|
AirSAR/MASTER
|
The Use of Airsar Data for Assessing the Potential of Future Spaceborne Sar for Regional Estimation of Woodland Biomass in Australia
7. Discussion
The results indicated the greater interaction of L- and P-band microwaves with the woody components of the biomass compared to C-band microwaves. The relationship between C-band and woody biomass is more likely to have resulted from the inherent relationships of leaf with branch, trunk and TAGB. Microwaves at both L and P band were related almost equally with both the trunk and the branch biomass. The correspondence in these relationships was attributed partly to the similarity in the size range and orientation of branches and trunks. However, the inherent relationship between branch and trunk biomass leads to uncertainties as to the interaction of these wavelengths with different vegetation components.
Figure. 1: Relationship between TAGB and a) SIR-C SAR C- and L band backscatter and
b) TOPSAR C-, L- and P-band backscatter
Unlike previous studies (e.g., [5]), where different P-band polarisations were related to differential scattering by trunks and branches, differences in the P-band backscatter at three polarisations were not noticeable in this study. This similarity was again attributed to the overlap in the size distribution/orientation of these components and a lack of layering within the uneven aged woodlands.
The TAGB saturation levels for both TOPSAR C-band HH and SIR-C SAR C-band VV and TOPSAR L-band VV and SIR-C SAR L-band VV were similar at 20-30 Mg
ha -1 and 50-60 Mg ha -1 respectively. Studies in Injune suggest a saturation of c 60-80 Mg ha -1 in the relationship with JERS-1 SAR L-band HH data. In spatially quantifying the TAGB of woodland vegetation, P-band HH data seems optimal due to the observed strength in the relationship with TAGB, the higher limit of biomass saturation (100 Mg ha -1), and the greater dynamic range of the data.
8. Conclusions And Future Work
Polarimetric C, L and P band data can be related to TAGB and different biomass components with varying degrees of significance. The better relationships were established using TOPSAR data. The main similarities in using TOPSAR and SIR-C SAR were the strength of the relationships between C- and L-band data and component mass, and the TAGB saturation levels for both C-band and L-band data. The study suggests that single-band SAR may be used to quantify the TAGB of woodlands in Australia, although better results are likely to be obtained using polarimetric SAR. This capacity will be fully tested using PACRIM AIRSAR data collected in August, 2000.
9. Acknowledgments
The authors thank the Bureau of Resource Sciences and the Australian Greenhouse Office, the Australian Research Council and the Queensland Department of Natural Resources for financial and logistical support.
10. References
- Beers T., and Miller, C., "Point Sampling: Research Results, Theory and Applications", Purdue University, Agricultural Experiment Station Lafayette, Research Bulletin, vol. 786, August 1965.
- Burrows, W.H., Compton, J.F., Hoffman, M.B., Back, P.V., and Tait, L.J., "Allometric relationships and community biomass estimates for some dominant eucalypts and associated species in Central Queensland woodlands", unpublished.
- Harrington, G., "Estimation of above ground biomass of trees and shrubs in Eucalyptus populnea F. Muell. woodland by regression of mass on trunk diameter and plant height", Australian Journal of Botany, vol. 2, pp. 135-143, 1979.
- Beauchamp, J.J., and Olsen, J.S., "Corrections for bias in regression estimates after logarithmic transformation", Ecology, vol. 54, pp. 1403-1407, 1973.
- Beaudoin, A., Le Toan, T., Goze, S., Nezry, E., Lopes, A., Mougin, E., Hsu, C.C., Han, H.C., Kong, J.A., and Shin, R.T., "Retrieval of forest biomass from SAR data", International Journal of Remote Sensing, vol. 15, pp. 2777-2796, 1994.
|
|
|
|
|
