1. Initial data and defining criteria for meteorological hazards

The basic data set used to study the climatology of hazardous meteorological phenomena (HMP) in the Russian Arctic Observation was obtained from 80 polar land and island stations over a 20-year period (1966-1985), and from 30 NP drifting stations for their entire duration of operation through 1991. A list of the polar stations is given in Table 1 and their locations are shown in Figure 1.

figure1

Figure 1: The network of polar observing stations. The numbers next to each point refer to the station numbers given in Table 1. The solid line denotes the boundary of the Arctic.

To estimate the frequency of hazardous meteorological phenomena in adjacent areas of the non-Russian Arctic, we have included observational data available from arctic stations in Canada, USA, Norway and Greenland ([32] Climate of the Polar Regions, 1973; [34] Climatic manual of the Northern America, 1985; [42] Meteorological Regimes of the foreign Arctic, 1971; [55] Putnins P., [71] 1973; Climatic Atlas Canada, 1984; [72]. Snowstorms along the northeastern coast of the United States, 1990). In addition, information published in meteorological monthly magazines, climatic manuals, atlases and other applied scientific literature (1, 29, 33, 43, 44, 46, 60, 61) have been used to describe dangerous meteorological phenomena and their influence on basic human activities under arctic conditions. Weather charts and barometric topography have also been included in the analysis of the synoptic characteristics of HMP.

Since 1966, when the network of hydrometeorological stations converted to eight observations daily and the recording procedures for atmospheric phenomena were improved, criteria for classifying meteorological phenomena as hazardous have been upgraded and refined several times.

According to the 1966 publication, "Statement about the collection of data on dangerous hydrometeorological phenomena"[52], meteorological phenomena were classified as HMP, when they produced or had the capability to produce even a small amount of damage. As a result of this broadened interpretation of HMP, less dangerous phenomena were classified as highly dangerous, and the number of HMP per year showed an apparent increase.

In 1972 an improved statement was issued [53], and the list of weather phenomena classified as HMP was significantly decreased. The criteria for classifying various phenomena was made more specific. In this protocol, considerably greater precision was introduced meaning that the threshold for HMP classification for wind, snow storms, precipitation, and a range of other weather phenomena was restricted to circumstances when one or more HMPs are observed over at least a third of a district or in a heavily populated area.

In 1986, a new document was approved. This was the "Statement of the procedure for preparation and broadcasting of warnings about highly dangerous natural hydrometeorological and helio-geophysical phenomena and environmental pollution hazards" [51]. According to this manual, hazardous natural hydrometeorological phenomena are "those meteorological, hydrological and marine hydrometeorological phenomena" whose intensity, areal coverage, and duration may bring about or have already brought about economic damage of national importance, harm to the population, and/or caused a natural disaster." In this document, the designation of HMPs has been extended to include "natural" phenomena.

Table 1
The primary set of meteorological stations in the Russian Arctic used for determining weather hazard phenomena.
Station Coordinates and Description of Location

#

Designation
North
Latitude
East
Longitude

Sea or Strait
Archipelago, Island, Peninsula, or Bay
1 Nagurskaya 80° 49' 47° 38' Northern Barents Sea Alexander Land
Franz-Joseph Land
2 Rudolph Island 84° 48' 57° 58' Northern Barents Sea Franz-Joseph Land
3 E.T.Krenkel Observatory,
Kheisa Island
80° 37' 58° 03' Northern Barents Sea Kheisa Island Franz Joseph Land
4 Little Karmakuli 72° 23' 52° 44' Eastern Barents Sea Novaya Zemlya, Southern Island
5 Russkaya Gavan' 76° 11' 63° 34' Eastern Barents Sea Novaya Zemlya, Northern Island
6 Cape Zhelaniya 76° 57' 63° 34' Eastern Barents Sea Novaya Zemlya, Northern Island
7 Amdyerma 69° 46' 61° 41' Southwestern Kara Sea Yugorskiy Peninsula
8 Ust'-Kara 69° 15' 64° 31' Southwestern Kara Sea Kara Gulf
9 Kharasavey 71° 08' 66° 49' Southwestern Kara Sea Yamal Peninsula
10 Marresalya 69° 43' 66° 49' Southwestern Kara Sea Yamal Peninsula
11 Ushakov Island 80° 49' 79° 33' Northeastern Kara Sea Ushakov Island
12 Vize Island 79° 30' 76° 59' Northeastern Kara Sea Vize Island
13 Popov 73° 20' 70° 02' Southwestern Kara Sea White Island
14 Tambey 73° 18' 73° 03' Gulf of the Ob Yamal Peninsula
15 Se-Yakha 70° 15' 72° 20' Gulf of the Ob Yamal Peninsula
16 Tadibeyakha 70° 22' 77° 08' Gulf of the Ob Gidanskiy Peninsula
17 Cape Kamenniy 68° 28' 73° 36' Gulf of the Ob Yamal Peninsula
18 Yar-Salye 66° 45' 70° 38' Gulf of the Ob Yamal Peninsula
19 Noviy Port 67° 42' 72° 52' Gulf of the Ob Yamal Peninsula
20 Gidoyamo 70° 55' 78° 31'   Gidanskiy Peninsula
21 Antipayuta 69° 02' 76° 24' Gulf of Tazovskiy Gidanskiy Peninsula
22 Cape Leskina 72° 21' 79° 33' Southwestern Kara Sea Yenisei Bay
23 Uyedineniye Island 77° 30' 82° 14' Northeastern Kara Sea Uyedineniye Island
24 Isvestiy TsIK Islands 75° 32' 83° 05' Northeastern Kara Sea Troynoy Island, Isvestiy TsIK Islands
25 Dickson Island 73° 30' 80° 14' Northeastern Kara Sea Dickson Island
26 Sopochnaya Karga 71° 52' 82° 42' Northeastern Kara Sea Yenisei Bay
27 Cape Sterlegova 75° 25' 88° 54' Northeastern Kara Sea Taymir Peninslua
28 Isachenko Island 77° 09' 89° 12' Northeastern Kara Sea Isachenko Island
29 Yst'-Tareya 73° 12' 89° 43' Inland Station Taymir Peninsula
30 Golomyanniy Island 79° 33' 90° 37' Northeastern Kara Sea Golomyanniy Island, Sedov Archipelago
31 Cape Peschaniy 79° 30' 103° 10' Laptev Sea Bol'shevik Island, Severnaya Zemlya Archipelago
32 Pravdy Island 76° 16' 94° 12' Northeastern Kara Sea Pravda Island
33 Russkiy Island 77° 10' 96° 26' Northeastern Kara Sea Russkiy Island, Nordenskold Archipelago
34 Krasnoflotskie Islands 78° 38' 98° 43' Northeastern Kara Sea Krasnoflotskie Islands
35 Heiberg Island 77° 36' 101° 31' Northeastern Kara Sea Heiberg Island
36 Solnechniy Bay 78° 12' 103° 16' Vil'kitskiy Strait Bol'shevik Island, Severnaya Zemlya Archipelago
37 E.K.Fyodorov Observatory,
Cape Chelyuskin
77° 43' 104° 17' Vil'kitskiy Strait Taymir Peninsula, Cape Chelyuskin
38 Lake Taymir 74° 37' 101° 25' Inland Station Taymir Peninsula
39 Khatanga 71° 59' 102° 28'   Khatanga, Dolina river
40 Little Taymir Island 78° 05' 106° 49' Laptev Sea Little Taymir Island
41 Andrei Island 76° 49' 111° 27' Laptev Sea Andrei Island
42 Pronchishchevoy Bay 75° 32' 113° 26' Laptev Sea Maria Pronchishchevoy Bay, Taymir Peninsula
43 Cape Kosistiy 73° 40' 109° 44' Laptev Sea Gulf of Khatanga
44 Preobrazheniya Island 74° 40' 112° 56' Laptev Sea Preobrazheniya Island
45 Cape Terpyay-Tumsa 73° 32' 118° 40' Laptev Sea Terpyay-Tumsa Peninsula
46 Tyumeti 72° 05' 123° 25' Inland Station Olenek River Valley
47 Dunae 73° 56' 124° 30' Laptev Sea Dunae Island
48 Kyusyur 70° 41' 127° 24' Inland Station Lower Lena River
49 Tiksi Bay 71° 35' 128° 55' Laptev Sea Sago Gulf
50 Muostakh Island 71° 33' 130° 01' Laptev Sea Buor-Khaya Bay
51 Kotel'niy Island 76° 00' 137° 54' Laptev Sea Kotel'niy Island, New Siberian Islands
52 Sannikov Strait 74° 40' 138° 54' Sannikov Strait New Siberian Islands
53 Stolb Island 72° 24' 126° 21' Laptev Sea Stolbovoy Island
54 Yubeleinaya, Kazachye 70° 45' 136° 13'   Lower Yani River
55 Cape Kigilyakh 73° 29' 139° 53' Laptev Sea Bol'shoy Lyakhovskiy Island
New Siberian Islands
56 Cape Svyatoy Nos 72° 50' 140° 44v Dmitri Laptev Strait  
57 Bunge Land 74° 53' 142° 07' Dmitri Laptev Strait Kotel'niy Island, New Siberian Islands
58 Cape Shalaurov 73° 11' 143° 56' Dmitri Laptev Strait Bol'shoy Lyakhovskiy Island
New Siberian Islands
59 Zhokhov Island 76° 09' 152° 50' East Siberian Sea De Long Islands
60 Chokurdakh 70° 37' 147° 53' Inland Station Lower Indigirka River
61 Indigirskaya 71° 15' 150° 18'   Lower Indigirka River
62 Alazeya 70° 46' 153° 44' Inland Station Bank of the Alazeya River, Kolyma Lowlands
63 Chetiryokh-stolbovoy Island 70° 38' 162° 24' East Siberian Sea Medvezh'i Islands
64 Cherskiy 68° 48' 161° 17' Inland Station Lower Kolyma River
65 Ambarchik Bay 69° 34' 162° 18' East Siberian Sea Ambarchik Bay
66 Rau-Chua 69° 31' 166° 35' East Siberian Sea Mouth of Rau-Chua River
67 Ayon Island 69° 55' 167° 59' East Siberian Sea Ayon Island
68 Cape Val'karkay 70° 05' 170° 56' East Siberian Sea  
69 Pevek 69° 45' 170° 36' East Siberian Sea Chaunskiy Gulf
70 Chaun 68° 20' 170° 16' East Siberian Sea Chaunskiy Gulf
71 Krasnoarmeyskiy 68° 10' 172° 25' Inland Station Chukotka
72 Cape Billings 69° 53' 175° 40' East Siberian Sea Chukchi Coast
73 Wrangel Island 70° 58' 181° 31' Chukchi Sea Rodgers Bay, Wrangel Island
74 Cape Schmidt 68° 55' 180° 31' Long Strait Chukchi Coast
75 Vankarem 67° 50' 184° 10' Chukchi Sea Cape Vankarem, Chukchi Coast
76 Kolyuchin Island 67° 29' 185° 22' Chukchi Sea Kolyuchin Island
77 Cape Netten 66° 57' 188° 04' Chukchi Sea Chukchi Coast
78 Uelen 66° 10' 190° 10' Bering Strait Chukchi Peninsula
79 Cape Chaplin 66° 24' 187° 45' Bering Sea Chukchi Peninsula
80 Provideniya Bay 64° 26' 186° 46' Bering Sea Provideniya Bay
Note: The numbers in the first column correspond to the station numbers shown in Figure 1.

It follows that a phenomenon is not dangerous per se; rather its consequences must be sufficiently severe. In singling out particular meteorological phenomena, we emphasize a strong negative influence on basic human activities and economic endeavors. If the interaction of HMPs with people is not considered, the entire range exhibited by meteorological phenomena is neither threatening nor highly dangerous; it is simply "natural." Previous classifications of such phenomena as dangerous or hazardous have been defined in subjective terms relative to the interaction between people and nature. The number of dangerous or hazardous phenomena is decreased to a large extent when new terminology is used.

There is, however, an inaccuracy in the previous assumption: a phenomenon itself is not dangerous or highly dangerous, but rather what is hazardous is the concrete effect as defined by objective criteria. That is why it is necessary to be specific about the extent of the hazard posed by a meteorological parameter or phenomenon.

There are several uncertainties present in the 1986 Manual for HMPs [51]. The following are not specified: the onset time of the phenomena, whether they occur out of season, how suddenly they arise, and whether they are atypical in nature. Damage to personnel has a lower priority than damage to the national economy; no priority is assigned to personal health and well being. Damage to wildlife and the natural environment is not considered. Undeveloped and uninhabited regions are not considered in detail. For example in many parts of the Arctic there is no damage now but meteorological phenomena reach dangerous levels, and could cause damage if the areas were inhabited.

There are also certain disparities between different criteria presented in available textbooks and instruction manuals. For example, early ice formation (1 time within 10 years) is classified as a hazardous hydrological phenomenon. The corresponding air temperature decrease to large negative values in autumn, which causes the ice formation, is not classified as dangerous. There is no exact classification of HMP, for example, by frequency or probability of occurrence (for example number of occurrences per year).

Anthropogenic influences have caused similar unfavorable phenomena that could not be classified as natural, but that nevertheless impact the environment. For example, industrial exhaust into the atmosphere in combination with normal meteorological conditions may increase the degree of contamination to levels that are hazardous to personal health or even lethal. This type of phenomenon cannot be classified as "natural," but may certainly be classified as highly dangerous.

According to the current classification (HMP, 1986 [51]), there are four criteria by which meteorological phenomena are classified as natural hazards: the influencing meteorological parameter or atmospheric phenomenon, the level (force, intensity), the duration, and the spatial extent. We describe these in order.

A list of meteorological elements and phenomena classified as HMP may be considered to be essentially complete for the Russian Arctic, however it is not exhaustive, and it doesn't take into account the specific characteristics of some areas. It is clear that the approach to defining the list of meteorological phenomena and especially quantitative criteria for HMP classification must be more regionally detailed, differentiating among influences to personal health, different branches of the economy, and large industrial processes. The interests of personal wellbeing must take priority over economic concerns.

The list of HMPs must be expanded to take ecological concerns into account. It is necessary to consider threatening conditions and the loss of people's ability to work caused by unfavorable weather phenomena, for example, sudden atmospheric pressure changes, which are characteristic of many areas of the Russian Arctic. At the present time, this parameter and its change are not included in the HMP list. Another example is as follows: the common meteorological arctic conditions of inversion and dead calm are not dangerous for uninhabited or sparsely inhabited regions, but in conjunction with industrial exhaust they can cause high contamination concentration in the near surface layers that hazardous to people's health and to the environment. It turns out that a particular atmospheric phenomenon can be considered hazardous when it is strong, as well as when it is weak under certain circumstances. Wind, for example, is considered as dangerous when its speed is high enough that buildings and watercraft may be damaged, but it also presents a hazard in cities when its speed is very low and there is a buildup of contamination due to industrial and vehicular activity. The combination of an inversion, no wind, and fog with industrial exhaust results in the HMP "smog." For a specific urban or suburban area it is also necessary to take into account the wind direction.

Heavy icing on the roads can increase the vehicular accident rates and cause a variety of associated problems. Heavy glaze must therefore be included in the list of HMPs.

With respect to the influence on wildlife [67], deep snow cover and ice crust may be classified as arctic HMPs because, in many cases, they severely restrict the movements and feeding of caribou, arctic foxes and other animals. Dense snow cover with ice crust layers also prevents partridges from finding cover during cold snaps.

Ambiguities in the definitions of HMPs are illustrated by the following examples. For the arctic coastal regions, wind speeds exceeding 30 meters per second (m/s) are defined as highly dangerous [51]. There are parts of the Arctic where such wind speeds are never observed, but there are regions, where 30 meters per second wind speeds are observed many times each year, thus constituting a normal and integral part of meteorological regime. The question arises as to whether a 30 meters per second wind is in fact hazardous. If damage will result because the high wind speed is not usual for a certain region, and the experience to prevent the damage is not available, the situation satisfies the criterion for an HMP. However, a 30 meters per second wind speed probably would not be considered an HMP in areas where high wind speeds are common.

To define the wind speed that qualifies as an HMP, the following circumstances have to be taken into account. Wind loads in arctic regions are basically the same as those in temperate zones when the wind speed is similar, however, the associated heat fluxes differ significantly. The impact of heavy wind differs significantly when air temperature is positive or slightly negative compared to very cold conditions when the air temperature drops to minus 30 or minus 40 degrees Celsius where conditions would be considered hazardous in any case. Taking into account only the dynamic effects of wind speed for the arctic and sub-arctic areas is not enough. At the moment, however, the experience needed to develop complex danger characteristics is limited.

According to the 1986 criteria [51], a phenomenon would have to extend over approximately 1/3 part of a region to be classified as an HMP. In our opinion, this criterion is mainly applicable in well-developed, densely populated regions. The criteria for HMPs include the the assumption that when the area over which an HMP is observed increases, the extent of the damage increases as well. In the Arctic, where the most of the population and materials are concentrated in separate settlements separated from each other by hundreds or thousands of kilometers, the relationship between the area affected and the amount of damage is more complex. If an HMP occurs in a populated area, the damage there can be extensive but in the surrounding uninhabited territory, as the area increases, the damage remains nearly constant. In addition, the sparsity of the arctic network of meteorological stations does not always make it possible to delineate the HMP region reliably. Since data analysis is performed after obtaining the data from stations, it is assumed that when phenomena that satisfy the requirements for HMP have occurred at a particular station, they have also occurred in the surrounding areas.

A variety of undesirable consequences may occur when a meteorological phenomenon shifts from a season where it is common to one where it is not. It is clear that a meteorological phenomenon that usually occurs in winter (it may be a climate indicator) can be very unusual if it occurs in the summer. As a result, such a meteorological phenomenon may become an HMP as a result of the time delay. For example, the summer snow storm with low temperatures, high wind speed winds, and icing of communication and power lines in Murmansk and other regions of the Kola peninsula on 21 and 22 June 1997 resulted in significant structural damage. The effects of temporal shifts and suddenness of onset are clearly more important than previously appreciated, both in determining the occurrences of HMPs and in defining the corresponding criteria.

In this work the enumeration of meteorological phenomena and their elements together with available quantitative classification criteria as HMP according to the classification system of 1986 [51] have been improved, taking into account the particular conditions specific to the Russian Arctic. The corresponding information on the observations of meteorological phenomena and their threshold values is presented in Table 2. The methods used for data processing are consistent with those universally accepted in climatology [24, 36, 48, 57, 59].

The first stage in data processing involved reviewing tables of daily meteorological observations, observation books, selection of terms, and days when meteorological conditions corresponded to established criteria (see Table 2). On the basis of these considerations, climatologies of HMPs were developed for each polar and drifting station over a 20-year period during which observations were carried out uniformly eight times per day. As a rule, for individual stations that had been closed, the climatologies were developed for shorter periods, but never less than 10 years. These intervals all fell within the specified 20-year period.

Further data processing was performed using physical and statistical analyses of time series [36, 48]. For each month when HMPs were observed, mean values and variances were obtained, extreme values were recorded, and temporal frequencies of the phenomena were calculated. On the basis of these calculations, the annual course of each element was determined, and the number of days with HMPs for each month and each year were evaluated, together with amplitudes and frequencies of interannual oscillations and their tendencies and trends. On the basis of the annual time series data, the correlation of certain HMPs were evaluated using spatial correlation analysis.

Contour maps of the number of days and the duration of HMPs per year (or season) were developed by selecting regions with different frequencies of specific meteorological phenomena occurrences. In a number of cases, the correlation between individual indices was studied, in particular, the relationship between the duration of an HMP and the number of days that the phenomenon occurred, the dependence of the HMP index and average magnitude of the meteorological parameter, etc. These relationships were used to develop maps of the spatial distribution of HMP duration in cases where data were sparse or lacking.

For various HMPs, maximum values of the meteorological elements were determined by a combination of graphical and analytical methods. To calculate probability distribution for the frequency of a meteorological element Chegodaev's formula [46] was used. For example, the volume of drifted snow (Q) is calculated from the empirical formula:

Q = C Va t,

where V is the wind speed in meters per second, t is the duration of the snowstorm in hours, and C and a are empirical coefficients (a is close to 3).

Corresponding values of V and t were taken from direct observational data at polar hydrometeorological stations. The empirical coefficients C and a were derived from experimental measurements with the "snow storm meter" in the Arctic (see section 4.5 and reference [5]). The magnitude of the error in these HMP estimates depends on length of the observation set and on the presence of significant variability in the HMPs. The largest errors most likely occur in the estimates of very infrequent events observed over a few individual months. Consequently, in order to develop the maps and the specific the regions affected, we have attempted to use HMP frequency values at particular stations selected as the most reliable over a full year. As new observational data are collected, the derived estimates of mean and extreme frequencies of occurrence and duration period of the HMPs need to be updated accordingly.

Table 2.
Criteria established for the classification of meteorological phenomena as HMP on the basis of data from the Russian Arctic stations
Meteorological parameter or phenomenon Threshold value
  Primary Supplementary
Wind speed 25 m/s 15 m/s, 30 m/s
Snowstorm Snow drifting
(a) wind speeds of 15-24 m/s for intervals of 9 hours or more, or
(b) wind speeds of- 25-29 m/s and higher for 6 hours or more
Snow drifting
When the wind speed is 30 m/s or higher, regardless of the duration
Low air temperature -30degreesC - 40degreesC, - 50degreesC
Snowfall 20 millimeters in 12 hours  
Rain 30 millimeters in 12 hours  
Fog Visibility of 100 m or less  
Icing 20 millimeters diameter or thickness 5 millimeters diameter or thickness
Complex deposition of ice 35 millimeters diameter or thickness