MONITORING SHORELINE CHANGE OF ACIGOL ANDBURDUR LAKES IN TURKEY OVER 44 YEARS USINGREMOTE SENSING AND GIS APPROACHES

摘要

In this study, the main aim was to analyze the spatial changes of Burdur and Acigol Lakes by us-ing Remote Sensing approaches and geographical in-formation system (GIS) applications. Five m ulti-temporal satellite images from Landsat 2, Landsat 5 TM and Landsat 8 O LI were used to monitor and map the shoreline changes for both lakes. Support Vector Machine (SVM ) classification was used as an effective way of monitoring the changes between specified time intervals accurately. Object-Based Image Analysis (O BIA) is the second classification approach that was applied to all data in this study. Besides both classification applications, spectral wa-ter indexes including the Normalized Difference Water Index (NDW I) and M odified Normalized D if-ference Water Index (M NDW I) were used for the ex-traction of the water body area and digitization of these areas in the GIS platform. The results of both SVM and O BIA classifications indicated that Burdur Lake lost 40% of its water body as the total area of Burdur lake was 209 km2 in 1975 and about 125 km2 in 2019, while Acigol Lake lost about 70% of its wa-ter body, w ith a total area of about 77 km2 in 1975 and 24.6 km2 in 2019. Spatiotemporal changes of Burdur and Acigol Lakes based on the applied meth-ods show a significant diminishing trend in surface area between the time period of 1975 and 2019 in this study. The results show that remote sensing ap-plications including water indexes and classification algorithms are effective ways of identifying changes between spatial and specific time intervals.