The analysis of surface water monitoring data of both the river Rhine and Maas revealed that melamine indeed is present in both rivers. Considering additional monitoring data from the Norman Empodat database, it can be concluded that the entry path via wastewater seems to be rather low (i.e., below the LOD). However, it has to be acknowledged that the number of measurements is also rather low. In surface waters, melamine was not detected in all samples, but in approximately 90% of the samples above the LOQ, indicating its presence in the aquatic freshwater environment, although the concentrations were in the Rhine between 0.30 and 3.23 µg/L and in the Maas 1.49–20 µg/L.
In a study from the USA, melamine and cyanuric acid were measured in a river and a lake in New York State. Melamine was the most abundant compound in river water with concentrations between 0.017 and 3.650 µg/L [42]. It can be seen that the concentrations are in a similar range compared to the Rhine and the Maas. Unfortunately, it is not clear from the paper by Zhu et al. what the flow rate of the river was and at what time the measurements were carried out. However, the data suggest that the rivers in large American cities have similar melamine concentrations as in Europe. The presence of cyanuric acid in American surface waters (river 0.147 µg/L, lakes 1.068 µg/L und seawater 0.157 µg/L) can be explained by the direct discharge of wastewater. In the USA, cyanuric acid is used in dishwashing detergents, sanitizer and drinking water treatment products [42].
In summary, it can be concluded that there is a good database concerning monitoring data in surface water along the rivers Rhine and Maas. Based on these measurements, it was concluded that there is a correlation between industrial production sites and melamine concentration in surface water [43,44,45]. Nevertheless, no clear trend in the annual melamine surface water concentrations during the entire observation period (from 2015 to 2020) was identified. Overall, it should be noted that the analysis of trends based on annual averages does not provide a complete picture of the melamine concentrations in surface water. Melamine concentrations are subject to strong fluctuations throughout the year. There might be a correlation between the melamine concentrations and the flow-rate of the river indicating low melamine concentrations a high flow-rates.
The number of available melamine groundwater detections are scarce, and the results are not consistent across various water bodies, as it was present in some areas, whereas it was not in others (i.e., below LOQ/LOD). This indicates that the occurrence of melamine in some groundwater areas is likely to be more dependent on local conditions and possible point sources than on a general soil leachability issue. Due to the lack of sufficient groundwater monitoring data and relevant additional information, however, no reliable conclusions can be drawn regarding the correlation between monitored surface and groundwater concentrations of melamine.
Considering the presence of melamine in drinking water, it has to be emphasized that there are currently no freely available monitoring data available as this substance is not part of measurements campaigns carried out on regular base. Despite this fact, some authors already stated that based on its substance properties melamine belongs to PM (persistent and mobile) substances, which cannot be removed by conventional or advanced treatment processes, such as activated carbon [46]. Due to this assumption, it was concluded that melamine may reach drinking water and associated resources. However, as mentioned previously this assumption lacks a causality check and/or a proof of concept as the few monitoring data available are not conclusive at this stage. Furthermore, it has to be emphasized that melamine may be considered as persistent in water using a standardized OECD 309 lab test but was shown to be non-persistent under realistic field conditions with half-life of 39 days [47].
In a study from the USA, melamine and cyanuric acid were detected in tap water and bottled water. Melamine was found with a mean concentration of 0.033 µg/L in tap water and 0.075 µg/L in bottled water. Cyanuric acid was detected at a mean concentration of 0.515 µg/L in tap water and 0.075 µg/L in bottled water [42]. According to the authors, the occurrence of melamine in tap water could be due to the presence of indoor dust [48]. Since it is not clear from the article what material the drinking bottles are made of, it cannot be ruled out that melamine or cyanuric acid has dissolved from the bottles. Drinking water in the USA is usually purified with chlorine [49]. Cyanuric acid is commonly used as a drinking water treatment product to stabilize the chlorine [42]. It has been also reported that drinking water can contain up to 1.6–3.2 mg/L cyanuric acid form the use of disinfectants (sodium dichloroisocyanurate) [50]. This use can explain the high concentrations in drinking water from the tap and the bottle.
Cyromazine, was not measured in drinking water at all. In summary, there is a lack of data regarding melamine and cyromazine measured concentrations in drinking water.
The presence of melamine in agricultural soil is due to use of the plant protection product active ingredient (AI) cyromazine and the use of fertilizers. Whereas, melamine is a relevant transformation product of the AI, it is also freely available within the fertilizer. Based on the results from a soil leachability study, a horizontal transfer of both cyromazine and melamine through the soil column was considered negligible [7]. If degradation to melamine already took place within the soil, melamine is primarily bound to the soil and hence, does not present a dominant entry path for melamine into waterways. Thus, it can be assumed that the presence of melamine in crops and vegetables is linked to the agricultural use of cyromazine and the use of fertilizers rather than by an uptake of melamine leached from surface water to the agricultural area [30]. A study from Korea shows that melamine and cyanuric acid were measured in high concentrations in the sediment of a lake near Seoul. The mean concentration was 0.182 µg/g dw for melamine and 0.0262 µg/g dw for cyanuric acid [51]. In a study from Japan, melamine concentrations between 10 and 400 ng/g dw were measured in river sediment (OECD, 1998, Primary source no longer available) [51]. The high concentrations can be explained by the fact that melamine and cyanuric acid were added to fish feed to increase the protein content [52, 53].
Releases to the soil from production have not been reported and is considered as negligible. If releases occur, melamine would primarily be bound to the soil reducing the risk of introduction into waterways. However, a run-off from the agricultural field straight into nearby surface waters after heavy rain falls could not be excluded. This was also demonstrated in another study, where 23.7% cyromazine was observed in the run-off water [7]. It can be assumed that the soil does not present a dominant entry path for melamine into waterways including groundwater taking into account that melamine release from the degradation of cyromazine is negligible.
A comparison with the potential impact of agriculture is limited due to the lack of available monitoring data. In this context, another author refers to the decreasing and low concentrations of cyromazine in surface waters, indicating that agricultural use of cyromazine most likely does not significantly contribute to the observed melamine concentrations in Dutch surface waters [1]. Another major gap is the lack of data for surface water in remote areas.
The available data on the occurrence of melamine due to the use of cyromazine as PPP suggest that the presence of melamine in plant food can rather be attributed to environmental contamination or residues from fertilizers and disinfectants than to the use of cyromazine as PPP [5, 54]. Cyromazine is commonly added to animal feed in concentrations up to 0.5 mg/kg to control the hatching from flies in the manure [55, 56]. However, data indicate that agricultural activities (manure/slurry treatment and veterinary purposes) might contribute to the occurrence of melamine in food and feed and hence, to melamine background exposure. Studies on consumer exposure to melamine indicate that consumers are exposed to low levels of melamine (migration from food contact materials and from feed ingredients). Data evaluated suggest that melamine can leach from melamine resin products, such as tableware. The rate increases with water temperature but seems to be not affected by acidity. This suggests that melamine resin products can contribute to releases into the environment, but only at a low emission rate. However, the available data did not allow for an assessment of specific quantities or shares.
Waste from melamine production is mainly incinerated, which leads to the thermic destruction of melamine. However, no detailed information on releases regarding waste from the manufacturing of intermediates, mixtures and articles was available but it is suggested to be also incinerated. Concerning professional and consumer use, the contribution to waste is considered negligible as products should typically be incinerated and not disposed of in landfills. However, releases from existing landfills may be relevant as they may even be connected to ground water resources [57,58,59].
In conclusion, quantified figures regarding releases of melamine into the environment can only be established for releases from the production of melamine, its use in the manufacturing of articles.
Although a correlation between releases from industrial point sources and the concentrations of melamine in surface water has been identified, the presence of melamine in surface water cannot exclusively be ascribed to a specific segment of the production and manufacturing chain. Further research should, therefore, also focus on other possible emission pathways of melamine to the environment.
However, when interpreting these shares, it must be kept in mind that no full picture including all emissions of melamine into the environment rather than a proportionate allocation of melamine emissions based on known (and assumed) emissions, i.e., a relative comparison of known sources and their (assumed) emissions is possible. Furthermore, it should be noted that the releases from the production of intermediates and manufacturing of articles using melamine might be overestimated as these are solely based on worst-case releases presented within the SDS for melamine and data submitted by EMPA members within the questionnaires.
Other contributions to melamine in the environment cannot be calculated or estimated due to the data gaps identified above (contributions resulting from the use of cyromazine) or as they were not the focus of this assessment (contribution from cyanamide containing fertilizers).
Based on the data presented in this report, the main contributions to surface water resulted from the production of melamine, intermediates and from the manufacture of melamine-containing products, while the significance of contributions from other sources (e.g., waste, landfill) to the release of melamine to water remains unclear.
This is further substantiated by the available monitoring data, which shows elevated melamine concentrations downstream of industrial point sources. In a study from China, wastewater from 37 production plants was analysed for its melamine content. Melamine was measured in only 9 samples with a content of 22–100 µg/L [30]. Therefore, a correlation between discharges of industrial point sources and the concentrations of melamine in surface water may be plausible. However, significant concentrations are already detected upstream of such point sources, indicating the presence of other sources.
Hence, while a connection between releases from industrial point sources and the concentrations of melamine in surface water has been identified, the presence of melamine in surface water and the environment as a whole may not exclusively be allocated to the production of melamine, the production of intermediates and the manufacturing of articles using melamine. Further research on sources and pathways of melamine to the environment should, therefore, also focus on other possible sources, such as:
-
i)
The degradation/metabolization of melamine from cyromazine resulting from its use as PPP and biocide and the emission to surface water via the feed and food chain and organic slurry.
-
ii)
Potential releases of melamine from the use of cyanamide-containing fertilizers.
-
iii)
Emissions from consumer uses such as tableware made from melamine–formaldehyde–resin.
-
iv)
Emissions from the degradation of incapsulated fragrances or.
-
v)
Emissions from waste.
Regarding the question whether melamine in the environment is ubiquitous and widely dispersive detectable or primarily occurs in direct relation to point sources, the following can be concluded.
As regards surface water, data are mainly available for monitoring stations related to the rivers Rhine and Maas. The data indicate that melamine concentrations at these locations frequently exceed the ERM target value of 1 μg/L and that higher melamine concentrations can be observed downstream of industrial point sources. Thus, the available data suggest a relationship between releases from industrial point sources and the measured concentrations of melamine in surface water. However, high melamine concentrations can also be observed in surface water, where no discharge of melamine is known. In the river Emscher (Germany) are high concentrations of melamine (max. value of 21 µg/L) detected. The river is characterised by a high proportion of wastewater, but there is no known industrial discharger of melamine into the Emscher [17, 21].
The available data also show that melamine is not detected in all surface water samples in the rivers Rhine and Maas above the corresponding LODs. It can, therefore, be concluded that melamine is not a ubiquitous substance in surface waters in Germany and the Netherlands.
Monitoring data for melamine in groundwater are scarce. The available data demonstrate that the presence of melamine in groundwater depends on local conditions (such as infiltration rates or the presence of point sources). Therefore, it is concluded that melamine is not ubiquitously found in ground water. Due to existing data gaps, no reliable conclusion can be drawn regarding the correlation between monitored surface and groundwater concentrations of melamine.
The rather large number of melamine detections in surface water did not comply with the low detection rate in groundwater (GW) and drinking water (DW), although it has to be acknowledged that the data from latter two are rather rare (i.e., 1 out of 12 for DW). In areas with high river water infiltration rates, detection of melamine in such filtrates (RWF) occur, but more on a local rather than on a general basis. The local occurrence in RWFs depends also on the melamine concentrations present in the surface water and as the concentrations may increase during the course of the river, it may result in exceedance of any threshold at the lower end of larger rivers (i.e., river Rhine: the Netherlands), rather than at the upper or middle region (i.e., Switzerland, Germany). However, river water infiltrations often lead to rather constant water flow from the river to the well. The associated soil, therefore, may be considered as fully saturated, which means that both physical and chemical absorption of substances may be limited. Furthermore, in case of high flow rates in combination with short distances (i.e., 1–10 m), the biodegradation does not play a relevant role. This result in the detection of substances such as caffeine typically not being considered as persistent or a substance of low absorption [8, 9, 11, 60, 61]. On the other hand, substances such as long-chain perfluoroalkyl carboxylic acids (PFOAs) considered as highly absorptive (log Koc of > 4) [21,22,23], may end up in groundwater, river filtrates and associated drinking water resources if they (a) are particle bound and/or (b) are highly persistent (i.e., half-life of several decades) so that they may travel through the horizontal and/or the vertical soil layer, respectively [21,22,23, 62].
Taking the available information into account, it remains questionable whether melamine is an abundant potential groundwater contaminant or whether the individual detections are due to high local emission rates in combination with high local river-water infiltrations, thus requiring further elaboration and research. If the additional data indicate that local sources are the main driver for melamine detections in groundwater further site-specific emission reduction are considered necessary.