The Challenges of Dealing with Nitrogen Pollutants in Groundwater

Introdution

The right nitrogen compounds circulating in Earth determine the existence of life. Human activities are causing the alteration of the global nitrogen circulation, because increasing quantities of nitrogen compounds are being introduced into air, soil and aquatic ecosystems in the last few years (Ghaly and Ramakrishnan, 2015; Fields, 2004). In the planet, nitrogen is found in non-reactive and reactive forms. The non-reactive form of nitrogen is N₂, which does not contribute to environmental impact. In contrast, reactive forms of nitrogen that come from the fixation of N_2, can be the root of strong environmental effects. The accumulation of reactive nitrogen compounds, due to especially the production of fertilizers and by-products of combusting fossil fuels, may cause grave impacts on human health, smog, acid rain, soil, forest dieback, biodiversity, global warming and water quality (Castner et al., 2017; Fields, 2004).

Particularly, the infiltration of reactive nitrogen in drinking water is carried out through runoff from livestock and fertilizers, which arrives to the aquifers that are the principal reservoirs of groundwater (Sampat, 2000). The growing pollution of groundwater sources by reactive nitrogen concentrations has been reported in the last three decades (Castner et al., 2017; Rivett, Buss, Morgan, Smith and Bemmet, 2008). This has set alarm bells ringing about the need to prevent the increase of reactive nitrogen concentrations in groundwater due to the effects on human health in most industrialized nations (Kobus and Kinzelbach, 1989). In order to preserve drinking water, the quantification of nitrogen loss rates in groundwater is paramount. However, the complex and dynamic biogeochemical network of groundwater has not enabled the development of effective methods that measure its nitrogen processes. Therefore, the aim of this review is to examine how nitrogen pollutants, principally nitrate and ammonium, are affecting the microbiological processes of the nitrogen cycle in groundwater and the strengths and weakness of developing effective methods for their detection.

Nitrogen pollutants: nitrate and ammonium

The pollution in groundwater caused by nitrate (NO₃ ^-) has been becoming an emerging concern worldwide since the 1970s (Spalding, Exner, Martin, and Snow, 1993; Rivett et al., 2008). Even though nitrate in lower concentrations is non-toxic in drinking water, the risk from long exposure may cause health problems such as hypertension, stomach cancer and methemoglobinemia in human infants (baby blue disease) (Almasri and Kaluarachchi, 2004) since it is suspected carcinogen (Kendall, Elliott, and Wankel, 2007). To improve the drinking water quality, the World Health Organization (WHO) established a limit of 10 mg-N/l for nitrate concentrations in 1993 (U.S EPA, 2017). Different developed countries have taken important septs; Germany, for instance, monitors the nitrate concentrations of potable water setting a limit value of 15 mg/l at 15% of all groundwater control points (UBA, 2014). In the case of developing countries, nitrate concentrations are not constantly monitored. Nitrate is also the most studied pollutant compared to ammonium (NH₄ ⁺) in groundwater (McMahon and Bölhke, 2006) due to low levels of ammonium in this natural waterbody (Schilling, 2002). For instance, the United States Environmental Protection Agency (U.S EPA) has not established a maximum contaminant level (MCL) for ammonia in drinking water. However, the National Academy of Science recommends, and many European nations have adopted, a drinking water standard of 0.5 mg/l (Twort et al., 2000). Ammonium levels are regulated via oxidation to nitrate in the unsaturated zone and attenuated by biological uptake and sorption (Schilling, 2002). According to Aravena and Robertson (1998), the high ammonium levels initially present in raw sewage cease to exist in shallow groundwater plumes emanating from septic system. The disinfection efficiency of drinking water systems may be altered by ammonium due to the formation of nitrite producing odour and taste problems on the water (Canter, 1996). The WHO reported that natural ammonium levels produced by the decay of buried organic matter are frequently less than 0.2 mg-N/l in groundwater (U.S EPA, 2017). Elevated ammonium levels in natural waters may indicate pollution caused by anthropogenic sources such as sewage effluent, leaking manure lagoons, and landfill leachate (Christensen et al., 2001). For instance, the high ammonium concentrations of about 61 mg/l are found on a leachate from the KL landfill in Kalamazoo, Michigan (Kehew and Passero, 1990). The primary consequence of the high ammonium concentration in groundwater is the alteration of the chemical and biological properties of water by a frequent eutrophication and hypoxia (Vitousek et al., 1997).

The principal sources contributing to the increase of nitrate and ammonium levels in groundwater are the agricultural practices such as mineralization of urea, degradation of organically bound nitrogen and dissociation of ammonium nitrate fertilizer (Clark et al., 2008), which also cause emission of greenhouse gases (i.e. N₂O) and climate change (Almasri and Kaluarachchi, 2004). There are other sources from localized activities that might contribute to the pollution of groundwater by nitrate and ammonium. They include septic effluent (Aravena and Robertson, 1998) atmospheric deposition, spreading of manure and sewage (Wakida and Lerner, 2005) and municipal wastewater (Christensen et al., 2000). Previous studies have reported that the attenuation of nitrate and ammonia is masked by dilution through dispersion and mixing (Domenico and Schwartz, 1998). For instance, the elevated nitrate and ammonium levels decrease over time in groundwater from North America, Caribbean and Europe during the time periods of one to thirty years (Widory et al., 2004). Furthermore, new reports suggest that denitrification and anaerobic ammonium oxidation (ANAMMOX) as the major responsible processes involved in the removal of these N-compounds in the nitrogen cycle, based on molecular and geochemical evidences (Clark et al., 2008; Seitzinger et al., 2006). Although denitrification has been extensively studied, the significance of anammox may have been largely overlooked. A better understanding of the nitrogen elimination in groundwater is essential to determine the potential impact of contaminants on water supplies, which is extremely important for public health, drinking water providers and society.

Nitrogen cycle

The global nitrogen cycle is defined by microbial processes produced in oxic and anoxic conditions, which maintain a small pool of fixed nitrogen in continues exchange with the atmospheric N₂ reservoir (figure 1) (Thamdrup and Dalsgaard, 2002).

Figure 1: Microbial nitrogen cycle and its influence upon the water environment.

Source: Rivett et al. (2008)

Redox reactions include the transfer of electrons between molecules by means of changing the molecular oxidation state. Groundwater involves a series of the redox reactions facilitated by microorganisms. Biologically, mediated redox reactions lead to nitrogen isotope fractionation and regulate complex nitrogen dynamics and release gases in soils as do lightning and biomass burning. The availability of nitrogen is principally represented by the existence of nitrate, nitrite, ammonium and organic nitrogen, which are generally grouped together as fixed N. Denitrification, anammox, dissimilatory nitrate reduction to ammonium (DNRA), and nitrification are the microbial processes involved in the removal of nitrogen in the nitrogen cycle.

Analytical methods in nitrogen compounds

The quantification of the mechanisms of nitrogen processing presupposes a critical role in determining the influence of groundwater nitrogen sources (Tobias, Macko, Anderson, Canuel, and Harvey, 2001). The study of biogeochemical reactions in groundwater, particularly denitrification and anammox, becomes a challenging task due to the difficulty of accessing or detecting the reaction sites, where the specific pathways that result in retention or accelerated losses of nitrogen by multiple ammonium and nitrate sources establish a complex and dynamic microbiological network. Unfortunately, the development of effective methods that guarantee adequate measurements of the involved microbiological processes has not been established so far. Available methods have shown serious drawbacks such as changes in substrate concentrations, lack of sensitivity, disturbances in the physical settings of the process, high spatial and temporal variation of the process or high background concentration (Jahangir et al., 2013). The identification of reaction rates from free living bacteria in groundwater has been enriched by new molecular and analytical approaches through the last years. Although the genetic characterization of new species established a possibility to understand the organism's function in groundwater, for instance quantitative PCR (qPCR), the measurements require to be calibrated as proxies for the reaction rate, and therefore, these rates do not necessarily correlate with their activity influenced by environment conditions (Song and Tobias, 2011). Therefore, chemists and hydrogeologists have tried to address the derivation of the reaction rates by different approaches. The examination of changes in chemical ratios along flow paths from stoichiometrically-related chemical concentrations of reactants and products is an analytical method to identify processes that are involved in groundwater. This method is not always sufficient due to the fact, that the chemical ratios do not provide insight into the sources of pollution.

Another analytical approach to derive reactions rates is the use of in situ tracer experiments coupled to advection-dispersion models, which are focused on the measurement of mass transport of compounds (Böhlke, Smith, and Miller, 2006). Isotope analysis is one such approach used to identify the relative importance of sources that are isotopically distinct such as nitrate and ammonium and the changes of isotope ratios that often occur in the biological cycling of nitrogen in predictable directions (Kendall, Elliott, and Wankel, 2007). However, there are some problems on the implementation of isotopes analysis that evoke controversy, such as the overlapping of isotopic compositions in different sources, the spatial and temporal variations in isotopic composition and the blurring of distinctive isotopic composition by isotope fractionation. New approaches try to minimize these problems, as in the case of a multi-isotope-multi tracer approach that tracks trace elements to different sources of nitrate giving a myriad of environmental variability inherent to natural systems (Kendall et al., 2007).

There are two different approaches applying nitrogen stable isotopes on groundwater, which are complementarity used: natural abundance and tracer methods, also known as ¹⁵N labeling experiments. The natural abundance approach reveals intrinsic information through rates of the predominant biogeochemical processes of the nitrogen cycle that incorporate large spatial and temporal variation of the processes with minimal alteration of the studied system (Hamilton and Ostrom, 2007). On the contrary, tracer methods offer primarily the study of rates experiments of elemental cycling on short incubations, which add around 10% of the ambient source concentration of ¹⁵N-compound and are often best conducted in situ with the aim to avoid any contamination of the samples (Hamilton and Ostrom, 2007). Studies have reported the use of ¹⁵N labelling experiments in a wide range of aquatic ecosystems (Mulholland et al., 2004). The discovery of anammox was also accomplished in natural systems by using direct δ¹⁵N tracer measurements. Currently, this ¹⁵N isotopic labelling has become a powerful tool to trace denitrification and anammox rates (Holtappels, Lavik, Jensen and Kuypers, 2011), but it cannot be used in the environment. The denitrification and anammox processes establish an unique labelling pattern, where anammox produces dinitrogen by one to one pairing of nitrogen from nitrite and ammonium (Strous et al., 1999), while denitrification generates nitrous oxide and dinitrogen through random pairing of nitrogen atoms during reduction of nitrogen oxide (Holtappels et al., 2011). Therefore, denitrification and anammox activity may be proved with the formation of ¹⁵N¹⁴N (²⁹N₂) rather than ¹⁵N¹⁵N (³⁰N₂) in the laboratory (Dale et al., 2009). It is important, however, to develop an analytical tool by using a natural label of ¹⁵N and ¹⁴N isotopes for the determination of these processes without disturbance in natural systems.

In addition, the invention of right analytical method becomes more complicated since the availability of NO₂ ^- pool may also cause the partitioning of nitrite between denitrification and DNRA. Thus, DNRA processes can be involved in anoxic tracer incubations due to the direct conversion of nitrite to ammonium in where the production of ³⁰N₂ may occur via anammox whether there is significant rates of DNRA by combining the DNRA substrate ¹⁵NO₂ ^- with the DNRA product ¹⁵NH₄ ⁺ (Kartal et al., 2007). The coupled DNRA-anammox may occur in two forms: intracellular coupled DNRA-anammox in which DNRA and anammox are effectuated within the same cell and intercellular coupled DNRA-anammox where distinct organisms carry out the coupled DNRA-anammox by using a small ambient ammonium pool (Holtappels et al., 2011).

In order to understand the complexity of nitrogen tracer experiments, it is important to clarify that there are ten known nitrogen isotopes, the majority of them being radioactive. The stable ones are ¹⁴N and ¹⁵N with concentrations of 99.634% and 0.368% of the global nitrogen pool of the earth’s atmosphere (Holtappels et al., 2011). Stable isotope compositions are defined in terms of delta (δ) values that are established in parts per mil (%_o) according to equation (10) .

The δ value is calculated by the comparison of the ratio (R) between the sample and standard. By convention, R is the ratio of the heavy (less abundant) to the light isotope (most abundant) (i.e. ¹⁵N/¹⁴N for nitrogen). The ratio standards are determined by the reference standard such as Vienna Pee Dee Belemnite (VPDB) or Vienna Standard Mean Ocean Water (VSMOW). The sample and standard have often very similar ratios of the two stable isotopes, which give a value very close to 1. To emphasize this difference, the deviation from the value 1 is represented in per mil. For instance, the nitrogen isotopes ratios are calculated as the %_o difference to atmospheric N₂ having a constant ¹⁵N/¹⁴N of 272 ±0.3 (Coplen, Krouse, and Bohlke, 1992). The isotopes from an element contain mass differences, which may be interpreted as a mass-dependent isotope fractionation effect (Kendall and Caldwell, 1998). Equilibrium isotope exchange reaction and kinetic processes produce isotopic fractionations. The equilibrium fractionation processes are involved in reversible equilibrium reactions redistributing isotopes of an element to various compounds. The equilibrium fractionation factor α is used to express the fractionation associated with the equilibrium exchange reaction between two substances as shown in equation (11) where k is the equilibrium constant and n the number of exchanged atoms. Kinetic fractionation processes are associated with irreversible unidirectional kinetic reactions such as evaporation, dissociation, biologically mediated reactions and diffusion that establish the difference in the motility of the isotopic species. The kinetic isotope fractionation factor α describes the relative difference in the reaction rates constants (k) of heavy and light isotopologues during the reaction, as shown in equation (12) for δ¹⁵N and δ¹⁴N (Mariotti et al., 1981).

Most biological reactions discriminate against the heavier isotope. Therefore, the definition of an isotope enrichment factor (ε) is more convenient considering that it clearly highlights the range of variation (eq. 13)

Many studies have taken advantage of isotope analysis of nitrogen (δ¹⁵N) and subsequently, of oxygen (δ¹⁸O) over the last decades (Kendall, 1998; Clark et al., 2008). Whereas the use of only nitrogen isotopes analysis is challenging, as the δ ¹⁵N values of fertilizers and atmospheric deposition of existing nitrate and nitrate in soil are similar to the nitrogen forms in the aquatic ecosystems, the combination of δ¹⁵N and δ¹⁸O analysis provides an important insight of nitrate sources involved in the nitrogen cycle (Kendall, 1998). Studies have reported the nitrogen isotopes fractionation of the different microbiological processes performed in the nitrogen cycle over the last decades. The fractionation of nitrification is often calculated from the rate determining step, which is the slower oxidation of ammonium to nitrite having a large δ¹⁵N value ranging from 38 to -14 ‰ (Mariotti et al., 1981). The transformation of nitrate to nitrogen gas via denitrification results in isotopic fractionations from +15 to +30‰ (Böttcher, Strebel, Voerkelius, and Schmidt, 1990). The reports of DNRA fractionation are lacking a range of isotopic effects (Kendall et al., 2007), whereas (McCready, Gould, and Barendregt, 1983) used a kinetic fractionation demonstrating that produced ammonium has a lower δ¹⁵N value than nitrate. There is little knowledge about the isotopic fractionation of anammox, but it is assumed that anammox and denitrification may have similar isotope fractionation. However, new evidences show that anammox preferentially removes dinitrogen from the ammonium pool with an isotope effect of + 23.5 to + 29.1 ‰ in marine ecosystems according to Brunner et al. (2013).

Conclusions

Bearing in mind the issues described in the parts of this review, it is evident that more research on the detection of biogeochemical reactions of the nitrogen cycle in groundwater is needed in order to determine the impact of nitrogen pollutants on water supplies for human consumption. However, the field of measurements in environmental chemistry has shown that despite the limitations of effective methods that detect and quantify microbiological processes of nitrogen loss, there are some advances on the quantification of the rates of nitrogen elimination in groundwater by developing new analytical methods in the last few years. While the first methods for measurement of nitrogen compounds involved in the nitrogen cycle that came to light evidenced serious drawbacks, the use of ¹⁵N-labelling experiments including in situ tracer experiments and multi-isotope-multi tracer approach has become a powerful tool investigating regulations of microbial redox reactions in the present. Nevertheless, although very elegant, this artificial isotopic labeling cannot be used in the environment, since introducing, for instance, ¹⁵NH₄+ or ¹⁵NOx- in sufficient quantities into aquifers is practically impossible in natural conditions. As anammox and denitrification processes in the nitrogen cycle are the principal responsable in the removal of nitrogen compounds, the study of these processes are a priority, principally anammox reaction, which has not largely been explored. In addition, further research to estimate the kinetic rate parameters of the biogeochemical transformations in this complex reactions environment by using numerical models would be beneficial to establish new critical insight for the development of new methods for anammox and denitrification determination in groundwater and other natural systems.