This paper introduces recent studies that measured the water retention curve (WRC) of snow in a cold laboratory. We then discuss the possibility of modeling the unsaturated hydraulic conductivity (K) of snow based on soil physics theories. Recent studies reported the dependence of the WRC of snow on both the sample grain size and density. In addition, they pointed out many similarities between the WRC of snow and sand, and then modeled the WRC of snow using two standard soil physics models, the Brooks and Corey model (BC model) and the van Genuchten model (VG model). Based on a comparison of the results of the models, they concluded that the VG model reconstructed the distribution of the measured WRC of snow better than the BC model did, although both models were in good agreement with the experimental WRC for snow. The modeled relationship between suction and the K of snow strongly depends on the model chosen for the WRC and on the value of the pore-connectivity coefficient (l) in the model. Therefore, the eventual determination of the hydraulic properties of snow will require future studies of the dependence of l on snow characteristics.

View full abstract

Snow hydrological studies were carried out in two experimental watersheds in Hokkaido to examine the mechanism of spring snowmelt runoff during a rain-onsnow event. A study of the runoff process within the snowpack showed that not less than 90% of the water that flowed out from the bottom of the snowpack had been stored within the snowpack before the event, and that this rate did not vary much based on whether it was a snowmelt-only event or a rain-on-snow event. A study of the peak streamflow under a rain-on-snow event showed that the peak streamflow was caused by the large initial streamflow just before the hydrograph rise and by the large inflow intensity comprised of the rainfall intensity and the snowmelt intensity. Furthermore, an artificial heavy rainfall on the snow surface was experimentally simulated, and although 200 L of water was sprinkled on 1 m² of snow surface for 6 hours, no outflow appeared from the bottom of the snowpack. This result indicated that a horizontal flow component became remarkable beyond expectation when a lot of water was supplied over the snow surface. To clarify the detailed processes of spring snowmelt runoff under a rain-on-snow event, it is crucial to know how much water is supplied to the ground surface under the snowpack. To do so, we need to accumulate more observation results regarding the amount of snowpack bottom outflows using a snow lysimeter.

View full abstract

The regions facing the Sea of Japan are known as the area that experiences the heaviest snowfall in the world. In those regions, precipitation due to snowfall is more important as a water resource than rainfall. During winter, an anticyclone is formed over the Siberian continent because of strong radiative cooling. The towering Tibetan Himalayas to the south block southward advance of this anticyclone. The cold and dry air mass blowing out of the Siberian anticyclone toward the east becomes unstable over the Sea of Japan, and then, this air mass absorbs heat and moisture vapor from the sea. This produces sequential cumulus convections, which land on the coast of the Japanese islands. In cumulus clouds, high moisture rates and cold temperature accelerate the formation of snow particles. In addition, because these clouds collide with the Japan Alps, they are forced to climb upward, and therefore, they cause a considerable amount of snowfall. Therefore, the mountain range experiences exceptionally heavy snowfall that is extreme by world standard, and in spring, the melting snow becomes a valuable water resource for the region. Snow plays the role of a natural white dam by accumulating in watersheds during winter.
It is said that mountainous areas are highly sensitive to global-scale environmental change. However, Mt. Fuji Weather Station, which used to be a symbol of meteorological observation in mountainous areas, has been unmanned since August 2004. Of the other observation sites under the Japan Meteorological Agency, the one at the highest elevation is the station at Nobeyama, at 1350 m elevation. It is necessary to evaluate the effect of a global-scale limatological change on environmental changes at a regional level, e.g., at the Japanese Alps at a high altitude of 3,000 m. We must note that the lack of meteorological observation data for high altitudes is a very serious limitation that affects the efforts for evaluating the effect of climatological changes on the water resources in mountainous ranges. Therefore, Institute of Mountain Science, Shinshu University has developed a network of meteorological observations in the Japanese Alps region.
It is difficult to observe snow depth in a mountain region with extremely heavy snowfall. Moreover, it is almost impossible to observe snowfall as precipitation in a mountainous location with no commercial power source. Therefore, a method for estimating winter solid precipitation in mountainous regions, using a snow chemical technique is being developed.

View full abstract

In the mountainous regions of northern Mongolia, there are peculiar different patches of the landscape "ecotone" with north-facing forest slope and south-facing grassland slope. The slopes are distinguished as marked contrast of permafrost distribution as well. The present study revealed ecohydrological processes in terms of evapotranspiration and water balance based on continuous observation since 2004 in the ecotone area of Shiljiree river watershed, upper part of Tuul River in Khentii Mountains. Soil moisture amounts following rainfall during early growing season have primary importance on interannual variations in transpiration from larch forest slope. In addition, timings of snow disappearance and thawing active layer affect subsequent foliation of trees and their transpiration. The estimation of water balance during summer 2006 demonstrated that the larch forest slope with underlying permafrost suppressed the evapotranspiration with half of amounts in grassland slope. It means that precipitation and soil water in forest slope can be partitioned to evapotranspiration and river runoff. On the other hand, grassland at south-facing slope with no underlying permafrost has remarkably large amounts of evapotranspiration and therefore both precipitation and soil water was consumed entirely to evapotranspiration. The results depicted that the coexistence between permafrost and forest in mountain slope has an important role in sustaining water resource.

View full abstract

According to the Köppen -Geiger climate classification, Hokkaido and central to southern Alaska are located at near southrn and northern borders of the subarctic region, respectively. River basins in Hokkaido and Alaska receive similar climatological and hydrological conditions of summer to autumn rainfall, winter snowfall and spring snowmelt. Here, hydrological characteristics of the Saromabetsu river basin (area, 277.0 km²) and the Oikamanai river basin (area, 62.0 km²) in Hokkaido are compared with those of the Tanana river basin (area, 6.63×10⁴ km²) in Alaska. Hourly time series of river discharge, Q, and suspended sediment concentration, C, in runoff events offers the relationship between Q and C for the three river basins. The C to Q relation for the Tanana and Saromabetsu rivers exhibited the clockwise hysteresis, while that for the Oikamanai river basin furnishes the counterclockwise hysteresis during rainfall runoffs. The counterclockwise loop suggests that the throughflow under the basin slope erodes and transports soil particles. The runoff analysis was conducted by the tank model to simulate daily discharge time series of the Saromabetsu and Tanana rivers in snowmelt and glacier-melt seasons, respectively. Optimizing the tank parameters for discharge time series of a certain year, the simulations for snowmelt and glacier-melt runoffs in the other years were reasonable with the root mean square error at 7.6 to 17 % of observed discharge and the Nash-Sutcliffe efficiency coefficient at 0.45 to 0.97. The high applicability of the tank model is probably caused by the consideration of water storage in snow and glaciers, and by each similar hydrological condition of initial soil water content and initial englacial water storage in the snowmelt and glacier-melt seasons.

View full abstract