Content uploaded by Dmitry A Streletskiy
Author content
All content in this area was uploaded by Dmitry A Streletskiy on Nov 12, 2019
Content may be subject to copyright.
Full Terms & Conditions of access and use can be found at
https://www.tandfonline.com/action/journalInformation?journalCode=tpog20
Polar Geography
ISSN: 1088-937X (Print) 1939-0513 (Online) Journal homepage: https://www.tandfonline.com/loi/tpog20
Assessment of the cost of climate change impacts
on critical infrastructure in the circumpolar Arctic
Luis Suter, Dmitry Streletskiy & Nikolay Shiklomanov
To cite this article: Luis Suter, Dmitry Streletskiy & Nikolay Shiklomanov (2019): Assessment
of the cost of climate change impacts on critical infrastructure in the circumpolar Arctic, Polar
Geography, DOI: 10.1080/1088937X.2019.1686082
To link to this article: https://doi.org/10.1080/1088937X.2019.1686082
Published online: 11 Nov 2019.
Submit your article to this journal
View related articles
View Crossmark data
Assessment of the cost of climate change impacts on critical
infrastructure in the circumpolar Arctic
Luis Suter
a
, Dmitry Streletskiy
a,b
and Nikolay Shiklomanov
a
a
Department of Geography, The George Washington University, Washington, DC, USA;
b
Institute of the Earth’s
Cryosphere of Tyumen Scientific Center of Siberian Branch of Russian Academy of Sciences, Tyumen, Russia
ABSTRACT
The Arctic is experiencing pronounced climatic and environmental
changes. These changes pose a risk to infrastructure, impacting the
accessibility and development of remote locations and adding
additional pressures on local and regional budgets. This study
estimates the costs of fixed infrastructure affected by climate change
impacts in the Arctic region, specifically on the impacts of permafrost
thaw. Geotechnical models are forced by climate data from six CMIP5
models and used to evaluate changes in permafrost geotechnical
characteristics between the decades of 2050–2059 and 2006–2015
under the RCP8.5 scenario. Country-specific infrastructure costs are
used to estimate the value of infrastructure affected. The results show
a 27% increase in infrastructure lifecycle replacement costs across the
circumpolar permafrost regions. In addition, more than 14% of total
fixed infrastructure assets are at risk of damages due to changes in
specific environmental stressors, such as loss of permafrost bearing
capacity and thaw subsidence due to ground ice melt. Regions of
Northern Canada and Western Siberia are projected to be particularly
affected and may require additional annual spending in the excess of
1% of annual GRP to support existing infrastructure into the future.
ARTICLE HISTORY
Received 10 May 2019
Accepted 24 October 2019
KEYWORDS
Arctic; infrastructure; climate
change; permafrost;
economics; GIS
Introduction
The circumpolar Arctic is home to about 4 million people –roughly 0.15% of the world
population –but it plays a disproportionally large role in the global economy, contributing
0.6% to global gross domestic product (GDP) (Heleniak & Bogoyavlensky, 2015; Larsen &
Huskey, 2015). The Arctic’s role in the global economy is only projected to increase, with
the region holding rich natural resources, including an estimated 25% of the world’s undis-
covered oil and gas reserves (Hossain, 2017). Resource and urban development often take
place in remote locations and require various types of infrastructure, including buildings,
roads, railroads, airports, ports, and pipelines to maintain large-scale economies (Laruelle,
2015; Orttung, 2016). This infrastructure is built at much greater cost relative to temperate
regions, due to remoteness and the absence of local production centers (Budzik, 2009;
Huskey, Mäenpää, & Pelyasov, 2015).
Meanwhile, the Arctic is warming at twice the rate of the global average (AMAP, 2017).
Changing climatic conditions have the potential to improve the accessibility of offshore
resources (Stephenson & Pincus, 2018; Stephenson, Smith, & Agnew, 2011) but may also
© 2019 Informa UK Limited, trading as Taylor & Francis Group
CONTACT Luis Suter lsuter@gwmail.gwu.edu Department of Geography, The George Washington University, 2036 H
St., NW, Washington, DC 20052, USA
POLAR GEOGRAPHY
https://doi.org/10.1080/1088937X.2019.1686082
negatively affect existing infrastructure, which is much larger in scale, and critical to the support
of marine activities (Anisimov & Reneva, 2006; Clarke-Sather et al., 2017;Huskeyetal.,2015).
Land-based infrastructure in the Arctic requires unique solutions in order to withstand severe
climatic conditions, including low temperatures, high snow loads, and the presence of perma-
frost (Frolov, 2015). Permafrost is critical, because the characteristics of the perennially frozen
ground are key variables in the design of foundations for various infrastructures.
Permafrost has highly variable temperature, extent, and thickness. This is due to differ-
ences in climatic conditions, land cover, and soil properties. The presence of permafrost
requires innovative construction techniques designed to decouple frozen ground from the
active and latent heat of structures. This is commonly accomplished by elevating buildings
and structures using piles anchored in the permafrost (Andersland & Ladanyi, 2003) and/
or installing technologies such as thermosyphons to maintain the temperature of soil. The
thermal state of permafrost and the thickness of the active layer –a layer just above the per-
mafrost which is subject to annual freezing and thawing –intrinsically affect the ability of
foundations to support structures (Shiklomanov, Streletskiy, Grebenets, & Suter, 2017a).
Permafrost warming and active layer thickening (ALT) decreases the bearing capacity of
the soil and results in ground subsidence in areas with ice-rich permafrost (Nelson, Anisi-
mov, & Shiklomanov, 2001; Streletskiy, Sherstiukov, Frauenfeld, & Nelson, 2015; Streletskiy
et al., 2016; Romanovsky et al., 2017). Where such changes are colocated with infrastructure,
their cumulative impacts may result in replacement and costs related to increased mainten-
ance (Grebenets, Streletskiy, & Shiklomanov, 2012; Schweikert, Chinowsky, Espinet, &
Tarbert, 2014).
Permafrost temperatures have steadily increased since the 1980s (AMAP, 2017; Streletskiy
et al., 2017). Following air temperature trends, increases in permafrost temperature have been
greatest in cold permafrost regions of the Alaskan Arctic, Canadian high Arctic, Svalbard, and
Northern Siberia (Aalto, Harrison, & Luoto, 2017; Aalto, Karjalainen, Hjort, & Luoto, 2018;
Biskaborn et al., 2019; Romanovsky et al., 2017). In lower latitudes, where discontinuous per-
mafrost is more common, permafrost temperature increases are less pronounced, however,
ALT is increasing (Shiklomanov et al., 2007; Streletskiy, Shiklomanov, & Nelson, 2012b).
Hazard-risk modeling of climate change impacts has been successfully applied to inform
policymakers and the public about potential future environmental conditions (Arndt, 2001;
Monmonier, 2008; Tol, 2018). Several studies have demonstrated methodologies to estimate
the increase of climate change impacts on infrastructure, within individual Arctic nations
including Alaska, Canada, and Russia (Dore & Burton, 2001; Larsen et al., 2008; Melvin
et al., 2017; Porfiriev et al., 2017; Streletskiy, Suter, Shiklomanov, Porfiriev, & Eliseev,
2019). Larsen et al. (2008) established relationships between permafrost extent, air tempera-
ture increase, and infrastructure’suseful lifespan. Melvin et al. (2017) and the Environmental
Protection Agency (EPA) (2017) implemented stressor-response models that estimated the
impacts of specific hazards using a GIS modeling approach (Chinowsky & Arndt, 2012).
Hjort et al. (2018) developed a hazard-risk index to assess the impact of permafrost degra-
dation on infrastructure at the circumpolar scale. Streletskiy et al. (2019) used a geotechnical
model to evaluate the cost of climate change impacts in Russian regions with permafrost,
based on government statistics on the value of fixed infrastructure by regions downscaled,
using gridded dataset of population distribution.
However, an assessment examining the costs needed to maintain and replace Arctic infra-
structure is not currently available at the circumpolar scale. Such economically-framed metrics
have proven to be effective tools for communicating with business, government, and public
2L. SUTER ET AL.
(Wihbey & Ward, 2016), all of whom are increasingly active stakeholders in the development
of the Arctic.
This study builds on previous works by integrating the infrastructure lifespan replace-
ments model (Larsen et al., 2008), the infrastructure stressor-response model (Melvin
et al., 2017), the permafrost geotechnical model (Shiklomanov, Streletskiy, Swales, &
Kokorev, 2017b; Streletskiy, Shiklomanov, & Nelson, 2012a), and a compiled geodatabase
of Arctic infrastructure, to quantitatively evaluate the cost of climate change impacts on criti-
cal infrastructure across the circumpolar Arctic.
Data and methods
Study area
The border to the Arctic is not explicitly defined. This is reflected by varying definitions
based on physiographic features, country borders, and various international organizations
delineations (CAVM, 2003; Husebekk, Andersson, & Penttilä, 2015; Klüter, 2000; NSIDC,
2018;O’Rourke, 2019; Orttung (Ed.), 2016). The study area of this assessment (Figure 1) con-
glomerates the maximum land area defined as ‘Arctic’by four major organizations: The
Arctic Human Development Report, Arctic Monitoring and Assessment Program, Conser-
vation of Arctic Flora and Fauna program, and Arctic Council Emergency Preparedness, Pre-
vention, and Response program. The study area extends further south to include areas with
significant permafrost within Alaska, Canada, Norway, Sweden, Finland, and Russia.
However, the Scandinavian countries, Finland, Iceland, Greenland, and the Faroe Islands
were excluded from the sub-national economic analysis on the basis of minimal fixed infra-
structure located on permafrost and a lack of necessary socioeconomic datasets at the appro-
priate scale and complexity.
Figure 1. Permafrost extent and administrative regions within the study area.
POLAR GEOGRAPHY 3
Arctic infrastructure inventory
At the small geographic scale of this study, infrastructures can be classified as linear infra-
structure or point-based infrastructure. Linear infrastructure includes roads, railroads, and
pipelines, while point-based infrastructure applies to buildings, airports, and ports. Publicly
available sources of governmental geodata, such as the Alaska Department of Natural
Resources, GeoGratis CanVec Series (Canada), the Norwegian Mapping Authority, the
Swedish Land Survey, the Finnish National Landsurvey, and OpenData Russia were
used to compile a circumpolar infrastructure database (Table 1). Governmental sources
were supplemented by publicly available repositories of geospatial data, namely DIVA
GIS, Natural Earth, and GeoFabrik (OpenStreetMap data). These open-source datasets
were cleaned and classified to extract relevant infrastructure types, with priority given to
the governmental datasets in cases of overlap.
Governmental GIS data on buildings was particularly variable in terms of attributes and
granularity. For Alaska, a dataset that was previously used by Melvin et al. (2017)ina
regional analysis of permafrost thaw impacts was employed. This dataset includes attributes
of public building size and value, among others, however, it does not include private
Table 1. Infrastructure inventory with sources by country.
Country
Infrastructure
type Source
Alaska Roads Alaska Department of Natural Resources
Railroads Alaska Department of Natural Resources
Pipelines Alaska Department of Natural Resources
Airports/Ports Arctic Marine and Aviation Transportation Infrastructure Initiative (AMATII)
Buildings Personal Correspondence with Dr. April Melvin (NAS, PRB)
Canada Roads GeoGratis CanVec Series –Transportation Features
Railroads GeoGratis CanVec Series –Transportation Features
Pipelines GeoGratis CanVec Series –Transportation Features
Airports/Ports Arctic Marine and Aviation Transportation Infrastructure Initiative (AMATII)
Buildings GeoGratis CanVec Series –Manmade Features
Iceland Roads National Land Survey of Iceland
Railroads National Land Survey of Iceland
Pipelines N/A
Airports/Ports Arctic Marine and Aviation Transportation Infrastructure Initiative (AMATII)
Buildings National Land Survey of Iceland –Structures
Norway Roads Norwegian Mapping Authority –Road Network
Railroads DIVA GIS
Pipelines Norwegian Petroleum Directorate
Airports/Ports Arctic Marine and Aviation Transportation Infrastructure Initiative (AMATII)
Buildings Norwegian Mapping Authority (Kartverket) –Official Addresses
Sweden Roads Swedish Land Survey GeoData Portal
Railroads Swedish Land Survey GeoData Portal
Pipelines N/A
Airports/Ports Arctic Marine and Aviation Transportation Infrastructure Initiative (AMATII)
Buildings Swedish Land Survey GeoData Portal
Finland Roads Finnish National Land Survey
Railroads Finnish National Land Survey
Pipelines Finnish National Land Survey
Airports/Ports Arctic Marine and Aviation Transportation Infrastructure Initiative (AMATII)
Buildings Finnish National Land Survey
Russia Roads Open Data Russia (Supplemented with DIVA GIS and OpenStreetMap)
Railroads Open Data Russia (Supplemented with DIVA GIS and OpenStreetMap)
Pipelines Proposed and Existing Russia Oil, Gas, and Product Pipelines (ArcGIS Online)
Airports/Ports Arctic Marine and Aviation Transportation Infrastructure Initiative (AMATII)
Buildings State Fund for Assistance to Reform housing and communal services (database accessed
through GitHub)
4L. SUTER ET AL.
residences. In Russia, detailed data on homes was accessible through the website of the State
Corporation Fund for Assistance with Housing and Communal Services, including infor-
mation on the number of dwellings per building (ReformaGKH, 2018). In Canada,
Iceland, Norway, Sweden, and Finland, point-based building and address datasets were gath-
ered from governmental statistical sources. While this aggregated inventory is not fully inte-
grated m in terms of common attributes or granularity, it represents some of the best
collated, publicly available, data at the circumpolar scale.
In the absence of high-quality and uniform geospatial data at the circumpolar scale, and
considering the small geographic scale of analysis, assumptions were made that the ratio of
infrastructure built on permafrost is proportional to the extent of permafrost within each
corresponding permafrost zone (Streletskiy et al., 2019). Permafrost extent is classified as
island (<10% coverage), sporadic (10%–50% coverage), discontinuous (50%–90% coverage),
and continuous (90%–100% coverage). Therefore, infrastructure assets within each grid cell
were weighted by the 0.1, 0.5, and 0.9 coefficients for sporadic, discontinuous, and continu-
ous permafrost zones, respectively, and no infrastructure was assumed to be present on
island permafrost.
Selection of climate input
This study utilized climate inputs from six preselected general circulation models (GCMs)
included in the CMIP5 project (WCRP, 2011). GCMs are computationally-driven simu-
lations of the global climate system that account for multiple climate components (i.e.
oceans, cryosphere, land cover, etc.), as well as the complex interactions and exchanges of
mass and energy between them (Gramelsberger & Feichter, 2011). The model ensemble
for this study was selected based on a documented ability to successfully reproduce historical
temperature trends in the Russian Arctic, (Anisimov, Kokorev, & Zhil’tsova, 2013) and was
previously applied for climate change impacts assessments of the Arctic regions (Melvin
et al., 2017; Shiklomanov et al., 2017b; Streletskiy et al., 2019). These CMIP5 models are:
CanESM2 (CanESM), CSIRO-Mk-3.6 (CSIRO), HadGEM2-ES (HadGEM), GFDL-CM3
(GFDL) IPSL-CM5A-LR (IPSL), and NorESM1-M (NorESM).
The RCP 8.5 scenario was used in order to provide the upper limit of impacts for which to
plan (Riahi et al., 2011). Based on continiual political trends, socioeconomic development,
and observed climate change, this scenario looks likely to occur (Lewis, King, & Perkins-
Kirkpatrick, 2017). The decade of 2006–2015 was used as a baseline of present climatic con-
ditions, while the decade of 2050–2059 represents the future climate scenario. These periods
were chosen based on the typical lifespan of most infrastructure types and the planned tra-
jectory of long-term development projects. For each reference period, daily means of temp-
erature and precipitation were used as modeling input. Precipitation co-occuring with
negative daily temperatures is assumed to accumulate as snow. The mean of the six
models was used to represent the best estimate of the projected climate change within
each climatic grid-cell, while standard deviation was used as a measure of variability
around the mean. The spatial resolutions of the models varied, so all climate inputs were
interpolated from the original grids of varying resolution to a unified 0.5 × 0.5-degree lati-
tude/longitude grid, using an inverse distance method with a linear decay function.
Changes to permafrost temperature and active layer thickness were assessed for two scen-
arios relevant to infrastructure stability. The first case assumed that snow and vegetative
cover are removed, as would be the case for buildings, roads, and railroads. The second
POLAR GEOGRAPHY 5
case assumed that snow and little vegetative cover are present, representing close to natural
conditions, which is further used to evaluate stability of above-ground pipelines.
Permafrost geotechnical modeling
The permafrost model employed in this study was developed by Streletskiy et al. (2012a) and
is used to estimate active layer thickness and permafrost temperature based on climatic and
environmental variables, which are then used to estimate the permafrost bearing capacity.
The permafrost component of the model is based on the modified solution to the general
Stefan problem of heat conduction with a moving phase boundary and accounts for the
effects of snow cover, vegetation, and soil properties (such as texture, moisture, organic
content, and ice content). The geotechnical part of the model is based on a set of formu-
lations developed to estimate the bearing capacity of frozen soils for different common foun-
dation types as a function of permafrost temperature and maximum annual thaw
propagation. The bearing capacity calculations are based on an experimentally-derived set
of equations available from Russian construction norms and regulations for common
types of soils, and accounts for salinity, peat presence, and ground ice content.
To allow the intercomparing between regions with permafrost regardless of local con-
struction designs, this study assumed all buildings are built on piling foundations and all
pipelines are built above ground within areas of permafrost. Two common types of ‘standard
pile’designs are assumed for buildings and pipelines, following approaches in studies by Stre-
letskiy et al. (2012a), Fedorovich and Targulyan (1991), and the U.S. Department of Defence
(2004).
Permafrost degradation is commonly associated with ground subsidence. Nelson et al.
(2001) developed a settlement index model based on the relative change of the maximum
seasonal thaw depth –or ALT –and the volumetric proportion of ground ice in the soil.
This study applies the approach developed by Nelson et al. (2001), using changes in ALT
calculated within the permafrost-geotechnical model and volumetric ice content data
obtained from the International permafrost map (Brown, Ferrians, Heginbottom, & Mel-
nikov, 2002). The permafrost map classifies ground ice as low, medium, or high. These
zones are assigned quantitative ground ice volume values of 10%, 20%, and 40%, based
on the ranges established by Brown et al. In this study, ground subsidence
was calculated as the change in ALT, multiplied by the ice content within each grid
cell-calculated from the IPA permafrost map, interpolated to a 0.5 × 0.5 grid to match
the climate input. Considering that the actual distribution of ground ice is highly hetero-
geneous, and may represent a major + source of uncertainty in the results of this study,
average ice content is the best estimate of ground ice content within each grid cell.
Modeling infrastructure costs
The value of roads, railroads, pipelines, (per km), ports, and airports (per unit) was obtained
from Larsen et al. (2008) and adjusted for inflation to 2017 US dollars ($). The value of infra-
structure was assumed to remain constant throughout the timespan of the study. Country-
level variations in infrastructure value were accounted for using country-specific comparative
price levels (OECD, 2018) as multiplication factors.
Data on the value of buildings is not available in a common format across the Arctic. For
example, in Alaska, the dataset included an attributeof value, but only included public buildings.
6L. SUTER ET AL.
In Russia, highly accurate spatial information on the location of housing units and the number of
dwellings within unit is available, but does not inlude any accessible value data. In this case,
regional statistics on average household size, square meter living space per person, and the
cost per square meter were used to estimate this value (Russian Federal State Statistics
Service, 2018). In the remaining countries, a similar methodology was applied, however, build-
ings or registered addresses were assumed to be single-family residences. These assumptions
were necessary due to the spatial scale of the study and lack of uniform geospatial data on
buildings.
The changing infrastructure lifecycle replacement costs from the present period to the
future period were estimated following the approach of Larsen et al. (2008). The baseline
annual cost of replacing infrastructure was calculated by dividing each infrastructure
type’s cost per unit by each infrastructure type’s designed useful lifespan.
The adjusted annual cost was calculated using the same formula, but with each infrastruc-
ture’s useful life adjusted for climate change. The ratio used to calculate infrastructures
adjusted useful life was based on relationships between permafrost extent and projected
air temperature increases, established by Larsen et al. (2008). As the useful lifespan of infra-
structure shortens, it requires replacement more often, representing increases in costs related
to addressing regular wear and tear.
Within stressor-response models, each infrastructure type is associated with specific
environmental hazards that impact it - in this case those hazards relevant to the Arctic
context - and the cost of addressing their impacts. Damages, set equal to the replacement
cost of infrastructure, were quantified using engineering & material science validated
relationships between each infrastructure type, and the stressors associated with projected
climate change, after given thresholds are crossed. The thresholds in this study were based
offpreviously established research when available (Larsen et al., 2008; Melvin et al., 2017),
and supplemented by consultation with engineering and infrastructure experts. The
thresholds were set to account for common safety factors in engineering design. If a threshold
was crossed, it was assumed that infrastructure would have to be repaired or replaced to
maintain functionality. In more detailed regional studies, these thresholds should be
further adapted to meet regional or local construction codes.
Costs were first analyzed on the national level, and further analyzed on the regional level
in particularly affected countries. On the regional level, costs were framed against measures of
economic prosperity, including gross regional product (GRP) and governmental spending. In
the absence of projected economic data for the study regions, GRP data for 2016
was assumed to remain constant. GRP for the US (Alaska), Canada, and Russian regions was
obtained from the national statistical databases. Government spending data for Alaska was refer-
enced from the National Association of State Budget Officers State Expenditures Reports for
2016 (NASBO, 2016).
Results
Value of infrastructure
The total value of six types of infrastructure assets within the study area is about $146 billion.
Russia, Norway, Canada, and US (Alaska) account for over 90% of infrastructure assets.
Norway has the highest asset value though their quantity of buildings on permafrost is rela-
tively minor. The high values of the Scandinavian countries are partially due to the extremely
POLAR GEOGRAPHY 7
high OECD cost-levels. Among the six infrastructure categories, buildings and structures
account for 67% of the total value of infrastructure assets, while roads account for an
additional 17%. Railroads and pipelines account for about 5% each, while airports and
ports combine for another 5%.
The distribution of infrastructure types varies. In Norway, buildings account for 90% of
national infrastructure asset value. However, building foundations in Norway are typically
anchored in bedrock and thus are less impacted by permafrost degradation. However, it is
not possible to filter these out of the GIS datasets, given the scale and quality of the data.
Norway is thus excluded from the regional analysis, based on expert opinion. On the
other hand, in Russia buildings account for 70% of infrastructure value, and it is typical
for buildings in Russia’s urban Arctic environments to have foundations anchored in non-
bedrock permafrost (Shiklomanov et al., 2017a). Buildings account for a further 58% of infra-
structure value in Alaska and 41% in Canada.
Roads account for about 10–17% of assets within Russia, the US (Alaska), and Sweden, yet
represent a more signficant proportion of infrastructure in Canada (37%), Finland (68%),
and Iceland (81%). Canada contains almost 42% of the total road assets in the study area,
while Russia accounts for 64% of pipeline assets in the study area. Russia also contains by
far the largest rail network in the Arctic.
The importance of airports and ports is clear in the context of Alaska and Canada. By
value, Alaska contains about 44% of circumpolar airport assets and Canada about 36%.
Within the US (Alaska), airports represent about 16% of total infrastructure assets; in all
the other countries, the share of the value of airports to total assets is under 10%. For
ports, Canada and Norway have the largest share, holding 23% and 32% of circumpolar
assets respectively.
Projected climate change in the Arctic
There is significant variability between the model projections of climate change at the cir-
cumpolar scale. The mean temperature increase derived from the combination of six
models employed is 3.6° C. The GFDL model shows the most pronounced warming of up
to 5°C across the study area, while the CSIRO model is the most conservative with 2.2 °C
projected across the study area. Overall, the six models employed are in greater agreement
regarding subarctic regions with maritime climates and areas with sporadic to discontinuous
permafrost. The model ensemble has a higher variability in areas with continental climate,
characterized by continuous permafrost and mountainous regions.
The largest temperature increases (Figure 2) are projected to occur over the interior of the
Siberian continental landmass, Northern Canada, and North/Central Alaska. Among these
regions, all models but CSIRO agree that Siberia will be most impacted, while CSIRO projects
the most warming in Alaska and Northern Canada.
The change in average annual precipitation (Figure 2) increase across the ensemble is
72 mm, but the range is considerable: some areas show slight drying (up to −34 mm)
while other areas show significant wetting, up to + 226 mm. The GFDL model predicted
the highest precipitation increases with a mean of 72 mm across the study area, with
regions of the Pacific sector, such as Chukotka, Alaska, and Yukon showing increases
greater than 200 mm. CSIRO is the most conservative model, projecting a 37 mm increase
(−149 mm / 271 mm) across the study area.
8L. SUTER ET AL.
Figure 2. Model outputs showing changes in Mean Annual Air Temperature and Precipitation between
2005–2010 and 2050–2059, under RCP8.5.
POLAR GEOGRAPHY 9
Changes in permafrost characteristics
Permafrost temperature and active layer thickness
In all models, there is strong agreement that permafrost will warm significantly across much
of the Arctic. In natural conditions - where snow and vegetation remain undisturbed -
ground temperature warming is expected to be less intense, increasing 3.7°C (1.3°/5°C) –
about 0.1°C lower than the scenario with snow removal.
Under conditions where snow and vegetation are absent, such as under buildings, the
ensemble mean projects an increase of 3.8°C, ranging from 1.6° to 6.5°C. All models show
by consensus that the most significant warming will take place in the High Arctic, which
is characterized by continuous permafrost. In the Subarctic, where discontinuous permafrost
is more common, the warming trend is less pronounced.
A negligible area of permafrost coverage in Iceland is projected to cool slightly by CSIRO,
which is the least pessimistic model. The CSIRO model (2.34°C; −1.1°/6.4°C) projects
warming to remain below 2°C across most of Siberia. Even this conservative model,
however, projects increases of up to 4°C in Alaska and Canada.
Given natural conditions, the ensemble mean predicts that ALT will decrease by 0.31 m
(−0.56 m/0.80 m), generally 0.1 m less than the scenario with snow and vegetation
removal (0.41 m; −0.78 m/0.76 m). Negative values indicate the loss of near surface perma-
frost during the time period of the study, and a shrinking of the seasonal freezing layer
(Melvin et al., 2017). The variability amongst the models is comparable across the Arctic,
with only small portions of central Alaska and Chukotka seeing a standard deviation
greater than 0.4 m.
Without snow, the active layer is expected to thicken significantly throughout the Arctic.
The ensemble mean predicts an increase in ALT of 0.41 m across the Arctic, with a range of
−.78 m to 0.76 m. The models generally agree on the spatial distribution of changes in ALT.
All the models agree that the greatest increases will occur in continental Siberia, northeast
Alaska, and Yukon. In southwest Alaska, near surface permafrost is expected to disappear
entirely in some models.
Loss of bearing capacity and ground subsidence
The ensemble mean reduction of bearing capacity under the natural scenario is 41%, with a
maximum of 69% reduction. The models generally agree on the spatial distribution of
changes, with the edges of permafrost zones more significantly impacted, and pockets of rela-
tive stability within central Siberia. Consensus is strongest within the continental interiors,
while higher standard deviations are notable near the edges of the study area.
The bearing capacity of soil around buildings –where vegetation is removed through
paving and snow is regularly cleared –is projected to decrease more than in the natural scen-
ario. The ensemble mean indicates a 43% loss of soil bearing capacity across the study area.
The maximum losses are projected around the southern limits of the permafrost zone. The
consensus between models in this case is relatively strong. Standard deviation is under 15%
for much of the study area, especially in the central areas of Siberia, Canada, and Alaska.
Yukon, Yamal-Nenets, and the southern Russian permafrost regions are particularly
impacted by bearing capacity losses (Figure 3), while in central Siberia, permafrost is pre-
dicted to remain slightly more stable. The highest projected decreases are in Eastern
Siberia, Northwest Alaska, Northeast Canada, and Chukotka. Around the edges of the
study area, in southwest Alaska and southern Russia, there is less model agreement due to
10 L. SUTER ET AL.
the variability in predictions of near-surface permafrost disappearance. In these areas, stan-
dard deviations above 40% are noted.
Given the natural scenario, significant portions of territory currently occupied by near-
surface permafrost –especially those in Alaska –are expected to become permafrost free.
The model ensemble mean ground subsidence is projected to be 0.1 m, with some regions
above 0.3 m. These changes are most intense in the areas of the highest
predicted subsidence, where large increases in ALT overlap with areas of high ground
ice content. There is greater agreement towards the edges of the study area, indicated
by a standard deviation lower than 0.1 m. Only in parts of central Alaska and
Russia does this standard deviation exceed 0.1 m (Figure 4).
Lifecycle replacement costs
The lifecycle replacement costs to maintain Arctic infrastructure (Table 2) are projected to
increase by 27.7%, or about $15.47 billion, by 2050–2059. Pipelines are projected to be the
most impacted infrastructure relative to baseline lifecycle costs, with an increase of over
60% by 2059. Significant cost increases, of over 40%, are projected for roads, railways, airports,
and ports. Building lifecycle costs are only projected to increase by about 12%, due to the long
intended lifespan of this infrastructure. Roads account for the largest share of all increased life-
cycle costs, at 39% of the total, while costs related to building maintenance account for a further
22% of total costs. The increased lifecycle replacement costs for pipelines, railroads, and air-
ports are $1.83 billion, $1.60 billion, and $2.05 billion, respectively, accountinig for 35% of
Figure 3. Model outputs for change in bearing capacity between 2005–2010 and 2050–2059, under
RCP8.5.
POLAR GEOGRAPHY 11
the total increase. Costs associated with ports are projected to account for the remaining 4% of
lifecycle cost increases. However, damage to ports is underestimated in this study, as the model
could not account for the environmental stressor of coastal erosion.
Figure 4. Model outputs for ground subsidence between 2005–2010 and 2050–2059, under RCP8.5.
Table 2. Baseline and climate forced lifecycle replacement costs by 2059, for infrastructure types
& countries.
Category
Baseline
Lifecycle Replacement Costs
Costs with Climate
Forcing Difference
Percent Increase from Climate
Change
Infrastructure ($ Millions) ($ Millions) ($ Millions) %
Roads (Paved) $15,055.97 $21,142.97 $6087.00 40.4%
Rail $3978.49 $5576.98 $1598.50 40.2%
Pipelines $3001.90 $4827.56 $1825.65 60.8%
Buildings $27610.02 $30948.61 $3338.58 12.1%
Airports $4951.72 $6999.09 $2047.37 41.3%
Ports $1340.23 $1913.52 $573.29 42.8%
Total $55,938.34 $71,408.73 $15,470.39 27.7%
Country ($ Millions) ($ Millions) ($ Millions) %
Canada $12,865.02 $17,190.46 $4325.44 33.6%
Russia $24,085.29 $30,716.65 $6631.36 27.5%
United States $7061.91 $9620.59 $2558.68 36.2%
Norway $10,718.64 $12,313.67 $1595.03 14.9%
Sweden $780.19 $988.47 $208.28 26.7%
Finland $381.50 $517.22 $135.72 35.6%
Iceland $45.79 $61.67 $15.87 34.7%
Total $55,938.34 $71,408.73 $15,470.39 27.7%
12 L. SUTER ET AL.
Russia incurred the greatest projected increases in lifecycle costs –about 43% of the total.
To mitigate the impacts of increased wear and tear, Russia may have to spend $6.63 billion
for lifecycle replacement by 2059, a 27.5% increase relative to their baseline. Canada is pro-
jected to incur $4.33 billion during the same time period, which represents a 33.6% increase
from baseline lifecycle replacement costs and 28.0% of total increased costs. Alaska and
Norway incur $2.56 billion and $1.60 billion in increased maintenance costs, respectively.
Finland and Iceland are projected to incur much lower total costs, but these still represent
35% increases, relative to the baseline.
Impacted infrastructure value
The value of damaged infrastructure (Table 3) across the entire Arctic is $21.6 billion dollars
by the begining of 2059. This damage represents about 15% of the total value of infrastruc-
ture assets included in this study. The impacts are projected to be highest in Russia, where
about 32% of infrastructure is estimated to be impacted. Comparatively, about one fifth of
total assets are estimated to be affected by in the US (Alaska) (22%) and Canada (19%).
Within the other Arctic countries, no substantial damages related to permafrost thaw and
precipitation increases were found.
Within each infrastructure category, about 15%–20% of assets are projected to be impacted,
except for airports, where about 26% of assets are projected to experience damages due to
climate change. In Russia buildings are at the greatest risk, while in North America
mainly roads and airports are impacted. This reflects historic development trends, with
Russia’s concentrated population in urban areas, while Canada and Alaska are characterized
by smaller, more dispersed settlements.
Regional costs
On a regional level, the lifecycle replacement and stressor-response models indicated that
Sakha Republic, Alaska, Yukon, Northwest Territories, Krasnoyarsk Krai, and Yamal-
Nenets AO are the most impacted. By 2059, Sakha Republic and Alaska are each expected
Table 3. Share of assets with climate change impacts by 2059, for infrastructure types & countries.
Category Infrastructure Total Asset Value ($ Millions) Percent of Total Assets Impacted by 2059 %
Roads (Paved) $25,027.67 18.92%
Rail $7663.67 12.30%
Pipelines $7553.11 15.02%
Buildings $98,179.46 13.29%
Airports $5878.75 25.89%
Ports $1630.50 13.65%
Total $145,933.17 14.80%
Country ($ Millions) %
Canada $27,977.44 18.75%
Russia $40,339.14 31.87%
United States $16,044.38 21.79%
Norway $51,760.96 0.01%
Sweden $3796.27 0.00%
Finland $2657.83 0.00%
Iceland $3357.15 0.00%
Total $145,933.17 14.80%
POLAR GEOGRAPHY 13
to incur over $6 billion in combined repair and replacement costs, while Yamal-Nenets is
expected to incur about $4 billion. Yukon and Krasnoyarsk Kari are projected to incur
$3.9 billion and $3.2 billion, respectively (Figure 5).
Costs relative to gross regional products
The ability of regional budgets to absorb these increased costs is tied to their economic pros-
perity, which is often measured in gross regional product (GRP). The mean annual costs to
address increased lifecycle replacement costs and direct damages due to climate change
exceed 3.7% of annual GRP in Yukon. This figure lies at 1.5% and 1% of GRP for the North-
west Territories and Nunavut. Within Russia, Magadan, Sakha Republic, and Zabaykalsky
Krai are particularly impacted, with mean annual combined costs accounting for between
1% and 1.4% of 2016 GRP. Yamal-Nenets, Komi Republic, and Krasnoyarsk are also
impacted, with about 0.3% of GRP needed to address mean annual cost increases (Table 4).
Figure 5. Total lifecycle costs & damages to infrastructure by 2060, by region.
Table 4. Annual Total Costs till 2059 as a Percent of Annual GRP (2016).
Regions (< 0.01% of GRP) Country Annual Total Costs till 2059 as Percent of Annual GRP (2016)
Yukon Canada 3.73%
Northwest Territories Canada 1.46%
Zabaykalsky Russia 1.41%
Nunavut Canada 1.06%
Sakha (Yakutia) Russia 1.05%
Magadan Russia 0.98%
Gorno-Altay Russia 0.38%
Nenets Russia 0.37%
Alaska USA 0.31%
Yamal-Nenets Russia 0.31%
Republic of Buryatia Russia 0.29%
Komi Republic Russia 0.28%
Krasnoyarsk Russia 0.28%
Chukotka Russia 0.25%
14 L. SUTER ET AL.
Discussion
These ratios may seem low, but the sums become consequential when considered in relation
to government spending, or the contribution of individual industries to GRP. For example,
the cumulative mining sector in Yukon contributed about 11% to GRP in 2015 (Yukon
Department of Economic Development, 2016). Addressing the impacts of climate change
may cost the region the equivalent of a third of total mining output. In a region of high
costs and small profit margins, these increased costs are significant.
In Russia, government spending accounted for 18% of 2016 GDP (Global Economy,
2018). The 1% of GRP incurred to address climate impacts thus represents almost 6% of
the government budget, which implies that the projected share of annual GRP going
towards mitigating climate damages in Sakha, Zabaykalsky, and Magadan represents a sig-
nificant portion of the government budget.
In Alaska, the mean annual costs of addressing some of these climate impacts registered as
about 0.3% of 2016 GRP, which was about $50 billion. In relation, government spending by
the state of Alaska in 2016 amounted to about $10 billion (NASBO, 2016), representing 20%
of GRP. This indicates that mean annual costs could account for close to 1.5% of annual
state government spending. This is more than Alaskan state government spending on
all public assistance (food programs, assistance for needy families) for fiscal year 2016,
which accounted for 1.4% of state budget expenditures (NASBO, 2016). This comparison
of fiscal impacts to regional budgets may present a clearer picture of the effects of climate
damages in a region, as GRP includes resource wealth that is moved out of the region.
Comparison to previous studies
As a comparison to this study, Larsen et al. (2008) predicted a 10% to 20% increase in infra-
structure lifecycle costs for Alaska by 2030, and 10% to 12% by 2080. This study estimates a
36% increase for the same region by 2059, almost triple the increase predicted by Larsen
et al. On a circumpolar level, the lifecycle replacement model predicted a 27% increase in
upkeep costs. The infrastructure stressor-response model used in this study predicted $3.5
billion of climate damages in Alaska, which corresponds well with the Melvin et al. (2017)esti-
mates for the region, which projected $5.7 billion (USD 2017) of direct damages in Alaska by
2100. The variability in results compared to the aforementioned studies is likely a result of the
differing time periods, GCM ensembles employed, and differences in the infrastructure inven-
tories used.
Within Russia, Streletskiy et al. (2019) estimated that costs related to permafrost thaw
impacts would amount to over $100 billion, vastly surpassing the $19.5 billion estimated in
this study. The infrastructure inventory vastly underestimated the value of infrastructure in
the Russian Arctic. The Streletskiy et al. (2019) study utilized Russian Federal Statistical Services
data, which provided a monetary breakdown of total infrastructure assets by region. This gov-
ernmental data indicated $249 billion of fixed infrastructure in Russian permafrost regions,
while the six types of infrastructure conglomerated for this study only estimated $40 billion.
This indicates that there remain significant gaps in the availability of Arctic infrastructure
data, especially in areas where governmental geospatial data is difficult or impossible to obtain.
Further studies estimated higher indirect costs of climate change, specifically within
Russia. Porfiriev et al. (2017) based their estimates on two previous studies on the indirect
economic impacts of climate change (Hope & Schaefer, 2016; Schuur et al., 2015). These
POLAR GEOGRAPHY 15
indirect costs, such as the release of methane into the atmosphere from thawing permafrost,
contributed significantly to the projected costs, adding $160 trillion by 2100, or about $1.9
trillion annually across the Arctic region. This could cost some Russian Arctic regions
upwards of 5 to 6% of annual GRP to address (Porfiriev et al., 2017).
Hjort et al. (2018) employed a hazard-risk model to produce a high-resolution circumpo-
lar geohazard map, assesing how changes in circumpolar permafrost conditions could impact
infrastructure. The areas of highest risk identified in that study correlate well to the regions of
highest-cost identified in this study, as Alaska, Yukon, YNAO, and Sakha (Yakutia)
were among the most directly impacted. However, this study shows that some of the
highest risk regions also contain the strongest economies (as measured by GRP), which
may help them offset some mitigation costs.
Further considerations
There are factors not resolved by this study, which are difficult to model at its scale; these
includes national construction codes, maintenance standards, and their associated costs.
Moreover, the model employed in this study focuses mainly on permafrost and does not inte-
grate several important environmental variables, such as coastal erosion (Fritz, Vonk, &
Lantuit, 2017; Melvin et al., 2017), frost-heaving (Abdalla, Fan, McKinnon, & Gaffard,
2015; Melvin et al., 2017), and hydrology (Instanes et al., 2016). Furthermore, the study is
unable to resolve the role of non-climatic factors, which have significant influence on infra-
structure stability and sustainability; these factors include maintenance capacity, social
capital, and political authority, all of which are important for infrastructure stability and
the ability to mitigate future climate impacts (Grebenets et al., 2012; Shiklomanov et al.,
2017a).
The results also reveal differences between the climate inputs used between the models,
which reflects the uncertainty inherent in most climate models (Räisänen, 2007).
There is significant spatial variability between the model results at the circumpolar scale.
Northeastern Siberia, the Canadian Arctic Archipelago, and Iceland are areas with relatively
high uncertainty in air temperature projections, while in Western Russian Arctic and Alaska,
the models show greater agreement. There is even more considerable uncertainty in the pro-
jections for precipitation. The improved climate models from upcoming releases, such as
CMIP6, could be used to improve this analysis (Eyring et al., 2016). The uncertainty in
climate projections and the coarse parametrization of soil characteristics, including
ground ice content, are major sources of uncertainty in the permafrost model outputs
used to calculate bearing capacity and ground subsidence.
With these considerable limitations in mind, the results of this study should be interpreted
with caution. However, the results provide a valuable estimate of the distribution of perma-
frost thaw related impacts on infrastructure at the circumpolar scale, as well as an analysis of
the costs related to those impacts at the national and regional scales. The results show that
there are several regions of the Arctic, including Yukon, Alaska, YNAO, and the Sakha
Republic (Yakutia) where more granular assessments should be conducted.
By modeling climate change impacts using economics, this study aims to scaffold more
effective policy and business advice metrics. These metrics are intended to guide sustainable
policy in a way that speaks to stakeholders who are not typically engaged with climate
science. Some academics, policymakers, and private industries have expressed optimism
about future resource production considering climate change, especially within Russia
16 L. SUTER ET AL.
(Anisimov & Reneva, 2006). As the Arctic is already a high-priced region, any additional
costs could impact budgets and fiscal priorities (Streletskiy et al., 2019). Estimates of the
costs of climate change should be considered by planners and developers interested in
expanding into the Arctic’s emerging market, to anticipate potential economic outcomes
and to ensure that private industries and governments alike are prepared to address the
costs of climate change.
Acknowledgements
Thanks to faculty and peers at the Department of Geography at The George Washington University,
Moscow State University, and many other for their useful inputs. A personal thanks to
Professor Robert Orttung for his support, and to Kate O’Brien for her invaluable time spent
reading and revising. Many thanks to all reviewers and editors for their constructive and insightful
comments.
Disclosure statement
No potential conflict of interest was reported by the authors.
Funding
This research was supported by U.S. National Science Foundation (NSF) grants ICER-1558389 and
1717770 ’Belmont Forum Collaborative Research: ARCTIC-ERA: ARCTIC climate change and its
impact on Environment, infrastructures and Resource Availability’, and OISE-1545913 ’PIRE: Pro-
moting Urban Sustainability in the Arctic’.The analysis of the Russian Arctic was funded by RFBR
project 18-05-60088 ’Urban Arctic resilience in the context of climate change and socio-economic
transformations’. The opinions, findings, conclusions, and recommendations expressed in this
paper are those of the authors, and do not necessarily reflect the views of NSF or RFBR.
References
Aalto, J., Harrison, S., & Luoto, M. (2017). Statistical modelling predicts almost complete loss of major
periglacial processes in Northern Europe by 2100. Nature Communications,8(1), 515. Retrieved
from https://www.nature.com/articles/s41467-017-00669-3.
Aalto, J., Karjalainen, O., Hjort, J., & Luoto, M. (2018). Statistical forecasting of current and future
circum-Arctic ground temperatures and active layer thickness. Geophysical Research Letters,45
(10), 4889–4898.
Abdalla, B. A., Fan, C., McKinnon, C., & Gaffard, V. (2015). Numerical evaluation of cyclic frost heave
growth in arctic pipelines and mitigation strategies. In OTC Arctic technology conference, 23–25
March, Copenhagen, Denmark.
Andersland, O. B., & Ladanyi, B. (2003). Frozen ground engineering (2nd ed.). Hoboken, NJ: Wiley.
Anisimov, O., Kokorev, V., & Zhil’tsova, Y. (2013). Temporal and spatial patterns of modern climatic
warming: Case study of Northern Eurasia. Climatic Change,118(3–4), 871–883.
Anisimov, O., & Reneva, S. (2006). Permafrost and changing climate: The Russian perspective.
AMBIO: A Journal of the Human Environment,35(4), 169–175.
Arctic Monitoring and Assessment Programme (AMAP). (2017). Snow, water, ice and permafrost in
the Arctic (SWIPA): Climate change and the cryosphere. Oslo: Author.
Arndt, C. (2001). Information measures: Information and its description in science and engineering.
Berlin, DE: Springer Science & Business Media.
Biskaborn, B. K., Smith, S. L., Noetzli, J., Matthes, H., Vieira, G., Streletskiy, D. A., …Allard, M. (2019).
Permafrost is warming at a global scale. Nature Communications,10(1), 264. Retrieved from https://
www.nature.com/articles/s41467-018-08240-4.
POLAR GEOGRAPHY 17
Brown, J., Ferrians, O. Jr., Heginbottom, J., & Melnikov, E. (2002). Circum-Arctic map of permafrost
and ground-ice conditions. Version 2. Boulder, CO: National Snow and Ice Data Center. Retrieved
from http://nsidc.org/data/GGD318
Budzik, P. (2009). Arctic oil and natural gas potential. US Energy Information Administration, Office
of Integrated Analysis and Forecasting, Oil and Gas Division. Retrieved from http://www.akleg.gov/
basis/get_documents.asp?session=28&docid=741
Chinowsky, P., & Arndt, C. (2012). Climate change and roads: A dynamic stressor–response model.
Review of Development Economics,16(3), 448–462.
Circumpolar Arctic Vegetation Mapping (CAVM) Team. (2003). Circumpolar Arctic vegetation
map.(1: 7,500,000 scale), conservation of Arctic flora and fauna (CAFF) map no. 1. US Fish and
Wildlife Service, Anchorage, Alaska. Retrieved from https://www.geobotany.uaf.edu/cavm.
Clarke-Sather, A., Crow-Miller, B., Banister, J. M., Anh Thomas, K., Norman, E. S., & Stephenson, S. R.
(2017). The Shifting Geopolitics of Water in the Anthropocene. Geopolitics,22(2), 332–359.
Dore, M. H. I., & Burton, I. (2001). Costs of adaptation to climate change in Canada: A stratified esti-
mate by sectors and regions: Social infrastructure. St. Catharines, ON: Brock University.
Environmental Protection Agency (EPA). (2017). Multi-model framework for quantitative sectoral
impacts analysis: A technical report for the fourth national climate assessment. U.S.
Environmental Protection Agency, EPA 430-R-17-001.
Eyring, V., Bony, S., Meehl, G. A., Senior, C. A., Stevens, B., Stouffer, R. J., & Taylor, K. E. (2016).
Overview of the coupled model intercomparison project phase 6 (CMIP6) experimental design
and organization. Geoscientific Model Development,9(5), 1937–1958.
Fedorovich, D. I., & Targulyan, Y. O. (1991). Pile foundation construction characteristics in the Alaska
oil pipeline. Soil Mechanics and Foundation Engineering,28(5), 204–207.
Fritz, M., Vonk, J. E., & Lantuit, H. (2017). Collapsing Arctic coastlines. Nature Climate Change,7(1),
6–7.
Frolov, I. E. (2015). Development of the Russian Arctic zone: Challenges facing the renovation of
transport and military infrastructure. Studies on Russian Economic Development,26(6), 561–566.
Global Economy. (2018). Russia: Government spending, percent of GDP. The Global Economy.
Retrieved from https://www.theglobaleconomy.com/Russia/Government_size.
Gramelsberger, G., & Feichter, J. (Eds.). (2011). Climate change and policy: The calculability of climate
change and the challenge of uncertainty. Berlin: Springer Science & Business Media.
Grebenets, V., Streletskiy, D., & Shiklomanov, N. (2012). Geotechnical safety issues in the cities of
polar regions. Geography, Environment, Sustainability Journal,5(3), 104–119.
Heleniak, T., & Bogoyavlensky, D. (2015). Arctic populations and migration. In J. N. Larsen & G.
Fondahl (Eds.), Arctic human development report: Regional processes and global linkages (pp.
53–104). Dordrecht: Springer.
Hjort, J., Karjalainen, O., Aalto, J., Westermann, S., Romanovsky, V. E., Nelson, F. E., …Luoto, M.
(2018). Degrading permafrost puts Arctic infrastructure at risk by mid-century. Nature
Communications,9(1), 5147. Retrieved from https://www.nature.com/articles/s41467-018-07557-
4#citeas.
Hope, C., & Schaefer, K. (2016). Economic impacts of carbon dioxide and methane released from
thawing permafrost. Nature Climate Change,6(1), 56–59.
Hossain, K. (2017). Arctic melting: A new economic frontier and global geopolitics. Current
Developments in Arctic Law,5,40–45.
Husebekk, A. R., Andersson, M., & Penttilä, R. E. (2015). Growth from the North. How can Norway,
Sweden and Finland achieve sustainable growth in the Scandinavian Arctic? Helsinki: Prime
Minister’sOffice Publications.
Huskey, L., Mäenpää, I., & Pelyasov, A. (2015). Economic Systems. In J. N. Larsen & G. Fondahl (Eds.),
Arctic human development report: Regional processes and global linkages (pp. 151–184). Dordrecht:
Springer.
Instanes, A., Kokorev, V., Janowicz, R., Bruland, O., Sand, K., & Prowse, T. (2016). Changes to fresh-
water systems affecting Arctic infrastructure and natural resources. Journal of Geophysical Research:
Biogeosciences,121(3), 567–585.
Klüter, H. (2000). Der Norden Russlands –vom Niedergang einer Entwicklungsregion. Geographische
Rundschau,52(12), 12–20.
18 L. SUTER ET AL.
Larsen, P. H., Goldsmith, S., Smith, O., Wilson, M. L., Strzepek, K., Chinowsky, P., & Saylor, B. (2008).
Estimating future costs for Alaska public infrastructure at risk from climate change. Global
Environmental Change,18(3), 442–457.
Larsen, J. N., & Huskey, L. (2015). The Arctic economy in a global context. In B. Evengård, L. Nymand,
& Ø. Paasche (Eds.), The New Arctic (pp. 159–174). Cham: Springer International Publishing.
Laruelle, M. (2015). Russia’s arctic strategies and the future of the Far North. New York: Routledge.
Lewis, S. C., King, A. D., & Perkins-Kirkpatrick, S. E. (2017). Defining a new normal for extremes in a
warming world. Bulletin of the American Meteorological Society,98(6), 1139–1151.
Melvin, A. M., Larsen, P., Boehlert, B., Neumann, J. E., Chinowsky, P., Espinet, X., …Nicolsky, D. J.
(2017). Climate change damages to Alaska public infrastructure and the economics of proactive
adaptation. Proceedings of the National Academy of Sciences,114(2), E122–E131.
Monmonier, M. (2008). Coast lines: How mapmakers frame the world and chart environmental change.
Chicago: University of Chicago Press.
National Association of State Budget Officers (NASBO). (2016). State expenditure report. Author.
National Snow and Ice Data Center (NSIDC). (2018). What is the Arctic? Author. Retrieved from
https://nsidc.org/cryosphere/arctic-meteorology/arctic.html
Nelson, F. E., Anisimov, O. A., & Shiklomanov, N. I. (2001). Subsidence risk from thawing permafrost.
Nature,410(6831), 889–890.
Organisation for Economic Co-operation and Development (OECD). (2018). Monthly Comparative
Price Levels. Author. Retrieved from https://stats.oecd.org/Index.aspx?DataSetCode=CPL
O’Rourke, R. (2019). Changes in the Arctic: Background and issues for congress (pp. 7–
5700). Washington, DC: Congressional Research Service.
Orttung, R. W. (2016). Sustaining Russia’s Arctic cities: Resource politics, migration, and climate change
(Vol. 2).New York: Berghahn Books.
Porfiriev, B. N., Voronina, S. A., Semikashev, V. V., Terentev, N. E., Eliseev, D. O., & Naumova, Y. V.
(2017). Consequences of climate change for economic Growth and development of individual
Sectors of the economy of the Russian Arctic. Arctic: Ecology and Economics,4(28), 4–17.
Räisänen, J. (2007). How reliable are climate models? Tellus A: Dynamic Meteorology and
Oceanography,59(1), 2–29.
ReformaGHK. (2018). My House Database. State Corporation: Fund for Assistance in Reforming
Housing and Communal Services. Retrieved from https://www.reformagkh.ru/myhouse
Riahi, K., Rao, S., Krey, V., Cho, C., Chirkov, V., Fischer, G., …Rafaj, P. (2011). RCP 8.5—A scenario
of comparatively high greenhouse gas emissions. Climatic Change,109(1–2), 33–57.
Romanovsky, V. E., Smith, S. L., Shiklomanov, I., Streletskiy, D. A., Isaksen, K., A, N., …Marchenko,
S. S. (2017). Terrestrial permafrost [in “state of the climate in 2016”]. Bulletin of the American
Meteorological Society,98(8), 147–149.
Russian Federal State Statistics Service (Rosstat). (2018). Database of indicators of municipalities.
Retrieved from https://gks.ru/free_doc/new_site/bd_munst/munst.htm
Schuur, E. A., McGuire, A. D., Schädel, C., Grosse, G., Harden, J. W., Hayes, D. J., …Natali, S. M.
(2015). Climate change and the permafrost carbon feedback. Nature,520(7546), 171–179.
Schweikert, A., Chinowsky, P., Espinet, X., & Tarbert, M. (2014). Climate change and infrastructure
impacts: Comparing the impact on roads in ten countries through 2100. Procedia Engineering,
78, 306–316.
Shiklomanov, N., Anisimov, O., Zhang, T., Marchenko, S., Nelson, F., & Oelke, C. (2007). Comparison
of model-produced active layer fields: Results for northern Alaska. Journal of Geophysical Research,
112,F02S10. Retrieved from https://agupubs.onlinelibrary.wiley.com/doi/pdf/10.1029/
2006JF00057129/2006JF000571.
Shiklomanov, N. I., Streletskiy, D. A., Grebenets, V. I., & Suter, L. (2017a). Conquering the permafrost:
Urban infrastructure development in Norilsk, Russia. Polar Geography,40(4), 273–290.
Shiklomanov, N. I., Streletskiy, D. A., Swales, T. B., & Kokorev, V. A. (2017b). Climate change and
stability of urban infrastructure in Russian permafrost regions: Prognostic assessment based on
GCM climate projections. Geographical Review,107(1), 125–142.
Stephenson, S. R., & Pincus, R. (2018). Challenges of sea-ice prediction for Arctic marine policy and
planning. Journal of Borderlands Studies,33(2), 255–272.
POLAR GEOGRAPHY 19
Stephenson, S. R., Smith, L. C., & Agnew, J. A. (2011). Divergent long-term trajectories of human
access to the Arctic. Nature Climate Change,1(3), 156–160.
Streletskiy, D. A., Biskaborn, B. K., Nötzli, J., Lanckman, J.-P., Romanovsky, V. E., Schoeneich, P., …
Zhao, L. (2017). Permafrost thermal state [in “state of the climate in 2016”]. Bulletin of the American
Meteorological Society,98(8), 19–21.
Streletskiy, D. A., Sherstiukov, A. B., Frauenfeld, O. W., & Nelson, F. E. (2015). Changes in the 1963–
2013 shallow ground thermal regime in Russian permafrost regions. Environmental Research
Letters,10(12), 125005. Retrieved from https://iopscience.iop.org/article/10.1088/1748-9326/10/
12/125005.
Streletskiy, D. A., Shiklomanov, N. I., Little, J. D., Nelson, F. E., Brown, J., Nyland, K. E., & Klene, A. E.
(2016). Thaw subsidence in undisturbed Tundra Landscapes, Barrow, Alaska, 1962–2015.
Permafrost and Periglacial Processes,28(3), 566–572.
Streletskiy, D. A., Shiklomanov, N. I., & Nelson, F. E. (2012a). Permafrost, infrastructure, and climate
change: A GIS-based landscape approach to geotechnical modeling. Arctic, Antarctic, and Alpine
Research,44(3), 368–380.
Streletskiy, D. A., Shiklomanov, N. I., & Nelson, F. E. (2012b). Spatial variability of permafrost active-
layer thickness under contemporary and projected climate in northern Alaska. Polar Geography,35,
95–116.
Streletskiy, D. A., Suter, L. J., Shiklomanov, N. I., Porfiriev, B. N., & Eliseev, D. O. (2019). Assessment
of climate change impacts on buildings, structures and infrastructure in the Russian regions on per-
mafrost. Environmental Research Letters,14(2), 025003. Retrieved from https://iopscience.iop.org/
article/10.1088/1748-9326/aaf5e6.
Tol, R. S. (2018). The economic impacts of climate change. Review of Environmental Economics and
Policy,12(1), 4–25.
U.S. Department of Defence. (2004). Foundations of structures: Arctic and subarctic construction.
Author. Retrieved from https://www.wbdg.org/FFC/DOD/UFC/INACTIVE/ufc_3_130_04_2004.
pdf
Wihbey, J., & Ward, B. (2016). Communicating about climate change with journalists and media pro-
ducers. Oxford Research Encyclopedia of Climate Science. Retrieved from https://oxfordre.com/
climatescience/view/10.1093/acrefore/9780190228620.001.0001/acrefore-9780190228620-e-407
World Climate Research Programme (WCRP). (2011). Coupled model intercomparison project –phase 5.
Special Issue of the CLIVAR Exchanges Newsletter, No. 56, Vol. 15, No. 2.
Yukon Department of Economic Development. (2016). Yukon Government: Whitehorse, Yukon.
Retrieved from http://www.finance.gov.yk.ca/pdf/YukonEconomicOutlook_May2016.pdf.
20 L. SUTER ET AL.
