Geospatial Analysis of Development Trends in the Mediterranean Coastal Landscape

Abstract

This paper demonstrates a methodology that aims to monitor landscape changes resulting from tourism development in the case of the Mediterranean coast of Turkey (Mersin). The methodology relies on geospatial analysis techniques for quantifying and mapping development trends. Digital image analyses supported by field works show that building construction along the coast, mainly for tourism purposes, is extensive. Though this trend occupies a very narrow strip of the coast (i.e., approximately 100-500m), it consumes productive soils and lead to landscape fragmentation.

Key Words: Geospatial analysis, change detection, development, Mediterranean coast, Mersin

Introduction

Mediterranean coastal landscape is subject to increasingly complex land use/land cover (LU/LC) changes (Alphan and Derse, 2013; Kuleli, 2010; Alphan and Yilmaz, 2005, Cetin et al., 1999). Majority of these changes occur as a result of urbanization, tourism, agriculture and transportation activities. One of the most important components of the contemporary landscape planning is to analyze these changes in terms of cause-consequence relationships and simulate future behaviors of the observed change.

Change information for landscapes serves as an ecological and geographical basis for environmental resource planning. Information derived from change detection facilitates understanding of underlying causes of landscape changes, while future projections based on the observed change lead to more sustainable management plans. These projections also help to monitor effectiveness of management plans and provide a basis for plan revisions. Diversity and extent of human activities on the coast results with complex changes in short term (Alphan, 2011, 2012). Therefore, high temporal and spatial resolution of change detection may facilitate analyzing landscape changes. This study focuses on monitoring Mediterranean landscapes using geospatial information and analysis techniques.

Study Area and Methodology

This study is conducted in the coastal zone of Mersin, located in the eastern Mediterranean region of Turkey. Typical Mediterranean climate with mild winters and dry summers prevail in the region. Due to its climate and proximity to major cities in the Eastern (Mediterranean) region of Turkey, the area has experienced problems due to development of multistory buildings serving as summer apartments near the coastline. This development tends to continue to the west along the coast, transforming agricultural lands and (semi)natural vegetation into large multistory apartment blocks. Simultaneous improvements of transportation infrastructure to/from this region also support this development. Rural settlements near the coastline tend to expand due to population growth and expanding agriculture.

The main character of the landscape change in this coastal zone is that building development occurs only in a narrow strip on the coast. Therefore, analysis of the change phenomena in this region requires datasets with high spatial resolution dating back to 1980s, acceleration of tourism development on the coastal zone. In this context, panchromatic SPOT (Satellite Pour l'Observation de la Terre) datasets that have advantages in terms of both high spatial resolution (10 m) and frequent temporal coverage for landscape monitoring and Landsat datasets with the advantage of high spectral capabilities (i.e., visible and near-infrared bands) were used for monitoring above mentioned change trends. SPOT and Landsat images were combined to enhance spectral and spatial capabilities for analyses that used four image datasets between 1989 and 2007. Images were classified and change information was derived using these classifications.

Results

This paper reflects preliminary results of an ongoing research project aiming to monitor qualitative and quantitative aspects of development in the coast of Mersin, Turkey. For this purpose digital image processing techniques including image enhancement and change analysis were employed to detect landscape changes. This change information may serve as baseline information, which may be used for analyzing negative impacts of changes on the coastal landscape and modeling future development. Quantitative methods such as landscape pattern analysis can be used for analyzing qualitative aspects of the change, while cellular automata may be used to create future predictions for development in the light of observed change trends.

Acknowledgements

            This project is supported by Scientific and Technological Research Council of Turkey (TUBITAK) under the grant number CAYDAG 111Y253.