|
چکیده
|
Northern Zagros forests are coppices and almost completely destroyed. Basal area allows us to determine the change in these forests. The estimation of this index using field work is difficult, take plenty of time and needs a good deal of money. The objective of this study was to evaluate the possibility of estimation of basal area using SPOT-5 data.320 circular plots (0,1 ha) were established using method of systematic random in investigated forest stands. Inventory grid was designed in GIS environment. On each plots all trees of all species with stem thickness more than 7.5 cm in diameter were callipered at breast height (i.e. 1.3 m above ground), and tree species were determined. Using field data basal area in plots was determined. Investigation on Satellite images shows that, these images have no radiometric distortion. These images were georeferecing by SPOT image Co. The geocoded images were checked for reliability in comparison with the digital topographic map. Diverse suitable spectral transformation such as rationing, PCA, and vegetation indexes (NDVI, DVI, RVI, SAVI and IPVI) transformation was performed on the images. The results from regression analysis imply that multispectral SPOT-5 data could be used for basal area estimation. Models for estimating basal area from SPOT-5 data were developed using multiple linear regression analysis. The set of predictor variables chosen consisted of the multispectral bands B1-B2, NDVI, DVI,RVI,SAVI AND IPVI.(R=0.73 R2 =0.53,P-value=0.000)
|