UCSB Biogeography Lab

UCSB Biogeography Lab Publications Abstracts

STAND DISCRIMINATION IN A WESTERN CONIFEROUS FOREST USING AIRSAR DATA
José L. Saleta
Masters Thesis, Department of Geography, University of California, Santa Barbara. 148 pp. 1995.

The capability of airborne synthetic aperture radar (AIRSAR) in the discrimination and classification of coniferous forest stands following the criteria that the US Forest Service uses to elaborate timber stand maps (i.e. percentage of canopy cover and average crown size) was evaluated. Discriminant analysis and a decision tree classifier were used to analyze data from 258 stands on a ponderosa pine forest near Mt. Shasta, California. The data analyzed consisted of backscatter coefficients at HH, VV and HV polarizations and like-polarization phase difference at C-, L- and P-bands. A set of ratios derived from these was also tested. Data from two dates with wet and very dry ground conditions were examined. Only limited discrimination of stands could be achieved using SAR images. C-HV, L-HV, P-HH and P-VV were the most useful channels, with classification accuracies that range from 42 to 49%. Among the derived ratios, the woody vs. herbaceous indices and the P-band biomass index were the most significant variables, with classification accuracies between 39 and 47%. The analyses were repeated aggregating the 8 stand types into the 3 timber volume classes considered by the USFS. The classification accuracies raised then to 72-77%, but class separation using discriminant analysis did not improve significantly.

Go to UCSB Biogeography Lab home page



What's New Research & Projects Remote Sensing & GIS Landscape Ecology Conservation Planning Publications People Data Links	\| Home \| Contact \| UCSB \| Bren \| ICESS \|

	UCSB Biogeography Lab Publications Abstracts STAND DISCRIMINATION IN A WESTERN CONIFEROUS FOREST USING AIRSAR DATA José L. Saleta Masters Thesis, Department of Geography, University of California, Santa Barbara. 148 pp. 1995. The capability of airborne synthetic aperture radar (AIRSAR) in the discrimination and classification of coniferous forest stands following the criteria that the US Forest Service uses to elaborate timber stand maps (i.e. percentage of canopy cover and average crown size) was evaluated. Discriminant analysis and a decision tree classifier were used to analyze data from 258 stands on a ponderosa pine forest near Mt. Shasta, California. The data analyzed consisted of backscatter coefficients at HH, VV and HV polarizations and like-polarization phase difference at C-, L- and P-bands. A set of ratios derived from these was also tested. Data from two dates with wet and very dry ground conditions were examined. Only limited discrimination of stands could be achieved using SAR images. C-HV, L-HV, P-HH and P-VV were the most useful channels, with classification accuracies that range from 42 to 49%. Among the derived ratios, the woody vs. herbaceous indices and the P-band biomass index were the most significant variables, with classification accuracies between 39 and 47%. The analyses were repeated aggregating the 8 stand types into the 3 timber volume classes considered by the USFS. The classification accuracies raised then to 72-77%, but class separation using discriminant analysis did not improve significantly. Go to UCSB Biogeography Lab home page
	Email stoms@bren.ucsb.edu