6. Heavy rain and snowfall

6.1. Uncertainties in precipitation measurements

6.2. General characterization of precipitation

6.3. Daily maxima of liquid precipitation

6.4. General data on heavy rain and snowfall

6.1. Uncertainties in precipitation measurements

Precipitation, whose intensity, duration period, and areal coverage can cause damage in both the economic and industrial sectors, can be classified as an HMP. It is particularly difficult to perform precise meteorological observations of precipitation in the Arctic due to significant measurement uncertainties, especially in winter during snowstorms. Actual values can exceed the apparent values recorded by a precipitation gauge by 200 to 400 percent during the cold part of the year and 20 to 30 percent during the warm season [59, 62, 65]. Three different types of inherent errors are associated with the standard precipitation gauge that is used by the arctic station network. These include moistening of the precipitation gauge vessel, evaporation of the precipitation from the collection bucket between observations, and the influence of wind on the precipitation collection process. The error due to moistening of the inner walls of the collection vessel is 0.2 millimeters for each measurement of liquid precipitation and 0.1 millimeters for solid precipitation.

The second error is caused by evaporation during the intervals between measurements of precipitation that has already been collected in the precipitation gauge. In the Arctic, this error is small. During the polar night it is equal to about 2 percent, but in spring it reaches 15 to 20 percent due to the intense solar radiation.

Table 37
Measured maximum amounts of solid precipitation and their corrected values
Station Date Sum of precipitation for 12 hours, millimeters
Measured Corrected
Russkaya Gavaní 16 Jan 1966 33.2 8.4
  19 Jan 1966 21.3 3.5
  11 Nov 1966 26.1 4.1
  23 Dec 1974 23.6 3.1
  24 Dec 1974 44.8 11.2
  25 Dec 1974 69.5 14.1
  26 Dec 1974 34.7 9.2
  27 Dec 1974 54.7 10.3
  16 Jan 1976 66.6 12.7
  17 Jan 1976 25.1 2.9
Malye Karmakuly 10 Feb 1966 29.7 2.9
  16 Jan 1967 28.8 3.7
  20 Nov 1968 37.0 4.5
  25 Dec 1968 30.9 4.2
  26 Jan 1971 45.9 5.9
  03 Nov 1971 59.3 6.1
  04 Jan 1972 69.0 6.9
  16 Jan 1972 35.5 4.3
Average   40.9 6.6

The principal measurement errors are due to the influence of the wind. On the one hand, precipitation measurements are not completely recorded due to the disturbance of the wind flow at the instrument, and on the other hand, when solid precipitation occurs in winter and the wind speed increases, blizzards start blowing snow up from the surface. This snow accumulates in the buckets of the precipitation gauges and is recorded as if it were precipitation. This is the largest error, and it is difficult to account for it. In the Arctic, all large snowfalls are recorded under these circumstances.

A method for correction of the precipitation readings has been developed at AARI [5]. Data from two arctic stations at which snowstorms are observed very frequently, and where snow accumulation into the gauges is known to occur, are presented in Table 37. These data illustrate how large the solid precipitation measurement errors are. On average, the corrected precipitation value is more than six times smaller than the recorded total. At other stations the analogous corrections in the precipitation sums are factors of two to three. From this it is clear that in the Arctic it is absolutely necessary to make appropriate corrections to obtain reliable results. Currently, however, the only error taken into account by the Arctic stations is that of moistening of the precipitation gauge vessel.

6.2. General characterization of precipitation

Atmospheric precipitation in the Arctic is characteristically two to three times less than in the midlatitudes [5]. For example, the total precipitation for the Arctic is 10 to 30 millimeters in January and 20 to 40 millimeters in July (Figures 40 a & b).

Figure 40a

Figure 40 (a). Contours of the amount of precipitation in January in millimeters.

Figure 40b

Figure 40 (b). Contours of the amount of precipitation in July in millimeters.

The greatest amount of precipitation is observed in the Barents and Kara Seas in the west and in the Bering and Chukchi seas in the east. The duration of precipitation in the Arctic is quite large. On Dixon Island, for example it is 2869 hours (100 to 250 hours per month). The maximum duration is recorded in October together with a second maximum in April. The minimum duration of precipitation is observed in summer (July to August) and occurs primarily as rain. In the western and eastern parts of the Arctic the duration is longer than in the central area. On Vize Island, the maximum duration of precipitation is greater than 4000 hours per year. However, over the entire year the daily amount of precipitation per day is observed to be only about 0.1 millimeters per day in 80 to 95 percent of cases.

Calculations show that on average, the precipitation intensity in winter is 0.1 millimeters per hour and in summer 0.3 millimeters per hour (Table 38). The minimum precipitation intensity occurs in April. During this month the weather is clear and cold, and for the most part the precipitation is in the form of ice needles and includes very little moisture.

Table 38
Average precipitation intensity, millimeters per hour
Station Month
  Jan. Apr. Jul. Oct.
Amderma 0.12 0.10 0.32 0.15
Belyi Island 0.06 0.06 0.20 0.10
Dixon Island 0.08 0.06 0.26 0.11
Uedineniya Island 0.06 0.06 0.19 0.09
Cape Chelyuskin 0.06 0.06 0.22 0.10
Tiksi 0.10 0.09 0.31 0.14
Chetyryokhstolbovoy Is. 0.08 0.05 0.24 0.09
Wrangel Island 0.09 0.08 0.28 0.10
Uelen 0.12 0.08 0.34 0.16
Average 0.09 0.07 0.28 0.12

In practice, three classification criteria are used to characterize the intensity of precipitation: weak, moderate, and heavy. These criteria are somewhat subjective and are characterized by the decrease in horizontal visibility as defined by visual observations. For weak precipitation, the visibility is as much as 8 to 10 kilometers, for moderate precipitation it is from several hundred meters up to 2 kilometers, and for heavy precipitation it is less than 100 meters. These precipitation types are recorded by the observer in the station's observation log. Precipitation intensity data (millimeters per hour) for these three criteria are presented in Table 39. They have been calculated using observational data from the arctic stations for the 10-year period 1966 to 1975. The intensity of heavy precipitation is more than an order of magnitude greater than that of weak precipitation, however, weak precipitation is observed more than 50 percent of the time. This is because of the particular processes involved in the development of precipitation in the Arctic.

Table 39
Average actual precipitation intensity in the Arctic according to type based on visual identification
Precipitation type Precipitation intensity, millimeters
weak moderate heavy
Solid 0.03 0.21 0.65
Mixed 0.08 0.35 1.25
Liquid 0.27 0.90 3.00

6.3. Daily maxima of liquid precipitation

According to Table 2, rain with an intensity of 30 millimeters in 12 hours is classified as HMP. Experience with the observations shows that if 30 millimeters of precipitation does not occur within 12 hours, then a total of 30 millimeters per day is also not observed. Since the daily total is one of that specified parameters of precipitation that is estimated directly at the stations, we considered it necessary to analyze the variability in maximum values of daily precipitation totals. According to this analysis, a daily precipitation maximum of 30 millimeters or more was observed at only 30 stations among all those operating in the Arctic. Values of the daily precipitation maximum at selected stations for the entire operation period from 1935 to 1990 are presented in Table 40.

Table 40
Daily maxima of liquid precipitation with values of 30 millimeters per day or more
Station Year Month Maximum, millimeters Monthly totals, millimeters
Malye Karmakuly 1974 July 30.4 103.6
  1975 July 40.3 69.8
Cape Zhelaniya 1964 August 42.5 82.0
  1969 July 59.2 85.1
  1979 August 32.1 70.5
Amderma 1969 July 47.4 87.1
  1982 July 33.3 81.8
Vilíkitskyi Island 1982 June 33.9 58.5
Gydo-Yamo 1988 August 41.5 85.1
Pravda Island 1979 July 36.0 55.0
Pronchishcheva Bay 1971 July 50.1 83.3
Preobrazheniya Island 1958 August 45.1 99.3
Khatanga 1981 July 32.7 49.1
Kotelínyi Island 1980 July 39.0 65.0
Cape Sannikov 1958 August 46.0 103.6
Yubileinaya 1968 July 49.6 104.3
Cape Cherskyi 1954 July 34.4 107.2
Ayon Island 1979 July 30.3 65.6
Cape Billings 1956 July 32.9 89.8
Rau-Chua 1954 July 35.8 67.5
Uelen 1938 August 48.8 150.5
  1942 September 36.3 124.5
  1955 August 46.1 99.6
  1957 August 33.0 77.6
  1961 July 30.1 60.6
  1966 July 31.6 70.9
Provideniya Bay 1951 July 40.9 103.5
  1952 June 36.4 49.1
  1954 September 54.5 224.0
  1955 September 43.1 149.1
  1960 August 67.9 117.5
  1962 July 39.8 106.4

Analysis of the precipitation data showed that precipitation totals from different time periods and especially extreme values are characterized by significant interannual variability. Thus, in addition to the standard statistical estimators applied to observational data sets (average, dispersion etc.), sometimes some additional parameters are used. On the basis of an empirical cumulative distribution, the daily maximum precipitation totals were calculated for different probability levels. The calculated results for stations with the longest continuous observation sets are presented in Table 41. For example, for a daily precipitation maximum of 25 millimeters, when the probability is equal to 5 percent, 25 millimeters of precipitation or more per day occurs on the average one time in 20 years. In the last two columns of this table, the actual observed maximum value is given along with the date when it was recorded.

The maximum daily precipitation amount at a probability level of 1 percent (one time in 100 years) reaches 67 millimeters at Cape Schmidt and 61 millimeters at Cape Zhelaniya. This is a sufficiently large value, as long as this maximum in subarctic regions does not exceed 80 to 85 millimeters.

Table 41
Maximum daily amount of liquid precipitation (millimeters) for various calculated duration periods
Station Cumulative probability, % Observed maximum, millimeters Date of Maximum
65 20 5 1
Rudolph Island 7 12 17 23 23 6 Sep 1932
Cape Zhelaniya 14 25 42 61 59 18 Jul 1969
Maliye Karmakuly 14 23 32 43 40 15 Jul 1975
Amderma 15 25 35 48 47 20 Jul 1969
Belyi Island 12 20 25 32 31 2 Aug 1940
Novyi Port 19 29 38 44 40 12 Aug 1942
Uedineniya Island 9 17 26 38 35 28 Jun 1956
Russkiy Island 9 17 27 40 38 5 Aug 1962
Lake Taimyr 15 25 37 48 47 8 Aug 1965
Dixon Island 14 22 28 32 31 23 Jun 1961
Khatanga 16 28 45 60 58 9 Jul 1960
Preobrazheniya Is. 12 23 39 51 48 14 Jul 1938
Cape Shalaurov 15 24 34 45 42 18 Aug 1952
Muostakh Island 13 22 31 39 34 5 Aug 1953
Kusyur 12 20 29 43 42 18 Aug 1952
Chokurdakh 10 18 27 38 37 6 Jul 1962
Wrangel Island 9 17 24 33 30 30 Jul 1979
Cape Schmidt 16 31 47 67 66 8 Sep 1956
Uelen 19 30 42 55 49 22 Aug 1938

To estimate the long-term variability of the maximum values of daily precipitation totals, the precipitation observations were analyzed from the coastal stations of the Kara, Laptev, East Siberian and Chukchi Seas during the summertime (July-August). The long-term variations of these totals for the western, central and eastern sectors of the Russian Arctic are presented in Figure 41. The values shown have been smoothed using a 5-year moving average. The maximum variability of this value was observed in the western sector, where Atlantic cyclones exert a significant influence.

Figure 41

Figure 41. Long-term variations of the maximum daily precipitation values in millimeters for (1) the western, (2) central, and (3) eastern Arctic sectors.

6.4. General data on heavy rain and snowfall

Observations show that the basic contribution to the maximum daily amount of precipitation is due to individual events, that is, it is strongly episodic. If 30 millimeters of precipitation falls in 12 hours, then the daily total is seldom greater than 30 millimeters.

Hazardous rainfall (more than 30 millimeters of precipitation per 12 hour interval) in the Russian Arctic is observed for the most part in July and August, but individual events are also observed in June and September. Over a 20-year period, 68 cases of hazardous rain storms were observed at 36 stations. Selected data are presented in Table 42. At the remaining stations (about 60 percent of the total number) heavy rains were not observed. In July and August heavy rains were recorded 30 times over the entire observation period. During this time the average amount of precipitation was 40 millimeters and 33 millimeters respectively for these two months. In June and September there were only 5 episodes of heavy rain over the total observation period, with average precipitation values of 38 and 44 millimeters respectively. No definite pattern has been observed in the interannual repeatability of heavy rainfall. This confirms that the occurrence of heavy rains in the Arctic is irregular. The distribution in the number of days with heavy rains shows that along the coasts of the Arctic seas, heavy rainfall was observed once every 20 years, and it was never observed further to the north. Heavy rains occurred more often at the Lake Taimyr station (11 cases in 20 years) and the Provideniya Bay station (10 cases). At the majority of stations there are 2 to 5 such cases (Figure 42a).

Table 42
Heavy rains (30 millimeters or more in 12 hours)
Station Year June July August September
Cape Zhelaniya 1969   18/ 50.6    
Maliye Karmakuly 1975   15/ 40.3    
Amderma 1969     20/ 30.7  
Tambey 1978 28/ 35.2      
Cape Kamennyi 1965   21/ 42.5    
  1979     3/ 43.1  
Gydo-Yamo 1988     3/ 36.2  
Taimyr lake 1965     8/ 40.1  
Preobrazheniya Is. 1958     4/ 40.0  
Khatanga 1934   28/ 35.1    
Ust-Olenek 1941     2/ 32.1  
Tyumeti 1976     5/ 42.2  
Kusyur 1970   2/ 40.7 2/ 40.7  
Kotelínyi Island 1980   21/ 34.1    
Tiksi 1942   25/ 40.2    
  1963   16/ 42.5    
  1976   31/ 36.5    
Muostakh Island 1953   22/ 30.2    
Cape Sannikov 1958     3/ 40.0  
Yubileynaya 1968   10/ 42.0    
Cape Shalaurova 1962   19/ 32.2    
Chokurdakh 1950     5/ 33.1  
Cape Schmidt 1956       3/ 52.5
  1973     28/ 35.0  
Uelen 1938     5/ 41.0  
  1955     10/ 40.1  
Provideniya Bay 1951   18/ 35.0    
  1955       6/ 37.5
  1960     10/ 53.4  
  1962   22/ 35.0    
  1965     8/ 34.1  
  1975   25/ 52.0    
  1979       2/ 36.2

Note: The day is given in the numerator, and the precipitation amount in millimeters is given in the denominator.

The distribution of intensity of heavy rain (Fig 38b) shows that rains with a total precipitation of 30 millimeters within 12 hours are observed along most but not all of the coast of the arctic seas. In the Kara Sea, such precipitation is observed only along the southwestern coast. On Novaya Zemlya, rains with precipitation of 40 to 50 millimeters in 12 hours are observed with a maximum on the northwestern coast. In the Laptev Sea 30 millimeters of precipitation occurs on the eastern coast of the Taimyr Peninsula, on the coast of the Laptev Sea, and on the New Siberian Islands. Heavy rains were not observed on the coast of the East Siberian Sea during the operation period of the stations there. In the Chukchi Sea, heavy rains with precipitation amounts of up to 40 millimeters in 12 hours occur at some stations. To the south of the arctic sea coast, rainfall of 40 and 50 millimeters in 12 hours is observed.

Figure 42a

Figure 42b

Figure 42. (a) Number of days with heavy rainfall of 30 millimeters or more in 12 hours; and (b) the intensity of heavy rain in summer (millimeters per 12 hours).

All the heavy rainfall is caused by cyclone activity. For example, at the Cape Zhelaniya station at the northernmost end of Novaya Zemlya, a rain storm was observed on 18 July 1969 with 50.8 millimeters of precipitation in 12 hours. This was connected with a series of cyclones traveling from the south of Western Siberia to the Kara Sea. These cyclones transported a large amount of moisture. Over the Kara Sea each cyclone filled up as it passed eastwards to the Laptev Sea under the influence of anticyclone located over the Barents Sea. The cyclones were not deep (993 mbar) and included secondary warm fronts whose moisture together with cold air from the Arctic Ocean maintained the influx of water vapor-laden air. A large amount of precipitation resulted and its duration was considerable. The rain continued for 12 hours with a mean intensity of 4.2 millimeters per hour. The precipitation also extended over the region from Vize Island and Uedineniya to Russkaya Gavan', however, the amount of precipitation at these stations was not as much as that at Cape Zhelaniya.

On 15 July 1975 at Maliye Karmakuly, an extremely hazardous amount of precipitation (40.3 millimeters) was deposited. It was caused by the following circumstances. A small cyclone formed over the Kola Peninsula on 13 July, passed to the southeast of the Barents Sea and then stopped, continuing to deepen until July 15. At that time a warm front passed over the southern part of Novaya Zemlya, causing a cloudburst and steady precipitation. Moisture was advected in from the heated northern part of Russia. Precipitation caused by this cyclone and its fronts extended from the north of the Ural mountains to the middle of Novaya Zemlya. The rain began along the coast of Baidaratskaya Bay, and shortly afterwards the maximum precipitation was observed on Novaya Zemlya. Forty millimeters of precipitation also fell in Malye Karmakuly within 12 hours. On that same day, the cyclone became stationary and filled up.

Heavy rains in the central and eastern Arctic regions were caused by similar synoptic situations, when cyclones laden with a great deal of moisture were advected from the heated continent. All cases of heavy rainfall were connected with warm fronts. However, many cases occur where the accumulation of moisture is due to local conditions and the precipitation falls at that location. Heavy rains were not observed as a result of cyclones coming from the Atlantic and Pacific oceans because in the summer these cyclones pass over the relatively cooler water surface and have only a small moisture content compared with cyclones coming from the continent.

There is a relationship between the daily precipitation maximum and heavy rains that can be used for practical calculations of hazardous levels of precipitation. The amount of highly dangerous precipitation (more than 30 millimeters in 12 hours) is linearly related to the daily precipitation maximum (Figure 43).

Figure 43

Figure 43. The dependence between the amount of precipitation for heavy rain and the daily precipitation maximum.

The following formula is obtained for calculating the amount of hazardous precipitation from observed daily maxima at the Arctic stations:

q12 = 0.76 q24 + 3.1

where q12 is amount of precipitation for 12 hours in millimeters, and q24 is the daily maximum. The associated correlation coefficient is equal to 0.86.

Heavy rains were observed only at some stations. There were no cases when precipitation fell on an area large enough to extend over even two stations simultaneously. It is difficult to specify the long-term variability in the frequency of heavy rains (30 millimeters and more for 12 hours), because these rains occur so rarely.

Snowfalls in the Arctic are always connected with synoptic cyclonic situations, except for weak snowfalls or ice needles (diamond dust), which occur during anticyclone activity. Snow falls during the passage of warm fronts (continuous snowfalls) and during cold fronts (often as showers of short duration). Heavy snowfalls of the kind observed in midlatitudes that stop the vehicular traffic of cars and railway transportation are not observed in the Arctic. Even wet snowfalls do not satisfy the criterion for HMPs.

Maximum intensity of falling snow in the Arctic is 1.6 millimeters/h [5]. Assuming a duration of 12 hours at this intensity, the snowfall amount would be 19 millimeters, which is less than the established criteria for an HMP.

Snowfall in the Arctic also occurs in the summer during the period that birds nest and raise their chicks on the tundra and among the coast cliffs. Summer snowfall amount also does not reach the HMP level of 20 millimeters in 12 hours. For example, on 9 July 1972 a cyclone from the North Atlantic with its center over the Barents Sea caused a snowfall on the Franz Josef Land which deposited only 12.5 millimeters in 12 hours. The same cyclonic situation was observed over the Kara Sea causing a snowfall whose maximum at one of the stations was 14.7 millimeters in 12 hours, the air temperature was -5 degrees Celsius, and it was recorded as a snowstorm.

Heavy snowfalls in the Arctic are observed in combination with snowstorms, and under these circumstances falling snow does constitute a hazardous phenomenon. However, the storms with blowing snow accompanied by snowfall do qualify as HMPs.