The primary sources of oxygen in the marine environment are atmospheric oxygen which enters the system via gaseous exchange across the air-sea surface interface and in situ production via photosynthesis of algae and other aquatic plants.

Of critical importance to marine organisms is the fate and behaviour of dissolved oxygen and the factors affecting fluctuations in dissolved (DO) levels. The principal anthropogenic activity resulting in changes in dissolved oxygen concentrations in the marine environment is the addition of organic matter.

DO is measured in estuaries and coastal waters in terms of either a concentration (mg l^-1) or as a percent saturation (%).

MPMMG (1998) reported summer and winter concentrations of DO at National Monitoring Programme sites in the UK in the range 4 to 11 mg l^-1 expressed as a median, with lowest concentrations occurring in estuaries during the summer.

Many estuaries have intensive monitoring programmes for DO and data will be available from the Environment Agency, SEPA or the Environment and Heritage Service.

Oxygen demand is also routinely measured in effluents discharging to estuaries and coastal waters and is a common condition of discharge consents for effluents with a high organic content. Oxygen demand can be estimated as the Biochemical Oxygen Demand (BOD) or the Chemical Oxygen Demand (COD). The oxygen demand of the sediment can also be estimated as the Sediment Oxygen Demand (SOD).

BOD is a standard analytical procedure involving the incubation of a sample of water or effluent for a standard period of time (5 days) at a constant temperature (20 °C) and measuring the dissolved oxygen concentration at the beginning and end of the incubation period with the difference between the two measurements being the oxygen demand expressed in mg l^-1. The oxygen demand results principally from the microbial degradation of organic matter and from nitrification, although some chemical oxidation may also be taking place (hence biochemical oxygen demand). Commonly, the nitrification process is suppressed by adding allylthiourea (ATU) to the sample and then BOD is expressed as BOD₅ (ATU).

BOD₅ is a useful measure of oxygen demand for comparative purposes. However, if an estimate of ultimate oxygen consumption is required, a modified test over a longer period is required to better estimate the oxygen demand arising from the breakdown of compounds, such as lignin, which are not readily broken down by aerobic bacteria. In this context, BOD₅ is used as a measure of fast BOD and the longer-term test of slow BOD. The combination of fast and slow BOD represents an estimate of medium-long term oxygen demand which falls somewhere between BOD₅ results and those obtained using total oxygen demand or chemical oxygen demand (COD) results (see Comber and Gunn 1994).

COD is a standard analytical procedure involving the addition of a chemical oxidising agent (potassium permanganate or dichromate) to a sample of water or effluent for a standard period of time at a constant temperature and measuring dissolved oxygen concentrations as for BOD. COD provides a more complete oxidation of both organic and inorganic compounds than BOD (although both are erratic in their response to aromatic organics), so provides higher estimates of oxygen consumption rates. COD therefore provides a better indicator of medium to long-term oxygen demand. However, COD results are affected by the presence of chlorides (Sherrard et al 1979), which casts doubt over COD results from saline waters.

SOD can be measured by numerous methods (Nixon 1990). It is the result of all biological respiration and nitrification, which may be measured either in situ or ex situ, depending on the method chosen. As such, it is more akin to BOD. The results are usually expressed as oxygen removal over a 24 hour period per m² of sediment, usually standardised to a temperature of 20 °C. Because of the different timescales involved in measuring BOD and SOD, it may be difficult to relate the water column and sediment oxygen demands to each other, particularly when one (BOD) excludes oxygen demand due to nitrification, while the other includes it. The reason for this difference appears to be the close coupling of nitrification and denitrification in many sediments, but not in the estuarine water column.

The principal natural physical factors affecting the concentration of oxygen in the marine environment are temperature and salinity. DO concentrations decrease with increasing temperature and salinity. The other major factor controlling DO concentrations is biological activity: photosynthesis producing oxygen and respiration and nitrification consuming oxygen.

Photosynthesis occurs in aquatic plants in the presence of adequate supplies of carbon dioxide and light. Oxygen is released as a by-product. Blooms of phytoplankton in surface waters can supersaturate the water with dissolved oxygen during the day in the presence of adequate supplies of nutrients and light.

Respiration consumes oxygen and occurs in all aerobic organisms. Blooms of phytoplankton in surface waters can deplete the water column of oxygen during the night in the presence of adequate supplies of nutrients. Microbial respiration can deplete the water column and sediments of dissolved oxygen in the presence of organic matter.

Nitrification (the conversion of ammonia to nitrate via nitrite) consumes oxygen, the process relying principally on two bacterial genera: Nitrosomonas and Nitrobacter. Except in regions with high ammonium concentrations, e.g. around sewage outfall discharges, nitrification in the water column of shallow marine and estuarine systems appears to be limited (Henriksen and Kemp 1988). However nitrification can constitute a large proportion of sediment oxygen demand (e.g. Rivsbech et al 1988). In estuaries such as the Tamar (Owens 1996), nitrification is closely coupled with the turbidity maximum, but in other estuaries (e.g. the Mersey, Reynolds et al 1994) nitrification rates may be greater in filtered than unfiltered samples.

Oxygen balance in estuaries was usefully reviewed by Nixon (1990).

The effects of changes in dissolved oxygen concentration on the marine environment can be sub-divided into direct effects (those organisms directly affected by changes in dissolved oxygen concentration) and secondary effects (those arising in the ecosystem as a result of the changes in the organisms directly affected).

The direct effects of changes in dissolved oxygen concentrations are primarily related to reduced DO levels and include:

• lethal and sub-lethal responses in marine organisms;
• release of nutrients;
• development of hypoxic and anoxic conditions.

The lethal and sub-lethal effects of reduced levels of dissolved oxygen were reviewed by Stiff et al (1992) for the purposes of EQS derivation. This review was updated by Nixon et al (1995) in order to derive a General Quality Assessment (GQA) scheme for dissolved oxygen and ammonia in estuaries for the Environment Agency in England and Wales. The reader is referred to these documents for a detailed assessment of the lethal and sub-lethal effects of dissolved oxygen on saltwater organisms.

The lethal and sub-lethal effects of reduced levels of dissolved oxygen are related to the concentration of dissolved oxygen and period of exposure of the reduced oxygen levels. A number of animals have behavioural strategies to survive periodic events of reduced dissolved oxygen. These include avoidance by mobile animals, such as fish and macrocrustaceans, shell closure and reduced metabolic rate in bivalve molluscs and either decreased burrowing depth or emergence from burrows for sediment dwelling crustaceans, molluscs and annelids.

Stiff et al (1992) and Nixon et al (1995) identified crustacea and fish as the most sensitive organisms to reduced DO levels with the early life stages of fish and migratory salmonids as particularly sensitive. For estuarine fish, Stiff et al (1992) suggested a minimum DO requirement of 3 to 5 mg l^-1. Based on the data in their review, EQSs for dissolved oxygen were proposed (see table below).

Nixon et al (1995) proposed the following class thresholds for exposure to levels of dissolved oxygen for a continuous period of greater than 1 hour in estuaries in England and Wales (see table below). This scheme has not been implemented but the class thresholds are a useful indication of the levels of DO that are likely to cause effects if organisms are exposed for a continuous period of greater than one hour.

Nixon et al (1995) reviewed information on reduced DO levels in bottom waters on benthic invertebrate communities. Josefson and Widbom (1988) investigated the response of benthic macro- and meiofauna to reduced DO levels in the bottom waters of a fjord. At DO concentrations of 0.21 mg l^-1, the macrofaunal community was eradicated and was not fully re-established 18 months after the hypoxic event. In contrast, the permanent meiofauna appeared unaffected. Jorgensen (1980) observed the response of macrofauna to reduced DO levels of 0.2 to 1 mg l^-1 for a period of 3 to 4 weeks in an estuarine/marine area in Sweden by diving. Mussels Mytilus edulis were observed to first close their shells and survived for 1 to 2 weeks before dying. Crabs Carcinus maenas and shrimp Crangon crangon were amongst the first to die from lack of oxygen. Hydrobia ulvae were observed to die first in the hollows in the sediment surface and were observed congregating on the ridges to find more oxygen-rich water. Polychaetes were observed to come to the surface, small specimens first. Hediste diversicolor and Lagis koreni were observed limp and motionless on the surface but could be revived in 30 minutes by placing in oxygenated water. Burrowing bivalves were first observed to extend their siphons further into the water column but, as oxygen depletion continued, emerged from burrows and laid on the sediment surface. Sea anemones were the last animals to succumb but eventually they loosened their attachment and were found lying on the sediment surface.

Reduced levels of dissolved oxygen in the water column can result in the release of phosphate from suspended particles and the sediment.

Sustained reduction of dissolved oxygen can lead to hypoxic (reduced dissolved oxygen) and anoxic (extremely low or no dissolved oxygen) conditions. In anoxic environments, anaerobic bacteria proliferate, with nitrogenous oxide reducers absorbing oxygen by reducing nitrate to nitrite and forming ammonia or nitrogen gas. In addition, sulphate-reducing bacteria reduce sulphate to hydrogen sulphide which, when liberated, increases mortality of marine organisms and increases the BOD as it permeates through the water column (Kennish 1986). Such conditions can occur under a cage fish farm installation where release of hydrogen sulphide has caused fish kills and sediment can become covered in filamentous fungi, such as Beggiatoa.

The indirect effects of reduced dissolved oxygen concentrations depend on the severity of the direct effects which, in turn, depend on extent and duration of the oxygen depletion. Sustained or repeated episodes of reduced dissolved oxygen has the potential to severely degrade an ecosystem. Reduced DO levels contributed significantly to the reported elimination of the fish populations of the Thames estuary and its recovery has resulted from strict management of water quality, including inputs of organic matter and the artificial injection of oxygen into the water column during low DO events.

The consequences for seabirds and sea mammals of such ecosystem degradation are likely to be significant as the supply of food organisms is affected.

Potential effects include:

• lethal and sub-lethal effects on marine organisms (in particular crustacea and fish) of reduced DO concentrations below the EQS values in Table C5.1;
• release of phosphate from suspended particles and sediment with potential contribution to the effects of eutrophication;
• establishment of anoxic conditions which can increase BOD, stimulate the release of ammonia and hydrogen sulphide which can be toxic to aquatic life;
• severe degradation of the ecosystem if reduced DO levels are sustained or repeated with potential adverse effects for sea birds and Annex II sea mammals.

Next Section                     References