Snow is a solid form of precipitation. The formation of snow crystals inside a cloud depends on the presence of ice nuclei, water vapor, motion of cloud and temperature of cloud. The snow crystals are formed when cloud temperature is below freezing temperature. These crystals experience diffusional, accretional and aggregational growth inside the cloud. At temperatures greater than about -5°C, crystals usually stick together and aggregation of ice crystals form snowflakes. Snowflakes obtain their maximum size when the temperature of cloud is near 0°C. Aggregation of ice crystals ceases at temperatures below -20°C. Due to changes in the ambient condition of the cloud, the size and form of the crystals change when they descend through a cloud. In a typical cloud, a snowflake having 1 mm diameter can grow to 10 mm in about 20 minutes. Because of irregular shapes of snowflakes, measurement of their linear dimensions is difficult. The maximum diameter of snowflakes may range from 0.1 mm to several centimeters. The snow falls from the clouds mainly in the form of branched hexagonal crystals or stars. Other observed shapes include hexagonal rods, needles and plates. Snowflakes have the large surface area and fall at slowly as compared with raindrops. Therefore, snowflakes scavenge more atmospheric aerosols than rain. Precipitation as snow occurs when the atmospheric temperature is near to 0°C or less than that. As compared to rainfall, snowfall occurs more uniformly, but its distribution on the ground is highly influenced by the wind. In order to obtain the water equivalent of the snow, actual depth of snow unaffected by wind and its density are needed. 

 

Accumulation of snow on the ground leads to the development of snowpack, which store the water in the form of snow. Depending upon the climatic conditions, the snow at ground can stay for a period as long as several months. The global distribution of snow shows that a major portion of the Northern Hemisphere is covered by snow during winter. Snow has a great scientific interest because of its practical importance and utility in many fields, specially, in water resources, climate and winter sports. For many countries, like USA, Canada, India, China, Pakistan, Afghanistan, Russia, Nepal and European countries, snow melt runoff is a vital source of water supply for the purpose of drinking, irrigation and hydropower generation. Several important rivers, like Columbia River in the U.S. and Canada, Rhine River in Europe, Indus River in Asia, get substantial contribution from snow melt runoff. Melting of snow provides over 70% of the water supply for the western United States (Chang et al., 1987).

 

In general, melting of snow starts in spring when temperature starts rising. The depth and extent of snowpack reduces due to melting of snow and water is released from the snowpack. Snowmelt runoff estimates are needed for forecasting seasonal water yields, river regulation, reservoir operation, determination of design floods, design of hydrologic and hydraulic structures planning flood control programs, etc. The snowmelt is estimated either using the energy balance approach or the temperature index method. An energy balance method requires the information on radiation energy, sensible and latent heat, energy transferred through rainfall onto the snow and heat conduction from the ground to the snowpack. When all the components for energy balance computation are known, the melt rate can be expressed as (Singh and Singh, 2001)

 

where M is the depth of melt water (mm d^-1), Q_m  is the net energy flux available for melting (kJm^-2d^-1),  L is the latent heat of fusion (333.5 kJ kg^-1), ?_w is the density of water (1000 kg m^-3), and  ß is the thermal quality of snow. The thermal quality of snow depends on the amount of free water content (generally 3-5%) and temperature of the snowpack. Assuming ß = 0.97 for a thermally ripened snowpack, the above melt equation reduces to

Snowmelt follows almost diurnal pattern of energy availability and thus, it is possible to determine the diurnal variation in snowmelt by applying the diurnal distribution of energy received on the snow surface. Albedo, the most important parameter controlling the absorption of solar radiation, is very high for snow. For fresh snow, albedo is about 0.80-0.90, suggesting that most of the short-wave radiation is reflected back to the atmosphere from the snow surface.

In the absence of availability of detailed energy data for computing snowmelt, the temperature index method is considered as the best substitute of the energy balance and is widely used. The air temperature, expressed as degree-days, is used for snowmelt computation. Since temperature is the most readily available data, therefore, the temperature index method is extensively used. The most common expression used for estimating snowmelt using temperature is given as

        M = D  (T_i - T_b)

where M is the depth of melt water (mm d^-1) produced in a unit time, D  is the degree-day factor (mm °C^-1 d^-1), T_i is the index air temperature (°C), and T_b is the base temperature (usually, 0°C). Daily mean temperature is the most commonly used as an index of temperature for snowmelt. The degree-day factor is used to convert the degree-days to the depth of snowmelt. A wide range of degree-day factor (0.7 – 9.2 mm°C^-1 d^-1) has been reported in literature. However, the majority of the reported values of degree-day factor range between 3-5 mm °C^-1 d^-1 (Singh and Singh, 2001). Degree-day factor is influenced by the physical properties of the snow and, therefore, changes with time. Usually, temperature data used for melt estimation are available at few locations in a basin. These available temperatures can be interpolated/extrapolated to the different altitudes using the temperature lapse rate, (usually, 0.65EC/100m).