Fatehi, Ahmad 
Effects of polluted rainfall related to burning Kuwaiti oil wells on vegetation 
in south of Iran 
Ahmad Fatehi , Nader. Jalali. 
Soil conservation and Watershed Management Research center, 
P.O. Box 13445-1136, Tehran, Iran 
KEY WORDS: Vegetation cover change, NOAA images, Pollution, NDVI 
Abstract 
The products of NOAA-AVHRR are an appropriate data to detect any changes in vegetation cover and to determine 
its distribution pattern. The most valuable effect of the use of NOAA-AVHRR is that this satellite source is the only 
instrument that is capable of demonstrating a change of vegetation cover over a large area. AVHRR data are useful 
for large scale land cover because of their multi-temporal nature; daily AVHRR data are formed into normalized 
difference vegetation index (NDVI) composites. NOAA-AVHRR were used to monitor the vegetation cover change 
due to pollution resulted from the Kuwaiti oil well fires. NOAA-AVHRR data sets, maximum 10-days NDVI 
images (GAC) time series for 1989-1993 were used. They have shown that NDVI is an excellent indicator to study 
green vegetation cover globally or locally. The NDVI profiles which were produced for many points for 1989 to 
1993 shows high reduction in NDVI values in most parts in 1991. Not only the peak of NDVI values in the growing 
season decreased, but also the area under the NDVI curve decreased in 1991. 
1 Introduction 
The NOAA-AVHRR (Advanced Very High Resolution Radiometer) is a part of a system which has been 
operational for over 15 years and which looks like being operational beyond the end of the present century. AVHRR 
was designed as an instrument of polar-orbiting satellite for meteorological purposes. The data collected by the 
AVHRR have come to be used for a whole range of purposes, not just in meteorology and oceanography for which 
it was planned but in a whole range of applications that were never envisaged when the system was originally 
specified and designed (Cracknell, P., 1996). The data from NOAA-AVHRR have been an appropriate tool to detect 
any changes in vegetation cover and to determine its distribution pattern. Using interpretation of time series, the 
main locations that changed are identified. The most valuable effect of the use of NOAA-AVHRR is that this 
satellite source is the only instrument that is capable of demonstrating a change of vegetation cover over a large 
area. In this study the NOAA-AVHRR were used to monitor the vegetation cover change due to pollution resulting 
from the Kuwaiti oil well fires since 1991. 
2 Objectives 
The main objective of this study is to demonstrate the changes in NDVI values, i.e. the index for detecting of 
changes of vegetation cover in the growing season (January to June) in south and southwest of Iran in 1991 using 
the NOAA-AVHRR data. 
3 Methods and approach 
NOAA-AVHRR 8km resolution (GAC) time series data for five years (1989-1993) were used to study vegetation 
cover, 10-days maximum NDVI values produced by NASA and re-scaled from 0 to 255. In order to distinguish 
variations in NDVI values in different years, the vegetation map was superimposed on the NDVI images. Sample 
points were selected in each vegetation cover type. The sample points were stratified and randomly distributed in 
different vegetation units. The vegetation types that cover the south and south-west of Iran are important for animal 
husbandry. Therefore, the investigation of vegetation change is focused on this region. For each sample point a 
NDVI profile was drawn for the whole year in the period of 1989-1993. These points were randomly selected and 
the NDVI profiles of these points during the five years (1989-1993) were plotted. In order to follow the variations of 
NDVI values in the five years, the 10 days maximum NDVI value for some were plotted points versus decades 
(from 1 January). To visually interpret the changes in NDVI values in different years, the values were calculated and 
the NDVI profiles were plotted for the sample points in the same scale for different vegetation cover types. In order 
to find the variation of NDVI values in relation to the distance from the source of the pollution, several samples 
  
  
International Archives of Photogrammetry and Remote Sensing. Vol. XXXIII, Part B7. Amsterdam 2000. 427