ardavan ghorbani; ardashir pournemati; mohsen panahande
Volume 24, Issue 1 , May 2017, , Pages 165-180
Abstract
The aim of this study was to estimate and map the plant group and total aboveground phytomass using Landsat 8 images in the rangelands of Sabalan Mountain. Images were selected on the 19th of July 2013 and field data were collected in April and July based on maximum matching with the phenology of the ...
Read More
The aim of this study was to estimate and map the plant group and total aboveground phytomass using Landsat 8 images in the rangelands of Sabalan Mountain. Images were selected on the 19th of July 2013 and field data were collected in April and July based on maximum matching with the phenology of the study area and in the closest date to the time of image acquisition. Twenty-four sampling sites on six vegetation types were determined. In each site, 9 sampling plots, based on previous studies, which are required for minimum sample number according to the variance of vegetation distribution, were determined in a systematic-random method, and the aboveground phytomass of vegetation groups, such as grasses, forbs, shrubs and total, were determined using the harvesting method. Initially, to calculate vegetation indices, the averages of 16 pixel values of the location of sample units from the corrected images were derived and transferred to the software environment. The correlation matrices between the derived pixel values and field collected data for the 24 selected vegetation indices were calculated and used for the estimation of grasses, forbs, shrubs and total aboveground phytomass. The results showed that indices such as RVI, TNDVI and GNDVI had the highest correlation with the aboveground phytomass of grasses, PD312, IO and PD311 with the aboveground phytomass of forbs, RDVI, DVI and RVI with the shrubs, and PD311, PD321 and PD312 with the total aboveground phytomass (P <0.01). In the second stage, three of the indices, having the highest correlation with the aboveground phytomass of each group and entire previous stage, were selected, and Landsat8 images were used to calculate the aboveground phytomass of each vegetation group and the total aboveground phytomass was calculated. The aboveground phytomass maps of each group and the total aboveground phytomass were controlled with sampling points to assess the accuracy. The results of this study showed that the best maps were obtained using the TNDVI index for grasses aboveground phytomass, PD312 for forbs, RVI for shrubs groups and PD311 for the total aboveground phytomass. Moreover, some indices, such as PD311 and RVI, could be used for all growth forms and estimation of total aboveground phytomass (P<0.01) and (P<0.01). In general, Landsat 8 data could be used to estimate and map the aboveground phytomass of vegetation groups and to determine the carrying capacity of the total aboveground phytomass in Sabalan rangelands, having advantages based on cost, time and the ability to monitor large areas with repeatability potential in comparison with the ground-based methods.