Ammonia is present in all natural waters, even if only at very low concentrations. It is derived either from the breakdown of organic nitrogen (mineralisation) or by the reduction of nitrate (a process known as denitrification). Ammonia also represents an intermediate stage in nitrogen fixation - the conversion of atmospheric N₂ to fixed nitrogen and subsequent incorporation into microbial proteins, etc. However, this remains a relatively unimportant source of ammonia compared to mineralisation. A substantial proportion of atmospheric nitrogen deposition is in the form of ammonia (Review Group on Acid Rain 1997), although this too is a relatively minor source.

However, although ammonia is produced in the nitrogen cycle, anthropogenic sources are more important, notably sewage treatment effluent and, in some situations, run-off from agricultural land (Seager et al 1988). In tidal waters, the primary source of ammonia is direct discharge from Secondary Treatment Work (STW) outfalls.

Total ammonia is most often measured in discrete samples. However, the need for continuous monitors to properly record the variability in levels has been addressed in a number of UK estuaries, such as the Forth, Humber and Thames. The following values are annual mean total ammonia levels recorded at continuous monitoring sites (Nixon et al 1995) measured as mg/ln:

Seager et al (1988) presented some older data for UK estuaries, with mean total ammonium values ranging between 0.017 mg N l^-1 in the Tay at Perth (at the tidal limit) to 2.094 mg N l^-1 in the Clyde at Glasgow. Much higher values were presented for two sites in the Tees Estuary, but these appear to be based on a single sample each.

The proposed GQA classification scheme for ammonia in the estuaries of England and Wales (Nixon et al 1995) consisted of a four tier system, with class boundaries at 0.86 mg N l^-1 (Class A/B); 4.7 mg N l^-1 (Class B/C) and 8.6 mg N l^-1 (Class C/D) using the upper 90%ile as the classification statistic. These boundary values were derived from a review of toxicity data. This scheme has not been implemented in England and Wales so far.

Total ammonia in aqueous solution comprises two principal forms: the ionised ammonium ion (NH₄⁺) and un-ionised ammonia (NH₃). There are technical difficulties in measuring the unionised form. However, total ammonium is usually monitored instead, despite the fact that the proposed EQS of 0.021 mg NH₃-N l^-1 for the protection of saltwater fish and shellfish is presented in terms of unionised ammonia. However, the proportion of ionised and un-ionised ammonia can be calculated from total ammonia, the relative proportions depending on salinity, temperature and pH. The proportion of unionised ammonia increases with increasing temperature and pH, but decreases with increasing salinity (Seager et al 1988). Of these three factors, salinity appears to be relatively unimportant, but at pH 8.5, the proportion of un-ionised ammonia is approximately 10 times that at pH 7.5. For every 9°C increase in temperature, the proportion of unionised ammonia approximately doubles.

An exhaustive literature review on the toxicity of ammonia to marine organisms has not been carried out for the purposes of this profile. The information provided in this section is taken from existing review documents (Seager et al 1988 and Nixon et al 1995). The most sensitive groups of organisms have been identified.

The toxicity of ammonia to marine organisms has not received the same level of study as freshwater organisms but there is sufficient information to indicate that the principal groups of organisms affected by ammonia toxicity in the marine environment are invertebrates and fish.

The un-ionised form of the ammonium ion (NH₃) is the most toxic. The toxicity of ammonia to aquatic life is affected by the following physico-chemical parameters: temperature, pH, dissolved oxygen and salinity. In general, ammonia toxicity is greater, the higher the temperature and pH and the lower the levels of dissolved oxygen and salinity.

Concerns about the toxicity of ammonia should be greatest in estuarine European marine sites and close to sewage outfalls in coastal waters.

A review of the effects of ammonium on estuarine and marine benthic organisms is given in Nixon et al (1995). Toxicity data are presented for shrimps, mysids and lobsters (in which ammonia appears to interfere with lobsters= ability to adjust to different salinities). Allan et al (1990) estimated 96 hour LC50s for juvenile school prawns Metapenaeus macleayi and leader prawns Penaeus monodon to be 1.39 and 1.69 mg un-ionised ammonia NH3-N/l (26.3 mg and 37.4 mg total ammonia-N/l respectively). Williams and Brown (1992) estimated a 96 hour LC50 of 0.787 mg NH₃-N/l (24.6 mg NH₄-H/l) for the nauplius of the marine copepod Tisbe battagliai and a No Observed Effect Concentration (NOEC) of 0.106 mg NH₃-N/l (3.34 mg NH₄-N/l) for a study comprising tests on several life stages. For invertebrates, toxicity appears to increase as salinity decreases (Miller et al 1990, Chen and Lin 1991), although too few data exist to indicate whether this pattern is typical for all or most invertebrates (Nixon et al 1995). Several studies indicate that ammonia toxicity is greatest to early life stages of invertebrates.

Acute toxicity of ammonia to fish increases with low dissolved oxygen concentrations. This has been shown in both fresh and marine water environments (Seager et al 1988, Nixon et al 1995). For this reason, the proposed GQA scheme for ammonia in estuaries was combined in a proposed joint scheme for dissolved oxygen and ammonia (Nixon et al 1995).

The majority of ammonium toxicity data relates to fish, although most of the species tested are freshwater species, with many coarse fish appearing to be as sensitive to ammonia as salmonids (Mallet et al 1992). Nevertheless, data are available for sole, turbot and larval inland silversides. Eddy and Twitchen (1990) suggested that high environmental sodium concentrations can decrease toxicity to fish.

In the Mersey Estuary at a mean unionised ammonia concentration of 0.008 mg NH₃-N/l, a diverse invertebrate population was present, and this region was passable by flounder and salmonids. For fish, ammonium toxicity appears to be less at lower salinity levels, gradually decreasing until it reaches a point similar to that found in freshwaters (Seager et al 1998, Miller et al 1990). This may be of relevance, especially in estuaries where DO sags occur at low salinities.

Ammonia exerts toxic effects in the water column and does not accumulate in the sediments. However, sediment dwelling organisms that use water in the boundary layer between the sediment and the water column (molluscs, crustacea and most annelids) for feeding or respiration could be at risk.

Ammonia does not bioaccumulate.

Potential effects include:

• toxicity to invertebrates and fish associated with Annex I habitat sub-features at concentrations above the EQS of 0.021 mg NH₃-N l^-1 in the water column;
• increased toxicity to these groups of animals with decreasing dissolved oxygen and salinity. These conditions are more likely in estuaries;
• Annex II species (other than anadromous fish) and birds are unlikely to be affected directly by ammonia toxicity. However, in estuaries, in particular, ammonia toxicity is a component of the effects of organic pollution which can degrade bird feeding habitats

Next Section                     References