LAND USE/ COVER CHANGE DETECTION IN GNEADA, TURKEY

Introduction

Igneada Longos (alluvial) forests have a multiple role in the landscape because of their ecological, biological, environmental and economic importance.

This research took place in floodplain forest in NW Turkey. The area is also known as one of the most important plant areas in Turkey. Igneada and the surrounding environment have been recognized as an important biodiversity hotspot due to their unique characteristics. Because, the area houses sensitive ecosystems, the parts of it were previously protected as Nature Protection Park, Natural Site, and Wildlife Protection Area.

Aim of Study

Determining land use/ land cover changes in Igneada between the year of 1984 and 2010 by using Landsat 5 TM data set.

Study Area

The Igneada Longos Forests National Park, located on the Black Sea coast far from 15 km from the Turkish-Bulgarian border. The study region lies on an area that is approximately 5757 ha located between the northern latitudes 41o44 ‘43'' and 41 o58 ‘27'' and the eastern longitudes 27o44 ‘52'' and 28 o39 ‘17'' (see Figure 1).

Figure 1

 

Igneada Longos (alluvial) forests with associated aquatic and coastal ecosystems include seawater, sand dunes, freshwater and saline lakes, wetlands, mixed forests of deciduous tall trees, and riparian ecosystems (Figure 1 and Figure 2). Upstream forests and water resources are critical to maintain the delicate balance of the alluvial forest and wetland ecosystem of Igneada. This area covers several rare and endemic plant species. Despite its ecological sensitivity and importance, Igneada has been under serious threats. Supplying drinking water project to Istanbul using upstream water sources is an important threat.

Method

1984, 1990, 2000 and 2010 dated Landsat 5 TM data were used in the study.

Figure 3

 

Image pre-processing was conducted to eliminate atmospheric distortions, sensor problems and geometric distortions (Figure 3). After image pre-processing Maximum Likelihood supervised classification method used to derive land use & land cover categories of the region. Accuracy assessment of classification determined using Kappa statistic and error matrix. Change detection applied by comparing classification results.

Conclusions

Different dated Landsat 5 TM data were classified using supervised classification method. Figure 4 shows classification results (thematic maps) and table I gives statistical results. According to classification results sand dunes have been increased and settlements have been decreased between 1984 and 2010.

 

Figure 4