Evaluating the thermal stratification of Koycegiz Lake (SW Turkey) using in-situ and remote sensing observations

Abstract

The goal of this study is to evaluate the thermal properties of Koycegiz Lake using in-situ measurements and satellite based thermal infrared imagery. In-situ measurements of surface water temperature and water depth, as well as meteorological data, were used in the analysis. Images of the Landsat 8 TIRS (Thermal Infrared (IR) Sensors) at IR channels were taken from the data archives of United State Geological Survey (USGS), and were validated with surface in-situ measurements. Specific Electrical conductivity (SEC) measurements done at the bottom of the lake were also used for a better understanding of the water thermal stratification and types (cold/hot) that impact on the underwater physical properties of the lake. In-situ measurements at the water surface were interpolated using a new method known as 'Empirical Bayesian Kriging (EBK)'. The 2D and 3D (dimensional) temperature maps were generated using the Delaunay triangulation technique which utilizes nearest neighbor interpolation method. The results show that the correlation coefficient (R-2) between water surface temperatures (WST) from in-situ measurements and IR channels is found to be greater than 0.86. The analysis of 2D and 3D temperature mapping of the lake water revealed evidences for subaqueous cold spring occurrence in the northern part and subaqueous hot springs in the south-western part of Koycegiz Lake. It is concluded that correlating in-situ water temperature measurements to Landsat 8 TIRS images is an efficient method for temporal and spatial evaluation of lake water characteristics.