In order to explore the mechanism of the expansion of supraglacial lakes in the Himalayas, the physical structure of three glacial lakes, Tsho Rolpa, Imja, and Lugge, increasing in surface area, was investigated by the filed observations. The lakes exhibited different thermal and sedimentary structures, depending on the wind-mixing condition and the turbidity and inflow of meltwater at the glacier terminus. The difference of the wind-mixing condition in the upper layer was judged to be due to the different degree of the topographic screening effect of the end moraine or the dead-ice zone on the valley winds near the water surface. Tsho Rolpa has little screening effect on the valley wind, since the crest of the end moraine upwind of the lake is almost level with the water surface. This topographic condition can induce strong wind mixing and wind-driven currents in the surface layer. Imja and Lugge have the dead-ice zone at downlake, the height of which is 10 to 25m (Imja) or ca. 28m (Lugge) above the water surface. This topographic situation produces a screening effect on the valley wind, which decreases the wind velocity near the lake surface and weakens the wind-mixing in the surface layer. This suggestion was certified by the numerical simulation of airflow, using the three-dimensional topographic models of actual size around Tsho Rolpa and Imja. A conceptual model for the hydrodynamics of Tsho Rolpa was proposed from in situ measurements of water temperature, turbidity, and flow velocity. The lake hydrodynamics were simulated by making up a three-dimensional lake basin in the calculation domain. The fields of flow velocity, pressure, water temperature and suspended sediment concentration in the lake were calculated in steady state. In the simulation, the strong return flow (Fig. 12A to C) was observed in the middle layer, where flow velocity was not monitored in the field observation. As the next step for observations in Himalayan lakes, referring to such simulated results, the location of observation sites should be determined. The simulated spatial distributions of concentration and particle size for suspended sediment were agreeable to the observation. The history of the lake expansion or the temporal development of the lake basin by the melting of debris-covered ice needs to be simulated by considering the seasonal and interannual variations of thermal environments around the lake and the upstream glacial movement, including the calving at the glacier terminus.
I am grateful to Dr. Tomomi Yamada, Representative, Snow and Ice Network, Nonprofit Organization (NPO) Corp. for consistent encouragement and invaluable advice to my Himalayan study. The field survey in Tsho Rolpa was supported as a GLOF (Glacier Lake Outburst Flood) research program by Japan International Cooperation Agency (JICA). The support team of CHAM-Japan, Inc. gave me advice to improve the numerical simulations.
Was this article helpful?