THE INFLUENCE OF URBAN GROWTH ON SURROUNDING MEDITERRANEAN LANDSCAPES WITH PARTICULAR REFERENCE TO DEGRADATION OF OLIVE ORCHARDS

摘要

Monitoring land use/land cover (LULC) changes is important in making effective decisions in landscape planning and management. Quantifying the changing pattern of land helps us to understand how ecological impacts reveal and affect ecosystem services. The changes in landscapes can be assessed in different qualitative and quantitative ways. One of the aims of this study is to classify land cover using object-based image analysis and quantify the changes of landscape patterns by applying landscape metrics in the Efeler district, an urban settlement of Aydin province, during the 25 years from 1986 to 2011. In this context, a combination of satellite imagery, geographic information systems (GIS), and landscape metrics was used. A Spot-1 image dated 1986 and a Worldview Ortoready pan-sharpened satellite image dated 2011 were respectively processed for LULC classification. Six classes of land use including agriculture, bare soil, built environment, forest, main road, and river were identified. According to the overall accuracy assessment of the 2 classified images, the accuracy values were computed as 85.6 (for the LULC map dated 1986) and 89.8 (for the LULC map dated 2011). Change-detection results showed that during the whole study period there was a remarkable increase in the built environment and the main road, while bare soil decreased dramatically. It is also interesting that although urban expansion increased, the total area of agricultural land remained almost the same. The location of the agricultural land changed notably, but the percentage of this area was unchanged. Within the boundary of urban settlement, agricultural areas, and especially olive orchards, were fragmented due to the developmental strategies of Aydin. Within the scope of change -detection analysis, degradation of olive orchards was highlighted. Heterogeneous structure and fragmented patterns of land cover were analyzed using several metrics at different levels.