Eutrophication

Background

The impact of atmospheric pollution by sulphur (S) and nitrogen (N) has historically been concerned with acidification of soils and freshwaters. However, inputs of atmospheric N are also a source of essential nutrients, which commonly limits growth in temperate ecosystems. This fertiliser effect results in increased plant growth and an increased demand for other plant nutrients. The gradual increase and enrichment of ecosystems by nutrients such as N and/or P is termed eutrophication.

Increased availability of N from enhanced atmospheric inputs impacts species composition, favouring those plants with a high demand for nitrogen. Where there are large inputs of reduced nitrogen (ammonia), which are not immobilised, in the soil, this may result in the suppression of the uptake of other essential plant nutrients such as potassium (K ⁺) or magnesium (Mg²⁺).

As most temperate natural and semi-natural ecosystems are N limited, additional N inputs in the first instance act to stimulate plant growth. However, there is a limit to how much additional N input can be utilised. Soils and ecosystems with N inputs in excess of plant nutritional requirements are often referred to as N saturated. Ultimately, increased losses of both inorganic and organic N from terrestrial systems may contribute to freshwater, coastal and marine eutrophication, where N is a limiting nutrient (Hornung et al., 1995).

In order to assess the impact of increased N deposition on eutrophication and the impact this has on ecosystems, an empirical approach for setting critical loads based on changes in plant communities is the most widely used.

The impact of the interaction between acidification and eutrophication on aquatic processes and biota is unclear (INDITE, 1994). It has been hypothesised by Hessen et al. (1997c), that N deposition may contribute to the oligotrophication of freshwaters, as acidification, in general, tends to decrease the mobility of P (Kopecek et al., 1995). However, this is a matter for debate and further research.

Eutrophication and terrestrial ecosystems

Although enhanced atmospheric N deposition has increased forest productivity, most boreal forests growing on mineral soils are still N-limited. Tamm et al. (1995) consider that, although many coniferous forests have not yet reached a state of N-saturation, it is likely that this will increase over the coming decades, with a threat to the conservation value of affected forests. The first indication of N saturation is the occurrence of elevated nitrate leaching (see below).

In forest ecosystems, subject to increased N deposition, effects on the tree component are uncommon and vary with species type and geographical location (Wilson and Emmett 1999). However changes to woodland ground flora have been recorded. Surveys of species composition of ground flora along a 0.5 Km transect from livestock buildings show adverse changes 50-300 m downwind of the buildings. Species such as Deschampsia flexuosa, Holcus lanatus, Rubus idaeus and Urtica dioica were abundant close to livestock units and their % cover decreased rapidly with distance from source, while the more N-sensitive species, such as Oxalis acetosella, Galium odoratum, mosses and ferns, which are found upwind and some distance downwind of the houses were scarce at all sites receiving > 25 Kg ha^-1 N year^-1. Visible injury to pine and spruce needles was observed immediately downwind of the buildings. Foliar nitrogen concentration of a number of species was large close to the buildings and declined with distance. Total N deposition at the woodland boundaries was estimated at 40 to 80 Kg N ha^-1 N year^-1 at the 4 sites, and exceeded critical loads for acidic coniferous forest, i.e. 15-20 kg ha^-1 N year ^-1 to protect ground flora (Pitcairn et al., 1995 and 1998).

In heathland communities, rapid changes in the species composition have occurred as a result of increased nutrient availability. In the Netherlands this has resulted in a dramatic decrease in species diversity, because many (rare) species, which were characteristic of the Calluna or Erica Sp. dominated heathlands, have disappeared and been replaced by the grasses Molinia caerulea and Deschampsia flexuosa. More than 35% of former Dutch heathland is estimated to have changed into grassland (Bobbink et al., 1993). Both increased N deposition (largely in the form of NH3 from intensive stock units) and heather beetle damage (Brunsting, 1982) have been implicated. Calluna decline has also occurred in a number of Breckland heaths in East Anglia, (Marrs, 1986; Pitcairn et al., 1991) and has been attributed to catastrophic death of Calluna caused by frost, drought or heather beetle attack. Such changes in these plant communities have also been linked to the disappearance of some butterflies, amphibians and birds in these habitats (Bobbink et al., 1995; Fangmeier et al., 1994).

Eutrophication and Surface waters

Phosphorus has long been considered the principal limiting nutrient in freshwaters (Schindler 1977). Nitrogen limitation was thought to be less common because of the potential for nitrogen fixation by certain species of cyanobacteria. Nitrogen limitation does, however, appear to be significant in many coastal waters and is now accepted as being more widespread in freshwaters, although generally as a consequence of elevated phosphorus concentrations (Moss et al. 1997). There is also growing evidence of temporally dynamic co-limitation by nitrogen and phosphorus in standing waters (Hessen et al., 1997c; Maberly et al., in press). In general, however, the impact of excess inorganic N on the trophic status of surface freshwaters is considered less significant (Hessen et al., 1997a). In Britain, a report on the impacts of N deposition in terrestrial ecosystems concluded that no biological impacts, directly attributable to N deposition, have been identified for lakes and that the impact of N deposition on the eutrophication of streams is even more difficult to quantify. Consequently, little information is available to make any robust assessment (INDITE, 1994).

Stoddard (1994) has proposed a scheme for classifying the degree to which freshwaters have been impacted. At Stage 0, the ecosystem is N-deficient, the N cycle is dominated by vegetation and microbial uptake, and the demand for N has the major influence on the seasonal NO₃^- pattern in surface waters. Thus, at this stage NO₃^- concentrations in surface waters should be negligible, except perhaps during the winter/spring period, when direct losses of atmospherically derived N, arising from either snowmelt or spring rains, may occur. The longer the period of small stream NO₃^- concentrations is, the greater the extent of N limitation within the ecosystem. With the progressive development of N saturation, amplification of the seasonal NO₃^- pattern in surface waters would be expected (Stage 1), as winter NO₃^- concentrations continue to increase and summer values remain small, compared with Stage 0. During Stage 2, the extent to which seasonal cycles of surface water NO₃^- concentrations occur has diminished, as significant loss of NO₃^- also occurs during the summer, primarily because NO ₃^- supply exceeds biological demand. By Stage 3, the catchment has become a net source of N rather than a sink, and the characteristics of surface waters in these system. Sites that fall into the later stages of N loss, Stages 2 and 3, have large NO₃^- concentrations throughout the year, and in some extreme cases concentrations may be larger in the summer than the winter (e.g. Stevens et al., 1993b). In Britain, moorland catchments and young forested sites generally fall into Stages 0 and 1 of Stoddard's stages of N loss to aquatic systems.

More recent attention has also been given to the role of N deposition and ecosystem nutrient status in relation to dissolved organic nitrogen (DON). Compared with inorganic forms of N, relatively few data are available for DON in upland streams. In those studies that have determined DON in stream water, it is usually a significant proportion of the total dissolved N present (Lepisto et al., 1995; Chapman et al., 1998), and may therefore be an important but often unaccounted component of total dissolved N in upland streams. The significance of DON as a potential bioavailable source of N for macrophytes and microflora has not yet been quantified.

Eutrophication and Nature Conservation

High rates of N deposition affecting species composition of low-nutrient status plant communities have been reported for lowland heaths in eastern Britain. Woodin and Farmer (1993) report that, for three National Nature Reserves in the same area, Calluna vulgaris is in decline, with a corresponding increase in grass cover. In the Breckland area of East Anglia, where N deposition ranges from 35 to 80 kg ha^-1 N year ^-1, Calluna cover declined by as much as 70% in some heaths between 1970 and 1990 (Pitcairn et al. 1991). Catastrophic events such as frost, drought and heather-beetle attack have led to an even-aged population susceptible to colonisation by Deschampsia flexuosa. As similar changes have occurred in upland moorlands and lowland heaths of UK, following N addition, it is likely that the changes observed in the Breckland, where the incidence of frost and drought are common, and where N deposition particularly from agricultural ammonia are high, are due to N deposition.

Numbers of bryophyte and lichen species declined in grazed and ungrazed plots at a number of sites at Moorhouse NNR between 1956 and 1989 (Pitcairn et al. 1991). Percentage changes in species number and cover in both grazed and ungrazed plots were largest in the base-rich grassland sites, compared with the intermediate grasslands and blanket bog sites. This indicates that the species rich base-rich grasslands are more susceptible to change. Atmospheric inputs of N and acidity are large at Moor House and may be implicated in the decline. Similar changes have been observed in the Derbyshire Dales following N addition.

Racomitrium moss heath has declined south of the Scottish Highlands during the last 50 years. Increased grazing pressure and acidic deposition are both likely to be responsible. Montane heaths receive larger inputs of cloud droplet deposition and higher concentrations of pollutant ions due to the seeder-feeder effect (Fowler et al.1988). Foliar nitrogen content of the moss was found to reflect atmospheric deposition of nitrogen, and transplant studies between sites within a mountain system demonstrated the importance of atmospheric N deposition in determining tissue N concentration of moss (Baddeley et al. 1994).

Brachypodium pinnatum expanded in calcareous grasslands in the UK during the last century, leading to reduced species diversity in many areas. Much of the early decline can be attribute to a decline in grazing. From studies of the pattern of change and management practices, the abundance of B pinnatum observed in 1990 in Surrey and Kent, and increases which occurred in Sussex and Hampshire, are likely to be partly due to large increases in N deposition

In a pilot study of the species composition of the ground flora of the Chiltern beechwoods (Ling et al. 1988), changes in species composition similar to those found in Sweden and the Netherlands were found. Species indicative of acid conditions had increased on calcareous soils and nitrogen-tolerant species had increased on acid soils.

Replacement of ombrotrophic Sphagnum species by minerotrophic species in ombrotrophic mires (Lee et al. 1993) is a definite response to increased N deposition. However, although the Moor House studies showed a decline in numbers of bryophyte and lichen species between 1956 and 1989 (Pitcairn et al.1991), some individual species increased in cover during the 1980s, and there is more recent evidence that certain species are increasing in response to increased N inputs from agricultural sources (Hill et al. 1994; Seaward 1992, 1998).

One Habitats Directive habitat thought to be particularly threatened by the combined acidification and eutrophication effects of N deposition are "oligotrophic waters containing very few minerals of Atlantic sandy plains with amphibious vegetation: Lobelia, Littorella and Isoetes" (Arts et al., 1990; Roelofs 1983).

References:

Arts GHP, Van der Velde G, Roelofs JGM, Van Swaay CAM. Successional changes in the soft-water macrophyte vegetation of (sub)atlantic, sandy, lowland regions during this century. Freshwater Biology. 1990;24:287-294.

Baddeley JA, Thompson DBA, Lee JA. Regional and historical variation in the nitrogen-content of Racomitrium lanuginosum in Britain In relation to atmospheric nitrogen deposition. Environmental Pollution. 1994;84:189-196.

Bobbink R, Boxman D, Fremstad E, Heil G, Houdijk A, Roelofs J. Nitrogen eutrophication and critical load for nitrogen based upon changes in flora and fauna in (semi)-natural terrestrial ecosystems. 1993:pp. 111-159.

Bobbink R, Roelofs JGM. Empirical nitrogen critical loads: update since Lokeberg (1992). (Hornung M, Sutton MA, Wilson RB). Bush: Institute of Terrestrial Ecology; 1995 pp. 9-19.

Bobbink R, Roelofs JGM. Nitrogen critical loads for natural and semi-natural ecosystems: the empirical approach. Water, Air and Soil Pollution. 1995;85:2413-2418.

Brunsting AMH. The influence of the dynamics of a population of herbivorous beetles on the development of vegetational patterns in a heathland system. Wageningen; 1982.

Chapman PJ, Edwards AC, Reynolds C, Cresser M, Neal C. The nitrogen content of upland rivers in Britain: the significance of organic N. IAHS Publ. 1998;248:443-450.

Faafeng B, Hessen DO. Nitrogen and phosphorus concentrations and N:P ratios in Norwegian lakes. Perspective on nutrient limitation.; 1993 pp. 505-507.

Fangmeir A, Hadwiger-Fangmeir A, Van der Eerden LJM, Jager HJ. Effects of atmospheric ammonia on vegetation - a review. Environmental Pollution. 1994;86:43-82.

Fowler D, Cape JN, Leith ID, Choularton TW, Gay MJ, Jones A. The influence of altitude on rainfall composition at Great Dun Fell. Atmospheric Environment. 1988;22:1355-1362.

Hessen DO, Hindar A, Holtan G. The significance of nitrogen runoff for eutrophication of freshwater and marine recipients. Ambio. 1997;26:312-320.