Fall velocity, mass and cross-sectional area of snowflakes are simultaneously observed in order to obtain the conditional probability function of fall velocity for a given mass.
The results are summarized as follow:
1) The frequency histograms of the mass, of the cross-sectional area and of the fall velocity are obtained for 4 snowfalls (see Fig. 3). The mode and average of the masses, of the cross-sectional areas and of the fall velocities over all snowfalls are 0.5 mg, 3.6 mg 1.5 mm
2, 48 mm
2 and 70-80 cm/s, 100 cm/s respectively. These values may be considered to represent the typical properties of snowflakes observed in the Hokuriku district.
2) The population containing the data of all snowfalls is divided into sub-populations by the interval of mass of 1 mg successively, and the conditional probability function of fall velocity is separately determined for each sub-population. These conditional probability functions can be expressed by a linear combination of several normal distributions with few exceptions. Of these normal distributions, there are two main normal distributions and each of them contributes to the corresponding conditional probability function at the rate of 30% or more.
3) For each sub-population, the average fall velocity of the normal distribution depends on the mass, though the standard deviation is roughly constant (12 cm/s) regardless of the mass.
4) The relationship between the cross-sectional area
S (mm
2) and the mass
M (mg) is shown by the following equation on the average:
M=0.012S 3/2.
View full abstract