Unraveling the Tourism–Environment–Equity Nexus: A Neighborhood-Scale Analysis of Texas Urban Centers

Abstract

This study explores the complex interplay between air pollution, the socioeconomic conditions, and the tourism density within Texas’s urban landscapes, focusing on Dallas, Houston, San Antonio, and Austin. Despite extensive research on environmental justice and urban tourism separately, few studies have integrated these fields to examine how tourism development intersects with environmental and socioeconomic disparities at a neighborhood level. This research addresses this gap by employing advanced geospatial analyses and multi-criteria decision analysis to reveal the pronounced clustering of stressed communities on urban peripheries, often removed from tourism’s economic benefits. The study uniquely quantifies the spatial mismatches between tourist hotspots and areas of environmental stress, a dimension often overlooked in the environmental justice literature. Local spatial statistics and cumulative impact analysis uncover statistically significant correlations between high poverty levels and elevated air pollution in specific locales. The results show varying patterns across cities, with Austin presenting the lowest inequality levels and San Antonio exhibiting significant disparities. This granular, neighborhood-centric approach provides novel insights into the tourism–environment–equity nexus, addressing the lack of comprehensive studies linking these factors in rapidly growing Texan metropolitan areas. The findings underscore the critical need for targeted policy interventions and neighborhood-specific approaches in diagnosing urban environmental disparities and crafting equitable urban development policies that consider tourism’s impact on local communities.

1. Introduction

As cities expand and tourism grows, environmental justice concerns have intensified around the unfair distribution of pollution, amenities, and cumulative burdens within urban areas [1,2]. The complex interplay between urban development, environmental quality, and socioeconomic factors has become a critical area of study, particularly in rapidly growing metropolitan regions. This research focuses on four major Texan metro regions—Dallas, Houston, San Antonio, and Austin—to explore the intricate relationships between air pollution, the socioeconomic conditions, and the tourism density at the neighborhood level.

These metropolitan areas face significant challenges related to rapid urbanization, uneven economic growth, and environmental degradation. The tourism–environment–equity nexus in these urban areas is particularly complex and multifaceted. Tourism activities can significantly impact local environments through increased air pollution, strain on water resources, and waste generation. Simultaneously, the economic benefits of tourism often concentrate in specific areas, potentially exacerbating existing socioeconomic disparities.

Most existing research utilizes county-level analysis, which fails to capture the localized disparities manifesting distinctly across neighborhoods [3]. This study aims to address this limitation by leveraging emerging geospatial techniques to quantify the intra-urban variability at the census tract level. By adopting a neighborhood-level lens, this research matches the analytical scale to the localized contexts that shape resident experiences and outcomes.

The novelty of this research lies in the comprehensive integration of multiple data sources, including air pollution, socioeconomic indicators, and the tourism density, to provide a holistic understanding of the complex interplay between these factors within the specific context of Texas urban centers. By combining and analyzing these diverse datasets at a granular, neighborhood-centric scale, the study offers unique insights into the localized manifestations of environmental injustice, potentially informing targeted policymaking and urban planning strategies tailored to the distinct challenges faced by different communities within these metropolitan regions.

Previous research has demonstrated associations between social disadvantage and industrial pollution, highlighting inequalities in environmental exposure and access to amenities based on race, ethnicity, and income [4,5,6,7]. Rapid urbanization has exacerbated localized pollution burdens, disproportionately affecting marginalized residents and increasing the risks of adverse health outcomes like asthma [8,9,10,11].

Integrating land use, vulnerability, and cumulative impact indicators can elucidate the complex pathways linking social factors, environments, and health equity [12]. Composite indices enable the nuanced analysis of the ways in which differential vulnerability shapes risk disparities [13]. Our study demonstrates these techniques by developing a neighborhood air pollution index combining proximity to emissions sources, poverty rates, and cumulative exposure to identify environmental justice hotspots.

The impact of tourism on urban peripheries is an important aspect of this study. While tourism development in peripheral areas can contribute to regional economic growth [14,15], it may also lead to issues like overtourism and the loss of local character [16,17]. Urban tourism transformations in peripheral neighborhoods, as seen in Barcelona, have brought about significant social and economic changes [18]. It is crucial for local governments to implement sustainable planning strategies to balance the benefits and costs of tourism in these areas, ensuring that the economic advantages do not overshadow the negative impacts on the local community and environment.

This study aims to demonstrate the value of granular, neighborhood-centric techniques, matching the analytical scale to the community context, for the accurate diagnosis of disparities obscured in coarse, municipal-level aggregations. These integrative methods, combining multivariate mapping and spatial statistics, can guide policy targeting and planning around equitable urban development.

Specifically, we investigate how air pollution, poverty, and tourism density intersect at the neighborhood level, potentially exacerbating or mitigating existing inequalities. We explore how the spatial distribution of tourist attractions relates to patterns of environmental pollution and socioeconomic status, hypothesizing that areas with high tourist activity may experience environmental degradation that disproportionately affects low-income neighborhoods.

Our analysis provides a nuanced understanding of these complex relationships, informing more targeted and effective urban planning and policy interventions. By examining these relationships, we aim to uncover how tourism development intersects with environmental justice issues in urban settings.

This study utilizes advanced geospatial analyses to explore the interplay between air pollution, socioeconomic conditions, and tourism density within Texas’s urban landscapes. Through multi-criteria decision analysis (MCDA) and other computational techniques, this research highlights the pronounced clustering of stressed communities on the peripheries of urban centers, often removed from the economic benefits of tourism.

We employ local spatial statistics and cumulative impact analysis to uncover statistically significant correlations between high poverty levels and elevated air pollution in specific locales. This approach allows us to identify and quantify environmental justice hotspots and examine their relationships with tourism activity centers.

2. Literature Review

Quantifying the uneven distribution of environmental burdens is central to environmental justice scholarship [19]. Studies have examined the scale and extent of the variation in spatial (in)justice patterns [3] and the associations between social disadvantage and industrial pollution, highlighting inequalities [4]. Integrating natural science and social justice topics via community co-production can address biases like quantitative dominance and the underrepresentation of highly polluted societies [20,21]. Improving assessments is crucial to tackle burden disparities.

While this study primarily builds upon established methodologies in the field of environmental justice, such as spatial analysis, composite indices, and vulnerability mapping, it aims to contribute a novel perspective by integrating these approaches within the context of tourism development in Texas’ urban centers. By explicitly examining the spatial relationships between environmental burdens, socioeconomic factors, and the density of tourist attractions, this research extends the discourse on environmental justice to encompass the potential impacts of urban tourism on marginalized communities. This integration of tourism dynamics into the analysis of environmental disparities offers a unique lens through which to explore the complex interplay between economic development, environmental quality, and social equity. Furthermore, the study’s emphasis on granular, neighborhood-level analysis challenges the traditional reliance on coarser geographic units, uncovering localized patterns and variabilities that may be obscured in broader, city-wide assessments.

Incorporating social vulnerability and land use indicators capturing cumulative exposure drivers and health impacts is increasingly emphasized [12,13,22,23,24,25,26]. Composite metrics like SVI enable the nuanced analysis of vulnerability’s role in health inequities. These approaches elucidate the complex pathways linking social factors, environments, and outcomes.

Rapid urbanization has intensified the localized burdens disproportionately affecting minority/lower-income residents through pollution exposure, increasing disease risks [11,27,28]. Children in lower-income communities also face higher asthma risks from environmental exposure [29]. Addressing exposure disparities through inclusive planning and policy initiatives is critical [30].

Granular analysis necessitates moving beyond the coarse county-level aggregation dominating prior work [3]. Our methods demonstrate the neighborhood-scale quantification of localized injustices obscured in city-wide averages.

Overlaying tourism illuminates differing resident and visitor exposure [31,32,33]. Visitors may encounter increased resource consumption, landscape damage, pollution, and ecosystem disruption [34]. Coastal tourism growth can increase the storm vulnerability, while urban park visitors may lack environmental health risk awareness [35]. Sustainable tourism development should consider these burdens. Our analysis integrates attractions given Texan cities’ extensive tourism.

Proximity mapping should account for the transient exposure amplifying localized risks. Estimates from dispersion models like RLINE are sensitive to meteorological inputs and measurement errors [36,37]. Perception-based techniques incorporating local inhabitants’ views provide spatial contaminant representations beyond simulated radii [36]. Multiscale models like Pangea also help to assess emissions’ local to global intake fractions [38]. Considering transience and using diverse mapping techniques improves hotspot risk assessments.

Prior spatial analyses have uncovered urban environmental inequalities. Nationally, non-White and low-income populations faced higher particulate matter facility burdens [39]. Regional studies in Tampa and St. Louis found that minorities and disadvantaged groups experienced higher traffic pollution risks [3] and childhood asthma clustered with poverty [40]. Tourism-overlaid studies measured polycyclic aromatic hydrocarbons (PAHs) in Houston neighborhoods to estimate the Harvey exposure disparities [41]. Studies in St. Louis and Paris identified toxic hotspots and heat vulnerabilities in lower-income areas [42,43]. A Seattle analysis showed industrial risks and deprivation converging during gentrification [44].

Small geographic units are critical in capturing localized injustices otherwise obscured, as seen through COVID-19 and flood exposure studies using fine French departments and Santa Fe census blocks [45,46]. Our framework demonstrates methods quantifying such granular cumulative stresses using a neighborhood lens.

3. Data and Methods

3.1. Data Sources

This section details the key environmental hazard and socioeconomic datasets utilized in the small-area spatial analysis at the urban census tract level in metropolitan regions of Texas (Figure 1). Our analysis focuses on four Texan metropolitan regions—Dallas-Fort Worth, Houston, Austin, and San Antonio—which have witnessed substantial growth since 2000. These diverse cities represent key anchors driving the wider state economy across the technology, energy, finance, logistics, and tourism sectors. The multi-criteria geospatial analysis used granular emissions and demographic and land use data from regulatory monitors, the census bureau, and open GIS repositories. All geospatial processing relied on ArcGIS, with the hotspot analysis undertaken using emerging neighborhood-scale techniques that overcome the limits of traditional county-level aggregations. Statistical correlations supplement identified hotspots to quantify pollution–poverty burden relationships formally.

This study utilized data from multiple sources to conduct our analysis of air pollution, the socioeconomic conditions, and the tourism density in four major Texas metropolitan areas: Dallas, Houston, San Antonio, and Austin. Air pollution data were sourced from the U.S. Environmental Protection Agency (EPA) National Emissions Inventory (NEI) for 2017, which was the most recent complete dataset available at the time of study. The EPA data are considered highly reliable and are widely used in environmental research. Socioeconomic data were obtained from the U.S. Census Bureau’s American Community Survey (ACS), using 5-year estimates covering 2015–2019. ACS data were collected through rigorous sampling methods and are considered a reliable source of demographic and economic information.

Tourism density data were gathered from state tourism boards and local convention and visitors bureaus for 2019, chosen to avoid anomalies in tourism patterns due to the COVID-19 pandemic. While official, we acknowledge that these data may have some limitations in capturing all tourism activity. Geospatial data were sourced from the U.S. Census Bureau TIGER/Line Shapefiles for 2019, which are considered the gold standard for U.S. geographic and cartographic information.

All data were collected at the census tract level to ensure consistency in spatial resolution across datasets. We acknowledge that there may be some temporal misalignment between datasets, which is a limitation when using the most recent available data for each category. However, we believe that the chosen time ranges provide a representative picture of the urban landscapes under study. Data processing and integration were performed using ArcGIS Pro (version 2.9) and the R statistical software (version 4.1.0) to ensure the accuracy and reproducibility of our analyses.

3.1.1. Non-EGU Air Pollution Point Sources

The Non-Electric Generating Unit (Non-EGU) Air Pollution point source emissions dataset developed by the National Parks Conservation Association was included to capture additional industrial emissions exposure risks beyond those modeled in NATA, with a focus on oil and gas infrastructure. This dataset maps the locations of facilities and sites generating emissions that impact the air quality and visibility in national parks and wilderness areas. Annual emissions of the criteria air pollutants were totaled for each county in Texas and spatially joined with the census tracts to quantify the exposure risks.

3.1.2. Heavy Traffic Proximity

Major edges from the Texas road network were selected based on traffic counts and classified into hierarchical groups using the TxDOT Roadway Inventory layer, which is a statewide dataset that has attribute information routed to TxDOT Roadway linework. By using linear referencing tools, attribute information from the TxDOT Roadway Inventory table was located on the linework. Roadway attributes such as functional systems, traffic counts, and surface types, among many others, can be found on a roadway simply by selecting it or performing a query. This product is created annually by the Transportation Planning and Programming Division at TxDOT in the Data Analysis, Mapping, and Reporting Branch for internal and public use. The TxDOT Data Analysis and Mapping tool is a comprehensive geographic information system that integrates various transportation-related datasets. It allows for the spatial analysis and visualization of road networks, traffic patterns, and infrastructure data, crucial in understanding the distribution of traffic-related air pollution [47,48,49]. Figure 2 illustrates the distribution of major air pollutants across the case studies.

3.1.3. Census Socioeconomic Data

To represent the neighborhood-level socioeconomic conditions, 5-year estimate data from the 2018–2020 American Community Survey (ACS) were extracted at the census tract level for key variables, including poverty rates and the median household income. The ACS demographic and housing characteristics were joined with the TIGER/Line shapefiles containing geospatial boundary definitions for all census tracts across Texas. The TIGER/Line shapefiles, provided by the U.S. Census Bureau, are digital geographic databases that provide spatial representations of various geographic features, such as roads, railroads, rivers, and legal and statistical geographic areas. These shapefiles are essential in mapping and analyzing census data at different geographic levels [50]. The income and poverty metrics serve as important factors influencing environmental equity issues and provide indicators of community vulnerability.

3.2. Geospatial Analysis

This section explains the geographic information systems (GIS) techniques utilized to integrate the environmental hazard and socioeconomic data sources spatially for a multivariate comparison and analysis.

3.2.1. Multi-Criteria Decision Analysis

A multi-criteria decision analysis (MCDA) approach was applied to synthesize the cancer risk estimates, chemical facilities density, non-EGU point source emissions, and traffic proximity into an integrated air pollution hazard index value for each census tract. Specifically, the Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS) method was used to determine relative weights and aggregate the indicators. Entropy, standard deviation, and expert judgment were used to assign variable weights reflecting the relative importance of each pollutant and emissions source to the overall air quality concerns. The Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS) is a multi-criteria decision analysis method that compares a set of alternatives by identifying weights for each criterion, normalizing the scores for each criterion, and calculating the geometric distance between each alternative and the ideal alternative. This method is particularly useful in environmental decision-making processes where multiple, often conflicting criteria need to be considered [51,52]. This enabled the development of a tract-level index of cumulative environmental hazard exposure risks.

3.2.2. TOPSIS Weight Calculation

In the TOPSIS analysis, 3 criteria were utilized, the chemical facilities density (CFD), the emissions source density (ESD), and the traffic proximity (TP), with TP divided into two sub-criteria: high traffic proximity (HTP) and moderate traffic proximity (MTP). Based on the entropy, standard deviation, and expert judgment, the following weights were assigned: CFD = 0.42, ESD = 0.31, TP = 0.27. For the sub-criteria, HTP = 0.7 and MTP = 0.3. These weights were input to the TOPSIS linear additive weighting model to generate an air pollution index value standardized from 0 to 1 for every census tract, representing the aggregated relative risks.

3.2.3. Bivariate Mapping

Bivariate choropleth map layouts were developed in ArcGIS Pro to symbolize each census tract based on values above or below the statewide median for the socioeconomic (income) and environmental hazard (air pollution index) variables. Four groups were mapped: low poverty–low pollution, high poverty–low pollution, low poverty–high pollution, and high poverty–high pollution. This technique facilitates spatial pattern visualization and the identification of potential hotspots with both social and environmental disparities.

3.3. Detailed Methodology and Objectivity Measures

To ensure the objectivity and reproducibility of our research, we provide the detailed steps of our methodological approach, the rationale for method selection, and the measures taken to minimize subjective influences.

3.3.1. Multi-Criteria Decision Analysis (MCDA)

• Selection Rationale. MCDA was chosen for its ability to integrate multiple environmental and socioeconomic factors into a single index, allowing for a comprehensive assessment of the neighborhood conditions. It is particularly suited to complex urban systems where various factors interact [53].
• Application Steps. a. Criteria Selection: We identified key criteria based on a literature review and data availability, including the air pollution levels, poverty rates, and proximity to pollution sources. b. Data Normalization: All criteria were normalized to a 0–1 scale to ensure comparability. c. Weighting: We used the analytical hierarchy process (AHP) to determine criteria weights, involving a panel of five experts in urban planning and environmental science to minimize individual bias. d. Aggregation: The weighted sum method was applied to aggregate the normalized criteria scores into a final index.
• Minimizing Subjectivity. To reduce potential bias in criteria weighting, we implemented the following. a. We employed the Delphi method with our expert panel to reach a consensus on the weights. b. We conducted a sensitivity analysis to assess the impact of weight variations on the final results. c. We compared our weights with those used in similar studies to ensure alignment with established practices.

3.3.2. Local Spatial Statistics

• Selection Rationale. Local spatial statistics, particularly the Getis–Ord Gi*, were chosen for their ability to identify statistically significant spatial clusters of high and low values, crucial in pinpointing environmental justice hotspots [54].
• Application Steps. a. Spatial Weight Matrix: We defined neighborhood relationships using a fixed distance band of 1 km, based on the literature suggesting this as an appropriate scale for urban environmental studies. b. Calculation of Gi* Statistic: We computed the Gi* statistic for each census tract using ArcGIS Pro’s optimized hotspot analysis tool. c. Statistical Significance: We used a 95% confidence level (p < 0.05) to identify significant hotspots and cold spots.
• Minimizing Subjectivity. To ensure objectivity in the spatial analysis, we implemented the following. a. We tested multiple distance thresholds (0.5 km, 1 km, 2 km) and selected the one that produced the most stable results. b. We applied false discovery rate (FDR) correction to account for multiple testing and spatial dependency. c. We validated the results using alternative local spatial statistics methods (e.g., Local Moran’s I) for consistency.

By following these detailed steps and implementing measures to minimize subjective influences, we aimed to ensure the highest possible level of objectivity in our analysis.

3.4. Controlling for External Variables

To enhance the robustness of our analysis, we considered several external variables that could potentially impact air pollution and the socioeconomic conditions.

• Industrial Activities: Using data from the U.S. Bureau of Labor Statistics, we tracked changes in the industrial composition and output in each metropolitan area. We created an industrial activity index and included it as a control variable in our analyses.
• Population Dynamics: We accounted for population growth and demographic shifts using annual estimates from the U.S. Census Bureau, incorporating these as control variables in our models.

By incorporating these external variables into our statistical models, we aimed to isolate the relationships between our primary variables of interest while controlling for potential confounding factors

3.5. Local Spatial Statistics

Advanced spatial statistics were utilized to quantify and visually analyze the relationships between the neighborhood socioeconomic conditions and environmental hazards. Specifically, the Local Bivariate Relationships tool in ArcGIS Pro was applied to evaluate the statistically significant correlations between the poverty rates and the air pollution index values at the census tract level. Each tract was computationally classified based on the type (positive, negative, none) and strength of the bivariate relationship. The spatial clusters of tracts with similar relationship types facilitated the identification of areas with a positive association between higher pollution and lower incomes, warranting further investigation.

3.6. Computational Analysis

The local bivariate analysis computed Moran’s I statistics to assess the significance of the spatial associations between socioeconomic disadvantage and environmental hazards. Moran’s I values range from −1 to +1, with positive values approaching +1 indicating the clustering of similar values (either high or low). Statistical significance is evaluated based on a randomization approach. In addition, false discovery rate correction accounts for multiple testing effects. Entropy values closer to 1 represent stronger clustered associations between two variables. Together, the localized Moran’s I and entropy metrics provided objective, robust techniques to detect and measure the strength of the linkages between the neighborhood income levels and the exposure risks at the small-area intra-urban scale.

The exact locations of leading tourist attractions compiled from city repositories informed an overlay mapping of the pollution levels in the visited hotspots. Multi-criteria indices aggregated the air quality indicators, poverty rates, and cumulative proximity burdens into an overall environmental justice score reflecting the relative neighborhood-level burden. A cluster analysis delineated the boundaries of the “most overburdened” communities using emerging spatially constrained techniques that overcome the limits of traditional municipal/county-level comparisons.

4. Results

4.1. Preliminary Analysis

The initial visualization of the chemical facilities density, emissions sources, and traffic proximity revealed noticeable geographic clustering patterns, with elevated values concentrated mainly in the core urban census tracts of major metropolitan regions, including Houston, Dallas, San Antonio, and Austin (Figure 3). The bivariate choropleth maps exhibited substantial overlaps between clusters of high air pollution values and clusters of low-income neighborhoods, as evidenced by darker-shaded tracts, symbolizing both socioeconomic disadvantage and concentrated environmental hazards.

These preliminarily observed spatial relationships point to potential pockets of disproportionately high exposure to poor air quality associated with economic disadvantage concentrated in the downtown areas of major cities. The visual correlations warrant more robust statistical analysis using computational geospatial techniques to quantify the strength, significance, and precise nature of the bivariate associations between the poverty rates and pollution levels at the neighborhood scale. Together, these preliminary spatial patterns point to potential pockets of concentrated exposure risks and socioeconomic disadvantage located predominantly in the peripheral census tracts of major cities rather than tourist-frequented downtown centers. Further statistical testing quantifies these relationships.

Contrary to our initial expectations, Figure 3 reveals that the urban centers in our study areas, particularly in Austin and San Antonio, show higher levels of pollution and poverty compared to the peripheries. The peripheral areas, especially in these two cities, are predominantly categorized as having low pollution and low poverty. This pattern suggests a more complex relationship between the urban structure, environmental quality, and socioeconomic status than initially hypothesized, highlighting the need for the nuanced interpretation of urban environmental justice issues.

At the next step, because this research aimed to focus on tourism’s impact and relationships with the other two factors (urban poverty and pollution burden at the tract level), major tourism attraction sites were collected and applied onto the previous geospatial layers. To do this, we used the 12 most related land use (activity) types, namely attractions, campsites, cinemas, fast food, hotels, motels, guest houses, museums, restaurants, theaters, parks, and zoos. After this, the point density method was used to calculate hotspots in the distributions of these attractions in all four case studies. Figure 4 displays the tourism attraction density in a relatively broad dimension range from minimum to maximum.

4.2. Local Bivariate Relationship Analysis

The advanced computational techniques available in the ArcGIS Pro 3.3.1 software were leveraged to rigorously determine the statistical relationships between the poverty rates and air pollution levels locally within each census tract’s neighborhood (Figure 5). The Local Bivariate Relationships tool in the latest version of ArcGIS Pro analyzes two attributes to identify statistically significant relationships using local entropy calculations. The tool categorizes each feature into one of six relationship types based on how reliably the first attribute predicts the values of the second attribute. The categories include not significant, meaning no statistically significant relationship; positive linear, where the second attribute increases linearly as the first attribute increases; negative linear, where the second attribute decreases linearly as the first attribute increases; concave, where the second attribute changes along a concave curve; convex, where the second attribute changes along a convex curve; and undefined complex, where a significant relationship exists but does not fit any other category. This tool helps to discern local patterns between pairs of attributes in a dataset. The analysis utilized the Local Bivariate Relationships tool to assess the statistical relationships between poverty and air pollution locally within the study area. It was run with a neighborhood size of 50 features, meaning that each feature was analyzed regarding its 50 nearest neighboring features. In total, 999 permutations were used to determine the reliability of the relationships. A 95% confidence level indicates that there is a 95% likelihood that the identified relationships are not due to random chance. False discovery rate correction adjusts the confidence level to account for multiple tests being run, to avoid too many false discoveries. Together, these parameters allowed for a rigorous localized analysis of the relationships between poverty and air pollution within the defined spatial neighborhoods. Figure 4 depicts the six different types of bivariate relationships.

The optimized analysis settings uncovered significant relationships in 34.17% of the tracts after properly adjusting for multiple testing effects, indicating non-random associations between social disadvantage and environmental hazards.

Convex and concave relationship patterns were most frequent, occurring in 8.10% and 6.35% of the tracts, respectively. Concave correlations signify disproportionately high pollution in lower-income areas. A moderate to strong average correlation strength (mean entropy 0.5004, max 1.2582) shows sizable effects in select locales. The mapping relationships uncovered spatial hotspots where poverty and pollution converge, mainly in tracts across Houston, Dallas, San Antonio, and El Paso. The statistically significant and clustered correlations provide robust evidence of potential environmental inequity pockets, requiring mitigation to alleviate the localized burden disparities.

With these settings, the tool found statistically significant relationships between the air pollution index values and poverty rates in 34.17% of the census tracts statewide after adjusting for multiple tests. This indicates non-random spatial associations between the socioeconomic and environmental variables.

In terms of the types of bivariate relationships, convex patterns were most frequent, representing 8.10% of the tracts. Concave correlations were also common, occurring in 6.35% of the tracts. Concave relationships suggest disproportionately high pollution burdens in lower-income neighborhoods.

The entropy results quantify the strength of the spatial clustering relationships. With a mean of 0.5004 and a maximum of 1.2582, the associations are moderately strong on average but highly pronounced in select locales.

The mapping of the significant local relationships showed geographic hotspots where poverty most closely corresponded to environmental risks, including tracts in Houston, Dallas, San Antonio, and El Paso, exhibiting strong concave or convex patterns.

The statistically significant and spatially clustered bivariate correlations provide evidence of potential pockets of environmental inequity, requiring further investigation and targeted mitigation policies to alleviate the pollution burdens.

4.3. Cumulative Impact Analysis

An integrated environmental justice score was computed for each tract, combining the air quality, poverty, and proximity burden indicators to determine relative cumulative exposure rankings (Figure 6). The cluster analysis then delineated the boundaries of highly overburdened neighborhoods. The overlay with the attraction locations showed differing visitor versus resident risks.

For instance, the percentage of overlapped areas with significant poverty–pollution relationships is considerably lower in tracts containing attractions than surrounding zones in three metro areas (

Table 1. Comparative calculations resulting from overlapping poverty, air pollution, and attraction density simultaneously.

Case Study	Percentage of Overlapped Areas for Which There Is a Direct Relationship between Poverty and Air Pollution in Each Metroplex	Average Point Density of Tourism Attractions in Overlapped Area	Average Point Density of Tourism Attractions in Whole Metroplex
Dallas Fort Worth	13.64%	17.61	42.11
San Antonio	34.98%	12.46	36.53
Houston	19.39%	32.33	49.6
Austin	9.22%	21.93	64.35

). The average attraction density metrics also exhibit substantial spatial imbalances, with inside–outside differentials of 2–3× across multiple regions. Together, these granular indicators offer standardized techniques to quantify and compare the localized health risks borne by tourists and nearby communities.

The results in Table 1 provide further insights into the research questions around the linkages between tourism activity and inequities. Despite having the highest mean attraction density, Austin exhibits the lowest degree of spatial mismatch between pollution–poverty clusters and amenity zones. With overlapped percentages under 10% and an inside–outside density differential barely exceeding parity, the risks appear relatively balanced. In contrast, San Antonio has a moderately low attraction density but the highest poverty–pollution overlaps (35%) and the widest inside–outside splits. This indicates that urban planning and tourism expansion in San Antonio may disproportionately concentrate the cumulative burdens compared to its neighboring regions. Houston represents a middle case, with moderate density imbalances and overlap extents. The findings showcase this methodology’s value for comparative benchmarking to diagnose the fine-grained dynamics shaping localized outcomes.

Figure 6 shows the areas where there is a direct relationship between poverty and air pollution and simultaneously depicts the density of tourist attractions. The quantitative results from Figure 6 are presented in Table 1.

Our analysis reveals a complex relationship between the tourism attraction density and environmental–socioeconomic disparities. Table 1 shows the percentage of areas where there is a direct relationship between poverty and air pollution, alongside the tourism attraction densities. Notably, San Antonio exhibits the highest percentage (34.98%) of overlapped areas with a direct poverty–pollution relationship, yet it has the lowest average tourism attraction density (12.46) in these overlapped areas. Conversely, Austin shows the lowest percentage of overlapped areas (9.22%) but has the highest overall tourism attraction density (64.35) across the metroplex. This suggests that while tourism may bring economic benefits to certain areas, its relationship with the environmental conditions and poverty is not uniform across cities.

In Houston, we see a different pattern, with a moderate percentage of overlapped areas (19.39%) but the highest tourism attraction density (32.33) within these areas. This could indicate that, in Houston, tourist attractions are more integrated into areas facing environmental and socioeconomic challenges. Dallas Fort Worth presents yet another scenario, with a relatively low percentage of overlapped areas (13.64%) and a significant difference between the tourism attraction density in overlapped areas (17.61) and that of the whole metroplex (42.11).

These findings indicate a potential disconnect between tourism development and the equitable distribution of its benefits and burdens across different urban contexts. The varying patterns across these Texas metropolitan areas underscore the complexity of the tourism–environment–equity nexus and the need for city-specific approaches in addressing these challenges.

4.4. Spatial Inequality Metrics

In addition to the observed patterns, the Moran’s I statistics derived from the computational analyses provide objective measures to quantify the statistical significance of the identified hotspots based on spatial autocorrelation principles. Positive Moran’s I values approaching +1 indicate the clustering of similar high or low attributes. Statistical p-values assess the cluster likelihood compared to random chance. False discovery rate correction accounts for multiple tests. Combined with entropy, gauging the cluster strength, these metrics enable standardized comparisons of the relationship types and intensities across metro areas. Leaders can apply the inequality estimates to prioritize mitigation resources towards communities with the greatest cumulative exposure and compounded vulnerability risks.

For example, Houston contained census tracts with the highest maximum Moran’s I (1.2582) and entropy (0.7231) values among the metro areas, signaling intense pockets of concentrated disadvantage unlikely to occur by chance alone (p < 0.05). Dallas, meanwhile, exhibited lower peak intensity statistics (Moran’s I = 1.0072; entropy = 0.3751), indicating more moderate spatial inequalities on average. The relative magnitude of the metrics can guide area-specific policy responses calibrated to the localized conditions. Automated cluster mapping further displays the boundaries where elevated cumulative burdens spatially converge, benefiting decision-makers seeking to geographically target overburdened neighborhoods for assistance programs.

Together, the multi-faceted results highlight the value of techniques used to uncover the variability in tourism’s links with environmental disparities. The integrated exposure index determining cumulative burden rankings, combined with the spatial mismatch estimates of the attraction density differentials, provides standardized indicators quantifying the nexus. Statistical tests reveal significant hotspots unlikely to arise randomly, while Moran’s I and entropy enable the benchmarking of the cluster strength across regions. Comparing these metrics serves to guide the diagnosis of the fine-grained dynamics shaping localized planning and tourism outcomes. The replicable bottom-up approach, matching the measurement scale to the neighborhood context directly influencing residents, promises a more nuanced insight into these inequities.

5. Discussion

5.1. Overview of Key Findings and Spatial Patterns

This study adopts a neighborhood-centric approach to characterize environmental health disparities regarding air pollution, poverty, and cumulative burdens in four major Texas metropolitan areas. Our granular spatial analysis elucidates localized inequalities that are often obscured within city-level averages. The integration of tourism attraction density maps reveals the variability in the presence of amenities within stressed communities, providing insights into the complex relationships between environmental justice, socioeconomic status, and tourism development.

Our findings reveal a significant spatial mismatch, whereby overburdened communities facing poverty, pollution, and cumulative burdens are predominantly positioned on the urban periphery, distant from the economically vibrant downtown regions that typically draw tourist activity [31]. This spatial arrangement has profound implications for distributive justice and equitable access to urban amenities [32].

The localized methods employed in this study uncover pockets where socioeconomic disadvantage and pollution closely intertwine, disproportionately affecting marginalized communities [4,11]. Spatial inequality statistics confirm that these clustered burdens are unlikely to occur randomly, signaling potential planning inequities that require attention [30].

Our analysis reveals that the percentage of overlapped areas with significant poverty–pollution relationships is considerably lower in tracts containing attractions than surrounding zones in three metro areas. The average attraction density metrics also exhibit substantial spatial imbalances, with inside–outside differentials of 2–3× across multiple regions. These granular indicators offer standardized techniques to quantify and compare the localized health risks borne by tourists and nearby communities.

5.2. Comparative Analysis of Metropolitan Areas

Our comparative findings showcase how tourism permeates the equity dynamics variably across regions [55]. Notably, Austin exhibits the lowest spatial mismatch despite having the highest overall attraction density. With less than 10% overlap between the pollution–poverty relationships and tourism zones, and near parity in the amenity availability between stressed and non-stressed neighborhoods, the risks appear relatively balanced in Austin [31].

In contrast, San Antonio shows the widest density differential inside versus outside stressed neighborhoods and the highest overlap percentage. This indicates that tourism planning and expansion in San Antonio may inadvertently concentrate the burdens in already disadvantaged areas [56]. There is a moderate density imbalance and overlap extent in Houston. Our analysis reveals distinct patterns in Dallas and Houston that warrant further examination. In Dallas, we observed moderate spatial inequality, with Moran’s I at 1.0072 and an entropy value of 0.3751. This suggests a more balanced distribution of environmental burdens and amenities compared to San Antonio, although it remains less equitable than Austin. The sprawling nature of Dallas contributes to this pattern, with areas of high pollution and poverty dispersed throughout the metroplex, rather than concentrated in specific locations. Conversely, Houston exhibited the highest maximum Moran’s I (1.2582) and entropy (0.7231) values among the metro areas, indicating significant clusters of concentrated disadvantage. This could be attributed to Houston’s lack of zoning laws, leading to a more chaotic development pattern. Additionally, the city’s strong ties to the petrochemical industry may create localized areas of high pollution. Both cities display a moderate degree of spatial mismatch between tourist attractions and areas of high pollution and poverty. However, Houston shows a higher percentage of overlap (19.39%) compared to Dallas (13.64%). These findings suggest that while both cities face challenges in equitable urban development, Houston may require more targeted interventions to address its pronounced environmental justice issues.

These inter-city differences warrant further investigation into the specific policies, historical development patterns, and economic factors that have shaped the current distribution of environmental quality, amenities, and socioeconomic status across neighborhoods.

Austin’s relatively balanced urban landscape may be attributed to several factors.

• Progressive Urban Planning: The Imagine Austin Comprehensive Plan, adopted in 2012, emphasized creating a compact and connected city, potentially contributing to the more equitable distribution of amenities and reduced environmental disparities [57].
• Tech Industry Influence: Austin’s booming technology sector has attracted a diverse, highly educated workforce, potentially leading to more distributed wealth and a demand for amenities across the city [58].
• Environmental Initiatives: Austin’s Green Building Program and Urban Forest Plan may have contributed to the more uniform environmental quality across neighborhoods (City of Austin, 2024).
• Dispersed Tourism Development: Unlike cities where tourism is concentrated in specific districts, Austin’s approach appears more dispersed, potentially leading to a more balanced distribution of tourism-related benefits and burdens [59].
• Socioeconomic Factors: Austin’s relatively young, educated population and strong job market may contribute to the lower overall poverty rates and more balanced economic opportunities across the city [59,60,61].

However, it is important to note that while Austin shows the most balanced picture among the studied cities, it is not free from inequalities. Recent studies have highlighted ongoing gentrification processes in East Austin, which may be displacing long-term residents and potentially shifting the patterns of inequality [60].

5.3. Alignment with Previous Studies and Unique Contributions

Our findings align with and extend previous studies on environmental justice and urban tourism in the USA and Europe. Similar to Humphrey et al. [62] in New York City, we found that lower-income neighborhoods often bear a disproportionate burden of environmental hazards. However, our research goes further by explicitly linking these disparities to tourism patterns, an often-overlooked aspect in the environmental justice literature.

In Europe, Scherber et al. [63] observed comparable spatial mismatches between tourist hotspots and areas of environmental stress in Berlin, echoing our findings in Texas cities. Our study contributes to this body of knowledge by providing a granular, neighborhood-level analysis of the tourism–environment–equity nexus in the context of rapidly growing Texan metropolitan areas.

The value of our research lies in its potential to inform evidence-based policymaking and urban planning. By providing a nuanced understanding of local dynamics, our study can lead to more targeted interventions, such as

• Developing green corridors to connect overburdened neighborhoods with tourist attractions;
• Implementing localized air quality improvement measures in areas with a high pollution burden;
• Crafting tourism development strategies that prioritize environmental justice and equitable access to urban amenities;
• Implementing policies to mitigate the potential negative impacts of tourism on vulnerable communities.

These strategies could help to balance the access and exposure disparities amplified by uneven tourism development patterns.

5.4. Methodological Contributions and Limitations

Our study demonstrates the value of integrating multiple data sources and advanced statistical techniques for a comprehensive analysis of environmental justice issues. The use of Local Bivariate Relationships analysis and a cumulative impact assessment provides a robust framework for the identification and quantification of environmental inequities at the neighborhood level.

The originality of the research lies in its integration of multiple data sources, advanced statistical techniques, and a neighborhood-centric analysis to unveil the intricate connections between environmental injustice, poverty, and tourism. While this study primarily builds upon established methodologies in the field of environmental justice, such as spatial analysis, composite indices, and vulnerability mapping, it aims to contribute a novel perspective by integrating these approaches within the context of tourism development in Texas urban centers.

By explicitly examining the spatial relationships between environmental burdens, socioeconomic factors, and the density of tourist attractions, this research extends the discourse on environmental justice to encompass the potential impacts of urban tourism on marginalized communities. This integration of tourism dynamics into the analysis of environmental disparities offers a unique lens through which to explore the complex interplay between economic development, environmental quality, and social equity.

However, we acknowledge several limitations of our study.

• Static Data: Our analysis relies on static data and does not capture the dynamic nature of tourist flows or temporal variations in environmental quality.
• Focus on Air Pollution: While air pollution is a critical environmental indicator, our study does not account for other forms of environmental degradation that may be relevant to environmental justice.
• Limited Qualitative Insights: Our quantitative approach, while providing valuable spatial insights, does not capture the lived experiences of residents in the affected communities.

5.5. Future Research Directions

To address these limitations and further advance the understanding in this field, we propose several directions for future research.

• Incorporate time-series data to capture the temporal dynamics of tourism flows and environmental quality. This could include mobile monitoring to capture pulsed amplification from periodic tourist influxes and events.
• Expand environmental indicators beyond air pollution to include factors such as noise pollution, water quality, and access to green spaces.
• Integrate qualitative methods to capture the lived experiences and perceptions of residents in the affected communities. This could include perception research to provide the experiential validation of the climate exposure and barriers shaping tourism participation gaps [64,65,66].
• Extend the analysis to other urban contexts, both within the United States and internationally, to test the generalizability of our findings.
• Investigate the impact of specific urban policies and planning decisions on the observed patterns of environmental justice and tourism development.
• Explore the potential of innovative technologies and smart city initiatives in mitigating environmental injustices and promoting sustainable tourism.
• Examine the role of community engagement and participatory planning in addressing environmental justice issues in the context of urban tourism development.
• Incorporate additional variables such as employment in the hospitality sector among residents of peripheral areas, unemployment rates, and other socioeconomic indicators to provide a more holistic view of the ways in which tourism impacts different urban neighborhoods.

In conclusion, this study provides valuable insights into the complex relationships between environmental justice, socioeconomic conditions, and tourism development in major Texas metropolitan areas. By adopting a neighborhood-centric approach and integrating multiple data sources, we have uncovered significant spatial mismatches and inequities that demand attention from policymakers and urban planners.

Our findings highlight the need for targeted, context-specific interventions to address environmental injustices and promote more equitable urban development. As cities continue to grow and evolve, it is crucial to ensure that the benefits of tourism and urban development are distributed fairly and that the environmental burdens are not disproportionately borne by vulnerable communities.

Future research building on this work will be essential in developing a more comprehensive understanding of these issues and informing effective, equitable urban policies in Texas and beyond. By continuing to refine our methodologies and expand our scope of inquiry, we can work towards creating more sustainable, just, and livable urban environments for all residents.

6. Conclusions

This comprehensive study illuminates the critical nexus between tourism development, environmental pollution, and socioeconomic disparity across major Texas urban centers. Through meticulous spatial analysis and the innovative integration of multiple data sources, our research reveals the disproportionate burden of environmental hazards borne by lower-income neighborhoods while highlighting nuanced differences in amenity access and pollution exposure among cities like San Antonio and Austin.

Our neighborhood-level analysis uncovers a significant spatial mismatch between areas of high tourist activity and neighborhoods facing the greatest environmental and socioeconomic burdens. The relationship between the tourism density, air pollution, and poverty varies considerably across different urban areas. Austin, for instance, demonstrates a more balanced distribution, while San Antonio exhibits more pronounced disparities. These findings underscore the importance of granular, neighborhood-centric approaches in capturing localized patterns of environmental injustice that broader county-level analyses often obscure.

By explicitly examining the spatial relationships between environmental burdens, socioeconomic factors, and the tourist attraction density, this study contributes a novel perspective to the environmental justice discourse. The integration of tourism dynamics offers a unique lens to explore the complex interplay between economic development, environmental quality, and social equity, extending our understanding of how urban tourism impacts marginalized communities.

Our findings highlight the urgent need for holistic and targeted policy interventions focused on mitigating the adverse effects of urban tourism and expansion on vulnerable populations. Based on our analysis, we propose several policy recommendations.

• Targeted Green Infrastructure: Implement green buffer zones between high-pollution areas and residential neighborhoods, potentially reducing particulate matter by 15–20% in adjacent areas. In San Antonio, prioritize creating green corridors connecting peripheral neighborhoods to tourist attractions.
• Low-Emission Zones (LEZs): Establish phased LEZs in high-pollution areas, which could reduce the NO2 levels by 10–15% and PM10 by 5–10%. In Houston, prioritize LEZs where industrial zones intersect with residential neighborhoods.
• Community-Based Air Quality Monitoring: Install low-cost air quality sensors in high-risk neighborhoods, potentially reducing personal exposure by up to 30%. In Dallas, implement a city-wide network of community-operated sensors to identify localized pollution hotspots.
• Green Job Training: Establish programs in high-poverty areas focused on green economy skills, potentially increasing the local employment rates by 5–10% over three years. In Austin, focus on innovative clean tech sectors.
• Targeted Public Transportation: Improve public transit in high-poverty, high-pollution areas, potentially reducing personal vehicle use by 15–20%. In San Antonio, develop dedicated bus rapid transit lines connecting peripheral neighborhoods to downtown tourist areas.
• Tourist Eco-Tax: Implement a tax in high-visitor-density areas, with revenues funding environmental projects in nearby high-pollution, high-poverty neighborhoods.
• Environmental Justice Scoring System: Develop a comprehensive scoring system for new urban development projects, incorporating air quality, socioeconomic factors, and proximity to tourist attractions.
• Neighborhood-Level Task Forces: Create environmental justice task forces composed of local residents, business owners, and city officials to develop and implement localized strategies.

These interventions should be implemented collaboratively with local communities, with regular monitoring and evaluation to assess their effectiveness and make necessary adjustments. While the expected effects are based on similar interventions elsewhere, the outcomes may vary and should be carefully measured in the local context.

By incorporating these neighborhood-centric approaches into urban planning and policy formulation, cities can foster more equitable distributions of environmental burdens and benefits. This study calls for a reimagining of urban development strategies that prioritize environmental justice and equal access to urban amenities, ensuring that the pursuit of sustainable and inclusive tourism growth does not come at the expense of the most vulnerable populations.

In conclusion, this research provides a foundation for the understanding and resolution of the complex relationships between tourism, environmental quality, and social equity in urban settings. As cities continue to grow and evolve, it is crucial to ensure that the benefits of tourism and urban development are distributed fairly and that the environmental burdens are not disproportionately borne by vulnerable communities. Future research building on this work will be essential for the development of more comprehensive, equitable, and sustainable urban policies in Texas and beyond.