Spatio-temporal dynamics of total mercury contamination in diving birds as bioindicators of tropical wetlands in India Authors: K. Fathima Shadiya, Vlatko Galić, Dora Bjedov, M. Anarkkaly, Jasmine Anand, Jorge Bernal-Alviz, Sabir Bin Muzaffar, K. M. Aarif Published: 2026 ABSTRACT The present study evaluates the spatio-temporal variation and magnitude of total mercury (THg) contamination in two diving bird species, the Little Cormorant (Microcarbo niger) and the Oriental Darter (Anhinga melanogaster), across diverse wetland habitats in India. Sampling was conducted seasonally from 2022 to 2024 across six sites: freshwater agricultural lands, saline agricultural lands, and both freshwater and coastal mangrove ecosystems. Bird droppings (n = 72) were collected using non-invasive methods, followed by acid digestion and atomic absorption spectrophotometry to quantify THg concentrations. Using a full-factorial ANOVA, we found strong main effects of year, species, and location, alongside significant interactions, suggesting a complex interplay of factors influencing Hg exposure in piscivorous birds. Coastal wetlands, notably Puthuvype and Ulavaipu, showed elevated THg concentrations and an increasing temporal trend, in both piscivorous species, with higher exposure in Little Cormorants than Oriental Darters, reaching up to 48.79 mg/kg and 36.72 mg/kg in droppings, respectively, by 2024. In contrast, inland sites like Maramangalam showed minimal contamination, likely due to reduced anthropogenic inputs. Temporal trends suggest a nearly three-fold rise in THg over three years, potentially associated with anthropogenic inputs and environmental changes in Hg cycling. This significant difference in mercury levels could be associated with species-specific foraging strategies and habitat use. Our findings demonstrate the utility of bird droppings as a non-invasive indicator of recent, diet-derived THg exposure, revealing pronounced site-specific contamination patterns that may pose neurotoxic risks to avian health and reproduction. These results emphasize the urgent need for targeted monitoring and mitigation strategies to address Hg contamination in rapidly urbanising tropical coastal ecosystems, safeguarding biodiversity and ecosystem integrity.Supplementary InformationThe online version contains supplementary material available at 10.1038/s41598-026-55158-9. Introduction Mercury (Hg) contamination represents a critical environmental and public health issue, largely due to its extreme toxicity and its persistence within ecosystems through bioaccumulation and biomagnification. As a globally pervasive contaminant, Hg ranks third on the U.S. Agency for Toxic Substances and Disease Registry’s list of hazardous substances1. It is a potent neurotoxin, with well-documented deleterious effects on both human and ecological health1. Anthropogenic sources are diverse and include agricultural runoff (e.g., pesticides and fertilizers), fossil fuel combustion, mining activities, and industrial or municipal discharges2,3. Additionally, improper disposal of Hg-containing products exacerbates soil and water contamination. Notably, Hg exhibits long-range atmospheric and hydrological transport, enabling its deposition in remote ecosystems far from emission sources4. These anthropogenic inputs introduce organic matter and sulfates that stimulate microbial activity, promoting the conversion of inorganic mercury into its more toxic form, methylmercury5,6 which may enter fishes and then to piscivorous birds through trophic transfer7–9. Avian species are especially valuable as bioindicators of Hg exposure due to their high trophic positions and sensitivity to contaminant levels10. In this context, our findings highlight that diving birds such as Little Cormorant (Microcarbo niger) and Oriental Darter (Anhinga melanogaster) are at heightened risk of Hg exposure, attributable to their fish-based diets and foraging behaviours. These species may serve as effective sentinels for monitoring Hg dynamics in freshwater ecosystems. The Little Cormorant and the Oriental Darter are widely distributed piscivorous diving birds widely distributed across South and Southeast Asia11 and serve as effective biomonitors of mercury contamination. Although both species are primarily exposed through consumption of contaminated fish, differences in foraging strategy and habitat use may result in species-specific Hg uptake. Increasing Hg concentrations was reported in piscivorous bird species12–14, and mangrove ecosystems15,16. This pattern is likely attributable to elevated THg burdens in mangrove-associated fish, which constitute the primary diet of these waterbirds. Mercury poses severe health risks to both avifauna and humans17,18. In waterbirds, chronic Hg exposure causes neurotoxicity and reproductive impairment, leading to reduced survival and population viability. Neurological effects such as ataxia impair essential behaviours including foraging and predator avoidance19–23 while reproductive impacts include reduced egg production, eggshell thinning, and lowered hatching success24,25. These effects are of particular conservation concern for the Oriental Darter, listed as Near Threatened on the IUCN Red List (BirdLife International26, as rising Hg levels may exacerbate existing threats and accelerate population declines. Sublethal physiological effects, including disruption of muscle and endocrine function, may further reduce fitness in both species27–29. Analysis of waterbird droppings is a widely accepted, non-invasive approach for assessing trophic transfer and bioconcentration of mercury within ecosystems4,30,31. Droppings reflect short-term, diet-derived mercury exposure and provide real-time information directly linked to active foraging habitats. In contrast, mercury concentrations in feathers, eggs, and blood integrate exposure over longer or physiologically constrained periods and may reflect mercury uptake from different locations or life stages, reducing spatial and temporal resolution32,33. While droppings primarily represent recent exposure, their non-invasive nature and strong linkage to local foraging make them particularly suitable for site-specific biomonitoring. Despite increasing concern over mercury contamination in tropical wetlands, spatio-temporal patterns of total mercury (THg) exposure in piscivorous waterbirds, particularly using non-invasive indicators, remain poorly documented in India. Moreover, comparative assessments of species-specific mercury exposure across contrasting wetland types, including inland and coastal mangrove systems, are largely lacking. A limited number of previous studies from this region have documented mercury contamination in waterbirds using non-invasive matrices4,30, but these have largely focused on wintering migrant shorebird species which primarily feeds on benthic macroinvertebrates. In contrast, the present study adopts a comparative framework across two piscivorous diving bird species with similar trophic roles but differing ecological traits, and spans a broader habitat gradient encompassing both inland and coastal wetlands. To address these gaps, the present study assessed spatial and temporal variations in THg contamination across multiple wetland habitats in Kerala, India, and evaluated species-specific exposure patterns in two widely distributed diving birds, the Little Cormorant and the Oriental Darter. By integrating species-level comparisons with multi-habitat assessment over time, this study provides a more comprehensive understanding of spatial and temporal variation in THg exposure and strengthens the use of waterbirds as bioindicators in tropical wetland ecosystems. Using droppings collected over three years (2022–2024) from six ecologically distinct freshwater, saline agro-ecosystem, and mangrove sites, we aimed to (I) quantify habitat-specific THg concentrations, (II) compare recent THg exposure between species, and (III) assess temporal trends. Site-level statistical comparisons were used to identify variation among locations, providing baseline data for tropical wetland biomonitoring. Materials and methods Study area Our study was conducted across six selected sites: Maramangalam freshwater agricultural lands, Nalupurakkal Atholi saline water agricultural lands, Kallampara inland mangroves, Ulavaipu inland mangroves, Puthuvype coastal mangroves, and Changaram coastal mangroves (Fig. 1; For more information about the study area, please refer to the Supplementary Information 1). Fig. 1Map illustrating the study area, where research was conducted between 2022 and 2024 across six different habitat types: freshwater agricultural lands, saline agricultural lands, and both saline and freshwater mangroves. The study focused on two species of diving birds, the Little Cormorant (M. niger) and the Oriental Darter (A. melanogaster), with the latter classified as Near Threatened by the IUCN. The map was prepared using ArcGIS Pro version 3.1.0 (Esri, Redlands, CA, USA;) with satellite imagery obtained from the basemap layer available within the software. Map illustrating the study area, where research was conducted between 2022 and 2024 across six different habitat types: freshwater agricultural lands, saline agricultural lands, and both saline and freshwater mangroves. The study focused on two species of diving birds, the Little Cormorant (M. niger) and the Oriental Darter (A. melanogaster), with the latter classified as Near Threatened by the IUCN. The map was prepared using ArcGIS Pro version 3.1.0 (Esri, Redlands, CA, USA;) with satellite imagery obtained from the basemap layer available within the software. Species account and sample collection The study focused on two piscivorous diving bird species, the Little Cormorant (Microcarbo niger), listed as Least Concern by the IUCN, and the Oriental Darter (Anhinga melanogaster), categorized as Near Threatened. Both species are shallow water divers that forage primarily in nearshore and inland wetlands and subsequently roost on exposed substrates such as mudbanks, embankments, and low-lying terrestrial edges adjacent to water bodies. These predictable roosting behaviours provided suitable opportunities for non-invasive sample collection. Given that both species may occur in mixed assemblages, particular care was taken to ensure accurate species-level assignment of droppings. Field observations were conducted using high resolution cameras and spotting scopes, supported by trained field assistants. Sampling was restricted to periods when these birds were actively roosting on land following foraging bouts. Droppings were collected only when the species identity of the individual or the immediate roosting group could be confidently established based on direct visual confirmation and spatial association. In cases where these birds were roosting in close proximity, samples were collected immediately after deposition and linked to the observed individual or clearly separated monospecific clusters. Any samples that could not be assigned with confidence were excluded from further analysis. This approach minimized the risk of misclassification in mixed-species roosting environments. Fresh droppings were collected aseptically using sterile spatulas and transferred into sterilized glass vials. Samples were immediately stored in airtight ice boxes for transport to the laboratory4,34–36. Upon arrival, all samples were preserved at − 40 °C until further analysis. Sampling was conducted across six sites during 2022, 2023, and 2024, covering four sampling months (January, March, October, and December) to capture seasonal variation. During each sampling period, three replicate droppings per species were collected at each site, ensuring consistent representation across species, locations, and years. Only freshly deposited droppings, identified based on moisture content, texture, and absence of environmental contamination, were included in the analysis. Laboratory processing for THg analysis The fresh droppings (including three replicates) were weighed (wet weight, ww) using an analytical balance (Mettler Toledo ME204E Series, Malaysia) and aseptically transferred into tetrafluoromethoxyl polymer bomb vessels. Each vessel was treated with 9 mL of nitric acid (Merck). Samples were then subjected to high-pressure, high-temperature microwave-assisted digestion (220 ᵒC, 30 min at 60–100 bar) using a Titan MPS system (Perkin-Elmer, USA). Following digestion, the digests were diluted to a final volume of 20 mL with acidified deionized water. Total Hg concentrations in the digested samples were quantified using an Atomic Absorption Spectrophotometer (PinnAAcle 500, Perkin-Elmer, USA) by Cold Vapour Atomic Absorption technique operated at a wavelength of 253.65 nm, as described by Rajaram et al.37. To enhance analytical sensitivity and lower the detection limits, an Electrodeless Discharge Lamp (EDL) was utilized as light source, providing optimized spectral output for accurate mercury determination4,38. Calibration curves were established using five Hg standard solutions (0.5, 1, 5, 10, and 20 µg/L) prepared from Perkin-Elmer stock standards. Each calibration point was run in triplicate, and the linearity of the calibration curve was verified (R² ≥ 0.999). Calibration was checked every 10 samples, and re-calibration was performed if deviation exceeded ± 5%. Quality analysis and quality control Quality assurance and quality control (QA/QC) procedures were implemented throughout all analytical steps to ensure data reliability and instrument stability (Supplementary table S1). These included the analysis of procedural blanks, replicate measurements, calibration verification, and spike recovery experiments. Due to the unavailability of certified reference materials specifically matching the bird droppings matrix, analytical accuracy was assessed using spike recovery tests. Spike recoveries were performed by adding known concentrations of a certified mercury standard solution to homogenized droppings samples prior to digestion. Spikes were applied at concentration levels comparable to the range observed in environmental samples. After digestion and analysis, recoveries were calculated as the percentage ratio between measured and expected mercury concentrations, following subtraction of background THg levels in unspiked samples. Mean recovery across all spiked samples was 97%, indicating good instrumental performance and consistency of the analytical procedure. While this approach follows commonly applied practices in environmental mercury analysis, spike recovery experiments do not fully account for potential matrix effects associated with complex biological materials such as bird droppings. Consequently, the absence of a matrix-matched certified reference material represents a limitation in the evaluation of absolute analytical accuracy. Blank samples were analysed at regular intervals to assess background contamination, and all glassware was acid-cleaned prior to use. The method detection limit for THg was 0.001 mg/kg. Challenges and limitations of the study The study was conducted across coastal and freshwater mangroves and agricultural lands, where both species commonly roosted on mangrove branches near land. Collection of fresh droppings was challenging because birds often roosted close to or over water, leading to sample loss through dispersion or dissolution. To minimise this, sampling was restricted to regularly used short-term roosting sites located as far from water as possible, where clean polythene sheets were placed for 2–3 h. Only freshly deposited droppings were collected using a steel spatula and transferred to sterilised glass vials; samples of insufficient volume were discarded. Individual birds could not be marked, creating a small risk of resampling the same individuals. This was mitigated by collecting droppings immediately after birds had moved from actively foraging areas, limiting sampling duration, sampling across multiple sites, and distinguishing fresh samples based on appearance, texture, and moisture. Given the number of birds present, any residual duplication is unlikely to have biased overall THg patterns. An additional limitation was the reliance on spike recovery experiments in the absence of certified reference materials specific to bird droppings. While this approach ensured analytical consistency, matrix effects associated with heterogeneous organic content and variable moisture may influence absolute THg quantification. Therefore, caution is required when comparing concentrations across studies or matrices. Nevertheless, the applied QA/QC framework supports the internal consistency of the dataset, and the observed spatial and temporal trends remain robust for evaluating relative mercury exposure across species, sites, and years. Statistical analysis Analyses were conducted using Python 3.11 with the pandas, statsmodels, scikit-learn, matplotlib, and seaborn libraries to process and visualize data. The dataset comprised variables including Year, Species, Location, Total Mercury (THg) concentration, Latitude (LAT), and Longitude (LON). Sampling month was included as a categorical temporal covariate (January, March, October, December; additionally grouped into seasonal windows for sensitivity checks) to evaluate whether within-year sampling timing influenced estimated THg levels. A small number of observations were at/near the detection limit. To retain these data without inducing undefined log values, non-detect/zero entries were handled using a standard substitution approach (LOD/2), and sensitivity analyses were performed to confirm that qualitative conclusions were unchanged. This treatment follows established guidance for non-detects in environmental monitoring data39. Descriptive statistics were calculated to summarize THg concentrations by species, location, and year. A full-factorial ANOVA was employed to evaluate the effects of Year, Species, and Location, along with their interactions, on THg concentrations, with Tukey’s Honest Significant Difference (HSD) test used for post-hoc comparisons among species. Prior to interpretation of ANOVA outputs, we assessed model assumptions using residual diagnostics (normality and homoscedasticity checks). Because THg concentrations were right-skewed and included near-zero values, we additionally fitted generalized linear models (GLMs) with a Gamma error distribution and log-link, which constrain predictions to positive values and are more suitable for concentration data. Inference focused on consistency of effects across model formulations (ANOVA/OLS and GLM). Model conclusions were consistent between ANOVA and Gamma log-link GLM formulations: Location remained the dominant predictor, Year effects persisted, and temporal amplification at coastal sites was retained. The GLM framework ensured strictly positive fitted concentrations and improved interpretability for right-skewed THg data. Linear regression was applied to model temporal THg trends across the study period and by subgroup. F-statistics indicate the relative contribution of factors within the fitted model but do not isolate independent causal pathways; therefore, effects are interpreted jointly with interaction structure and ecological context. Results Total mercury concentrations showed clear spatial and temporal variation across wetlands and between species from 2022 to 2024. Coastal habitats consistently exhibited higher THg levels and stronger increasing trends than inland sites, with the Little Cormorant generally showing higher exposure than the Oriental Darter. Tables 1 and 2 show mean THg values with standard deviations, LSD results, and slope values for each location and year from 2022 to 2024. For Little Cormorant (Table 1), coastal sites such as Puthuvype exhibited the highest THg concentrations, with a slope of 15.435, indicating an increase over time. Similarly, Ulavaipu showed an increasing pattern with a slope of 11.733. In contrast, inland sites such as Maramangalam showed minimal change, with a slope of 0.384, indicating relatively stable and low THg levels. For Oriental Darter (Table 2), Puthuvype also showed increasing THg concentrations, with a slope of 10.694, while Maramangalam remained low, with a slope of 0.318. These LSD tests confirmed significant within-species variation, with coastal habitats consistently showing higher THg levels. Slope values indicate increasing THg concentrations across most locations, with higher rates of increase at coastal sites. These patterns are descriptive, while statistical inference is supported by the ANOVA and GLM analyses. Table 1Mean values ± standard deviation of THg in droppings (mg/kg ww) of Little Cormorant over the six screened locations. Slopes represent descriptive annual changes in THg concentration (mg/kg/year) estimated from linear regression across years within each location. Different letters in the THg column denote significant differences between years within that specific location, as determined by Tukey’s Honest Significant Difference (HSD) post-hoc test (α = 0.05).LocationYearTHgSlopeChangaram202211.89 ± 1.28 b9.115202318.64 ± 2.76 c202430.12 ± 3.33 dKallampara202213.41 ± 2.34 b10.057202319.25 ± 2.37 c202433.52 ± 5.07 eMaramangalam20220.08 ± 0.14 a0.38420230.50 ± 0.12 a20240.85 ± 0.07 aNalupurakkal Atholi20220.62 ± 0.27 a1.05520231.68 ± 0.20 a20242.73 ± 0.34 aPuthuvype202217.92 ± 2.68 c15.435202330.62 ± 3.74 d202448.79 ± 4.52 fUlavaipu202211.79 ± 1.31 b11.733202319.07 ± 3.2 c202435.25 ± 5.3 e Table 1Mean values ± standard deviation of THg in droppings (mg/kg ww) of Little Cormorant over the six screened locations. Slopes represent descriptive annual changes in THg concentration (mg/kg/year) estimated from linear regression across years within each location. Different letters in the THg column denote significant differences between years within that specific location, as determined by Tukey’s Honest Significant Difference (HSD) post-hoc test (α = 0.05).LocationYearTHgSlopeChangaram202211.89 ± 1.28 b9.115202318.64 ± 2.76 c202430.12 ± 3.33 dKallampara202213.41 ± 2.34 b10.057202319.25 ± 2.37 c202433.52 ± 5.07 eMaramangalam20220.08 ± 0.14 a0.38420230.50 ± 0.12 a20240.85 ± 0.07 aNalupurakkal Atholi20220.62 ± 0.27 a1.05520231.68 ± 0.20 a20242.73 ± 0.34 aPuthuvype202217.92 ± 2.68 c15.435202330.62 ± 3.74 d202448.79 ± 4.52 fUlavaipu202211.79 ± 1.31 b11.733202319.07 ± 3.2 c202435.25 ± 5.3 e Mean values ± standard deviation of THg in droppings (mg/kg ww) of Little Cormorant over the six screened locations. Slopes represent descriptive annual changes in THg concentration (mg/kg/year) estimated from linear regression across years within each location. Different letters in the THg column denote significant differences between years within that specific location, as determined by Tukey’s Honest Significant Difference (HSD) post-hoc test (α = 0.05). Table 2Mean values ± standard deviation of THg in droppings (mg/kg ww) of Oriental Darter over the six screened locations. Slopes represent descriptive annual changes in THg concentration (mg/kg/year) estimated from linear regression across years within each location. Different letters in the THg column denote significant differences between years within that specific location, as determined by Tukey’s Honest Significant Difference (HSD) post-hoc test (α = 0.05).LocationYearTHgSlopeChangaram202213.64 ± 2.80 d9.583202326.07 ± 3.05 g202432.81 ± 1.94 hKallampara202214.12 ± 1.63 d9.331202319.31 ± 1.81 e202432.78 ± 5.57 hMaramangalam20220.18 ± 0.05 a0.31820230.40 ± 0.10 a20240.81 ± 0.12 aNalupurakkal Atholi20220.73 ± 0.18 a1.44120231.43 ± 0.28 a20243.61 ± 1.03 bPuthuvype202215.33 ± 3.49 d10.694202323.81 ± 2.41 f202436.72 ± 4.93 iUlavaipu202210.81 ± 0.89 c8.120202314.84 ± 2.44 d202427.05 ± 4.40 g Table 2Mean values ± standard deviation of THg in droppings (mg/kg ww) of Oriental Darter over the six screened locations. Slopes represent descriptive annual changes in THg concentration (mg/kg/year) estimated from linear regression across years within each location. Different letters in the THg column denote significant differences between years within that specific location, as determined by Tukey’s Honest Significant Difference (HSD) post-hoc test (α = 0.05).LocationYearTHgSlopeChangaram202213.64 ± 2.80 d9.583202326.07 ± 3.05 g202432.81 ± 1.94 hKallampara202214.12 ± 1.63 d9.331202319.31 ± 1.81 e202432.78 ± 5.57 hMaramangalam20220.18 ± 0.05 a0.31820230.40 ± 0.10 a20240.81 ± 0.12 aNalupurakkal Atholi20220.73 ± 0.18 a1.44120231.43 ± 0.28 a20243.61 ± 1.03 bPuthuvype202215.33 ± 3.49 d10.694202323.81 ± 2.41 f202436.72 ± 4.93 iUlavaipu202210.81 ± 0.89 c8.120202314.84 ± 2.44 d202427.05 ± 4.40 g Mean values ± standard deviation of THg in droppings (mg/kg ww) of Oriental Darter over the six screened locations. Slopes represent descriptive annual changes in THg concentration (mg/kg/year) estimated from linear regression across years within each location. Different letters in the THg column denote significant differences between years within that specific location, as determined by Tukey’s Honest Significant Difference (HSD) post-hoc test (α = 0.05). Linear regression was used to describe temporal changes in THg concentrations across locations between 2022 and 2024. The results indicate increasing THg levels at most sites, with higher rates of increase observed in coastal locations such as Puthuvype and Ulavaipu compared to inland sites like Maramangalam. These slope values represent descriptive indicators of temporal change (mg/kg/year), while statistical inference is supported by the ANOVA and GLM analyses. The ANOVA output confirmed that Year, Species, and Location each significantly influenced THg concentrations, and strong interactions were found among these factors. Table 3 shows the full-factorial ANOVA summary, including the main effects and interactions. The full-factorial model (C(Year) * C(Species) * C(Location)) revealed that Location had an especially high F-value, indicating that some locations exhibit notably higher contamination. Species and Year also interacted significantly, supporting the conclusion that THg toxicity is increasing over time and that different species accumulate THg at different rates. Table 3Results of the full-factorial three-way ANOVA testing the effects of Year, Species, and Location, and all possible two-way and three-way interactions, on the response variable. The table reports degrees of freedom (df), sum of squares (SS), mean square (MS), F-statistics (F), and associated p-values (PR(> F)). Main effects indicate overall differences among years, between species, and among locations, while interaction terms evaluate whether species-specific responses vary across years and locations. The significant three-way interaction (Year × Species × Location) indicates that temporal patterns differ among species depending on location. Residuals represent unexplained variation after accounting for all modelled factors. Statistical significance was assessed at α = 0.05.TermdfSum of SquaresMean SquareFPR(> F)Year215558.237779.111051.300.0000000Species1165.57165.5722.380.0000031Location550245.3110049.061358.070.0000000Year: Species2156.1078.0510.550.0000344Year: Location106823.07682.3192.210.0000000Species: Location51398.14279.6337.790.0000000Year: Species: Location10393.4739.355.320.0000002Residual3962930.217.40-- Table 3Results of the full-factorial three-way ANOVA testing the effects of Year, Species, and Location, and all possible two-way and three-way interactions, on the response variable. The table reports degrees of freedom (df), sum of squares (SS), mean square (MS), F-statistics (F), and associated p-values (PR(> F)). Main effects indicate overall differences among years, between species, and among locations, while interaction terms evaluate whether species-specific responses vary across years and locations. The significant three-way interaction (Year × Species × Location) indicates that temporal patterns differ among species depending on location. Residuals represent unexplained variation after accounting for all modelled factors. Statistical significance was assessed at α = 0.05.TermdfSum of SquaresMean SquareFPR(> F)Year215558.237779.111051.300.0000000Species1165.57165.5722.380.0000031Location550245.3110049.061358.070.0000000Year: Species2156.1078.0510.550.0000344Year: Location106823.07682.3192.210.0000000Species: Location51398.14279.6337.790.0000000Year: Species: Location10393.4739.355.320.0000002Residual3962930.217.40-- Results of the full-factorial three-way ANOVA testing the effects of Year, Species, and Location, and all possible two-way and three-way interactions, on the response variable. The table reports degrees of freedom (df), sum of squares (SS), mean square (MS), F-statistics (F), and associated p-values (PR(> F)). Main effects indicate overall differences among years, between species, and among locations, while interaction terms evaluate whether species-specific responses vary across years and locations. The significant three-way interaction (Year × Species × Location) indicates that temporal patterns differ among species depending on location. Residuals represent unexplained variation after accounting for all modelled factors. Statistical significance was assessed at α = 0.05. Linear regressions across 2022–2024 indicated an upward THg trend at nearly all locations, with slopes ranging from 0.318 (Maramangalam, Oriental Darter) to 15.435 (Puthuvype, Little Cormorant) (Supplementary Figs. 1 & 2). These slope values represent descriptive indicators of temporal change (mg/kg/year) and should not be interpreted as inferential effect estimates. Fig. 2Comparison of mean total mercury (THg; mg/kg ww) concentrations between 2022 and 2024 across sampling locations, shown separately for Little Cormorant and Oriental Darter. For each location, points represent location-wise mean THg in 2022 (blue) and 2024 (red), and connecting lines indicate direction and magnitude of change over time. Locations are ordered by 2024 mean concentration within each species to improve readability. . Comparison of mean total mercury (THg; mg/kg ww) concentrations between 2022 and 2024 across sampling locations, shown separately for Little Cormorant and Oriental Darter. For each location, points represent location-wise mean THg in 2022 (blue) and 2024 (red), and connecting lines indicate direction and magnitude of change over time. Locations are ordered by 2024 mean concentration within each species to improve readability. . Figure 2 shows THg distributions for each species and across years, revealing that coastal habitats displayed notably higher average THg in some areas, consistent with the location-based ANOVA results. Figure 2 further elucidates the temporal dynamics of THg concentrations, presenting mean THg levels with standard error bars across six locations (Maramangalam, Nalupurakkal Atholi, Kallampara, Ulavaipu, Puthuvype, Changaram) from 2022 to 2024 for both Little Cormorant and Oriental Darter. The plot reveals a consistent upward trend in THg concentrations over the study period, with coastal locations such as Puthuvype and Ulavaipu exhibiting the most pronounced increases, reaching the highest levels by 2024. This aligns with the steep slopes observed in Tables 2 and 3 (e.g., 15.435 for Little Cormorant at Puthuvype and 11.733 at Ulavaipu), reflecting the significant Year effect (F = 1051.30, p < 0.0001) and Year: Location interaction (F = 92.21, p < 0.0001) from the ANOVA. Inland sites like Maramangalam and Nalupurakkal Atholi show minimal increases, remaining below 5 mg/kg, consistent with their low slopes (0.384 and 1.055 for Little Cormorant, respectively) and the Location effect (F = 1358.07, p < 0.0001). The differentiation by Species (solid lines for Little Cormorant, dashed lines for Oriental Darter) highlights a slightly higher proxy of contamination in Little Cormorant at coastal sites, supporting the Species: Location interaction (F = 37.79, p < 0.0001). These temporal trends corroborate the spatial patterns in Fig. 2, where coastal mangroves consistently show elevated THg levels, likely due to higher Hg availability in fish prey, as suggested by the ecological context (Supplementary Figs. 1 & 2). Discussion This study demonstrates that THg concentrations reflected in diving bird droppings are significantly shaped by temporal, spatial, and biological factors, with notable interactions (Fig. 3). The full-factorial ANOVA revealed strong main effects of Year, Species, and Location, as well as significant interactions, highlighting the complex dynamics associated with Hg exposure in avian top predators. These findings align with global concerns about escalating Hg contamination in coastal ecosystems, particularly in rapidly urbanizing tropical and subtropical regions30,40,41. Fig. 3Visual depiction of study outcomes showing that THg concentrations in the droppings of Little Cormorant and globally threatened Oriental Darter were lower in inland wetlands compared to coastal wetlands. For both species, THg concentrations in droppings from all sites showed a marked increase over the study years. Visual depiction of study outcomes showing that THg concentrations in the droppings of Little Cormorant and globally threatened Oriental Darter were lower in inland wetlands compared to coastal wetlands. For both species, THg concentrations in droppings from all sites showed a marked increase over the study years. Spatio-temporal variation in THg levels A key limitation of this study is the low replicate number per species-location-time cell (n = 3), which reduces statistical precision and may underrepresent intra-specific variability linked to age, sex, physiological state, and individual foraging behaviour. Accordingly, species-level contrasts and short-term temporal differences should be interpreted as relative patterns rather than definitive population-level estimates. The spatio-temporal analysis revealed that variation in THg exposure across the studied wetlands is primarily structured by location and habitat type, with coastal systems consistently exhibiting higher contamination than inland sites. These spatial contrasts were accompanied by a clear temporal signal, with THg levels increasing over the three-year study period, particularly in coastal habitats. Together, these patterns indicate that mercury contamination in wetlands of west coast of Kerala is not static but is characterized by both persistent spatial heterogeneity and ongoing temporal intensification, underscoring the dynamic nature of mercury exposure in these ecosystems. Because birds were not individually marked, residual dependence among samples collected close in time is possible, and uncertainty may be underestimated in such clusters. A similar trend of THg contamination was also documented in the fresh droppings of Kentish Plover (Charadrius alexandrinus), a small shorebird species, within the Kadalundi-Vallikkunnu Community Reserve (ranging between ~ 26 and ~ 65 mg/kg in mudflats and ~ 46 and ~ 73 mg/kg in mangroves), a vital coastal habitat for waterbirds, over a six-year monitoring period3042,43,. In addition to this species, total mercury concentrations were also reported in ten other wintering shorebird species across three distinct habitats (highest mean concentration of Hg reported from mudflats, mangroves, and sand beaches were approximately 41, 81 and 59 mg/kg, respectively) within KVCR over multiple years4. These consistent patterns of mercury contamination are particularly concerning as it bioaccumulates and biomagnifies in aquatic food webs, leading to higher concentrations at higher trophic levels9,44. Piscivorous birds, as apex predators, are especially vulnerable to such accumulation. In contrast, inland sites such as Maramangalam displayed minimal or no increase in THg levels, suggesting a lower current impact from anthropogenic Hg sources (Fig. 2). These spatial patterns may be associated with multiple anthropogenic inputs, including industrial discharge, shipping activity, and land-based runoff, which have been identified as important sources of mercury in coastal and estuarine systems6,45. However, these potential drivers were not directly measured in the present study, and therefore their role should be considered as hypothetical and requiring further validation through targeted environmental and trophic-level assessments. Future studies should incorporate site-specific measurements, including prey fish THg concentrations, sediment methylmercury profiles, and detailed environmental variables, to better evaluate the relative contribution of these potential drivers. Though there were no point-sources, increased concentrations of heavy metals were reported in the faeces of wintering shorebirds along the west coast of India, particularly KVCR and surrounding habitats4,35. Conversely, increased concentrations of Hg were reported in the inshore foraging marine birds than the offshore foraging species14. Even though, mudflats and mangroves of KVCR were considered as key natural foraging habitats for wintering shorebirds along the west coast of India43,46,47, heavy metal contamination has been proposed to be one among several causes for the decline in shorebird population and abundance in the natural habitats, forcing them to alter their habitat use35,48. These previous studies reported from the nearby geographical area of the present study points to the critical situation of habitat degradation associated with heightened heavy metal contamination which may also affect the prey population and assemblages of these waterbirds31,34,35,43. Apart from the heavy metal contamination31, environmental changes and changes in sediment and water properties was also identified to be associated with the declines in waterbird populations and abundance49,50. Thus, the present study sheds light to the urgent need for conservation measures to preserve the natural habitats. The significant Year × Species interaction further indicates that THg in the droppings differs among species, likely reflecting variations in trophic level, foraging strategy, and habitat use9,10. Across all shared locations, Little Cormorants consistently exhibited higher THg levels than Oriental Darters, possibly associated with their preference for nearshore, estuarine fish that are more prone to Hg contamination, especially under low-oxygen, nutrient-rich conditions conducive to Hg methylation51. This pattern is consistent with earlier findings in tropical piscivorous birds such as the Neotropic Cormorant and Roseate Spoonbill, which also show elevated Hg in tissues compared to herbivorous or omnivorous species19. Temporal analysis supports these spatial trends, revealing a consistent rise in THg levels from 2022 to 2024 across most sites. Coastal locations saw nearly a threefold increase during this period, a pattern that may be associated with escalating industrial activities, shifts in land use, and climate-driven changes in the mercury cycling, such as enhanced methylation under warmer and wetter conditions, as reported in previous studies6,52, although such processes were not directly evaluated in the present study. Site-based comparisons and regression analyses further reinforce these findings, confirming that mercury exposure is both species-specific and habitat-dependent. These patterns are derived strictly from observed data across sampled locations, without extrapolation beyond the study sites. Notably, Fig. 2 illustrates the strong coastal dominance in Hg contamination, with Little Cormorant at Puthuvype reaching nearly 49 mg/kg in 2024, a concentration that far exceeds established ecological safety thresholds for piscivorous birds53. Biological implications The THg concentrations recorded in this study were consistently high, particularly at coastal wetland sites, and therefore have important biological and conservation implications for piscivorous waterbirds. Species differences were context-dependent and in some locations relatively small, potentially reflecting overlap in prey resources and foraging zones between the two piscivorous taxa. Diving bird droppings serve as a non-invasive, short-term proxy for THg exposure, primarily reflecting recent dietary intake54. In this context, our study provides robust evidence of significant spatial and temporal variation in THg concentrations in the droppings of the Little Cormorant across six wetland sites in India. Linear regression analyses revealed consistent annual increases in THg at most locations, with high model fit, indicating persistent and site-specific contamination. Particularly high THg concentrations were observed at Puthuvype (48.79 ± 4.52 mg/kg) and Ulavaipu (35.25 ± 5.30 mg/kg), indicating substantial mercury exposure in these coastal environments. When compared with existing literature, these values appear markedly elevated. For example, faecal THg concentrations in Little Penguins (Eudyptula minor) from St. Kilda ranged from 0.14 to 0.31 mg/kg, with blood concentrations reaching up to 4.20 mg/kg55. Similarly, studies on cormorants and other piscivorous birds from coastal and estuarine systems generally report values within low to moderate ranges56reported 14.56 mg/kg,57;. Recent studies from southwestern India have also highlighted environmental degradation and population declines in waterbirds associated with habitat alteration and contamination pressures4,30,34,35,49,58. In this context, the THg concentrations observed at these sites suggest that these values may indicate elevated local contamination associated with industrial and estuarine inputs. However, the magnitude of these concentrations exceeds most, although limited, reported values and therefore warrants further investigation and validation across complementary biological matrices. In comparison, Eeva et al.54, assessed a broader suite of trace elements, including arsenic (As), cadmium (Cd), cobalt (Co), copper (Cu), nickel (Ni), and lead (Pb) in urate and faeces of European Pied Flycatcher (Ficedula hypoleuca) nestlings in a temperate environment. While their reported geometric means in the faecal fraction ranged from 2.86 mg/kg (Cd) to 382 mg/kg (Cu), THg concentrations in our study, analysed in whole droppings, were not only consistently high but also more narrowly focused on a contaminant of high toxicological concern. Moreover, unlike Eeva et al., who found that only a small proportion of metal variation was explained by faecal mass proportion, our models demonstrated highly predictive trends in THg concentration over time. This pattern may be associated with stable foraging behaviour of Little Cormorants and potential localised anthropogenic inputs. The persistently elevated THg levels found in Little Cormorant and Oriental Darter droppings, especially at Puthuvype and Kallampara, suggest persistent site-specific exposure patterns. A similar study conducted at Kallampara reported the presence of heavy metals in the droppings of kingfishers, a plunge-diving species that primarily preys on fish59. These studies employ avian droppings as non-invasive biomonitoring tools, and our findings extend this approach to tropical piscivores, revealing striking inter-site contrasts, from < 1 mg/kg in relatively undisturbed Maramangalam to > 48 mg/kg in industrially impacted Puthuvype. This highlights the value of faecal THg as a sensitive, non-invasive indicator of local mercury exposure in aquatic ecosystems. However, faecal mercury reflects short-term dietary exposure and can be influenced by individual feeding behaviour and matrix-related variability. Therefore, droppings are best interpreted as indicators of relative spatial and temporal patterns rather than exact body burdens. Future studies combining faecal THg with mercury measurements in prey and blood would help further validate this approach as a proxy for recent dietary exposure. These findings have critical implications for conservation and risk assessment, particularly for long-lived piscivorous birds that serve as effective bioindicators of aquatic contamination. The progressive THg loading seen here raises concerns about potential sub-lethal effects on reproduction, behaviour, and population viability, especially in protected wetlands and Ramsar-designated areas60,61. Targeted mitigation strategies, including mercury source identification, reduction in emissions, and monitoring of high-risk zones, are urgently needed to safeguard both biodiversity and ecosystem health in coastal Kerala and comparable habitats. Conclusion This study demonstrates clear spatial, temporal, and species-specific patterns of Hg contamination in piscivorous waterbirds along the coastal zones of Kerala. Elevated THg concentrations in Little Cormorant and Oriental Darter droppings, particularly at coastal sites, indicate increasing mercury exposure over the three-year period, while inland wetlands currently show comparatively lower levels. The species- and habitat-specific concentration patterns revealed in this study highlight the importance of using top avian predators as effective bioindicators for Hg contamination. Furthermore, the significant interaction effects between year, species, and location emphasize the need for spatiotemporally explicit conservation strategies. Given the ecological importance of Kerala’s wetlands, many of which fall under the Ramsar Convention, the findings highlight the need for careful consideration in environmental management and monitoring. In particular, they underscore the value of strengthening oversight of mercury inputs, supporting wetland protection and restoration initiatives, and developing long-term biomonitoring programmes that incorporate avian indicators. While the present study documents elevated mercury exposure in waterbirds, which may have implications for their health and reproduction, it also points to broader environmental concerns that warrant further investigation. However, given the relatively limited sample size, the observed patterns should be interpreted with caution and require confirmation through broader long-term studies incorporating larger sample sets and additional wetland systems. Future research should prioritise identifying potential sources of mercury, elucidating methylation processes within these wetland systems, and linking contaminant burdens to reproductive and fitness outcomes in birds. Supplementary Information Below is the link to the electronic supplementary material. Supplementary Material 1 Supplementary Material 1