https://orcid.org/0000-0001-8849-8517
Figures
Abstract
Background
Indigenous, mountain communities residing upstream of Bhumibol Dam, Thailand, rely on vulnerable natural water sources for their water supply, yet remain unaware of the associated health risks.
Methods
This study assessed the water quality, usage patterns, and contamination pathways across six villages upstream of Bhumibol Dam to shed light on the obstacles to sustainable water security. Samples from 38 water sources of drinking and/or non-drinking water, soil, and the edible parts of crops were subjected to analyses of physical, chemical (NO3-N, pH), and qualitative pesticide-related variables, alongside a 6-month assessment of a community water filter system.
Results
Principal component analysis identified a “at-risk group” of preferred drinking water sources all exhibiting high NO3-N, highly alkaline pH, and substantial pesticide contamination, which was found to likely be caused by agricultural run-off. This was reinforced by the detection of pesticide residues in all soil samples and, critically, in the below-ground edible parts of crops (taro, lemongrass, arrowroot), confirming dietary exposure in the local communities. Further compounding the risks posed by the unsafe water supply, the community water filter was found to be ineffective throughout the 6-month analysis with there being no significant difference in water quality between before and after filtration. The residents’ paradoxical preference for high-risk, still water (from sand-filtered puddles) for drinking, rather than water from flowing sources, which they used only for cooking and cleaning.
Conclusion
Overall, these findings reveal compounded environmental health risks emerging from the convergence of mineral-rich hydrogeology, agrochemical runoff, and vulnerable water-collection behaviours. To mitigate these threats, the study suggests that decentralized water management strategies must prioritize both the implementation of advanced filtration technologies capable of mineral and chemical removal and targeted public health education to shift community reliance away from high-risk, stagnant water sources.
Citation: Trivalairat P, Phiwchai I, Chaichan M, Trivalairat K, Sripo N (2026) Compounded environmental health risks in mountain communities upstream of Bhumibol Dam, Thailand. PLoS One 21(6): e0340557. https://doi.org/10.1371/journal.pone.0340557
Editor: Muammar Qadafi, National Research and Innovation Agency, INDONESIA
Received: December 22, 2025; Accepted: May 12, 2026; Published: June 9, 2026
Copyright: © 2026 Trivalairat et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the manuscript.
Funding: Chulabhorn Royal Academy (Grant number RFCRA-2557-007) The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. There was no additional external funding received for this study.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Access to clean water is essential for human health, supporting hydration and overall body function, as well as enabling hygienic cooking and cleaning practices. According to the principles of the Ottawa Charter and the tenets upheld by the World Health Organization (WHO), access to clean drinking water is a fundamental component of public health and a basic human right that promotes health equity and well-being [1,2]. In rural areas, particularly those at high altitudes, communities depend on water sources derived from mountain for their daily life. Water safety in these highland areas is governed by basic physicochemical parameters (temperature, pH, dissolved oxygen) and mineral content (specific conductance, total dissolved solids, salinity), which dictate both pathogen survival and the long-term risk of urolithiasis (bladder stones). This mountain water is renowned for its purity and high mineral content, which can help to maintain balanced pH and support hydration, while also being essential for agriculture. However, the high levels of alkaline minerals such as calcium and magnesium can also lead to health complications, including the formation of bladder stones [3–7]. Moreover, agricultural markers like nitrate-nitrogen (NO3-N) serve as critical indicators of anthropogenic runoff in these vulnerable watersheds. Beyond natural mineralization [8–10], many rural communities around the world are facing challenges to their water security exacerbated by contamination (with herbicides, pesticides, and heavy metals) and climate change (such as long droughts and less precipitation), which has influenced the availability of clean fresh water. This necessitates the development of effective water management strategies to ensure that populations continue to have safe access to this vital resource [3,11–14]. Indeed, the responsible stewardship of mountain water is critical for safeguarding the well-being of mountain communities and securing their sustainable access to clean drinking water.Bhumibol Dam, the second-largest dam in Thailand, was constructed in 1953 and is located within the Mae Tuen Wildlife Sanctuary in Sam Ngao District, Tak Province, northwest Thailand. Currently, almost 2,000 members of the Paganyaw (S’gaw Karen) ethnic group live in mountain communities located above the dam. Given the long distances to the dam and its reservoir and the difficulty of traveling through this mountainous region, these communities’ water supply relies in part on systems derived water from mountain stream developed through community collaboration as part of a rural development program. While preliminary testing of the quality of drinking water from these systems was shown to be superior to that of natural sources, subsequent, more thorough testing revealed contamination. Specifically, water samples were found to breach safe drinking standards via the presence of 260 MPN/100 mL (MPN: most probable number) of coliform bacteria (including 7 MPN/100 mL of fecal coliform bacteria). Moreover, some residents of this region are unable to access the improved drinking water provided by these systems due to their remoteness, forcing them to consume untreated water from natural sources, particularly from sand-filtered puddles near streams [15–20]. This exposes them to various waterborne diseases, including acute abdominal pain, diarrhea, gastrointestinal conditions, and urinary issues, linked to contaminated water [21–23]. Moreover, widespread agricultural practices in this area, especially monoculture farming, exacerbate these health risks because of the intensive use of pesticides, which contaminate local water supplies and enter the food chain via fish. Given the remote geographical location, lack of reliable telecommunications and electricity, and limited access to healthcare in these vulnerable communities, there is an urgent need to improve water quality and sanitation.
The above background highlights the need to comprehensively assess the water sources used by mountain communities upstream of Bhumibol Dam in terms of their water quality and contamination levels. Such assessment is essential for effective management of the water supply for rural residents who live far from modern water infrastructure. Accordingly, this study was implemented with the aims of characterizing the quality of natural drinking water sources and water storage facilities used by the mountain communities upstream of Bhumibol Dam; investigating the pathways of agrochemical contamination into water, soil, and crops; and assessing the effectiveness of community water filtration systems. Additionally, compounded environmental health risks are defined here as the intersection of natural mineral-rich hydrogeology, agricultural chemical contamination, and local water-use behaviours. The overarching goal of this work is to enhance sustainable access to safe drinking water and improve health outcomes for these vulnerable communities.
Materials and methods
Study location and water sources
Mountain communities upstream Bhumibol Dam have been divided into six discrete “villages” since the building of the dam within Mae Tuen Wildlife Sanctuary, Ban Nar Sub-district, Sam Ngao District, Tak Province, northwest Thailand (Fig 1): Aum Haum (V1), Aum Vab (V2), Hin Lat (V3), Nahi (V4), Nam Sub (V5), and San Pa Puai (V6). Access to these field sites was granted by the Sam Ngao District Public Health Office (Tak Province, Thailand) under the “Promoting health and preventing water–borne disease of people living in the area of Bhumibol Dam” project. No additional collection permits were required as the study involved sampling from communal water sources and agricultural lands with the cooperation of local health authorities. The location of each village was recorded and plotted on a map using QGIS (v.3.44), with subsequent editing of legends and additional information using Adobe Photoshop (v.13.0). Map data was retrieved from Department of Water Resources (DWR) GIS and Geo-Informatics and Space Technology Development Agency (GISTDA). Given the absence of tap water, residents of these villages rely on water sources such as waterfalls, rivers, canals, rainwater, or sand-filtered puddles. They use the water from these sources directly or store it in containers in their homes. In this study, water samples were collected from various water resources used by local residents across these six villages between 19 and 21 November 2024, to assess water quality. This period was selected to capture the water‑quality profile during the dry season, when agricultural activity (post‑harvest) and reliance on stagnant water sources are both high. Sampling at this time allows identification of compounded risks during a period of increased community vulnerability and when access to sand‑puddle sources is limited to the dry season. All water sources in study areas were purposively selected as they represented the total land area accessible to and used by the community at the time of data collection. This approach was chosen to ensure comprehensive coverage, given the geographical dispersion and limited number of water sources available in these remote villages.
Water status at each site was categorized based on its use and characteristics, including: (1) source (SOURCE)—natural source vs. stored water (in containers); (2) drinking status (DRINK)—water used for drinking (drinking water) vs. water used only for other daily activities (non-drinking water); and (3) flow of the water (FLOW)—flowing vs. still. These factors were recorded using the YSI EXO multiparameter instrument (YSI Inc., Yellow Springs, OH, USA), along with physical parameters including height above sea level (ASL, m), specific conductivity (SPC, μs/cm), nitrate nitrogen (NO3-N) (NO3N, mg/L), mass of dissolved oxygen (DOm, mg/L), percentage of dissolved oxygen (DOp, %), pH, salinity (SAL, ppt), total dissolved solids (TDS, mg/L), and temperature (TEMP, °C). The data were subsequently analyzed to explore the relationships between water sources, environmental conditions, and water uses among the residents of each village. Additionally, samples from each water source were collected in two polyethylene bottles for pesticide detection and toxicity level assessment by purposive sampling.
Water samples were also collected from a single community water filter at Hin Lat village, which filters water from a mountain stream and supplies it to each household. The maintenance records of this community filtration system indicated that washing and filter replacement were infrequently performed, suggesting that its performance was suboptimal. Samples were collected from both the pre-filtration and post-filtration tanks of this system, in line with the sampling procedures outlined in the Standard Methods for the Examination of Water and Wastewater [24] for assessing the quality of drinking water. The samples were then separately collected into three sterile polyethylene bottles for laboratory analysis of physical and chemical indicators of water quality, microbial contamination, and heavy metals (Table 1). An acid-washed sterile bottle was used specifically for heavy metal analysis to prevent contamination and preserve metal ions. Samples were stored at 1–4°C and transported immediately to the Public Health Laboratory Division, Department of Health, Ministry of Public Health, Thailand, within 24 h for further examination. Furthermore, water samples from this filter were tested and assessed three times over a 6-month period with 3-month intervals to ensure consistency and reliability of the results.
All protocols in this study were deposited in protocols.io (http://dx.doi.org/10.17504/protocols.io.4r3l2zpz3l1y/v1).
Pesticide analysis
To assess potential pesticide exposure and associated risk factors, samples of the edible part of crops, soil, and drinking water were collected and tested for the presence and level of pesticides. Specifically, the testing of targeted pesticides from the organophosphate, carbamate, and cholinesterase inhibitor groups was performed using the GT-Pesticide Residual Test Kit (Bsmartscience Co., Ltd., Mueang Nonthaburi, Nonthaburi, Thailand). This kit extracts pesticides from water, soil, and agricultural products and indicates their presence or absence (TOX) by comparing the colour of the extracted sample tube with that of a control tube. A colour that is lighter than or similar to that of the control indicates the absence of pesticides (score 0), while a darker colour suggests their presence (score −1). Additionally, the kit provides qualitative assessment of the toxicity levels (SAFE). Specifically, by comparing the sample’s colour with that of a standardized comparison tube, a lighter colour indicates that the pesticide is present but within a safe level (score 0), whereas a darker colour indicates the presence of pesticides at an unsafe level (score 1).
Statistical analysis
Data were analyzed using SPSS 27.0. Before analysis, normality tests conducted using the Shapiro–Wilk method indicated that most water quality parameters did not follow a normal distribution within each village (p < 0.05). Box-and-whisker plots were used to visualize and analyze the spatial variation and statistical distribution of water quality variables across the six study villages. Therefore, non-parametric statistical methods were employed, with Kruskal–Wallis tests to compare differences among villages. To identify specific pairwise differences, Mann–Whitney U tests were performed as post hoc analyses. These tests were also applied to data collected from the water filter at Hin Lat at three different timepoints to assess the consistency and reliability of the water quality measurements over time.
In addition to the non-parametric tests, Pearson’s or Spearman’s correlation analyses were conducted to examine the relationships between water status (SOURCE, DRINK, FLOW) and water parameters (SPC, NO3N, DOm, DOp, pH, SAL, TDS, TEMP, TOX, SAFE). The choice between Pearson’s and Spearman’s correlation depended on the data distribution. Spearman’s rank correlation was used for non-normally distributed variables.
Multivariate analysis
Principal component analysis (PCA) was performed using Python libraries and dependence was analysed via the Google Colab environment, allowing exploration and description of the relationships among various water status and water quality parameters. Variables included water status (SOURCE, DRINK, FLOW), physical parameters (SPC, NO3N, DOm, DOp, pH, SAL, TDS, TEMP), and toxicity indicators (TOX, SAFE). PCA helped to identify underlying patterns and key factors influencing water quality and usage across the study sites.
Results
Water use
This study assessed water quality across six villages located upstream of Bhumibol Dam in Sam Ngao District, Tak Province, between 19 and 21 November, 2024. The villages spanned an altitudinal range of 675.7 to 758.0 m ASL (mean 729.88 ± 16.10). A total of 38 water sources utilized by residents were sampled and analyzed for water status and various physical parameters as follows: Aum Haum (V1, n = 5), Aum Vab (V2, n = 6), Hin Lat (V3, n = 17), Nahi (V4, n = 6), Nam Sub (V5, n = 3), and San Pa Puai (V6, n = 1). The results indicated that 13 sources of natural flowing water were primarily used for non-drinking purposes (such as cleaning and cooking) (Fig 2). The other 25 sources were used for drinking, with residents most commonly showing a preference for obtaining their drinking water from natural still water (sand-filtered puddles) from 10 sources, still water stored in containers (jars or concrete tanks) from 7 sources, flowing water derived through pipeline from mountain stream from 6 sources, and natural flowing water from 2 sources (Fig 3). Regarding water status, the residents generally preferred natural water over water in containers (25:13) and flowing water over still water (21:17) for drinking purposes (25:13). However, the results of independent sample t-tests indicated that there was no difference in water use between each village for the parameters SOURCE, DRINK, and FLOW (p-values 0.562, 0.935, and 0.907, respectively).
Water quality
Across the six villages, the water quality parameters were as follows: mean TEMP, 25.87 ± 2.59°C (range 16.77–29.44°C); mean DOp, 89.26 ± 18.25% (range 21.30%–105.20%); mean DOm, 7.37 ± 1.51 mg/L (range 1.68–8.94 mg/L); mean NO3N, 0.76 ± 0.87 mg/L (range 0.26–5.79 mg/L); mean SPC, 282.08 ± 188.73 μs/cm (range 2.00–583.00 μs/cm); mean TDS, 183.33 ± 122.64 mg/L (range 1.32–378.77 mg/L); mean SAL, 0.13 ± 0.90 ppt (range 0.00–0.28 ppt); and mean pH, 7.81 ± 0.44 (range 6.20–8.46) (Table 1) (Fig 4).
Box-and-whisker plots represent data from 38 water sources sampled between 19 and 21 November 2024. The parameters include: (A) temperature (°C); (B) nitrate-nitrogen (mg/L); (C) mass of dissolved oxygen (mg/L); (D) percentage of dissolved oxygen (%); (E) specific conductivity (µS/cm); (F) total dissolved solids (mg/L); (G) pH; and (H) salinity (ppt). The boxes indicate the interquartile range (IQR), the horizontal line within the box represents the median, and the whiskers indicate the range. Outliers are represented by individual points.
Normality tests (Shapiro–Wilk) revealed that most water quality parameters were not normally distributed within each village (p < 0.05). Therefore, non-parametric statistical methods were used. Kruskal–Wallis tests revealed significant variations in various water quality parameters across the six villages surveyed. Specifically, statistically significant differences were observed for TEMP (χ2 = 12.558, p = 0.028), DOp (χ2 = 13.805, p = 0.017), DOm (χ2 = 15.189, p = 0.01), SPC (χ2 = 16.905, p = 0.005), TDS (χ2 = 16.903, p = 0.005), SAL (χ2 = 17.079, p = 0.004), and pH (χ2 = 16.343, p = 0.006) (Table 2). No significant differences were detected for NO₃N across the villages.
To identify specific differences between the villages, post hoc Mann–Whitney U tests were performed. These analyses indicated significant differences in TEMP between V1 and V3 (p = 0.031), V3 and V6 (p = 0.007), and V4 and V6 (p = 0.020). For DOp, significant differences were found between V1 and V2 (p = 0.006), V1 and V3 (p = 0.025), V1 and V6 (p = 0.024), V2 and V4 (p = 0.025), and V3 and V4 (p = 0.027). Significant differences in DOm were observed between V2 and V3 (p = 0.042), V2 and V4 (p = 0.025), V3 and V6 (p = 0.021), and V4 and V6 (p = 0.030). Regarding SPC, significant differences occurred between V1 and V2 (p = 0.006), V1 and V3 (p = 0.014), V1 and V4 (p = 0.006), V2 and V6 (p = 0.039), and V4 and V6 (p = 0.020). “Similarly, TDS, SAL, and SPC showed significant differences between villages (see Table 2). Significant differences in pH were found between V1 and V2 (p = 0.006), V2 and V3 (p < 0.001), V2 and V4 (p = 0.004), and V2 and V6 (p = 0.038). Moreover, NO3N and SAFE did not show significant differences across villages, except for between V2 and V3 for NO3N (p = 0.039) and between V3 and V6 for SAFE (p = 0.010). No other statistically significant differences were observed between the remaining pairs of villages for the parameters examined.
The PCA biplot illustrates the clustering of six villages, divided into two groups based on their water status, water quality parameters, and safety levels: a risk group (purple dotted circle) and a safe group (green dotted circle) (Fig 5). The risk group, represented particularly by water sources from all villages, is characterized by higher loadings of NO3N, pH, TEMP, and TOX. This group is positioned opposite to the directions of FLOW and SAFE in the biplot, which indicates that it might involve unsafe, stagnant water sources. Unfortunately, the water sources in this group are often used for drinking, suggesting a potential health risk. The findings for the risk group indicate that residents using water sources in this group might be exposed to pesticides via their water and consume water of low quality.
Variables include source (SOURCE), drinking status (DRINK), flow condition (FLOW), height above sea level (ASL), specific conductivity (SPC), nitrate nitrogen (NO3N), mass of dissolved oxygen (DOm), percentage of dissolved oxygen (DOp), pH, salinity (SAL), temperature (TEMP), presence of pesticides (TOX), and toxicity levels (SAFE).
Conversely, the safe group, including sites such as Aum Haum, Aum Vab, Hin Lat, and Nahi, exhibited clustering associated with variables such as FLOW and SAFE, indicating that these sites tend to have safe, flowing water with low loading of pH, NO3N, and TOX. However, in the PCA biplot, this group is positioned away from the SOURCE and DRINK vector directions, suggesting that the residents of these villages may prefer not to drink water from these natural sources, and instead use it primarily for activities such as cooking and cleaning. Overall, this PCA plot effectively profiled water sources based on key water quality parameters, relating specific site conditions to the ecological and health risks that they pose.
Pesticide contamination in the environment
In addition to assessing water sources, soil and agricultural products were collected and analyzed using the GT-Pesticide Residual Test Kit to determine contamination and toxicity levels. This kit specifically detects organophosphate and carbamate compounds. In this study, seven soil samples were collected across four villages: Aum Haum (3 samples), Aum Vab (1 sample), Hin Lat (1 sample), and Nahi (2 samples) (Table 3). The results indicated that all samples were contaminated with pesticides; however, the contamination levels remained within safe limits.
Regarding agricultural products, various edible parts of crops were collected from the same sites as the soil samples across the four villages. Most edible plant parts showed no detectable pesticide contamination, especially those above ground (Table 4). However, three plants, arrowroot (Canna edulis), lemongrass (Cymbopogon citratus), and taro (Colocasia esculenta), from Hin Lat (V3) had edible below-ground parts that contained pesticide residues. Despite this, all samples remained within safe levels (Fig 6).
These results suggest that the soil in these villages is contaminated with pesticides, which may influence the plants growing there. This poses a potential health risk to residents who consume these crops, particularly the below-ground parts of them.
Water filter efficacy
A water filter, installed at Hin Lat in 2018, is frequently used by residents of that village and its surroundings to obtain water for drinking and cooking. Water samples collected at that site before and after filtration were stored at 1–4°C and immediately transported to the Public Health Laboratory Division, Department of Health, Ministry of Public Health, Thailand, within 24 h for analysis. The analyses included assessments of apparent colour (Pt-Co), turbidity (NTU), pH at 25°C, TDS (mg/L), hardness as reflected by CaCO3 (mg/L), sulfate (mg/L), chloride (mg/L), nitrate as NO−₃ (mg/L), fluoride (mg/L), nitrite as NO−₂ (mg/L), iron (mg/L), manganese (mg/L), copper (mg/L), zinc (mg/L), lead (mg/L), total chromium (mg/L), cadmium (mg/L), arsenic (mg/L), and mercury (mg/L), as well as bacteria in the form of coliforms (MPN/100 mL) and E. coli (MPN/100 mL). Water samples were tested three times over a 6-month period at approximately 3-month intervals. Descriptive statistics based on the results of these analyses are presented in S1 Table. However, owing to the data not being normally distributed, a Wilcoxon signed-rank test was performed. The results indicated no significant differences in any of the analyzed water-related variables between before and after filtration, suggesting limited effectiveness of this water filter in reducing physical, chemical, microbial, and heavy metal contaminants (Fig 7).
Discussion
Patterns of water use in Thai mountain communities
Mountain ecosystems in Thailand, particularly the ranges like the Tenasserim Hills, are indispensable as regional headwaters, supplying perennial flow to various rivers and canals that sustain downstream lowland communities, evergreen nature of the mountainous regions year-round, and the upstream communities [25–29]. Mae Tuen Wildlife Sanctuary serves a similar critical function, acting as an upstream source for Mae Tuen Stream, which eventually feeds Ping River, a primary tributary of Bhumibol Dam. The people living in the catchment area upstream of the dam are entirely dependent on these localized, natural water sources within the sanctuary for their daily needs, including drinking, cooking, and cleaning. The people of this region universally refuse to use or consume water directly from Ping River or the reservoir behind Bhumibol Dam. This is primarily because of the water’s high turbidity, which has led to the widespread perception that it is unclean. This aversion to highly turbid water is common among rural populations in Thailand and elsewhere who rely on surface water sources, where high sediment load is perceived as being strongly correlated with poor quality and biological contaminants [2,15,17,18,30–33]. Analysis of the patterns of water use confirmed that 34% of sampled sources were flowing natural water (mountain streams and waterfalls), which residents tend to favor for high-volume, non-drinking household activities such as washing and cleaning. This preference is rooted in the practical and perceived benefits of continuously moving water: its lower sediment load, reduced propensity for microbial contamination compared with stagnant pools, and an appearance suggestive of higher purity [2,32,33]. Furthermore, the ease of access to these sources of flowing water has allowed residents in places such as Aum Haum, Aum Vab, Nahi, and Nam Sub to establish settlements nearby. This in turn facilitates the transfer and storage of large volumes of water without the need to navigate kilometers of steep, difficult terrain to reach Bhumibol dam. However, despite this general preference for flowing water for use in everyday tasks like cooking and cleaning, the findings also raise concerns about the water used for drinking in this region, with still, sand-filtered puddles commonly being used for drinking water.
The analysis of water-use patterns in these mountain communities reveals a critical vulnerability gap where traditional knowledge and modern environmental risks converge. While residents correctly identify high turbidity as a proxy for poor water quality in larger rivers, their reliance on untreated, stagnant sources, specifically the sand-filtered puddles near streams, highlights a preference driven by accessibility and perceived traditional safety. However, this behaviour functions as the foundational layer of a compounded environmental health risk. These preferred drinking sources, as shown in results, act as primary collection points for both high mineralization (TDS, SPC) and anthropogenic runoff (NO3N). This suggests that the term of community water-use behaviour effectively concentrates exposure to multi-category stressors, rendering traditional filtration methods insufficient against modern chemical and geological hazards.
Choice of drinking water and water quality heterogeneity
While the communities in this study exhibit a general preference for natural, flowing water sources for domestic uses such as cooking and cleaning, the analysis of drinking water sources revealed that more than half of the sampled drinking water (54.2% of drinking sources) was still water from dug sand-filtered puddles/storage containers (Fig 2). This is particularly alarming because still or stagnant waterbodies, especially shallow, open pools such as these sand-filtered puddles (typically 30–60 cm in diameter and 5–10 cm in depth), inherently have a high risk of microbial contamination, exhibit increased concentrations of dissolved impurities, and can promote the transmission of vector-borne diseases [2,34–37].
The persistence of this practice appears to stem from a convergence of accessibility and indigenous knowledge. Generally, the sand-filtered puddles are easily dug close to residences, enabling the convenient collection of drinking water on a daily basis. More importantly, residents operate under the belief that the sand and soil naturally filter out “dirt, microbes, and toxins,” mimicking the principles of a slow sand filter (SSF) [38,39]. However, the efficacy of these rudimentary, unmanaged pools differs vastly from that of engineered SSFs, which require deep sand beds and a stable biological layer (dirty layer, schmutzdecke) to effectively remove pathogens [38–41]. The practice of using the same shallow puddle for 2–8 weeks before creating a new one risks the accumulation and possible fermentation of organic matter, leading to a potential increase in microbial load over time, creating a localized, ongoing health hazard [36,38,42,43].
Furthermore, statistical analysis supports the existence of significant heterogeneity in water quality across the six villages, as revealed by the Kruskal–Wallis test showing significant variations in SPC, TDS, and pH (Table 2). Specifically, villages V2 (Aum Vab), V3 (Hin Lat), and V4 (Nahi) exhibited significantly higher SPC and TDS, which aligns with the PCA risk-based clustering and indicates mineral-rich water conditions. In mountain hydrogeology, elevated SPC and TDS are directly indicative of high mineral content, predominantly in the form of alkaline minerals such as calcium and magnesium, which are dissolved as water percolates through the bedrock [44–46]. Water with this geochemical characteristic, while occasionally valued for its taste, might be directly linked to the health complication of bladder stone formation (urolithiasis), which is common in this region. In fact, according to community health reports, between 2019 and 2023, 22 people from Hin Lat reported having difficulty urinating (5 of whom received treatment for urinary tract stones). While this study did not directly measure the prevalence of urolithiasis among residents, the high levels of nitrate-nitrogen and alkaline pH observed in preferred drinking sources align with established environmental risk factors for renal calculi. In line with this, studies have consistently demonstrated that consuming hard water (high in Ca2+ and Mg2+) can increase the concentration of stone-forming minerals in urine, significantly raising the risk of calcium oxalate stones, particularly in individuals with a genetic predisposition for them or inadequate hydration [15,36,37,47,48]. The differences in pH of the water obtained from the analyzed water sources among the villages in this study, such as the high mean pH of 8.32 ± 0.10 at V2, further suggest that local geological or hydrogeological variations are the primary drivers determining the mineral composition, thereby generating distinct, site-specific health risks across the communities. The evidence suggests that geological profile of the region is not an isolated variable but a constant environmental stressor that compounds the risk of urolithiasis through the community’s specific choice of highly mineralized drinking sources.
A limitation of this study is the brief sampling window. Future research involving longitudinal monitoring across both monsoon and dry seasons would provide more comprehensive view of seasonal agrochemical runoff patterns.
Evidence of pesticide contamination and associated risk
The second critical issue emerges from the intersection of monoculture farming and water safety, where the Principal Component Analysis (PCA) identifies a distinct at-risk cluster that exemplifies compounded environmental health risks. Rather than facing isolated issues, the communities in villages V2 and V3 are exposed to a triad of stressors: (1) high mineral toxicity indicated by elevated SPC and TDS; (2) high alkaline pH, and (3) significant nitrate (NO3-N) and pesticide residuals. The outcome of this compounding effect is an elevated physiological burden on the residents, potentially manifesting in the high reported rates of urolithiasis observed in these specific clusters of the high-water quality.
Additionally, the PCA provides strong empirical evidence that the water used by the communities in the studied region constitutes a critical environmental health hazard. The analysis effectively clustered certain water sources into a “risk group” defined by high factor loadings for (NO3N), alkaline pH, and TOX (Fig 5). Alarmingly, many of the sources categorized into this cluster are those actively used by residents for daily drinking water. The strong correlation between NO3N and TOX in preferred sources, such as sand–filtered puddles, identifies agricultural runoff (e.g., of agricultural chemicals) as a dominant anthropogenic stressor. This contamination pathway is further substantiated by findings from analyses of soil and water in the environment. Specifically, pesticide residues (organophosphate and carbamate compounds) were detected in all tested soil samples and within some collected water sources (Table 3). Critically, this contamination extended to the next trophic level, as evidenced by detectable levels of contaminants in the below-ground edible parts of perennial crops (arrowroot, lemongrass, taro) from V3 (Hin Lat) (Table 4). Although rice was not tested in this study because the work was performed outside of its growing season, conversations with residents showed that agrochemicals are used in paddy fields in the region. These findings support the hypothesis that prevalent monoculture farming practices in the region, particularly agrochemical use in paddy fields, have resulted in the environmental partitioning and bioaccumulation of chemical pollutants, exposing residents through both their primary drinking water sources and their staple foods [49–51].
While the qualitative field kit determined the pesticide levels to be “safe,” this classification only reflects the kit’s detection limits and does not imply toxicological safety. Quantitative laboratory methods (e.g., gas chromatography-mass spectrometry (GC-MS)) are required to determine actual exposure levels. Nonetheless, the persistent detection of organophosphate and carbamate residues in water, soil, and crops signals a clear risk of chronic exposure. These compounds are known cholinesterase inhibitors, for which even low-level, long-term exposure, which is common in agricultural communities, has been linked to neurological, developmental, and endocrinological abnormalities [52–55]. Given the findings indicating such persistent chemical loading, there is an urgent need for quantitative assessments using laboratory-grade analytical methods in order to determine actual concentrations and better estimate the burden in terms of disability-adjusted life years (DALY).
Furthermore, the high NO3–N levels observed in the risk group pose a separate and immediate toxicological threat. Excessive nitrate exposure can oxidize hemoglobin in the blood, leading to the severe condition of methemoglobinemia (or “blue baby syndrome”), which is particularly dangerous for infants and vulnerable populations [8–11,56]. The co-occurrence of these two major agricultural pollutants (NO3N and pesticides) in the communities’ preferred sources of drinking water underscores the existence of a compounded environmental health crisis exacerbated by inadvisable choices when selecting sources for drinking water and its inadequate treatment.
Inadequate water infrastructure and public health implications
Behavioral and technical dimensions of water security address a vulnerability gap where residents often prefer untreated, stagnant sources, such as sand–filtered puddles, over community–provided systems due to accessibility or perceived taste. This suggests that health risks are not only environmental but are exacerbated by a mismatch between community needs and decentralized water treatment capabilities. The community water infrastructure, specifically the filter system installed at Hin Lat, represents a profound shortcoming in terms of efforts to protect public health. In fact, specialized laboratory analysis of samples taken over a 6-month period demonstrated its complete lack of efficacy: There were no significant differences in physical, chemical, microbial, or heavy metal contaminant levels between the pre-filtration and post-filtration tanks. This directly supports the conclusion that the filter does not function effectively at removing compounded risks of high mineralization and chemical runoff, leaving the community under the misapprehension that their water supply is safe. This failure is doubly concerning, because it was previously shown that the water supply supposedly improved by the installation of this filter actually contained high levels of coliform bacteria (260 MPN/100 mL), including fecal ones (7 MPN/100 mL) far exceeding national and WHO safety levels for drinking water [2,56]. The sustained presence of pathogens coupled with the ineffective filtration system means that the community is not only relying on contaminated water but also under the misapprehension that its water supply is safe. The confluence of this failed infrastructure and risky water-related practices create a compounded health burden for the Paganyaw communities. Specifically, people living in this region are simultaneously subjected to a double exposure risk. On the one hand, they risk acute exposure to microbial pathogens from the ongoing presence of fecal coliform bacteria in their water sources, leading to conditions such as diarrhea and gastrointestinal illnesses [36,38,43,44]. On the other hand, they suffer chronic exposure to alkaline minerals (linked to urolithiasis) and pesticide nitrate residues (linked to long-term systemic and neurological issues) because of their reliance on still sand-filtered puddles as sources of drinking water [8–10,15,36,37,47,48,56].
This situation highlights the urgent need for effective, sustainable water management in isolated, mountainous parts of Thailand like that upstream of Bhumibol Dam. Merely installing infrastructure to provide clean water, without subsequent technical validation, rigorous maintenance, and robust water safety plans, is insufficient to ensure water security in remote, vulnerable mountain communities. The findings of this work reveal the need for a shift toward context-appropriate, decentralized water treatment solutions that are locally manageable and resilient to the environmental and operational challenges of particular regions [57].
Supporting information
S1 Table. Descriptive analysis of water quality parameters of samples from a community water filters (Village V3, Hin Lat), evaluated across three sampling points over 6 months at approximately 3-month intervals.
https://doi.org/10.1371/journal.pone.0340557.s001
(DOCX)
Acknowledgments
Access to YSI EXO multiparameter instruments was supported by the Department of Zoology, Faculty of Science, Kasetsart University. Additionally, we thank Tom Buckle from Scribendi (www.scribendi.com) for editing a draft of this manuscript.
Declaration of generative AI and AI-assisted technologies in the writing process: During the preparation of this work, the authors used ChatGPT and Gemini in order to check the grammar and improve the language of the manuscript. After using these tools, the authors reviewed and edited the content as needed and took full responsibility for the content of this manuscript.
References
- 1.
World Health Organization. Ottawa Charter for Health Promotion: First international conference on health promotion, Ottawa, 21 November, 1986. Geneva: World Health Organization; 1986.
- 2.
World Health Organization (WHO), United Nations Children’s Fund. Progress on household drinking water, sanitation and hygiene 2000-2020: five years into the SDGs. Geneva: World Health Organization, United Nations Children’s Fund; 2021.
- 3. Møller K, Christensen B, Förster J, Piskur J, Nielsen J, Olsson L. Aerobic glucose metabolism of Saccharomyces kluyveri: growth, metabolite production, and quantification of metabolic fluxes. Biotechnol Bioeng. 2002;77(2):186–93. pmid:11753925
- 4. Chin NP, Dye TD. The health and wealth of mountain communities. Matern Child Health J. 2016;20(12):2413–4. pmid:27853922
- 5.
Rosborg I, Kozisek F. Drinking water minerals and mineral balance. London: Springer International; 2016. pp. 175.
- 6. Solár J, Tomaškovič J. Physicochemical properties of mountain streams in the High and Western Tatras. Environ Monit Assess. 2023;195(12):1543. pmid:38012457
- 7. Pop MS, Cheregi DC, Onose G, Munteanu C, Popescu C, Rotariu M, et al. Exploring the potential benefits of natural calcium-rich mineral waters for health and wellness: a systematic review. Nutrients. 2023;15(14):3126. pmid:37513544
- 8. Fan AM, Steinberg VE. Health implications of nitrate and nitrite in drinking water: an update on methemoglobinemia occurrence and reproductive and developmental toxicity. Regul Toxicol Pharmacol. 1996;23(1 Pt 1):35–43. pmid:8628918
- 9. Fewtrell L. Drinking-water nitrate, methemoglobinemia, and global burden of disease: a discussion. Environ Health Perspect. 2004;112(14):1371–4. pmid:15471727
- 10. Francis OA, Fagbayide DS. Nitrate in drinking water: how safe is this in Africa and Asia. Ann Univ Oradea Geogr Ser. 2018;1:28–35.
- 11. Robson M, Schneider D. Environmental health issues in rural communities. J Environ Health. 2001;63(10):16–9.
- 12. Sen SM, Kansal A. Achieving water security in rural Indian Himalayas: a participatory account of challenges and potential solutions. J Environ Manage. 2019;245:398–408. pmid:31163377
- 13. Drenkhan F, Buytaert W, Mackay JD, Barrand NE, Hannah DM, Huggel C. Looking beyond glaciers to understand mountain water security. Nat Sustain. 2023;6(2):130–8.
- 14.
Ghosh D, Chaudhary S, Sarkar S, Singh P. Water pollution in rural areas: primary sources, associated health issues, and remedies. In: Madhav S, Izah SC, van Hullebusch ED, Srivastav AL, editors. Water resources management for rural development. Elsevier; 2024. pp. 15–28.
- 15. Mirzazadeh M, Nouran MG, Richards KA, Zare M. Effects of drinking water quality on urinary parameters in men with and without urinary tract stones. Urology. 2012;79(3):501–7. pmid:22173182
- 16. Namsai M, Charoenlerkthawin W, Sirapojanakul S, Burnett WC, Bidorn B. Did the construction of the Bhumibol Dam cause a dramatic reduction in sediment supply to the Chao Phraya River? Water. 2021;13(3):386.
- 17. Sripo N, Chaichan M. Risk factors for waterborne diseases in communities upstream of Sirikit Dam. Lanna Public Health J. 2021;17(2):37–48.
- 18. Sripo N, Chaichan M, Buoloy P, Yuttaitra P. Water usage behaviors and drinking water quality among residents in communities upstream of Sirikit Dam. Boromarajonani Coll Nurs Sukothai J. 2021;4(2):58–71.
- 19. Tabucanon A, Rittima A, Raveephinit D, Phankamolsil Y, Sawangphol W, Kraisangka J, et al. Impact of climate change on reservoir reliability: a case of Bhumibol dam in Ping river basin, Thailand. Environ Nat Resour J. 2021;19(4):266–81.
- 20. Al Sas O, Prambudi H, Syafitri M. Examination of calcium oxalate crystals in mountain water consumers In Kuningan Regency. J Health Sains. 2024;5(4):278–83.
- 21. Chaturongkasumrit Y, Techaruvichit P, Takahashi H, Kimura B, Keeratipibul S. Microbiological evaluation of water during the 2011 flood crisis in Thailand. Sci Total Environ. 2013;463–464:959–67. pmid:23872186
- 22. Forstinus N, Ikechukwu N, Emenike M, Christiana A. Water and waterborne diseases: a review. Int J Trop Dis Health. 2016;12(4):1–14.
- 23. Maneeprakorn W, Tumcharern G, Bamrungsap S, Chansaenpak K, Segkhoonthod K, Rattanabut C, et al. Addressing water contamination and associated health issues through community-based interventions: a case study in Khon Kaen Province. Int J Environ Res Public Health. 2024;21(6):729. pmid:38928976
- 24.
American Public Health Association, American Water Works Association, Water Environment Federation. Standard methods for the examination of water and wastewater. 24 ed. Baxter TE, editor. Washington DC: APHA Press; 2023.
- 25. Dearden CWP. Decision-making arrangements in community-based watershed management in northern Thailand. Soc Nat Resour. 1999;12(7):673–91.
- 26.
Chomchai P. Public participation in watershed management in theory and practice: A Mekong River Basin perspective. In: Bruch C, Jansky L, Nakayama M, Salewicz KA, editors. Public participation in the governance of international freshwater resources. Tokyo: United Nations University Press; 2005. pp. 139–55.
- 27. Saipothong P, Preechapanya P, Promduang T, Kaewpoka N, Thomas DE. Community-based watershed monitoring and management in northern Thailand. Mt Res Dev. 2006;26(3):289–91.
- 28.
Wuthiwongyothin S. Assessment of water resources during the 21st century in northern Thailand with focus on Ping River Basin [dissertation]. Davis (CA): University of California, Davis; 2016.
- 29. Uthaipan K, Yuan Z, Zhao Y, Intararuang W, Vasinamekhin V, Sangmanee C, et al. Spatiotemporal variations in socioeconomic and hydrological factors impacting water quality in Thailand’s Major Rivers. Ecosyst Health Sustain. 2025;11.
- 30. Roth R. Spatial organization of environmental knowledge: conservation conflicts in the inhabited forest of northern Thailand. Ecol Soc. 2004;9(3).
- 31. Waltham NJ, Reichelt-Brushett A, McCann D, Eyre BD. Water and sediment quality, nutrient biochemistry and pollution loads in an urban freshwater lake: balancing human and ecological services. Environ Sci Process Impacts. 2014;16(12):2804–13. pmid:25384753
- 32. Jiang C, Zhang L, Li D, Li F. Water discharge and sediment load changes in China: change patterns, causes, and implications. Water. 2015;7(10):5849–75.
- 33. Manandhar S, Pratoomchai W, Ono K, Kazama S, Komori D. Local people’s perceptions of climate change and related hazards in mountainous areas of northern Thailand. Int J Disaster Risk Reduct. 2015;11:47–59.
- 34. Barakat A, Meddah R, Afdali M, Touhami F. Physicochemical and microbial assessment of spring water quality for drinking supply in Piedmont of Béni-Mellal Atlas (Morocco). Phys Chem Earth. 2018;104:39–46.
- 35. Stupar Z, Levei EA, Neag E, Baricz A, Szekeres E, Moldovan OT. Microbial water quality and health risk assessment in karst springs from Apuseni Mountains, Romania. Front Environ Sci. 2022;10:931893.
- 36. Sudsandee S, Fakkaew K, Keawdounglek V, Laor P, Worakhunpiset S, Apidechkul T. Drinking water investigation of hill tribes: a case study in Northern Thailand. Int J Environ Res Public Health. 2020;17(5):1698. pmid:32150932
- 37. Sudsandee S, Fakkaew K, Siriratruengsuk W, Worakhunpiset S, Apidechkul T. Quality of sources of drinking water and health among the hill tribe people of northern Thailand. Environ Geochem Health. 2022;44(3):873–92. pmid:34114160
- 38. Ray C, Grischek T, Schubert J, Wang JZ, Speth TF. A perspective of riverbank filtration. J Am Water Works Assoc. 2002;94(4):149–60.
- 39.
Ray C, Melin G, Linsky RB. Riverbank filtration: improving source-water quality. Berlin: Springer Science & Business Media; 2003.
- 40.
Ray C. Riverbank filtration concepts and applicability to desert environments. In: Shamrukh M, editor. Riverbank filtration for water security in desert countries. Dordrecht: Springer Netherlands; 2010. pp. 1–4.
- 41. Sandhu C, Grischek T, Kumar P, Ray C. Potential for Riverbank filtration in India. Clean Techn Environ Policy. 2010;13(2):295–316.
- 42. Warner NR, Levy J, Harpp K, Farruggia F. Drinking water quality in Nepal’s Kathmandu Valley: a survey and assessment of selected controlling site characteristics. Hydrogeol J. 2008;16(2):321–34.
- 43. Nisa FU, Umar R. Evaluation of physicochemical and microbiological parameters, and their correlation in Himalayan Spring Water Systems: A case study of District Kulgam of Kashmir Valley, India, Western Himalaya. Environ Monit Assess. 2023;195(4):441. pmid:36869254
- 44. Frei M, Bielert U, Heinrichs H. Effects of pH, alkalinity and bedrock chemistry on metal concentrations of springs in an acidified catchment (Ecker Dam, Harz Mountains, FRG). Chem Geol. 2000;170(1–4):221–42.
- 45. Pisciotta A, Tiwari AK, De Maio M. An integrated multivariate statistical analysis and hydrogeochemical approaches to identify the major factors governing the chemistry of water resources in a mountain region of northwest Italy. Carbonates Evaporites. 2018;34(3):955–73.
- 46. Jiang G, Chen Z, Siripornpibul C, Haryono E, Nguyen NX, Oo T, et al. The karst water environment in Southeast Asia: characteristics, challenges, and approaches. Hydrogeol J. 2021;29(1):123–35.
- 47.
Valyasevi A, Dhanamitta S. Clinical and field studies in Thailand. In: Van Reen R, editor. Proceedings: WHO Regional Symposium on Vesical Calculus Disease; Jan 1972; Bangkok, Thailand. US Department of Health, Education, and Welfare, Public Health Service, National Institutes of Health, National Institute of Arthritis, Metabolism, and Digestive Diseases; 1976. pp. 84.
- 48. Ryu HY, Lee YK, Park J, Son H, Cho SY. Dietary risk factors for urolithiasis in Korea: a case-control pilot study. Investig Clin Urol. 2018;59(2):106–11. pmid:29520386
- 49. Hugenschmidt C, Ingwersen J, Sangchan W, Sukvanachaikul Y, Uhlenbrook S, Streck T. Hydrochemical analysis of stream water in a tropical, mountainous headwater catchment in northern Thailand. Hydrol Earth Syst Sci Discuss. 2010;7(2):2187–220.
- 50.
Fröhlich HL, Ingwersen J, Schmitter P, Lamers M, Hilger T, Schad I. Water and matter flows in mountainous watersheds of Southeast Asia: processes and implications for management. In: Fröhlich HL, Schreinemachers P, Stahr K, Clemens G, editors. Sustainable land use and rural development in Southeast Asia: innovations and policies for mountainous areas. Springer; 2013. pp. 109.
- 51. Jidauna SB, Edziyie R, Campion BB. Spatio-temporal distribution of organochlorine pesticide residues in water and sediments of tropical reservoirs: a case study of Bui Reservoir, Ghana. Afr J Aquat Sci. 2020;45(4):421–30.
- 52. Brown MA, Brix KA. Review of health consequences from high-, intermediate- and low-level exposure to organophosphorus nerve agents. J Appl Toxicol. 1998;18(6):393–408. pmid:9840747
- 53. Androutsopoulos VP, Hernandez AF, Liesivuori J, Tsatsakis AM. A mechanistic overview of health associated effects of low levels of organochlorine and organophosphorous pesticides. Toxicology. 2013;307:89–94. pmid:23041710
- 54. Sánchez-Santed F, Colomina MT, Herrero Hernández E. Organophosphate pesticide exposure and neurodegeneration. Cortex. 2016;74:417–26. pmid:26687930
- 55. Antonangeli LM, Kenzhebekova S, Colosio C. Neurobehavioral effects of low-dose chronic exposure to insecticides: a review. Toxics. 2023;11(2):192. pmid:36851066
- 56. Fossen Johnson S. Methemoglobinemia: Infants at risk. Curr Probl Pediatr Adolesc Health Care. 2019;49(3):57–67. pmid:30956100
- 57. Mintz ED, Reiff FM, Tauxe RV. Safe water treatment and storage in the home. A practical new strategy to prevent waterborne disease. JAMA. 1995;273(12):948–53. pmid:7884954