Retention Time in the Golden Horn Estuary

This paper presents the results of a numerical study on the retention time in the Golden Horn, a natural estuary that connects with the Bosphorus Strait at the point where the Bosphorus meets the Sea of Marmara. The numerical model is a three-dimensional model, incorporated with the large eddy simulation model for the horizontal turbulence closure and a modified mixing-length model for the vertical turbulence closure. The retention time was based on the so-called volume-averaged concentration, and the latter quantity was calculated for an initially uniformly distributed instantaneous volume source by solving the turbulent diffusion equation. The retention time was calculated for two cases, namely, (1) for the entire volume of the Golden Horn, and (2) for only the shallow-depth zone of the estuary, extending 4,300 m interior from the Kagithane Creek. In both cases, four different scenarios were tested in terms of the head difference between the Black Sea and the Marmara Sea, Δh, the main parameter that governs the flow in the Golden Horn Estuary: (1) the case of mean flow with Δh = 33 cm; (2) that of mean flow but with inflow from Kagithane and Alibeykoyu Creeks present; (3) that of extreme flow, with Δh = 50.0 cm; and (4) that of another extreme flow, with Δh = 0cm. It was found that the retention time of the Golden Horn Estuary appears to be O(3) days, irrespective of the scenarios studied, with the exception that it is O(2) days in the scenario where the inflows from the Kagithane and Alibeykoyu Creeks are also present. (Here, O is the order of magnitude.) Regarding the retention time of the shallow-depth zone alone of the Golden Horn, it was found that this latter time is increased by a factor of 1.5–4, being in the range of O(4)–O(12) days. This is with the exception of the scenario where the inflows from the Kagithane and Alibeykoyu Creeks are also present, in which case the retention time is reduced to only O(1) day. Overall, the self-cleaning of the Golden Horn occurs over a time scale of O(3) days, on average, with no inflow from outside pollutants, while this time scale is reduced to O(2) days when there exists, on average, a constant inflow from the Kagithane and Alibeykoyu Creeks. The results have been interpreted in terms of dispersion caused by the variation of the velocity in the transverse direction combined with the turbulent diffusion in the same direction.