Regular land-based snow surveys are carried out since late nineteenth century. They consist of regular measurements of snow depth using a stake and of snow density by a snow sampler which is a cylinder tube made of steel pushed into the snow pack cautiously to dig it outwards. All the snow survey courses (routes) must be marked by distance, and the survey must be carried out regularly each 10 days. Only then the data series obtained will be representative and susceptible to any statistical procedures. To obtain the accurate data, the snow pack cores should be sampled in regularly located points along a survey route assigned as 1km for an open area, and 0.5km under a forest canopy. Based upon these data, the numerous regional summaries were performed to investigate regional climate.

Except for weighed sampler there are devices called snow pillow which are to detect what the pressure of snow cover is on the surface. The pillow itself is a tank about 10 m² in area made of rubber. It is filled of admixture of alcohol and fresh water to prevent freezing. Level of the liquid is measured by a float recorder in well linked with the tank by a tube laid under soil.

Another method is aerial gamma-survey based upon extinction of natural terrestrial gamma-rays. The method is widely operated with use of airplane-borne technique to detect snow water equivalent over large river basins of European Russia. As was found, the gamma radiation from underlying surface vary particularly during the "episodic" snowmelt in winter. It was also found, that the snow water equivalent at early spring can be overestimated as including that melt water collected within deep pools under snow. Besides the considerable errors occur due to spatial non-homogeneity of the snow distribution.

In Russia and many other countries the radar precipitation measurements are used to detect snowfall intensities and snow water content. Some previous research were undertaken since the 1980s in Central Asian mountains of former Soviet Union to retrieve snow water equivalent by absorption of the space neutron flux within a snow pack. This advanced method was proposed as one of the tools for detection of avalanches in high mountains with very deep snow pack.

For the first time, the snow cover distribution and its dynamics were detected remotely by "Tiros" satellite at mid 1960s, and the global monitoring system was developed since that time. The Russian Hydro-meteorological Service uses also the imagery from Russian satellites "Meteor" and "Kosmos". Land surface is continually scanned in visual (0.4-0.7 mm) and near infra-red (to 1.3 mm) spectral bands. At present, the Advanced Very-High Resolution Radiometer (AVHRR) is used as a satellite-based sensor to detect sea ice and snow cover extent.

Digital processing of the imagery consists firstly of subdivision of an area onto the snow-covered and no snow. By contrasting the picture, we can define, as far as possible, a several degree of snow depth by some threshold in a white-gray scale. The question arises how the snow cover can be separated from clouds. As was found by examining the pictures, the infra-red brightness of snow cover is 60 % less than that for the clouds which are white like snow only in visual spectral band. It is much more difficult to recognize the snow hidden beneath cloudiness which obscure land surface. For this aim, the skillful operator must make an examination the pictures as arranged in time.

The main problem of how to evaluate the water equivalent of snow by satellite pictures does not resolved hitherto. Micro-wave survey is the most suitable satellite based remote sensing of snow, for instance, the records of natural radiation of the snow cover in diapason from 6 to 37 GHz. This radiation flux well correlates to the snow water equivalent. To detect it more correctly, the additional factors must be examined: the liquid water content, texture of snow pack, size of snow crystals, thickness of crusts and snow surface contamination. Measurements showed that the temporal variations in micro-wave brightness occur during a "melt-freezing" cycle. As was found previously, a moist soil and snow crusts distort back-scattering from their interfaces. This has not been enough explored till now and requires further investigations in snow physics.

Global grid snow cover data are available owing to satellite imagery receiving each day. The most complete data have been collected at the National Snow and Ice Data Center (NSIDC) in Boulder, Colorado, USA. One of these sources is, for instance, the monthly global snow depth data. They were allocated into the grid cells each of 4 degree in latitude by 5 degree in longitude. Such global grid data sets are too coarse to draw up the snow accumulation at local scale, in particular, for small river basins, but they are very useful for climate research, e.g. to update the climatic and atmospheric circulation models.

Hydrosphere © 1990 — 2005 Vladimir A. Shutov
Design by Timothy (Azs) Shutov © 2001