Summary

Historically, the use of snow for human well-being has been mainly for recreation, such as skiing. Recently in Japan, there is a mounting interest in using snow for various purposes that better serve life in snowy areas. This paper reports a feasibility study of a totally new idea of gathering and storing snow as a “snow dam”, and using the delayed meltwater to supplement water demand in summer.

The idea is not to build a structure that stands like a concrete dam, but to gather snow in gorges where siting conditions and technology are appropriate, for example in a region of very heavy snow where there is an urban area to be served which in summer runs short of water.

Western parts of Japan’s main island receive a large quantity of snow due to the seasonal wind blowing across the Sea of Japan. The city of Arai in this area is chosen to be the study site, on the western side of which is the Nakano River basin, bounded by 2000 m high mountains where the erection of snow dams is proposed.

Technology is needed both to gather snow in gorges and to reduce rates of melting, so that water can be extracted to provide a supply when it is most needed.

Various kinds of technologies are reviewed. It is concluded that none can at present furnish a quantitative answer as to how much snow can be gathered.

Technology to control and delay the melting consists of treating snow dams and predicting snowmelt runoff. The latter requires research on small-scale runoff analysis in a dam located deep in a gorge, and the calculation has to be done well through the summer season. A model developed for this purpose is applied to the study site.

Snow and avalanche data have been collected in the Nakano River basin. Several gorges have been selected as possible snow-dam sites, and the model assumes that a reasonable amount of snow has been gathered in them. A distribution model has been applied to these conceptual snow dams, which takes into account both topographical and meteorological distributions, as well as that of solar radiation within the basin. The model combines a calculation of net heat influx into the snow surface (Reference Koike, Takahasi and YosinoKoike and others, 1985) with a calculation of direct and indirect runoff components (Reference Lu, Koike and HayakawaLu and others, 1989), routing the former down the meshed channel network and the latter with the lumped model.

A snowmelt calculation started on 3 April 3 1990, demonstrates that the conceptual snow dams increase summer stream flow well into August.