Scaling of Forest Biomass and Structural Attributes through Integration of ALOS PALSAR data: Case Studies from Australia

Richard Lucas(1), John Armston(2), Joao Carreiras(3) and Peter Bunting(1)

(1) Aberystwyth University, Llandinam Tower, Aberystwyth, United Kingdom
(2) Queensland Department of Natural Resources, 80 Meiers Road, 4068 Brisbane, Australia
(3) Tropical Research Institute, Travessa do Conde da Ribeira, 1300-142 Lisbon, Portugal


The development and/or validation of algorithms for retrieving forest biomass and structural attributes across large areas using Advanced Land Observing Satellite (ALOS) Phased Arrayed L-band Synthetic Aperture Radar (SAR) data is dependent upon detailed quantitative information on these attributes at the field level. In this paper, and focusing on sites in Queensland, Australia, we illustrate how field and airborne remote sensing have been used to quantify and scale-up such attributes to the wider landscape thereby informing the classification of ALOS PALSAR data and increasing understanding of the information content of such data.

Initial studies focused primarily on complex, native, mixed species forests near Injune in central east Queensland. In 2000, airborne hyperspectral, LiDAR and polarimetric AIRSAR data were acquired along with field-based estimates of species composition, structural attributes and biomass. Algorithms were developed subsequently for delineating tree crowns and differentiating these to a species type using hyperspectral data and retrieving tree/stand height and density from LiDAR. The integration of these data products facilitated estimation of total above ground and component biomass at scales ranging from the individual tree to the landscape. Such information was used subsequently to establish empirical relationships with AIRSAR and ALOS PALSAR backscatter and biomass and to parameterise SAR backscatter models that led to a better understanding of the interaction of microwaves with different components of the forest volume.

These and associated algorithms have subsequently been applied to airborne data acquired at other sites across Queensland and representing a greater range of forest types. Preliminary assessments of ALOS PALSAR data indicate a close correspondence between structural attributes (namely density, height and crown cover) and backscattered. Such observations provide some explanation of the variability in backscatter across the forested regions of Queensland and form the basis for subsequently characterising and mapping forest structural types and changes in these associated with deforestation, degradation, disturbance, woody thickening and regeneration.


Symposium presentation


  Higher level                 Last modified: 07.05.06