Literature Review of Socioeconomic and Environmental Impacts of High-Speed Rail in the World

: Countries considering high-speed rail (HSR) developments face enormous challenges because of their high deployment cost, environmental obstacles, political opposition, and their potentially adverse effects on society. Nevertheless, HSR services are importantly sustainable that can have positive and transformative effects on the economic growth of a nation. This paper systematically reviews and classiﬁes impact areas of HSR deployments around the world as well as the analytical methods used to evaluate those impacts. We have utilized the scholarly scientiﬁc database to ﬁnd articles in HSR systems. By deﬁning some rules, we select 116 articles between 1997 and March 2020. The approach revealed interesting patterns and trends in space, time, and sentiment of the analyzed impacts on society, the economy, and the environment. The ﬁndings can inform decision-making about HSR developments and deployments, and the gaps identiﬁed in the literature can propose new research opportunities for future studies.


Introduction
High-speed rail (HSR) is categorized as a transport mode that uses a combination of specialized rolling stock and dedicated tracks to transfer passengers between origins and destinations. HSR runs significantly faster and smoother than traditional trains. The first HSR systems appeared in Tokyo, Japan in 1964, known as bullet trains that connected cities with the aim of travel time savings and the mitigation of capacity constraints on existing railways. France was the next country to make high-speed rail available to the public in 1981-the system connected Paris and Lyon with speeds up to 124 mph. Today, the French high-speed rail network grew to more than 2800 km with the Lignes à Grande Vitesse (LGV) allowing speeds of up to 200 mph [1].
Some of the acknowledged benefits of HSR are shorter travel time across long distances, lower transportation costs, reduced fuel consumption, lower highway, and air traffic congestion, rising land values, and economic development in large and small communities [2]. According to a study by Chinese and U.S. economists, extending bullet trains can increase the value of the real estate, improve the quality of life, reduce air pollution, reduce traffic congestion, and provide more travel options for crowded cities, especially in the developing world [3]. The Chinese study is based on the nation's high-speed rail network, but the researcher's project is that the benefits experienced there would be the same for California's proposed HSR system. A study coordinated by Tsinghua University and the University of California found that bullet train systems connecting China's largest cities to nearby smaller cities have made those cities more attractive for workers, alleviated traffic congestion, and reduced pollution in megacities [4].

Methodology
The systematic literature review utilizes a quantitative analysis of article appearances [7][8][9]. The next subsections describe the databases used for article identification, the article screening and selection criteria, the classification of impact areas from HSR deployments, the classification of analytical methods used to assess those impacts, and the dimensions of the quantitative analysis.

Article Identification and Selection
The search was limited to articles indexed by the three most popular sources of identifying scholarly academic publications, namely the Scopus database, the Google Scholar search engine, and the Web of Science database. The search focused on scholarly academic research with good rigor by excluding books, conference proceedings, and student dissertations. Scopus produced more articles that were relevant to the fields of social, physical, health, and economic sciences [10,11], and from databases of the largest publishers of scholarly academic work such as Routledge, Elsevier, Springer, and Taylor and Francis. Google Scholar has emerged as a contemporary tool to scour the Internet for any source of scholarly academic work that a focused search using Scopus, or the Web of Science may miss.
The search focused on articles published by international peer-reviewed journals in the English language. The initial search revealed that there were more than 2100 arti-cles with the exact phrase "High-Speed Rail" in the title, abstract, or keywords. After examining the temporal pattern of relevant and continuous growth in publications, the search narrowed the selection to articles from 1997 to March 2020. This temporal filter reduced the number of articles to just under 2000. Further limiting the subject areas to social sciences, environmental science, business, energy, economics, decision sciences, and multidisciplinary reduced the count to just under 1000 articles. A similar search conducted using the "Advanced search" feature of Google Scholar resulted in more than 2000 articles. Unfortunately, the Google Scholar search did not allow for filters to be defined for article type, subject areas, or to limit the exact phrase to the title, abstract, or keywords. Therefore, the articles needed to be manually cross-referenced to further scrutinize those that did not overlap. Next, a careful review of the title, abstract, and keywords reduced the number of articles to just below 200. Finally, a more careful review of all the articles relative to the research objectives resulted in the selection of 116 articles.

Impact Area Classification
This subsection introduces the classified impact areas, derived by an empirical process. The resulting impact areas were eight broad but distinct categories: The next subsections summarize the significance of each impact area in the studies of HSR development and deployment.

Energy
Compared with aviation and automobiles, railroads are eight and four times more energy-efficient, respectively [6, 12,13]. Although some authors posit that a modal switch to HSR will substantially reduce energy demand [14], others found that the energy demand from HSR is significant [15].

Environment
Despite advancements in alternative zero-emission fuels [16], the transport sector continues to increase its carbon dioxide (CO 2 ) emissions. Hence, critics argue that HSR will have a relatively low impact on the reduction of CO 2 [17]. In contrast, Horvath, and Chester (2012) found that a transition from automotive trips to HSR would positively affect the atmosphere [18]. Yue et al. (2015) found that vehicle operation in China dominated environmental impacts [19]. Several studies concluded that shifting traffic to rail will positively benefit the environment [20][21][22][23].

Land
The effects of HSR on urban growth include interactions with land use and land price [24,25]. Studies associate investments in transport infrastructure with an increase of nearby land value [26]. The theory posits that greater accessibility for people and goods will increase urbanization [27], and the value of the nearby real estate, which in turn will increase property tax revenues [28]. Conversely, there are several adverse consequences such as noise and an increase in crime [28].

Economy
HSR promotes competition and economic cohesion [29]. The convergence of HSR with traditional railways can accelerate urban growth [30]. HSR has the potential to Sustainability 2021, 13, 12231 4 of 27 enable more rapid output flows and to optimize the distribution of resources to sustain economic growth [31]. HSR can help to increase productivity and interdependency among companies and cities [32][33][34]. The HSR effects of space-time compression can accelerate economic growth by reducing the transport and trading costs between cities [35][36][37]. The employment and knowledge economy will also grow because HSR can help to enhance the flow of information [38].

Health
The rapid and systematic urbanization from HSR deployments presented challenges such as the depletion of limited land resources, environmental degradation, air pollution, noise pollution, congested roadways, and poor living conditions, which ultimately affected public health [39][40][41].

Resilience
HSR infrastructure is vulnerable to various threats, including the wearing of parts, terrorist attacks, natural disasters, and spikes in demand [5,[42][43][44][45]. Even minor disturbances from wind speed variations can affect the normal operation of HSR [46].

Policy
Western governments often tie HSR initiatives to policymaking that relate to funding, political leadership, and national goals [47][48][49]. Applications for federal aid often include independent reviews of potential ridership, modal competition, and other forecasts [50]. Differences in political regimes can influence differences in various aspects of HSR deployments [51].

Innovations
Scientific and technical advancements have spurred economic growth [52]. HSR is an important part of contemporary transportation technology innovation to enhance mobility and accessibility [53][54][55][56][57]. Although global economic integration has led to shorter technological innovation cycles, HSR technology developments outside of China were less aggressive.

Analysis Technique Classification
Many analysis methods were applied to assess impacts from HSR deployments. The ten most prevalent methods assessed were difference-in-differences (DID), regression analysis (Regression), simulation-of-scenarios (Simulations), dependencies network analysis (Networks), spatial analysis (Spatial), optimism sentiment analysis (Sentiment), benefitcost-analysis (BCA), game theory (Games), and graph theory (Graph). Table 1 summarizes and describes each method that we listed here. The two most prevalent methods used were DID and Regression. The DID technique is a quantitative and statistical method that was used by most authors [36,39,56,[58][59][60][61][62]. The second most-used method was Regression, another statistical method used by authors [24,35,[63][64][65][66][67] to measure and analyze the relationships between various factors. In Table 1, we have explained each method separately as we mentioned early.
Optimism is linked to proactive measures for safeguarding health whereas pessimism is associated with behaviors that adversely affect health. [93] BCA A systematic prediction of social benefits and social costs to assess the economic effects of suggested public policies over time. The Benefit-Cost Analysis technique may incorporate principles of willingness-to-pay and opportunity cost. The outcome of a BCA is often policy decisions that would maximize net community benefits in monetary equivalents. [27,83,87,[94][95][96][97] Games Study of ways that the interactions among agents and economic factors produce outcomes relative to choices made. [76,98,99] Graphs Representation of transport networks as nodes and weighted links to model and understand flows, throughput, distributions, mode selection, noise, bottlenecks, connectivity, accessibility, and other factors. [66,68,74,[100][101][102] 2.4. Bivariate Frequency Distributions Patterns are exposed by using pair-wise combinations of bivariate frequency distributions for seven dimensions of the publications. The selected dimensions are time, frequency, location, publisher, impact area, analytical method, and sentiment. Time was simply the year of the publication. Frequency was a count of the number of publications. Location was the country or region where the article focused the study. The publisher was the journal where the article appeared. The impact area was one of the eight defined and explained previously. The analysis method was one of the ten explained in Table 1 above. Finally, the sentiment was the optimistic or pessimistic outlook of the article, classified as positive, negative, both, and unknown. The sentiment analysis is utilized to assess the effects of impact area as we defined in Section 2.2 on HSR systems. For instance, environmental effects on the HSR system were positive [103][104][105]. It means that implementing HSR could cause the reduction of CO 2 in two cities in Turkey. The appendix has all the article mapping tables (AMT) used to create either the univariate or bivariate frequency distributions.

Results
This section discusses the results of the review based on the combinatorial distributions of articles in time, frequency, location, methods, and sentiment. Figure 1 shows that there has been an exponential increase in the number of relevant articles published since 1997. As we can see in Figure 1, the number of papers published since 1997 has considerably increased from one study to 26. The reason that the number of articles in 2020 is decreased is due to the limitation that we have considered in this paper. So, the number of papers in 2020 is considered until 1 March 2020. Finally, the sentiment was the optimistic or pessimistic outlook of the article, classified as positive, negative, both, and unknown. The sentiment analysis is utilized to assess the effects of impact area as we defined in Section 2.2 on HSR systems. For instance, environmental effects on the HSR system were positive [103][104][105]. It means that implementing HSR could cause the reduction of CO2 in two cities in Turkey. The appendix has all the article mapping tables (AMT) used to create either the univariate or bivariate frequency distributions.

Results
This section discusses the results of the review based on the combinatorial distributions of articles in time, frequency, location, methods, and sentiment. Figure 1 shows that there has been an exponential increase in the number of relevant articles published since 1997. As we can see in Figure 1, the number of papers published since 1997 has considerably increased from one study to 26. The reason that the number of articles in 2020 is decreased is due to the limitation that we have considered in this paper. So, the number of papers in 2020 is considered until 1 March 2020.    Moreover, Table 2 summarizes the scope of the top four journals that contained most of the articles. The policy implications of theoretical analysis and the spatial implications of deployments were dominant themes. Moreover, Table 2 summarizes the scope of the top four journals that contained most of the articles. The policy implications of theoretical analysis and the spatial implications of deployments were dominant themes.

Transport Policy
Design of innovative policy, management practices, and applications bridging the gap between transport theory and practice.
Government, management, and industry strategies across all transport modes. Topics include international transport economics, strategies that influence policy, and effects monitoring.

Transport Geography
Issues and technological advances spanning the spatial dimensions of transportation, organization, structure, and operations that affect the globalization of economies. Deals with the effects of transport policy and governance on regions and places.
Articles analyze the impacts of transport investment on mobility, livelihoods, social networks, the spatial economy, patterns of development, travel behavior, and accessibility. Policy analysis, design, formulation, and evaluation; planning; interaction with the political socioeconomic and physical environments; management and evaluation of transport systems.
The paper focuses on a clear policy concern with research covering topics from disciplines including economics, engineering, psychology, sociology, and urbanism.

Transportation Research Part D: Transport and Environment
The environmental impacts of transportation and policy responses, implications for the design, planning, and management of transportation systems.
Covers all factors relating to environmental impacts including, air quality, ecosystems, global climate, economic development, quality of life, public health, and land use. Figure 3 shows the distribution of articles by the region analyzed and the cumulative proportion, ranked by frequency. The cumulative line conveys the dominance of studies within the first few countries and then the gradual increase of studies focused on other countries. The category "Other Countries" includes an analysis of HSR deployments in several countries. China and Spain accounted for 27% and 13% of the published articles, respectively. Several authors included multiple countries in their analysis [12,41,44,47,63,73,76,101]. In those cases, the articles were mapped to multiple regions.  The first high-speed rail system, known as the bullet train, was created in 1964 to link the Japanese cities of Tokyo and Osaka. China has the longest length of HSR in the world with 26,869 km; by comparison, the United Kingdom has about 1377 km of rail [106]. In 2009, there were more than 12,158 km of HSR in the world, which increased to more than 78,850 by 2018. Italy [107] the United Kingdom [108] and France [1] were the next countries to make HSR available to the public. In 1981, the French system connected Paris and Lyon with speeds up to 200 km/h. Today, the French high-speed rail network has expanded to more than 3220 km of rail networks that allow for speeds of up to 320 km/h. Germany's development lasted nearly two decades [109] whereas the other developments were complete within 10 years. HSR in China has developed rapidly over the past 15 years because of generous funding from their government [110]. In the early 1990s, China started planning the current HSR network as well as the country of Norway [111], modeling it after the Shinkansen system in Japan. Korea began construction shortly after China [112]. The first high-speed rail system, known as the bullet train, was created in 1964 to link the Japanese cities of Tokyo and Osaka. China has the longest length of HSR in the world with 26,869 km; by comparison, the United Kingdom has about 1377 km of rail [106]. In 2009, there were more than 12,158 km of HSR in the world, which increased to more than 78,850 by 2018. Italy [107] the United Kingdom [108] and France [1] were the next countries to make HSR available to the public. In 1981, the French system connected Paris and Lyon with speeds up to 200 km/h. Today, the French high-speed rail network has expanded to more than 3220 km of rail networks that allow for speeds of up to 320 km/h. Germany's development lasted nearly two decades [109] whereas the other developments were complete within 10 years. HSR in China has developed rapidly over the past 15 years because of generous funding from their government [110]. In the early 1990s, China started planning the current HSR network as well as the country of Norway [111], modeling it after the Shinkansen system in Japan. Korea began construction shortly after China [112]. Table 3 summarizes the length of a line in operation, under construction, and not started construction [113]. Table 3 is ordered from the largest network to the smallest. Table 3 also includes the year of construction start to the commencement of operations for the first HSR deployment in that country. The last column of Table 3 lists the number of years taken to develop and deploy the first HSR system in each country. The average time is 7.48 years, and the standard deviation is 4.82 years. The average development time for countries in the Asian continent was 8.83 years as compared with 8.25 years for European countries.  3038  330  0  1991  1973  300  18  Sweden  1706  180  0  1985  1979  200  6  UK  1377  230  320  1976  1969  201  7  South Korea  1104  376  49  2004  1992  300  12  Italy  999  116  0  1977  1970  250  7  Turkey  802  1208  1127  2007  2003  250  4  Russia  845  0  770  2012  2009  250  3  Finland  609  0  0  2006  2005  200  1  Uzbekistan  344  0  0  2011  2011  250  0  Austria  352  208  0  2008  1999  230  9  Taiwan  354  0  0  2007  1998  300  9  Belgium  326  0  0  1997  1997  260  5  Poland  224  0  484  2014  2011  200  3  Netherlands  175  0  0  2009  2000  300  9  Switzerland  144  15  0  2005  2000  200  5  Luxembourg  142  0  0  2007  2002  320  5  Norway  64  54  0  1998  1997  210  1  USA  54  192  1710  2030  2020  240  10  Saudi Arabia  453  453  0  2018  2009  299  9  Denmark  60  56  0  2019  2012  180  7  Thailand  0  873  615  2023  2017  250  6  Iran  0  410  1351  2021  2015  270  6  Indonesia  0  0  712  2021  2015  250  6  India  0  0  508  2023  2018  250  5  Malaysia  0  0  350  2031  2019  250  12  Israel  0  0  85  2018  2001  250  17  Portugal  0  0  550  2015  2000  250  15  Czech Rep.  0  0  660  2030  2017  250  13  Greece  0  500  200  2021  2018  250  3 Amtrak is a government-owned railroad company that provides high-speed service across only small portions (55 km) of the system [114 -116]. The state of California recently secured funding to solidify plans for a route between Los Angeles and Las Vegas [117]. Figure 4 shows a frequency distribution in the order of impact area ranking. Figure 5 shows the distribution of articles from 2011 to 2020, subcategorized by impact area. Figure 6 shows the distribution of articles by impact area, subcategorized by the deployment location analyzed.

Analysis Technique Distribution
Assessments included a wide range of factors such as deployment cost, travel time, accessibility, life cycle, energy consumption, economic impacts, emissions, fares, speed comfort, demand, impacts on tourism, spatial interactions, changes in land cover, and the opportunity costs of travel time savings. Table A1 in Appendix A summarizes the main assessment from each article. Where articles assessed multiple factors in different regions, using several different techniques, the table lists the dominant factors assessed. The discussion section provides a synthesis of the table with a discussion of the advantages and disadvantages of each method. Figure 7 shows the distribution of analysis techniques used in HSR analysis. As we can see in Figure 7, the two methods of DID and Regression are among the methods that have been used by many researchers. Figure 8 shows the distribution of the articles by analysis technique, subcategorized by year. Figure 9 shows the distribution of the articles by analysis technique, subcategorized by impact area. Figure 10 shows the distribution of the articles by analysis technique, subcategorized by location.

Analysis Technique Distribution
Assessments included a wide range of factors such as deployment cost, travel time, accessibility, life cycle, energy consumption, economic impacts, emissions, fares, speed comfort, demand, impacts on tourism, spatial interactions, changes in land cover, and the opportunity costs of travel time savings. Table A1 in Appendix A summarizes the main assessment from each article. Where articles assessed multiple factors in different regions, using several different techniques, the table lists the dominant factors assessed. The discussion section provides a synthesis of the table with a discussion of the advantages and disadvantages of each method. Figure 7 shows the distribution of analysis techniques used in HSR analysis. As we can see in Figure 7, the two methods of DID and Regression are among the methods that have been used by many researchers. Figure 8 shows the distribution of the articles by analysis technique, subcategorized by year. Figure 9 shows the distribution of the articles by analysis technique, subcategorized by impact area. Figure 10 shows the distribution of the articles by analysis technique, subcategorized by location.

Analysis Technique Distribution
Assessments included a wide range of factors such as deployment cost, travel time, accessibility, life cycle, energy consumption, economic impacts, emissions, fares, speed comfort, demand, impacts on tourism, spatial interactions, changes in land cover, and the opportunity costs of travel time savings. Table A1 in Appendix A summarizes the main assessment from each article. Where articles assessed multiple factors in different regions, using several different techniques, the table lists the dominant factors assessed. The discussion section provides a synthesis of the table with a discussion of the advantages and disadvantages of each method. Figure 7 shows the distribution of analysis techniques used in HSR analysis. As we can see in Figure 7, the two methods of DID and Regression are among the methods that have been used by many researchers. Figure 8 shows the distribution of the articles by analysis technique, subcategorized by year. Figure 9 shows the distribution of the articles by analysis technique, subcategorized by impact area. Figure 10 shows the distribution of the articles by analysis technique, subcategorized by location.

Article Sentiment Distribution
The result of the assessed article sentiment was based on whether the article suggested or implied that impact or impacts of the investigated factor or factors were positive, negative, both, and unclear. Overall, positive, negative, both, and unknown sentiments accounted for 59.6%, 18.4%, 12.3%, and 9.6%, respectively. Figure 11 shows the distribution of publications by year, subcategorized by sentiment. Negative sentiments dominated before 2011, after which positive sentiments began to dominate. Figure 12 shows the distribution of locations, subcategorized by sentiment. Figure 13 shows the distribution of impact areas, subcategorized by sentiment. There were no positive sentiments for articles within the impact area of health. Conversely, there were no negative sentiments for articles addressing policy. Figure 14 shows the distribution of the analysis methods, subcategorized by sentiment. There were no negative sentiments for methods involving graph theory and sentiment analysis.

Article Sentiment Distribution
The result of the assessed article sentiment was based on whether the article suggested or implied that impact or impacts of the investigated factor or factors were positive, negative, both, and unclear. Overall, positive, negative, both, and unknown sentiments accounted for 59.6%, 18.4%, 12.3%, and 9.6%, respectively. Figure 11 shows the distribution of publications by year, subcategorized by sentiment. Negative sentiments dominated before 2011, after which positive sentiments began to dominate. Figure 12 shows the distribution of locations, subcategorized by sentiment. Figure 13 shows the distribution of impact areas, subcategorized by sentiment. There were no positive sentiments for articles within the impact area of health. Conversely, there were no negative sentiments for articles addressing policy. Figure 14 shows the distribution of the analysis methods, subcategorized by sentiment. There were no negative sentiments for methods involving graph theory and sentiment analysis.           Figure 15; Figure 16 are a visualization of the frequency and relationship of keywords from all the articles analyzed. HSR studies involved both reviews and case studies related to impacts and implications on the airline industry and tourism.    Figure 15; Figure 16 are a visualization of the frequency and relationship of keywords from all the articles analyzed. HSR studies involved both reviews and case studies related to impacts and implications on the airline industry and tourism.

Discussion
This section provides observations about the distributions of the journal articles, the location focus, the impact area assessed, the analysis technique used, and the sentiments conveyed as well as some future research gaps for future research studies. Figure 1 shows that there has been an exponential increase in the number of articles published since 1997, with only five articles from 1997-2010. HSR analysis is published in a wide range of journals, but publications were more highly skewed towards policy and environmental implications. Figure 2 shows that the journals, Transport Policy, Transport

Discussion
This section provides observations about the distributions of the journal articles, the location focus, the impact area assessed, the analysis technique used, and the sentiments conveyed as well as some future research gaps for future research studies. Figure 1 shows that there has been an exponential increase in the number of articles published since 1997, with only five articles from 1997-2010. HSR analysis is published in a wide range of journals, but publications were more highly skewed towards policy and environmental implications. Figure 2 shows that the journals, Transport Policy, Transport

Discussion
This section provides observations about the distributions of the journal articles, the location focus, the impact area assessed, the analysis technique used, and the sentiments conveyed as well as some future research gaps for future research studies. Figure 1 shows that there has been an exponential increase in the number of articles published since 1997, with only five articles from 1997-2010. HSR analysis is published in a wide range of journals, but publications were more highly skewed towards policy and environmental implications. Figure 2 shows that the journals, Transport Policy, Transport Geography, Transportation Research Part A: Policy and Practice, and Transportation Research Part D: Transport and Environment published most of the articles. Figure 17 graphically illustrates the timelines for the first deployments in the selected countries. The pattern indicates that construction can last between 1 and 18 years, and there is no obvious regional correlation to the length of development time. In summary, there was very little difference in the development times between the two regions of the world.  Figure 17 graphically illustrates the timelines for the first deployments in the selected countries. The pattern indicates that construction can last between 1 and 18 years, and there is no obvious regional correlation to the length of development time. In summary, there was very little difference in the development times between the two regions of the world. The distribution of articles by region analyzed (Figure 3) shows that China, Spain, and Italy accounted for more than half of the articles published. The pattern indicates that there was an acceleration of papers after the second wave of European deployments were complete. Figure 4 revealed that HSR impacts of the highest and the lowest interest were economic development and health, respectively. Articles directed at policymaking ranked next to last even though most of the journals in which articles appeared dealt with policy implications. The pattern in Figure 5 shows that studies about HSR impacts on economic development, the environment, and land use were consistent over the years. Topics in Figure 17. HSR development from funding approval to launch. The distribution of articles by region analyzed (Figure 3) shows that China, Spain, and Italy accounted for more than half of the articles published. The pattern indicates that there was an acceleration of papers after the second wave of European deployments were complete. Figure 4 revealed that HSR impacts of the highest and the lowest interest were economic development and health, respectively. Articles directed at policymaking ranked next to last even though most of the journals in which articles appeared dealt with policy implications. The pattern in Figure 5 shows that studies about HSR impacts on economic development, the environment, and land use were consistent over the years. Topics in innovation emerged after 2014. The pattern in Figure 6 highlights those studies about the impact of HSR on the economic development of China and Spain dominated.

Analysis Technique Distribution
All methods used in this review have advantages and disadvantages as summarized in Table 4. All the methods used were quantitative. The DID and Regression methods were the most popular overall (Figure 7), likely because of their statistical rigor. Although the DID method was most popular overall, it began to appear consistently since 2016 as observed in Figure 8. Similarly, the Regression method began to appear consistently in 2014. The use of game theory emerged after 2017. Simulations were used consistently throughout the years.

DID
A simple method of comparing changes between a treated and an untreated time series, assuming that the latter is unaffected but similar.
Requires similar and stable data from both series.

Regression
The coefficients of the functional relationship model can convey the relative importance of variables.
Assumes that the cause-and-effect relationship between variables remains unchanged. May lead to erroneous and misleading results if the assumptions do not hold.

Simulation
Provides practical feedback when designing real-world systems, before construction. Allows for the efficient exploration of various scenarios before investments and construction.
Can be computationally intensive, which could delay results. May not accurately model the phenomenon.

Network
Accounts for dependencies and relationships to produce insights for improving management, planning, scheduling, and project control.
Evaluations could involve complex interactions that are difficult to visualize. Could be computationally complex.

Spatial
Accounts for locational aspects of attributes and considers spatial dependencies and correlations.
Spatial interactions destroy the assumption of variable independence.

Sentiment
Associates' population growth with patterns in technology adoption based on waves of optimism, pessimism, and uncertainty in the marketplace.
Does not account for failures in innovation due to fear, uncertainty, doubt, or reduction of funding and political support.
BCA Well-studied, simple, objective, and requires goal settings, for example, payback period.
Benefits often require quantification of intangibles and require determination of a discount rate, which may be unknown.

Graphs
Well-studied with fertile theoretical constructs that can represent complex flows and interactions to solve a wide variety of transport and logistical problems.
Representation of the network may be too simplified to account for unexpected disruptions such as terrorism or natural disasters.

Games
Often arrives at an optimal strategy when rules are well-defined.
Complete information is required for all parties involved. It can involve recursive thinking and dependencies among opponents, which can be intractable for humans to follow.
As observed in the pattern of Figure 9, all methods except Sentiment were applied to the analysis of HSR economic impacts. Except for Sentiment, all other methods were applied in articles that discussed HSR innovation. Only DID, Regression analysis, spatial analysis, and network analysis were used to study the health impacts of HSR. DID and game theory were the only methods not applied to the study of HSR resilience. Figure 10 revealed some interesting patterns. Curiously, articles that studied HSR deployments in China applied all methods except BCA and sentiment analysis. Similarly, articles about French deployments applied all methods except DID and sentiment analysis. For Spain, the only method not used was sentiment analysis. The Regression method was used to study HSR deployments in all countries except the United States and Australia. Interestingly, the network analysis method was applied to studies of deployments in all countries, except the United Kingdom. Sentiment analysis was applied only to the assessments of U.S. deployments.

Article Sentiment Distribution
Interestingly, the number of articles reporting positive impacts was more than triple the number of those reporting negative impacts. However, most sentiments were negative before 2011 and started to grow in the positive direction thereafter (Figure 11). The sentiments were most balanced for articles published in 2019. Positive comments dominated in most countries. Interestingly, sentiments about deployments in China were more balanced than for any other country ( Figure 12). There were no positive and no negative sentiments for studies about the health impacts and policy implications of HSR deployments, respectively ( Figure 13). Sentiments about economic impacts were more balanced than for the other impact areas studied ( Figure 13). Studies using the DID method had a more balanced sentiment than those using the other methods ( Figure 14). The sentiment was only positive for studies using sentiment analysis and graph theory.

Gap Assessment
There has been a lack of studies on the short-and long-term effects of HSR deployments on health, including microscopic and macroscopic studies on psychological effects. Figure 4 illustrates that the study of economic impacts dominated with 62 papers whereas only four papers covered health impacts. Furthermore, only one study used sentiment analysis whereas 17 papers used regression as an analytical method. Techniques such as text mining are continuously evolving as an effective method to analyze sentiments on largescale datasets, including social media feeds. Although positive sentiments outweighed negative sentiments, more needs to be understood about underlying reasons for the latter. Another observation is that there were very few assessments about how cultural, social, or political positions, for example on labor laws, right-of-way acquisition, and environmental regulations, correlated with deployment decisions and timelines. Furthermore, there were no tests of a hypothesis that the sudden acceleration in Chinese deployments during the 1990s was associated with enhanced trading with Western nations.
The effective supply of labor is of great significance to economic development. China has the world's largest HSR market and the second-largest air passenger market. Therefore, with regards to the effect of HSR competition on other means of transportation the literature has promising gaps. When the accessibility of a tourist destination is considered, it is unavoidable to face the problem of multiple transport combinations. The opening of HSR directly increases the transportation modes available to tourists. The calculation of the fastest mode of transportation between two places has become the primary task of the measurement of accessibility. These innovations may lead individual rail companies to adopt different solutions depending on their needs and budgets. Additionally, rail lines that anticipate implementing new rail technologies, such as high-speed rail, or that focus on specific types of freight, may implement new technologies optimized for those options. It is unclear how the operational performance of transportation systems may vary among different weather events from an ex-post perspective. The indicator approach may help emergency management agencies recognize the vulnerability of transportation to severe weather events. Although several studies have examined the impact of extreme weather on transportation, the investigation of this issue from the supply-side with a focus on assessing how the on-time performance of transport systems is affected by severe weather has been scant. Future transportation planning agencies need to develop appropriate strategies to further improve system resilience of different modes of transportation to reduce negative consequences caused by future extreme weather events. Future transportation planning needs to facilitate the development of an integrated transportation system to minimize trip disruptions under severe weather events, researchers say. Although some studies have evaluated the effect of High-Speed Rail on the urban tourism economy, gaps in the literature remain. The impact of HSR on tourism has been inadequately analyzed from a comprehensive perspective.
Another thing is that authors should not only pay attention to the development of the central developed cities, but also the development of surrounding underdeveloped cities. The indirect benefit analysis of the new HSR project should include the impact on airfares. Several studies have found significant monopoly premiums on many routes in China's airline market. The introduction of HSR can increase external competition for the airline industry, thus reducing the monopoly and increasing the competition in the high-speed passenger transport market. First, in the short and medium distance of 50-700 km, high-speed rail has overwhelming advantages. Second, the improvement of accessibility of small and medium-sized cities is the most significant. The results showed that the impact of HSR on the number of inbound tourists is significantly positive because the attraction of HSR to inbound tourists measures 12% higher than the attraction of HSR to domestic tourists. Moreover, the impact coefficient on total inbound tourism revenue is 11.2%, but the impact coefficient on domestic tourism revenue is not significant.

Conclusions
Although extremely expensive to deploy, operate, and maintain, high-speed rail (HSR) positively benefits society. This work revealed that the effects of space-time compression through faster travel provided several economic benefits. Such benefits included lower fuel consumption, job creation, tourism growth, congestion reduction, and enhancements of real estate value along service routes. However, there were also adverse impacts on the environment such as increased pollution, community severance, and risks to biodiversity.
The patterns of article appearance and subject matter covered indicated that interest in the societal effects of HSR sustained an exponential growth since 2011. The exponential growth appeared after the second wave of HSR deployments across Europe. The coverage of HSR deployments in China, Spain, and Italy was dominant. Those countries also had the most extensive HSR infrastructure. The focus of articles on economic, environmental, and land-use impacts remained consistent across time. The articles reviewed revealed mostly positive sentiments about impacts on society. Techniques known for their statistical rigor, such as DID method and Regression analysis, dominated all cross-sections of time, location, impact area, and sentiments assessed.
The authors used all methods except benefit-cost analysis (BCA) and sentiment analysis to analyze deployments in China. Conversely, those methods were used more extensively to analyze deployments in the United States and Europe, suggesting that there were regional differences in influence. These findings suggest that cultural, social, and political differences may influence decisions about HSR deployments in a variety of ways. Therefore, future research will examine how such differences may have influenced HSR deployments in Western, Eastern, and European countries. Factors will include environmental laws, labor laws, land ownership, terrain, and influences by political lobbying from competing modalities such as aviation and trucking.

Conflicts of Interest:
The authors declare no conflict of interest.

Appendix A
This appendix contains the article mapping tables used to produce the bivariate categorical frequency charts. Emerging travel trends and the resurgence of rail transporthigh-speed rail and its station hubs Yue et al. [19] Vehicle, infrastructure, operation Krishnan et al. [14] Interstate passenger transportation portfolio, fuel, and electricity consumption, and 40-year cost and carbon dioxide emission Delaplace et al. [79] Cost of HSR relative to air and rail alternatives Clauzel et al. [101] The potential long-distance effect of a high-speed railway line Jones et al. [120] Train manufacturing, train operation, train maintenance, train disposal, track construction, track operation, and maintenance, and track disposal Alvarez-sanjaime et al. [49] Vertical structure of the rail sector Liorca et al. [121] The economic, social, and environmental impacts of transportation Co-authorship, co-citation, and keywords co-occurrence in the preceding part of the text Yu et al. [55] Accessibility scores of the future HSR corridor during peak and off-peak hours Fengjun et al. [125] Transport circle and accessibility of HSR in East Asia Zheng et al. [53] The spatial spillover effect of HSR stations Yin et al. [126] Tourism spatial interactions Raturi et al. [99] Speed and passengers' characteristics