ИСТИНА

The presence of dissolved salts in water brings new physical effects into the thermal and ice regime of a lake. However, only a limited number of studies has been conducted to simulate saline lakes. This presentation addresses two distinct effects caused be salinity which have not been well understood on a quantitative level before, and where a numerical model provides new insight into natural phenomena. The mid-depth temperature maximum (TeM) was measured in an estuarine Bol’ shoi Vilyui Lake (Kamchatka peninsula, Russia) in summer 2015. We applied 1D k–ε model LAKE to the case, and found it successfully simulating the phenomenon. We argue that the main prerequisite for mid-depth TeM development is a salinity increase below the freshwater mixed layer, sharp enough in order to increase the temperature with depth not to cause convective mixing and double diffusion there. Given that this condition is satisfied, the TeM magnitude is controlled by physical factors which we identified as: radiation absorption below the mixed layer, mixed-layer temperature dynamics, vertical heat conduction and water-sediments heat exchange. In addition to these, we formulate the mechanism of temperature maximum ‘pumping’, resulting from the phase shift between diurnal cycles of mixed-layer depth and temperature maximum magnitude. Based on the LAKE model results we quantify the contribution of the above listed mechanisms and find their individual significance highly sensitive to water turbidity. Relying on physical mechanisms identified we define environmental conditions favoring the summertime TeM development in salinity-stratified lakes as: small-mixed layer depth (roughly, ∼ < 2 m), transparent water, daytime maximum of wind and cloudless weather. We exemplify the effect of mixed-layer depth on TeM by a set of selected lakes. The saline Lake Uvs (Mongolia) is not stratified by salinity, but salinity plays an important role in ice cover formation here. In order to simulate this lake, the LAKE model was supplemented by a compartment calculating vertical distribution of salinity in the ice cover. The model was compared to data of in situ and satellite measurements of temperature and ice on Lake Uvs for the period from 2000 to 2015. It is demonstrated that neglecting water salinity in the model leads to ice build-up 16-17 days earlier compared to real dates. This error diminishes if the model takes into account the dependence of water density and freezing point on salinity; however, in this case, the model underestimates the maximum ice thickness, on average, by ≈0.2 m. The later error, in turn, is reduced by an order of magnitude, if the model accounts for vertical distribution and dynamics of salinity in ice. The work is supported by Russian Foundation for Basic Research (grant 16-55-44057, Lake Uvs study) and Russian Science Foundation (grant 17-17-01210, Bol’shoi Vilyui Lake study).