Silberman and Rees (2010)

Applied Geography 30 (2010) 36–49
Contents lists available at ScienceDirect
Applied Geography
journal homepage: www.elsevier.com/locate/apgeog
Reinventing mountain settlements: A GIS model for identifying
possible ski towns in the U.S. Rocky Mountains
Jordan A. Silberman a, Peter W. Rees b, *
a
b
P.O. Box 391, Unionville, PA 19375, USA
Department of Geography, University of Delaware, 222 South Chapel Street, Newark, DE 19716, USA
a b s t r a c t
Keywords:
Ski towns
Resort development
Rocky Mountains
GIS
Former mining and ranching settlements in the U.S. Rocky Mountains frequently seek to
reinvent themselves as the industries that created them have declined. Redevelopment as
ski resorts is a common strategy that can successfully revive the economies of mountain
settlements but this approach, if undertaken without careful advanced planning, can also
have negative consequences, damaging fragile alpine environments, overwhelming social
and housing services and distorting local economies. This study develops a GIS-based
model that follows a systematic sequential elimination procedure to identify those Rocky
Mountain settlements most likely to be attractive to ski resort development, based on the
location criteria of existing ski areas. Results show that while no single settlement is an
obvious candidate for development by the ski industry, a number of places are contenders
in a way that can be systematically measured and evaluated. Moreover, the methodology
used can be applied to additional areas subject to winter sports growth worldwide.
Ó 2009 Elsevier Ltd. All rights reserved.
Introduction
The location of ski resorts in mountain communities often produces major economic, environmental and cultural consequences
(Chipeniuk, 2005; Gill, 2000; Kariel,1989; Lasanta, Laguna, & Vicente-Serrano, 2007; Orens & Seidl, 2009; Purdue, 2004; Riebsame,
Gosnell, & Theobald, 1996; Theobald, Gosnell, & Riebsame, 1996). Some communities may possess local entrepreneurs who
actively organize their appeal to developers (Miller & Blevins, 2005). Alternatively, developers may scour the landscape in search of
suitable locations for ideal ski conditions (Rothman, 1998: pp.187–188; Smith, 2003). In either case, the resulting selection of sites
for winter resort development does not appear to be subject to any systematic process (Hudson, 2000). Yet, given the transformative effects of ski resort development, there is value in anticipating where environmental and accessibility characteristics are
most likely to attract investment. The purpose of this paper is to present a model to identify potential sites that could be attractive
to ski resort development in the Rocky Mountains. Anticipation of such development may allow communities time to plan for the
inevitable consequences resorts bring: revitalizing economies while impacting fragile mountain environments and creating social
strains among the population (Blevins and Jensen, 1998; Hansen et al., 2002; Riebsame et al., 1996: p. 398; Ringholz, 1996).
Background
The United States ski industry is increasingly concentrated, having undergone considerable contraction and corporate
reorganization. While nationally, the number of resorts has declined from a peak of 622 in 1987 to 481 in 2007, annual ski
* Corresponding author. Tel.: þ1 302 821 8270.
E-mail address: [email protected] (P.W. Rees).
0143-6228/$ – see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.apgeog.2009.10.005
J.A. Silberman, P.W. Rees / Applied Geography 30 (2010) 36–49
37
visits have remained stable during the past decade, averaging 56.7 million, with the Rocky Mountain region the most popular,
accounting for 35% of the total (NSAA, 2008). Estimates suggest the 50 largest firms now earn 80% of the $2 billion annual
revenue generated by the ski industry while owning 20% of ski facilities (Business Wire, 2007). Despite this concentration,
operating profit margins for the industry as a whole for the 2006–2007 season were 24.8%. For the largest resorts, this figure
rose to 29.2%. Overall, pre-tax profit for the industry rose two-thirds over the previous five-year period (NSAA, 2007).
Although no systematic data are available at the regional and state level, several studies on the impact of the ski industry
report that in 2005–2006, the industry produced $692 million in revenue and generated a further $281 million in indirect
earnings in Utah (State of Utah, 2007); Colorado’s ski industry generated 31,000 jobs in 2002–2003 (Walsh, 2004); and in
2005–2006 Montana received 1.35 million ski visits (Nickerson & Bruns-Dubois, 2008).
The greatest impact of the ski industry is found at the local level. The original 19th century communities in the Rocky
Mountains were prompted by the discovery of gold, silver, copper and other metals, ranching in the high mountain basins to
supply food and animal power, and timber extraction. A few owed their origin to the growth of mountain tourism when a later
reassessment of the meaning of wilderness produced recreation centers such as Estes Park, Colorado and health spas such as
Colorado Springs (Wyckoff & Dilsaver, 1995: p. 41). But the dominant reliance on natural resources subjected most Rocky
Mountain settlements to the boom and bust cycles of exogenous commodity markets. After the 1880s, a long structural
decline in mining set in that saw a slow shrinking of many mountain settlements in the Rocky Mountain region (Power, 1996:
pp. 31–38; Lorah & Southwick, 2003: pp. 258–259; Wyckoff, 1999: p. 73). Some sites became ghost towns, devoid of
inhabitants while others struggled to reinvent themselves through reinterpretation of their natural resources and distinctive
architecture. (Dorward, 1990; Ringholz, 1996: pp. 113–123; Stoehr, 1975). In the forefront of this renaissance was the ski
industry (Coleman, 2004; Fry, 2006; Rothman, 1998).
Skiing began as a practical means of communication for snowbound residents of towns like Steamboat Springs and
developed as a pastime for wealthy visitors who sought a wilderness experience with luxurious accommodations epitomized
by the first major ski resort in Sun Valley, Idaho, opened in 1936 (Rothman, 1998: pp. 186–201). Following World War II, skiing
became an increasingly middle-class recreation, popular with both locals and a growing national market. The world’s longest
chairlift opened in 1947 in Aspen, a mining town that had foundered after the silver panic of 1893 but revived as a ski center
(Rothman, 1998: pp. 206–211). In the 1950s, a new inward flow of people seeking leisure activities appeared in the mountain
region. These ‘‘amenity migrants’’ were drawn by opportunities for hunting, outdoor sports, and visions of a clean environment and peaceful mountain existence (Chipeniuk, 2004; Jobes, 1995; Price, Moss, & Williams, 1997: p. 249). They were
followed by suburban escapists seeking second homes, later converted to permanent residences (McHugh, 1990). Many of
these newcomers were first exposed to the region as skiing visitors (Cuba, 1989). The demand for skiing and residential
development combined in a new type of resort, increasingly controlled by regional and national corporate, rather than local,
capital (Clifford, 2002). Vail in 1962 and Snowmass in 1967 were early examples of this new wave of ski resort that began to
proliferate across the Rocky Mountain region.
The consequences have been significant. Throughout the Mountain West, per capita income in counties with ski areas
exceeded those without by $3900 in 1997 (Booth, 2002: p. 55). In Summit County, Colorado, home to six major resorts, winter
tourism in 2003–2004 contributed 39% of basic spending and 46% of county employment (NCCG, 2004). One estimate
suggests that the development of Bitteroot Resort in Missoula county, Montana, would be the county’s fifth largest employer
by 2020, adding over 2000 jobs and nearly $110 million to the local economy (Moore, Wyman, Whelan, & Josephson, 2007).
But as new resort development returned some old mountain towns ‘‘from oblivion to eminence’’ (Dorward, 1990: p. 325),
cultural and economic cleavages emerged. Town populations experienced conflicting objectives between the desire to sustain
pastoral landscapes and small-town community life while feeding the economic engine created by the recreational economy
(Muller, Yin, & Alexandrescu, 2008: p. 1740). The influx of visitors and new permanent residents often clashed with the
interests of old-timers who saw the loss of open space, traffic congestion, pollution and dramatic property inflation as too
high a price to pay (Ringholz, 1996: pp. 64–85; Vias & Carruthers, 2005). In Vail, only one-fifth of the police and firefighters
can afford to live in the town while three-quarters of the dwellings are second homes, occupied for only a few months each
year (Howe, McMahon, & Propst, 1997). In Jackson Hole, low-paid hotel workers unable to pay rents in town are forced to
commute across Teton Pass from more affordable accommodations in Driggs and Victor in Idaho (Beyers & Nelson, 2000:
p. 470; Power, 1996). Careful planning in advance of resort development might help ameliorate these impacts but often town
leaders believe they can address growth problems once they occur, or they lack the expertise and ability to resist the demands
of corporate developers (Chipeniuk, 2005).
The history of Telluride, Colorado illustrates these trends (Fetter and Fetter, 1982; Ringholz, 1996: pp. 93–97). Silver
deposits in a remote region of Colorado’s San Juan Mountains were discovered in 1878, prompting the establishment of
several mining camps but not until rail access was achieved in 1890 did the town of Telluride begin to flourish. Three years
later the Silver Panic closed mines and population dwindled. Mining continued into the first half of the 20th century as
deposits of copper, gold and zinc were uncovered but they were rapidly exhausted and on April 27th, 1953, the last mine
closed, forcing 230 miners into unemployment. Two weeks later another company purchased the Telluride mines and
re-started production. However, local citizens were apprehensive for the town’s future and sought other ways to diversify its
economy. National Historic Landmark status was obtained and discussion began about a winter resort. A wealthy Californian,
Joe Zoline, seized the opportunity to turn this remote but beautiful setting into a ski resort. In 1968, rejecting what he
considered the excesses of growth and sprawl in Aspen (Coleman, 2004: p. 184), Zoline developed a master resort plan with
strict controls on land development, residential growth and the number of skiers. Building codes limited building height and
38
J.A. Silberman, P.W. Rees / Applied Geography 30 (2010) 36–49
restricted signage. As the resort began to grow, Main Street attracted new shops with restaurants and boutiques filling its
Victorian streetscape. However, despite attempts to control resort development, land prices soared, and long-time residents
began to be forced out. Old Victorian houses were selling for $900,000 and some resort employees could only afford car
camping in nearby forests and city parks (Ringholz, 1996: p. 95). Others were forced to commute from Montrose, CO, 75 miles
away (Beyers & Nelson, 2000: p. 470). By 2003, Tellurides’s permanent population had grown to 2500 and despite its
remoteness daily visitors reached into the thousands. Growth spilled beyond the historic box-canyon town site and Telluride
Mountain Village, a sprawling new community that evolved to service the resort, lacked the original town’s strict controls
(Coleman, 2004: p. 196).
Telluride’s rich history raises the question: how many other Rocky Mountain settlements are likely to be transformed from
economies based on resource extraction to ones relying on recreation with the inevitable policy conflicts over the value of
economic revival, the social costs of property inflation and the need for environmental protection? Expansion of existing ski
resorts and development of new projects continue, despite the global economic downturn that began in 2008. In addition to
the aforementioned Bitteroot resort in Montana, planned for the size of Park City in Utah, other projects under discussion
include a world-class resort in the Oquirrh mountain range, 15 min from the Salt Lake City airport, and Battle Mountain resort
near Pueblo, CO (Denver Post, 2009; Fullmer, 2008). A number of existing resorts continue to grow. Real estate expansion and
new ski runs are planned or underway in Sun Valley, Breckenridge, Vail, Grand Targhee and Steamboat Springs in 2009,
despite the continuing poor national economy (Berwin, 2009; Skiresortsmarket.com, 2009). Anticipating where this transformation of Rocky Mountain communities by the ski industry might next occur offers both a local and regional planning tool.
Building a GIS model of ski resort prediction
Modeling the location of economic phenomena has a long history (Isard, 1956; Owen & Daskin, 1998). Studies consider the
geographic variation of manufacturing cost factors (Bae, 2009; Rathelot & Sillard, 2008), the role of proximity to regions of
similar economic specialization (Andersson and Hellerstadt, 2009; Porter, 2003) and criteria influencing the behavior of
decision-makers (Hätönen, 2009; Strauss-Kahn & Vives, 2009). Use of Geographic Information Systems/Science (GIS) allows
an even more robust analysis of spatial behavior. Examples include Weber and Chapman (2009) who rated London districts by
labor supply, facilities and business support structure and used GIS to identify business sector clusters suitable for different
types of industry; and Cheng, Heng, and Ling (2007) who created an interactive inquiry model for potential shopping malls,
based on minimum distance, maximum demand coverage and maximum incomes to identify the location of optimal sites.
In both examples, GIS allows evaluation of the location of criteria businesses would use to make site choices.
Only one study using GIS was found that addressed the ski industry. Geneletti (2008) assessed the environmental impact
of proposed ski resorts in the northeastern Italian Alps. A geographical environmental sensitivity index was mapped against
the location of proposed ski resort sites. However, in this study, the location of the ski resorts is already known. In the present
paper, a GIS model is proposed that predicts the potential sites of future resorts using criteria that existing resorts appear to
have chosen as demonstrated by the location they select. As the UNEP-WCMC (2002: p. 65) study advocated, ‘‘GIS databases
can be used as decision support systems.to develop scenarios of possible future changes [in mountain environments]. These
can be produced for different management options, providing assessment of possible consequences.’’
The model developed for this study covers the U.S. Rocky Mountain region, defined as parts of New Mexico, Colorado,
Utah, Wyoming, Idaho and Montana, and potential sites for ski resorts include former mining towns, ranching communities
and other still-populated mountain settlements. Presently unpopulated locations could also have been evaluated but were
excluded from the present study because of interest in how existing settlements lacking resorts might reinvent themselves
and be affected by new development.
There is no standard definition of a ski ‘‘resort’’ and no definitive database of existing resorts. Two sources were used in this
study: The White Book of Ski Areas (Enzel, 2005) and Ski Town Resort Guide (Mountain Resort Guides, 2007). They list 81 and 85
resorts, respectively, operating in the Rocky Mountain region. Because the location of a ski resort may be some distance from
the named settlement listed as its corporate address, the exact coordinate locations for each existing resort were determined
through a visual aerial survey using Google EarthÒ. During this process, duplicate sites, as well as some missing from the
inventories, were identified. A modified listing was compiled of 85 stand-alone ski resorts believed to be operating during the
2006–2007 season in the Rocky Mountain region. These are used as the basis for the present study (Fig. 1). Each resort was
evaluated according to four criteria: seasonal snowfall quantity, potential ski season, proximity to National Forest land, and
accessibility to population centers that would represent the market for skiers and other visitors.
Estimating snowfall
Adequate snowfall is the minimum requirement for a ski resort, and the advertisements for many resorts make extravagant
claims for the amounts of snow they receive. Snow quantity depends on sufficiently low temperatures, the number of precipitation events and local climate factors such as the lake effect snowfall experienced east of the Great Salt Lake. But skiing is also
affected by snow quality. Low relative humidity of air masses during precipitation produce the fine powder snow skiers seek while
more humid air creates wetter snow that is less favored and can make snow-covered slopes more susceptible to avalanches. Slope
aspect can influence surface temperatures although Lazar and Williams (2008) found that whether north- or south-facing, aspect
had little influence on snow density. The number of freeze/thaw cycles that occur early in the season influence the snow pack,
J.A. Silberman, P.W. Rees / Applied Geography 30 (2010) 36–49
39
Fig. 1. Active ski resorts in the Rocky Mountain region, 2007. Source: The White Book of Ski Areas (Enzel, 2005), Ski Town Resort Guide (Mountain Resort Guides,
2007) and Google EarthÒ.
which should be at a minimum depth of 30 cm (11.8 inches) for downhill skiing and snowboarding (Breiling & Charamaza, 1999:
p. 12). Human action can influence these natural processes. Snow pack compression can be aided with grooming machines that
often soften snow, break up ice and remove rocky obstacles, a practice common in New England but also followed by many resorts
in the Rockies. Trees can be strategically planted to retain snow and shade slopes, thereby reducing surface melt (Scott & McBoyle,
2007: p. 1418). Machine-made snow can supplement the natural supply if nighttime temperatures are ideally less than
24F (4 C), but at least below freezing, and sufficient water supplies are available (Gosnell & Preston, 2006: p. 59). The variability
of factors influencing snow quality prohibit its evaluation at the scale of the six-state Rocky Mountain region and instead should be
reserved for micro-scale analysis of particular locations.
The present study therefore relies on snow quantity, although even here, snowfall measurement is an uncertain science
(Doesken & Judson, 1997). Given the diversity of terrain in the Rockies, snowfall amounts over a short horizontal distance can
be highly variable, depending on whether data are collected at the peak or the base of a mountain slope or some mid-way
point where drifts can exaggerate inaccuracies. Moreover, the location of the two main sources for snow data, National
Weather Service first order and co-operative weather stations (NCDC, 2008) and SNOTEL (NRCS-USDA, 2009) sites, are
inadequate for evaluating ski resort sites. Snowfall measurements from co-operative sites are reported by trained volunteers,
the stations are frequently found some distance from ski resorts and recorded snowfall can vary substantially between the
two locations. The mean distance between existing ski areas and the nearest co-operative station in the Rocky Mountain
region was approximately 9.6 km, using a nearest neighbor analysis in ArcGIS, while the maximum distance was over 35 km.
SNOTEL collection points are located on mountain crests and operated by the National Resource Conservation Service
(NRCS). Designed to help calculate annual snowmelt run-off as a means of estimating water supply, SNOTEL sites record snow
depth rather than snowfall. Furthermore, the mean distance from existing ski areas to the nearest SNOTEL site was found to be
10.2 km with a maximum distance of over 80 km. Resorts themselves are another potential source of data. Of the 85 resorts
evaluated, a majority claim to receive more than 100 inches (254 cm) of snowfall annually, however, this information is based
on self-reporting and cannot be considered reliable because a standardized technique prescribing specifically how snowfall
data is to be collected by trained resort employees is not used. Thus, existing resort reports may inadvertently introduce bias
because it is beneficial for a resort to report a high snowfall amount for marketing purposes.
To achieve consistency, therefore, what is needed is a source that interpolates snowfall amounts over a continuous surface.
For the present study, Climate Atlas of the Contiguous United States data from the National Climatic Data Center is used (NCDC,
2007). The Climate Atlas data provides interpolated snowfall amounts for the entire United States, derived from existing
40
J.A. Silberman, P.W. Rees / Applied Geography 30 (2010) 36–49
station measurements (NCDC, 2008). While these stations do create the issues as previously noted, the public dataset is run
through a quality control process that attempts to account for elevation and terrain changes using the PRISM methodology
(NCDC, 2008; PRISM, 2009). A limitation of the data is that snowfall amounts greater than 36 inches (91.4 cm) in any given
month are not reported. Assuming the probability that most significant snowfall in the Rocky Mountains is received between
the months of October and May, the maximum value of snowfall for a location using NCDC data is 288 inches (731.5 cm) per
season. An attempt was made to obtain the original data used by NCDC to create the Climate Atlas that contained actual
maximum snowfall but the dataset is unavailable (Stevens, 2009). The available NCDC data are reported in ranges (0–3, 3–6,
6–9, 9–12 inches, etc.). Combining the lowest number in each range for each eight-month season provided a conservative
estimate of the amount of snowfall received at a given location. Using this method, the seasonal average snowfall for all 85
resorts, computed from measurements over a 30-year period from 1961 to 1990, was 149 inches (378.5 cm). While some sites
receive far in excess of that amount, it is certainly sufficient to support a ski resort and 63% of resorts fell within one standard
deviation of that average (Table 1).
Potential ski season
Besides snowfall amount, the length of the skiing season is a prominent feature of resort literature. For some resorts,
National Forest Service contracts require them to close on specific dates. For others, the persistence of adequate snow cover is
the governing factor (Scott and Kaiser, 2004). The ability of a location to maintain snow can be influenced by micro-scale
changes in solar radiation, aspect, orientation, prevailing winds, rain-shadow effects and temperature. Only the latter variable
is easily determined across a region as large as the Rockies and the length of season for a site is influenced by the number of
days that temperatures remain below freezing, limiting surface snow melt and allowing the ability for machine-made snow.
The need for interpolation of temperature over intervening spaces between station points poses the same problem as
described for snowfall. Therefore, for the period 1961–1990, Climate Atlas data were again used. A measurement of the
potential length of the skiing season was calculated by taking the mean daily minimum temperatures for each month at or
below 32 F (0 C). The potential season for a resort was determined to be the sum of the months. For the 85 ski resorts in the
dataset, the mean number of months between September and May (the outside limits of measurable snowfall in the Rocky
Mountains) was 7.6, with 58% of resorts falling within one standard deviation. Comparing this result with the actual experience of resorts is not possible because many are required to close at a specific date determined by their Forest Service
permits, which are based on non-climate factors such as environmental preservation (Clifford, 2002).
Influence of global warming on future snow data
The issue of global warming raises a question about the use of snowfall averages from a 30-year past record to evaluate
locations for future ski resort choices. Beniston, Keller, Kofi, and Goyette (2003a) states that in the Swiss Alps, snow cover
duration is reduced 15–20 days for each 1 C of wintertime warming and an anticipated 3 C warming by 2050 would push
the snowline upwards by 400 m (Beniston, Keller, Kofi, & Goyette, 2003b; Elsasser & Bürki, 2002; Haeberli & Beniston, 1999).
Temperatures in New England during the past century rose 2.1 C, much of the rise since 1970, and the center of gravity of
New Hampshire’s ski industry has moved northwards as a consequence (Hamilton, Rohall, Brown, Hayward, & Keim, 2003).
A study in Australia suggests climate change will decrease the ski season length by 10–60% at sites below 1600 m and from
5 to 30% above that elevation by 2020 (Hennessey et al., 2008).
In the Rocky Mountains, however, the impact of global warming scenarios is less certain. Global climate models (GCMs)
evaluated by Smith and Wagner (2006) were in general agreement that temperature increases would occur in the Rockies
below 9000 feet, but more in summer than in winter. Less agreement was found regarding precipitation. Some GCMs predicted an increase in winter precipitation while the one regional climate model considered predicted drier winters and wetter
summers. Five models used by Lazar and Williams (2008) for Aspen, CO produced general agreement about temperature,
with estimates of increased warming ranging from 1.8 to 2.5 C by 2030. Predicting precipitation was again more variable,
with estimates by 2030 ranging from one to eighteen% drier. These findings contrast with those of Beniston et al. (2003b:
p. 27) whose models for the European Alps predicted that warmer winters would bring more precipitation as snow above
1700–2000 m.
Table 1
Location criteria values for existing ski resorts.
Resort location criteria and units of
measurement
Mean value
of all resorts
Standard
deviation (s)
Percent of resorts
with values within 1s
Annual snowfall (inches)
Potential ski season (months)
Distance to national forests (miles)
Accessibility index (min)a
Distance of existing ski areas to
nearest settlement (min)
149
7.58
0.75
255
16
62
1.10
1.78
116
15
63
58
89
61
79
a
Calculated using ESRI network analyst closest facility tool.
J.A. Silberman, P.W. Rees / Applied Geography 30 (2010) 36–49
41
Pepin and Losleben (2002) criticize GCMs for relying on free air temperature lapse rates for their calculations rather than
surface temperature recordings. They argue that the high alpine regions of the Rockies are becoming a progressively stronger
heat sink, perhaps because of increased snow cover, enhanced air movement over the surface or decreased solar radiation
input. Their findings support Barry’s (1990) theory of climate change that enhanced warming of free air in higher elevations is
not being transferred to a surface that experiences cooler temperatures with increased precipitation and cloud cover.
The complexity associated with predicting future patterns of snowfall and seasonal longevity with any accuracy that could
impact the behavior of the ski industry is reinforced by the position that the high elevation and northerly latitude of much of
the Rocky Mountain region makes the area less susceptible to global warming effects than other parts of North American and
European ski destinations (Breiling and Charamaza, 1999: p. 12; Bustic, Hanak, & Valletta, 2009: p. 32). Mote, Hamlet, Clark,
and Lettenmaier (2005: p. 45) observe that while some locations in the interior West and Rocky Mountains are susceptible to
warming, most are so cold that a warm winter has little effect on spring snowpack. This position is reiterated by Gosnell and
Preston (2006: p. 74) who state ‘‘ski areas in the Central Rockies are probably the most resilient to climate change simply
because, in this continental climate, winter storms are accompanied by very cold temperatures well below the threshold for
snow instead of rain.’’
Two additional reasons for deferring attempts to incorporate climate change predictions into the present model involve
human actions. First, several studies have examined how machine-made snow can help offset the impact of warmer
temperatures on the ski industry. Scott and McBoyle (2007) quote the NSAA that reports 89% of Rocky Mountain ski areas now
have snowmaking equipment although it is used on only 11–13% of the skiable terrain. In New England and Quebec,
snowmaking extended the average season from 55 to 120 days over the period 1960–1991 (Scott, McBoyle, Minogue, & Mills,
2006; Scott, McBoyle, & Minogue, 2007). Other studies in Europe and Australia also support the ability of machine-made
snow to slow if not completely offset the impacts of global warming in the near-term to 2020 (Bicknell and McManus, 2006;
Hennessey et al., 2008; Hoffman, Sprengel, Ziegler, Kolb, & Abegg, 2009). Second, there is considerable evidence that skiers
recognize and react to short-term daily-to-annual conditions of snowfall and season length that are subject to far greater
fluctuations than the gradual average long-term trend of global warming (Gosnell and Preston, 2006: p. 57; Shih, Nicholls, &
Holecek, 2009). This observation is reflected in the ski industry’s own actions (Hoffman et al., 2009). Scott and McBoyle (2007:
p. 1426) ‘‘found little evidence that ski areas were engaged in long-term business planning in anticipation of future changes in
climate.’’
Proximity to national forest land
In addition to space for the ski lodge and parking, resorts require substantial amounts of land to support snow sports. The
geographical co-existence of resorts and federal forest land has often been noted (Clifford, 2002; Coleman, 2004: pp.
138–142). The U.S. Forest Service permits ski areas to operate on 62 national forests in 18 states and even promotes the
operation of private ski activities for profit on public land through low-cost, long-term renewable leases. Reforms in 1996
lengthened and standardized leases to 40-year terms with flexibility to interpret environmental review. Regulations,
however, do still restrict construction of base facilities on public land such that in many instances ski resort companies own or
acquire private land adjacent to the publicly owned Forest Service. They then seek permits for public land where most skiing
occurs. The permitting process has become more onerous for the ski industry as opposition from environmentalists has
provided a counterbalance to corporate interests (Casselman & McLaughlin, 2008; Tenenbaum, 2001).
Existing ski resorts were evaluated by measuring the straight-line distance between each resort location and the edge of
the nearest National Forest. Distances were calculated using ET Geo Wizards (ÓIanko Tchoukanski) on a U.S. Equidistant Conic
map projection that best preserved distances and minimized errors. Sixty-five percent of Rocky Mountain ski areas were
found bordering National Forest land. Because of complex land agreements, variable lease terms and the precise location of
base facilities, it is difficult to determine how many resorts utilize Forest Service property, however, the mean distance from
public lands for all resorts was 0.75 miles (1.2 km), with 89% falling within one standard deviation. Given this close association, the distribution of National Forests was used in this analysis to represent the available land on which most ski areas
were likely to operate.
Accessibility to the market for skiers
Most skiers originate in major metropolitan areas within the Rocky Mountain region or come from more distant parts of the
country and even overseas. Information regarding the exact market areas of ski resorts is proprietary so a surrogate measure was
necessary to determine accessibility. A ski resort’s market was considered to be composed of local visitors, who resided in
settlements of at least 10,000 population, regional visitors from cities with at least 50,000 population, and national and international visitors. Because the latter group’s origins are so indeterminate, it was assumed that airports within the region that
accommodated commuter or long-distance aviation services would serve as the point of contact for those more distant visitors.
Thus, while the outer edge of a ski resort’s market area could theoretically be the entire world (i.e. a skier could come from
anywhere), access to reach that market was represented by a resort’s proximity to the nearest settlement of 10,000, city of 50,000
and available airport. Information on the location of commercial airports were taken from the National Atlas of the United States and
U.S. Geological Survey; local population centers (>10,000 people) and regional population centers (>50,000 people), were
obtained from ESRI Cities 2000 Census (ESRI, 2006).
42
J.A. Silberman, P.W. Rees / Applied Geography 30 (2010) 36–49
An accessibility index was determined by a combined measure of travel time and distance to the three locations. The closest
facility tool in ESRI Network AnalystÒ was used to calculate an accessibility value. The tool first calculates the shortest distance
highway route between each ski resort and various market points and then the average time required to traverse that route, based
on a road-speed classification. The result, expressed in minutes, is considered more representative of actual accessibility than
direct ‘‘as-the-crow-flies’’ miles between points. The mean driving time from all existing resorts to commercial airports was
55 min. Equivalent figures for local population and regional population centers were 77 and 123 min, respectively. Ideally, the
contribution of each measure to the accessibility index should be proportional to the degree to which ski resorts draw a local,
regional and national population for their clientele, but this information is not systematically available. Orens and Seidl (2009: p.
225) sampled the ski population of Crested Butte and found that the median one-way travel time (excluding local Gunnison
County residents) was 12 h with a median travel distance of 950 miles. Two-thirds of visitors came from places 1000–1200 miles
away. Bunting, Wagner, & Jones (2005) found that 71% of skiers at Mt. Spokane, WA and Lookout Pass, Silver Mountain and
Schweitzer Mountain resorts in Idaho originated from the Inland Northwest region during the 2000–2004 seasons. But such
isolated studies provide no broadly useful indicators of the geographical breakdown of the ski visitor market. Lacking this
information, the index in this study assumes that local, regional and national/international markets each contribute one-third.
Should more precise measurement of the ski market origin become available, the model can easily accommodate a different
percentage contribution. Combining these scores provided a representation of the level of accessibility of each resort. Among the
85 resorts, accessibility varied from 530 to 17 min. The mean index score for all resorts was 255 min, with 61% falling within one
standard deviation.
Applying the model and analysis
Refining the database of potential settlements
The model is composed of two parts (Fig. 2). In the first, previously described, a database of 85 existing ski resorts was
created, and the resorts ranked according to the four described criteria to determine standard measures. In the second part,
the results were applied to all populated settlements in the Rocky Mountain region to evaluate each place as a potential for
Fig. 2. GIS model for identifying potential ski resorts.
J.A. Silberman, P.W. Rees / Applied Geography 30 (2010) 36–49
43
new ski resort development, based on the standard measures of existing resorts. However, not all existing ski resorts are
located at the site of named settlements. Some can be found up to 70 min driving time away. Among existing ski resorts, the
mean driving time between a resort and the closest named settlement is 16 min with 79% of resorts falling within one
standard deviation (1–31 min) (Table 1). In the application of the model, therefore, communities were considered potentially
subject to new resort development if criteria reflecting conditions at existing ski resorts were found not just at a given named
settlement but at any location up to 31 min driving time from its center.
Procedures for identifying potentially impacted settlements began with a database obtained from ESRI’s Populated Place
dataset (ESRI, 2006). Within the six Rocky Mountain States, 1555 places were found with population totals ranging from one
in Lost Springs, WY to over one-half million in Denver. As constituted, the database included places already impacted by
existing ski resorts. The database was queried and 214 locations were found that were within 31 min driving time of a ski
resort. These were removed and the remaining 1341 settlements in the database subjected to the model’s analysis.
The next step involved querying the database to identify all areas in the Rocky Mountain region where the minimum
criteria for a ski resort could be found. Areas designated were those where seasonal snowfall, potential ski season and
distance to National Forests fell within one standard deviation of the scores of existing ski resorts. All 1341 settlements were
then measured by driving-time minutes to the nearest area boundary. Those with values greater than 31 were eliminated
because they exceeded one standard deviation of the driving-time distance of existing ski resorts to the nearest settlement
and were considered to be too far from locations offering the minimum criteria for ski resorts. This action reduced the
database to 467 settlements.
An accessibility index was then calculated for the settlements remaining in the database and 72 places were identified
with values that exceeded one standard deviation of the accessibility scores of existing ski resorts. These were removed
because while they possessed the minimum criteria for a ski resort, they were considered too remote from the ski market,
leaving a final reduced list of 395 communities. These remaining settlements exhibited the minimum attributes of existing ski
areas at a location somewhere within 31 min driving time and were within reach of a ski visitor market, yet were not
presently operating as ski towns.
Identifying settlements most likely to be impacted by ski resort development
An additional sequential elimination procedure was then applied to the reduced list of 395 settlements to identify those
with the greatest potential for resort development. For this next step in the analysis, the criteria were strengthened and
possible sites were compared with the values of existing ski resorts that fell above one standard deviation for snowfall
(211 inches) and potential ski season (8.68 months), and at or less than the mean distance from National Forest land. The
decision to stiffen the criteria allowed identification of the top candidates for development but the database as constructed for
this study could be queried for any set of qualifying values deemed appropriate. Repeating the analytical steps described
above with a higher threshold of criteria values eliminated an additional 371 settlements, leaving a final list of 24 places
(Fig. 3).
Further refining of the database of 24 candidate settlements depends on the business model followed by a ski resort
company with regard to market area accessibility and distance from competing ski resorts. One scenario might emphasize
proximity to the market, relying on a high volume of day skiers from nearby communities and regional metropolitan areas to
generate income. In such a case, a company would want to consider those settlements with the fewest total driving-time
minutes to reach the market for skiers. A second scenario, one similar to that taken by the developers of Telluride, might
emphasize geographical exclusivity, the objective being to find a remote location that would require longer overnight stays
and appeal to higher income skiers whose greater expenditures would offset lower customer volume. In this scenario,
a company would select from settlements with the highest number of driving-time minutes to reach their markets. A third
scenario might emphasize the importance of reducing competition by locating a new resort as far as possible from existing ski
resorts. A fourth scenario could involve the opposite strategy in which a company would chose to locate as close as possible to
existing ski resorts, thereby sharing an enlarged geographical market. Such a strategy takes advantage of the tendency of
skiers to enjoy skiing in a variety of neighboring nearby resorts, utilizing reciprocal lift fees and other discounts.
Querying the reduced settlement database for these four scenarios, based on the two variables of distance from existing ski
resorts and the accessibility score, identified Dayton, Wyoming as the settlement most likely to be susceptible to ski resort
development (Table 2, Fig. 4). It had the third best accessibility score and thus was close to the market for skiers (scenario one)
and was also third most distant from existing resorts that might produce competition (scenario two). Dayton is a small town
of 713 inhabitants located close to the Montana-Wyoming border in the eastern foothills of the Big Horn range. Founded in
the 1880s near a branch of the Tongue River, its main economy today involves support for nearby cattle ranching and alfalfa
feedlot operations. Although it houses a high school for the more than nine-hundred square mile Sheridan County School
District, Dayton’s retail establishment contains only a corner grocery store, café and bar. It has no hotel and no bank. Although
possessing a city government, the town’s zoning ordinance has no published provisions for subdivision development. As such,
Dayton is the classic small community that would be quite unprepared for the economic and social impact of a nearby major
ski resort.
While the database analysis identifies Dayton as a likely candidate for growth, it does not suggest that it fulfills all the
criteria perfectly. In fact, to a considerable extent the database indicates that accessibility to market and reduced competition
are mutually exclusive (Fig. 4). For instance, seven settlements were above the median of 24 for higher levels of accessibility
44
J.A. Silberman, P.W. Rees / Applied Geography 30 (2010) 36–49
Fig. 3. Rocky Mountain settlements fulfilling criteria for ski resort development.
(scenario one) but also were above the median for close competition from existing resorts. An additional seven settlements
were among the most remote, being above the median for driving-time minutes to reach their markets (scenarios two and
three), but were below the median for market accessibility. Finally, five settlements were above the median for proximity to
existing resorts (scenario four) but were also below the median for accessibility to the ski visitor market.
Identifying other potential candidates for development in addition to Dayton therefore required considering those
settlements that had a balanced ranking of relatively strong accessibility and distance from competing resorts. This approach
yielded four additional potential sites within the Rocky Mountain region. Island Park, Idaho had the sixth best accessibility
index and ranked second best in distance from competing ski resorts. Located west of Yellowstone National Park, Island Park
J.A. Silberman, P.W. Rees / Applied Geography 30 (2010) 36–49
45
Table 2
Selected settlements ranked by accessibility index scores and distance from competitors.
Settlement
State
Accessibility
index (distance-time mins)
Accessibility
rank
Nearest ski
area (distance-time mins)
Nearest ski
area rank
Albany
Auburn
Bedford
Buena Vista
Central City
Dayton
Etna
Fairplay
Grover
Island Park
Kremmling
Pagosa Springs
Philipsburg
Red Feather Lakes
Ridgway
Samak
Sheridan
Star Valley Ranch
Thayne
Turnerville
Victor
Walden
West Yellowstone
Woods Landing-Jelm
WY
WY
WY
CO
CO
WY
WY
CO
WY
ID
CO
CO
MT
CO
CO
UT
CO
WY
WY
WY
ID
CO
MT
WY
229.66
342.45
329.61
293.88
117.64
167.5
270.08
276.16
334.01
205.78
319.74
323.38
214.69
215.82
358.67
169.12
30.83
300.07
297.92
359.26
231.43
357.21
231.93
180.82
9
21
19
14
2
3
12
13
20
6
17
18
7
8
23
4
1
16
15
24
10
22
11
5
44.98
80.98
76.7
48.82
35.4
104.58
56.86
39.77
78.17
104.78
35.92
35.64
48.41
131.32
36.21
33.99
53.08
66.85
66.14
86.58
41.35
71.6
78.23
55.59
8
20
17
10
2
22
13
6
18
23
4
3
9
24
5
1
11
15
14
21
7
16
19
12
serves as a center of fly-fishing on the Henry’s Fork of the Snake River and a winter cross-country skiing destination in the
nearby Caribou-Targhee National Forest. Its early history is associated with the establishment of hunting lodges and guest
ranches beginning in the late 19th century. Today, it has a small population of only 215, extensive areas of open land around
the town center and is touted by its chamber of commerce as ‘‘an ideal winter destination.’’ Although governed by a county
planning commission comprehensive plan and zoning code, controls on development are mainly concerned with environmental mitigation rather than the pressure on infrastructure and housing that would accompany a major resort development
(Fremont County Comprehensive Plan, 2008; Fremont County Development Code, 2003).
Red Feather Lakes, Colorado ranked eighth best in accessibility to the market while occupying a location farthest from
existing Rocky Mountain ski resorts among the 24 considered. An unincorporated community in Larimer County, Red Feather
Lakes was founded in 1871 as a center of logging and ranching but grew in the 1920s when local landowners envisaged
development of a large resort community and subdivided much of the surrounding land. Since then, about half the residential
lots have been developed but over 4000 acres of land surrounding the community remains in private hands and subject to
potential development (Red Feather Lakes Area Plan, 2006). At present, with a population of 596, Red Feather Lakes serves
residents, many from Fort Collins, 40 miles to the south-east, who are attracted by the nearby scenic mountain lakes within
the Roosevelt National Forest. A broad spectrum of housing ranges from historic mountain cabins to a gated community
surrounding an 18-hole golf course. The settlement offers lodging, a few restaurants, three churches, a supermarket,
elementary school, library and post office. Development regulations are controlled by Larimer County and the State of
Colorado but absent changes to the present codes, major development in the Red Feather Lakes area would severely overload
existing water, sewer and local road infrastructure (Null Alternative, Red Feather Lakes Area Plan, 2006).
West Yellowstone, Montana, ranked 11th in accessibility and fifth most distant from existing ski resorts, is already a well
developed town of 1177 permanent residents (2000 Census), although that number swells ten-fold with visitors as an
estimated two million annually enter Yellowstone National Park via the west gate. The incorporated town already contains
a privately-developed Grizzly Discovery Center composed of a natural habitat for grizzly bears and gray wolves, numerous
motels, art galleries, retail stores and an IMAX movie theater (Geske, 1997). In winter, it becomes the self-styled ‘‘snowmobile
capital of the world’’ with over 1000 vehicles for rent to support the 60% of all visitors who use this form of transportation to
enter Yellowstone Park (Sacklin, Legg, Creachbaum, Hawkes, & Helfrich, 2000). The area is noted for unrestrictive land use
policies and large-lot subdivision that will likely further stress the surrounding rural environment bordering the town should
a nearby ski resort be added to the already substantial development (Gude, Hansen, Rasker, & Maxwell, 2006).
A final candidate settlement, Etna, Wyoming, ranked 12th in accessibility and 11th in distance from competition, is a small
community of 136 people, part of a number of settlements that make up the Star Valley along the Wyoming-Idaho border,
30 miles as the crow flies to the southwest of Jackson. Etna is surrounded by the Bridger-Teton, Targhee and Caribou national
forests and sits at an elevation of 5827 feet, yet is less than 5 miles from Stewart Peak with an elevation of 10,026 feet. Settled
in the 1870s by Morman Pioneers, the valley’s communities relied on dairying in the early 20th century before the agricultural
economy gave way to tourism and recreation. There is a Mormon church, a motel and an elementary school in the town,
although it is also within several miles of Star Valley Ranch, a self-contained community with restaurants, two golf courses
and its own planning authority. Etna’s unincorporated status places its development policies under those of Lincoln County,
46
J.A. Silberman, P.W. Rees / Applied Geography 30 (2010) 36–49
Fig. 4. Accessibility index values and driving-time distances from competitors for 24 selected settlements.
Wyoming. A recent evaluation of the major highway through the Star Valley identified the need for significant upgrades as
communities served shifted from agricultural to residential development, yet also noted that room for an alternative route
was absent (Lincoln County Comprehensive Development Plan, 2009). Any significant resort development would therefore
require major infrastructure improvements beyond the scope presently planned.
Conclusion
The purpose of the present study was to develop a GIS-based model and analytical procedure to identify settlements in the
Rocky Mountain region that possess the attributes of existing ski areas and might be suitable candidates for reinvention as ski
resort towns. Each step in the procedure required a degree of investigator judgment. The available data on annual snowfall
and potential ski season both have limitations in the way information is reported and because interpolation values for
intervening locations are constructed from a number of fixed-point data collection sites, considerable room for error can
result. Accessibility criteria were based on reasonable assumptions that could not be confirmed because actual ski area market
data is proprietary and difficult to obtain. Selection of the measures of existing ski areas using values above or below one
standard deviation of the group as a whole were statistically valid but not selected on the basis of any factual criteria. On the
other hand, the GIS-based model allowed these parameters to be transparent and was constructed so that the task of running
the model with different criteria and improved data would be relatively simple. Similarly, the sequential elimination
procedure could be easily run with different values, depending on the objectives of a particular investigation.
An additional characteristic of the present study was the evaluation of existing Rocky Mountain populated settlements as
possible sites for ski resort development, rather than considering all locations in the region. Many additional sites that once
supported settlements are now ghost towns yet if they possess the attributes of current ski resorts, their revival is possible.
While the lack of data points can pose difficulties in interpolating accurate snowfall and potential length of season throughout
the entire region, the logic of the model would still allow identification of sites meeting the criteria for ski resort development, regardless of whether or not they were occupied by an existing settlement. For instance, should climate change
J.A. Silberman, P.W. Rees / Applied Geography 30 (2010) 36–49
47
ultimately impact industry decision-making, the model could be calibrated to identify climatically advantaged ski locations
with high elevation terrain potential where development would be more likely.
A final caveat concerning the model involves the selection of specific named settlements as those most likely to experience
development. The study identified five possible candidates and suggested four possible business scenarios that could be
followed, which would each result in different solutions. Moreover, from the brief descriptions of those communities
identified as final candidates based on a balanced ranking approach, it is clear that at this point the model produces an
important initial sorting function but detailed field observation would still be needed to judge the ultimate suitability of
a particular site for resort development. Further investigations could consider localized climates, terrain potential, town and
site aesthetics and environmental issues. Nonetheless, by providing a systematic analytical procedure, the study does offer
environmental and preservationist advocacy groups, as well as town boosters seeking to reinvent the fortunes of economically declining communities in the Rocky Mountain region, the opportunity to anticipate prospects for change. Equally, the
procedures followed in this study have applicability to other winter sports regions worldwide or to other forms of landscape
change where identifying environmental, cultural and economic location criteria of existing phenomena can be the basis for
evaluating alternative but yet-to-be developed sites.
Acknowledgements
The authors wish to thank Daniel Leathers and Wayne Gibson for help with climatic data, David Legates for reading the
manuscript and commenting on statistical measures, and Tracy DeLiberty and John and Tina Callahan for help with computer
programming and analysis. The thoughtful comments of two anonymous reviewers were also appreciated and prompted
substantial improvements to the manuscript.
References
Andersson, M., & Hellerstadt, K. (2009). Location attributes and start-ups in knowledge-intensive business services. Industry and Innovation, 16(1), 103–121.
Bae, S. (2009). The responses of manufacturing business to geographical differences in electricity prices. Annals of Regional Science, 43(2), 453–472.
Barry, R. (1990). Changes in mountain climate and glacio-hydrological responses. Mountain Research and Development, 10, 161–170.
Beniston, M., Keller, F., Kofi, B., & Goyette, S. (2003a). Estimates of snow accumulation and volume in the Swiss Alps under changing climatic conditions.
Theoretical and Applied Climatology, 76, 125–140.
Beniston, M., Keller F., Kofi B., & Goyette S. (2003b). Snow pack in the Swiss Alps under changing climatic conditions: an empirical approach for climate
impact studies. pp. 19–31.
Berwin, B. (February 19, 2009). USFS studies lynx habitat around Peak 6 in Breckenridge. Denver Post.
Beyers, W. B., & Nelson, P. B. (2000). Contemporary development forces in the non-metropolitan West: new insights from rapidly growing communities.
Journal of Rural Studies, 16, 459–474.
Bicknell, S., & McManus, P. (2006). The canary in the coalmine. Australian ski resorts and their response to climate change. Geographical Research, 44(4), 386–401.
Blevins, A., & Jensen, K. (1998). Gambling as a community development quick fix. Annals of the American Academy of Political and Social Science, 556, 109–123.
Booth, D. E. (2002). Searching for paradise. Economic development and environmental change in the Mountain West. Lanham, MD: Rowman and Littlefield
Publishers.
Breiling, M., & Charamaza, U. P. (1999). The impact of global warming on winter tourism and skiing. A regionalized model for Austrian snow conditions.
Regional Environmental Change, 1, 4–14.
Bunting, D., Wagner, M., & Jones, D.P. (2005). The economic impact of ski areas represented by the Inland Northwest Ski Association. Executive summary.
Spokane, WA: Institute for Public Policy and Economic Analysis, Eastern Washington State University.
Business Wire. (July 30, 2007). Research and markets: the U.S. Ski Industry. Business Wire.
Bustic, V., Hanak, E., & Valletta, R. G. (2009). Climate change and housing prices. Hedonic estimates for North American ski resorts. San Francisco: Federal
Reserve Bank.
Casselman, B., & McLaughlin, K. (May 16, 2008). Politically correct developments. Toronto Globe and Mail.
Chipeniuk, R. (2004). Planning for amenity migration in Canada. Current capacities of interior British Columbia mountain communities. Mountain Research
and Development, 24(4), 327–335.
Chipeniuk, R. (2005). Planning for the advent of large resorts: current capacities of interior British Columbian mountain communities. Environments, 33(2), 57–69.
Clifford, H. (2002). Downhill slide: why the corporate ski industry is bad for skiing, ski towns and the environment. San Francisco: Sierra Club Books.
Coleman, A. G. (2004). Ski style. Sport and culture in the Rockies. Lawrence: University Press of Kansas.
Cheng, E. W. L., Heng, L., & Ling, Y. (2007). A GIS approach to shopping mall location selection. Building and Environment, 42(2), 884–892.
Cuba, L. (1989). From visitor to resident. Retiring in vacation land. Generations, 13, 63–67.
Denver Post. (May 26, 2009). Pueblo may sell water rights to new ski resort. Denver Post.
Doesken, N. J., & Judson, A. (1997). A guide to the science, climatology and measurement of snow in the United States (2nd ed.). Fort Collins: Department of
Atmospheric Science, Colorado State University.
Dorward, S. (1990). Design for mountain communities: a landscape and architectural guide. Norman: University of Oklahoma Press.
Elsasser, H., & Bürki, R. (2002). Climate change as a threat to tourism in the Alps. Climate Research, 20(3), 253–257.
Enzel, R. G. (2005). The white book of ski areas (30th ed.). Washington, D.C.: Inter-Ski Services. http://www.inter-ski.com/whitebook30/whitebook.htm
[accessed 04.08.09].
ESRI. (2006). ArcGIS 9. ESRI data & maps and street map USA. Redlands, CA: ESRI Inc.
Fetter, R. L., & Fetter, S. (1982). Telluride: from pick to powder. Caldwell, CO: Caxton Printers.
Fremont County Comprehensive Plan. (January 1, 2008). Draft. http://www.co.fremont.id.us/departments/planning_building/Comp_Plan/index.htm
[accessed 04.08.09].
Fremont County Development Code. (2003). http://www.co.fremont.id.us/departments/planning building/Dev Code New. pdf [accessed 04.08.09].
Fry, J. (2006). The story of modern skiing. Hanover, NH: University Press of New England.
Fullmer, B. (2008). Sustainable growth: Kennecott Land has a vision. Intermountain Contractor, 64(2), 32.
Geneletti, D. (2008). Impact assessment of proposed ski areas. A GIS approach integrating biological, physical and landscape indicators. Environmental
Impact Assessment Review, 28(2–3), 116–130.
Geske, L. D. (1997). Montana utility brings LNG to West Yellowstone. Pipeline & Gas Journal, 224(1), 49–50.
Gill, A. (2000). From growth machine to growth management. The dynamics of resort development in Whistler, British Columbia. Environment and Planning
A, 32, 1083–1103.
48
J.A. Silberman, P.W. Rees / Applied Geography 30 (2010) 36–49
Gosnell, H. T., & Preston, G. (2006). Socio-economic impacts and adaptation. In J. Katzenberger, & K. Crandall (Eds.), Climate change and Aspen. An assessment
of impacts and potential responses (pp. 57–80). Aspen: Aspen Global Change Institute.
Gude, P. H., Hansen, A. J., Rasker, R., & Maxwell, B. (2006). Rates and drivers of rural residential development in the Greater Yellowstone. Landscape and
Urban Planning, 77(1–2), 131–151.
Haeberli, W., & Beniston, M. (1999). Climate change and its impacts on glaciers and permafrost in the Alps. Ambio, 27, 258–265.
Hamilton, L. C., Rohall, D. E., Brown, B. C., Hayward, G., & Keim, B. D. (2003). Warming winters and New Hampshire’s lost ski areas. An integrated case study.
International Journal of Sociology and Public Policy, 23(10), 52–73.
Hansen, A. J., Rasker, R., Maxwell, B., Rotella, J. J., Johnson, D., Parmenter, A. W., et al. (2002). Ecological causes and consequences of demographic change in
the New West. BioScience, 52(2), 151–162.
Hätönen, J. (2009). Making locational choice. A case approach to the development of a theory of offshore outsourcing and internationalization. Journal of
International Management, 15(1), 61–79.
Hennessey, K. J., Whetton, P. H., Walsh, K., Smith, I. N., Bathols, J. M., & Hutchinson, M. (2008). Climate change effects on snow conditions in Mainland
Australia and adaptation of ski resorts through snowmaking. Climate Research, 35(3), 255–270.
Hoffman, V. H., Sprengel, D. C., Ziegler, A., Kolb, M., & Abegg, B. (2009). Determinants of corporate adaptation to climate change in winter tourism.
An econometric analysis. Global Environmental Change, Human and Policy Dimensions, 19(2), 256–264.
Howe, J., McMahon, E., & Propst, L. (1997). Balancing nature and commerce in gateway communities. Washington, D.C.: Island Press.
Hudson, S. (2000). Snow business. A study of the international ski industry. London: Cassell.
Isard, W. (1956). Location and space-economy; a general theory relating to industrial location, market areas, land use, trade, and urban structure. In: Regional
science studies 1. Cambridge: Technology Press of Massachusetts Institute of Technology and New York:Wiley.
Jobes, P. (1995). Migration in the West. A Gallatin Valley, Montana case study. The Western Planner, 16(3), 10–13.
Kariel, H. G. (1989). Socio-cultural impacts of tourism in the Austrian Alps. Mountain Research and Development, 9, 59–70.
Lasanta, T., Laguna, M., & Vicente-Serrano, S. M. (2007). Do tourism-based ski resorts contribute to the homogeneous development of the Mediterranean
mountains? A case study of the Central Spanish Pyrenees. Tourism Management, 28(5), 1326–1339.
Lazar, B., & Williams, M. (2008). Climate change in western ski areas. Potential changes in the timing of wet avalanches and snow quality for the Aspen ski
areas in the years 2030 and 2100. Cold Regions Science and Technology, 51(2–3), 219–228.
Lincoln County Comprehensive Development Plan. (2009). Appendix 6B: U.S. 89 corridor study. http://www.lcwy.org/docs/Planning/Comprehensive_Plan/
Comprehensive_Plan/Appendix%206B%20-%20US89%20Corridor%20Study%20Comprehensive%20Plan.pdf [accessed 16.10.09].
Lorah, P., & Southwick, R. (2003). Environmental protection, population change, and economic development in the rural western United States. Population
and Environment, 24(3), 255–272.
McHugh, K. E. (1990). Seasonal migration as a substitute for, or precursor to, permanent migration. Research on Aging, 12, 229–245.
Miller, C. M., & Blevins, A. (2005). Battlement Mesa. A case study of community evolution. Social Science Journal, 42(1), 1–12.
Mote, P. W., Hamlet, A. F., Clark, M. P., & Lettenmaier, D. P. (2005). Declining mountain snowpack in western North America. Bulletin of the American
Meteorological Society, 86(1), 39–49.
Moore, T., Wyman, R., Whelan, B., & Josephson, A. (2007). Bitteroot resort impact analysis. Prepared for the Missoula Area Economic Development
Corporation and Missoula Area Chamber of Commerce. Portland, OR: ECONorthwest.
Mountain Resort Guides. (2007, 2008). Ski town resort guide. http://www.skitown.com [accessed 04.08.09].
Muller, B., Yin, L., & Alexandrescu, F. (2008). The dynamics of land development in resort communities. A multiagent simulation of growth regimes and
housing choices. Environment and Planning A, 40(7), 1728–1743.
NCDC (National Climatic Data Center). (2007). Climate atlas of the contiguous United States. http://cdo.ncdc.noaa.gov/cgi-bin/climaps/climaps.pl [accessed 04.08.04].
NCDC (National Climate Data Center). (2008). Climate atlas of the United States. Data documentation. http://www.ncdc.noaa.gov/oa/about/cdrom/climatls2/
datadoc.html [accessed 30.08.09].
National Ski Areas Association [NSAA]. (2007). Ski resort industry research compendium. http://www.nsaa.org/nsaa/marketing/docs/0607-researchcompendium.pdf [accessed 28.07.09].
National Ski Areas Association [NSAA]. (2008). Industry stats. http://www.nsaa.org/nsaa/press/industryStats.asp [accessed 28.07.09].
Nickerson, N. P., & Bruns-Dubois, M. (2008). Outlook and trends 2008: Montana travel and recreation. ITTR Newsletter, 13(1), 2, Institute for Tourism and
Recreation Research, University of Montana.
Northwest Colorado Council of Governments[NCCG]. (2004). Job generation in the Colorado Mountain resort community. http://www.nwc.cog.co.us/docs/
resources/nwcog_reports_studies/job_generation_mtn_resort_economy_2004.pdf [accessed 03.08.09].
NRCS-USDA (Natural Resources Conservation Service, United States Department of Agriculture). SNOTEL data and products. http://www.wcc.nrcs.usda.gov/
snow/ [accessed 05.08.09].
Orens, A., & Seidl, A. (2009). Working lands and winter tourists in the Rocky Mountain West. A travel cost, contingent behavior, input-output analysis.
Tourism Economics, 15(1), 216.
Owen, S. H., & Daskin, M. S. (1998). Strategic facility location: a review. European Journal of Operation Research, 111, 422–447.
Pepin, N., & Losleben, M. (2002). Climate change in the Colorado mountains. Free air versus surface temperature trends. International Journal of Climatology,
22(3), 311–329.
Price, M. F., Moss, A. G., & Williams, P. W. (1997). Tourism and amenity migration. In B. Messerli, & J. Ives (Eds.), Mountains of the World: a global priority (pp.
409–446). New York: The Parthenon Publishing Group.
Porter, M. E. (2003). The economic performance of regions. Regional Studies, 37(6/7), 549–578.
Power, T. M. (1996). Lost landscapes and failed economies. The search for a value of place. Washington, D.C.: Island Press.
PRISM. (2009). PRISM (parameter-elevation regressions on independent slopes model). PRISM Climate Group, Oregon State University. http://prism.
oregonstate.edu/ [accessed 29.08.09].
Purdue, R. R. (2004). Skiers, ski bums trust fund babies, migrants, techies and entrepreneurs. The changing face of the Colorado ski industry. In K.
Weiermair, & C. Mathies (Eds.), The tourism and leisure industry. Shaping the future (pp. 209–225). New York: Haworth Hospitality Press.
Rathelot, R., & Sillard, P. (2008). The importance of local corporate taxes in business location decisions. Evidence from French micro data. Economic Journal,
18(527), 499–514.
Red Feather Lakes Area Plan. (2006). http://www.co.larimer.co.us/redfeather/ [accessed 30.09.09].
Riebsame, W. E., Gosnell, H., & Theobald, D. M. (1996). Land use and landscape change in the Colorado Mountains I. Theory, scale and pattern. Mountain
Research and Development, 16(4), 395–405.
Ringholz, R. C. (1996). Paradise paved. The challenge of growth in the West. Salt Lake City: University of Utah Press.
Rothman, H. K. (1998). Devil’s bargain. Tourism in the twentieth-century west. Lawrence: University Press of Kansas.
Sacklin, J. A., Legg, K. L., Creachbaum, M. S., Hawkes, C. L., & Helfrich, G. (2000). Winter visitor use planning in Yellowstone and Grand Teton National Parks. In:
USDA Forest Service Proceedings, vol. 4. USDA. pp. 243–250.
Scott, D., & Kaiser, D. (2004). Variability and trends in United States snowfall over the last half century. In Fifteenth symposium on global climate variations
and change. Seattle: American Meteorology Society.
Scott, D., & McBoyle, G. (2007). Climate change adaptation in the ski industry. Mitigation and Adaptation Strategies for Global Change, 12(8), 1411–1431.
Scott, D., McBoyle, G., Minogue, A., & Mills, B. (2006). Climate change and the sustainability of ski-based tourism in eastern North America. A reassessment.
Journal of Sustainable Tourism, 14(4), 376–398.
Scott, D., McBoyle, G., & Minogue, A. (2007). Climate change and Quebec’s ski industry. Global Environmental Change. Human and Policy Dimensions, 17(2),181–190.
Shih, C., Nicholls, S., & Holecek, D. F. (2009). Impact of weather on downhill ski lift ticket sales. Journal of Travel Research, 47(3), 359–372.
J.A. Silberman, P.W. Rees / Applied Geography 30 (2010) 36–49
49
Skiresortsmarkets.com. (2009). Ski communities and new developments. Monthly. January–August.
Smith, S. (2003). One lift, one man. Silverton: Colorado’s anti-resort. Frequency. The Snowboarder’s Journal, 2.2, 84–90.
Smith, J. B., & Wagner, C. (2006). Climate change and its implications for the Rocky Mountain region. Journal of American Water Works Association, 98(6),
80–92.
State of Utah. (2007). Utah’s ski industry. Economic report of the Governor. pp. 215–221. http://governor.utah.gov/dea/ERG/ERG2007/ [accessed 05.08.09].
Stevens, S. Personal communication, August 29, 2009.
Stoehr, C. E. (1975). Bonanza Victorian: architecture and society in Colorado mining towns. Albuquerque: University of New Mexico Press.
Strauss-Kahn, V., & Vives, X. (2009). Why and where do headquarters move? Regional Science and Urban Economics, 39(2), 168–186.
Tenenbaum, D. J. (2001). Land use. The slippery slope of ski expansion. Environmental Health Perspectives, 109(3), A112.
Theobald, D. M., Gosnell, H., & Riebsame, W. E. (1996). Land use and landscape change in the Colorado mountains II. A case study of the East River Valley.
Mountain Research and Development, 16(4), 407–418.
United Nations Environment Programme, World Conservation Monitoring Centre (UNEP-WCMC). (2002). Sustainable development. In Mountain Watch
launch and Bishkek briefing. A report on changing mountain environments for the Bishkek Global Mountain summit. pp. 64–73. http://www.unepwcmc.org/mountains/mountain_watch/pdfs/ [accessed 03.08.09].
Vias, A. C., & Carruthers, J. I. (2005). Regional development and land use change in the Rocky Mountain West, 1982–1997. Growth and Change, 36(2),
244–272.
Walsh, C. (April 2004). State ski industry a peak producer. Rocky Mountain News, 1, 113.
Weber, P., & Chapman, D. (2009). Investing in geography. A GIS to support inward investment. Computers Environment and Urban Systems, 33(1), 1–14.
Wyckoff, W. (1999). Creating Colorado. The making of a Western American landscape 1860–1940. New Haven: Yale University Press.
Wyckoff, W., & Dilsaver, L. M. (Eds.). (1995). The mountainous west. Explorations in historical geography. Lincoln: University of Nebraska Press.