Ice, Cloud and land Elevation Satellite for large-scale biomass inventories. GLAS is the first spaceborne lidar sensor that will provide global estimates of forest height. We compared accuracy and regional variability of GLAS height estimates with data from the US Forest Service Inventory and Analysis (FIA) program and found that current GLAS algorithms provided generally accurate estimates of height. GLAS heights were on average 2–3 m lower than FIA estimates. To translate GLAS-estimated heights into forest biomass will require general allometric equations. Analysis of the regional variability of forest height-biomass relationships using FIA field data indicates that general nonspecies specific equations are applicable without a significant loss of prediction accuracy. We developed biomass models from FIA data and applied them to the GLAS-estimated heights. Regional estimates of forest biomass from GLAS differed between 39.7 and 58.2 Mg ha−1 compared with FIA.