Printer Friendly

The Relationship between Airport Infrastructure and Flight Arrivals in Remote Northern Canadian Communities.

ABSTRACT. Much of Canada's northern population resides in communities that are inaccessible by road for a substantial portion of the year. Residents of these "fly-in" communities rely on aircraft to provide a wide range of social, economic, and transportation services. However, for numerous reasons, including the often extreme environmental conditions in the circumpolar regions of Canada, a substantial number of flights to these communities are cancelled or diverted. Using a dataset from two airlines that serve these regions with information about schedules, delays, and cancellations of more than 18500 flights, we examined the links between airport infrastructure, flight arrival reliability, and a variety of socioeconomic variables in 23 northern communities. Results show that runway length has a significant impact on the reliability of flight arrival, but also that the reliability of flights may not affect the cost of food in the communities included in our analysis. These findings provide evidence that lengthening runways could improve air service in the Canadian North.

Key words: Canadian North; fly-in communities; airport infrastructure; air cargo; arrival reliability; food

RESUME. Une grande partie de la population du Nord canadien reside dans des localites inaccessibles par voie terrestre pendant une grande partie de l'annee. Les habitants de ces localites desservies par voie aerienne dependent des avions pour une vaste gamme de services sociaux, economiques et de transport. Toutefois, pour maintes raisons, dont les conditions environnementales souvent extremes qui sevissent dans les regions circumpolaires du Canada, un grand nombre de vols a destination de ces localites est annule ou devie. En nous appuyant sur des donnees en provenance de deux societes aeriennes qui desservent ces regions, donnees portant sur les horaires de vol, les retards et les annulations concernant plus de 18 500 vols, nous avons examine les liens entre les infrastructures aeroportuaires, la fiabilite de l'arrivee des vols et un eventail de variables socioeconomiques propres a 23 localites nordiques. Les resultats ont permis de constater que la longueur des pistes exerce une grande incidence sur la fiabilite de l'arrivee des vols, mais aussi, que la fiabilite des vols n'a pas necessairement d'influence sur le cout des aliments dans les localites visees par notre analyse. Grace a ces constatations, nous pouvons soutenir que l'allongement des pistes pourrait ameliorer les dessertes aeriennes dans le Nord canadien.

Mots cles : Nord canadien; localites desservies par voie aerienne; infrastructure aeroportuaire; fret aerien; fiabilite des arrivees; aliments

Traduit pour la revue Arctic par Nicole Giguere.


The Canadian North is home to numerous remote communities, dependent for much or all of the year on aircraft for transporting people and cargo into and out of the area. Given this dependency, it is important to understand the reliability of air service to these communities, as their residents rely on these flights for everything from routine services (e.g., food and postal delivery, primary and public health care, and legal services) to emergency medical transport. Regular air service to the Canadian North, which allows these communities to participate in the broader national system, is constrained by basic airport infrastructure, challenging environmental conditions such as inclement weather and extreme cold, and the high cost of moving people and goods.

This paper investigates the relationship between airport infrastructure and the reliability of service to communities in northern Canada, with a particular focus on selected airports in the circumpolar North (specifically fly-in communities in Nunavut, Nunavik, and the Northwest Territories). Using data provided directly from two airlines that served this area with more than 18 500 scheduled flights in 2015, we demonstrate the links between airport infrastructure, community-level variables (e.g., population and income), and the reliability of service. In addition, we used publicly available data on food costs in these remote communities to understand how airport infrastructure can affect the daily lives of residents. Through investigating these topics, we identify potential strategies that would allow airlines to provide improved service to the tens of thousands of people who live across the northern territories and provinces of Canada (henceforth referred to as "the North").


Canada's northern communities are scattered across a vast geography that represents approximately 60% of Canada's land mass. Most residents of these communities are Indigenous Canadians, whose territorial homelands constitute the majority of Canada's northern jurisdictions. Communities range in size from 313 in Chesterfield Inlet, Nunavut, to 6699 in Iqaluit, Nunavut. Many of these communities lack surface transportation and are accessible by marine routes only during the short (and highly weather-dependent) open-water season (August to October), when ice is not a barrier. Thus, residents must access the vast majority of services via commercial aviation. Airlines bring service providers to communities, and residents fly south to access many of their health, social, legal, and education services (Adelson, 2000; Mitton et al., 2011; Young and Chatwood, 2011).

Most of the limited number of airlines that serve these remote communities are partly or wholly owned by Indigenous organizations, as stipulated in a number of northern land-claim settlements that require territorial governments to prefer contracts for the supply of goods and services from corporations that are majority Inuit-owned (Nunavut Tunngavik Inc. et al., 2010). These airlines operate fixed-wing aircraft equipped to land on gravel or ice strips in remote locations (Canadian North, 2016; First Air, 2016a). Fleets are generally composed of combination aircraft, which can be modified to carry varying load configurations of both cargo and passengers.

Community airports are maintained by the individual provinces and territories, with assistance from Transport Canada's Airports Operations and Maintenance Subsidy and Airports Capital Assistance programs (Transport Canada, 2016). Existing infrastructure is a combination of purpose-built facilities and legacy facilities built to serve Cold War airbases and surveillance stations such as the Distant Early Warning (DEW) line, a chain of 67 radar and communications centres stretching from Alaska to Greenland (Lajeunesse, 2007). For example, the airport in Cambridge Bay was originally built in 1955 to service a local DEW line radar station (Collier, 2012) that was decommissioned in 1985, but it continues to support the long-range radar component of the North Warning System, a joint defense initiative of Canada and the United States (Government of Canada, 2013).

One result of such arrangements is considerable variability among northern communities in airport infrastructure. In a 2013 report, the Canadian Chamber of Commerce (2013) articulated numerous challenges facing the northern transportation sector. One acute issue is the aging infrastructure that has not kept up with changes in technology, even simple improvements such as paving runways. The report notes that changing this situation would "entail a significant financial investment that industry, territorial and local governments simply are not capable of funding themselves" (Canadian Chamber of Commerce, 2013:1). The Northern Air Transport Association (2016) echoed these calls for increased federal support of airport infrastructure in a resolution passed at their 2016 annual conference.

While airport infrastructure varies, all communities under investigation in this study rely on the air network across the North. Larger hub communities (Yellowknife, Rankin Inlet, Iqaluit, Kuujjuaq) tend to be served by jet aircraft connecting with other hub communities or with southern cities such as Edmonton, Winnipeg, Ottawa, and Montreal (First Air, 2016b). Turbo-propeller aircraft fly to smaller communities, generally via circular routes that originate in hub communities.

The degree of dependence on aviation services cannot be overstated. Statistics Canada examined the annual number of passengers getting on or off aircraft as a proportion of the city's population and found that this ratio is much higher in northern hub communities like Iqaluit and Yellowknife than in southern cities, where alternative forms of transport exist. The ratio of passengers per capita is 15.1 in Yellowknife and 17.9 in Iqaluit (Dunlavy et al., 2009), compared to only 5.8 at Canada's busiest airport, Lester B. Pearson in Toronto. Passengers who must travel in the North include service providers (e.g., physicians and lawyers) flying to and from regional centres and remote communities, as well as residents flying to and from regional hospitals, treatment centres, schools, justice facilities, and other institutions and referral centres in southern cities (Browne, 2010; Canadian Polar Commission, 2014). In addition to northern communities, the resource-based industries that are prevalent throughout the region are highly dependent on reliable air passenger and cargo services. Further, outside the short open-water season, all mail, food, fuel, medical supplies, and other goods must be brought in and out by plane. A recent Conference Board of Canada report described the import of northern aviation: "Small airports are not only vital to their community's prosperity--ultimately, they may also determine whether or not a town or an industry is viable" (Gill and Raynor, 2013:26).

Perishable food is one example of cargo that is reliant on aircraft arrival and relates to both the well-being and health of residents in the North, as well as the economic viability of businesses in these communities (Sharma et al., 2010). Fresh fruit and vegetables, fresh and frozen meat, dairy products, and other perishable foods sold in retail stores arrive in remote northern communities as cargo shipments on commercial airlines. Given the significant value of these deliveries to communities, freight costs for many of these foods are subsidized under the Government of Canada's Nutrition North Canada (NNC) retail subsidy program (Burnett et al., 2015; Government of Canada, 2016). Because of the consistent need for food products, nearly all flights to the North carry food cargo.

As a way to examine how airport infrastructure in the North affects the well-being and economies of residents and businesses in these communities, we focus on the link between runway length and the proportion of scheduled flights that arrived as scheduled in 2015. To have "arrived as scheduled," a flight must have taken off and landed at its scheduled destination anytime on the scheduled day. This criterion includes delayed flights, but not flights cancelled at their origin or diverted to another airport. Additionally, we explore the relationship between the reliability of air transport and food cost as a preliminary case study to illustrate the potential impact of airport infrastructure on the provision of essential services and, more generally, on overall social and economic viability in remote northern communities.


We obtained data directly from two air carriers on northern flights scheduled during 2015, flights that arrived, and those that were cancelled or diverted. These carriers maintain extensive networks in northern Canada and are responsible for a substantial portion of the passenger and cargo traffic to and from northern communities. These flight data were processed, standardized, and combined for further analysis using a Python script that created a database documenting the number of flights scheduled and flights that arrived each day at each airport served. From the database, we calculated the percentage of scheduled flights that arrived in northern communities and were not diverted or cancelled. The dataset was reduced to a subset of 23 relevant communities that met the following criteria:

1) Each was a fly-in community, where access is restricted to air travel for more than six months of the year.

2) Each had at least 100 scheduled flights through the year. Some remote fly-in communities do not have scheduled service meeting this threshold serviced by the two studied airlines, but the data show an increased number of landings due to diversion; these communities are not considered in the analysis.

3) Each was eligible for a full NNC subsidy, which permits the use of food cost as an outcome measure of the reliability of air transportation.

Of the 23 communities selected for analysis, 20 are in the Canadian territory of Nunavut (NU), two are in the Northwest Territories (NT), and one is in northern Quebec (QC) (Fig. 1). We excluded Old Crow, Yukon, from the analysis because during the study year it had an ice road for several months and because it is served by a separate, Yukon-based airline from which we had no data.

With the study communities established, we used the Python script to compute the percentage of scheduled flights that arrived for every month of 2015, calculated as the total number of flights that arrived at an airport in a given month divided by the total number of flights scheduled to arrive at that airport in the same month. We then made the same calculation for the entire year (Table 1). In a few cases, the percentage of arrivals is greater than 100% because of unanticipated arrivals of flights diverted from other airports. Despite the "noise" introduced by diverted flights, these proportions capture the general reliability of service given the large total number of flights arriving at the study airports. If there was no service during a month, there is no monthly percentage calculation, and the yearly arrival percentage is calculated using only the months that have service.

In addition to flight arrival data, we collected data on socioeconomic status via the 2011 Census of Canada (Statistics Canada, 2011) and on airport infrastructure (represented by runway length) from the March 2016 NAV CANADA Canadian Airport Charts document (NAV CANADA, 2016). Finally, food costs were acquired from reports on the Revised Northern Food Basket (RNFB) published by the NNC subsidy program (Government of Canada, 2017). The RNFB is a 67-item instrument designed to monitor the average cost of food for a family of four for one week. It is used as a benchmark for subsidy efficacy in the North since the majority of food items in the RNFB are perishable foods delivered by air. Retailers in 91 northern communities report RNFB costs quarterly, in March, June, September, and December (Government of Canada, 2014; Galloway, 2017). We used data from 2014 RNFB reports because at the time of our analysis, data were available for only the first quarter of 2015. Basic descriptions of variables are presented in Table 2, and maps of median income, RNFB cost, runway length, and the percent of scheduled flights that arrived are shown in Figure 2. While the data come from different years, the values of these variables change relatively little from year to year. Therefore, we assume that the signal provided by each data source is relevant to our analysis.

In order to establish the relationships between arrival reliability, airport infrastructure, and community socioeconomic variables, we constructed correlation matrices and ordinary least squares regression models in R 3.3.1 (R Development Core Team, 2015) using the data described above. It is important at this point to mention the uniqueness of the community of Iqaluit, NU, which is substantially larger than the other communities studied. With 6699 residents, it is 2.8 times the size of Kuujjuaq, the next largest community (2375 residents). The runway at Iqaluit airport (YFB) is also substantially longer (8605 feet) than those in other communities considered for this study (average length: 4576 feet). The larger population and longer runway are a result of Iqaluit's having served as a major military airbase for the United States during the mid-20th century (Eno, 2003). Given Iqaluit's outlier status, we discuss our results both with and without the Iqaluit data.


Preliminary Analysis of Arrival Data

In examining the monthly percentage of flights that arrived as scheduled (Table 1), we found that, in general, more than 80% of all scheduled flights had arrived as planned. Comparable statistics on southern Canadian airports could not be found; however, as an example for comparison, the arrival rate at Chicago O'Hare International Airport in 2014 was 95% (Morris, 2015). A noticeable drop in arrivals at many study airports (e.g., YZS, YTE, and YUX) during July 2015 is likely related to particularly difficult weather conditions that plagued Nunavut during that month (Bell, 2015; Momin and Statham, 2015). When calculated for the entire year, the mean percentage of flights that arrived as scheduled was 88% (SD = 6%), with individual airports reporting averages as low as 75% (Kimmirut) and as high as 100% (Kuujjuaq).

Correlation Analysis

The two correlation matrices computed are displayed in Figure 3, which includes Iqaluit, and Figure 4, which excludes this outlier. In both figures, the boxes above the diagonal show the Pearson's correlation coefficient (r) and the significance levels between the corresponding variables, while those below the diagonal display the scatter plots between the two variables. It is immediately apparent that the inclusion of Iqaluit results in many more significant correlation coefficients, with eight correlations significant at p [less than or equal to] 0.05 and 10 at p [less than or equal to] 0.10. The correlation matrix that excludes Iqaluit shows only eight significant correlations, three at p [less than or equal to] 0.05 and five at p [less than or equal to] 0.10. Given the great influence Iqaluit has on these measures, we used only the results displayed in Figure 4 for interpretation.

Median income (a variable meant to capture social capital) is significantly and positively correlated with three other variables: runway length, the proportion of flights that arrive as scheduled, and the total number of scheduled flights. In addition to these relationships, runway length is significantly and positively associated with the proportion of flights that arrive as scheduled, and population size is significantly and negatively correlated with food costs.

These correlations all make intuitive sense, but it is interesting to point out two other findings from this matrix. First, the total number of scheduled flights to the study area airports was not correlated with the proportion of these flights that arrived as scheduled. This means that airports with more flights do not necessarily see more reliability in air traffic. Second, there is a strong and highly significant correlation (r = 0.59, p [less than or equal to] 0.01) between runway length and the proportion of flights that arrive as scheduled. This correlation is important because it implies that investments in lengthening runways may improve service to these remote communities.

Regression Analysis

Of four regression models, Models l and 2 (Table 3) explore the relationship between the dependent variable "average proportion of scheduled flights that arrive" and independent variables population, median income, runway length, and the total number of scheduled flights from 2015. Median income and population are included in the models to control for any general demographic and socioeconomic differences in communities. Models 3 and 4 (Table 4) explore the relationship between the dependent variable food cost and independent variables "average proportion of scheduled flights that arrive," population, median income, runway length, and the total number of flights scheduled in 2015. The communities of Chesterfield, Clyde River, Igloolik, and Pangnirtung were excluded from all models because there were no data on 2010 median incomes, while Kugaaruk was excluded from Models 3 and 4 because of missing data on food costs. Only Models 1 and 3 include Iqaluit.

Models 1 and 2 both have F statistics that indicate they represent moderately significant (p < 0.10) relationships between the dependent and independent variables. The r2 value for both models also suggests that more than 40% of the variance in the proportion of scheduled flights that arrived is explained by the independent variables. However, the only significant predictor in both models is runway length: an increase in runway length of 1000 feet is associated with a 4% increase in the number of flights that arrive as scheduled.

Models 3 and 4 are included in order to demonstrate the link between a ubiquitous social need (food), airport infrastructure, and arrival reliability. Again, residents in northern communities have multiple social, health, and institutional needs, but we used food cost as an example, controlling for previously discussed socioeconomic variables, because of the large amount of data available on the topic from the Government of Canada and because the cost of food affects all residents.

Models 3 and 4 both have a significant F statistic, indicating there is an overall significant relationship between the dependent and independent variables, and the [r.sup.2] values of 0.602 for Model 3 and 0.555 for Model 4 suggest that the independent variables can account for more than half of the variability in food costs. The average proportion of scheduled flights that arrive (p < 0.10) and the population in 2011 (p < 0.05) are significant predictors (p < 0.10 or less) for both models, while runway length (p < 0.10) and the total number of scheduled flights in 2015 (p < 0.05) are significant in the model including Iqaluit.

The direction and magnitude of the beta coefficients are consistent across both models. A positive relationship exists between arrival reliability and food costs, with both models indicating that a 1% increase in the proportion of scheduled flights that arrive corresponds to an increase of approximately $2.00 in food costs. Both models also show a negative relationship between population size and food costs: an increase of 100 people corresponds to a reduction in food costs of approximately $2.00. Finally, Model 3 also shows a significant negative relationship between runway length and food costs and a significant positive relationship between the total number of flights and food costs.

Summary of Findings

Overall, the correlation and regression analyses indicate that there is a link between airport infrastructure, as represented by runway length, and the average proportion of scheduled flights that arrive in the North. These findings suggest that the lengthening of runways in northern communities could lead to improved overall service, a particularly important outcome for residents, who rely on air travel for a large number of social and economic services. Longer runways not only allow for larger aircraft, but also for heavier aircraft (i.e., more cargo) and safer operations in poor weather conditions (Transport Canada, 1996; Federal Aviation Administration, 2005; St. John's International Airport Authority, 2015).

A second and important finding relates to the impact of runway length and the proportion of scheduled flights that arrive on the cost of food in northern communities. Interestingly, the models presented in Table 4 suggest a positive association between arrival reliability and food costs, but a negative association between runway length and food costs, controlling for population size, median income, and the total number of flights. Similar results are found in the correlation matrices. It makes intuitive sense that food costs would decrease with longer runways, as longer runways can handle larger aircraft with more cargo, allowing for more food deliveries. However, the positive association with arrival reliability is more difficult to explain. Our analysis actually demonstrated that food prices are higher in places where flight arrivals are relatively more reliable. It is worth noting that recent work (Galloway, 2014) has found little evidence that the NNC subsidy program (which the author critiques for a lack of transparency in how rates are calculated) is efficiently and effectively achieving its goal of providing available and affordable healthy foods. The intersection of retail and policy sectors could potentially lead to this counterintuitive finding. An additional confounding factor is weather. Airport operations may be impeded by extreme cold, precipitation, fog, and high winds. Future investigation will include a more nuanced approach to understanding the northern air transportation environment, including policy, human services, and weather patterns in study communities.


This research examined the relationships between airport infrastructure, socioeconomic variables, and the proportion of scheduled flights that arrive in the Canadian North, a region where residents are highly dependent on air transportation for cargo and a broad range of social, economic, and institutional services. Using a unique data source from two airlines that service dozens of airports in the North, we show that there is a significant and positive correlation between runway length and the reliability of scheduled flight arrivals. As northern communities are so dependent on air travel, the national and regional governments should consider investing in lengthening runways as a way to improve multiple facets of northern residents' lives. A second interesting finding is that there is conflicting information about how food costs, an economic consideration for every northern resident, are affected by air service and airport infrastructure. While longer runways were significantly associated with lower food costs, more reliable air service was significantly associated with higher food costs. This second, counterintuitive finding requires follow-up, as a more comprehensive analysis is required to understand how infrastructure, environmental conditions, and government subsidies affect food costs.

This paper is the first, to our knowledge, that explicitly explores how transportation infrastructure affects both air service reliability and a number of socioeconomic factors in remote, northern communities. Since economic pressures make maintaining these services difficult, it is critical that more research on this topic be conducted to improve our understanding of how to make air transport more efficient and effective.


The authors would like to acknowledge the anonymous reviewers and the journal editor for their help in making this paper stronger.


Adelson, N. 2000. Re-imagining Aboriginality: An Indigenous peoples' response to social suffering. Transcultural Psychiatry 37(1):11-34.

Bell, P. 2015. Iqaluit gets a month's worth of rain in 9 days. CBC News North, July 15. n-in-9-days-1.3147605

Browne, A. 2010. Issues affecting access to health services in northern, rural and remote regions of Canada. Northern Article Series. Prince George: University of Northern British Columbia.

Burnett, K, Skinner, K., and LeBlanc, J. 2015. From Food Mail to Nutrition North Canada: Reconsidering federal food subsidy programs for northern Ontario. Canadian Food Studies 2(1): 141- 156.

Canadian Chamber of Commerce. 2013. Air transportation infrastructure in Canada's Territories. Ottawa. 1-93a5-000c29c04ade

Canadian North. 2016. Our fleet. Calgary, Alberta: Canadian North.

Canadian Polar Commission. 2014. Health and well-being in the Canadian North: Recent advances and remaining knowledge gaps and research opportunities. and wellbeing_summary.pdf

Collier, K. 2012. To the 69th parallel (Part 2): On the DEW Line. The Newfoundland & Labrador Independent, March 26.

Dunlavy, J.P., Lipai, M., and Baldwin, G. 2009. Transportation in the North. Ottawa: Statistics Canada. 10820-eng.htm

Eno, R.V. 2003. Crystal Two: The origin of Iqaluit. Arctic 56(1):63-75.

Federal Aviation Administration. 2005. AC 150/5325-4B Runway length requirements for airport design. Washington, D.C.: U.S. Department of Transportation. circulars/ index, cfm/go/document.information/documentID/22809

First Air. 2016a. Our fleet. Kanata, Ontario: First Air.

--. 2016b. Route map. Kanata, Ontario: First Air.

Galloway, T. 2014. Is the Nutrition North Canada retail subsidy program meeting the goal of making nutritious and perishable food more accessible and affordable in the North? Canadian Journal of Public Health 105(5):e395-e397.

--. 2017. Canada's northern food subsidy Nutrition North Canada: A comprehensive program evaluation. International Journal of Circumpolar Health 76(1), 1279451.

Gill, V., and Raynor, R.N. 2013. Growing Canada's economy: A new National Air Transportation Policy. Ottawa: The Conference Board of Canada. airpolicy.aspx

Government of Canada. 2013. North Warning System: Backgrounder. Ottawa: National Defence and the Canadian Armed Forces.

--. 2014. Eligible communities. Ottawa: Nutrition North Canada. 1415540791407

--. 2016. Nutrition North Canada. Ottawa. 1415385790537

--. 2017. Nutrition North Canada: Reports on the cost of the revised northern food basket. 1415647437113

Lajeunesse, A. 2007. The Distant Early Warning Line and the Canadian battle for public perception. Canadian Military Journal 8(2):51-59.

Mitton, C., Dionne, F., Masucci, L., Wong, S., and Law, S. 2011. Innovations in health service organization and delivery in northern rural and remote regions: A review of the literature. International Journal of Circumpolar Health 70(5):460 472.

Momin, A., and Statham, S. 2015. What happens when there's ridiculous fog in Iqaluit for 11 days straight. Finding True North Blog.

Morris, C. 2015. These are the airports where your flight will be delayed or canceled. Fortune, April 15.

NAV CANADA. 2016. Canadian airports charts. Ottawa. aeronautical-in formation-products-canadian-airports-charts. aspx

Northern Air Transport Association. 2016. Funding for northern airports. Resolution Number 2016-3. Yellowknife: NATA.

Nunavut Tunngavik Inc., Minister of Indian Affairs and Northern Development, and Federal Interlocutor for Metis and NonStatus Indians. 2010. Agreement between the Inuit of the Nunavut Settlement Area and Her Majesty the Queen in right of Canada. Ottawa: Indian and Northern Affairs Canada. CLAIMS AGREEM ENT_NUNAVUT.pdf

R Development Core Team. 2015. R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing.

Sharma, S., Gittelsohn, J., Rosol, R., and Beck, L. 2010. Addressing the public health burden caused by the nutrition transition through the Healthy Foods North nutrition and lifestyle intervention programme. Journal of Human Nutrition and Dietetics 23(Suppl. 1):120 127. http://doi:10.1111/j.l365-277X.2010.01107.x

Statistics Canada. 2011. 2011 Census of Population Program: Data products. Ottawa. index-eng.cfm

St. John's International Airport Authority. 2015. Airfield accessibility and safety initiative FAQs. http://stjohnsairport.eom/wp-content/uploads/2015/07/FAQsAirfield-Accessibility-Safety-Initiativc2.pdf Transport Canada. 1996. Canadian aviation regulations (SOR/96433).

--. 2016. Funding programs.

Young, T.K., and Chatwood, S. 2011. Health care in the North: What Canada can learn from its circumpolar neighbours. Canadian Mcdical Association Journal 183(2):209-214.

Michael J. Widener, (1) Shoshanna Saxe (2) and Tracey Galloway (3)

(1) Department of Geography and Planning, University of Toronto - St. George, 100 Saint George Street, Toronto, Ontario M5S 3G3, Canada;

(2) Department of Civil Engineering, University of Toronto - St. George, 35 Saint George Street, Toronto, Ontario M5S 1A4, Canada

(3) Department of Anthropology, University of Toronto - Mississauga, 3359 Mississauga Road, Mississauga, Ontario L5L 1C6, Canada

(Received 21 November 2016; accepted in revised form 9 June 2017)

Caption: FIG. 1. Map of the 23 fly-in communities in the North examined in this study. Population values come from the 2011 Canadian Census (Statistics Canada, 2011).

Caption: FIG. 2. Maps of income, food basket cost, runway lengths, and percentage of arrivals.

Caption: FIG. 3. Correlation matrix between variables with Iqaluit (YFB) included. The magnitude of values is provided on the respective row or column of a given variable. The units of a row or column are provided along the diagonal of the figure.

Caption: FIG. 4. Correlation matrix between variables with Iqaluit (YFB) excluded. Details as in Figure 3.
TABLE 1. The monthly and annual percentages of flights scheduled in
2015 that arrived at the 23 communities included in this study.

Airport    Place                    Runway    Jan%   Feb%   Mar%
code                                 (ft)

YAB        Arctic Bay, NU            3935      91     94     91
YBB        Kugaaruk/Pelly Bay, NU    5000      75     86     90
YBK        Baker Lake, NU            4195      70     84     72
YCB        Cambridge Bay, NU         5076      83     95     90
YCO        Kugluktuk, NU             5502      84     82     87
YCS        Chesterfield Inlet, NU    3600      75     93     83
YCY        Clyde River, NU           3501      94     93     96
YEK        Arviat, NU                4000      73     73     78
YFB        Iqaluit, NU               8605      94     89     95
YGT        Igloolik, NU              4095      91     91     89
YHI        Ulukhaktok/Holman, NWT    4300     100     79     94
YHK        Gjoa Haven, NU            4400      87     90     93
YIO        Pond Inlet, NU            4006      93     92     92
YLC        Kimmirut, NU              1899      88     87     89
YRT        Rankin Inlet, NU          6000      75     79     86
YTE        Cape Dorset, NU           3988      81     83     90
YUX        Hall Beach, NU            5410      84     82     85
YVM        Qikiqtarjuaq, NU          3803      91     90     89
YVP        Kuujjuaq, QC              6000      98    104    102
YVQ        Norman Wells, NWT         5998      99     98     93
YXP        Pangnirtung, NU           2920      86     89     92
YYH        Taloyoak, NU              4009      85     92     96
YZS        Coral Harbour, NU         5006      85     86     78
                                     Mean      87     88     91

Airport    Place                    Apr%   May%   Jun%   Jul%   Aug%

YAB        Arctic Bay, NU            96    100    100     94    100
YBB        Kugaaruk/Pelly Bay, NU    90     89     80     95     97
YBK        Baker Lake, NU            90    100     --     --     --
YCB        Cambridge Bay, NU         98     94     91     94     96
YCO        Kugluktuk, NU             96     95     95     98     95
YCS        Chesterfield Inlet, NU    93     97     --     --     --
YCY        Clyde River, NU           93     89     92     88     94
YEK        Arviat, NU                91     93     --     --     --
YFB        Iqaluit, NU               93     87     87     76     91
YGT        Igloolik, NU              93     87     86     59     92
YHI        Ulukhaktok/Holman, NWT   100     93     88    100    100
YHK        Gjoa Haven, NU            95     88     90     93     97
YIO        Pond Inlet, NU           100     92     87     90     91
YLC        Kimmirut, NU              71     65     76     59     83
YRT        Rankin Inlet, NU          87     98     94     87    100
YTE        Cape Dorset, NU           85     85     73     44     81
YUX        Hall Beach, NU            90     85     82     55     91
YVM        Qikiqtarjuaq, NU          86     66     57     76     69
YVP        Kuujjuaq, QC             100    102    100     98    100
YVQ        Norman Wells, NWT         97     97    100     98    100
YXP        Pangnirtung, NU           82     70     56     70     80
YYH        Taloyoak, NU              94     91     85     98     96
YZS        Coral Harbour, NU         89     94    100     25     92
                                     92     89     86     80     92

Airport    Place                    Sep%   Oct%   Nov%   Dec%

YAB        Arctic Bay, NU            98     73     88     95
YBB        Kugaaruk/Pelly Bay, NU    86     81     97     88
YBK        Baker Lake, NU            --     --     --     --
YCB        Cambridge Bay, NU         91     96     94     98
YCO        Kugluktuk, NU             89     98     92     97
YCS        Chesterfield Inlet, NU    --     --     --     --
YCY        Clyde River, NU          100     90     98     91
YEK        Arviat, NU                --     --     --     --
YFB        Iqaluit, NU               92     89     88     87
YGT        Igloolik, NU              88     86     89     80
YHI        Ulukhaktok/Holman, NWT    93    100     81     93
YHK        Gjoa Haven, NU            84     90     77     87
YIO        Pond Inlet, NU           100     86     97     95
YLC        Kimmirut, NU              82     74     71     75
YRT        Rankin Inlet, NU          99     98     82     81
YTE        Cape Dorset, NU           84     92     75     72
YUX        Hall Beach, NU            93     98     90     84
YVM        Qikiqtarjuaq, NU          88     90     86     86
YVP        Kuujjuaq, QC             102    100     98     96
YVQ        Norman Wells, NWT         98     98     97     98
YXP        Pangnirtung, NU           94     93     85     88
YYH        Taloyoak, NU              85     78     83     90
YZS        Coral Harbour, NU         83     93     64     69
                                     92     90     87     88

Airport    Place                    Weighted
code                                  mean

YAB        Arctic Bay, NU              94
YBB        Kugaaruk/Pelly Bay, NU      88
YBK        Baker Lake, NU              84
YCB        Cambridge Bay, NU           93
YCO        Kugluktuk, NU               92
YCS        Chesterfield Inlet, NU      88
YCY        Clyde River, NU             93
YEK        Arviat, NU                  81
YFB        Iqaluit, NU                 89
YGT        Igloolik, NU                85
YHI        Ulukhaktok/Holman, NWT      93
YHK        Gjoa Haven, NU              89
YIO        Pond Inlet, NU              93
YLC        Kimmirut, NU                75
YRT        Rankin Inlet, NU            88
YTE        Cape Dorset, NU             80
YUX        Hall Beach, NU              84
YVM        Qikiqtarjuaq, NU            80
YVP        Kuujjuaq, QC                100
YVQ        Norman Wells, NWT           98
YXP        Pangnirtung, NU             81
YYH        Taloyoak, NU                90
YZS        Coral Harbour, NU           84

TABLE 2. Descriptive statistics for the six variables examined
in 23 northern communities.

                           N      Mean        SD      Minimum

Runway length (feet)       23   4576.00    1336.74      1899
Number of scheduled        23    816.35     963.10       142
 flights (2015)
Arrivals as percentage     23      0.88       0.06      0.75
 of scheduled flights
Population (2011)          23   1428.65    1304.70       313
Median income (2010)       19   27028.21   14400.00    16080
Average RNFB cost (2014)   22    434.40      29.20    377.590


Runway length (feet)         8605
Number of scheduled          4968
 flights (2015)
Arrivals as percentage       1.00
 of scheduled flights
Population (2011)            6699
Median income (2010)        67082
Average RNFB cost (2014)   475.998

TABLE 3. Results from regression models exploring arrival
reliability. Standard errors are presented in parentheses beneath
their corresponding beta coefficients.

Dependent variable: Average proportion of scheduled flights that arrive

                                   Model 1            Model 2
                                (with Iqaluit)   (without Iqaluit)

Population (2011)                  0.00000            0.00000
                                  (0.00002)          (0.00002)
Median income (2010)               0.00000            0.00000
                                  (0.00000)          (0.00000)
Runway length (ft)                0.00004 **         0.00004 *
                                  (0.00002)          (0.00002)
Number of scheduled flights        -0.00004          -0.00002
 (2015)                           (0.00003)          (0.00010)
Constant                           0.698***          0.695 ***
                                   (0.061)            (0.063)
Observations                          19                18
[r.sup.2]                           0.437              0.441
Adjusted [r.sup.2]                  0.276              0.269
Residual SE                         0.055              0.056
                                  (df = 14)          (df = 13)
F statistic                        2.715 *            2.566 *
                                 (df = 4, 14)      (df = 4, 13)

* p < 0.1; ** p < 0.05; *** p < 0.01

TABLE 4. Results from regression models exploring food cost.
Standard errors are presented in parentheses beneath their
corresponding beta coefficients.

Dependent variable: Average RNFB cost 2014

                                Model 3            Model 4
                             (with Iqaluit)   (without Iqaluit)

Average arrival proportion     215.717 *          214.706 *
                               (106.689)          (111.639)
Population (2011)              -0.023 **          -0.022 **
                                (0.008)            (0.009)
Median income (2010)            -0.0004            -0.0004
                                (0.001)            (0.001)
Runway length (ft)              -0.014 *           -0.015
                                (0.008)            (0.009)
Number of scheduled             0.035 **            0.038
 flights (2015)                 (0.012)            (0.021)
Constant                      326.809 ***        327.117 ***
                                (78.060)          (81.509)
Observations                       18                17
[r.sup.2]                        0.602              0.555
Adjusted [r.sup.2]               0.436              0.352
Residual SE                      21.664            22.611
                                (df= 12)          (df = 11)
F statistic                     3.625 **          2.738 **
                              (df = 5, 12)       df = 5, 11)

* p < 0.1; ** p < 0.05; *** p < 0.01
COPYRIGHT 2017 Arctic Institute of North America of the University of Calgary
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2017 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Widener, Michael J.; Saxe, Shoshanna; Galloway, Tracey
Article Type:Report
Geographic Code:1CONT
Date:Sep 1, 2017
Previous Article:Seasonal Patterns in Ocean Ambient Noise near Sachs Harbour, Northwest Territories.
Next Article:Evaluating Evidence for Historical Anadromous Salmon Runs in Eklutna Lake, Alaska.

Terms of use | Privacy policy | Copyright © 2021 Farlex, Inc. | Feedback | For webmasters