6.3 Snow cover
Snow measurements were conducted at the meteorological site of the Suntar-Khayata Station at the altitude of 2070 m. The triangle plot with the sides of 12 m was instrumented with three rails installed in the corners where snow height and volume weight were measured each 10 days (Koreisha, 1963). The snow data are available only for two years (1958-1959).
At the altitude of the Suntar-Khayata Station, snow cover is formed in early September and melts only in the second part of June – early July. The snow-free period lasts about 2 months, (on average 56 days in 1957-1959).
Snow water equivalent (SWE) varies greatly from year to year. Maximum observed SWE reached 348 mm in 1958 and only 153 mm in 1959. Maximum simulated SWE amounted to 363 mm in 1958 and 171 mm in 1959. Mean simulated SWE by the end of winter for the period of 1957-1964 was 251 mm.
Mean deviation value between simulated and observed daily SWE is 18.4 mm. The maximum deviation value reached 43 mm on November 20th, 1958. The absolute error of simulated maximum SWE is 5-12% during 1958-1959. Mean simulated SWE by the end of winter for the period of 1966-2012 was 213 mm and range from 114 mm (1983) to 379 (1967). The height of snow changed from 44 cm to 99 cm and the mean value was 70 cm.
The comparison of simulated and observed values of SWE during winter seasons 1958-1959 is presented in Fig. 4 and shows general model adequacy. It is noteworthy that modelling results show that the snow cover formed within 5 days from 5 to 10 August 1957 and then melted (Fig. 4). The modelling results of the same variable state at the Canadian Rocky Mountains watersheds shows that overestimation or underestimation of the peak SWE ranged from 2.4 to 16% for the upper watershed landscapes (Fang et al., 2013).
Also snow surveys were conducted at the territory adjacent to the Suntar-Khayata Station along the 3-km long route with elevation range over 400 m (2068-2477) in 1957-1959. Spatial variation coefficient of snow redistribution was calculated based on the data as 0.60. A normal distribution (Vinogradov et al. 2011) was used to statistically account for snow redistribution in the goltsy landscape based on this estimation. In general, this value is consistent with the data on snow variation for mountainous landscapes in the Yukon River Basin (SWE variation is 0.69 by McCartney et al. (2006), 0.48 by Pomeroy et al. (2004) and 0.69-0.84 in the upstreams of Kolyma River based on the data by Makarieva et al. (2018a).