The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences. Vol. XXXVII. Part B7. Beijing 2008
Figure 3. Appearance of different land covers on PALSAR scene (VV-> Red, VH-> Green and HV-> Blue)
4. CONCLUSION
Current study prepared (i) an interpretation key to delineate
forest and other land cover types and (ii) land/forest cover map
for Southern Chittagong using ALOS PALSAR polarimetrie
data. This study used visual interpretation and digitizing on
screen technique to prepare land cover map. Further study
should concentrate on digital classification for forest mapping
using both polarimetrie SAR intensity and texture information.
ACKNOWLEDGEMENTS
Japan Society for the Promotion of Science (JSPS) is
acknowledged for granting fellowship to the first author.
Landsat ETM+ Satellite imageries were downloaded from
http://glcf.umiacs.umd.edu/index.shtml website.
REFERENCES
Champion, H. G., Seth, S. K. and Kattak, G. M., 1965. Forest
Types of Pakistan. Pakistan Forest Institute, Peshawar, 238 p.
Dobson, M. C., Pierce, L. E. and Ulaby, F. T., 1996.
Knowledge-based land-cover classification using ERS-l/JERS-
1 SAR composites. IEEE Transactions on Geoscience and
Remote Sensing, 34(1), pp. 83-99.
Ford, J. P. and Casey, D. J., 1988. Shuttle radar mapping with
diverse incidence angles in the rainforest of Borneo.
International Journal of Remote Sensing, 9(5), pp. 927-943.
Jiyuan, L., Xuyan, T. and Jinkai, X., 1986. Application of
Shuttle Imaging Radar data for land use investigations. Remote
Sensing of Environment, 19, pp. 291-301.
Khan, S. A., 1979. Revised working plan for the forests of
Chittagong Division (for the years 1978-79 to 1987-88).
Volume 1. Forest Department, Government of Bangladesh,
Dhaka.
Landcover/
Landuse
Area (ha)
Percentage
Forest
5 955
18.54
Plantation
2 349
7.32
Forest mosaic
5 678
17.68
Shrubs
2 209
6.88
Crops/Homesteads
4 082
12.71
Bare soil
6 479
20.18
Shrimp farms/Salt-
beds
5 107
15.90
Water
253
0.79
Total
32 111
-
Table 3. Extent of different land covers in the study area
Polarimetrie PALSAR data could distinguish forest from other
land covers based on colour and texture. Plantation could be
interpreted using texture; this class has smoother texture than
natural forest. The separation of shrubs and forest mosaics from
other land covers is possible using colour and contextual
information (i.e. terrain feature). Single polarization (HH) L
band JERS-1 SAR image could separate forest and degraded
forest (sometimes may be represented as forest mosaic) in the
South-eastern Bangladesh (Rahman and Sumantyo, 2007);
PALSAR polarimetrie data distinguished additional classes-
shrubs and homestead forests (intermixed with agricultural
crops). On the other hand, Landsat ETM+ optical data
interpreted four different categories of natural vegetation
(primary forests, secondary forests, bamboo and shrubs) and
four types of plantations (acacia, rubber, indigenous species and
teak/scattered trees) in that region (Rahman, 2008).
188
