Calculation and Forecast of Glacial Feeding in River Basins

A two-stage method has been developed for calculating and forecasting the annual volumes of glacial runoff feeding mountainous rivers. At the first stage, the series of morphological characteristics of glaciers are reconstructed using limited data from regional glaciation monitoring. An example of a numerical description and analysis of the annual reconstructed dynamics of glaciation parameters in the upper Rhone River (Switzerland) is presented. Similar results of reconstruction of annual values of the morphological characteristics were obtained for the basins of tributaries of the Terek River (North Caucasus) and the Western Kyzylsu River (Pamir). At the second stage, the calculation and forecast of the time series of the average summer air temperature T_s(Z_mean) at the height of Z_mean is performed, which is used as an argument for determining the ablation layer by the formula Ab = f(T_s) on the glaciation area F_gl. The annual vertical profiles of mean air temperature of April T₄ = T₄(Z), summer ones T_s = T_s(Z), and formulas for calculating T_s as a function of T₄ are constructed and used for the calculations. Thus, on a regional scale, it was established for the first time that the April air temperature T₄ allows calculating a thickness of the annual ablation layer Ab = f(T_s) with a month earliness the forecast at the average height Z_mean of the glacier. The reconstructed F_gl(t) series is used to obtain annual volumes of glacial alimentation. A regional study of variability of the index of glacial alimentation δ (Schultz, 1965) with time t was carried out using long-period measurements of runoff in the river basins of Eurasia, North America, Central Europe, and Central Asia. The index δ is equal to the ratio between the volumes W of flow or the average water discharges Q for the periods July-September (Q_7–9) and March-June (Q_3–6). As a result of the analysis of the expression δ = δ(t), it was found that the gradient of the linear trend equation for the δ index in all the above river basins is negative, which is indicative of a reduction in glacial-snow alimentation, or more precisely – in only its glacial component. Notwithstanding this, the annual runoff Q_year decreased only in three basins, and in the others Q_year, increased due to the growth of Q_3–6, which overlapped decreasing of Q_7–9. Index δ for the upper reaches of the Rhone River turned out to be not only a representative characteristic of changes in the vegetation period and annual runoff of the river, but also an efficient argument for the super-long-range prediction of these variables for 2025–2054 years.