maryam khosravian; AliReza Entezari; Abolfazl Rahmani; Mohmmad Baaghide
Volume 4, Issue 13 , March 2018, , Pages 99-120
Abstract
Expanded Abstract
Introduction
It is very difficult and costly to collect information about the continuous changes of vegetation by conventional methods. With the development of the satellite technology, the satellite imagery has provided extensive access to information about ground sources. The detection ...
Read More
Expanded Abstract
Introduction
It is very difficult and costly to collect information about the continuous changes of vegetation by conventional methods. With the development of the satellite technology, the satellite imagery has provided extensive access to information about ground sources. The detection of changes is one of the main factors in the study of the relationship between human activities and the environment. Most remote sensing products are used to evaluate and estimate the bio facial and biochemical parameters of plants from broad bands such as NOAA, AVHRR, SPOT, and TM / ETM. Landsat sensors often consist of three to seven bands. Vegetation indicators are mathematical transformations that are defined based on different gauges of the sensors and designed to evaluate and investigate plants in multi-spectral satellite observations. Lake Parishan Basin is located in Fars Province (Kazeroun city). In recent years, many environmental and human factors have had an adverse effect on its ecosystem. The interference of the agricultural lands with the marginal lands of this lake, the disposal of chemical fertilizers, the flow of pesticides from plant pests to the lake, burning of grasslands, and bogs of the surrounding countryside by profitable individuals in order to expand agricultural lands, are factors that affect its ecosystem. The purpose of this study was to investigate Lake level changes between the years 1989 and 2004.
Materials and Methods
Lake Parishan Basin is located in the geographic coordinates of 29˚ 25ˊ 12˝ to 29˚ 36ˊ 15˝ north latitude and 51˚ 40ˊ 50˝ to 51˚ 48ˊ 20˝ east longitude in Kazeroon, Fars Province. In this study, multispectral sensor data was used to investigate the changes in the lake level. Landsat ETM + 1: 50000 satellite and digital elevation data were used to carry out atmospheric and geometric corrections on satellite data. The Satellite data was interpreted and processed in ENVI 4.1 software and performed in ARC GIS 9.3 Cartographic Mapping Software. The datasets of this study were from Landsat satellite imagery of 1989 and 2004. There has been a very intense occurrence since 2004. The process of drying the lake so that nothing remains in 2007 can be checked using remote sensing techniques. The data processing process was carried out in three stages of pre-processing, processing, and post-processing. Regarding the necessity of geometric correction in detecting water level changes, the topographic maps of 1: 50000 and GPS-controlled points were collected. Using the reference grounding function in the PCI-Geomatica software environment on each image, the control points were made. The pixels were re-evaluated using the nearest neighbor method. In addition, images with an error of 0.39 and 0.42 RMS were correlated.
Results and Discussion
Different indices (SAVI), (NDWI), (NDMI), (MNDWI), (NDVI), and (AWEI) were separately extracted in order to identify distorted lake changes in the period of 1989-2008.The detection and identification of water level was done to study the relevant indicators of water level changes. The level of lake changes between the years 1989 and 2004 was evaluated and extracted in various indices. In addition, an algorithm was designed in ENVI software environment. The results showed that it was incapable of extracting the surface water of the lake. The NDMI index (with the lowest overall accuracy and kappa coefficient) was also considered due to its excessive sensitivity to blue areas.
Conclusion
The changes in the water level of the distressed lake using vegetation and water indices during the period of 1984-2008 were investigated. The results of this study showed that the normalized dispersion index (NDMI) with the lowest overall accuracy and kappa coefficient was unable to extract water from the surface, while the NDWI had the highest accuracy for the extraction of lake water level changes. Accordingly, the Normed Water Difference Index (NDWI) was used to model the spatial and temporal changes in the level of the disturbing lake in the period of 1989-2008. The results of this study also indicated that the decline in the lake water level between 2000 and 2004 was 19,160 m2. Moreover, the decline between 1989 and 1991 and between 1991 and 2000 were respectively 1344010 m2 and 1313000 m2. It was also estimated that the total changes in the water level of the lake amounted to 1494470 m2 between 1989 and 2004.