Atmospheric Transport Patterns
from the Kola Nuclear Reactors
Alexander Mahura
University of Alaska, Fairbanks, USA
University of Washington, Seattle, USA
Kola Science Center, Apatity, Russia
Robert Andres
University of Alaska, Fairbanks, USA
University of North Dakota, Grand Forks, USA
Daniel Jaffe
University of Alaska, Fairbanks, USA
University of Washington-Bothell, USA
Northern Studies Working Paper
No. 24:2001
cerum, Centre for Regional Science
se-90187 Umeå
Sweden
The Project “Nuclear Problems, Risk
Perceptions of, and Societal Responses to,
Nuclear Waste in the Barents Region”
- an Acknowledgement
Since the late eighties, CERUM has developed research with a focus on
the shaping of and development within the Barents Region. Two specific features have characterised this research. First of all our ambition
has been to develop research projects in close collaboration with international and especially Russian researchers. This has materialised as an
exchange of researchers at conferences both in Sweden and in Russia.
Secondly, our view has been that the Barents region must be analysed
by researchers that represent a broad set of competences. Especially our
ambition is to develop a deeper and more integrated collaboration
between researchers from social sciences and arts on one hand and natural sciences on the other.
With the Swedish Board for Civil Emergency Preparedness (ÖCB)
as the main finacier, CERUM has for a couple of years developed research within the project “Nuclear Problems, Risks Perceptions of, and
Social Responses to, Nuclear Waste in the Barents Region”.
This report is produced within the afore-mentioned project. The
project deals with vulnerability as a response to the latent security questions associated with the existence of nuclear power and nuclear waste
in the Barents region. Clearly there is within the project a large scoop
for an analysis with its roots in natural sciences of the size and dispersion of various types of waste from the region. The project also has produced a set of such papers. Those papers raise questions that immediately lead to other papers and a discussion with its roots in social sciences, of civil emergency preparedness in a broad and spatially delimited sense as well as a discussion of the need for an enlarged concept of
safety. The pattern of spatial risk dispersion, which in this case not halts
at the national borders, and the associated construction of governance
in the Barents region also imply that trans-border negotiation, conflict,
and cooperation become key words in the discourse.
Gösta Weissglas
Project leader
Lars Westin
Director of CERUM
2
List of Figures and Tables
Figure 1.2.1 Geographic relationship of KNPP
Figure 1.3.1 Position of the Arctic front
5
6
Figure 2.1.1
Figure 2.2.1
Figure 2.2.2
Figure 2.3.1
Geographical regions of interest and model grid domain
Examples of trajectories showing consistent air flow
Examples of complex trajectories
5-day isentropic forward trajectories originated over the KNR region during
1992
Figure 2.3.2 Cluster mean trajectories or transport pathways from KNR during 1992
8
9
1
11
Figure 3.1.1 Monthly variations in the average transport time (in days) from the KNR region
to geographical regions based on the forward trajectories during 1991-1995
Figure 3.1.2 Monthly variations in the average percentage of trajectories originated over the
KNR region at lower altitudes within the boundary layer and reached the
geographical regions during 1991-1995
Figure 3.2.1 Atmospheric transport pathways (cluster mean trajectories) from the KNR
region based on the forward trajectories during 1991-1995
Figure 3.3.1 Airflow probability field within the boundary layer for the trajectories,
originated over the KNR region during 1991-1995 (isolines are shown every
10%)
Figure 3.4.1 Precipitation factor probability field for the boundary layer based on the relative
humidity fields for the trajectories, originated over the KNR region during
1991-1995 (isolines are shown every 5%)
Figure 3.5.1 Fast transport probability field within the boundary layer for the trajectories,
originated over the KNR region during April of 1991-1995 (isolines are shown
every 10%)
14
Figure A1
Figure A2
Figure A3
Figure A4
Atmospheric transport pathways from KNR during spring
Atmospheric transport pathways from KNR during summer
Atmospheric transport pathways from KNR during fall
Atmospheric transport pathways from KNR during winter
Figure B1
Figure B2
Figure B3
Figure B4
Spring airflow probability field within the boundary layer
Summer airflow probability field within the boundary layer
Fall airflow probability field within the boundary layer
Winter airflow probability field within the boundary layer
Figure C1
Figure C2
Figure C3
Figure C4
Spring precipitation probability field for the surface-1.5 km asl layer
Summer precipitation probability field for the surface-1.5 km asl layer
Fall precipitation probability field for the surface-1.5 km asl layer
Winter precipitation probability field for the surface-1.5 km asl layer
Figure D1
Figure D2
Figure D3
Figure D4
Spring fast transport probability field within the boundary layer
Summer fast transport probability field within the boundary
Fall fast transport probability field within the boundary layer
Winter fast transport probability field within the boundary layer
Table 1.3.1
Table 3.1.1
Seasonal average meteorological variables (Meteorological station at Zasheek)
Transport summary from KNR region to geographical regions
15
3
EXECUTIVE SUMMARY
This report has a purpose to show a methodology and possibilities for future studies of the
possible impact to different geographical areas of the Barents Euro-Arctic Region from the
numerous nuclear risk objects on the Kola Peninsula.
The Kola Peninsula (Murmansk region, Russia), being the nearest neighbor to the Nordic
countries, plays a considerable role in the airborne pollution problem in the Barents Euro-Arctic
region. There are two main types of airborne pollution sources in this area. From one side, it is an
airborne pollution from largest Cu-Ni metallurgical companies of the Kola Peninsula. From another
side, there is a potential risk of radioactive pollution under hypothetical accidents at numerous
objects of the radiation risk in the region.
It is known that among the reactors of the Former Soviet Union and Eastern Europe, the 15
RBMK and 6 VVER-440/230 are reactors of the greatest concern. Beside that, the high number of
naval and commercial nuclear reactors on and along the Kola Peninsula in the Northwest Russia
exceeds by far the concentration in any other region of the world. There are more than 170 nuclear
reactors in operation, and about 140 reactors waiting to be decommissioned. Finally, the Russian
government plans additional reactor units, despite public opposition and limited financial resources.
The Kola nuclear power plant (KNPP) on the Kola Peninsula is the closest Russian nuclear
power plant to the Scandinavian countries. It is located just 100 km from the Finnish and 200 km
from the Norwegian border. Because this plant uses an older VVER-440/230 design reactors, has
problems with storage and disposal of existing radioactive wastes and it was built in an area of the
medium seismic activity, there are concerns about potential accidents at this facility and a
subsequent release of radionuclides.
For these reasons, we have examined atmospheric transport patterns, airflow, precipitation
factor, fast transport probabilities, and possible impact from the plant’s reactors based on 19911995 meteorological data sets on several geographical regions -- Scandinavian, European, Central
Former Soviet Union (FSU), and Taymyr -- using an isentropic trajectory model, cluster analysis,
and probability fields techniques.
We found, that highly populated regions, Scandinavian and Central FSU, are at the greatest
risk in comparison with the European region. Radionuclide transport to the Central FSU region can
occur in one day, but averages 2.7 days with 45% of the cases of the free troposphere transport. To
the Scandinavian region it can occur in 0.5 days, but averages 1.2 days with most of occurrences
(up to 70%) resulting from the transport in the boundary layer. Since rapid transport (1 day) could
bring relatively undiluted air parcels containing significant concentrations of radionuclides, these
events were considered in greater detail by analyzing probabilistic fields.
4
1. INTRODUCTION
1.1. Introduction Background
An accident at the Chernobyl nuclear power plant (51°17' N, 30°15' E), located in Ukraine
(30 km south of the Belorussian border), was the most severe in the nuclear industry. The emissions
continued from 26 April until 6 May 1986. This accident caused the rapid death of 31 plant
employees and brought the evacuation of 116,000 people within a few weeks. About half a million
workers and four million others have been exposed, to some extent, to radiation doses resulting
from the Chernobyl accident (Bouville, 1995). During the weeks after the initial explosion,
radioactive material was observed almost everywhere in the Northern Hemisphere due to longrange atmospheric transport.
The main sources of radioactive pollution in northern Russia are similar to those in many
western countries (Bridges and Bridges, 1995; Nuclear Wastes in the Arctic, 1995). The most
common source of anthropogenic radioactivity has been the global fallout from nuclear weapons
testing in the atmosphere (Aarkrog, 1994; Bocharov et al., 1995). The ecosystems may contain
relatively high levels of radioactive contamination. Beside global fallout they have received
significant amounts of radioactivity from a number of other sources. These sources were studied in
detail by Baklanov et al., 1993; Nilsen and Bohmer, 1994; Bergman et al., 1996; Nilsen et al.,
1996. Nuclear power plants (NPPs) are one of these sources and they have a rank of relatively high
radiological consequences in terms of doses to total population and influence on environment, if
substantial release occurs (Baklanov et al., 1996).
There are 29 operating nuclear power reactors, including 4 at the Kola Nuclear Power
Plant (KNPP), situated at 9 sites within Russia (Improving Safety of Soviet-Designed Nuclear
Power Plants, 1996). All sites, except Bilibino, are in the European part of the former Soviet Union
(FSU) and in areas with high population density. An accident at any of them could have an effect on
both far and near territories.
The KNPP is a potential radioactive risk site that could have impact on environment. There
are other nuclear power plants in Scandinavia, Russia and England located within 1000 km of the
Arctic Circle. Routine discharges from these plants are small. The main concern regarding NPPs is
the possibility of accidents. Although construction and operation of nuclear power plants are
monitored and regulated, accidents are possible. The possible ways of radionuclide releases are
accidents: at nuclear reactors, storage facilities for radioactive wastes and during transport of the
spent nuclear fuel and waste to or from plants (Shvyryaev et al., 1992, Baklanov et al., 1994;
Elokhin and Solovei, 1994). Human error or natural hazardous conditions can cause the releases.
The most immediate danger from an accident is exposure to high levels of radiation. There
may be radiation hazard in the surrounding areas both near and far, depending on the type of
accident, amounts of radiation released, and weather conditions. The influence on the environment
will vary with temporal and spatial conditions.
In this study the atmospheric transport patterns, probabilistic fields, possible impact to
different geographical regions -- Scandinavian, European, Central FSU and Taymyr -- are examined
from the Kola Nuclear Reactors (KNR). Atmospheric transport pathways are analyzed using 5-day
forward isentropic trajectories originated over the KNR region during 1991-1995. We assume that
obtained results can be used in the event of an accident to estimate the probability of the
radionuclide transport from KNR. It is very important to have a probabilistic assessment of a
possible long transport of pollutants, especially for further study of social and economical
consequences of the radioactive risk for the Euro-Arctic region.
1.2. Reactors of the Kola Nuclear Power Plant – Source of the Radiation Risk
The Kola nuclear power plant (67° 28' N, 32° 28' E; 131 m asl) is located 15 km
north-west of the urban settlement Polyarnye Zori on the shore of Iokostrovskaya Imandra lake. It is
5
the first NPP in the FSU to be built north of the Arctic Circle. The distance to the Finnish border is
about 100 km, to the Norwegian border is 220 km and to the Swedish border is 330 km. The
geographic relationship of the Kola NPP is shown in Figure 1.2.1.
Figure 1.2.1. Geographic relationship of KNPP
The main road from St.-Petersburg to Murmansk follows the Kola Peninsula and crosses
Imandra Lake close to the plant. A railway line passes a few kilometers south of the plant along the
southern part of the lake with a sidetrack connecting the plant and main line. The closest airport,
Khibiny, is located about 40 km north of the plant. The plant’s personnel live at Polyarnye Zori.
The town is situated 10 km south of Imandra lake on the Niva river. The population is more than
20000 inhabitants with nearly 3000 people attached to KNPP.
The operational parts of KNPP are situated on the cape of the narrow peninsula, which juts
into the Imandra lake. Imandra is the largest lake in the center of the Kola Peninsula with an
average depth around 13 m. The surrounding area is mountainous with ridges up to 1000 m above
sea level. The immediate area around the plant is marshy and the nearest hills reach heights of about
200 m. The Kola Peninsula belongs to a stable geological formation (Baltic Shield), but there is low
to medium probability for seismic activity (Bune et al., 1980).
The plant was built to supply power to a growing mining and metallurgical industry and to
support the increasing population due to buildup of activities in the Murmansk Region. The KNPP
has two older reactors of type VVER-440/230 and two reactors of type VVER-440/213. First two
units (VVER-440/230) were put into operation between 1973-1974, third and forth (VVER440/213) -- between 1981-1984. The electric capacity of each unit is 411 MW. The expected
service time is 30 years. The plant produces 1644 MW with all units on full load. It is connected by
four high-tension 330 kW lines to the network of the "Kolenergo" Power Company. Approximately
60% of the electric energy produced at the plant are supplied to local heavy industry.
The cooling water is discharged into the western part of Babinskaya Imandra lake through a
one-kilometer long channel. As a result, the vast water surface of Molochnaya bay does not freeze
throughout the winter. Due to the cool climate and correspondingly low temperature of lake water,
the plant has access to a particularly good supply of water. Parts of the cooling water are used for
heating greenhouses and were used for a fish farm. Radioactive wastes of low and medium level
6
radiation are stored by the plant (Nilsen and Bohmer, 1994). Used fuel assemblies are stored up to 3
years and then transported for reprocessing to South Ural (Mayak) by railway.
There are plans for the construction of three 640 MW NP-500 reactor units near Imandra
lake. The lake’s water supply is sufficient for additional energy units. Russian authorities have
announced plans to build two or three new NP-500s (a 640-megawatt VVER with enhanced safety
features). The first of the units is scheduled to begin operating in 2003, when the first two units of
the existing plant, which are VVER-440/230s, will be decommissioned (“Soviet-Designed Nuclear
Power Plants in Russia, Ukraine, Lithuania, the Czech Republic, the Slovak Republic, Hungary,
and Bulgaria”, 1996).
1.3. Conditions Affecting Atmospheric Transport on the Kola Peninsula
The complexity of the Arctic climate derives from the varying radiative effects of the
underlying surface, non-equal distribution of land and water, horizontal transport of the heat from
lower latitudes. These regions are synoptically active (Walsh and Chapman, 1990). The Aleutian
and Icelandic Lows and Siberian High have influence on the transport of air masses within the
Arctic regions. The Kola Peninsula lies on the boundary of the Arctic front (Krebs and Barry,
1970). Its mean position, as shown in Figure 1.3.1, varies as a function of season over the Kola
Peninsula: from Greenland to the middle territories of the Scandinavian Peninsula (in winter) and
from Spitsbergen Islands to the Barents Sea (in summer) (Shaw, 1988). The front divides two air
masses with different characteristics: cold and dense Arctic air and the warmer and moister
maritime air of the Northern Atlantic. On the synoptic time scales, the front in the Kola and
Scandinavian Peninsula sector may be located anywhere from the Arctic to the inland at any season.
Figure 1.3.1. Position of the Arctic front
For most of the Arctic territories, summer precipitation is up to three times the amount of
winter precipitation. Increased precipitation favors the washout of pollutants and levels of the
deposition. Atmospheric circulation over these regions is affected by strong Coriolis effects. During
winter, cold air temperatures, limited low-level moisture supply and high static stability characterize
these areas. Such conditions will limit the intensity of the vertical circulation. During summer-fall,
when the background stratification is not excessively strong, intensity of cyclone development over
7
the regions will depend critically on the conditions of the underlying surface (e.g. water surface or
land, open or covered by snow or ice). The main cyclone and anticyclone pathways are associated
with activity of the Icelandic Low throughout the year (Klein, 1957).
Although KNPP is located north of the Polar Circle, the climate is rather mild. Observations
for wind direction and velocity, humidity and temperature are conducted and recorded at the plant,
but not available. The nearest meteorological station is situated at Zasheek (67° 42' N, 32° 45' E;
151m asl), 7 km from the plant. Average monthly temperature is from -13.7° C (January) to 13.9° C
(July), humidity from 86% (January) to 72% (July). During winter south and southwest winds
prevail in the area, in summer time north and northeast winds. The extreme atmospheric phenomena
for the Kola Peninsula are snowstorms, thunderstorms, hails and fogs. Only one tornado was
observed in February 2, 1993 with a wind speed more than 40 m/s, which had impact on the normal
operation of the plant. A summary of meteorological variables by season is presented in Table
1.2.1.
Season vs.
Meteorological variable
Temperature, ° C
Humidity, %
Cloudiness, units of 10
total
low
Velocity of wind, m/s
Direction of wind, %
N
NE
E
SE
S
SW
W
NW
Winter
Spring
Summer
Fall
Year
-12.8
86
-2.0
75
+12.1
72
+0.3
85
-0.6
80
7.4
6.1
3.4
7.3
5.2
4.0
7.2
5.5
3.8
8.0
6.8
3.8
7.5
5.9
3.8
10
7
8
9
19
28
9
10
17
9
5
6
15
24
9
15
23
18
7
6
15
16
6
9
10
7
6
7
19
24
17
10
15
10
7
7
17
23
10
11
Table 1.3.1. Seasonal average meteorological variables (Meteorological station at Zasheek)
8
2. METHODOLOGY
2.1. Specification of the Impact Regions
As we mentioned in Chapter 1, the Kola Nuclear Power Plant (KNPP or Kola NPP) located
at 67.45°N and 32.45°E; and, in particular, its reactors of the older generation (PWR-440/230
design) are subject to possible severe accidents. Bergman et al., 1996 mentioned that KNPP is one
of the main potential radiation risk sites in the Russian European North. An accident at this plant,
following by the subsequent radionuclide’s releases, may have a significant impact on the
environment and population of the Barents region.
In this part of the study, we had examined the probabilistic atmospheric transport patterns
from the Kola Nuclear Reactors (KNR) region. These patterns, we assumed, may be useful in
estimation of the possible radiological impact on different geographical regions. For this study, we
selected (as shown in Figure 2.1.1) the following impact regions: Scandinavian, European, Central
Former Soviet Union (FSU) and Taymyr regions. Scandinavian region covers the Scandinavian
Peninsula and surrounding seas (60-72° N and 0-30° E). European region includes practically
territories of all European countries and attaches surrounding seas (35-55° N and 10° W-25° E).
Central FSU region is situated between 45-60° N and 25-70° E and covers the European part of
Russia and territories of the FSU republics -- Ukraine, Belorussia and north-western areas of
Kazakhstan. Taymyr region has boundaries between 60-80° N and 70-120° E and includes the
Taymyr Peninsula, the land south and seas north of the peninsula.
The boundaries of the impact regions were chosen based on the selection’s assumption of
the most populated geographical areas and various climatic regimes.
Figure 2.1.1. Geographical regions of interest and model grid domain
2.2. Isentropic Trajectory Calculation
Although atmospheric trajectory models use an assumption of adiabatically moving air
parcels and neglect various physical effects, they are a useful tool for evaluating common air flow
patterns within meteorological systems on various scales (Merrill et al., 1986; Harris & Kahl,
1990; Harris & Kahl, 1994; Jaffe et al., 1997a; Mahura et al., 1997a; Jaffe et al., 1997b; Mahura
et al., 1997b). Some uncertainties in these models are related to the interpolation of meteorological
data, applicability of the considered horizontal and vertical scales, and assumptions of vertical
transport (Merrill et al., 1985; Draxler, 1987; Kahl, 1996). In general, the computed atmospheric
9
trajectory represents the pathway of an air parcel motion in time and space and is considered as an
estimation of the mean motion of a defusing cloud of material.
In our study we interpolated the original National Center for Environmental Prediction
(NCEP) gridded wind fields from the database DS.082 (NCEP Global Tropo Analysis, daily, from
July 1976 to present) to potential temperature (isentropic) surfaces. Interpolation procedure has
been performed for the period of 1991-1995 and applying a technique described by Merrill et al.,
1985. Then, we used the wind fields on these surfaces to calculate trajectories in the model domain.
The model grid domain is located between 20°-80°N and 60°W-127.5°E as shown in Figure 2.1.1.
All forward isentropic trajectories for the KNR region we computed twice per day (at 00 and 12
UTC, Universal Coordinated Time) at different potential temperature levels. These levels ranged
from 255°K to 330°K with a step of 5°K. The National Center for Atmospheric Research (NCAR,
Boulder, Colorado) and Arctic Region Supercomputing Center (ARSC, Fairbanks, Alaska) CRAY
computer resources were used to compute more than 200 thousand trajectories. Less than one
percent of the trajectories were missing because of the absence of archived meteorological data and
processing problems. In this study, we used for every calculation four trajectories.
The initial points of trajectories are located on the corners of a 1° by 1° of latitude-longitude
box, where KNR is in the center of the box. Only cases, when the wind field was reasonably
consistent along the transport pathway (i.e. all four trajectories have shown a similar direction of
transport for one time period) were included in further analysis (as shown in Figure 2.2.1).
Trajectories, showing a strong divergence of flow, were assigned to the category of the complex
trajectories (as shown in Figure 2.2.2). Such trajectories reflect more uncertainties in the air parcel
transport and, therefore, we excluded them from our further analysis.
Figure 2.2.1. Examples of trajectories showing consistent air flow
As shown in Figure 2.2.1 (left), the air mass originated over the KNR region at February 3,
1991, 12 UTC. Beginning at 60° N it moves to west crossing the Karelia, Baltic Sea and the
northern territories of the Baltic states during next day. Air parcel reached the Great Britain Islands
on February 7, 1991. During all this time the air parcel traveled within the boundary layer.
10
Figure 2.2.2. Examples of complex trajectories
Baklanov et al., 1996 performed analysis of different scenarios for the hypothetical KNPP
accidents. Their study noted that the height of radionuclide releases and its duration may vary from
100 to 500 m and from several hours to several days. Therefore, from all trajectories we decided to
choose only less than 13 thousand trajectories (from original more than 200 thousand trajectories).
All chosen trajectories start in the range of these heights and have duration of 5 days or longer.
Additionally, to study altitudinal variations in flow patterns (in particular, within the boundary layer
and free troposphere), we also considered trajectories originated over the KNR region at 1.5 and 3
km asl.
2.3. Cluster Analysis Technique
Cluster analysis refers to a variety of multivariate statistical analysis techniques used to
explore the structure within data sets (Romesburg, 1984). The specific purpose is to divide a data set
into clusters or groups of similar variables or cases. Miller (1981) initiated the use of trajectories to
determine the flow climatology, particularly over long time periods. Studies using cluster analysis
techniques on trajectories were conducted by Moody (1986), Moody and Galloway (1988), and later
by Harris and Kahl (1990), Harris (1992), Harris and Kahl (1994), Moody et al. (1994); Dorling
and Davies (1995), Jaffe et al. (1997a), Mahura et al. (1997a) and Mahura et al. (1997b).
SAS/STAT software has tools for many types of statistical analysis including cluster
analysis. We used FASTCLUS procedure, which performs a disjoint cluster analysis on the basis of
Euclidean distances computed from one or more quantitative variables. The observations are
divided into clusters such that every observation belongs to, at least, one cluster. In the case of
separate analyses for different numbers of clusters we run FASTCLUS once for each analysis. This
procedure is intended for use with large data sets, up to 100000 observations. It is possible to print
brief summaries of the clusters it finds. For more extensive examination of the clusters it was
necessary to request output data sets containing a cluster membership variable. Hence, we decided
that FASTCLUS procedure is more useful for our analysis.
Figure 2.3.1 shows all trajectories starting at KNR region during 1992. Another Figure 2.3.2
shows obtained after the cluster analysis the atmospheric transport pathways from the KNR region
11
for 1992. The mean trajectory for each cluster is given with the points indicating 12-hour intervals
between points. The clusters are marked by 2 corresponding numbers. The first number is the
number or identifier of the cluster. The second is the percentage of trajectories within that cluster.
The cluster numbers are used to separate the possible types of transport and are arbitrary.
Figure 2.3.1. 5-day isentropic forward trajectories originated over the KNR region during 1992
We used cluster analysis to divide the calculated trajectories into groups. The following
criteria were used: latitude and longitude values for each time step, which represent the direction
and velocity of air parcel motion. Similarity among trajectories in each cluster is maximized
considering the full length of each 5-day forward trajectory. Within each cluster, individual
trajectories may be averaged to obtain the mean cluster trajectory (or transport pathway). Thus, the
initial large data set may be reduced to a small number of mean cluster plots. These plots then could
be interpreted, based on common synoptic conditions and features.
Figure 2.3.2. Cluster mean trajectories or transport pathways from KNR during 1992
Using the cluster analysis technique we summarized the flow climatology for the KNR
region for 1991-1995, each year, season and month. Details of the clustering and discussion of the
results are presented in consequent Section of the “Results and Discussion” Chapter, and Appendix
12
A. We should also note that trajectories, which cross the top boundary at 80°N (i.e. showing
transport to the Central Arctic territories) of the model grid domain, were not used in cluster
analysis. It is due to their short duration and termination at the border. In particular, these
trajectories did not pass the most populated regions, and hence, have no interest for our study
conditions.
2.4. Probability Field Analysis
Probabilistic analysis is one of the ways to estimate the likelihood in occurrence of one or
another phenomena or event. As we mentioned, in this study we calculated a large number of
trajectories passed other various geographical regions. Each calculated trajectory contained
information about longitude, latitude, altitude, pressure, temperature, relative humidity and other
variables on each time step. The probability fields for these characteristics, individual or combined,
may be represented by a superposition of probabilities for the air parcels reaching each grid region
in the chosen domain or on the geographical map. The most interest for an analysis have the
following probabilistic fields: a) airflow patterns, b) contribution of precipitation, and c) fast
transport. The first type of the fields shows the general common features in the atmospheric
transport patterns, i.e. it may provide a general insight on the possible main direction of the
radioactive cloud’s transport as well as probability to reach or pass any geographical area. The
second type describes a possibility for removal processes from the polluted air mass when air
parcels passed over the particular geographical area. The last type has indication for probability of
the rapid air parcels’ movement during the first day of transport. It is important information,
especially, for the estimation of the short-lived radionuclides impact. I.e. it shows, which territories
may be reached during the first day, and which areas are at the most danger due to higher transport
probability.
In our study probabilistic fields were constructed for the mentioned above three areas of
interest. Fields for the airflow and fast transport were constructed using latitude, longitude, altitude,
and time step for 5 and 1 day trajectories, respectively.
These probability fields reflected 1991-1995, individual year, seasonal, and monthly
variations for the boundary layer. Fields for the precipitation factor were constructed based on the
values of the relative humidity at trajectory’s latitude, longitude, and altitude on each time step. As
the previous fields, these fields have seasonal variations, but additionally it was calculated for
several atmospheric layers between situated between surface, 1.5, 3, 5 km asl, and higher. Results
of the probability field analysis are presented in consequent Section of the “Results and Discussion”
Chapter as well as in Appendixes B,C, and D.
13
3. RESULTS AND DISCUSSION
3.1. KNR Possible Impact
In this study we analyzed all forward trajectories originated over the nuclear reactors
locations to investigate the likelihood of the reactors’ impact on several remote geographical
regions. We assumed that any isentropic trajectory, which crosses into chosen geographical region,
could bring relatively undiluted air parcels containing radionuclides. Therefore, only trajectories
crossed boundaries of these regions were used for further analysis. To analyze the probability of the
plant’s impact, we estimated several parameters: 1) number and percentage of trajectories reached
the boundaries of the geographical regions; 2) number and percentage of days if at least one
trajectory had reached the region; 3) average transport time of air parcels; 4) probability of transport
within different atmospheric layers; 5) likelihood of very rapid (fast) transport - i.e. transport in one
or less days. Such evaluation was performed over the studied period, by individual year, season and
month. A summary of transport from the region of interest to four chosen geographical regions is
presented in Table 3.1.1. Monthly variations in the average time of transport (in days) and number
of trajectories reached regions during 1991-1995 are shown in Figures 3.1.1 and 3.1.2.
Parameter vs. Region
Scandinavia
Trajectories reached the regions
# vs. % of trajectories
3834 (28.4)
Days when trajectories reached the
regions
# vs. % of days
812 (44.5)
Transport time: average ± std.dev.
1.3 ± 1.2
(in days)
Apr,
Highest occurrence of transport
(months)
Jun-Sep
72.8
Fast transport events
% trajectories
Highest occurrence of fast transport Mar-Nov
(months)
70
Transport within boundary layer
(% trajectories)
Europe
Central FSU
Taymyr
363 (2.7)
3192 (23.7)
4535 (33.6)
148 (8.1)
4.5 ± 2.2
788 (43.2)
2.7 ± 1.8
1014 (55.5)
3.4 ± 2.0
Jul-Aug,
Nov-Feb
0.8
Aug-Feb
Sep-Mar
16.0
1.9
Dec
Dec-Mar,
May
55
Dec-Jan,
Apr-May
27
60
Table 3.1.1. Transport summary from KNR region to geographical regions
The winter shows the lowest probability of transport (less than 20% of trajectories) from the
KNR area to the Scandinavian region with the highest during spring and summer. The Taymyr
region reflects the higher occurrence of the transport during October-January (more than 40%), and
lowest occurrence during May-August with a minimum in June (19.1%). For the European region,
there are two periods when probability of transport is higher: during November-February
(approximately 4%) and during July-August (3.5%). The lowest probability – less than 1% - is
observed during May and October. The Central FSU region is characterized by the lower
probability during spring and summer (on average less than 20%). Then this probability gradually
increases during fall and winter reaching absolute maximum in February (36.2%).
It is well known that in the cases of the transport within the boundary layer the higher
concentrations of radionuclides are more likely at the surface. As seen from Table 3.1.1, the
boundary layer transport, on a yearly basis, is predominant for the Scandinavian, European and
Central FSU regions. The free troposphere transport prevails for the Taymyr region. It should be
noted also that the European region is characterized by the boundary layer transport during all
14
seasons, except in summer, when the transport occurs half of the time on the border between the
boundary layer and free troposphere. Although the Taymyr region shows transport on the border
between layers and in the free troposphere throughout the year, during summer the transport pattern
is shifted completely to the higher layers in the free troposphere.
Figure 3.1.1. Monthly variations in the average transport time (in days) from the KNR region to
geographical regions based on the forward trajectories during 1991-1995
Figure 3.1.2. Monthly variations in the average percentage of trajectories originated over the KNR
region at lower altitudes within the boundary layer and reached the geographical regions during
1991-1995
The cases of the fast transport (1 day or less), as shown in Table 2.3.2.1, account on average
for approximately 73, 16, 2.1 and less than 1% for Scandinavian, Central FSU, Taymyr and
European regions, respectively. In this study we found that the lowest probability of the fast
transport events is observed for the European region, and, in particular, all these events took place
during December. In comparison with other regions, the Scandinavian region is characterized by the
highest probability of the fast transport due to its proximity to the KNR area. On average this
probability is 72.8% of the total number of trajectories reaching the region, and it decreases during
winter to 61%. For the Central FSU region, the probability of such transport is higher during winter
(up to 30%) and it decreases by factor of 1.5-2 in other seasons. The Taymyr region as well as
European has the lower probability of the fast transport events. Although on average this probability
is approximately 1.8%, during fall months it is less than 1%, and there is a flat maximum of 4%
during December-January.
In this study, we found that highly populated Scandinavian and Central FSU regions are at
the greatest risk in comparison with the European region. Moreover, the transport of radionuclides
in the case of an accident to the Central FSU region could occur within a day, but it averages 2.7
days, and almost half of the time resulting from the free troposphere transport. To the Scandinavian
region this transport may occur within 0.5 days, but it averages 1.2 days, and 70% of the time
resulting from boundary layer transport.
3.2. Atmospheric Transport Patterns
In this study we had an interest in the fast transport from KNR region. We found also that
the average transport times to chosen geographical regions (as shown in Table 3.1.1) are less than 5
days. Thus, we decided to use in the cluster analysis only forward trajectories of this duration.
Figure 3.2.1 shows the atmospheric transport pathways from the KNR region using trajectories
during 1991-1995, which had an origin within the boundary layer. The mean trajectory for each
cluster is given with points indicating 12-hour intervals. Two numbers were used for each cluster.
15
The first is the identifier of a cluster. The second is the percentage of trajectories within a cluster.
The cluster numbers are only used to separate the possible types of transport and are arbitrary.
In our study six clusters were identified for the KNR region (Figure 2.3.2.3). Four of them
(#1, 2, 3 and 4 with 27, 22, 10 and 16% of occurrence, respectively) show westerly flow. These
observed about 75% of the time. Cluster #6 (7%) shows easterly flow toward the Northern Atlantic
both within the boundary layer and free troposphere. Cluster #5, which occur 18% of the time, is
transport to southwest through the Scandinavian Peninsula into the Baltic Sea. As was noted
previously, we did not consider trajectories, which cross the top of model grid domain and move
into the Central Arctic.
Figure 3.2.1. Atmospheric transport pathways (cluster mean trajectories) from the KNR region
based on the forward trajectories during 1991-1995
Throughout the year, westerly flow is predominant for the KNR region. Transport from the
west varies from 68% (in fall) to 94% (in spring) of the time. Transport from the east occurs from
3% (in winter) to 26% (in summer) of the cases. Transport with the southward component take
place 15% of the time (in winter) increasing up to 25% (in fall). Analyzing trajectories at the higher
altitudes - 1.5 and 3 km asl - we also found that within the free troposphere the probability of
transport from west increases up to 90%. Detail seasonal variations in the mean transport pathways
within the boundary layer are shown in Appendix A.
3.3. Airflow Probability Fields
To test and compare the results of clustering we calculated the airflow probability field
using all forward trajectories during 1991-1995. Such probability fields show geographical
variations of airflow patterns from the chosen site. In a climatological sense, the pathway of airflow
from the chosen site could be represented by a superposition of the probability of air parcels
reaching each grid region on a geographical map (Merrill, 1994). The regions with higher
occurrence of trajectory passages are areas where the probability of KNPP impact will be higher.
Figure 3.3.1 shows the airflow probability field for KNR constructed using 1991-1995
trajectories. Each probabilistic field is presented using isolines, given with interval of 5%, on the
background of the geographical territories. The areas of higher probability, which are located close
to the KNR region, indicate that trajectories have spent more time in this geographical area.
Because all trajectories start near the site, the cumulative probability is 100% there. Thus, the field
was altered using a correction factor. This factor accounts the increase with distance from the site
therefore it takes into account the contribution at greater distances. The airflow probability field
16
also shows that westerly flow is predominant for the KNR region. It is in agreement with the results
of the trajectory’s cluster analysis. Our detail analysis showed that transport toward east is clearly
dominates throughout the year. The southward components of flow reveal themselves starting in
January and continued until September. The westward flow components show up in April and
summer months. Detail seasonal airflow probability fields within the boundary layer are shown in
Appendix B.
Figure 3.3.1. Airflow probability field within the boundary layer for the trajectories, originated over
the KNR region during 1991-1995 (isolines are shown every 10%)
3.4. Precipitation Factor Probability Fields
To account for the contribution of radionuclide wet removal during transport of an air parcel
we calculated simultaneously for each trajectory point (i.e. at latitude, longitude and pressure level
point) the value of relative humidity. Based on calculated temporal and spatial distribution of the
relative humidity we constructed the relative humidity (we named it “precipitation factor”)
probability fields over the studied geographical areas. Several atmospheric layers - surface - 1.5,
1.5-3, 3-5, and above 5 km asl - were examined to determine altitudinal differences in the
possibility of removal processes. We assumed, that areas with the relative humidity above 65%
were areas, where water vapor could be condensed and later removed in the form of precipitation.
Figure 3.4.1 shows the relative humidity probability field for the boundary layer. Appendix
B shows the seasonal distribution of the precipitation factor within the boundary layer. Detail
seasonal variations in the precipitation factor probability field for various layers are shown in
Appendix C. Our detail analysis of the probabilistic fields concluded that contribution of the
precipitation factor dominates within the low troposphere layers in the areas associated with the
activity of the Icelandic low and along main tracks of the cyclone systems.
17
Figure 3.4.1. Precipitation factor probability field for the boundary layer based on the relative
humidity fields for the trajectories, originated over the KNR region during 1991-1995
(isolines are shown every 5%)
3.5. Fast Transport Probability Fields
Although atmospheric transport from the radiation risk site or region to another geographical
area may occur at any time, the fast transport is the greatest concern. It is especially valid and
important issue for the measures of the emergency response and preparedness. To study the
contribution of the fast transport we evaluated all available trajectories during the first day of their
transport. I.e. all trajectories were terminated routinely after 1 day of transport. In a similar way, as
for the flow probability, we constructed the fast transport probability fields. All these fields show
the probability of air transport from the region of interest during the first day. An analysis has been
done for the entire period and each year, as well as by season and month. We analyzed the lowest
altitude of trajectories, i.e. trajectories starting below 0.5 km asl. This is due to major importance of
the boundary layer transport, especially for the short-lived radionuclides, during the first several
hours after an accident.
Our analysis showed that on a yearly basis the highest probability of the fast transport is
with the eastward and southward components. During summer the southwestern component of the
fast transport dominates within the boundary layer. During fall and winter the eastward and
southeastward components are predominant, which show transport toward less populated territories
of the Northern Arctic Russia. February is associated with the highest probability of transport
toward the Central Russia. During spring the southward components exist. The southwestward
component toward the Scandinavian countries is well pronounced during April (as shown in Figure
3.5.1). November and December represent the clear eastward component in the fast transport.
Appendix D shows seasonal variations in the fast transport pattern.
18
Figure 3.5.1. Fast transport probability field within the boundary layer for the trajectories,
originated over the KNR region during April of 1991-1995 (isolines are shown every 10%)
19
CONCLUSIONS AND RECOMMENDATIONS
The main purpose of this study was the testing of the methodology to examine probabilistic
atmospheric transport patterns, precipitation factor’s contribution, fast transport, and possible
impact of the Kola Nuclear Reactors, located in the North Western Russia region, on several
geographical regions: Scandinavia, Europe, Central FSU and Taymyr – following a hypothetical
accident. The period studied is 1991-1995. An isentropic trajectory model has been used to
calculate forward isentropic trajectories which originated over the studied region. Atmospheric
transport patterns were identified using a cluster analysis of the 5-day forward trajectories
originating over the KNR region. Probabilities for the airflow, precipitation, and fast transport
patterns were constructed using probabilistic field’s analysis.
The main results of this part of the study are the following:
1. The methodology to estimate potential nuclear risk and vulnerability levels – i.e. examination of
the atmospheric transport patterns, contribution of the precipitation factor, fast transport, and
possible impact - was applied and tested. As an example, we selected the Kola Nuclear Reactors
in the North Western Russia. Examination has been done for the following geographical
regions: Scandinavia, Europe, Central FSU and Taymyr.
2. Scandinavian and Central FSU regions, as the most highly populated regions, are at the greatest
risk of the radioactive pollution in the case of an accident comparing to the European region.
Although transport from these regions could occur in 1-1.5 days, it averages in 1.3 and 2.7 days
for the Scandinavian and Central FSU regions, respectively. It should be noted also that for
these regions transport mostly (65-70% of the time) occurs within the boundary layer.
3. For the studied region, during all seasons the westerly flows are predominant within the
boundary layer, and on average it occurs 75% of the time throughout the year. In the free
troposphere it could occur about 90% of the time.
4. Precipitation factor’s contribution dominates in the low troposphere layers, and in the areas
associated with the Icelandic Low activity as well as along the main tracks of the cyclone
systems.
5. The fast transport pattern shows the dominance of the eastward and southward components
throughout the year. February is associated with the highest probability of fast transport toward
the Central Russia regions. April has a clear dominance of the fast transport toward the
Scandinavian countries.
Our results has been used for the further analysis of the possible consequences as a result of
the hypothetical accident at the Kola nuclear power plant. Our obtained results could be used in the
event of an accident to estimate, and especially at the first stages, the probability of the radionuclide
transport from the accident location. They may be used as input data for the high-level models in
order to estimate detail impact on the environment and population within the Barents region
territories. They also could be important for the probabilistic assessment of the pollutant’s longrange transport as well as in the studies of the social and economical consequences from the
radioactive risk sites in the Barents Euro-Arctic region.
From our point of view, several areas need continue attention and recommended for an
additional study:
20
1. Due to the possibility of rapid transport of radionuclides to Central FSU and Scandinavian
regions, current and future response plans should include provisions for rapid event notification.
2. Modeling of transport, deposition, removal processes and an estimation of consequences should
be continued, especially focusing on the rapid transport events which could bring radionuclides
to Scandinavian and Central FSU regions as quickly as 24 hours following an accident.
3. For a future study, it will important to develop a method for calculation of probabilistic
trajectories and direct impact, taking into account the precipitation factor, the estimation of the
boundary layer influence and other parameters.
4. For a future study, it will important to conduct a probabilistic estimation of possible effects on
the population in the Barents Euro-Arctic Region and territories of Central Arctic from KNPP,
other radioactive risk objects and industrial pollution source regions.
5. We suppose that plant safety at these reactors must be improved to reduce the likelihood of a
serious accident as well as monitoring for radionuclide release should be continued in the KNPP
region as well as surrounding territories of Russia and Scandinavian countries.
We believe that the use of mentioned methodology -- trajectory modeling with high-resolution
grid and cluster analysis technique -- might be successively applied for similar studies.
21
ACKNOWLEDGMENTS
The authors are very grateful for collaboration with Drs. Alexander Baklanov (Danish
Meteorological Institute, Copenhagen, Denmark), Sergey Morozov (Institute of the Northern
Ecological Problems, Kola Science Center, Russian Academy of Sciences, Apatity, Russia), and
John Merrill (Center for Atmospheric Chemistry, Graduate School of Oceanography, University of
Rhode Island). Thanks to Glenn Shaw and Sue-Ann Bowling (Geophysical Institute, University of
Alaska, Fairbanks) for constructive discussions. Thanks to Robert Huebert and Dennis Fielding
(ARSC) for computer assistance.
The computer facilities from the National Center for Atmospheric Research (NCAR,
Boulder, Colorado, USA), Arctic Region Supercomputing Center (ARSC, Fairbanks, Alaska,
USA), University of Washington (UW) of the Atmospheric Sciences Department, and the Institute
of the North Ecological Problems (INEP, Apatity, RUSSIA), as well as databases of NCAR and
European Center for Medium-Range Weather Forecast (ECMWF, Reading, UK) have been used in
this work.
The isentropic trajectory model has been developed and modified using FORTRAN-77 and
90 languages. Partially, some model blocks and blocks for data statistical analysis were developed
in FORTRAN and C++. Trajectory’s visualization was done using NCAR Graphics and General
Mapping Tools (GMT) software packages. Cluster analysis of trajectories we performed using SAS
software. Programs for analysis were written in SAS language. Probabilistic fields were constructed
using MATLAB-v.5 software, which includes the geographical maps’ addition.
22
REFERENCES
Aarkrog, A. : Radioactivity in Polar Regions -- Main Sources. Journal of Environmental Radioactivity, Vol 25, Iss 1, p
21-35, 1994
Baklanov, A., Bakulin, L., Zolotkov, A., Kluychnikova, E. : Ecological Radiation Situation in the Arctic Region of
Russia. In:Ekonord-Inform, Special Information Issue of International Ecological Ekonord-Magazine, n1, p 3-16, 1993
Baklanov, A., Mahura, A., Morozov, S. : The Simulation of Radioactive Pollution of the Environment after an
Hypothetical Accident at the Kola Nuclear Power Plant. Journal of Environmental Radioactivity, Vol 25, Iss 2, p 65-84,
1994
Baklanov, A., Bergman, R., Segerstahl, B. : Radioactive Sources in the Kola region: Actual and Potential Radiological
Consequences for Man. Research Report of International Institute for Applied Systems Analysis, Laxenburg, Austria, p
1-255, 1996
Bergman, R., A. Baklanov and B. Segerst?hl. : Overview of nuclear risks on the Kola peninsula. Summary report. KAS
of the RAD Project, International Institute for Applied Systems Analysis, Laxenburg, Austria 110 p, 1996
Bocharov, M.V., Loborev, V.M., Matveichuk, I.P., Sudakov, V.V. : The global Radioactive Contamination of the
Environment of the Northern Hemisphere and the Contribution to it of Soviet Nuclear Tests. Journal Atomic Energy,
Vol 78, Iss 1, p 48-51, 1995
Bouville, A. : The Chernobyl Accident. Journal of Radiation Protection Dosimetry, Vol 60, Iss 4, p 287-293, 1995
Bridges, O. and Bridges, J.W. : Radioactive waste problems in Russia. Journal of Radiological Protection, Vol 15, Iss 3,
p 223-234, 1995
Bune, V.I. et al. (Ed) :Seismic Region Division of the USSR territory. Moscow, Printing House “Science’, p 1-307, (in
Russian), 1980
Dorling, S.R. and Davies, T.D. : Extending Cluster Analysis -- Synoptic Meteorology Links to Characterize Chemical
Climates at Six Northwest European Monitoring Stations. Atmospheric Environment, Vol 29, Iss 2, p 45-167, 1995
Draxler, R.R. : Sensitivity of a Trajectory Model to the Spatial and Temporal Resolution of the Meteorological Data
During CAPTEX. Journal of Climate and Applied Meteorology, Vol 26, p 1,577-1,588, 1987
Elokhin, A.P. and Solovei, A.F. : Assessment and Prediction of the Scales of Radioactive Contamination of the
Environment by Emissions from a Nuclear Power Plant. Journal Atomic Energy, Vol 77, Iss 2, p 635-640, 1994
Harris, J.M. and Kahl, J.D. : A Descriptive Atmospheric Transport Climatology for Mauna Loa Observatory, Using
Clustered Trajectories. Journal of Geophysical Research, Vol 95, p 13,651-13,667, 1990
Harris, J.M. : An Analysis of 5-day Midtropospheric Flow Patterns for the South Pole: 1985-1989. Tellus, 44B, p 409421, 1992
Harris, J.M. and Kahl, J.D. : Analysis of 10-day Isentropic Flow Patterns for Barrow, Alaska: 1985-1992. Journal of
Geophysical Research, Vol 99, p 25,845-25,855, 1994
Jaffe, D., Mahura, A., Andres R. : Atmospheric Transport Pathways to Alaska from Potential Radionuclide Sites in the
Former Soviet Union. Research Report, UAF-ADEC Joint Project 96-001, p 1-71, Feb 1997a
Jaffe, D.A., Mahura A., Kelley, J., Atkins J., Novelli P.C., Merrill J. Impact of Asian Emissions on the Remote North
Pacific Atmosphere: Interpretation of CO Data from Shemya, Guam, Midway and Mauna Loa. Journal of Geophysical
Research, Vol. 23, pp.28627-28636, 1997b
Kahl, J.D. : On the Prediction of Trajectory Model Error. Atmospheric Environment, Vol 30, p 2,945-2,957, 1996
Klein, W.H. : Principal Tracks and Mean Frequencies of Cyclones and Anticyclones in the Northern Hemisphere.
Research Paper 40 of the USA Weather Bureau, p 1-60, 1957
23
Krebs, J.S. and Barry, R.G. : The Arctic Front and the Tundra-Taiga Boundary in Eurasia. Geographical Review, Vol
60, Iss 4, p 548-554, 1970
Mahura, A., Jaffe D., Andres, R., Dasher, D., Merrill, J. : Atmospheric Transport Pathways to Alaska from Potential
Radionuclide Sites in the Former Soviet Union. Proceedings of American Nuclear Society Sixth Topical Meeting on
Emergency Preparedness and Response, San Francisco, California, April 1997, Vol 1, p 173-174, 1997a
Mahura, A., Jaffe D., Andres, R., Dasher, D., Merrill, J. : Atmospheric Transport Pathways from the Kola Nuclear
Power Plant. Extended abstracts of Intentional Symposium on Environmental Pollution of the Arctic and The Third
International Conference on Environmental Radioactivity in the Arctic, Tromso, Norway, June 1-5, 1997, Vol 2, p 5254, 1997b
Merrill, J., Bleck, R. and Avila, L. : Modeling Atmospheric Transport to the Marshall Islands. Journal of Geophysical
Research, Vol 90, p 12,927-12,936, 1985
Merrill, J., Bleck, R. and Boudra, D.B. : Techniques of Lagrangian Trajectory Analysis in Isentropic Coordinates.
Monthly Weather Review, Vol 114, p 571-581, 1986
Merrill, J. : Isentropic Airflow Probability Analysis. Journal of Geophysical Research, Vol 99, p 25,881-25,889, 1994
Miller, J.M. : A Five-Year Climatology of Back Trajectories from the Mauna Loa Observatory, Hawaii. Atmospheric
Environment, Vol 15, Iss 9, p 1,553-1,558, 1981
Moody, J.L. : The Influence of Meteorology on Precipitation Chemistry at Selected Sites in the Eastern United States.
Ph.D. thesis, University of Michigan, Ann Arbor, p 1-176, 1986
Moody, J.L. and Gallow, J.N. : Quantifying the Relationship Between Atmospheric Transport and the Chemical
Composition of Precipitation on Bermuda. Tellus, 40B, p 463-479, 1988
Moody, J.L., Oltmans, S.J., Hiram Levy II, Merrill, J.T. : Transport Climatology of Tropospheric Ozone: Bermuda,
1988-1991. Journal of Geophysical Research, Vol 100, p 7,179-7,194, 1994
Nilsen, T. and Bohmer, N. : Sources of Radioactive Contamination in the Murmansk and Arkhangel’sk Counties.
Bellona Report #1, 1994 WWW address -http://www.ngo.grida.no/ngo/bellona/ehome/russia/murmark/index.htm
Nilsen, T., Kudrik, I. and Nikitin, A. : The Russian Northern Fleet -- Sources of Radioactive Contamination. Bellona
Report #2, 1996 WWW address - http://www.ngo.grida.no/ngo/bellona/ehome/russia/nfl/index.html
Nuclear Wastes in the Arctic: An Analysis of Arctic and Other Regional Impacts From Soviet Nuclear Contamination.
OTA-ENV-632 (Washington, DC: U.S. Government Printing Office, Sep 1995), p 159-163, 1995
Romesburg, C.H. : Cluster Analysis for Researches. Lifetime Learning, Belmont, California, p 1-334, 1984
SAS/GRAPH Software: Introduction, Version 6, First Edition, 1990
SAS Language and Procedures: Introduction, Version 6, First Edition, 1991
SAS/STAT User's Guide, Version 6, Fourth Edition, Vol 1 and Vol 2, 1990
SAS Technical Report P-229, SAS/STAT Software: Changes and Enhancements. Release 6.07, 1992
Shaw, G.E. : Chemical Air Mass Systems in Alaska. Atmospheric Environment, Vol 22, p 2,239-2,248, 1988
Shvyryaev, Yu.V., Derevyankin, A.A., Tokmachev, G.V. : Probabilistic Modeling of Accident Sequences for Power
Plants with VVER-440 Reactors. Journal Atomic Energy, Vol 73, Iss 1, p 593-597, 1992
Soviet-Designed Nuclear Power Plants in Russia, Ukraine, Lithuania, the Czech Republic, the Slovak Republic,
Hungary, and Bulgaria. Source Book, Fourth Edition, Nuclear Energy Institute, 1996, WWW address http://www.insc.anl.gov/neisb/NEISB_C.html
Walsh, J.E. and Chapman, W.L. : Short Term Climatic Variability of the Arctic. Journal of Climate, Vol 3, p 237-250,
1990
24
APPENDIX A
Seasonal Atmospheric Transport Pathways from KNR
Figure A1. Atmospheric transport pathways from KNR during spring
Figure A2. Atmospheric transport pathways from KNR during summer
25
Figure A3. Atmospheric transport pathways from KNR during fall
Figure A4. Atmospheric transport pathways from KNR during winter
26
APPENDIX B
Seasonal Airflow Probability Fields within the Boundary Layer
Figure B1. Spring airflow probability field within the boundary layer
Figure B2. Summer airflow probability field within the boundary layer
27
Figure B3. Fall airflow probability field within the boundary layer
Figure B4. Winter airflow probability field within the boundary layer
28
APPENDIX C
Seasonal Precipitation (Relative Humidity) Probability Fields
Figure C1. Spring precipitation probability field for the surface-1.5 km asl layer
Figure C2. Summer precipitation probability field for the surface-1.5 km asl layer
29
Figure C3. Fall precipitation probability field for the surface-1.5 km asl layer
Figure C4. Winter precipitation probability field for the surface-1.5 km asl layer
30
APPENDIX D
Seasonal Fast Transport Probability Fields within the Boundary Layer
Figure D1. Spring fast transport probability field within the boundary layer
Figure D2. Summer fast transport probability field within the boundary layer
31
Figure D3. Fall fast transport probability field within the boundary layer
Figure D4. Winter fast transport probability field within the boundary layer
8
Northern Studies Working Papers
1 Andreev, Oleg A. and Olsson, Mats-Olov (1992). Environmentalism in Russia’s
North-West.
2 Svensson, Bo (1994). Opportunity or Illusion? Prospects for Foreign Direct Investment in North-West Russia.
3 Andreev Oleg A. and Olsson Mats-Olov (1994). Regional Self-government in Russia - The Situation in the County of Murmansk.
4 Bröms, Peter (1994). Changing Stands? The negotiations of an International Environmental Security Regime in Barents Euro-Artic Region.
5 Eriksson, Johan (1994). Security in the Barents Region: Interpretations and Implications of the Norwegian Barents Initiative.
6 Svensson, Bo (1994). Barentsregionen, dess regionala byggstenar och den transregionala utmaningen.
7 Bröms, Peter (1995). Living on the Edge: “The Russian Problem” of the Barents
Region’s Security Problematique.
8 Svensson, Bo (1995). The Political Economy of East-West Transnational Regionalization.
9 Olofsson, Ebba (1995). Samer utan samiska rättigheter och icke-samer med samiska
rättigheter - en fråga om definition.
10 Monsma, Mark (1995). Winds of Change Within the Barents Organization: An
Institutional Analysis of Transnational Regionalizations in the North.
11 Nilsson, Per Ola (1996). Republiken Karelen. En översikt över dess ekonomiska och
politiska förhållanden.
12 Masegosa Carrillo, José Luis (1998). Regional Security Building in Europe. The Barents Euro-Artic Region.
13 Gidlund, Janerik; Wiberg, Ulf and Gunnarsson, Malin (1998). Knowledge based
strategies for Sustainable Development and Civic Security. A North-Swedish Initiative in the Northern Dimension.
14 Masegosa Carrillo José Luis (1999). The County of Västerbotten and its Recent
Membership in the Barents Euro-Artic Region. Fact-Finding Report.
15 Hallström, Marie-Louise (1999). Risks and Nuclear Waste. Nuclear problems, risk
perceptions of, and societal responses to, nuclear waste in the Barents region.
16 Axensten, Peder (2001). Modelling and Visualizing Short Term Impact of a Nuclear
Accident on Transportation Flows.
17 Hedkvist, Fred (2001). Great Expectations. Russian Attitudes to the Barents Region
Co-operation.
18 Tønnesen, Arnfinn (2001). Perception of Nuclear Risk at the Kola Peninsula.
19 Namjatov, Aleksey (2001). Modern Level of Radioactivity Contamination and Risk
Assessment in the Coastal Waters of the Barents Sea.
20 Arkhanguelskaia, V. Guenrietta and Zykova, A. Irina (2001). Social Scientific
Nuclear Waste Risk Assessment in the Barents Region.
21 Morozov, Sergey and Naumov, Andrey (2001). Assessment of Potential Risk for
Kola’s Population from Radiological Impact of Accident on Spent Nuclear Fuel Facilities.
22 Lundström, Christoffer (2001). Simulering av radioaktiv beläggning vid utsläpp på
Kolahalvön.
23 Baklanov, A.; Bergman, R.; Lundström, C. and Thaning, L. (2001). Modelling of
Episodes of Atmospheric Transport and Deposition from Hypothetical Nuclear Accidents on the Kola Peninsula.
24 Mahura, Alexander; Andres, Robert; Jaffe, Daniel (2001). Atmospheric Transport
Patterns from the Kola Nuclear Reactors.
25 Hedkvist, Fred; Weissglas, Gösta (2001). Regionalisation in North-Western Europe:
Spatial Planning or Building a Frame for Development Cooperation. The Case of the
Barents Region.