A maximum a posteriori Bayesian classifier is used to perform a supervised classification of multifrequency, polarimetric, airborne, SAR observations of boreal forests from the Bonanza Creek Experimental Forest, near Fairbanks, Alaska, into six categories: 1) white spruce; 2) black spruce; 3) balsam poplar; 4) alder; 5) treeless areas; and 6) open water. Tree classification accuracy is highest (86%) using L- and C-band fully polarimetric combined on a date where the forest just recovered from river flooding. The SAR map compares favorably with a vegetation map obtained from digitized aerial infra-red photos. C-band frequency and HV-polarization are, respectively, the most useful frequency and polarization for mapping tree types using SAR. Combination of multi-date SAR observations does not improve classification accuracy, and SAR data acquired on different dates, under different environmental conditions, yield classification accuracies 16% to 41% lower. Single-frequency, single-polarization, SAR data show limited mapping capability. Multispectral SPOT observations of the same area on a single date yield a classification accuracy of 78%. Combining optical and SAR data is useful for identifying tree species, independent of ground truth verification, using biomass estimates from SAR, at L-band HV-polarization, NDVI from SPOT red and infra-red radiances, and an unsupervised segmentation map of the SAR data.<<ETX>>
[1]
Eric Rignot,et al.
Monitoring environmental state of Alaskan forests with AIRSAR
,
1992
.
[2]
John A. Richards,et al.
The effect of changing environmental conditions on microwave signatures of forest ecosystems - Preliminary results of the March 1988 Alaskan aircraft SAR experiment
,
1990
.
[3]
Rama Chellappa,et al.
Segmentation of polarimetric synthetic aperture radar data
,
1992,
IEEE Trans. Image Process..
[4]
Eric S. Kasischke,et al.
Connecting forest ecosystem and microwave backscatter models
,
1990
.
[5]
Stephen Wall,et al.
Multiple Incidence Angle SIR-B Experiment Over Argentina: Mapping of Forest Units
,
1986,
IEEE Transactions on Geoscience and Remote Sensing.
[6]
Thuy Le Toan,et al.
Relating forest biomass to SAR data
,
1992,
IEEE Trans. Geosci. Remote. Sens..
[7]
Gordon B. Bonan,et al.
Importance of leaf area index and forest type when estimating photosynthesis in boreal forests
,
1993
.
[8]
S. Sader.
Forest biomass, canopy structure, and species composition relationships with multipolarization L-band synthetic aperture radar data
,
1987
.