FALL 2026
Quantifying Environmental Inequities in Texas Drinking Water: A Time-Series Analysis of Contaminant Exposure and Demographic Trends
Milagro Salinas
Lone Star College-Montgomery
ABSTRACT
To evaluate environmental inequities in drinking water quality across demographic groups, this study conducts a time-series analysis of trends and variability in concentrations of three regulated contaminants across seven Texas counties from 2015 to 2025. While prior studies on drinking water contamination in Texas have centered on regulatory exceedances, this study examines how long-term trends in drinking water quality differ across demographically diverse Texas counties. County-level community drinking water system data for HAA5, TTHM, and lead were obtained from the Texas Commission on Environmental Quality (TCEQ), along with demographic data from the U.S. Census Bureau. This investigation applies a one-way ANOVA to test differences in mean contaminant concentrations across counties and computes county-specific regression slopes to evaluate improvement rates over time. Then, concentration variability within each county is analyzed through distributional measures and formally tested with Levene’s test to conduct a comparative analysis of stability over time. Statistical results are compared against county-level demographic variables, including Hispanic population percentage, population size, and median household income, to explore patterns in water quality inequity. Results indicate that mean concentrations differ significantly across counties, showing higher averages in counties with higher Hispanic percentage and smaller total population. Trend and variance analyses reveal unequal progress and stability, as counties with higher Hispanic populations and lower-income showed slower improvement and greater variability in water quality compared to larger, higher‑income counties. These findings suggest that incorporating time-series trends reveals demographic patterns in drinking water quality that are not captured by traditional metrics based on compliance. Future research should include additional counties to capture greater demographic diversity.
INTRODUCTION
According to state and federal monitoring databases, Texas has among the highest numbers of violations of the Safe Drinking Water Act (SDWA) in the United States. (TCEQ, 2024). National analyses of Environmental Protection Agency (EPA) data further indicate that more than 20% of Texas residents were served by water systems with at least one SDWA violation over the most recent five-year period, emphasizing the extent of exposure risks (Texas Commission on Environmental Quality, 2024). Furthermore, drinking water safety in Texas reflects broader patterns of environmental inequity, with rural and Hispanic communities often facing higher contaminant exposure and weaker regulatory compliance (McDonald, 2018). However, while there has been increased attention paid to drinking water contamination in Texas, these studies have centered on urban regions, leaving rural and border counties underexamined. Additionally, although TCEQ maintains extensive monitoring databases, these datasets are rarely incorporated in equity analyses. Prior analyses also tend to rely on regulatory exceedances and mean contaminant concentrations, overlooking year-to-year fluctuations and how water quality actually changes over time. These limitations constrain a full understanding of how contamination patterns vary across regions and demographic groups.
​
Significance of Study
​
As Texas community water systems provide drinking water for more than 30 million residents, contamination represents a critical public-health concern (Texas Commission on Environmental Quality, 2024). Moreover, many rural and predominantly Hispanic counties face chronic underinvestment in water infrastructure, leading to aging distribution systems and slower remediation of contamination issues. Therefore, understanding where water quality improvements are slower or more unstable can help direct state and federal resources toward the communities most in need of infrastructure support.
​
Purpose and Scope of Study
​
Therefore, with an aim to evaluate environmental inequities in drinking water quality across demographic groups, this study conducts a time-series analysis of trends and variability in contaminant concentrations across seven Texas counties from 2015 to 2025. The analysis focuses on three regulated contaminants: haloacetic acids (HAA5), trihalomethanes (TTHM), and lead. Specifically, this research performs a cross-county comparative analysis in terms of mean contaminant concentrations, the rate of improvement over time, and the stability of contaminant levels. By examining contaminant patterns across multiple years instead of only focusing on violations or annual averages, this study provides a more detailed examination of how water quality changes over time, offering a meaningful indicator of whether communities are receiving equitable progress in drinking water safety. This study contributes a time-series framework that integrates trends and variability into water quality assessment, enabling a more comprehensive evaluation of inequities in exposure, improvement, and stability across communities.
LITERATURE REVIEW
Demographic Inequities in Drinking Water Quality​
Research consistently shows that drinking water quality in the United States exhibits demographic and structural disparities. Several area-specific studies have found that minority and lower-income communities are disproportionately affected by drinking water violations (McDonald, 2018). For example, previous scholarship indicates that historically, populations most at risk for lead exposure in the United States are families living at or below the federal poverty level and ethnic minority groups, particularly because poverty is associated with substandard housing, aging plumbing, and inadequate corrosion control (Fawkes, 2021). Furthermore, a national 2018 study found that small systems serving fewer people are more likely to have persistent or repeated violations of water quality, suggesting that failures originate from limited technical and financial capacity instead of being occasional incidents (Allaire, 2018). The study indicated that the counties with the highest prevalence of violations are rural regions, primarily located in Texas, Oklahoma, and Idaho (Allaire, 2018). These patterns are also reflected in public opinions within Texas. A recent statewide survey of 650 households found that Black and Hispanic Texans, particularly in low-income communities, were significantly more likely than white respondents to distrust the safety of their tap water or avoid drinking it altogether (Lozano Carver & Salhotra, 2023). Collectively, these studies demonstrate how water quality problems may be influenced by structural disadvantages, including poverty, minority status, and small system capacity, emphasizing the need to examine how these issues occur across different Texas communities.
​
Key Drinking Water Contaminants and Associated Health Implications
​
Previous studies on drinking water contamination in Texas have identified trihalomethanes (TTHM), haloacetic acids (HAAs), and lead as three of the most persistent and health-relevant pollutants in public water systems (Texas Commission on Environmental Quality, n.d.; Texas Water Resources Institute, 2020). Disinfection byproducts (DBPs) are organic chemicals that form during drinking water treatment and distribution when naturally occurring organic matter reacts with disinfectants, such as chlorine, chloramine, and chlorine dioxide, used to kill pathogenic organisms (Texas Commission on Environmental Quality, n.d.). Since several of these chemicals pose documented health risks, the Environmental Protection Agency (EPA) regulates them under the Safe Drinking Water Act, establishing Maximum Contaminant Levels (MCLs) of 80 µg/L for TTHM and 60 µg/L for HAA5.
​
Epidemiological studies indicate that TTHM exposure is associated with an increased risk of urinary-bladder cancer in populations served by chlorinated water, as well as potential kidney, neurological, and reproductive effects (Weisman et al., 2022). Similarly, HAAs, particularly HAA5, are considered mutagenic, cytotoxic, genotoxic, teratogenic, and carcinogenic (Parvez et al., 2019). National risk assessments estimate that lifetime exposure to disinfection byproducts may result in approximately three cancer cases per thousand people (Evans et al., 2020). In Texas, exceedances of DBP standards have been recorded across multiple rural and border counties, particularly in regions with high temperatures, elevated organic carbon, and aging distribution systems (Texas Commission on Environmental Quality, n.d.).
​
Lead contamination presents a different mechanism and risk profile. Lead typically enters drinking water through the corrosion of pipes, plumbing materials, and service lines. The EPA has set the maximum contaminant level goal for lead in drinking water at zero because it is a toxic metal that can be harmful to human health even at low exposure levels, enforcing an action level of 15 µg/L under the Lead and Copper Rule (U.S. Environmental Protection Agency, 2024). Researchers indicate that children and adults exposed to lead regularly experience an elevated risk for nervous system damage, cardiovascular disease, various cancers, and, at higher levels, death (Bardot et al., 2025). Additionally, lead is persistent, and it can bioaccumulate in the body over time. In Texas, detections of lead have been documented in several community water systems, especially in smaller and rural municipalities, where aging infrastructure and limited corrosion control contribute to elevated risk.
METHODOLOGY
County-level datasets of community water system samples for analysis are collected from the Texas Commission on Environmental Quality (TCEQ) for annual concentrations of HAA5, TTHM, and lead, and from the U.S. Census Bureau (2020 Census) for Hispanic/Latino population percentage, total population size, and median household income. The selected analysis period is 2015 to 2025 because it captures the most significant demographic shifts in Texas, particularly rapid changes in Hispanic population distribution, making it an accurate representation of current demographic characteristics. The following seven Texas counties are selected to capture variation in contaminant exposure and demographic composition (Table 1).
The sample includes a mix of urban, suburban, and rural areas with varying socioeconomic and demographic characteristics. For example, Hidalgo, El Paso, and Zapata Counties represent predominantly Hispanic and lower-income communities, while Rockwall and Comanche Counties represent higher-income or majority White regions. Dallas and Guadalupe Counties are included to represent more urbanized areas with larger populations. County locations are shown in Figure 1.
All datasets are inspected and standardized to μG/L units before analysis to ensure consistency across counties and contaminants. However, lead concentrations are kept in MG/L units to align with TCEQ’s raw datasets. For visualization purposes, values are converted to μG/L when needed to allow comparison with EPA’s 15 μG/L Lead Action Level.
​
Before conducting the inferential analysis, descriptive statistics (mean, median, standard deviation, minimum, and maximum) are calculated using Microsoft Excel for each contaminant (HAA5, TTHM, and lead) within each of the seven selected Texas counties from 2015 to 2025. Additionally, the mean concentrations for each contaminant and county are visualized in a bar chart for comparison. To assess differences in mean contaminant levels across counties, a one-way Analysis of Variance (ANOVA) is conducted using raw concentration data with a significance threshold of α = 0.05. This test evaluates whether the average contaminant concentrations of at least one county vary significantly from those of other counties. Then, a time-series analysis is performed in Microsoft Excel to quantify long-term changes during the study period. For each contaminant and county, annual sample concentrations were regressed on time (year) in Excel to estimate slope coefficients (µg/L per year). The resulting 21 slope values are used as indicators of improvement or deterioration in water quality, which are interpreted in relation to demographic variables (Hispanic %, population, and median income) to contextualize observed patterns. Lastly, to examine the stability of contaminant concentrations over time, variability is assessed for each contaminant and county using boxplots summarizing variability metrics, including distributional width and interquartile range. To statistically test whether variability differed significantly across counties, Levene’s test is applied to the raw concentration data for each contaminant. This test evaluates whether the variances are equal by analyzing the absolute deviations from each county’s median. The results are used for a cross-county comparative analysis of contaminant stability and to interpret these differences in relation to demographic characteristics.
RESULTS
Results for descriptive statistics for each contaminant and county are shown in Tables 2 to 8 in the appendix.
The computed contaminant-specific trendlines for each contaminant provided the following 21 slopes:
​Boxplots illustrating county-level variability for each contaminant are shown on Figures 12 through 15.
Distribution of Lead concentrations is shown again in Figure 15, excluding Comanche to better illustrate the variability in the rest of the counties.
DISCUSSION
Cross-County Comparison of Mean Concentrations
​
The comparison of mean concentrations of HAA5, TTHM, and lead across the seven selected Texas counties, shown in Figure 2, shows clear cross-county differences in contaminant levels. Although all mean values remain below federal regulatory thresholds, counties with smaller populations (Comanche and Guadalupe) and higher Hispanic population percentages (El Paso) tend to show higher mean concentrations than larger, more urban counties. For example, Dallas, with the largest population size, displays the lowest average concentrations for all three contaminants, suggesting better water quality conditions. Furthermore, the one-way ANOVA results shown in Table 9 indicate that the p-values for all three contaminants are substantially below 0.001, confirming that at least one county’s mean levels differ from the others. The results suggest that the differences in mean concentrations between the seven counties are not due to random variation, providing evidence of unequal water quality conditions.
​
Time-Series Trends in Water Quality
​
In Table 10, negative slopes represent declining concentrations over time, reflecting gradual improvement, while positive slopes represent increasing concentrations, indicating deterioration in drinking water quality. Across the seven counties, most slopes are negative, suggesting overall reductions in HAA5, TTHM, and lead throughout the 10-year period. However, the magnitude of these slopes varies significantly, as some counties have steeper negative slopes, such as Guadalupe for HAA5 and TTHM, indicating faster improvement, while others show slopes near zero, showing slower progress. Notably, Comanche has a steeper positive slope for HAA5, indicating increasing concentrations over time. Although these slopes are weak, their differences across counties highlight that communities are not experiencing the same rate of improvement, even though they are regulated under the same federal and state standards. This uneven progress suggests that factors such as the size of the system, available resources, and community demographics may influence how effectively water systems reduce contaminant levels.
​
Cross-County Comparative Analysis of Variability in Contaminant Concentrations
​
The results of variability for each contaminant reveal that water quality stability differs significantly across counties. For HAA5 in Figure 12, distributions are mostly similar across counties, but Zapata and Comanche show slightly wider interquartile ranges, indicating less consistent contaminant levels over time. In contrast, TTHM in Figure 13 shows clearer differences, with Comanche, Guadalupe, and El Paso showing significantly wider distributions. Notably, these counties share demographic characteristics: Comanche and Guadalupe have smaller populations, while El Paso has a higher Hispanic percentage and lower median household income. This indicates that demographic and socioeconomic factors may be linked to reduced stability in water quality. Moreover, lead concentrations in Figure 14 reveal the most pronounced differences, particularly in Comanche County, which was excluded from Figure 15 due to its outlier behavior. Comanche’s significantly wider distribution suggests sudden spikes in contamination or inconsistent remediation efforts. Additionally, Figure 15 shows that Zapata and El Paso continue to demonstrate wider distributions, highlighting persistent instability in counties with higher Hispanic populations, lower incomes, and small populations. The results reinforce the pattern that communities with fewer resources experience greater fluctuations in contaminant concentrations. In contrast, Dallas consistently displays narrower distributions across all three contaminants, representing more stable water quality in urban areas.
​
The results from Levene’s test confirm that the differences in variability are statistically significant, with p-values less than 0.001 for all three contaminants. This means that at least one county experiences significantly different variability in contaminant levels compared to others. This indicates that water quality stability is not uniform across counties, even though they are monitored by the same federal and state regulations. Thus, some communities face more unpredictable spikes than others, making risks harder to anticipate and manage. The evidence suggests that demographic and socioeconomic characteristics may shape not only average contaminant levels but also the predictability of water quality over time, emphasizing inequities in regulatory oversight and water infrastructure investment. These findings demonstrate that evaluating long-term trends and variability provides a more comprehensive framework for identifying environmental inequities than reliance on regulatory compliance metrics alone.
CONCLUSION
​By comparing mean contaminant concentrations, time‑series slopes, and variability across seven counties, the analysis reveals that communities with smaller populations, lower incomes, and higher Hispanic percentages consistently face greater exposure, slower improvement, and less stable water quality. In contrast, larger and wealthier counties have lower contaminant levels and more reliable stability. The results demonstrate that inequities persist despite uniform regulatory oversight, suggesting that structural and resource‑based differences influence how effectively water systems protect public health. These findings underscore the need for policy approaches that address not only regulatory compliance but also disparities in long-term water quality outcomes across communities.
​
Nevertheless, this study is restricted to only seven Texas counties, which limits the generalizability of results to the state as a whole, as they do not represent all communities across Texas. Additionally, only three contaminants are examined. While they are among the most frequently reported and health‑relevant contaminants, excluding others may overlook additional inequities in water quality. Thus, future research should include additional counties and contaminants to strengthen generalizability.
​
​
APPENDIX
REFERENCES
Allaire, M., Wu, H., & Lall, U. (2018). National trends in drinking water quality violations. Proceedings of the National Academy of Sciences of the United States of America, 115(9), 2078–2083. https://doi.org/10.1073/pnas.1719805115
Bardot, T., Austin, W., & El-Khattabi, A. R. (2025). Disparities in Drinking Water Quality Across the United States. PLOS Water. https://journals.plos.org/water/article?id=10.1371%2Fjournal.pwat.0000362#pwat.0000362.ref011
Lozano Carver, J., & Salhotra, P. (2023). Black and Hispanic Say They Don’t Trust the Quality of Their Water. The Texas Tribune. https://www.texastribune.org/2023/02/16/black-hispanic-texas-water-quality-survey/
Evans, S., Campbell, C., & Naidenko, O. V. (2020). Analysis of Cumulative Cancer Risk Associated with Disinfection Byproducts in United States Drinking Water. International journal of environmental research and public health, 17(6), 2149. https://doi.org/10.3390/ijerph17062149
Fawkes, L., & Sansom, G. (2021). Preliminary Study of Lead-Contaminated Drinking Water in Public Parks-An Assessment of Equity and Exposure Risks in Two Texas Communities. International journal of environmental research and public health, 18(12), 6443. https://doi.org/10.3390/ijerph18126443
McDonald, Y. J., & Jones, N. E. (2018). Drinking Water Violations and Environmental Justice in the United States, 2011-2015. American journal of public health, 108(10), 1401–1407. https://doi.org/10.2105/AJPH.2018.304621
Texas Commission on Environmental Quality. (n.d.). Disinfection by-products in drinking water. Texas Commission on Environmental Quality. https://www.tceq.texas.gov/drinkingwater/chemicals/dbp
Texas Commission on Environmental Quality. (2025). Texas Drinking Water Watch (DWW). https://dww2.tceq.texas.gov/DWW/
Texas Commission on Environmental Quality (TCEQ). (2024). State of Texas Public Drinking Water Program 2024 Annual Compliance Report . https://www.tceq.texas.gov/downloads/drinking-water/epa-acr-2023.pdf
Texas Legislative Budget Board. (2019). Staff report: Texas water systems. Legislative Budget Board. https://www.lbb.texas.gov/Documents/Publications/Staff_Report/2019/5464_Water_Systems.pdf
Texas Water Resources Institute. (2020, January 1). Lead in Texas water. Texas Water Resources Institute. https://twri.tamu.edu/blog/2020/01/01/lead-in-texas-water/
U.S. Census Bureau. (n.d.). Census.gov. U.S. Department of Commerce. https://www.census.gov/
U.S. Environmental Protection Agency. (2024). Basic information about lead in drinking water. U.S. Environmental Protection Agency. https://www.epa.gov/ground-water-and-drinking-water/basic-information-about-lead-drinking-water
Weisman, R. J., Heinrich, A., Letkiewicz, F., Messner, M., Studer, K., Wang, L., & Regli, S. (2022). Estimating National Exposures and Potential Bladder Cancer Cases Associated with Chlorination DBPs in U.S. Drinking Water. Environmental health perspectives, 130(8), 87002. https://doi.org/10.1289/EHP9985
CONTINUE READING
L’drina Aldonza
Texas A&M International University
While many only begin to concern themselves for their health in later stages of life or when complications begin to arise, the ideal time to care for one's health is during youth, before any problem begins. This care may take on many forms, but one often overlooked, or more importantly, underemphasized for its grand effects on overall health in the body is maintaining adequate dental health. Touted more often than not to children to avoid cavities or plaque, proper dental health can lessen the likelihood of major health concerns such as health disease and diabetes (Shmerling 2024). Proper dental care can mediate periodontal disease, a risk factor in major health complications...
Pride as the Antithesis to Virtue: More, Spenser, and Milton’s Differing Definitions of Sinful Pride
.jpeg)
Matthew Garner
University
The sixteenth and seventeenth centuries saw England thrown into religious and political turmoil from the tumultuous reign of King Henry VIII to the chaos of the English Civil War. Because of the deep theological roots of much of the day’s discourse and disagreement, this period of English history (later known as the English Renaissance) is the original historical context of countless transformative and thought-provoking literary works that continue to shape the way theologians think about Christianity and virtue. Many of the most influential writers of the English Renaissance were directly involved with religion through their lives and careers, and as such, they became intrinsically connected to the politicized and often controversial facets of their own belief systems...
.jpeg)
Oluchi Ogbonna
State University Affiliation
To understand the role of culture in peer-initiated sexual abuse (PISA) perpetrated by adolescents aged 13-18, this study analyzes PISA risk factors identified in research from 8 countries over the past 15 years. Juveniles accounted for approximately 17% of sexual crimes in the U.S. in 2015 - a trend mirrored by rising rates in other nations - making PISA a significant global public health issue. Yet most research is Western-centric and focuses on victimization rather than perpetration. This study employs a socio-ecological framework to analyze research on PISA from 8 countries: the United States, South Korea, Switzerland, Norway, Sweden, Nigeria, Ethiopia, and Sierra Leone...