Analysis of satellite plant spectral indices in the detection of health stress in pistachio orchards

Document Type : Original Article

Authors

Science and Research Branch, Daneshgah Blvd, Simon Bulivar Blvd, Tehran

Abstract

This research was carried out in order to identify healthy pistachio plant from stressed varieties by using Sentinel 2A satellite image of MSI sensor and Landsat 8 sensor of OLI satellite in Varamin city. Satellite images help to identify a healthy plant by using the reflection of electromagnetic waves in the apparent structure of the plant. The amount of this reflection varies depending on the leaf surface, type of disease, vegetation density, distribution and vegetative state of the plant during a year. In the current research, NDVI, NDRE, NDMI, and SAVI indices were used in the two months of May and August during the years 2015-2016 as a factor in describing the vegetation status. In the first stage, the images related to the study area were prepared and the desired layers were placed on top of each other in the GIS environment. In the next step, using the artificial neural network algorithm, the studied area was classified into three parts: pistachio, non-pistachio and soil, and continued research on the pistachio area. Then the correlation of NDVI with other three indices was calculated in four specified times. The results showed that in May, there was a correlation of 0.9819 between SAVI soil index and NDVI freshness index compared to August, and this value was higher than the other two indices in both years. Considering that an increase in the SAVI index indicates a decrease in the freshness of vegetation, this index can be used to evaluate the health of pistachio orchards.

Keywords

Main Subjects