Despite engineering problems with the GLAS sensor, over 300 million ICESat waveforms have been collected, with nearly 1 million currently available for the Amazon basin. With no other global lidar data collection scheduled for the near future, GLAS data represents a unique source of information on regional-to-global scale forest canopy height and aboveground biomass. GLAS waveforms have an approximate footprint diameter of 70 m and have been generated at 170 m intervals for tracks about 30 km apart at the equator. Processing of GLAS data to create reliable estimates of forest height is complicated by elements of sensor design related to its primary mission—the topographic mapping of the ice sheets of Greenland and Antarctica – and on-going issues associated with instrument performance. Our work to date on the problem has resulted in an algorithm that uses topography from 90 m Shuttle Radar Topography Mission (SRTM) digital elevation models to correct for ground slope effects and create unbiased estimates of stand height. To achieve spatially continuous coverages of biomass, we will employ both multi-phase sampling to create statistical summaries of lidar-estimated attributes over the study region, and statistical data fusion with MODIS data. Comparisons with existing biomass estimates will include surfaces derived from inventory data, interpolations based on climatic gradients, and ecosystem carbon models driven by remote sensing data. An evaluation of the sensitivity of the regional carbon flux associated with land cover change to the new biomass estimates will use the Woods Hole Research Center Bookkeeping model. Besides contributing towards the scaling of biomass related field measurements, the mapping the canopy height and biomass of the Amazon has other practical and theoretical justifications, including the testing of various regional carbon models and supplying aerodynamic roughness to climate models.

Back to Project Overview index