Spectator Travel and Carbon Savings: Evaluating the Role of Football Stadium Relocation in Sustainable Urban Planning

Abstract

Environmental consciousness has become increasingly important in the professional sports industry as it often hosts large-scale events that have significant environmental impacts. While the economic benefits of locating stadiums in city centers have been discussed, especially in terms of neighborhood revitalization, there has been limited empirical research on whether stadium relocation affects the transportation choices of spectators and reduces carbon dioxide emissions. Through a case study of a Japanese professional football club that relocated its home stadium from the suburb to the city center, this study quantitatively elucidated the change in spectators’ transportation choices and resulting reductions in carbon emissions achieved by the stadium relocation. Analysis indicated variations in behavioral changes among groups based on their loyalty levels to the club. It also highlighted the varying influence of the different residential areas within the metropolitan area on the modal choice. This study demonstrates the potential contribution of stadium relocation to sustainable urban planning by providing empirical evidence of these behavioral changes and policy implications for restructuring the urban public transportation network.

1. Introduction

As climate change emerges as a significant global challenge, all industries are expected to take responsibility for ensuring environmental sustainability [1,2]. In alignment with the United Nations Sustainable Development Goals (SDGs), the professional sports industry widely recognizes the need to adopt sustainable practices and environmentally responsible transitions as they operate large-scale events with significant environmental impacts. As a result, many clubs, leagues, and governing bodies have published strategies incorporating climate actions [3,4]. A notable example is the “UEFA Football Sustainability Strategy 2030”, developed by the Union of European Football Associations (UEFA) in collaboration with the European Commission [5]. Similarly, the English Premier League has published the “Environmental Sustainability Strategy”, intending to reaffirm its commitment to addressing environmental issues [6]. Under these strategies, initiatives for improving the environmental performance of stadiums and promoting environmentally friendly behavior were conducted [7,8]. Such initiatives enhance the legitimacy of professional sports as well as the subjective well-being of the stakeholders involved [9,10,11,12]. Given the current climate and social circumstances, it is becoming increasingly impossible to sustain the professional sports industry without environmental awareness.

When assessing the environmental impact of professional sports, attention must be paid to greenhouse gas emissions, particularly carbon dioxide. The emissions are classified as Scopes 1, 2 and 3 based on the Greenhouse Gas (GHG) Protocol developed by the World Resources Institute (WRI) and the World Business Council for Sustainable Development. The protocol has been widely adopted as an international standard for companies and organizations to measure and report greenhouse gas emissions [13]. Scope 1 refers to GHG emissions directly from facilities and equipment owned or controlled by a company. Scope 2 refers to indirect emissions associated with the energy purchased by a company. And Scope 3 refers to other indirect emissions that occur throughout a company’s value chain. Sports clubs contribute to carbon dioxide emissions through activities, including energy usage in venues such as stadiums (Scope 1), district heating (Scope 2), and merchandise manufacturing and team travel (Scope 3) [14,15]. Notably, spectator travel, classified under Scope 3, is the predominant source of emissions, as indicated in various studies [15,16,17]. Therefore, environmental initiatives must prioritize reducing emissions from spectator travel and adopt responsible measures to address this issue [18]. Against this background, various methodologies have been proposed to estimate carbon emissions accurately [19,20].

Wilby et al. [21] reviewed the literature from the past three decades on sports and carbon emissions, identifying four key themes: carbon emissions and their measurement, emission mitigation and decarbonization, carbon sinks and offsets, and behavioral change. Among these, behavioral change, which relates to carbon emissions associated with spectator travel, has attracted research interest. Previous studies have shown that carbon emissions are substantially influenced by spectators’ choice of transport mode [22,23]. For example, a study of German football supporters revealed that more than 300 kg of carbon dioxide emissions are produced each season, with car travel accounting for 70% of these emissions [24]. Similarly, a study examining an Austrian professional football team found that carbon dioxide emissions per game reached 6.0 kg per spectator, with car users generating significantly more emissions than public transport users [25].

Although modal choices are subject to individual free will, they are constrained by many factors, including the location of destinations and transport infrastructure availability [26,27,28]. The impact of relocating residential areas, workplaces, and factories on carbon emissions has been discussed [29,30,31]. Large-scale urban facilities such as stadiums also induce a significant amount of travel, making them comparable in importance. Meanwhile, sports venues serve entertainment purposes, unlike residences and workplaces, which may result in different environmental impacts and behavioral patterns. However, the impact of stadium location on spectator modal choice has not been explored in some respects. Firstly, research on stadium location and carbon dioxide emissions has largely overlooked professional sporting events that typically attract a broader demographic of spectators. The few studies that have identified differences in carbon emissions for different stadium locations include a study on modal choices for college football games on- and off-campus [32]. Still, this study also did not focus on different stadium locations in terms of urban settings. Secondly, previous studies have made limited assumptions when asking spectators about their modes of transportation. Studies in North America tend to consider only cars and airplanes [32,33,34]. More transportation options were usually included in European studies, but respondents were always asked to select one mode, and no study has conducted a survey assuming that spectators would combine multiple modes of transportation to attend the games [22,24,25,35].

The impact of professional sports and the location of their home stadiums must also be discussed from social and economic perspectives. Regarding the social impact of stadium location, there have been studies on gentrification and displacement which point out that stadium construction does not just passively push out long-time residents by increasing rents; it is sometimes used intentionally as a catalyst for gentrification [36,37,38]. Professional sports franchises within urban areas have also frequently been analyzed through an economic lens, particularly in North America, where substantial public funding has been allocated for constructing home stadiums. Empirical studies have indicated that professional sports clubs fail to produce macroeconomic benefits for their host cities [39], with an increasing consensus characterizing such impacts as “searching for a needle in a haystack” [40]. However, sports facilities can serve as centers for emergency or community resilience, providing a rationale for public funding [41]. Conversely, despite the lack of substantial macroeconomic effects on metropolitan areas, empirical evidence indicates beneficial impacts at the micro level near stadiums. For example, real estate values and commercial rental prices [42,43,44], food service employment [44], and restaurant sales [45] have been found to show statistically significant increases within a one-mile radius of stadiums. This evidence provides an adequate rationale for policymakers to prioritize specific areas within metropolitan areas [40]. With growing recognition that stadium locations yield district-scale benefits, the focus on stadiums within the context of urban regeneration, a global issue concerning inner-city decay, has increased. In North America, constructing professional sports facilities, including stadiums, has become a promising urban redevelopment strategy [46]. A survey of practitioners found that many experts are particularly optimistic about the potential of professional sports facilities to drive place-based changes in city centers rather than suburban areas [47]. These pieces of empirical evidence have influenced the research and practice of sport cities [48], a conceptual framework that incorporates into urban planning the rationalization of the cost of sports infrastructure development, the increased economic benefits from the legacy of sports events, and the social benefits of the presence of sports in cities.

Based on the argument above, combining efforts to reduce environmental impacts with strategies for revitalizing city centers suggests that locating stadiums in city centers may be an effective strategy for achieving sustainable urban planning from both environmental and economic standpoints. Specifically, the question should be whether stadium location and a regional transport network strategy that encourages the use of public transportation can effectively change spectator travel behavior and reduce carbon dioxide emissions while maintaining the urban regeneration benefits produced by sporting venues. Therefore, this study aimed to quantitatively investigate whether relocating stadiums from suburbs to city centers influences the transport modal choices of spectators and reduces carbon dioxide emissions to offer implications for achieving sustainable urban planning. Whether gentrification and displacement of existing residents will happen in the surrounding area after the stadium relocation is an important issue, but since it requires continuous long-term monitoring and this paper focuses on environmental sustainability, this topic will not be directly addressed.

This paper is structured as follows: Section 2 presents a case study of a professional football club that relocated its home stadium from the suburbs to the city center. Section 3 describes the impact of stadium relocation on spectators’ transport mode choices and the subsequent changes in carbon dioxide emissions. Section 4 examines the characteristics of carbon emissions in Japan, where the case study was conducted, the breakdown of carbon emissions change by the places of residence of spectators, and their relationship to the structure of the public transportation network. Finally, Section 5 addresses the advantages of the emission estimation method introduced in this study and future research directions, which include implications for sustainable urban planning.

2. Materials and Methods

To examine whether stadium relocation can reduce carbon dioxide emissions by influencing spectators’ transport choices, we focused on Hiroshima, Japan, where a professional football club relocated its home stadium from the suburbs to the city center. The following sections introduce the context of Hiroshima’s stadium relocation, survey methods, and data aggregation and analysis.

The professional football league, J. League, established in the early 1990s, resulted in the emergence of professional football clubs across Japan. Many clubs initially used athletic stadiums, which were constructed as the main venues for the National Sports Festival. The Japanese government conducts this annual event, which rotates among prefectures. These stadiums were often located in suburban areas because of a focus on automotive accessibility, which stemmed from land acquisition difficulties and the widespread use of family-owned cars. However, as the stadiums are getting older, there has been growing interest in relocating them to city centers, which are primarily influenced by the enactment of laws encouraging urban regeneration and the introduction of government subsidies supporting mixed-use urban development. Hiroshima, a city in western Japan globally renowned for its historical significance, illustrates this trend.

Sanfrecce Hiroshima F.C., a professional football club based in Hiroshima, was established in 1992 and is one of the ten founding clubs that participated in the inaugural season of the J. League. Until 2023, the club’s home stadium was the Hiroshima Big Arch, an athletic stadium constructed in September 1992 and used as the main venue for the 1994 Asian Games and 1996 National Sports Festival. The stadium has been criticized for poor accessibility and lack of an immersive atmosphere, as it is located on a hilly terrain approximately 8 km from the city center, and its versatile design for athletics creates an unignorable distance between the pitch and the stands.

Discussions on constructing a football-specific stadium in the city center occurred in the early 2000s. After extensive deliberations among Hiroshima Prefecture, Hiroshima City, and local stakeholders, a basic policy for building a new stadium within an urban park in the city center was finalized in February 2019 [49]. In October 2020, Hiroshima City called for proposals for the “Football Stadium Development Project”. Presentations from four groups of joint ventures were conducted on 30 March 2021, and a business contract for design and construction was signed between Hiroshima City and a private consortium on 29 June 2021.

The new stadium, Hiroshima Soccer Stadium, occupies an area of approximately 49,000 square meters, with a seven-story structure and a seating capacity of 28,520. Designed for multifunctional use, it included emergency storage facilities on the first floor to support displaced individuals for up to three days during a disaster [50]. The stadium opened on 1 February 2024, with its inaugural match, a J. League pre-season game, on 10 February. The first official match occurred on 23 February, the opening round of the J1 League (the first division of the J. League) season. Figure 1 displays the locations of the old and new stadiums.

In collaboration with Sanfrecce Hiroshima F.C., we conducted several surveys to examine the effects of stadium relocation from the suburbs to the city center from different viewpoints. The survey dates included a match day at the old stadium and five match days at the new venue. Compared to previous studies, in which data collection was conducted over multiple matches [24,25], the number of surveys conducted at the old stadium is relatively limited, and the results may deviate from the average conditions there. This could represent a limitation of the study. However, it is worth noting that carbon emission estimates for initiatives such as zero-carbon games have also been reported based on a single match. Therefore, it is reasonable to consider that the approach in this study retains a certain validity. A total of 16,478 responses were collected, which accounted for approximately 10.5% of the attendance during these six matches. Since many transportation options are available in Hiroshima, the questions regarding the transportation modes were designed to allow multiple selections. In addition, the feedback was obtained from club officials who were familiar with local transportation conditions. The provided options included cars (as a driver), cars (as a passenger), drop-off, taxis, motorbikes, bicycles, local buses, highway buses, chartered buses, shuttle buses, trams, Astram Line [an automated guideway transit system operating in Hiroshima], railways (local), Shinkansen bullet train system, airplanes, ferries, walking and others. Shuttle buses operated exclusively for matches at the old stadium and ran a single route from a railway station in Hiroshima to the stadium.

Additional data included respondents’ demographics, such as their place of residence (municipality in Hiroshima Prefecture or the region or country, for those outside the prefecture), age, and sex; factors potentially related to loyalty to the sport or club, such as the annual number of matches attended in person and fan club membership status, were also considered. In two matches at the new stadium, open-ended descriptions of their routes using each transport mode were asked, with the aim of analyzing the distances traveled by different transport modes in detail.

Table 1. Overview of the survey, questionnaire items, and response options.

Dates	25 November 2023; 10 February 2024; 23 February 2024; 30 March 2024; 26 June 2024; and 5 July 2024
Method	Online (survey forms sent to respondents via email)
Respondents	All spectators of the “Meiji Yasuda J1 League” official matches held on the survey dates (except 10 February 2024) and the pre-season match held on 10 February 2024
Questionnaire items and response options	Age Sex Place of residence (For residents of Hiroshima City, ward; for residents of Hiroshima Prefecture, municipality; for residents of the Chugoku region, prefecture; for other regions, region; for foreign respondents, country.) Frequency of attendance (For the survey conducted in 2023, history of attendance at matches in the 2023 season; for surveys in 2024, history of attendance at the 2023 and 2024 season matches. Respondents selected one of the following: “0 times/first time”, “once”, “twice”, “3 times”, “4 times”, “5 times”, and “6 times or more”.) People accompanying the respondent (Respondents selected one from the following: Alone, family (with their spouse), family (parent-child), other family (with siblings, grandchildren, grandparents), three-generation family, friends/acquaintances/partner, school/club teammates, colleagues/business associates, fellow fans, and other.) Fan club membership status (Membership categories set by the club for the 2023 and 2024 seasons were given as the options.) Transportation modes used (Respondents selected all that applied from the following options: car (as a driver), car (as a passenger), drop-off, taxi, motorbike, bicycle, local bus, highway bus, chartered bus, shuttle bus, tram, Astram Line [an automated guideway transit system operating in Hiroshima], railway (local), Shinkansen bullet train, airplane, ferry, on foot, and other.)

presents an overview of the survey, questionnaire items, and response options.

The surveys were administered online. For each match, an invitation containing a link to the questionnaire was emailed to all ticket purchasers on the evening of the match day and two days thereafter. As Sanfrecce Hiroshima restricts ticket sales to the club’s online platform, referral filter bias owing to ticket-purchasing variations was avoided.

Survey responses were systematically compiled. Initially, for participants who specified their modes of transportation and the routes traveled by each mode, the respective travel distances were determined based on the operational distance data provided by the transport operators between stations and stops. Subsequently, assuming that each respondent’s place of residence aligns with the population centroid of the reported municipality, the proportion of the total travel distance using each mode of transport was computed relative to the distance required to travel entirely by car from the respondent’s residence to either the old or new stadium. Data published by Japan’s Ministry of Internal Affairs and Communications [51] were used to identify population centroids of municipalities. For municipalities outside the Hiroshima Prefecture, the population centroid of the most populous or central city was used as the representative location.

Next, for groups of respondents who provided identical answers to the multiple-choice question regarding transport modes, we calculated the average proportional share of travel distance for each mode. For instance, respondents choosing “train (local) and tram” as their transportation reported an average trip of 66.5% of the total distance from their residence to the stadium using the local train. Based on these data, the estimated travel distances for each mode of transport were calculated for the respondents who did not provide free-text responses regarding travel sections.

Because a “shuttle bus” operated only for matches held at the old stadium, for respondents who selected the shuttle bus as a mode of transport, the proportional share of travel distances for each mode was applied to the route from their place of residence to the departure station (JR Yokogawa Station). Emissions from shuttle bus use were then added. Finally, per-person and total emissions were calculated based on these estimations. Emission intensity values for each transport mode, in terms of CO₂ emissions, were obtained from the data published by Japan’s Ministry of the Environment [52].

During the result analysis, spectators were categorized by their place of residence into three groups: spectators residing in Hiroshima City, spectators residing in Hiroshima Prefecture excluding Hiroshima City, and spectators not residing in Hiroshima Prefecture. The emission per capita was calculated for each group. Additionally, we analyzed whether age, sex, number of matches attended annually, and fan club membership status influenced the transportation modal choices and their changes before and after the stadium relocation.

3. Results

This section describes the results of the survey conducted before and after the stadium relocation, the calculation of travel distances by transportation mode based on the responses, the estimated carbon footprint, and the effects of different variables on the modal choice.

3.1. Spectators’ Modal Choice

Table 2. Number of survey responses on each survey date.

Date	Opponent	Attendance (n)	Responses (n)	Response Rate
25 November 2023	Gamba Osaka	29,097	3480	12.0%
10 February 2024	Gamba Osaka	26,418	4257	16.1%
23 February 2024	Urawa Reds	27,545	3704	13.4%
30 March 2024	Gamba Osaka	25,683	1949	7.6%
26 June 2024	Albirex Niigata	22,774	1671	7.3%
5 July 2024	Vissel Kobe	25,119	1417	5.6%

presents the number of survey responses for each match. The highest response rate was observed for the match held on 10 February 2024, which marked the opening of a new stadium. Of the 26,418 spectators, 4257 responded to the survey (16.1%). Conversely, the lowest response rate was recorded for the match held on 5 July 2024, for which 1417 responses were received from 25,119 spectators, accounting for only 5.6%.

Among the transportation modes, allowing for multiple selections, the most common answer was “car (as a driver)” with 2461 respondents (14.9%), followed by “local bus” with 2009 respondents (12.2%), “railway (local)” with 1375 respondents (8.3%), “Astram Line” with 1347 respondents, and “car (as a passenger)” with 1089 respondents (6.6%). The most frequently selected combination of transportation modes was “railway (local) and Shinkansen bullet train”, chosen by 196 respondents (1.2%).

Table 3. Transport modes used by spectators (An excerpt of most frequent responses).

Transport Modes	Responses in Total (Rate)	Ratio Before Relocation	Ratio After Relocation
Car (as a driver)	2461 (14.9%)	17.6%	14.2%
Local bus	2009 (12.2%)	3.5%	14.5%
On foot	1428 (8.7%)	3.1%	10.2%
Railway (local)	1375 (8.3%)	0.1%	10.6%
Astram Line	1347 (8.2%)	18.7%	5.4%
Car (as a passenger)	1089 (6.6%)	9.7%	5.8%
Tram	879 (5.3%)	0.0%	6.8%
Bicycle/Motorbike a	814 (4.9%)	4.7%	5.0%
Drop off	469 (2.8%)	6.2%	1.9%
Highway bus	342 (2.1%)	1.0%	2.4%
Shinkansen bullet train	232 (1.4%)	0.0%	1.8%
Shuttle bus	217 (1.3%)	6.2%	-
Bicycle a	208 (1.3%)	-	1.6%
Astram Line, Railway (local)	196 (1.2%)	3.6%	0.5%
Railway (local), Shinkansen bullet train	174 (1.1%)	0.0%	1.3%
Car (as a driver), Astram Line	170 (1.0%)	3.4%	0.4%
Taxi	166 (1.0%)	0.8%	1.1%
Tram, Shinkansen bullet train	161 (1.0%)	0.0%	1.2%

^a Before the match on 30 March 2024, the option for “Bicycle/Motorbike” was provided; however, from the match on 26 June 2024, onwards, the options were split into “Bicycle” and “Motorbike”. For the responses to “Bicycle/Motorbike” before 30 March 2024, the number of responses was proportionally allocated according to the ratio of responses for “Bicycle” and “Motorbike” from 26 June 2024, onwards.

summarizes the top responses before and after relocation. Notably, the combined percentage of respondents using cars decreased from 44.9% before relocation to an average of 28.4% across the five matches held after relocation. The combined percentage of respondents using trams and local buses increased from 1.7% and 8.7% to 12.0% and 18.3%, respectively. The percentage of respondents traveling exclusively on their feet increased from 3.1% to 10.2%.

3.2. Distance Covered by Transportation Modes

Five hundred ninety-four respondents provided answers in the free-text section regarding their travel routes, enabling route identification. The distance from the population centroid of their residential area to the stadium was determined by selecting the shortest route, as determined using Google Maps, with the departure time set as a Sunday at noon. For airplanes, trains, and buses, the operational distances were obtained by referring to the Japanese website “Yahoo! Transit” or the official websites of the transportation companies. For example, if a respondent from outside Hiroshima Prefecture stated that they used only the Shinkansen bullet train, it was assumed that they traveled by local train (if residing in one of Japan’s major metropolitan areas) or by car (if residing elsewhere) from their population centroid to the Shinkansen station, making reasonable estimates based on Japanese transportation circumstances wherever possible. Based on this information, the percentage of each mode of transport used for the total distance was calculated for each of the 594 respondents, and the average for respondents who provided identical responses regarding their combination of transport modes was computed. The transportation modes and their respective distance-share percentages are listed in

Table 4. Transport modes used and their respective distance-share percentages.

Transport Modes	Distance-Share Percentages for Each Transport Mode Emitting Carbon Dioxide (Proportion)
Car (as a driver)	Car (as a driver) (97.6%), Airplane (1.9%), Railway (local) (0.08%)
Local bus	Local bus (95.3%)
On foot
Railway (local)	Railway (local) (115.9%)
Astram Line	Astram Line (102.9%)
Car (as a passenger)	Car (as a passenger) (100.5%)
Tram	Tram (76.5%)
Bicycle/Motorbike a	Motorbike (24.9%)
Drop off	Dropped off (100.0%)
Highway bus	Highway bus (85.4%), Airplane (9.9%), Railway (local) (0.51%)
Shinkansen bullet train	Shinkansen bullet train (93.2%), Car (as a driver) (2.1%), Railway (local) (1.1%)
Shuttle bus b	Shuttle bus (between JR Yokogawa Station and the Hiroshima Big Arch stadium)
Bicycle a	Railway (local) (3.2%)
Astram Line, Railway (local)	Railway (local) (63.4%), Astram Line (27.0%)
Railway (local), Shinkansen bullet train	Shinkansen bullet train (70.8%), Railway (local) (0.88%)
Car (as a driver), Astram Line	Car (as a driver) (101.9%), Astram Line (19.0%)
Taxi	Taxi (109.5%)
Tram, Shinkansen bullet train	Shinkansen bullet train (108.6%), Tram (1.0%)

^a For the “Bicycle/Motorbike”, the distance was proportionally allocated between bicycle and motorbike based on the response ratios for “Bicycle” and “Motorbike” after 26 June 2024. Consequently, 24.9% of the total distance traveled was attributed to motorbike travel. ^b For the “Shuttle bus”, as only one route operated from JR Yokogawa Station to the old stadium, the same amount of carbon dioxide emissions was assumed for all users based on the operating distance.

.

3.3. Estimated Carbon Dioxide Emissions Derived from Spectator Travel

Based on these results, carbon dioxide emissions per capita and total emissions per match were calculated. For combinations of transport modes for which no respondents provided information on their routes, the distance share for each transport mode could not be calculated. Therefore, they were excluded from the analysis. Of the 16,478 survey respondents, emissions were estimated for 91.8% (15,135 individuals). The average estimated emissions per capita for a one-way journey was 4039 g, with the highest being 177,299 g. The emissions per capita and total emissions for each survey match are listed in

Table 5. Emissions per capita and total emissions for each match.

Date	Attendance	Emissions Per Capita (gCO2)	Reduction Rate	Total Emission (tCO2)	Reduction Rate
25 November 2023	29,097	10,670	-	310.5	-
10 February 2024	26,418	7528	29.5%	198.9	36.0%
23 February 2024	27,545	8716	18.3%	240.1	22.7%
25 March 2024	25,683	5966	44.1%	153.2	50.6%
26 June 2024	22,774	5096	52.2%	116.0	62.6%
5 July 2024	25,119	9022	15.4%	226.6	27.0%

. The carbon dioxide emissions per capita for the suburban stadium before relocation was 10,670 g, whereas the average emissions across five matches for the city center stadium after relocation was 7480 g, marking a reduction of 29.9%. The total emissions per match were estimated as 310.5 tCO₂ before relocation and at an average of 190.8 tCO₂ (ranging from 116.0 tCO₂ to 240.1 tCO₂) after relocation.

The significant differences in the estimated emissions between matches are likely because of the distance from the franchise city of the opposing team and the number of fans traveling from distant areas. Therefore, the residences of the spectators were classified into three distinct categories: Hiroshima City, Hiroshima Prefecture excluding Hiroshima City, and Areas outside of Hiroshima Prefecture. The calculated values for each residential category are listed in

Table 6. Emissions per capita for each residential category of spectators.

Date	Spectators Residing in Hiroshima City		Spectators Residing in Hiroshima Prefecture		Spectators Residing Outside Hiroshima Prefecture
	Emissions Per Capita (gCO2)	Reduction Rate	Emissions Per Capita (gCO2)	Reduction Rate	Emissions Per Capita (gCO2)	Reduction Rate
25 November 2023	1892	-	11,508	-	42,410	-
10 February 2024	714	62.2%	8224	28.5%	33,900	20.1%
23 February 2024	696	63.2%	8460	26.5%	37,362	11.9%
25 March 2024	702	63.0%	8484	26.3%	26,740	36.9%
26 June 2024	828	56.3%	8230	28.5%	34,888	17.7%
5 July 2024	880	53.5%	8404	27.0%	51,124	−20.6%

. For Hiroshima City resident spectators, the emissions per capita before relocation were 1892 g, decreasing to 744 g, representing a 60.7% reduction after relocation. For Hiroshima Prefecture residents outside Hiroshima City, the emissions per capita dropped from 11,508 g to 8352 g, indicating a reduction of 27.4% after relocation. For spectators residing outside Hiroshima Prefecture, the emissions before and after relocation were 42,410 g and 35,950 g, respectively, marking a reduction of 15.2%. The carbon emissions per capita for spectators visiting from outside Hiroshima Prefecture for the match on 5 July 2024 are an outlier. This may be because Kobe, the hometown of the opposing team, Vissel Kobe, is closer to Hiroshima than those of other opponents on other survey dates, encouraging more fans to travel by car.

The shares of total emissions by residential category before relocation were 10.0% for Hiroshima City residents, 31.0% for Hiroshima Prefecture residents, excluding Hiroshima City, and 59.0% for spectators from outside Hiroshima Prefecture, and after relocation these figures shifted to 6.0%, 29.6%, and 64.4%, respectively (

Table 7. Share of total emissions by residential category of the spectators.

Date	Spectators Residing in Hiroshima City		Spectators Residing in Hiroshima Prefecture		Spectators Residing Outside Hiroshima Prefecture
	Emissions (tCO2)	Share	Emissions (tCO2)	Share	Emissions (tCO2)	Share
25 November 2023	15.5	10.0%	48.1	31.0%	91.6	59.0%
10 February 2024	5.5	5.5%	29.7	29.9%	64.2	64.6%
23 February 2024	5.5	4.6%	31.2	26.0%	83.4	69.5%
25 March 2024	5.5	7.1%	29.7	38.8%	41.4	54.1%
26 June 2024	6.4	11.1%	22.8	39.3%	28.8	49.6%
5 July 2024	6.9	6.1%	26.6	23.4%	79.8	70.5%

). These results show that although Hiroshima City residents accounted for the smallest share of total emissions, their reduction rate was the highest. Conversely, spectators from outside the Hiroshima Prefecture had the largest share of total emissions but achieved the lowest reduction rates. Consequently, the share of total emissions from Hiroshima City residents decreased following the relocation.

3.4. Factors Affecting the Modal Choice and Its Changes Following Relocation

The impact of the variables on the modal choice and its changes before and after relocation was analyzed. The variables included (1) sex, (2) age, (3) frequency of attendance, (4) companions of the respondent at the match, and (5) fan club membership status.

When analyzed by sex, age, and respondent companionship, the proportion of spectators using cars decreased significantly across all categories (

Table 8. Changes in car usage rates before and after relocation by selected attributes.

Category	Attribute	Rate Before the Relocation	Rate After the Relocation	Difference (Percentage Points)
Sex	Female	49.9%	31.7%	−18.2
	Male	42.1%	27.0%	−15.1
Generation	Children (17 years or below)	58.5%	27.6%	−30.9
	Young (18–39 years)	43.8%	25.7%	−18.1
	Middle (40–59 years)	43.4%	29.4%	−14.0
	Old (60–69 years)	48.8%	28.6%	−20.2
	Older (70 years or above)	49.4%	27.3%	−22.1
People who attended the match with the respondent	Alone	25.9%	18.7%	−7.2
	Spouse	51.8%	31.8%	−20.0
	Parents/Children	55.9%	37.6%	−18.3
	Other family members	57.6%	37.4%	−20.2
	Others	37.4%	22.4%	−15.0
Frequency of attendance	Core (six or more for 2023)	49.6%	30.2%	−19.4
	Middle (one to five for 2023)	40.2%	27.4%	−12.8
	Light (no attendance or only for the last game of 2023)	26.0%	24.9%	−1.1
Fan club membership status	Paid members	45.7%	28.3%	−17.4
	Non-paid members a	N/A	28.0%	N/A
	Non-members	41.9%	35.4%	−6.5

^a The status of “non-paid members” was introduced in the 2024 season.

). Simultaneously, the proportion of those traveling solely on foot increased markedly. By contrast, two variables related to club loyalty (frequency of attendance and fan club membership status) revealed distinct trends. Fans were divided into three groups based on the number of matches they attended in the 2023 season (Table 8): core fans (those who attended six or more matches prior to the last game), middle fans (those who attended one to five matches prior to the last game), and light fans (those who either did not attend any matches or only attended the last game). The proportion of spectators using cars decreased by 19.4 percentage points for core fans and 12.8 percentage points for middle fans, whereas the decrease was only 1.1 percentage points for light fans. The core fans tended to travel by car to the old stadium more than other groups; however, no clear difference was observed between the frequency of attendance and the use of cars at the new stadium. Regarding fan club membership status, paid members reduced car use by 17.4 percentage points, compared to a decrease of only 6.5 percentage points for non-members. Consequently, paid members currently use cars at a lower rate than non-members.

4. Discussion

In this study, we investigated whether relocating a football stadium from the suburbs to the city center influenced spectators’ modal choices and reduced carbon dioxide emissions by conducting a comprehensive survey before and after the stadium relocation. During professional football league matches in Hiroshima, Japan, the carbon dioxide emissions of spectators per capita were 10,670 g at the suburban stadium and 7480 g at the city center stadium. This estimate after the relocation is lower than those for other events, such as 30 kg for Major League Baseball (MLB)’s Tampa Bay Rays games, 12 kg for the zero-carbon game held by Tottenham Hotspur in the English Premier League (EPL) in 2021, and approximately equal to 6.0 kg for matches in Austria’s Bundesliga at Rapid Vienna [25,53].

Two factors could explain the lower emissions in Hiroshima compared to those in the events held by the MLB and EPL. First, EPL matches attract substantial international interest, with many spectators traveling from outside England, as noted by Herold et al. [25]. In contrast, only six participants (0.036%) of the survey at Hiroshima resided outside Japan. This result aligns with the view expressed by club officials that “almost no spectators come from abroad”. Second, Japan has a high share of public transportation in urban areas. According to a 2020 survey report, Tokyo’s public transport modal share is 36%, compared to just 1–2% in U.S. cities hosting MLB games [54]. Therefore, the small number of international spectators and the high usage of public transportation might explain the low carbon emissions from transportation by spectators in Japanese cities.

Next, we examined the breakdown of the differences in emissions before and after the stadium relocation. These changes could be attributed to the modal choices of spectators and changes in travel distances to the stadium. For example, when comparing the distances from the population center to the old and new stadiums, the travel distance increased from 8.6 km to 12.9 km in Saeki Ward, Hiroshima City. However, in Naka Ward, Hiroshima City, it decreased from 10.3 km to 2.5 km. Carbon dioxide emissions were calculated based on the emission factors for each mode of transportation and were proportional to travel distances. Therefore, before assessing the reduction effects of changes in transportation modes, the changes in travel distances to stadiums before and after relocation must be considered.

In this study, the reduction rate (R%) in per capita carbon dioxide emissions by place of residence is expressed as the sum of the reduction rates due to distance shortening (Distance-R%) and behavioral changes (Choice-R%). Figure 2 shows a graph plotting Choice-R% against Distance-R% for 34 residential areas with 50 or more respondents in this survey. Plotting a point on the line y = −x passing through the origin indicates that R% is zero, indicating no reduction. Points above the line y = 0 indicate a positive Choice-R%, signifying a reduction effect owing to behavioral change. Behavioral change effects were observed in a positive direction in 29 of the 34 areas (

Table 9. R%, Distance-R%, and Choice-R% by point of residence.

Point of Residence	R%	Distance-R%	Choice-R%
Saeki Ward, Hiroshima City	18.6%	−50.0%	68.6%
Hatsukaichi City, Hiroshima Prefecture	30.5%	−20.5%	51.0%
Kyushu/Okinawa Region	41.7%	−1.4%	43.2%
Saka Town, Hiroshima Prefecture	78.6%	46.5%	32.1%
Asa-Kita Ward, Hiroshima City	56.5%	26.5%	30.1%
Onomichi City, Hiroshima Prefecture	30.0%	0.5%	29.5%
Takehara City, Hiroshima Prefecture	43.4%	14.1%	29.3%
Ohtake City, Hiroshima Prefecture	22.3%	−4.1%	26.4%
Tokai Region	26.7%	0.8%	25.8%
Nishi Ward, Hiroshima City	73.6%	48.0%	25.5%
Fuchu Town, Hiroshima Prefecture	84.4%	60.7%	23.8%
Kinki Region	24.9%	1.5%	23.4%
Kaita Town, Hiroshima Prefecture	73.2%	50.7%	22.5%
Yamaguchi Prefecture	12.7%	−9.4%	22.0%
Mihara City, Hiroshima Prefecture	27.2%	5.5%	21.7%
Higashi-Hiroshima City, Hiroshima Prefecture	36.3%	16.7%	19.7%
Asa-Minami Ward Ward, Hiroshima City	40.8%	22.0%	18.7%
Naka Ward, Hiroshima City	90.3%	75.7%	14.6%
Minami Ward, Hiroshima City	78.4%	64.4%	14.0%
Kanto Region	14.0%	0.5%	13.5%
Aki Ward, Hiroshima City	57.3%	47.6%	9.7%
Kure City, Hiroshima Prefecture	35.2%	26.2%	9.0%
Tottori Prefecture	3.1%	−5.0%	8.1%
Akitakata City, Hiroshima Prefecture	13.5%	6.2%	7.3%
Miyoshi City, Hiroshima Prefecture	16.9%	9.7%	7.2%
Higashi Ward, Hiroshima City	64.2%	57.2%	6.9%
Shikoku Region	8.5%	1.9%	6.6%
Fuchu City, Hiroshima Prefecture	10.8%	4.6%	6.1%
Shobara City, Hiroshima Prefecture	−9.4%	−10.5%	1.2%
Okayama Prefecture	−0.9%	−0.6%	−0.2%
Fukuyama City, Hiroshima Prefecture	1.1%	3.8%	−2.7%
Kumano Town, Hiroshima Prefecture	32.1%	36.1%	−4.0%
Shimane Prefecture	−14.6%	−9.7%	−4.9%
Kita-Hiroshima Town, Hiroshima Prefecture	−10.6%	−4.1%	−6.5%

).

Figure 3 delineates the administrative boundaries, railway and bus route networks, arterial roads, and the distribution of sports, cultural, and entertainment venues. In Saeki Ward, Hiroshima City, and Hatsukaichi City, the former stadium was connected by limited public transport routes but was easily accessible via arterial roads. By contrast, the new stadium benefits from multiple direct public transport routes. These two municipalities demonstrated the highest Choice-R% among municipalities in the survey, reflecting the most substantial behavioral change effects.

In contrast, Naka and Minami Wards in Hiroshima City, which are well-connected to the old stadium by public transport routes and are areas where tram use is convenient, were among the top three regions in R%. However, this was primarily because of the high Distance-R%, with the Choice-R% values remaining at 14.6% and 14.0% for Naka and Minami Wards, respectively, placing them in the mid-range.

The results of this study are significant for policymaking from several perspectives. First, it opens up the potential for restructuring public transportation networks prompted by the construction of large-scale urban facilities. Studies such as those of Pereira [55] discussed the societal significance of reorganizing public transport networks around new sports facilities; they emphasized ensuring accessibility to landmark facilities as a crucial issue associated with social inclusion. In Hiroshima, many facilities are located outside the city center, and access to these facilities remains easier by car than by public transport. For instance, as revealed in this study, the Hiroshima Big Arch, located in Hiroshima Regional Park, is more conveniently accessible by car than by other transport choices from the western parts of the Hiroshima metropolitan area. Similarly, the Nishi Ward Sports Centre, another large sports complex located in the southern part of Nishi Ward, is more accessible by car than by public transport from the western and southern parts of the metropolitan area. It has been argued that public transportation policies should be implemented with an awareness of social equity through a participatory and evidence-based process [56,57]. The data obtained in this study provide empirical evidence for municipalities to better understand and improve accessibility conditions in different districts, even within the same municipality.

Second, it has the potential to promote evidence-based discussions on the development of urban facilities that balance environmental and economic considerations. Regarding the economic effects of a stadium in the city center, Sanfrecce Hiroshima and the Hiroshima University of Economics estimated that each league match generates approximately 1.1 billion yen (equivalent to seven million U.S. dollars) [58]. For a comparison based on figures published by another Japanese prefecture, we found data showing that an event held to promote the consumption of agricultural products generated an economic impact of 628 million yen and emitted 924 tons of carbon dioxide [59]. While more detailed comparisons with other types of events are needed, the findings from Hiroshima indicate an economic impact of 1.1 billion yen resulting from approximately 550 tons of carbon dioxide emissions (assuming that spectator travel accounts for 34.65% of total carbon emissions, as in the case of VfL Wolfsburg [15]), which suggest that sporting events may generate more economic benefits with fewer emissions than other types of events. These results indicate that both environmental and economic impacts must be considered in the policymaking process that discusses the types of facilities to locate and the types of events to hold in the city center.

5. Conclusions

This study conducted an extensive questionnaire survey in Hiroshima, where a professional football club relocated from the suburbs to the city center, and found that the stadium relocation to the city center significantly affected spectators’ modal choices and significantly reduced both per capita and total carbon emissions on match days. These results are the first quantitative evidence of the significance of stadium relocation in sustainable urban planning. Moreover, the geographical analysis of reduction rates (R%) based on a residential area, segmented into the impacts of decreased travel distance to the stadium (Distance-R%) and behavioral changes in transport modal choices (Choice-R%), showed notable differences among districts. The results indicated that spectators were differently incentivized or disincentivized towards the sustainable choice, according to their places of residence.

The emissions estimation method proposed in this study, based on responses by individual spectators regarding the modes of transportation they use, could be further utilized. Conventional methodologies that require respondents to select modes of transportation to stadiums from general categories such as “airplane, train, automobile, or on foot” have low estimation accuracy, particularly in cases where public transportation usage is considerable and when spectators combine multiple modes of transportation. Conversely, methods requiring spectators to provide comprehensive information on both transportation modes and their respective routes may lead to diminished response rates. Hence, we employed a hybrid approach; although selecting transportation modes was obligatory, respondents could delineate their travel routes in an optional free-text section. Even though only 3.6% of the survey participants provided detailed information, at least one detailed route was obtained for each transport mode combination selected by more than 1% of the respondents. Thus, the study achieved coverage of 91.8% in emission estimates by averaging their detailed responses and applying the modal share distribution to others. This approach balances data volume with estimation accuracy and can be applied to any city.

However, future research should consider the disparities in modal choice trends that are influenced by the varying characteristics of public transport networks across different countries and regions. In Japan, metropolitan areas like Hiroshima have well-developed public transport systems, including trams, and many spectators travel by the Shinkansen, a high-speed rail network that links cities, resulting in relatively low emissions, whereas regional cities with less public transportation also continue to debate stadium locations in the suburbs or city centers. Thus, although our findings from Hiroshima offer insights relevant to such discussions, the circumstances surrounding them may differ considerably. Globally, while public transport modal shares are high in many cities in East Asia, Latin America, and Europe, similar to Hiroshima, regions such as North America exhibit different characteristics. It is imperative to consider the characteristics of the existing transportation network in each region and their potential outcomes when determining the appropriate locations for stadiums in a specific city.

In addition, the impact of carbon emissions from the professional sports industry on the natural environment needs to be discussed more comprehensively. Though this study focused on carbon emissions from spectator travel, which is classified under Scope 3 of the GHG protocol and is the largest contributor associated with professional sports, direct carbon emissions at stadiums and indirect ones from electricity and water usage must also be considered. Attention should also be paid to the negative impact of traffic congestion caused by the concentration of automobile traffic in the city center rather than in the suburbs. It should also be noted that although this study assumed the case where constructing a new stadium is a prerequisite when the previous stadium gets old, it is necessary to analyze the whole lifecycle’s carbon footprint. To test the above issues, a long-term study should be conducted to track changes after the relocation. Long-term research is also needed to examine whether stadium relocation will cause social impacts like gentrification and displacement of existing residents.

Developing a system to simulate the carbon footprint and economic impact of stadium locations would also benefit public decision-making. Such a system would allow local governments and other stakeholders to conduct interactive comparisons of multiple location options, help communities make decisions based on quantitative evidence, and be applied to large-scale projects beyond stadiums.