Utility of the Normalised Difference Vegetation Index (NDVI) for land/canopy cover mapping in Khalkhal County (Iran)

Normalized Difference Vegetation Index (NDVI) is commonly used for assessing and monitoring ecological variables such as vegetation cover, above-ground biomass and Leaf Area Index. This paper examines the utility of NDVI for mapping the land/canopy cover characteristics in Khalkhal County in north-west of Iran. Images were selected by considering seasonality and phenological patterns, and three images from three sensors including: Landsat TM (30/09/1987), ETM+ (29/07/2002), and IRS P6LISSIII (26/05/2008) were selected. Preporccessing stages including geometric and radiometric corrections and topographic normalization were conducted, and NDVI of those images were derived. Land / canopy cover of 270 sites using point intercept method were estimated and sample sites using GPS were recorded. Collected points transferred on image and using zonal-based attributes average of 16 pixels around the GPS points transferred to excel file and correlated with field collected data. Derived maps were evaluated for accuracy assessment and correlation analyses. Results showed that: although there is some considerable evidence that the results of the derived NDVI are acceptable, however by comparison of the derived maps with the field data, there are considerable differences. Accuracy assessment showed that the results are not acceptable. Finally, the correlation analyses between individual land/canopy covers showed that there are no significant (p>0.05) relationships between data extracted from images versus field collected data. Therefore, the use of NDVI in the routine procedure in the same regions of Khalkhal County is not suitable for canopy/land cover mapping. Results of this study suggest that further analysis are required to determine the usefulness of the NDVI calculated from course resolution satellite data for estimation of land/canopy cover when there is high heterogeneity exists on the study area.