Forest Dynamics and Carbon Cycle Studies from Multi-sensor Data Fusion

Guoqing Sun, Jon Ranson (GSFC), V. I. Kharuk (SFI), Katalin Kovacs (SSAI)

The man-induced and natural forest disturbances (wildfires, cuttings, pest invasions, air pollution) have caused changes in the biomass level, forest composition structure, and succession patterns. These changes affect the forest spatial structure, ecosystem function, and the region and the global carbon balance. Mapping and understanding the impact of the disturbances on forests are very important for global climate change studies.

A group of scientists from UMD, GSFC, and Sukachev Forest Institute of Russia have been working on forest mapping and structural parameters retrieval from multi-sensor (optical, radar, lidar) data and data fusion methods. 3D optical reflectance, radar backscatter and lidar waveform models are being used to explore the information contents of the data and the potentials of combined use of optical, lidar and radar data in forest biomass and forest structure parameter estimation.

A unique three dimensional, high-resolution radar backscatter model and a lidar waveform model have been developed at UMD and GSFC. These models were parameterized using the same forest physical model, so the simulated signatures are derived from the same forest physical parameters and spatial structure.

The group is now developing methods to map boreal forest type and structure in Siberia using Terra instrument data (MODIS, MISR) aided by point measurements of canopy structure inferred from satellite lidar data (GLAS). The radar data from ENVISAT (ASAR) and future ALOS (PALSAR) and other high-resolution optical data (Landsat, ASTER) are being used in the project. The key results of the project will be improved cover type maps of Siberia boreal forests that utilize canopy structure information and derived biomass maps. This work should also provide an analysis framework suitable for use in other parts of the boreal forest.

Example publications: Sun, G., K. J. Ranson, and V. I. Kharuk, 2002, Radiometric slope correction for forest biomass estimation from SAR data in Western Sayani mountains, Siberia, Remote Sensing of Environment, Vol. 79:279-287.

Sun, G. and K. J. Ranson. 2000, Modeling lidar returns from forest canopies IEEE Transactions on Geoscience and Remote Sensing, Vol. 38, pp.2617-2626.