Occurrence, sources and ecological risk assessment of per- and polyfluoroalkyl substances (PFASs) in water and sediment from urban rivers in Dhaka, Bangladesh

Study area and sampling locations

This study focused on three major rivers - Turag, Buriganga, and Shitalakkha - around Dhaka City, Bangladesh [Figure 1]. The banks of these rivers have turned into unplanned industrial areas. The metropolitan area of Dhaka is about 815.8 km² and is located at the center of Bangladesh. Dhaka City is one of the most densely populated cities in the world, home to approximately twelve million people, of which less than 25% are served by a sewage treatment facility^[34]. These rivers are used as a convenient means for disposing of domestic raw sewage as well as untreated industrial waste effluents from surrounding habitation and the nearby industrial belt. No information is available regarding the levels of PFAS contamination in these rivers. Therefore, these three rivers have been selected for the present study. Furthermore, sampling sites were strategically selected along the upstream, midstream, and downstream sections of each river to capture spatial variations in PFAS contamination. This division enables assessment of potential changes in PFAS levels due to differing anthropogenic inputs along the river course, such as industrial effluent discharges, urban runoff, and municipal wastewater inputs, which typically intensify in the midstream and downstream zones of urban rivers.

Figure 1. Map of Dhaka city, Bangladesh, showing the three urban rivers sampled in this study: BR, TR, and SR. Sampling sites were distributed across upstream (1), midstream (2), and downstream (3) locations for each river (e.g., BR1–BR3, TR1–TR3, SR1–SR3). BR: Buriganga River; TR: Turag River; SR: Shitalakkha River.

Collection of samples

Sampling was carried out at three points (up-, mid-, and downstream) along each river. A total of 27 surface water samples and 27 sediment samples were collected during the dry season, in January-April 2017. For water, triplicate composite samples (~100 mL each) were taken at each site using pre-cleaned polypropylene (PP) bottles. The bottles were rinsed first with deionized water, then methanol, and finally with site water before sample collection. Immediately post-collection, samples were passed through 0.45 µm membrane filters to eliminate suspended particles and stored in clean PP containers prewashed with methanol and Milli-Q water. Samples were transported on ice and kept at -20 °C until further processing, which occurred within 48 h. Sediment samples were collected using a handheld Ekman grab sampler targeting the top 0-5 cm of surficial sediment. At each site, triplicate composite samples (approximately 200 g each) were placed in polyethylene ziplock bags and kept chilled during transport. In the laboratory, sediments were freeze-dried, ground, sieved to < 2 mm, and stored in PP containers at -20 °C. For both matrices, each composite sample was then analyzed individually, and no further analytical replicates were performed on each composite. Thus, each reported value represents a single analytical determination of a site-specific composite sample. All materials used during sample collection and processing were handled to avoid external contamination.

Chemicals and reagents

Fifteen PFAS compounds were analyzed: PFBA, PFPeA, PFHxA, PFHpA, PFOA, PFNA, PFDA, PFUnDA, PFDoDA, PFTrDA, PFTeDA, PFBS, PFHxS, PFOS, and PFDS. Calibration and isotope-labeled internal standards (e.g., ¹³C₄–PFBA, ¹³C₂–PFHxA, ¹³C₄–PFOA, ¹³C₂–PFDoDA, and ¹³C₄–PFOS) were obtained from Wellington Laboratories (Canada) and Kanto Chemical (Japan). All standards were prepared in methanol and stored in PP vials at 4 °C. Solid-phase extraction (SPE) was performed using Oasis® WAX cartridges (6 cc, 150 mg) and ENVI-Carb^TM clean-up tubes (1 mL, 100 mg). High-purity solvents (methanol, acetonitrile, ammonium acetate, acetic acid) were purchased from Wako Pure Chemical Industries (Japan), and ultrapure Milli-Q water (18.2 MΩ) was used throughout. Filter membranes (0.45 µm, 47 mm) were sourced from ADVANTEC® (Japan). Detailed instrument parameters are provided in Supplementary Table 1.

Sample pretreatment

Pretreatment followed a modified SPE protocol previously described in the study “Occurrence and distribution of PFAAs in surface water and sediment of a tropical coastal area (Bay of Bengal coast, Bangladesh)” by Habibullah-Al-Mamun et al.^[20]. Briefly, 800 µL of a 5 ng/mL internal standard mix was added to 100 mL of filtered water. SPE cartridges were conditioned sequentially with 0.1% ammonium hydroxide in methanol (4 mL), methanol (4 mL), and Milli-Q water (4 mL). Samples were loaded at ~1 drop/sec. After discarding the initial flow-through and washing with 4 mL of 50% methanol in water, the cartridges were centrifuged to remove residual water. Target analytes were eluted with 4 mL of methanol followed by 4 mL of 0.1% ammonium hydroxide in methanol. The eluates were passed through an ENVI-Carb^TM cartridge for additional clean-up and collected in 15 mL PP tubes. Extracts were concentrated to 400 µL under nitrogen at 40 °C and filtered through a 0.2 µm nylon membrane prior to analysis. To avoid PFAS losses or contamination, samples were stored in PP autosampler vials with PP septa. Approximately 1 g of freeze-dried, homogenized sediment was weighed into 50 mL PP centrifuge tubes and spiked with the same internal standards used for water. After adding 10 mL of 1% acetic acid, samples were vortexed and sonicated at 50 °C for 25 min. Following centrifugation (3,000 rpm, 5 min), the supernatant was transferred to a clean tube. The sediment residue was sequentially extracted twice using 2.5 mL of 90:10 methanol–acetic acid (1%), followed by an additional 10 mL rinse with 1% acetic acid. All extracts were combined, cleaned, and processed identically to the water samples.

Instrumental analysis

PFASs were quantified using high-performance liquid chromatography coupled with tandem mass spectrometry (HPLC–MS/MS). An Agilent 1100 HPLC system was interfaced with a Micromass Quattro Ultima Pt mass spectrometer operating in negative electrospray ionization (ESI–) mode. Separation was performed on a Zorbax XDB–C18 column (150 mm × 2.1 mm, 5 µm) maintained at 40 °C. A 10 µL sample was injected, and elution was achieved using a gradient of methanol and 10 mM ammonium acetate at a flow rate of 0.2 mL/min. The column was maintained at 40 °C during the sample analysis. The gradient started at 10% methanol, increased to 62% after 5 min, held for 10 min, then raised to 65% and maintained for another 10 min. It was further increased to 70% methanol, held for 6 min, then adjusted to 100% and maintained for 5 min. Finally, the gradient was returned to its initial condition. Analyte detection used multiple reaction monitoring (MRM), and data analysis was performed using MassLynx™ software (v4.1). Detailed instrument parameters are provided in Supplementary Table 2.

Quality assurance and quality control

All labware was rigorously pre-cleaned with methanol, and contact with PTFE or other fluorinated materials was strictly avoided. Quantification was based on internal standard calibration curves (0.01 to 50 µg/L), with r² values consistently above 0.99. Calibration accuracy was verified for each analytical run. Detection limits (LOD) were defined as the concentration giving a signal-to-noise ratio ≥ 3, while LOQs corresponded to a S/N ratio ≥ 10. LOQs ranged from 0.016-0.4 ng/L (water) and 0.006-0.102 ng/g (sediment). Instrumental blanks (solvent only) and procedural blanks (Milli-Q with internal standards) were run at regular intervals. All blanks were below LOQ. Method accuracy was evaluated through spike recovery tests (n = 3), with average recoveries of 74%-114% for water and 87%-108% for sediment. Detailed recovery, LOD, and LOQ data are presented in Supplementary Table 3.

Ecological risk assessment

To evaluate the potential ecological hazards associated with PFAS contamination in the Buriganga, Turag, and Shitalakkha rivers, an environmental risk assessment (ERA) was conducted using the widely accepted risk quotient (RQ) approach. The RQ was calculated as the ratio of the measured environmental concentration (MEC) to the predicted no-effect concentration (PNEC), according to Equation (1).

where: MEC - measured environmental concentration [ng/L or ng/g dw]; PNEC - predicted no effect concentration [ng/L or ng/g]. Owing to the absence of comprehensive freshwater PNEC values for all 15 PFASs, relevant values were sourced from published studies^[20,35,36]. The collected PNEC values are presented in Supplementary Materials. Ecological risk thresholds were categorized as: very low (RQ < 0.01), low (0.01 ≤ RQ < 0.1), moderate (0.1 ≤ RQ < 1), and high (RQ ≥ 1)^[37].

Statistical analysis

Data analysis was conducted using SPSS (v23.0, IBM Corp., NY, USA). Prior to statistical evaluation, concentrations below LOQ were assigned values equal to LOQ/2, while those below LOD were set to zero^[38]. Normality of distribution was assessed using P–P plots. One-way ANOVA was applied to detect differences in PFAS levels across rivers. Principal component analysis (PCA) was employed to interpret potential sources and distribution patterns of PFAS contamination.

Occurrence, concentration and composition of PFASs in surface water

All fifteen targeted PFASs, including perfluorocarboxylic acids (PFCAs) and perfluorosulfonic acids (PFSAs), were detected in surface water samples from the Buriganga, Turag, and Shitalakkha rivers. PFHxA, PFHpA, PFOA, PFNA, PFDA, and PFOS were consistently present across all sites. Concentrations ranged from below the limit of detection (< LOD) to 52.63 ng/L [Table 1 and Figure 2]. One-way ANOVA analysis for the river water samples showed that there were no significant differences (P > 0.05) in the mean concentrations of PFASs among the Buriganga, Turag, and Shitalakkha rivers, indicating a broadly similar level of overall PFAS contamination. However, when individual PFAS compounds were considered, significant differences were observed. For example, PFHxA concentrations were significantly higher in the Buriganga River than in the Shitalakkha River (P < 0.05). Similarly, PFNA concentrations were significantly higher in both the Buriganga and Turag rivers compared to the Shitalakkha River (P < 0.001). PFHxS levels were also significantly higher in the Buriganga River than in the Shitalakkha River (P < 0.05). These findings suggest that, while diffuse sources contribute to a relatively uniform overall PFAS burden, specific compounds are more strongly influenced by local industrial discharges and wastewater inputs. The higher concentrations of PFHxA, PFNA, and PFHxS in the Buriganga and Turag rivers may reflect the impact of intensive industrial and urban activities along these riverbanks, whereas the lower levels in the Shitalakkha River could be due to fewer direct discharges or greater dilution capacity. The total PFAS concentration (ΣPFASs) was highest in the Buriganga River (mean: 71.12 ng/L), followed by the Turag (42.22 ng/L) and Shitalakkha Rivers (37.32 ng/L). ΣPFCAs were most abundant in the Buriganga (mean: 64.74 ng/L), followed by the Turag (38.97 ng/L) and Shitalakkha (32.47 ng/L). Similarly, ΣPFSAs were highest in the Buriganga (6.39 ng/L), suggesting more severe contamination in this river system [Table 1].

Figure 2. ΣPFASs concentrations (ng/L) in surface water samples from the BR, SR, and TR. ΣPFASs represents the sum of all quantified perfluoroalkyl and polyfluoroalkyl substances across each site. ΣPFASs: Total per- and polyfluoroalkyl substances; BR: Buriganga River; SR: Shitalakkha River; TR: Turag River.

PFOA was the dominant PFAS, accounting for 34%-59% of ΣPFASs [Figure 3], with concentrations between 8.13 and 52.63 ng/L. PFOS contributed only 1%-13% of the total PFAS burden. The PFOA-to-PFOS concentration ratio ranged from 4:1 to 40:1, likely due to higher solubility and lower sediment affinity for PFOA^[39]. Regulatory gaps and continued industrial use in rapidly urbanizing regions like Dhaka may also explain its prevalence. Similar PFOA dominance has been reported in rivers globally, including the Tennessee^[40], Yangtze^[41], and Elbe^[42]. However, some rivers such as the Jiulong and Júcar show PFOS dominance, highlighting variable source profiles and contamination histories^[43,44]. Other moderately detected PFASs included PFHxA (7%-32%), PFHpA (4%-9%), and PFNA (3%-12%). Less prevalent compounds, including PFBA, PFPeA, PFDA, PFUnDA, PFDoDA, PFTrDA, PFTeDA, PFBS, PFHxS, and PFDS, contributed minor fractions. These patterns reflect global shifts from long- to short-chain PFASs due to regulatory phase-outs^[45-47], consistent with increased short-chain usage in Bangladesh^[20]. Importantly, long-chain PFASs (e.g., PFOA, PFOS, PFNA) exhibit greater environmental persistence, bioaccumulation potential, and protein/lipid affinity, increasing risks of biomagnification and chronic health effects (e.g., carcinogenicity, reproductive/endocrine toxicity)^[48,49]. Short-chain alternatives (e.g., PFBA, PFBS) are less bioaccumulative but remain mobile and persistent, posing concerns about continuous exposure, cumulative mixture effects, and ecological impacts^[50].

Figure 3. Relative composition (%) of individual PFAS compounds in surface water samples. PFAS: Per- and polyfluoroalkyl substance.

Compared to the previous studies [Supplementary Table 4], PFAS concentrations in rivers around Dhaka were lower than in highly contaminated sites such as the Liuxi and Xiaoqing Rivers in China^[51,52] but comparable to moderately polluted rivers like the Jiulong, Oder, and Seine Rivers^[44,53,54]. However, levels in the Buriganga exceeded those in rural or less industrialized regions, including the coastal area of Bangladesh^[20], the Yellow River^[51], and Lakes Superior and Huron^[55].

Notably, PFCAs in this study were over three times higher than those in the Yangtze Estuary^[56] and Yodo River^[57], though lower than in more industrialized systems such as the Las Vegas Wash^[58]. PFOA concentrations in this study surpassed many international reports but were comparable to those from the Liuxi River^[52] and Taihu Lake^[59]. PFOS levels (0.95-5.95 ng/L) were moderate, exceeding values in the Oder^[54] and Ganges Rivers^[60] but lower than those found in the Swan-Canning Estuary^[47] and Júcar River^[43]. Overall, elevated PFAS concentrations, especially in the Buriganga River, highlight the influence of untreated industrial and domestic effluents and emphasize the need for targeted management and consistent monitoring in urban waterways around Dhaka city.

Occurrence, concentration and composition of PFASs in sediment

All fifteen PFASs were detected in sediment samples from the three rivers [Table 2 and Figure 4]. The one-way ANOVA analysis revealed that differences in the concentrations of individual PFAS and the overall mean PFAS concentrations in sediment samples among the Buriganga, Turag, and Shitalakkha rivers were not statistically significant (P > 0.05). This suggests that, despite variations in local sources and industrial activities, the levels of PFAS contamination in sediments are relatively comparable across these river systems. Concentrations ranged from < LOD to 8.25 ng/g dry weight (dw). The highest total PFAS burden was observed in the Buriganga (mean: 7.29 ng/g dw), followed by the Turag (5.38 ng/g dw) and Shitalakkha (4.14 ng/g dw). Mean ΣPFCAs in sediments were similar across rivers: Shitalakkha (2.26 ng/g dw), Turag (2.25 ng/g dw), and Buriganga (1.99 ng/g dw). Conversely, ΣPFSAs were pointedly higher in the Buriganga (5.30 ng/g dw), supporting its role as the most impacted site. Elevated levels likely result from high industrial and municipal discharge, historical accumulation, and limited regulation.

Figure 4. ΣPFASs concentrations (ng/g dry weight) in sediment samples from the Buriganga, Shitalakkha, and Turag Rivers. ΣPFASs: Total per- and polyfluoroalkyl substances.

PFHxA, PFNA, PFDA, PFUnDA, and PFOS were consistently found across all sites [Figure 5]. PFOS concentrations (0.21-8.25 ng/g dw) dominated, contributing up to 71% of ΣPFASs, especially in the Buriganga (mean: 4.01 ng/g dw). This reflects historical usage, persistence, and affinity of PFOS for sediments^[9]. PFUnDA and PFDS also contributed notably (2%-37% and 0%-21%, respectively), while others such as PFNA, PFHpA, and PFDoDA were present at moderate levels. The prevalence of PFOS in sediment aligns with findings from other urban rivers including the Chaobai^[61], Tangxun Lake^[62], and Júcar^[43]. Exceptions exist, such as the Huangpu River, where PFOA dominated^[38].

Figure 5. Relative composition (%) of individual PFAS compounds in sediment samples. PFAS: Per- and polyfluoroalkyl substance.

Compared to international datasets [Supplementary Table 5], sedimentary ΣPFAS levels in Dhaka were moderate. They were lower than those in highly contaminated sites (e.g., Liuxi, Júcar, Tangxun Lake)^[43,52,62], but comparable to or slightly lower than levels in the Orge River^[63], Conasauga River^[64], and Jiulong River^[43]. Higher levels were observed compared to the Chaobai, Elbe, Savannah, and Pearl Rivers^{[38,61,62,65]}. PFOS concentrations in sediments observed in the present study were similar to those in the Júcar^[43] and Conasauga Rivers^[64] but exceeded values from the Chaobai, Elbe, and Pearl Rivers^[38,61,62]. In contrast, PFOA levels in sediments in this study (< LOQ to 0.18 ng/g dw) were significantly lower than international reports from the Uji and Xiaoqing Rivers^[66,67]. The widespread presence of PFASs, particularly PFOS, in this study reflects persistent contamination driven by historical and ongoing discharges. Though levels are moderate globally, they indicate the need for stricter regulatory action and comprehensive monitoring in urban river systems.

Spatial distribution and source analysis of PFASs

To elucidate the potential sources and environmental behaviors of PFASs in the Buriganga, Shitalakkha, and Turag Rivers, both compositional analysis and multivariate statistical modeling were employed. Specifically, a comparative assessment of PFAS concentration patterns was conducted alongside PCA using the mean concentrations of 15 target PFASs in both water and sediment matrices.

PCA

To further differentiate PFAS source signatures and understand environmental partitioning behaviors, PCA was applied to standardized mean concentrations of PFASs in both water and sediment matrices. The first two components explained 81.1% of the total variance, with PC1 accounting for 54.47% and PC2 for 26.63% [Figure 6]. PC1 was strongly associated with the overall PFAS burden, particularly long-chain compounds such as PFHxA, PFHpA, PFDA, and PFOS. PC2 captured variability linked to partitioning behavior between water and sediment, highlighting PFTrDA and PFUnDA, which may indicate contributions from niche industrial applications.

Figure 6. PCA biplot of PFAS concentrations in water and sediment, illustrating spatial clustering of sampling sites (Buriganga, Turag and Shitalakkha) and associations with specific PFAS compounds. PCA: Principal component analysis; PFAS: per- and polyfluoroalkyl substance.

The PCA biplot revealed clear spatial clustering of river systems. Buriganga was situated on the negative axes of both PC1 and PC2, influenced by sediment-associated legacy PFASs (e.g., PFOS, PFHxS, PFNA), consistent with long-standing industrial pollution. This aligns with the role of the Buriganga River as the primary recipient of untreated industrial and domestic wastewater in Dhaka^[68,69]. The river is flanked by tanneries, dyeing plants, metal workshops, and plastic manufacturers^[70], with significant PFAS input likely from textile and garment processing. The lack of adequate waste management exacerbates PFAS accumulation and sediment resuspension continues to release these legacy contaminants^[71].

Shitalakkha River clustered strongly on the positive side of PC1, driven by short-chain PFASs (PFBA, PFUnDA) predominantly in water. This composition indicates recent inputs of alternative PFASs used in modern consumer and industrial goods. The river flows through industrial zones in Narayanganj where textile, chemical, paper, and food processing units discharge effluents - often untreated - into the water system. Although less industrialized than Buriganga or Turag, persistent discharges and lack of centralized treatment contribute to measurable PFAS levels. Turag River, moderately negative on PC1 and strongly positive on PC2, exhibited a unique profile dominated by mid-chain PFASs such as PFPeA, PFDoDA, and PFTrDA. These compounds may originate from specialized or emerging sources including electronics or high-performance coatings. The Turag receives discharges from various industries in the Tongi-Gazipur zone, including tanneries and textile factories, as well as emissions from brick kilns that partially combust PFAS-containing waste^[72]. Drainage channels carrying industrial effluents directly into the river further increase contamination risks^[73].

The spatial variation in PFAS composition across the rivers supports the conclusion that localized industrial practices are key determinants of PFAS signatures. Legacy long-chain PFASs in Buriganga align with chronic contamination from historical industrial activity. Shitalakkha reflects a transition to short-chain alternatives, while Turag’s profile indicates inputs from emerging or niche PFAS sources.

Ecological risk assessment

In this study, the risk assessment of PFASs for aquatic organisms was conducted based on the RQ method. In surface water samples, most PFASs exhibited very low ecological risks, with RQ values generally below 0.01 across all sampling sites [Table 3]. However, PFUnDA presented moderate ecological concern at specific locations, particularly at station BR3, where the RQ reached 0.634. PFOA, a well-documented persistent pollutant with significant toxicological relevance, showed consistently higher RQs than other compounds, ranging between 0.0057 and 0.0369 - indicating low to potentially moderate risk. Other compounds, including PFDoDA and PFTrDA, also demonstrated elevated RQs at isolated sites, although these remained within the low-risk category. Sediment samples exhibited a similar risk pattern, with most PFASs presenting negligible ecological risk. Notably, PFOS posed a significant threat, with RQ values exceeding 1 at multiple locations and reaching a peak of 1.68 at station BR3, thus falling into the high-risk category [Table 3]. PFUnDA and PFDoDA also exhibited low to moderate risk, with RQs ranging from 0.0052 to 0.0373 and 0.0056 to 0.0085, respectively. Although PFOA was consistently detected in sediment, its RQ values (0.00013-0.00207) remained well below levels of concern.

The spatial distribution of elevated RQs, particularly at station BR3, suggests the presence of localized point-source contamination. This pattern aligns with intensive industrial activity near the sampling locations, including garment washing, textile dyeing, tanning, and metal processing facilities - many of which are known to discharge PFAS-containing effluents directly into the river systems^[74,75]. High concentrations of PFDoDA in water and PFOS in sediment at these sites point to persistent industrial discharges and poor wastewater management.

Although the majority of PFASs did not individually exceed ecological risk thresholds, the presence of PFOS and PFOA at elevated levels in both environmental compartments raises concerns due to their well-established persistence, bioaccumulative potential, and toxicity to aquatic biota^[76,77]. Specific ecological impacts include endocrine disruption in fish by PFOS/PFOA, causing altered hormones, impaired reproduction, and developmental defects^[78,79]. Aquatic invertebrates exhibit reduced growth, survival, and feeding rates under PFAS exposure^[80]. Such effects threaten species survival and may disrupt trophic dynamics and community structures in freshwater ecosystems. Moreover, the risk associated with chronic, low-level exposure and mixture effects may not be adequately captured by traditional RQ-based assessments. Limited toxicity data for emerging PFASs, combined with uncertainty in PNEC derivation, further complicates accurate risk characterization^[81]. To address these limitations, future assessments should incorporate more comprehensive toxicological datasets, consider sublethal and chronic exposure scenarios, and employ probabilistic and mixture-based frameworks to better account for cumulative effects^[82,83]. Long-term monitoring and regulatory interventions are essential to manage and mitigate the ecological risks posed by PFASs in freshwater ecosystems, particularly in urban areas experiencing rapid industrial expansion.

Additionally, the ecological risk assessment of PFASs in the studied rivers highlights not only the threat to aquatic organisms but also the potential implications for human health. As PFASs are highly persistent and bioaccumulative, their presence in water and sediment may lead to trophic transfer through aquatic food webs, ultimately reaching humans via the consumption of contaminated fish and other consumable aquatic organisms. Additionally, direct exposure pathways such as the use of river water for irrigation, bathing, or other domestic purposes may increase the likelihood of human exposure. Extensive epidemiological studies link PFAS exposure (notably PFOA/PFOS) to adverse human health outcomes, including elevated cholesterol, hypertension, altered liver enzymes, reduced vaccine response, thyroid disruption, and reproductive effects (e.g., lower fertility and birth weight)^[84,85]. Long-term exposure correlates with kidney and testicular cancers^[86], underscoring the need to monitor freshwater systems that serve as indirect exposure pathways.

Limitations of the study

While this study provides critical baseline data on PFAS contamination in urban rivers of Dhaka, several limitations should be acknowledged. First, sampling was conducted only once during the dry season, which precludes assessment of temporal or seasonal variability in PFAS concentrations. Second, although 15 PFAS compounds were analyzed, only five isotope-labeled internal standards were used as surrogates; while this approach is widely accepted, it may introduce uncertainty for compounds lacking compound-specific surrogates. Third, key sediment characteristics such as total organic carbon content and grain size distribution - both of which influence PFAS sorption - were not measured, limiting our ability to fully interpret the spatial patterns in sediment contamination. Finally, while the study focused on a representative set of legacy and emerging PFASs, a broader suite of PFASs and their precursors remains uncharacterized. These limitations highlight the need for more comprehensive, long-term studies incorporating seasonal sampling, expanded PFAS analyte lists, and additional environmental parameters to better understand the fate and transport of PFASs in urban aquatic systems.