A regional comparison of the implications of climate change for the golf industry in Canada.
Outdoor recreation and related tourism in Canada is inherently sensitive to climate conditions, and consequently, climate change could have potentially important and disparate implications for specific outdoor recreation industries. Recent studies have demonstrated that projected changes in the climate are relevant to the future economic success of Canada's skiing industry (Scott et al. 2003; Scott et al. 2007), natural seasonality in visitation to national parks (Jones and Scott 2006) and even the timing and quality of recognized tourism events (Scott et al. 2005; Jones et al. forthcoming). Yet, in comparison to other economic sectors of Canada's economy (e.g., agriculture, forestry), our understanding of the implications of climate change for outdoor recreation and tourism is much less advanced.
One of the largest recreation sectors in Canada is the golf industry. There are approximately 2,000 golf courses (Andrew 1998) and some 5 million amateur golfers in Canada (Royal Canadian Golf Association 2004). Canadians played an estimated 65 million rounds of golf in 2001 (Royal Canadian Golf Association 2002). As of 2003, the Canadian golf industry had operating revenues in excess of CDN$1.9 billion (Statistics Canada 2005).
An important determinant in the annual success of the Canadian golf industry is the climate. Variations in the climate from year to year can have positive and negative impacts on the industry. For example, above normal seasonal autumn temperatures allowed many Winnipeg-area golf courses to re-open for several weeks in November 2002 (Winnipeg Free Press 2002). However, a cool, wet spring in 2002 had a discernable negative impact on the number of rounds played and golf course revenues in many parts of Canada. Rounds played in Edmonton in May (2002) were 23 percent lower than average and 10 percent to 20 percent lower in parts of Ontario (Doey and Wong 2002). Cool and wet conditions during the spring and summer of 2004 in many parts of Canada also contributed to reductions in rounds played. In southern Ontario, rain occurred on 78 of the 142 days between 1 April and 21 August, contributing to reductions in rounds played by as much as 20 percent at some courses (Herbert 2004). Cool, wet conditions resulted in reductions in rounds played by as high as 40 percent at some golf courses in southern Manitoba (Canadian Press 2004; Winnipeg Free Press 2004). A hot summer and several prolonged heat waves (humidex values ~+35[degrees]C) in Ontario in 2005 contributed to reductions in rounds played during the warmest part of the day (Herbert 2005).
Climate is the long-term average weather at a given location, and is important to the golf industry because it has a direct role in influencing the length and quality of operating seasons at golf courses in a given location. Scott and Jones (2006) argued that any direct changes in the length and quality of golf operating seasons (i.e., where golf seasons are currently less than 365 days) induced by climate change would present opportunities to increase the number of rounds played. Increases in golf participation generated by a warmer climate would have benefits for golf course revenues and the economies of seasonal golf-based tourism destinations. However, longer golf seasons in some Canadian locations would also alter the competitive advantage among golf destinations, and would have important implications for golf course management (i.e., irrigation, turf grass selection and turf disease, and pest management).
Limited progress has been made in assessing the potential impact of climate change on the Canadian golf industry. Lamothe & Periard Consultants (1988) used climatological criteria developed by Crowe et al. (1977) for Ontario to define a desirable golf day in Quebec. Under a doubled atmospheric carbon-dioxide scenario (~2050s), they projected that the golf season in southern Quebec would be extended 6 weeks from its current 29-week average season. This early study of climate change and golf, however, had a number of limitations. First, the criteria used to define a desirable golf day were based on expert opinion (from interviews with golf course managers and officials from the Toronto Department of Parks and Recreation), not observations of how actual golf participation varied with weather conditions. Second, their method was unable to provide insight into the projected change in rounds played, which is an important economic indicator for the golf industry. Third, it is unclear how monthly global climate models (GCMs) scenarios were downscaled to provide daily climate change scenarios for this study. Finally, validation of the climate change assessment results for the Quebec area (e.g., using a spatial analogue) was not considered.
More recently, Scott and Jones (2006) attempted to address the weaknesses in this earlier climate change and golf study by developing an empirical regression relationship between daily weather variables and observed rounds played in order to examine the potential impacts of a changed climate on the golf industry in the Greater Toronto Area (GTA) (Canada). They concluded that as early as the 2050s, the golfing season in the GTA could be extended 12 weeks and rounds played at an average 18-hole golf course would increase between 27 percent and 61 percent under a range of climate change scenarios; by the 2080s, golf in the GTA could be a year-round intermittent activity under the warmest climate change scenario. Under the warmest climate change scenario for the 2080s, the GTA was projected to have a climate similar to Columbus, Ohio (40[degrees]0'N, 83[degrees]'1'W) today. When Scott and Jones (2006) compared their modelled climate change participation projections for the GTA in the 2080s with current average golf operating seasons in Columbus, their modelled season length was within 3 percent of Columbus' current average season length and modelled average annual rounds played were within 2 percent.
Canada is large and climatically diverse. Consequently, a limitation of previous case studies is that they are not generalizable to other regions of Canada where the climate, and consequently, golf supply and participation differ. In spite of the importance of climate to the Canadian golf industry, no study has yet examined the regional impacts of projected changes in the climate on the industry. This article is concerned with the impacts of climate change for the golf industry in Canada and explores if and when (2020s, 2050s and 2080s) projected climate change would have a meaningful impact on the golf industry in three climatically different regions of the country. The two specific objectives of this study were to: (1) develop a model of climate and golf participation for each study region and (2) use the model to assess how projected climatic changes could affect the future of the golf industry in each region through changes in the length of the golf season and the number of rounds played. Implications of the projected changes for the Canadian golf industry are also explored.
Canada is a large country with many prevailing climates. The marine climate of Canada's west coast (Koeppen classification--Cfb) is the mildest and wettest in the country. Moderated by the Pacific Ocean, mean temperatures seldom fall below freezing, even in winter, and precipitation is disproportionately higher during the fall and winter seasons. The southern Prairies generally experience a semi-arid, mid-latitude steppe climate (Koeppen classification BSk), characterized by hot, dry summers and very cold winters in which wind chills of -40[degrees]C are not an uncommon occurrence. The Great Lakes region tends to experience a humid continental climate (Koeppen classification--Dfa). The summers are typically warm with frequent episodes of high humidity (associated with advancing warm fronts from the U.S. south); winters tend to be cold, but not to the extent of the Prairie climate, with large amounts of lake-effect snow in some areas. Canada's east coast climate (Koeppen classification--Dfc) is typified by short, mild summers and cold winters; the mixing of cold and warm ocean currents off the Atlantic coast tend to also produce the region's characteristic advection fogs.
Canada's different climates play an important role in influencing regional golf operations. Considering the climate sensitivities that would occur among golf courses in different parts of the country, many (60) mid- to high-quality courses were contacted to provide data for this study. Several course superintendents expressed interest in the study, bur were unable to release what was considered proprietary business information. Inconsistency in record keeping was another important data barrier. Some golf courses recorded participation as the number of rounds paid for each day, bur not necessarily played (i.e., purchases of multiple green fee packages or tournaments were recorded when they were paid for not when they were actually played). Technological limitations in course management software posed another barrier, as some golf courses could only provide monthly averages for rounds played or averages for specific days of the week (i.e., all Mondays, Saturdays, or holidays).
Only four golf courses were able and willing to provide the necessary data, and course officials did so with the agreement of anonymity (i.e., officials agreed to provide data provided that the names of the courses were not identified). Although the sample was small, it was the opinion of the participating course superintendents, who have expertise in their marketplace and which courses are comparable in quality and business conditions, that their course was fairly typical of course operations for high-quality courses in their respective locations. The data are less representative of some municipally run courses that sometimes operate under marginal weather conditions (e.g., early spring and late autumn) when mid-to-high-quality courses would normally not operate because of the risk of damage to the course. As this study was the first regional impact assessment of the golf industry, we acknowledge it would be beneficial if future studies assessed additional courses in these and other areas of Canada to provide a more detailed national perspective on the potential impacts of climate change.
Table 1 summarizes the characteristics of the four golf courses used in this study, including the prevailing Koeppen climate classification for each study region. The two golf courses in the Great Lakes region are located approximately 250 km apart. Although these two golf courses are located in close proximity to one another, they represent different golf markets within the Great Lakes region and different sub-regional climates within the Province of Ontario. All four golf courses are located within or in close proximity (~10 minutes) of urban areas, operate at between 80 percent and 90 percent of capacity during their respective peak seasons, and have been distinguished as among Canada's best golf courses by Score Magazine and Golf Digest.
Data and analysis
This study utilized daily, recorded golf participation data (number of rounds played) from the four golf courses. Table 1 outlines the observed record of golf data available. Analysis of the influence of climate on golf participation was undertaken using station data from the nearest Meteorological Service of Canada climate station with a high-quality record of climate data (1) (Table 1).
The climate change scenarios for each location were developed from monthly GCMs available from the Canadian Climate Impacts and Scenarios (CCIS) Project. The scenario data available from CCIS are constructed in accordance with the recommendations set out by the United Nations Intergovernmental Panel on Climate Change (IPCC) Task Group on Scenarios. Three future timeframes were examined, each of which were based on a 30-year period of climate data (i.e., the 2020s represent the period 2010 to 2039; the 2050s represent 2040 to 2069; and, the 2080s represent 2070 to 2099). All scenarios represent climate changes with respect to the 30-year baseline climate (i.e., 1961-1990).
In accordance with IPCC recommendations, two GCMs and climate change scenarios were used in this study to capture the range of plausible future climates in Canada. The two GCMs and climate change scenarios utilized in this study are the National Center for Atmospheric Research (NCARP) GCM with the B2 scenario (NCARPCM B21--a low greenhouse gas emission future) and the Center for Climate System Research (CCSR) GCM with the Al scenario (CCSRNIES All--a high greenhouse gas emission future). The two climate change scenarios were selected from 19 available scenarios to represent a broad range of projected temperature and precipitation changes in Canada. These two scenarios effectively represent the high and low ends of anticipated climate change in Canada, and in this study are referred to as the 'least-change' climate change scenario (NCARPCM B21) and the 'warmest' climate change scenario (CCSRNIES All).
A difficulty noted by the climate change impacts research community is that many impact assessments, including this study, require climate change information at finer temporal and spatial scales than are generally available from GCMs (Wilby et al. 2004). There are several methodological approaches to producing higher-resolution climate change scenarios, including regional climate models (RCMs), statistical downscaling, spatial and temporal analogues, and simple application of climate change factors to a reference climate. As noted by a report prepared by Wilby et al. (2004), all have strengths and weaknesses depending on the application. RCM scenarios were not available for all of the study regions in this analysis, so a weather generator for statistical downscaling was used. Monthly GCM scenarios were downscaled to the daily level and parameterized to the respective climate stations using the Long Ashton Research Station (LARS) stochastic weather generator (Semenov et al. 1998). Weather generators are inexpensive computational tools that replicate the statistical attributes of a local climate and can be used to produce site-specific, multiple-year, climate change scenarios at the daily level (Semenov et al. 1998; Wilby et al. 2004). LARS was selected for this study because it has been found to simulate precipitation statistics in the regions of Canada examined in this study better than other weather generators (Qain et al. 2004).
As defined in the introduction to this article, climate is the long-term average weather at a given location. To understand the impact of climate on golf operations, the influence of weather on daily rounds played at each golf course was first examined using methods consistent with Scott and Jones (2006). These methods involved calculating the average number of rounds played daily with respect to a range of daily precipitation amounts (i.e., 0 mm, <2.5 mm, 2.5 to 5 mm, 5 to 10 mm) and the timing of precipitation (morning, afternoon, all day). The weather characteristics (e.g., >30 mm of precipitation in the previous 24 hours) associated with course closures (e.g., 2 days, 1 week) during the operating season were also identified. The relationship between daily rounds played and temperature and wind speed were also examined using regression analyses in order to identify climatic thresholds (i.e., maximum temperature or maximum wind speed) beyond which rounds played declined. Since the primary focus of this article is to assess the potential impact of climate change on regional golf participation in Canada, a summary of the impact of weather on daily rounds is provided in Table 2, but not elaborated on further in this article.
To assess future golf participation at the four golf courses under a changed climate, regression analysis was used to develop an empirical relationship between current patterns of daily golf participation and weather variables that were available as daily outputs in the current generation of climate change scenarios (i.e., daily timing of precipitation, humidity, and fog are not available from all GCMs). The regression analysis was performed on daily rounds played at each golf course for the years of golf participation data available (Table 1) using three weather variables (daily maximum and minimum temperature and daily total precipitation) and one temporal variable (day of the week). Day of the week was included because it plays an important role in influencing the timing of peak participation on short time scales and it could be used in the climate change impact assessment. Days of the week were given an ordinal classification based on the level of participation (the pattern of daily rounds played at individual golf courses). For example, if the average number of rounds played at the West Coast golf course was highest on Fridays and lowest on Tuesdays, Fridays were given an ordinal rank of 2 in the regression analysis, while Tuesdays were ranked as 0 and all other days of the week were ranked as 1. The resulting regression models for each golf course were then run with weather data from their respective climate station for the 1961-1990 period to establish the 30-year climatological baseline, against which future climate change scenarios would be compared. This methodological step should not be construed as an attempt to model actual participation in any given year between 1961 and 1990. Rather, the purpose of running the regression models with the climate data from 1961 to 1990 was to establish the number of rounds played in a climatologically average year during the baseline period and 30 years is the standard used by climatologists to establish climatic averages. The regression models were then run with both climate change scenarios (NCARPCM B21 and CCSRNIES All) to assess changes in annual golf participation under climates projected for the 2020s, 2050s, and 2080s at each of the study areas.
The number of rounds played daily varies throughout the golf operating season in Canada, and are typically lowest during the opening and closing weeks of the operating season and highest during the summer period. Since daily data on rounds played were used in the regression models developed in this study, a participation indicator was used to assess changes in the average length of golf seasons. The average number of rounds played during the first and last 14 days of the golf season in the observed data was calculated and used to assess changes in season length at the four golf courses used in this study in the 2020s, 2050s and 2080s.
The potential impact of climate change on the golf sector was assessed exclusively in this study. It is recognized that other major social (e.g., population growth) and economic (e.g., cost of participation) factors could affect golf participation in the study regions in the future. However, with few golf market analyses available to project future market growth, it was assumed current levels of participation would not change. In this respect, the projections of rounds played in the future are likely conservative, as available market research suggests golf demand is likely to increase in Canada (Region of Halton 2000; Royal Canadian Golf Association 2002) and the U.S. (National Golf Foundation 2000, 2004).
Seasonality of golf supply and participation in Canada is climate limited. Communications with a number of golf course mangers revealed that operating seasons can be as long as 365 days in some regions of Canada (e.g., Victoria and Vancouver, BC), but in most geographic regions, climate plays a important role in limiting golf operating seasons to less than 250 days. Regionally, West Coast golf courses tend to have the longest operating seasons in the country, while East Coast courses can have some of the shortest.
Figure 1 illustrates the operating seasons of the four golf courses in this study. As anticipated, the mild marine climate of Canada's west coast contributed to the region's representative golf course having the longest operating season (365 days). Although West Coast golf courses have the potential for year-round operation, many courses close for several weeks in late January to perform grounds maintenance before opening up for the official start of the region's golf season in early February. Golf operating seasons in the Great Lakes region were shorter, with openings typically occurring in early-to-mid April and closures occurring in mid-to-late November. The two-golf courses analyzed in this region typified this seasonal pattern. During each of 2002 and 2003, the golf course in southern Ontario opened on 16 and 24 April (Julian days 106 and 114), and closed on 15 and 24 November (Julian days 320 and 338), respectively, resulting in a season length of 214 days. In each of 2002 and 2003, the central Ontario golf course operated for 192 days, opening on 18 April (Julian day 108) and closing on 27 and 26 October (Julian days 331 and 330). Of the four golf courses analyzed, the East Coast golf course had the shortest season-approximately 168 days. The East Coast golf course opened (mid-May; ~Julian day 135) and closed (late October; ~Julian day 300) on approximately the same day over the 4 years of observed participation data.
[FIGURE 1 OMITTED]
It is also evident from Figure 1 that golf participation in Canada varies throughout the year regardless of the region. In ali bur the West Coast golf course, golf participation during the spring (April and May) and fall (September to November) months was found to be lower than during the summer peak months of June, July and August. In the first few weeks of the season, fewer than 100 rounds of golf were played per day on average at the Great Lakes and East Coast courses, which is approximately 50 percent less than the summer peak (~180 to 200 rounds per day). In the last few weeks of the season, only 50 rounds of golf were played per day (~25 percent of peak participation). The distinction between participation levels in the first and last few weeks of the season likely reflects pent-up desire to play golf in the spring after the winter off-season. By comparison, golf participation was highest during the spring and summer (April to August) months and lowest during the winter (November to January) months at the West Coast golf course. The effect of natural seasonality on average rounds played at the West Coast golf course was also less noticeable than in the other three regions: average rounds of golf played per day during the winter months were approximately 50 percent of peak participation (~240 rounds per day).
The principal objective of this study was to explore the regional impacts of projected changes in climate on annual golf participation in Canada. In order to assess the potential impact of climate change on golf participation, a multiple regression analysis that included three weather variables (maximum and minimum temperature, total precipitation amount) and one temporal variable (day of the week) was completed. The regression analysis revealed that all four variables were important predictors of daily rounds played throughout the year in three of the four study regions. Minimum temperature was revealed to have limited predictive power in rounds played at the West Coast golf course. This finding is to be expected as temperatures rarely fall below freezing and golf is a year-round activity in this region. Table 3 summarizes the climate-golf participation regression models developed for each of the four golf courses in this study.
An important limitation of multiple linear regression models is that they continue to project an increase in rounds played as the maximum temperature increases. To account for the effect of an upper temperature threshold on golf participation, a high heat adjustment factor was added to each of the four models. A base temperature of 32[degrees]C was used because it is considered a reasonable proxy for heat-related physical discomfort and physiological heat stress in Canada in vulnerable populations (e.g., the elderly) (Mills et al. 2002). Based on the observed relationship between declining participation and increasing maximum temperature at all four golf courses, modelled and projected rounds played were reduced a certain percentage for every 1[degrees]C increase over 32[degrees]C--West Coast (-2 percent), Great Lakes 1 (-6 percent), Great Lakes 2 (-4 percent) and East Coast (-8 percent). When the adjusted regression models were run with climate data for the 30-year climatological baseline period, the difference between modelled and observed annual rounds played at all four golf courses was small. Modelled average number of rounds played per year was within 10 percent of observed annual rounds at three of the four golf courses (West Coast, Great Lakes 1 and East Coast); modelled annual rounds for the fourth course (Great Lakes 2) were within 15 percent of observed values. These models were deemed suitable for climate change impact assessments, as some climatological (e.g., inaccurate forecasts, high winds, fog) and non-climatological (e.g., promotional events, competition among businesses to open the earliest and close the latest) factors could not be accounted for in the models.
Impact of Climate Change on the Golf Industry
Due to Canada's geographic size, the nature and magnitude of climatic changes are projected to vary from one region of the country to another, potentially contributing to unequal regional impacts on golf season lengths and number of rounds played. Table 4 summarizes the projected changes in mean annual temperature and total annual precipitation projected for the four regions in this study based on the least-change (NCARPCM B21) and warmest (CCSRNIES All) climate change scenarios.
In this study, it was assumed that golf course managers in Canada would adapt to any new opportunities provided by a changed climate in their respective regions. Consequently, the four climate-participation models developed in this study were run with the assumption that golf courses would open and close when participation reached levels typical of current opening and closing weeks. Table 5 summarizes the projected changes in annual golf participation at the four golf courses used for this study. Comparison of the projected changes under the two climate change scenarios consistently indicated a trend toward longer golf seasons and higher golf participation across the study areas, largely because of more conducive weather conditions during the months of April to May and October to November (Figure 2). However, there were some important regional differences in the magnitude of change.
[FIGURE 2 OMITTED]
In the West Coast region, there was no extension to the golf season length, as the golf course operates year round at the present time. The West Coast golf course also benefited the least of the four golf courses studied from the projected regional climate changes (i.e., warmer and somewhat drier). There was no noticeable increase (i.e., less than 5 percent) in the average number of rounds played per season under the least-change climate change scenario (NCARPCM B21) in the 2080s. Under the warmest climate change scenario (CCSRNIES All), average number annual rounds played were projected to increase minimally (2 percent in the 2020s, 11 percent in the 2050s and 18 percent in the 2080s), with most of the increases occurring during the summer months of July and August.
Golf participation was projected to experience substantial growth in the Great Lakes region, although there was an important difference in the magnitude of change between central and southern parts of the region. In central Ontario, there was only a moderate (10 to 18 days) increase in the average golf season under the NCARPCM B21 scenario through the end of the twenty-first century, but under the much warmer CCSRNIES All scenario, the climate-change-adapted golf season was projected to be approximately 16 days longer in the 2020s and 37 days longer in the 2050s. In the 2080s, golf seasons in the region were projected to be extended by 68 days, creating an opportunity for a potential 260-day golf season. By comparison, in southern Ontario, golf season lengths were projected to be extended by 17 days in the 2020s under the least-change scenario (NCARPCM B21), with only modest additional increases in the 2050s and 2080s (up to 27 additional days longer). Under the much warmer CCSRNIES All scenario, however, the climate-change-adapted golf season was projected to be 51 days longer in the 2020s and 86 days longer in the 2050s. In the 2080s, golf was projected to have a potential season length of 323 days, approximately 16 weeks longer than at present. Although the potential for a year-round golf season exists in the study area under the warmest scenario for the 2080s, it would remain intermittent at best as daily mean temperatures during the winter months of December to February are projected to be remain relatively cool (~2.5 C) with extended periods of frost and snow cover in some years.
With respect to average rounds played per season, 18-hole golf courses in the Great Lakes region were projected to experience similar rates of increase in golf participation under a climate-adapted operating season. In central Ontario, increases in annual rounds season were projected to range from 21 percent (NCARPCM B21) to 35 percent (CCSRNIES All) in the 2020s, 25 percent (NCARPCM B21) to 59 percent (CCSRNIES A11) in the 2050s, and 30 percent (NCARPCM B21) to 74 percent (CCSRNIES All) in the 2080s. In southern Ontario, the increase in the average number of rounds played per season was projected to range from 23 percent (NCARPCM B21) to 37 percent (CCSRNIES All) in the 2020s and 27 percent (NCARPCM B21) to 61 percent (CCSRNIES All) in the 2050s. In the 2080s, an 18-bole golf course in the southern Ontario could experience between 32 percent (~8,500 rounds) and 73 percent (~19,700 rounds) more rounds annually.
Golf participation in the East Coast region was also projected to benefit substantially from a warmer climate, even under the least-change climate change scenario. Golf season lengths were projected to be extended between 28 days (NCARPCM B21) and 45 days (CCSRNIES A11) in the 2020s and between 28 days (NCARPCM B21) and 56 days (CCSRNIES A11) in the 2050s. Under a climate-adapted operating season, annual rounds were projected to increase between 40 percent and 49 percent (NCARPCM B21) and between 48 percent and 74 percent (CCSRNIES All) in the 2020s and 2050s, respectively, over current conditions. In the 2080s, an average 18-hole golf course in the East Coast region could have a golf season length of 207 days (39 additional days) under the least-change climate change scenario (NCARPCM B21) and a season length of 253 days (85 additional days) under the warmest climate change scenario (CCSRNIES A11). Such extensions in the golf season translate into a 53 percent increase in rounds played over current levels of participation under the least-change scenario (NCARPCM B21) and a 94 percent increase under the warmest climate change scenario (CCSRNIES A11).
Discussion and Concluding Comments
This study examined the influence of climate on golf participation by developing regression models of the daily weather-rounds played relationship to explore the potential impact of climate change on the golf industry in three geographic regions of Canada with different prevailing climates. The potential impact of climate change was assessed exclusively in this study. While it is recognized that other major factors could affect golf participation in the study regions decades into the future (e.g., population growth, demographic change, the supply of golf courses, the costs of participation [equipment, green fees, time]), this study focused on climate change only in order to isolate and assess the importance of a changed climate to the golf sector. An important challenge of future research will be the integration of these other factors in to climate change impact assessments of the golf sector.
The climate change impact assessment indicated that golf participation would increase in Canada even under the most conservative climate change scenario, and that the golf industry in some regions in Canada would benefit (in terms of season length and rounds played) from projected changes in the climate more than other regions. Golf courses on the West Coast were projected to benefit the least from a changed climate, as golf is already a year-round activity in many parts of the region. Golf operating seasons in the Great Lakes region were projected to be at least 260 days in the 2080s (baseline is 192 to 214 days) under the warmest climate change scenario (CCSRNIES All), with the potential for golf to be a year-round, but intermittent, activity in southern Ontario. Annual rounds played were projected to increase, on average, between 30 percent and 74 percent in central Ontario and between 32 percent and 73 percent in southern Ontario in the 2080s. The East Coast region was projected to benefit the most from a changed climate, experiencing more than a doubling in annual rounds played and a climate-adapted golf season length in the 2080s comparable to that of central Ontario (260 days).
If these findings are suggestive of the longer-term effects of climate change on golf participation in Canada, to say nothing of future increases in participation from population and economic growth, the implications for the country's golf industry are substantive. Individual golf courses and golf tourism destinations (e.g., Muskoka Tourism Region [Ontario] and Prince Edward Island) in Canada would benefit economically, as longer golf seasons and increases in rounds played could contribute to higher golf course revenues (e.g., from green fees, cart rentals, pro shop, food/beverage sales) and related tourism spending.
Although Canadian golf courses nationwide could benefit financially from projected climate change, there could also be other important impacts in the decades to come. Longer golf season lengths could alter competitive relationships between major golf destinations within the country and internationally. For example, as golf season lengths are extended in the Great Lakes and East Coast regions, the attractiveness of winter golf-tourism destinations in the southern U.S. for Canadians may diminish. As for other aspects of day-to-day golf operations, the warmer climate projected by the CCSRNIES All scenario in particular could lead to greater demand for turf grass irrigation in most regions of Canada. A 1998 survey of Canadian golf courses found water availability or cost was currently or was soon expected to be a problem for 22 percent of courses (Royal Canadian Golf Association 1999). With increased competition for water in the future, climate change is anticipated to exacerbate the challenge of water supply for the industry. Golf course managers also need to consider the potential impact of climate change on grass maintenance issues, such as turf grass selection, turf diseases and insect pests. Pests that have only one life cycle at present in many parts of Canada could adapt to new climate regimes and have two life cycles (Vittum 2003). Perhaps more importantly, there is the potential for turf grass diseases and pests currently limited to more southerly latitudes to expand northward and require management interventions in the future. Future analysis of these operational issues is essential to providing insight into the adaptive capacity of golf courses across Canada to take advantage of the opportunities projected climate change could bring.
Climate change has been identified as an important issue by leading golf organizations around the world. Drawing on the input of over 250 stakeholders, including course mangers, union leaders and professional organizations, the Golf Course Advisory Panel at the Royal and Ancient Golf Course of St. Andrews (Scotland) identified climate change as one of six strategic issues facing the golf industry over the next 20 years (Royal and Ancient Golf Club of St. Andrews 2000). More recently, the governing body responsible for staging the Open Championship (R & A Rules Limited) stated, 'every national governing body and golf facility should be finding out more about climate change predictions for their region, and starting to anticipate the effects for golf' (R & A, 2005, 1). Now that it has been established that climate change is relevant to the future of the golf industry across Canada, larger regional studies are needed. In addition, future research will need to assess how climate change may interact with other social (population growth, ethnic diversity, aging society) and economic (cost of participation, golf course supply, and operation) factors to determine the net impact on the future of Canada's golf industry.
The authors would like to thank the golf courses that provided data for this study. At the request of golf course officials who agreed to provide data for this study, their identity and the names of the golf courses they manage were kept confidential. Partial funding for this study was provided by the Government of Canada's Climate Change Action Fund, Impacts and Adaptation Programme (Project A715).
ANDREW. R. 1998 'On a roll' Ca Magazine, 131(3). Accessed 15 November 2005, available at www.camagazine.com
CANADIAN PRESS 2004 'Chilly weather is keeping people off the links at several Winnipeg-area golf courses so far this year'. The Canadian Press, 25 June, available at http://www.broadcastnews.com, accessed 21 November 2005.
CROWE, R., MCKAY, and BAKER, W. 1977 The Tourist and Outdoor Recreation Participation of Ontario. Volume 2: The Summer Season, REC-1-73 (Downsview, Ontario: Atmospheric Environment Service)
DOEY B., and WONG R. 2002 'Weathering the storms: does mother nature control our industry?' Golf Business Canada, Winter. Accessed 12 March 2004, available at http://www.ngcoa.ca/shtml/golf_business_magazine/online/2002winter
HERBERT. J. 2004 'The summer that wasn't; par for the course' London Free Press, 21 August, E7
--. 2005 'Fore the record' London Free Press, 27 August, E5
JONES. B, and SCOTT. D. 2006 'Climate change, seasonality and visitation to Canada's national parks' Journal of Park and Recreation Administration 24(2), 42-62
JONES, B., SCOTT, D., and AB KHALED, H. 2006 'Implications of climate change for outdoor event planning: a case study of three special events in Canada's national capital region' Event Management 10(1), 63-76
LAMOTHE & PERIARD CONSULTANTS. 1988 Implications of Climate Change on Municipal Water Use and the Golfing Industry in Quebec (Downsview, ON: Atmospheric Environment Service, Environment Canada)
MILLS. B., SCOTT, D. and BASS b. 2002 'Exploring uncertainties in climate change health impacts', Proceedings of the 15th Conference on Biometeorology and Aerobiology (Kansas City, MO: American Meteorological Society)
NATIONAL GOLF FOUNDATION. 2000 Golf Practice in the US: A Summary of Supply and Demand (Boca Rotan, FL: National Golf Foundation)
--. 2004 Golf 20/20 Vision for the Future: Industry Report for 2003 (Boca Rotan, FL: National Golf Foundation)
QAIN, B., GAMEDA, S. HAYHOE. H., DE JONG, R., and BOOTSAM. A. 2004 'Comparison of LARS-WG and AAFC-WG stochastic weather generators for diverse Canadian climates' Climate Research 26, 175-191
R & A, THE OPEN CHAMPIONSHIP. 2005 Climate Change. R & A Course Management. Best Practices Guidelines (Scotland, UK: The R & A, Open Championship) Accessed 15 February, 2006, available at http://www.bestcourseforgolf.org/ content/environment/key_environment/climate_change
REGION OF HALTON. 2000 Halton Golf Course Study: An Analysis of Future Demand of Golf Courses in the Greater Toronto Area and Hamilton-Wentworth Region (Region of Halton, Ontario: Halton Region Planning Partnership Project)
ROYAL AND ANCIENT GOLF CLUB OF ST ANDREWS. 2000 On Course For Change: Tackling the Challenges Facing Golf in the First Decades of the New Millennium (Scotland, UK: Royal and Ancient Golf Club of St Andrews)
ROYAL CANADIAN GOLF ASSOCIATION. 1999b 1998 Canadian Golf Course Operations Survey (Toronto: Royal Canadian Golf Association)
--. 2002 2002 Golf Participation in Canada Survey (Toronto: Royal Canadian Golf Association)
--. 2004 Statistics (Toronto, Ontario: Royal Canadian Golf Association), Accessed 15 November 2005, available at www.rcga.org
SCOTT, D., and JONES. B. 2006 'The impact of climate change on golf participation in the Greater Toronto Area (GTA): a case study' Journal of Leisure Research 38(3), 363-380
SCOTT, D., JONES, B., and ABI KHALED. H. 2005 Climate Change--A Long-term Strategic Issue for the NCC. Implications for Recreation-Tourism Business Lines. Report prepared for the National Capital Commission (Waterloo, ON: University of Waterloo)
SCOTT, D., MCBOYLE. G., and MILLS, B. 2003 'Climate change and the skiing industry in southern Ontario (Canada): exploring the importance of snowmaking as a technical adaptation' Climate Research 23, 171-181
SCOTT, D., MCBOYLE. G., and MINOGUE. A. 2007 (in press) 'Implications of climate change for the Quebec ski industry' Global Environmental Change
SEMENOV, M., BROOKS, R., BARROWS, E., and RICHARDSON, C. 1998 'Comparison of the WGEN and LARS-WG stochastic weather generators in diverse climates' Climate Research 10, 95-107
STATISTICS CANADA. 2005 Annual Survey of Arts, Entertainment and Recreation (Ottawa: Statistics Canada)
VITTUM, P. 2003 'Insects like it hot' Golf Course Management December, 113-115
WINNIPEG FREE PRESS. 2002 'Golf links are open one last time' Winnipeg Free Press, 7 November, A10
--. 2004 'Wet season drowns local golf retailer' Winnipeg Free Press 23 December, B4
WILBY, R., CHARLES, S., ZORITA, E., WHETTON, P., and MEARNS, L, 2004 Guidelines For Use of Climate Scenarios Developed From Statistical Downscaling Methods (Switzerland: United Nations Intergovernmental Panel on Climate Change, Task Group on Data and Scenarios Support for Impacts and Climate Analysis)
Department of Geography, University of Waterloo, Waterloo, Ontario N2L 3G1, Canada (e-mail: email@example.com)
Department of Geography, University of Waterloo, Waterloo, Ontario N2L 3G1, Canada
(1) Climate data met quality standards for inclusion in the Meteorological Service of Canada's national archive and there were no prolonged periods of missing data.
Table 1 Summary of golf courses and climate stations in this study Study regions West Coast Great Lakes Golf course features Course location British Columbia Ontario (ON) (BC) (southern) Course type Private Private Number of holes 18 holes 18 holes High-season green fee (CONS) $125 $89 Number of nearby 18-hole golf 25 176 courses (-50 km radius) Data used in study Golf participation 2004/05 * 2002 and 2003 (years of record) Climate station data Victoria, BC Toronto, ON Dominant climate type ([dagger]) Cfb Dfa Study regions Great Lakes East Coast Golf course features Course location Ontario (ON) Nova Scotia (central) (NS) Course type Semi-private Public Number of holes 18 holes 18 holes High-season green fee (CONS) $89 $83 Number of nearby 18-hole golf 45 9 courses (-50 km radius) Data used in study Golf participation 2002 and 2003 1996 to 1999 (years of record) Climate station data Orillia, ON Ingonish, NS Dominant climate type ([dagger]) Dfa Dfc * Golf season officially begins in February and ends the following January. ([dagger]) Koeppen classification: Cfb (marine west coast); Dfa (humid continental); Dfc (sub-Arctic). Table 2 Impact of weather on average daily golf participation Study regions Great West Lakes l Coast (southern) Daily precipitation No mm Average rounds 183 150 <2.5 mm % [DELTA] -19% -22% 2.5 to 5 mm % [DELTA] -26% -48% 5 to 10 mm % [DELTA] -45% -36% Timing of rain More rounds played in the in the when rain occurs afternoon afternoon Wind Rounds played begin to decline at wind speeds >17 kph >20 kph Maximum temperature Rounds played level off at 30[deggres]C 28[deggres]C Minimum temperature Rounds played level off at -6[deggres]C -5[deggres]C Study regions Great Lakes 2 East (central) Coast Daily precipitation No mm Average rounds 168 161 <2.5 mm % [DELTA] -17% -8% 2.5 to 5 mm % [DELTA] -17% -32% 5 to 10 mm % [DELTA] -18% -39% Timing of rain More rounds played in the in the when rain occurs afternoon afternoon Wind Rounds played begin to decline at wind speeds >20 kph >18 kph Maximum temperature Rounds played level off at 29[deggres]C 30[deggres]C Minimum temperature Rounds played level off at -6[deggres]C -5[deggres]C Table 3 Regression models used in this study Predictors of rounds played * Study region (t-statistic) [r.sup.2] West Coast T-max (1 5.60) 0.58 Precip (-8.66) Day of week (8.03) ([dagger]) Great Lakes 1 T-max (6.37) 0.60 (southern) T-min (2.00) Precip (7.08) Day of week (9.88) ([double dagger]) Great Lakes 2 T-max (7.98) 0.55 (central) T-min (2.45) Precip (-3.07) Day of week (10.97) ([double dagger]) East Coast T-max (9.76) 0.48 T-min (5.61) Precip (-6.87) Day of week (9.58) (#) Study region Equation used (daily rounds =) West Coast = 46.27 + 5.72 (T-max) - 3.64 (Precip) + 25.40 (Day of the week) Great Lakes 1 = -16.38 + 4.64 (T-max) + 1.72 (T-min) (southern) - 3.25 (Precip) + 36.52 (Day of the week) Great Lakes 2 = -1 7.84 + 5.13 (T-max) + 1.83 (T-min) (central) - 1. 12 (Precip) + 43.51 (Day of the week) East Coast = -31.84 + 4.70 (T-max) +3.25 (T-min) - 1.54 (Precip) + 39.49 (Day of the week) * Day of the week-classified based on day of the week when the highest and lowest average number of rounds were played. ([dagger]) Weekend (highest-Friday to Sunday), Tuesday (lowest), weekday (all other days). ([double dagger]) Weekend (highest-Saturday and Sunday), Tuesday (lowest), weekday (all other days). (#) Weekend (highest-Friday to Sunday), weekday (all other days). Table 4 Projected annual climatic changes in Canada by study region 2020s 2050s Study region T P T P West Coast NCARPCM B21 +1[degrees]C -1% +2[degrees]C -5% CCSRNIES All +1[degrees]C -3% +4[degrees]C -5% Great Lakes NCARPCM B21 +1[degrees]C +1% +2[degrees]C +3% CCSRNIES All +3[degrees]C +6% +6[degrees]C +4% East Coast NCARPCM B21 +1[degrees]C -1% +3[degrees]C +1% CCSRNIES All +2[degrees]C -1% +5[degrees]C -1% 2080s Study region T P West Coast NCARPCM B21 +2[degrees]C 4% CCSRNIES All +6[degrees]C -7% Great Lakes NCARPCM B21 +3[degrees]C -- CCSRNIES A11 +8[degrees]C 1% East Coast NCARPCM B21 +3[degrees]C -- CCSRNIES A11 +7[degrees]C -2% T-Mean annual temperature change with respect to the 1961-1990 baseline. P-Total annual precipitation change with respect to the 1961-1990 baseline. Table 5 Projected rounds played for 'climate-adapted' operating seasons (maximum 365 days) Modelled (1961-1990) 2020s Season [DELTA] Annual length % [DELTA] in golf rounds (days) rounds days * West Coast Current average season 56,120 365 NCARPCM B21 2% -- CCSRNIES All 2% -- Great Lakes 1 (southern) Current average season 27,050 214 NCARPCM B21 23% +17.0 CCSRNIES All 37% +51.0 Great Lakes 2 (central) Current average season 25,309 192 NCARPCM B21 21% +10.0 CCSRNIES A11 35% +16.0 East Coast Current average season 23,269 168 NCARPCM B21 40% +28.0 CCSRNIES All 48% +45.0 2050s 2080s [DELTA] [DELTA] % [DELTA] in golf % [DELTA] in golf rounds days * rounds days * West Coast Current average season +2% -- 4% -- NCARPCM B21 +11% -- 18% -- CCSRNIES All Great Lakes 1 (southern) Current average season +27% +24.0 32% +27.0 NCARPCM B21 +61% +86.0 73% +109.0 CCSRNIES All Great Lakes 2 (central) Current average season +25% +10.0 30% +18.0 NCARPCM B21 +59% +37.0 74% +68.0 CCSRNIES A11 East Coast Current average season +49% +28.0 53% +39.0 NCARPCM B21 +74% +56.0 94% +85.0 CCSRNIES All * Based on the average number of daily rounds played during the first 2 weeks and last 2 weeks of current operating seasons: West Coast (not applicable-year round season); Great Lakes 1 (opening-75 rds/day; closing-50 rds/day); Great Lakes 2 (opening-50 rds/day; closing-75 rds/day); East Coast (opening-75 rds/day; closing-75 rds/day).
|Printer friendly Cite/link Email Feedback|
|Author:||Scott, Daniel; Jones, Brenda|
|Publication:||The Canadian Geographer|
|Date:||Jun 22, 2007|
|Previous Article:||Comment reorganiser le reseau postal suisse? Une application de l'aide a la decision territoriale.|
|Next Article:||The making of an 'arcane' infrastructure immigrant practitioners and the origins of professional engineering regulation in Ontario.|