Occurrence and risk assessment of pharmaceutical and personal care product pollution in Urban Aquatic Ecosystems of South India

Abstract

Pharmaceuticals and personal care products are emerging pollutants since they persist for for an extended period and can harm the environment. This study assessed the prevalence, spatial distribution, and ecological risk of triclosan, naproxen, diclofenac, norfloxacin, and caffeine in surface and groundwater at 30 locations in Bengaluru and Mysuru, India. We used SPE–HPLC to look at the compounds and the risk quotient method to look at the ecological risk across different aquatic trophic levels. All target PPCPs were detected across the study area. Ecological risk assessment showed that the highest risk quotient values for norfloxacin, diclofenac, naproxen, triclosan, and caffeine were about 2, 13, 0.7, 8, and 16, respectively. Caffeine, diclofenac, and triclosan consistently exceeded over the high-risk threshold (RQ > 1), while norfloxacin was exhibited moderate risk and naproxen was in the low to middle of the risk range. The results show that urban aquatic ecosystems in southern India indicate substantial pharmaceutical contamination. This is primarily due to untreated sewage from households, hospital waste, and poor wastewater management. The high ecological risks linked to multiple compounds underscore the necessity for ongoing monitoring, enhanced wastewater treatment technologies, and risk-based management strategies to mitigate pharmaceuticals and personal care products pollution and safeguard aquatic ecosystems and public health.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-026-53365-y.

Introduction

Over the past few decades, the manufacturing and consumption of pharmaceuticals and personal care products (PPCPs) have grown significantly due to improvements in healthcare systems, rapid urbanization, and changing lifestyles¹. Because conventional wastewater treatment only partially eliminates many PPCPs, their residues are continuously discharged into surface and groundwater through municipal sewage, hospital effluents, agricultural activities, and improper disposal practices². Consequently, PPCPs are now widely detected in rivers, lakes, groundwater, and even potable water sources globally, leading to concerns about ecological toxicity, antimicrobial resistance, and prolonged human exposure^3,4.

PPCPs are especially concerning in water because they are biologically active and persistent, and can have harmful effects even at very low levels. Antibiotics may modify the structure of microbial communities and promote the proliferation of antibiotic resistance genes to antibiotics⁵. Non-steroidal anti-inflammatory drugs and antimicrobial agents have been observed to contribute to physiological stress, reproductive problems, and mortality in aquatic organisms⁶. Physicochemical properties like hydrophobicity, mobility, and resistance to biodegradation strongly influence their environmental behavior and ecological effects⁷. These properties govern their transport, persistence, and bioavailability, persist in water bodies, and become readily accessible to organisms⁸.

Rapid population growth, poor wastewater treatment infrastructure, and limited regulatory oversight make urban areas in developing countries especially vulnerable to PPCP contamination. India has one of the largest pharmaceutical manufacturing and consumption sectors in the world⁹. Consequently, there is a higher risk of pharmaceutical emissions being released into freshwater systems¹⁰. Although awareness of PPCPs as contaminants of emerging concern has increased, there remains insufficient field-based data from Indian cities, especially for groundwater systems that people believe are safe from contamination¹¹.

Previous research has documented the presence of PPCPs in various rivers, lakes, groundwater sources, and wastewater systems throughout India and other areas. These studies have provided baseline occurrence data and preliminary ecological insights. However, most previously published studies are limited in scope. They typically focus on a single city, a single source of water, or a small group of pharmaceuticals, and they often use different risk assessment frameworks and analytical targets. Consequently, direct spatial comparisons among urban areas and a comprehensive understanding of cumulative ecological risks remain limited.

Furthermore, previous studies have frequently been constrained by low sampling density, specific geographic coverage, and inadequate integration of surface water and groundwater systems. In many instances, ecological risk assessments depend on insufficient datasets or concentrate on specific compounds, neglecting variations in physicochemical behaviour that dictate fate, transport, and exposure pathways. As a result, the relationship between the presence of pharmaceuticals, their movement in the environment, and the observed ecological risks has not been thoroughly investigated, especially in developing countries that are quickly becoming more urbanised. Table 1 summarises reported concentrations from previous studies, but these data highlight inconsistencies in compound selection and methodological approaches, rather than providing one method to evaluate them.

Table 1.

PPCPs detected from previous studies.

PPCPS	Country	City	Location	Concentration	References
Triclosan	India	Uttar Pradesh	River Gomti	1100–9650 ng/L	12
Naproxen	India	Karnataka	Gurupura River	11,000–217,000 ng/L	13
Caffeine	India	Telangana	Musi River	16,600 ± 2,570 ng/L	14
Triclosan	China	Jiangsu	Taihu Lake	1.1 ng/L	15
Diclofenac	Europe	–	Surface water	2–41 ng/L	16
Norfloxacin	China	Qingshitan	Surface water	3.49–660.13 ng/L	17
Naproxen	South Africa	KwaZulu-Natal	Ladysmith River	2770 ng/L	18
Diclofenac	Pakistan	Makhad	Sawan River	116,000 ng/L	19
Diclofenac	Germany		Lake	6.3 × 10⁹ ng/L	20
Caffeine	Brazil	Manaus	Amazonian water bodies	69,650 ng/L	21
Diclofenac	Colombia	Barranquilla,	Magdalena River	680–7350 ng/L	22
Caffeine	Iran	–	Treated wastewater	19,860 ± 2,080 ng/L	23
Naproxen	Czech Republic	Bohemia	Stenava River	160 ng/L	24
Norfloxacin	Kenya	–	Surface water	2.66 × 10⁴ ng/L	25
Caffeine	Egypt	Sudan	Nile River	5.385 × 10⁴ ng/L	26
Naproxen	India	Nagpur	Nag River	5.0 × 10² – 1.59 × 10⁴ ng/L	27
Caffeine	Brazil	São Paulo	Cotia River	17,970 ng/L	28
Triclosan	India	Mumbai	Mithi River	1.68 × 10³ ng/L	29
Norfloxacin	South Brazil	Paraná	Barigui river	1.3 × 10² ng/L	30
Triclosan	China	Wuhan	Surface water	466 ​ng/L	31

To address these gaps, this study is the systematic, cross-city evaluation of five commonly used drugs—triclosan, naproxen, caffeine, diclofenac, and norfloxacin—at 30 surface water and groundwater sites in the Bengaluru and Mysuru urban ecosystems. This study advances previous research by integrating high-resolution occurrence data with multi-trophic ecological risk assessment, spatial hotspot analysis, and a mechanistic interpretation grounded in physicochemical properties. Integrating fate and transport behavior with risk quotient analysis improves the assessment of chronic exposure and ecological stress. This integration provides new insights into the occurrence and ecological risks of pharmaceutical pollution.

In urban India, drinking water comes from both surface water and groundwater, raising concerns about potential human exposure to PPCPs. However, thorough evaluations of human health risks are limited by the absence of compound-specific regulatory thresholds and exposure data. Therefore, this study predominantly focuses on occurrence and ecological risk, while qualitatively addressing implications for human exposure.

Material and methodology

Sampling locations and preparatory work

This study was conducted at multiple sampling sites identified as potential PPCP contamination hotspots within and around Bengaluru and Mysuru, Karnataka, India. Thirty sampling sites were strategically selected to provide representative coverage of surface and groundwater sources (rivers, lakes, and aquifers) that serve as major water supply sources.

The chosen sites comprised rivers such as the Arkavathi, Vrishabhavathi, Kabini, Dakshina Pinakini, and Cauvery, all of which are subject to anthropogenic pressures including industrial effluent discharge, sewage inflow, and urban runoff. Lakes such as Bellandur, Madiwala, Hebbal, Kukkarahalli, and Karanji were also chosen based on their ecological sensitivity and historical importance as water bodies surrounded by dense human settlements. Groundwater was sampled from peri-urban and urbanised regions such as Electronic City, Whitefield, Yelahanka, Rajajinagar, Koramangala, and JP Nagar, some of the most densely populated areas that receive a major portion of the municipal water supply. These areas were selected because they are highly prone to leaching from unregulated septic systems and landfills and are located in close proximity to untreated sewage drains^32,33.

All sampling sites were geo-tagged using a handheld GPS unit to accurately record geographic coordinates, as presented in Table S1. Figure 1 displays the spatial distribution of sampling sites across the Bengaluru and Mysuru regions. Grab samples were collected at all sites using thoroughly cleaned high-density polyethylene (HDPE) containers to maintain sample integrity and prevent contamination. Sampling was performed under dry weather conditions of April–May 2024 to minimise dilution effects caused by rainfall. Samples were stored in insulated cool boxes (4 °C) and transported to the laboratory on the same day for analysis. One grab water sample was collected at each site, resulting in a total of 30 environmental samples. Field duplicates were not collected, but all samples were analyzed in triplicate in the laboratory, and the reported concentrations represent the mean of the replicates. Physicochemical parameters (pH, temperature, and conductivity) were measured in situ for each sample.

Fig. 1.

Map displaying sampling sites around Bengaluru and Mysuru regions, generated using ArcGIS software (Version 10.8; ESRI, Redlands, CA, USA) under the JSSAHER institutional license (Subscription ID: 4907031121). Software URL: https://www.arcgis.com/index.html.

Pharmaceuticals and laboratory chemicals

All PPCPs and reagents used were of analytical and HPLC grade. Methanol (≥ 99.9%, Sisco Research Laboratories (SRL), India) and acetonitrile (≥ 99%, SRL, India) were used for HPLC analysis. Norfloxacin, diclofenac, naproxen, triclosan, and caffeine (purity ≥ 98%) were purchased from Sigma-Aldrich. Ultrapure water (18.2 MΩ·cm) was obtained using a Milli-Q system (Millipore, USA). All reagents were of analytical grade and used without further purification. Table S2 summarizes the physicochemical properties of the target PPCPs. There are no public city-level prescription registries for certain drugs. To prioritise target compounds, we used several quantitative proxies, such as market/sales data showing that diclofenac is used a lot in India (market value ~ USD 312 million in 2025). (ii) There is evidence of over-the-counter (OTC) sales of antibiotics in Indian retail pharmacies, and a recent survey in Bengaluru found that about 26% of pharmacies dispensed antibiotics without a prescription^34,35. This supports the concept of prioritising fluoroquinolone antibiotics like norfloxacin. (iii) National statistics on the consumption of caffeinated drinks (coffee production and consumption are approximately 91,000 tonnes of green beans in 2023), which is in line with widespread caffeine exposure.(iv) Prior occurrence studies from Bengaluru and other Indian urban rivers that identified diclofenac, naproxen, triclosan, and norfloxacin to be common PPCPs, which support their environmental relevance in the study region. The choice of norfloxacin, diclofenac, naproxen, triclosan, and caffeine as target analytes was based on these data-driven criteria, as well as the compounds’ physicochemical persistence and toxicological concern^36,37.

Analysis and extraction

The extraction procedure and HPLC analysis followed US EPA Method 1694: Determination of Pharmaceuticals and Personal Care Products in water samples by solid-phase extraction (SPE)³⁸. An autosampler, a photodiode array detector (PDA), and a deuterium (D2) detector were integrated into the AHT2010 HPLC system (Shimadzu, Japan), were used for chromatographic analysis.

Separation was performed using a 250 × 4.6 mm Phenomenex C-18 column. Chromatographic conditions (solvent ratio, mobile phase composition, flow rate, and pH) were optimized, and system suitability parameters (capacity factor, column efficiency, and asymmetry factor) were subsequently evaluated. Before selecting a final method, the HPLC conditions were methodically optimised to ensure precise quantification of the identified PPCPs. The mobile-phase composition (ACN and MeOH in various aqueous ratios), pH adjustment, flow rate, and detection wavelengths for each analyte were compared as part of method optimisation. Each 500 mL water sample was passed through a 0.45 μm membrane filter and then through Oasis HLB SPE cartridges (6 mL, 200 mg, Waters Corporation) at a flow rate of 5 mL/min. Before loading the samples, the cartridges were preconditioned with 5 mL of methanol and 5 mL of ultrapure water. After loading the samples, the cartridges were cleaned with 5 mL of ultrapure water, dried under vacuum for 10 min, and then eluted with 10 mL of methanol and 5 mL of acetonitrile. The correlation coefficients for norfloxacin, diclofenac, naproxen, triclosan, and caffeine of the standards were assessed within the ranges of 1–50, 0.5–20, 1–25, 0.1–10 and 0.5–25 µg/ml, respectively. All calibration curves showed excellent linearity (R² = 0.998–0.999). The norfloxacin, diclofenac, naproxen, triclosan and caffeine in the samples were identified by peaks at retention times 3.5, 6.2, 5.1, 7.3 and 2.5 min, respectively, based on standard retention times. The formula y = mx + c, where x is the analyte concentration in the samples and y is the peak area, was then used to calculate the concentrations. The limit of detection (LOD) and limit of quantification (LOQ) were calculated based on the standard deviation of the response and the slope of the calibration curve using the following equations:

1

2

where σ represents the standard deviation of the response (y-intercept) and S represents the slope of the calibration curve. Table 2 summarizes the calibration and standardization parameters. HPLC–PDA was employed due to its robustness and cost-effectiveness, and efficacy in quantifying PPCPs with significant UV absorbance, especially in extensive monitoring studies where access to LC–MS/MS facilities was limited. Fig. S1 represents the chromatogram of pharmaceutical compounds using HPLC and Fig. S2 represents the calibration curve of the targeted analytes.

Table 2.

HPLC measurement: calibration and standardization report.

Parameters	Norfloxacin	Diclofenac	Naproxen	Triclosan	Caffeine
Wavelength (nm)	278	276	230	280	272
Mobile Phase	Water: ACN (70:30)	MeOH: Water (60:40)	ACN: Water (50:50)	MeOH: Water (80:20)	Water: MeOH (60:40)
Linear regression (R2)	0.999	0.999	0.998	0.998	0.999
Range (µg/ml)	1–50	0.5–20	1–25	0.1–10	0.5–25
Tailing factor	1.1	1.2	1.1	1.0	1.2
Retention time (min)	3.5	6.2	5.1	7.3	2.5
Calibration Range (µg/ml)	1–60	2–15	1–30	1–10	2–25
Concentration Range (µg/ml)	2–35	1–15	2–20	0.2-8	1–20
LOD	0.19	0.11	0.25	0.13	0.15
LOQ	0.65	0.45	0.70	0.25	0.40

ACN: Acetonitrile, MeOH: Methanol.

Quality Assurance and Quality Control (QA/QC)

Quality assurance and quality control protocols were implemented during sampling and instrumental analysis in order to ensure data integrity. Field and procedural blanks were analyzed to determine whether contamination occurred during sampling, extraction, and analysis. No target PPCPs were detected above method detection limits in blank samples. We used multi-point external calibration curves (R² ≥ 0.998) and replicate injections to evaluate the performance of the instruments and the accuracy of the analysis. Relative standard deviations remained below 5%. Table S3 and S4 show the method detection and quantification limits for each analyte. Spike recovery experiments, which are explained in detail in the next section, were carried out to assess the accuracy of the analytical method.

Recovery studies

To evaluate the analytical method’s accuracy, ultrapure and environmental water samples were spiked with known concentrations of the target PPCPs (10, 20, 50, 80, and 100 ng/) in recovery experiments. The extraction and HPLC techniques adopted for real water samples were also used for the spiked samples. The recovery results for each analyte are summarised in Table S3 and S4. Relative standard deviations (RSDs) ranged from 4.3 to 7.1%, and mean recoveries varied from 83.2 to 96.8%, the method provides respectable precision and good accuracy. The analytes with higher average recoveries were triclosan (96.8%) and norfloxacin (92.4%), which demonstrated low matrix interference and accurate quantification at all tested concentration levels.

Ecotoxicological assessments

The risk quotient (RQ) approach was used to assess the environmental risk of each analyte. The RQ value was determined by dividing the measured environmental concentration (MEC) by the predicted no-effect concentration (PNEC) (RQ = MEC/PNEC)³⁹. PNEC values were calculated using toxicity data in accordance with Organisation for Economic Co-operation and Development (OECD) guidelines. The acute toxicological values (EC50 or LC50) were divided by a safety margin of 1000 to calculate the PNEC, as per previous studies⁴⁰. The acute toxicity data for the target organisms in this study, including fish, amphibians, mollusks, and algae were obtained using ECOTOX Knowledgebase⁴¹. An RQ value greater than 1.0 indicates a significant amount of environmental contamination, whereas RQ values 0.01–0.1 and 0.1–1 imply low and moderate risk to the target species, respectively. Values below 0.01 indicate negligible risk. Table S5 represents the PNEC value for the targeted PPCPs⁴². Amphibians were incorporated as a sensitive surrogate group owing to their permeable integument and aquatic life stages, which heighten susceptibility to PPCPs, despite being infrequently evaluated in PPCP studies.

In this study, the quantitative risk assessment was confined to ecological risk, applying the RQ methodology. Human health and socioeconomic risks were not quantitatively assessed; instead, they were discussed qualitatively based on observed environmental concentrations, documented health effects, exposure scenarios, and consequences for water utilisation, fisheries, and wastewater management.

The ecological risk assessment utilised a single-compound RQ methodology, and the potential additive, synergistic, or antagonistic effects resulting from PPCP mixtures were not quantitatively assessed.

Results and discussion

Physicochemical analysis

Table 3 presents the physicochemical characteristics of the sampled water. Physicochemical analysis of Bengaluru and Mysuru water samples identified considerable variation in major water quality parameters. The pH levels ranged from 6.2 to 9.3 in Bengaluru and from 6.4 to 8.9 in Mysuru, indicating neutral to slightly alkaline conditions that may influence microbial activity and pollutant interactions. Turbidity was considerably higher in Bengaluru (145–300 NTU) than in Mysuru (109–269 NTU), perhaps due to higher suspended particulate matter resulting from urban runoff or industrial outfall. Chemical oxygen demand (COD) varied between 100 and 600 mg/L in Mysuru and reached up to 900 mg/L in Bengaluru, indicating a greater abundance of oxidizable organic material in the water bodies of Bengaluru’s water bodies. Interestingly, levels of chlorides were much greater in Bengaluru (152.4–251.2 mg/L) compared to Mysuru (22.7–133.3 mg/L), possibly due to intrusion of domestic wastewater or localized salt sources. Biochemical oxygen demand (BOD) levels were relatively low in both cities, being between 0.1 and 0.7 mg/L (Bengaluru) and 0.2–0.6 mg/L (Mysuru), indicating limited biological activity or organic pollution. Agar plate counts confirmed the presence of high microbial loads in both locations, with Bengaluru at 60,000–80,000 CFU/mL and Mysuru a little higher at up to 80,000 CFU/mL. These findings indicate possible microbial and chemical contamination harm in both urban water ecosystems. The presence of low BOD values with high microbial counts indicates that microbial abundance was not directly associated with the presence of readily biodegradable organic matter. BOD measures the short-term oxygen demand associated with biodegradable organic carbon, but microbial counts could remain high because of continuous contamination from wastewater, biofilm detachment, or microorganisms adapted to low-carbon environments. For this study, a specific set of physicochemical parameters was selected: temperature, pH, electrical conductivity, turbidity, biochemical oxygen demand, and chemical oxygen demand. These parameters are widely recognized as key indicators governing the environmental behaviour of PPCPs, as they directly influence compound solubility, mobility, microbial activity, redox conditions, and organic matter availability in aquatic systems. The selection was therefore based on the study’s principal aim of correlating pharmaceutical presence with fate, transport, and ecological risk, rather than conducting a comprehensive water quality assessment. Comprehensive physicochemical and nutrient profiles of these regions have been reported previously, and the present work builds upon this existing knowledge by integrating pharmaceutical contamination and risk-based evaluation^43–45.

Table 3.

Results of physicochemical characterization of samples.

Parameter	Unit	Bengaluru	Mysuru
pH	–	6.2–9.3	6.4–8.9
Turbidity	NTU	145–300	109–269
COD	mg/l	120–900	100–600
Chlorides	mg/l	152.4-251.2	22.7-133.3
BOD	mg/l	0.1–0.7	0.2–0.6
Agar plate count	CFU/ml	60,000–80,000	70,000–80,000

Analysis of PPCP residues

The analysis of 30 sampling sites at Bengaluru and Mysuru identified that the concentration of PPCPs such as norfloxacin, diclofenac, naproxen, triclosan, and caffeine was site-specific and influenced by factors such as the nature of the water source (river, lake, or groundwater), the urbanisation level, the distance from healthcare facilities, and the presence of wastewater discharge systems. The concentrations detected at each site are presented in Table S6 and depicted in Fig. 2. The highest levels of contamination were observed in surface waters, especially in locations receiving direct industrial effluents or sewage. Certain sites exhibited very high cumulative PPCP concentrations, including 540.35 µg/L of norfloxacin, 302.78 µg/L of diclofenac, and 712.6 µg/L of naproxen. Site 6, a eutrophic urban lake with residential and commercial surroundings, revealed a very high naproxen concentration (650.1 µg/L), suggesting sewage inflow as a probable source of contamination. Site 17, exposed to sewage inflow from peripheral Bengaluru regions, revealed elevated concentrations of norfloxacin (447.66 µg/L) and caffeine (181.6 µg/L), pointing towards the impact of domestic discharges. Groundwater sources, in general, had lower levels of PPCPs but were not free from contamination. Industrial or densely populated regions, such as site 3 and site 7, had quantifiable amounts of all five PPCPs, i.e., norfloxacin (30.81 µg/L and 27.52 µg/L, respectively), suggesting potential leaching from surface sources or septic tank effluent infiltration. Similar trends were noted at other groundwater locations like site 19, with significantly elevated caffeine levels at 32.9 µg/L, indicating chronic household wastewater inputs. In contrast, less urbanized groundwater sites, such as site 2, site 14, and site 15 exhibited relatively lower levels. Site 2 exhibited low to moderate levels of PPCPs, and site 14 had minimal contamination, perhaps due to hydrogeological barriers or reduced population pressure. Nevertheless, caffeine remained detectable in most of these sites, reflecting extensive domestic sewage influence.

Fig. 2.

Represents the concentration of targeted PPCPs obtained from Bengaluru and Mysuru samples. Sites 1–15 are sampling locations within the Bengaluru city area, whereas 16–30 are sampling sites in the Mysuru region.

Other significantly affected surface water locations include site 23, which had 701.1 µg/L of naproxen and 210.4 µg/L of triclosan. This may be attributed to the site’s proximity to pharmaceutical waste disposal areas. Site 21, a semi-urban surface water location, had equally high levels of Norfloxacin (396.13 µg/L) and Naproxen (545.9 µg/L), indicating combined domestic and hospital effluent intrusion. Notably, the Cauvery River at site 29, which was anticipated to be cleaner given its volume of flow, still contained high PPCP concentrations (e.g., Norfloxacin 322.43 µg/L, Naproxen 426.7 µg/L), indicating upstream contamination from settlements and towns.

Caffeine appeared as the most ubiquitous compound, found nearly in all sites. Higher levels of caffeine were noted not only in surface waters but also in urban groundwater locations (e.g., site 16: 21.2 µg/L, site 19: 32.9 µg/L), indicating possible inefficiencies in existing waste management systems. Values reported as not detected (n.d.) indicate concentrations below the method detection limit, either through rapid degradation, or sorption on sediments^46,47.

High PPCP levels were primarily observed in areas with considerable urban land use, near to major sewage drains, and water bodies known to receive untreated or partially treated wastewater. In Bengaluru and Mysuru, rivers and lakes that contain wastewater possessed consistently higher PPCP levels than groundwater sites. This suggests that domestic sewage is a major contributing source. The regular finding of substances linked to human use and clinical use, like caffeine, diclofenac, and norfloxacin, supports the assumption that wastewater is a primary source. Although direct discharge volumes and source apportionment modeling were not conducted, the elevated PPCP concentrations in sewage-impacted areas strongly suggest anthropogenic inputs.

Assessment of the ecological risks associated with PPCPs

The risks to the environment posed by the presence of norfloxacin, diclofenac, naproxen, triclosan, and caffeine in the surface and groundwater bodies of Bengaluru and Mysuru were evaluated using the RQ approach. A distinct pattern of increased ecological risks across trophic levels is shown by the RQ values computed for each PPCP across the 30 sampling locations (Table 4) and visualized in Fig. 3, with several compounds frequently exceeding the PNEC thresholds for susceptible aquatic species. The overall risk profile suggests that untreated household sewage, hospital effluents, and urban runoff are likely contributors of the chronic and compound-specific toxic pressure observed in the aquatic ecosystems under study.

Table 4.

Risk assessment of PPCP based on their PNEC and MEC calculation.

Site	Norfloxacin	Risk	Diclofenac	Risk	Naproxen	Risk	Triclosan	Risk	Caffeine	Risk
1	1.24188	HR	6.2	HR	0.3104	MR	3.75333	HR	3.5	HR
2	0.21938	MR	1.18696	HR	0.0426	LR	0.77	MR	5.9	HR
3	0.1925	MR	1.05217	HR	0.0337	LR	0.61333	MR	3.7	HR
4	3.37688	HR	13.16087	HR	0.7126	MR	8.91667	HR	12.9	HR
5	0.16	MR	0.86087	MR	0.0285	LR	0.49	MR	2.8	HR
6	3.05	HR	n.d	n.d	0.6501	MR	6.49333	HR	0.9	MR
7	0.17188	MR	0.91304	MR	0.0314	MR	0.63333	MR	1.3	HR
8	0.19875	MR	1.02174	HR	n.d	n.d	0.74	MR	5.3	HR
9	0.18188	MR	0.96087	MR	0.0345	LR	0.67	MR	1.7	HR
10	0.2075	MR	1.10435	HR	0.039	LR	0.83	MR	15.5	HR
11	1.81563	HR	8.05652	HR	0.3702	MR	5.18667	HR	15.5	HR
12	2.87375	HR	11.54783	HR	0.6125	MR	6.78	HR	19.7	HR
13	2.61313	HR	10.52174	HR	n.d	n.d	6.23667	HR	18.6	HR
14	0.16875	MR	0.8913	MR	0.0294	LR	0.53	MR	3.9	HR
15	0.14	MR	0.77826	MR	0.0248	LR	0.44	MR	2.6	HR
16	0.15313	MR	0.82174	MR	0.0271	LR	0.58	MR	9.2	HR
17	2.7975	HR	10.06957	HR	0.582	MR	6.28667	HR	7.6	HR
18	0.13	MR	n.d	n.d	0.0227	LR	0.41667	MR	n.d	n.d
19	0.1475	MR	0.81304	MR	0.0265	LR	0.54333	MR	32.9	HR
20	0.12125	MR	0.68696	MR	0.0206	LR	0.43333	MR	11.3	HR
21	2.47563	HR	9.4	HR	0.5459	MR	5.76	HR	17.2	HR
22	0.11125	MR	0.63913	MR	0.0195	LR	0.39667	MR	10.6	HR
23	3.18875	HR	12.56087	HR	0.7011	MR	7.01333	HR	11.8	HR
24	n.d	n.d	0.78696	MR	0.0251	LR	0.52	MR	13.1	HR
25	2.3	HR	8.69565	HR	0.498	MR	5.34	HR	16.3	HR
26	0.20063	MR	1.03913	HR	n.d	n.d	0.7	MR	n.d	n.d
27	0.17875	MR	0.95217	MR	0.032	LR	0.64333	MR	14.5	HR
28	1.14063	HR	5.67391	HR	0.3002	MR	3.62333	HR	12.1	HR
29	2.015	HR	8.37826	HR	0.4267	MR	4.84	HR	15.7	HR
30	0.18375	MR	0.98261	MR	0.0342	MR	0.63	MR	14.2	HR

HR- High Risk, MR- Medium Risk, LR- Low Risk, n.d-Not Detected.

Fig. 3.

Ecotoxicological evaluation of (a) norfloxacin, (b) diclofenac, (c) naproxen, (d) triclosan and (e) caffeine across the investigated sampling sites based on the risk quotient (RQ) approach.

The fluoroquinolone antibiotic norfloxacin showed RQ values > 1 at a number of locations, including 4, 6, 12, 17, and 23, indicating high ecological risk, particularly for algae, which are among the most sensitive organisms to this compound. The persistence of fluoroquinolones in aquatic matrices can change the structure of microbial communities, reduce primary productivity, and interfere with nutrient cycling because they inhibit bacterial DNA gyrase. Because of its extreme preference for organic matter and sediments, norfloxacin is more persistent in benthic environments and can cause prolonged sub-lethal toxicity to photosynthetic microorganisms and periphytic algae. Moderate groundwater levels (e.g., sites 3, 7, 15) suggest possible sewage-impacted infiltration, extending its ecological footprint beyond surface waters. The accumulation of fluoroquinolones is significantly associated with the emergence of antibiotic-resistant genes (ARGs) in aquatic biofilms, further increasing long-term ecological risk⁴⁸.

Diclofenac, a widely used NSAID, (RQ > 1 in 22 of 30 sites), indicating high risk particularly to fish species that are sensitive to its endocrine and immunotoxic effects. Even at trace amounts, diclofenac has been shown to cause renal lesions, oxidative stress, and disruption of prostaglandin production in fish. The higher RQ values detected in urban lakes (sites 12, 13, 17, 21, 23, 25) indicate recurrent inputs from household wastewater and hospital effluents. Diclofenac persists due its low biodegradability and partial resistance to traditional wastewater treatment, which raises concerns about long-term bioaccumulation within fish tissues. The consistent high-risk category across regions indicates that diclofenac is a significant driver of ecotoxicity in both cities⁴⁹.

Naproxen, a further NSAID, showed moderate RQ values (0.3–0.7) in most surface water sites and above 0.5 in several high-load locations (sites 4, 11, 12, 21, 23, 29). Although naproxen’s average PNEC (1,000 ng/L for amphibians) is higher than other PPCPs, its widespread detection at higher levels indicates that amphibians and early-life stages of aquatic life may experience chronic exposure. Naproxen’s phototransformation products have been shown to exhibit additional ecological toxicity, particularly under high ultraviolet radiation—conditions prevalent in South Indian urban water bodies. Thus, even moderate RQ values might overlook long-term ecological impacts, particularly around shallow lakes such as Bellandur and Kukkarahalli, where retention periods are high and natural photodegradation produces secondary toxicants⁵⁰.

The majority of surface water sources had high RQ values for triclosan, especially at sites 12, 13, 17, 21, 23, and 25, where RQ values varied between 3 and 9. Because it inhibits the synthesis of fatty acids and interferes with the stability of cell membranes and metabolic processes, triclosan is extremely toxic to aquatic invertebrates. Its hydrophobicity (log Kow 4.76) promotes sorption to sediments, leading to gradual release back into the water column and prolonging the exposure duration. Triclosan is also known to exacerbate endocrine disruption, hinder the development of shells in molluscs, and inhibit algal growth. Elevated triclosan levels in lakes that receive untreated greywater sources indicate substancial contamination from personal care products such as toothpaste and detergents⁵¹.

Caffeine, though frequently regarded as a low-toxicity compound, emerged as the most widespread PPCP with exceptionally high RQ values (RQ > 10 at several sites; PNEC = 1 ng/L). It indicates persistent domestic sewage infiltration in almost all groundwater and surface water locations. Caffeine occurs as a metabolic stimulant in aquatic organisms, resulting in physiological stress, altered aquatic behaviour, and increased oxygen consumption. Its low PNEC reflects the sensitivity of embryonic and early life stages to prolonged exposure. The extremely high RQ values identified at sites 10, 11, 12, 13, 19, 21, 25, and 29 confirm persistent contaminant pressure and indicate caffeine as an efficient indicator component for untreated domestic wastewater discharge⁵².

Overall, the ecological risk assessment shows that the aquatic ecosystems in Bengaluru and Mysuru are under multi-compound chemical stress, with high-risk conditions being driven by diclofenac, triclosan, norfloxacin, and caffeine. The hotspot distributions shown in Figs. 4 and 5 visually illustrate spatial contamination patterns, corresponding strongly with areas impacted by sewage inflow, a significant population density, and inadequate treatment of sewage. These outcomes underscore the need for improved treatment strategies, specific pharmaceutical disposal policies and continuous monitoring of the environment.

Fig. 4.

Spatial distribution of PPCP hotspots in Bengaluru’s urban water bodies based on ecological risk assessment.

Fig. 5.

Spatial distribution of PPCP hotspots in Mysuru’s urban water bodies based on ecological risk assessment.

The sampling occurred during a single dry-season period and was therefore insufficient to account for seasonal variability attributed to monsoon-driven hydrological changes. WWTP influent and effluent were not directly characterised, which limited precise source attribution and treatment performance evaluation and how effectively the treatment performed. The study only looked at aqueous matrices, and therefore did not assess the accumulation of PPCPs in sediments and biota or how they move up the food chain. Although strict QA/QC rules were followed, HPLC–PDA analysis of complex urban water matrices may be subject to co-elutionor matrix interferences. Future research should include multi-season monitoring, characterisation of WWTPs, analysis of sediment and biota, and the application of high-resolution mass spectrometric techniques.

Environmental risks and management implications

PPCPs like norfloxacin, diclofenac, naproxen, triclosan, and caffeine are harmful to aquatic ecosystems due to their continuous release, remain in the environmental persistence, and biological activity. These compounds may negatively impact aquatic life even at low levels once they enter surface water and groundwater systems. Diclofenac, a non-steroidal anti-inflammatory drug, and triclosan are known to cause physiological stress, reproductive problems, and mortality in fish and algae. Antibiotics like norfloxacin can change the structure of microbial communities and promote the proliferation of antibiotic-resistant bacteria. This poses long-term risks to ecosystem integrity and public health. Although naproxen exhibits lower acute toxicity, its frequent use and potential for long-term exposure remain concerning. Caffeine, frequently regarded as a low-toxicity substance, was widely detected and can substantially influence metabolic processes and behaviour in aquatic organisms at elevated concentrations.

Long-term co-exposure to these contaminants may disrupt nutrient cycling, the dynamics of food webs, and the quality of freshwater. This can have a cumulative effect that affects agriculture, animal health, and human populations. The ecological risks identified in this study underscore the inadequacies of traditional wastewater treatment methods, which frequently fail to eliminate both hydrophilic and hydrophobic pharmaceuticals. Biological treatment methods can significantly diminish biodegradable substances like caffeine and naproxen; however, substances with enhanced chemical stability or hydrophobicity, such as diclofenac and norfloxacin, demonstrate reduced removal efficiencies. Triclosan is also a long-term environmental problem because it does not readily degrade and builds up in biosolids, which can harm land-based ecosystems when sludge is reused.

To reduce pollution from drugs, an integrated management approach is required that is based on science and works together. Source control measures, such as proper disposal of unused medications, raising public awareness, and regulating hospital and household waste, are all cost-effective ways to cut down on pharmaceutical inputs at the source. At the same time, advanced treatment technologies like advanced oxidation processes, photocatalysis, and adsorption-based systems have been shown to be effective at removing long-lasting drugs. These technologies may be considered for targeted wastewater treatment upgrades. Regulatory frameworks bolstered by risk-based assessments utilising predicted and measured environmental concentrations can enhance the prioritisation of compounds of concern These findings underscore the need for integrated source control, treatment optimization, regulatory enforcement, and continuous monitoring in order to protect public health and aquatic ecosystems^53–57. The viability of PPCP mitigation strategies has been determined based on compound-specific removal characteristics and their compatibility with current wastewater treatment systems. Compounds that are highly mobile in water and partially biodegradable, like caffeine and naproxen, can be reduced by improving biological treatment processes. However, persistent or hydrophobic compounds, like diclofenac, triclosan, and fluoroquinolones, need tertiary treatment due to their chemical stability and resistance to degradation and interact strongly with sediments and organic matter. Adsorption-based systems and oxidation-driven processes can be used as polishing steps for these compounds. They perform best when used on high-load effluents instead of all treatment pipelines. implementing this strategy at the identified hotspots makes it possible to effectively lower PPCP loads while maintaining operational demands manageable. A risk-prioritized treatment framework that combines biological optimisation with targeted tertiary intervention is technically possible for managing PPCP discharge in urban wastewater systems^58,59.

While heavy metals were not identified in this study, existing evidence from Indian river systems highlights the importance of considering co-occurring contaminants in the assessment of PPCP-related ecological risks^35,60,61.

The high levels of PPCPs identified in Bengaluru and Mysuru indicate that current wastewater management practices have failed and therefore we need more effective yet practical ways of dealing with this issue. Both WWTPs mostly use traditional secondary treatment methods like UASB and activated sludge systems. These technologies effectively remove organic matter, but are less effective at removing persistent pharmaceuticals that remain in the body for a long time. This study does not support the widespread use of expensive new technologies. Instead, it suggests a treatment framework that is based on risk and is affordable.

At the municipal level, optimising current secondary treatment processes—by improving aeration, increasing sludge retention time, and better monitoring of operations—can greatly increase the removal of easily biodegradable compounds like caffeine and naproxen, with little additional investment. In high-risk discharge areas like urban lakes and river stretches that receive untreated sewage or hospital waste, targeted tertiary polishing units are more economical than upgrading the whole system. Activated carbon or inexpensive sorbents that are available in the area, like biochar made from agricultural waste, can effectively get rid of long-lasting compounds like diclofenac and triclosan.

Interventions that are specific to the source are very important for antibiotics like norfloxacin. Decentralised pre-treatment of hospital effluents can significantly reduce pharmaceutical loads entering municipal sewers. In peri-urban Mysuru, land-intensive nature-based solutions, such as constructed wetlands, can offer long-term tertiary treatment. In general, a tiered management strategy that includes process optimisation, treatment focused on hotspots, source control, and nature-based solutions is technically possible and is technically feasible in Indian urban contexts.

Fate and transport of PPCPs in aquatic environment

The physicochemical properties of pharmaceuticals and personal care products, especially the octanol-water partition coefficient (log Kow), have significant effects on their fate and behaviour in aquatic systems. The log Kow determines hydrophobicity, mobility, sorption behaviour, and bioavailability. The present research exhibited a distinct differentiation between compounds with negative log Kow values, such as caffeine (log Kow = − 0.07) and norfloxacin (log Kow = − 0.28), as well as compounds with positive log Kow values, including diclofenac (log Kow = 4.51), naproxen (log Kow = 3.18), and triclosan (log Kow = 4.76). The negative log Kow values of caffeine and norfloxacin indicate high hydrophilicity, indicating that they persist in the dissolved aqueous phase longer, move more freely, and are not attached to sediments as much. This behaviour is consistent with their widespread detection in both surface water and groundwater sites, in addition to their frequent occurrence at elevated concentrations, especially for caffeine, which was found at almost all sampling locations. These compounds are extremely mobile in water, which means that aquatic organisms are always exposed to them. This is why their risk quotient values are always high, even though they exhibit low bioaccumulation potential.

On the other hand, compounds with high positive log Kow values, like triclosan and diclofenac, are very hydrophobic and frequently partition into sediments and organic matter. Sediment sorption limits spatial dispersion but enhances localized persistence, but they tend to build up in one place and stay there for a long time, which means that people who live in benthic and near-shore areas are exposed to them for a long time. The high risk quotient values for triclosan and diclofenac at some surface water sites can be explained by the fact that they can build up in living things and are toxic over time, especially to algae, invertebrates, and fish. Naproxen, which has a log Kow that is in the middle, showed moderate mobility and risk, which is an indication of a balance between how long it stays in water and how strongly it sorbs to sediments and organic matter. In broad terms, combining log Kow values with measured environmental concentrations and ecological risk quotients shows that hydrophilic compounds (negative log Kow) cause widespread, long-term exposure risks, while hydrophobic compounds (positive log Kow) cause severe but spatially localized ecological effects. This correlation highlights the necessity of integrating physicochemical parameters into fate–transport and risk assessment frameworks to achieve a more mechanistic understanding of PPCP behaviour in urban aquatic ecosystems^62,63. The mechanistic interpretation of fate and transport predicated on log Kow values is explicitly shown in the ecological risk quotient results (Table S4). Here, hydrophilic compounds with negative log Kow values (caffeine and norfloxacin) consistently demonstrated elevated risk quotient values due to significant aqueous persistence and bioavailability. Conversely, hydrophobic compounds with positive log Kow values (triclosan and diclofenac) exhibited site-specific yet severe ecological risks related to sediment partitioning and bioaccumulation⁶⁴.

Site-specific physicochemical conditions significantly influence on the fate and persistence of PPCPs in the urban waters that were studied. The broad pH range (6.2–9.3) influences the way compounds develop as well as their mobility. For example, diclofenac and naproxen, which are weakly acidic drugs, are primarily ionised in alkaline conditions, reducing their tendency to sorb to sediments and more probable to remain in water. Very high turbidity levels (145–300 NTU) make it hard for light to penetrate through, which prevents photolytic degradation of photo-sensitive compounds even though southern India experiences high solar irradiance. The elevated organic load, which is demonstrated by COD values of 900 mg/L, may suppress PPCP biodegradation due to preferential microbial utilization of labile organic matter rather than structurally complex PPCPs like fluoroquinolones and triclosan. High levels of suspended solids and organic matter also make it simpler for hydrophobic compounds to stick to things, which helps sediment build up and stay there for a long time, with the possibility of being transported again. In general, these conditions promote PPCP persistence and less likely to change, which explains the higher levels and ecological risks observed in the Bengaluru and Mysuru water bodies^65,66. The presence of PPCPs in groundwater probably indicates that individuals have affected urban areas; however, lacking site-specific hydrogeological analyses, these pathways should be considered indicative rather than definitive.

Conclusions

This research demonstrates the prevalent presence of PPCPs in the urban surface water and groundwater systems of Bengaluru and Mysuru, signifying ongoing anthropogenic contributions from domestic sewage, hospital effluents, and urban runoff. Caffeine, diclofenac, triclosan, and norfloxacin have been identified in concentrations that often exceeded ecological safety thresholds, posing risks across multiple aquatic trophic levels.

Risk quotient analysis identified caffeine and diclofenac as the primary contributors to ecological risk due to their high prevalence and persistence. Additionally, the presence of PPCPs in groundwater underscores limitations in subsurface protection and wastewater management.

This study confirms the widespread occurrence of PPCPs in the water bodies of Bengaluru and Mysuru, underscoring the necessity of immediate action. Risk assessments reveal that ecological risks are linked with the identified PPCPs. Additionally, this study addresses removal inefficiencies, persistence, mobility, and ecological risks while concentrating on the fate and transport of PPCPs in the aquatic environment as well as management and treatment strategies to control PPCP pollution. The results highlight the significance of a multidisciplinary strategy that includes source control, advanced treatment technologies, and sustainable policy frameworks in order to mitigate PPCP contamination and the risks that it poses to human health and ecosystems.

The study provides a thorough ecological risk assessment of PPCPs; however, it does not include a quantitative evaluation of human health risk due to the absence of regulatory guideline standards and exposure parameters. The results highlight the necessity for future multidisciplinary initiatives that combine environmental monitoring with treatment optimisation, regulatory evaluation, and human exposure assessment. Further study ought to combine drinking water exposure modelling, food-chain bioaccumulation, and wellness standards.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

Authors’ contributions**Divya Vinod** : Writing original draft, Data collection, Methodology, Investigation**B D Vishwajit, J Divya, Satisha AD, Prashant Basavaraj Bhagawati, Baburao Gaddala** : Data curation, sampling, analysis, visualisation.**Kotermane Mallikarjunappa Anilkumar** , **Harini R** : Resource and software**Mohammad Khalid, Shadma Wahab** : Resource, proof reading**Shivaraju Harikaranahalli Puttaiah** : Conceptualisation, methodology, review and editing, resources, supervision.

Funding

Open access funding provided by JSS Academy of Higher Education and Research, Mysore. This study was not funded by any specific grants from funding agencies.

Data availability

All data generated or analysed during this study are included in this published article.

Declarations

Competing interests

The authors declare no competing interests.