Identification of Canebrake level changes of the Zarivar Lake between 1984 to 2011, using the images of Landsat TM and ETM +

Abstract

In the past decade, remote sensing has been widely used to identify surface changes of different vegetation and their classification. Increasing the level of Canebrake of the Zarivar Lake and its risks for aquatic organisms living in the lake has become one of the most important issues in recent years. Therefore, the aim of this study was to identify surface changes of this Canebrake in the past three decades using Landsat TM and ETM+. For this purpose, bands 3, 4, and 5 of images were geo-referenced. RMSE were less than one pixel for all bands. The supervised classification method with a maximum likelihood algorithm was also applied to detect the changes of water area on the combined images (bands 5, 4, and 3) of months with full water in the lake. NDVI index was utilized to identify the surface changes of Canebrake on the images taken in the months with low water in the lake. The results show that the rise and fall of water area and surrounding canebrake has a direct correlation with a rainfall and increase in both levels maybe occur at the same time. Study on the coastal strip of water area with GPS and combined images showed that the coastal line had not a significant change in the past three decades.

Keywords

• remote sensing
• ndvi index
• classification
• water zone
• maximum likelihood algorithm