Search

Nowadays, machine learning (ML) algorithms have been widely chosen for classifying satellite images for mapping Earth's surface. Support Vector Machine (SVM) and Random Forest (RF) stand out among these algorithms with their accurate results in the literature. The aim of this study is to analyze the performances of these algorithms on land use and land cover (LULC) classification, especially wetlands which have significant ecological functions. For this purpose, Sentinel-2 satellite image, which is freely provided by European Space Agency (ESA), was used to monitor not only the open surface water body but also around Marmara Lake. The performance evaluation was made with the increasing number of the training dataset. 3 different training datasets having 10, 15, and 20 areas of interest (AOI) per class, respectively were used for the classification of the satellite images acquired in 2015 and 2020. The most accurate results were obtained from the classification with RF algorithm and 20 AOIs. According to obtained results, the change detection analysis of Marmara Lake was investigated for possible reasons. Whereas the water body and wetland have decreased more than 50% between 2015 and 2020, crop sites have increased approximately 50%. [ABSTRACT FROM AUTHOR]

Advances in remote sensing technology enable monitoring and detection of these vulnerable water bodies that bear numerous functions and ecological services beyond their intended use. As such, valuable data and information may be provided for long and short‐term analyses and stored in a database for future projections. This paper initially grouped 18 natural lakes of the Konya Closed Basin of Turkey according to their salinity level. Freshwater, saline and brackish lakes have then undergone temporal analysis for every 5‐year intervals via Landsat satellite images. Freshwater lake surfaces have not changed noticeably during the inspection period; however, saline waters have decreased in surface area by 32%. Beysehir Lake constituting 94% of the freshwater category and Tuz Lake representing 97% of the saline group were selected for the short‐term analysis that was monthly conducted for years 2017 and 2018 with optical and SAR images to better verify cause and effect relationship. [ABSTRACT FROM AUTHOR]

Lakes and wetlands have a variety of uses in keeping the ecological balance and the well-being of a basin. Turkey has almost 303 water bodies among which Konya Closed Basin bears 29 of those where two of them have been registered as Ramsar sites. In this study, one of these sites, Meke Lake, and its nearby Akgol Wetland and Acigol Lake that are of international value are temporally assessed by means of satellite images regarding their water surfaces. Meke Lake has been declared as a Ramsar site in 2005, whereas Acigol is one of the volcanic lakes of the world with a depth of 300 m. Akgol has been registered as 1st level Natural Site and Natural Protection Area. Four classified images belonging to years 1990, 2000, 2010 and 2016 are further correlated with the ground truth measurements to assess the temporal variation of the sites. Producing such scientific data and conducting similar analyses are important for especially the local, regional and national authorities and planners in charge of the rehabilitation and restoration of natural sites that have lost a significant amount of land and water in the past years. [ABSTRACT FROM AUTHOR]