Hidden Contaminants on the Menu – Market and Socioeconomic Profiles of Emerging Contaminants in Fish of Nutritional Importance Revealed by µ-Raman

1. Introduction

The increased production, consumption, and disposal of synthetic materials have intensified the release of emerging contaminants into aquatic environments worldwide. Among these contaminants, microplastics and other associated anthropogenic components have attracted growing scientific attention because of their persistence, ubiquity, and ability to interact with physical, chemical, and biological processes in marine and freshwater ecosystems [1,2]. In addition to conventional plastic polymers, environmental contamination increasingly includes carbonaceous materials, regenerated fibers, industrial dyes, pigments, pharmaceutical residues, and other anthropogenic compounds introduced through urban runoff, wastewater discharges, industrial activities, maritime transport, and atmospheric deposition [3,4,5].

Marine organisms are continuously exposed to these contaminants through direct ingestion, trophic transfer, and interactions with contaminated sediments and water columns. Fish are particularly relevant because they occupy different trophic levels, are widely distributed throughout coastal ecosystems, and provide an important source of animal protein for human populations worldwide [6]. Consequently, commercially important fish species have been increasingly recognized as valuable indicators of environmental contamination and as sentinels of contaminant occurrence within aquatic food webs, providing important information on the distribution, accumulation, and potential transfer pathways of anthropogenic materials in marine ecosystems [7,8].

The presence of emerging contaminants in seafood poses a challenge that extends beyond environmental pollution and directly intersects with food security, public health, and sustainable development. Current evidence suggests that microplastics and associated contaminants can act as vectors for additives, trace metals, persistent organic pollutants, and other hazardous substances, raising concerns about food safety and long-term human exposure [9,10]. These concerns are closely aligned with the United Nations Sustainable Development Goals (SDGs), particularly SDG 2 (Zero Hunger), SDG 3 (Good Health and Well-being), SDG 12 (Responsible Consumption and Production), and SDG 14 (Life Below Water), which emphasize food security, the protection of human health, the sustainable use of resources, and the reduction of marine pollution [2,11,12,13,14,15].

The increasing complexity of contamination pathways has also reinforced the relevance of the "One Health" framework, which recognizes the interconnectedness between environmental integrity, animal health, food systems, and human well-being [9,16]. From this perspective, seafood contamination is a critical interface linking environmental pollution, seafood quality, and potential exposure pathways in food systems, highlighting the need for integrated approaches that simultaneously address ecosystem health, food safety, and human well-being.

Despite growing evidence of anthropogenic materials in aquatic organisms, comparatively little attention has been given to whether their occurrence in commercially important fish varies among seafood commercialization systems [5,8,17]. Market categories may integrate differences in fishing origin, supplier networks, transportation routes, storage conditions, and other characteristics of seafood supply chains [6,15,18]. Consequently, comparisons among markets can reveal distributional patterns in contaminant occurrence, but they cannot identify the source or timing of gastrointestinal exposure.

The selected markets represent the principal seafood commercialization channels in Santos and Maputo and encompass distinct commercialization systems, infrastructure, seafood handling practices, supply chains, and consumer socioeconomic profiles. In Santos, sampling included the Municipal Market (MSM), Open-Air Street Fairs (MSF), and Municipal Markets and Supermarkets (MSAP), representing traditional fish markets, informal retail channels, and formal commercial establishments. In Maputo, fish were collected from the Central Market (MC), Zimpeto Market (MZ), and Xipamanine Market (MX), which serve upper-middle-, lower-middle-, and low-income consumers, respectively.

This sampling strategy was designed to investigate whether different market environments and commercialization practices are associated with distinct patterns of emerging contaminant occurrence throughout the seafood supply chain. The selection of these markets was hypothesis-driven, as differences in infrastructure, storage conditions, handling practices, packaging materials, supplier networks, and consumer profiles may be associated with differences in material occurrence during commercialization. By encompassing distinct seafood retail systems in Brazil and Mozambique, the study provides a framework for exploring whether market characteristics are associated with differences in material-occurrence patterns.

In this context, the present study investigates the occurrence and composition of emerging contaminants in nutritionally important fish marketed in Santos (Brazil) and Maputo (Mozambique). Integrating µ-Raman spectroscopy, multivariate analysis, market categories, and socioeconomic profiles of consumers, the study seeks to identify contaminant signatures and explore potential relationships between seafood marketing systems and the occurrence of these contaminants. Through a comparative analysis between two important coastal cities in the South Atlantic and Western Indian Ocean regions, this research contributes to ongoing debates on environmental contamination, seafood safety, sustainable development, and One Health approaches.

2. Methodological Approach

This study adopted an interdisciplinary approach to investigate the occurrence and composition of emerging contaminants in nutritionally important fish marketed in Brazil and Mozambique. The study developed a comparative framework using seafood sourced from markets that represent distinct socioeconomic contexts and consumer profiles in Santos and Maputo. By integrating environmental, socioeconomic, and food-supply perspectives, the study evaluates how market characteristics and consumer contexts may relate to contaminant occurrence in seafood. A broader understanding of contaminant occurrence and potential exposure pathways in seafood systems contributes to discussions on food safety, environmental pollution, and human health in coastal urban systems (Figure 1).

2.1. Study Area

2.1.1. Santos, Brazil

Santos is located on the central coast of São Paulo State, southeastern Brazil, and is part of the Baixada Santista Metropolitan Region (Figure 2). The municipality covers about 281 km² and has a population of over 418,000, making it one of Brazil's most densely populated coastal cities [19].

Santos hosts the largest port complex in Latin America, which plays a central role in regional and national trade [20] and supports an extensive seafood supply chain connecting artisanal fisheries, wholesale distributors, retail markets, and consumers [21,22]. Fish samples were obtained from the Santos Municipal Market (MSM), Open-Air Street Fairs (MSF), and municipal markets and supermarkets (MSAP) in the metropolitan area [23]. These commercial establishments serve distinct consumer profiles, ranging from traditional fish consumers and local residents to tourists and higher-income households [22]. The city is characterized by high levels of urbanization, extensive commercial infrastructure, and a strong dependence on marine resources for food and economic activities [21,22].

2.1.2. Maputo, Mozambique

Maputo, the capital and largest city of Mozambique, is located on the western margin of Maputo Bay in the southwestern Indian Ocean (Figure 2). The metropolitan area has a population exceeding 1.1 million inhabitants and functions as the country's principal economic, administrative, and commercial center [24]. The city is strongly influenced by coastal and estuarine environments and supports important artisanal and small-scale fisheries that contribute substantially to local food security and livelihoods. We collected fish samples from three major seafood markets: Central, Zimpeto, and Xipamanine. These markets represent different socioeconomic segments of the urban population. The Central Market (MC) primarily serves upper-middle-income consumers, whereas the Zimpeto Market (MZ) mainly serves lower-middle-income consumers. The Xipamanine Market (MX) is predominantly associated with low-income consumers and is characterized by a stronger presence of informal seafood trade [25,26].

2.2. Fish Sampling

This study investigated the occurrence of emerging pollutants in commercially important fish species, namely croaker (Micropogonias furnieri) and horse mackerel (Trachurus trachurus), at fish markets in Santos (Brazil) and Maputo (Mozambique). These species are ecologically relevant, highly consumed, and may serve as bioindicators of plastic pollution and contaminant occurrence within aquatic food webs [6]. We analyzed the gastrointestinal tract, including the stomach and intestine, to detect meso- (5–25 mm) and microplastics (<5 mm). Samples were then chemically digested to remove organic matter (H₂O₂), density-separated (ZnCl₂), and visually screened under a stereoscopic magnifying glass [27,28].

2.3. Material Characterization by µ-Raman Spectroscopy

Material characterization was performed using µ-Raman spectroscopy [29] with a HORIBA Scientific, controlled by LabSpec software for spectral acquisition and processing. The system is equipped with laser excitation wavelengths of 473, 532, 633, 785, and 1064 nm and a long-working-distance objective with a numerical aperture (NA) of 0.55 and magnification of up to 100×. For each measurement, the laser power was initially adjusted to optimize spectral acquisition while avoiding visible damage to the analyzed material. Raman spectra were acquired over the spectral range of 200–3200 cm⁻¹ to enable the characterization of polymeric and non-polymeric materials [30,31,32,33].

For each sample, we adjusted the integration time, number of accumulations, and slit width to optimize the signal-to-noise ratio and prevent detector saturation. The acquired spectra were subjected to baseline correction and noise filtering using a computational routine implemented in MATLAB® v. 23.2. Following spectral preprocessing, the Raman spectra were compared with reference spectra available in the Knowitall® v2024.1 database to obtain spectral assignments for polymers and other materials [3,27,28]. Assignments were evaluated based on the correspondence between characteristic Raman bands of the samples and reference spectra. We interpreted library-based spectral matches cautiously, particularly for uncommon or chemically specific compounds, and considered them tentative when independent analytical confirmation was unavailable. We used images of the analyzed particles acquired with the Raman microscope to complement their morphological characterization.

2.4. Correspondence Analysis and Association Structure

Correspondence analysis (CA) explored distributional relationships between material components characterized by µ-Raman spectroscopy and categorical variables related to market type and market-associated socioeconomic profiles. We initially assessed overall associations using Pearson's chi-square test (χ²; p < 0.05). We then calculated adjusted standardized residuals (ASRs) to identify localized deviations from independence, using |ASR| ≥ 1.96 as an exploratory threshold corresponding to approximately the 5% level under the standard normal distribution. Because we examined multiple contingency-table cells, we interpreted ASR-based deviations as exploratory local patterns rather than confirmatory evidence of statistically significant associations. We then performed CA via singular value decomposition (SVD) of the standardized residual matrix to generate low-dimensional representations of the association structure [34,35,36]. We performed all analyses and visualizations in Python using NumPy, Pandas, SciPy, Matplotlib, and OpenPyXL [37].

3. Results

3.1. Frequency of Emerging Contaminant Components

The µ-Raman analysis yielded 65 material and compound spectral assignments, revealing a chemically diverse assemblage in the analyzed gastrointestinal samples. Frequency analysis showed a predominance of carbonaceous materials and synthetic polymers among the recorded spectral assignments (Figure 3). Carbon was the most frequent component, with 18 occurrences (17.8%), followed by graphite, with 10 occurrences (9.9%) (Figure 3). Together, these carbonaceous materials accounted for approximately 27.7% of all recorded contaminant occurrences.

Synthetic polymers were the second most frequently observed contaminant group. Poly(ethylene terephthalate) (PET) was detected on nine occasions (8.9%), while Polyester Film*2000 Series was detected on seven occasions (6.9%). In addition to synthetic polymers, regenerated cellulosic fibers such as Lyocell were also identified, expanding the diversity of anthropogenic fiber-related materials detected in the dataset (Figure 3).

The class distribution emphasizes that seafood contamination extends beyond conventional plastic polymers (Figure 4). Although plastics and carbonaceous materials were the most frequent individual components, the overall contaminant profile was dominated by a chemically diverse set of organic compounds, industrial dyes, pigments, pharmaceutical residues, and specialized industrial substances. Overall, the frequency distribution shows a contaminant profile characterized by the coexistence of carbonaceous particles, synthetic polymers, textile compounds, industrial pigments, and chemically diverse organic substances (Figure 3 and Figure 4). The results reinforce the notion that nutritionally important fish integrate multiple contamination pathways operating in coastal urban systems and seafood marketing networks, highlighting the complexity of exposure to emerging contaminants through consumption of these products.

3.2. µ-Raman Signatures

Figure 5 presents representative µ-Raman spectra of the main materials characterized in fish sold in Santos (Brazil) and Maputo (Mozambique). The spectra reveal high chemical diversity among the detected materials, including synthetic polymers, carbonaceous materials, regenerated fibers, and complex mixtures of organic and inorganic compounds. The spectra attributed to polyethylene terephthalate (PET) (Figure 5a and Figure 5d) exhibit high agreement with the respective reference standards, highlighting intense bands near 630, 860, 1290, 1615, and 1725 cm⁻¹, consistent with the characteristic vibrations of the aromatic and carbonyl groups present in the molecular structure of the polymer. The spectra obtained from the Brazilian and Mozambican samples show very similar profiles, indicating the presence of PET in both sample sets.

The carbonaceous materials identified as graphite (Figure 5b and Figure 5f) exhibit the G band centered between 1580 and 1600 cm⁻¹, characteristic of organized graphitic structures. Although the spectrum in Figure 5f corresponds to a mixture of compounds, the graphite signature remains clearly identifiable. The regenerated fibers identified as Lyocell (Figure 5c) exhibit bands characteristic of cellulosic materials, particularly in the regions of 1090–1120 cm⁻¹, 1330–1380 cm⁻¹, and approximately 2900 cm⁻¹. The composite spectra obtained in Mozambique (Figure 5e and Figure 5f) show the coexistence of different materials, including cellulose, industrial organic compounds, silicate minerals, dyes, and carbonaceous materials. Overall, the two countries show similar spectral patterns, especially for PET components, carbonaceous materials, and cellulosic compounds, indicating the recurring presence of these contaminant classes in the analyzed samples.

3.3. Market and Consumer Associations

Correspondence analysis revealed distinct distributional patterns between the contaminant components identified by µ-Raman spectroscopy and the categorical variables describing seafood markets and consumer socioeconomic profiles. Table 1 summarizes the statistical results. Pearson's chi-square tests did not indicate statistically significant dependence in any of the contingency tables (Market: χ² = 75.70, p = 0.171; Consumer: χ² = 23.31, p = 0.616). However, the adjusted standardized residuals (ASRs) revealed localized deviations from independence in specific cells of the contingency tables, suggesting exploratory associations between specific contaminant components and market categories, as well as with low-income consumers.

For the Market vs. Components analysis, we identified seven statistically significant associations (|ASR| ≥ 1.96) (Table 1). Municipal Markets and Supermarkets (MSAP) showed positive associations with 5-Nitroisatin (ASR = 1.99) and Carbon (ASR = 2.46), while Open-Air Street Fairs (MSF) were associated with Cellulose (ASR = 3.10) and Lyocell (ASR = 2.16). Similarly, the Santos Municipal Market (MSM) exhibited significant associations with Poly(ethylene terephthalate) (ASR = 2.02) and Polyester Film*2000 Series (ASR = 2.56), whereas the Xipamanine Market (MX) was positively associated with Graphite (ASR = 2.33). The perceptual map (Figure 6) visually represents these relationships, where proximity between markets and contaminant components reflects the strength of their associations. The first two dimensions of the market correspondence analysis explained 69.5% of the total inertia, with Dimension 1 accounting for 42.2% and Dimension 2 accounting for 27.3% (Table 1). The perceptual map (Figure 6) demonstrates a clear separation among market categories and highlights the contaminant components that contributed most strongly to the observed structure. Components such as Cellulose, Lyocell, Poly(ethylene terephthalate), Polyester Film*2000 Series, and Graphite were positioned near their associated markets, reinforcing the ASR-based results.

In contrast, the Consumer vs. Components analysis revealed a substantially simpler association structure. As summarized in Table 1, we detected only one statistically significant association. The Low-Income (LI) consumer category was positively associated with Graphite (ASR = 2.33), whereas we found no significant relationships for the Lower-Middle (LMI) and Upper-Middle (UMI) groups. This pattern is evident in the perceptual map (Figure 7), where LI is positioned in close proximity to Graphite, while the remaining consumer categories occupy regions with no significant associations. The consumer correspondence analysis explained 100% of the total inertia within the first two dimensions, with Dimension 1 and Dimension 2 accounting for 59.1% and 40.9%, respectively (Table 1). The resulting perceptual map (Figure 7) captured the full variability in the contingency table and revealed limited differentiation among socioeconomic groups, despite a significant association with the Low-Income category.

4. Discussion

4.1. Market Dynamics, Urbanization, and Contaminant Pathways

The occurrence of anthropogenic and other detected materials in the gastrointestinal tracts of fish marketed in Santos and Maputo highlights the complexity of exposure pathways operating in coastal urban systems. Because these materials were detected in gastrointestinal tissues, their occurrence primarily indicates environmental exposure and ingestion before capture. At the same time, differences among marketed fish may also reflect distinct fishing origins and seafood supply chains.

Urbanization plays a central role in this context, since coastal metropolitan areas concentrate high population densities, industrial activities, transport infrastructure, solid waste generation, and effluent discharges, which favor the entry of anthropogenic materials into aquatic ecosystems [1,2]. In cities like Santos and Maputo, characterized by intense coastal occupation, maritime activities, and high fish consumption, anthropogenic materials and contaminants can reach marine environments through surface runoff, river transport, sewage, port activities, and atmospheric deposition, exposing aquatic organisms throughout their life cycle [4,5].

Port activities may also add anthropogenic materials to coastal environments. Santos hosts Latin America's largest port complex, while Maputo is a major maritime gateway to Southeast Africa. The intense movement of vessels, cargo, containers, and fishery products is associated with potential inputs of plastic debris, synthetic fibers, packaging-derived materials, and particles originating from industrial and transportation activities [38,39].

Beyond environmental exposure, post-capture handling may also contaminate fish products. Materials used in storage, refrigeration, transportation, and packaging, including plastic containers, synthetic ropes, and protective films, may release particles during handling and commercialization [17,40]. However, such post-capture processes do not readily explain materials detected within an intact gastrointestinal tract. Therefore, gastrointestinal occurrence in this study should primarily be interpreted as evidence of pre-capture exposure. In contrast, market identity may reflect differences in fishing grounds, suppliers, transportation routes, and broader supply-chain characteristics [18,41].

Although the global chi-square tests did not indicate significant dependence between the evaluated variables, adjusted standardized residuals revealed localized patterns in which certain materials occurred more frequently in specific market categories. Interpret these patterns cautiously because market identity may integrate multiple factors, including fishing origin, supplier networks, transportation routes, storage conditions, and commercialization practices [5,15,18]. Consequently, correspondence analysis provides an exploratory representation of distributional patterns rather than evidence of causal relationships or significant overall associations between market categories, socioeconomic profiles, and material occurrence [34,42].

The detection of PET, polyester-related materials, graphite, lyocell, and cellulosic signatures demonstrates the chemical diversity of materials associated with gastrointestinal exposure [2,5,43]. Although these materials may occur in urban, industrial, textile, and packaging-related contexts, spectral identification alone cannot determine their specific sources. Fish markets therefore represent relevant interfaces between aquatic resources and urban food systems, and integrated monitoring should consider both aquatic environments and seafood distribution systems when investigating contaminant occurrence in coastal regions [2,15].

The occurrence of these materials in commercially important fish also raises broader sustainability questions related to urban development, resource management, and marine conservation. These issues are relevant to Sustainable Development Goals (SDGs) 11 (Sustainable Cities and Communities), 12 (Responsible Consumption and Production), and 14 (Life Below Water), which emphasize waste reduction, marine pollution prevention, and sustainable coastal systems. From a One Health perspective, anthropogenic materials in commercially important fish highlight the interconnectedness of environmental quality, aquatic organisms, and food systems [1,11,16]. However, potential implications for human exposure and public health require further investigation, particularly through analyses of edible tissues and dietary exposure.

4.2. Fish Viscera – Nutritional Benefits and Emerging Contaminant Exposure

Fish viscera are by-products of fish processing and include organs such as the stomach, intestines, liver, spleen, gonads, and pancreas [44]. These structures represent a significant proportion of the animal's biomass, averaging 12–18% of total body weight depending on species, developmental stage, and environmental conditions [45,46]. In the present study, we analyzed only gastrointestinal tissues, which are recognized as important compartments for retaining ingested microplastics and other anthropogenic particles in fish [1,6]. Therefore, these findings should be interpreted as evidence of material occurrence within the gastrointestinal tract and environmental exposure, rather than as direct evidence of contamination of edible muscle tissues or human dietary exposure.

Nevertheless, gastrointestinal tissues provide valuable information on the ingestion and retention of anthropogenic materials and on environmental exposure processes operating within aquatic ecosystems. Although direct human consumption of fish viscera remains limited in many regions, these by-products are increasingly valorized for fish oils, nutraceuticals, pharmaceuticals, biodiesel, and animal feed, contributing to waste reduction and more efficient use of fishery resources within circular economy frameworks [45,47]. Their potential use in food- and feed-related applications makes characterizing retained anthropogenic materials particularly relevant.

Fish offal also has substantial nutritional and biotechnological value, containing proteins, lipids, vitamins, minerals, and enzymes. The lipid fraction can range from 5% to 36% and may contain polyunsaturated fatty acids, particularly eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA), which are associated with cardiovascular, neurological, and immunological functions [46]. Protein concentrations can range from 5% to 22%, while digestive tissues are important sources of proteases, lipases, and amylases used in several industrial applications [44,45]. In some species, the nutritional density of offal can exceed that of traditionally consumed muscle tissues, supporting growing interest in its valorization.

Although the present study did not evaluate edible muscle tissues, previous investigations have reported that microplastics and associated substances may translocate from the gastrointestinal tract to secondary tissues, including the liver, gills, bloodstream, and, in some cases, muscle, particularly for smaller particles and nanoplastics [7,48,49,50,51]. Such evidence should not be interpreted as demonstrating translocation in the fish examined here. Instead, the materials detected in gastrointestinal tissues provide evidence of exposure and ingestion, while their possible distribution to other tissues remains to be investigated. Future studies combining gastrointestinal and edible-tissue analyses are therefore necessary to evaluate tissue distribution and potential implications for seafood safety.

The gastrointestinal tract represents a major interface between fish and particles ingested from aquatic environments, including microplastics, mesoplastics, synthetic fibers, pigments, chemical additives, and other xenobiotic materials [1,6]. Microplastics retained in digestive tissues may also interact with trace metals, persistent organic pollutants, plastic additives, and microorganisms, increasing the complexity of contaminant behavior and potential transfer processes within organisms and aquatic food webs [4,7,52,53,54]. However, the presence of these materials in gastrointestinal tissues should be distinguished from their bioavailability, translocation, toxicological effects, and presence in edible tissues, none of which the present study directly assessed.

From a food safety perspective, the occurrence of microplastics and associated contaminants in fish by-products warrants attention when these materials are intended for human food, nutraceutical, pharmaceutical, or animal-feed applications. Although the toxicological consequences of microplastic exposure remain incompletely understood, experimental studies have associated exposure under specific conditions with oxidative stress, inflammatory responses, metabolic alterations, and interactions with adsorbed chemical compounds [8,9,10,53,55,56,57]. Fish viscera may also accumulate other contaminants, including trace metals, persistent organic compounds, parasites, and pathogenic microorganisms, supporting the need for appropriate monitoring and quality-control procedures when these tissues are valorized or incorporated into food and feed products [44,52].

The dual role of fish viscera as potentially valuable biological resources and compartments that retain ingested anthropogenic materials raises important questions for sustainability and resource utilization. Valorizing them supports SDG 2 (Zero Hunger) and SDG 12 (Responsible Consumption and Production) by improving resource efficiency and reducing waste in fisheries and seafood processing. Potential concerns about contaminants are also relevant to SDG 3 (Good Health and Well-being), although the current gastrointestinal findings do not establish human exposure or health risk. From a One Health perspective, these results reinforce the importance of considering environmental contamination, aquatic organisms, seafood production, and resource valorization within integrated monitoring frameworks [8,10,11,16,54].

4.3. Methodological Contributions, Limitations, and Future Perspectives

Because fish gastrointestinal tissues are important compartments for retaining ingested microplastics and associated materials, reliable analytical methods are essential to assess environmental exposure and investigate potential implications for seafood systems [6,10]. In this context, µ-Raman spectroscopy combined with multivariate statistical analysis provides a useful framework for characterizing, tentatively identifying, and exploring the distribution of diverse materials in fish gastrointestinal tissues. Such approaches may also support future monitoring of fish by-products increasingly used in food, nutraceutical, pharmaceutical, and animal-feed applications [45,46].

One of the principal methodological contributions of this study is the use of µ-Raman spectroscopy as a high-resolution analytical tool to characterize both polymeric and non-polymeric materials associated with gastrointestinal exposure. Unlike visual classification approaches, which are largely restricted to morphological characteristics, µ-Raman spectroscopy provides molecular-level spectral information that enables discrimination among synthetic polymers, carbonaceous materials, pigments, dyes, fibers, and other chemical signatures occurring in environmental samples [31,32]. However, library-based spectral matches, particularly for uncommon or chemically specific compounds, should be interpreted cautiously and regarded as tentative when independent analytical confirmation is unavailable. This capability remains particularly relevant because materials occurring in aquatic organisms can extend beyond conventional plastic polymers.

Recent studies have further demonstrated the versatility of µ-Raman spectroscopy for investigating anthropogenic materials across different environmental matrices. Integrating µ-Raman spectroscopy with geospatial analysis, multivariate statistics, machine learning, and environmental monitoring approaches has supported the characterization of polymeric particles, associated contaminants, and complex mixtures of materials in coastal sediments, urban rivers, and beach deposits [3,27,28,33]. These advances highlight the potential of µ-Raman spectroscopy as a multidisciplinary analytical platform for material characterization, environmental monitoring, and exploratory investigations of contaminant occurrence and distribution.

Another methodological contribution of the present study is integrating µ-Raman spectroscopy with correspondence analysis (CA). While spectroscopy provided spectral characterization of the detected materials, CA provided a statistical framework for exploring their distribution across market categories and market-associated socioeconomic profiles. Although the global chi-square tests did not show statistically significant dependence, adjusted standardized residuals identified localized deviations from independence that frequency analyses alone did not reveal. Accordingly, this combined approach is valuable for exploratory characterization of distributional patterns, but it should not be interpreted as demonstrating causal relationships, contaminant sources, or exposure pathways [3,27,28,33,34,36].

The results also support analytical approaches that extend beyond conventional polymer identification. The detection of PET and polyester-related signatures, carbonaceous materials, cellulosic signatures, pigments, dyes, and other spectral matches illustrates the chemical diversity of materials occurring in gastrointestinal samples [3,27,28,33,58]. However, these spectral signatures do not independently establish anthropogenic origin, specific environmental sources, or biological effects, and uncommon compound assignments require additional validation. Consequently, analytical frameworks that characterize multiple material classes while explicitly accounting for uncertainty in spectral identification are important for investigating environmental exposure and generating hypotheses regarding potential implications for seafood and environmental health [9,17,18,28].

Despite these methodological contributions, the study has several limitations. The analysis focused on material occurrence rather than concentration, preventing estimates of contaminant loads or exposure levels. Furthermore, environmental weathering, fluorescence, spectral overlap, sample preparation, and library-matching uncertainty can complicate Raman-based identification, particularly for complex environmental samples [31,32]. The exclusive analysis of gastrointestinal tissues prevents direct inference regarding contamination of edible muscle, dietary exposure, or human health risk. Moreover, the study design cannot determine the specific sources or timing of gastrointestinal exposure or distinguish among potential pre-capture environmental pathways. Although post-capture handling, transportation, storage, packaging, and commercialization may introduce additional contamination to fish products [17,18], these processes do not readily explain materials detected within intact gastrointestinal tissues. Future studies should therefore incorporate quantitative measurements, edible-tissue analyses, procedural blanks and strengthened QA/QC, complementary analytical confirmation where necessary, and improved traceability of fishing origin and supply chains.

Beyond these analytical considerations, the methodological framework adopted here aligns with SDGs 9 (Industry, Innovation and Infrastructure), 3 (Good Health and Well-being), 12 (Responsible Consumption and Production), and 14 (Life Below Water) by supporting analytical innovation and environmental monitoring. Integrating µ-Raman spectroscopy with multivariate statistics can improve the characterization of complex material assemblages and support hypothesis generation for subsequent environmental and seafood investigations. From a One Health perspective, reliable and validated analytical approaches are important for investigating connections among environmental contamination, aquatic organisms, and food systems [11,27,28,32,33]. However, establishing implications for human exposure and health requires dedicated analyses of edible tissues, exposure levels, and toxicological relevance.

5. Conclusions

This study revealed a chemically diverse assemblage of materials in the gastrointestinal tracts of commercially important fish marketed in Santos, Brazil, and Maputo, Mozambique. µ-Raman spectroscopy showed the recurrent presence of carbonaceous materials and synthetic polymers, particularly graphite and PET, as well as cellulosic and other materials with distinct spectral signatures. Similar occurrences of PET, graphite, and cellulosic materials between the two study areas indicate that comparable contaminant classes appear in fish marketed in geographically distinct coastal urban systems.

Correspondence analysis identified localized patterns between specific materials and market categories; however, the global chi-square tests were not statistically significant. These relationships should therefore be interpreted as exploratory associations rather than evidence of market or socioeconomic effects or causal contamination pathways.

Overall, integrating µ-Raman spectroscopy and multivariate analysis provides a useful framework for characterizing the diversity and distribution of materials associated with gastrointestinal exposure in commercially important fish. Because we analyzed only gastrointestinal tracts, the results do not provide direct evidence of contamination in edible muscle tissues or human dietary exposure. Further studies incorporating edible tissues, quantitative contaminant measurements, expanded sampling, and strengthened analytical validation are needed to evaluate contaminant distribution, sources, and potential implications for seafood safety.