Unpolluted river, River Ticino, Switzerland

Source: Michael St Maur Sheil


INTRODUCTION

Human health and development are threatened in many places because of insufficient or poor quality water. Flooding is a serious problem in some countries, and destruction of aquatic habitats by channelisation, rough maintenance schemes and damming of rivers can lead to an overall impoverishment of native plant and animal species.

By tradition most evaluations of the quality of the aquatic environment have been based on measurements of a set of concentrations, speciations, and physical partitions of inorganic or organic substances in the water. Other elements such as the amount of water and the physical conditions of the waterbody have only recently been considered to be of equal importance as the water quality per se for determining the ecological quality of the aquatic environment.

Demand for water of good quality has increased with the advent of industrialisation and rapid population growth. This trend has continued over time and has become more widespread geographically. In addition to domestic and industrial use of water, other requirements have become increasingly important. These include improved personal hygiene, agricultural irrigation and livestock supply, hydropower generation, cooling water for power plants and industry, as well as recreational purposes such as boating, swimming and fishing. Each of these intentional water uses affects, more or less, the quality of the water. Together with increased intensity of water use, discharge of untreated domestic and industrial wastes, excessive application of fertilisers and pesticides in agriculture, and accidental spills of harmful substances (including radioactive substances) have led to increasing pollution of many European waterbodies: groundwaters, rivers, lakes, coastal areas and seas. Figure 5.1 illustrates in chronological order the sequence of water pollution problems found in European freshwaters since 1850. Most pollution problems have evolved unrecognised over time until they have become apparent and measurable. Recognition of a problem, therefore, took considerable time and control measures took, in most cases, even longer.

Not all water quality problems are due solely to human impacts. Locally, natural geochemical conditions may cause high content of reduced iron (eg, in the Russian Federation and Denmark), fluoride (in Moldova and Bavaria, Germany), arsenic and strontium (in some mountainous countries), and salts in the groundwater, reducing its use as a source of drinking water. Natural events like volcano eruptions and subsequent mud flows, floods and droughts can lead to serious local and regional deterioration of the aquatic environment. The impact of some of these events, however, can be made worse by human activities; for example, by landuse changes, deforestation and river channelisation.

Water pollution used to be primarily a local problem, with identifiable sources of pollution by liquid waste. Up to a few decades ago most of the wastes discharged to waters came from animal and human excreta and other organic components from industry. In areas with low population density without sewerage systems, such problems are to a great extent alleviated by the natural self-purification capacity of the receiving water. However, with the increasing urbanisation of the 19th and early 20th centuries, and subsequent expansion of sewerage systems without any or adequate treatment, liquid waste loads have become so large that the self-purification capacity of receiving waters downstream of large human settlements can no longer prevent adverse effects on water resources. The results of discharges of such materials include dying fish, offensive smells and the risk of infection. In addition, the widespread channelisation of rivers that took place over this period contributed significantly to the reduction of the natural self-purification capacity of rivers.

Over the years, the pollution load of most receiving waters has further increased. In addition to impacts from point sources, pollution from non-point (diffuse) sources, for example leaching and runoff from agricultural areas and long-range transported air pollutants, have become increasingly important. Consequently, the associated problems are no longer just local or regional, but have become continental in scope.

For Europe, no general overview of water quality exists. Therefore, it is the main objective of this chapter and the following one to compile a picture, as comprehensive as possible, of the present state of European groundwaters, rivers, lakes, coastal areas and seas. Although the information available in many cases is anecdotal, an evaluation of water quality trends has been undertaken, and the present state of the waterbodies has been related to natural processes, and to human activities in the catchments overlying the aquifers or draining to the waterbodies in question.

THE WATER RESOURCE

A thorough understanding of water in the hydrosphere is necessary if we are to appreciate its role in the Earth system and provide a solid basis for rational water management to meet human demands, while at the same time preserving the integrity of the environment. The International Hydrological Programme and the Hydrology and Water Resources Programme of UNESCO and WMO, respectively, have provided important contributions to knowledge about the hydrology and water resources of the Earth and its continents. Other significant contributions to the understanding of global water resources have been provided by Lvovich (1973), Baumgartner and Reichel (1975) and Shiklomanov (1991).

The hydrological cycle

The Earth's salt water and freshwater have been formed in the course of the evolution of the planet as a by-product of numerous chemical processes transforming rock matter at large depths. It penetrated to the surface of the Earth as water vapour in particular as a result of volcanic activity, after having leaked from the interior of the Earth.

In total, it has been estimated by UNEP (1991) that the Earth's total water amount equals approximately 1360 million km3 of which less than 3 per cent is present as liquid freshwater or locked up temporarily in ice caps. As shown in Figure 5.2, 8 million km3 are present as groundwater, 0.2 million km3 as fresh surface water, and 29 million km3 form the polar ice caps and minor glaciers elsewhere. For Europe the approximate figures are, according to original estimates and UNESCO (1978): 1 million km3 of groundwater (very approximate), 2580 km3 surface water (131 km3 in rivers, 2027 km3 in lakes and 422 km3 in reservoirs) and 4090 km3 of water locked up in glaciers.

Although the total amount of water on Earth is fixed on a short time-scale, the physical state of the water is continuously changing between the three phases (solid in ice, liquid, and as atmospheric water vapour), circulating through the different environmental compartments (ocean, atmosphere, glaciers, rivers, lakes, soil moisture and groundwater) and renewing the resources. The typical average renewal rates of water resources show considerable differences, ranging from thousands of years for the ocean and polar ice down to fortnightly or even weekly renewal of water in rivers and the atmosphere.

Average annual global evaporation from the ocean is six times higher than the evaporation from land (0.43 versus 0.07 million km3), whereas the ratio between precipitation over ocean and land is 3.5 (0.39 against 0.11 million km3), showing that approximately 40 000 km3 of water each year is transported from the ocean via the atmosphere to renew the freshwater resources. This amount of water constitutes the natural water resource that appears as global river runoff and which is potentially available for consumption each year.

Water resources

As shown in Figure 5.3, the annual average renewable water resources per capita show very large variability between different geographical regions in Europe. The populations of Nordic countries have in general between six and eight times as much water available for consumption per capita than the population of the other three geographical regions: Eastern, Southern and Western Europe.

Regional averages, however, provide only a very general measure of the adequacy of water availability, and obscure any variation between and within countries. It is interesting to examine the per capita distribution of available water resources in Europe, which is highly uneven, as shown in Map 5.1.

Relative water availability can be classified in the seven categories shown in Table 5.1.

Low, very low or extremely low water availabilities are found in 36 per cent of the European countries, in particular in the southern countries, with Malta having the lowest amount of water available of all European countries ­ 100 m3/capita per year. Densely populated Western countries with moderate precipitation (eg, Belgium, Germany, Denmark and the UK) also fall in these categories. Water availability is also low in some Eastern countries (Moldova and Ukraine), mainly due to low precipitation.

Thirty-two per cent of European countries fall in the category of medium availability.

For Europe as a whole, above medium, high or very high water availabilities are found in another 32 per cent of the countries. Plenty of water is found either in sparsely populated countries where precipitation is very high, like the Nordic countries (Iceland, for example, has more than 600 000 m3 of water available per year for each of its citizens), or in countries with large transboundary rivers running through them. For example, the per capita water availability in Spain is only one third of that for Portugal, which receives 48 per cent of its water in transboundary rivers from Spain. Many other countries are dependent on external contributions of water for meeting their demands. The importance of transboundary rivers as water suppliers is summarised in Figure 5.4 for those countries which receive a significant proportion (more than 10 per cent) of their water from other countries.

In this context, 20 European countries receive a high proportion of their water from transboundary rivers. Five countries ­ Hungary, Moldova, Romania, Luxembourg and The Netherlands ­ are heavily dependent on external inputs, receiving more than 75 per cent as external contributions, while three countries ­ Latvia, Ukraine and former Czechoslovakia (CSFR) ­ receive between 50 and 75 per cent of their renewable water resources from abroad. Bulgaria and Romania share the water resources of their boundary river, the Danube.

For countries with major rivers running through them, estimates of total renewable water resources tend to overestimate sustainable water resources. In humid, temperate climates, a good estimate of a country's sustainable freshwater resources is the long-term average river runoff generated within the country itself. Therefore, consideration of the spatial and temporal variability of river flows is of utmost interest for the assessment and management of European water resources.

River flow characteristics in Europe

River runoff is one of the main sources of freshwater from which the various water demands are satisfied.

There is considerable spatial and temporal variation in river flow across Europe. The seasonal flow regime of a Mediterranean catchment, with its winter high flows, dry periods in summer and frequent flash floods, is very different from that of an English lowland chalk stream, where flows vary relatively little over the year. They both have a very different seasonal regime from a catchment in Poland, which has a maximum flow following the spring snow-melt. Two factors, climatic and physical properties of the catchment, lead to such variations in seasonal flow regimes. In addition, European rivers also experience considerable variation from year to year. Finally river flow regimes are also affected by human activities, such as river impoundment and landuse changes.

European flow regimes are described here, with special emphasis on:

The information for this section of the report has been extracted primarily from four databases:

  1. the FRIEND database for Northern and Western Europe;
  2. the AMHY database for Alpine and Mediterranean Europe;
  3. the CORINE river flow database of the EC; and
  4. the GRDC database operated by the WMO.

Supplementary information has been obtained from hydrological yearbooks and monographs (Korzun et al, 1977; Stancik et al, 1988), or has been prepared by hydrological agencies across Europe specifically for this report.

Average annual runoff

The average annual runoff in Europe (Map 5.2) follows very closely the pattern of average annual rainfall ­ and topography. Annual runoff is greater than 4500 mm in western Norway, and decreases to less than 25 mm in parts of Spain, central Hungary and eastern Romania, and in large regions of Ukraine and the southern part of the Russian Federation. The greater variation of runoff in Western Europe, compared with Eastern Europe, reflects the greater variability in topography, and hence rainfall.

Across most of lowland Europe, between 25 and 45 per cent of rainfall runs off into waterbodies. In high rainfall areas, such as the Alps, western Norway and western Scotland, over 70 per cent of the rainfall may become runoff. In drier regions, particularly southern Spain, runoff may amount to less than 10 per cent of the annual rainfall.

The annual average runoff for Europe is estimated from Map 5.2 at 3100 km3 over a territory of 10.2 million km2, that is, 304 mm/year, or 9.6 l/second per km2. This is the equivalent of 4560 m3/capita per year for a population of 680 million. Compared to the present average total water abstraction of 700 m3/capita per year (1920 l/capita per day) in Europe, the immediate conclusion would be that Europe faces no water shortage problems.

Even when average runoff, evenly distributed, would suffice to cover the needs of water users, the natural variation from one season to another, or from year to year, can set constraints. The part of runoff which can serve as a continuous supply source, even under drought conditions, is obviously much less than average runoff, and depends primarily on the local hydrological regime. For most of Western Europe, it has been estimated that this runoff, taken as the mean annual minimum 10-day flow, is between 1 and 3 l/second per km2.

Seasonal river flow regimes

Variations in the distribution of river flows through the year define different flow regimes. All are based, to a large extent, on the timing of maximum and minimum flows, the number of flow seasons, and the origin of the river flows (Gottschalk et al, 1979; Krasovskaia and Gottschalk, 1992). Climate is usually the dominant influence on river flow regimes, but there are cases where the effect of physical characteristics of the catchment can dominate. Such 'atypical' regimes, with very little variability in flow from month to month, include those of catchments with very large lakes, or where river runoff is controlled by seepage from an aquifer.

River flow regimes vary considerably across Europe, but insufficient information is available to produce maps showing their geographical distribution.

The river flow regimes of large catchments (Map 5.3) can be different from those of small catchments. In particular they are much less variable because they integrate runoff over a large area, and can include sub-catchments with very different characteristics.

The differences in river flow regimes are apparent between Western Europe (where flows are at a minimum in summer and late autumn), the mountain-fed catchments (where flows are at a maximum in summer), and Eastern and Northern Europe (where most runoff occurs during the spring snow-melt period).

Inter-annual variability

The droughts of the early 1990s, which have been affecting several parts of Europe, illustrate clearly the variability in river runoff from year to year. From an analysis of runoff data from 14 gauging stations with long records (between 40 and 150 years) across Europe, three important conclusions can be drawn:

  1. Consistent patterns in the variability of river flow regimes from year to year exist across Europe. Droughts, for example, tend to affect large parts of Europe at the same time. The regional consistencies and differences between wide areas reflect the correlation with large-scale weather patterns.
  2. It appears that extreme years tend to cluster. Several low flow years occurred in succession in the early 1970s in Western Europe, followed by a period in the 1980s where several years were above average.
  3. There is no evidence of any consistent trend in river flow characteristics. Some records show upward or downward trends for a part of the record, but in each case these periods are followed by periods with no trend, or an opposing trend.

Human influences on river flow regimes

Many European river flow regimes are heavily affected by human activities, ranging from direct abstraction of water through regulation of flow regimes by reservoirs, to changes in catchment landuse. It is, however, difficult to generalise about the impacts of human activities on flow regimes. However, it is important to recognise that these impacts are not limited to areas with high population densities (examples are reservoirs and river channelisation).

Main sources of water for abstraction

Surface water is the main source for water abstraction by all utilisation sectors in Europe. On average, 70 per cent of total abstraction is drawn from this source, but with large variation between countries (Figure 5.5). Groundwater is by far the next most important source, and other sources include desalinated sea water.

Countries like Spain, Belgium, The Netherlands, Finland and Moldova, with insufficient groundwater supplies, abstract more than 90 per cent from surface water sources. However, in Cyprus, Switzerland, Slovenia, Iceland and Denmark, which are countries with extensive groundwater reservoirs, more than 75 per cent is drawn from groundwater.

The public water supply systems serve primarily domestic users and some industrial demands, and the source of water for this sensitive use is predominantly groundwater. As a source of public drinking water groundwater is of enormous importance. For Europe as a whole, about 65 per cent of the public supply is provided from groundwater (Zektser et al, 1992) which normally is of a better quality than surface water. On average, the total water abstraction in Europe amounts to approximately 480 km3/year (or 700 m3/capita per year), ranging from below 200 m3/capita per year in Luxembourg, Malta and Switzerland to above 1000 m3/capita per year in Bulgaria, Estonia, Lithuania and Spain.

Over the last two decades total water abstraction has in general increased in Europe (Figure 5.6). This trend, however, masks great variability between countries. Abstraction increases have been particularly marked in Southern European countries, but also in the majority of countries in Eastern and Western Europe more water was abstracted in the late 1980s than before. Stabilisation or even an abstraction decrease has occurred in some of these countries, including Austria, Bulgaria, The Netherlands and Switzerland. This is also the case in the Nordic countries: Sweden and Finland.

For Europe as a whole, 53 per cent of the abstracted water (surface and groundwater) is used for industrial purposes, 26 per cent in agriculture and only 19 per cent for domestic purposes (WRI, 1990).

There is large variability in sectoral water abstraction between countries (Figure 5.7). Interpretations of such statistics, however, should be carried out with caution because it is not always clear how they have been derived and what they comprise. Water use in the industrial sector is particularly difficult to evaluate, since it is not always clear whether cooling water and water used for power generation is included in the industrial share. Water used for these purposes often accounts for about 70 to 80 per cent of industrial water use, and most of this is used in power generation (see Chapters 19 and 20).

Agriculture's share is highest (30 to 70 per cent) in a number of Southern and Eastern European countries with low net precipitation and in Denmark and The Netherlands where agriculture is very intensive (see Chapter 22). Water use in agriculture is mainly for irrigation and livestock.

Water abstraction for public use shows less variability between European countries than other sectoral abstractions, ranging between 15 and 25 per cent of total abstractions in the majority of countries to more than 40 per cent in Luxembourg and the UK.

The intensity of water use in a country can, according to the OECD (1992), be represented as the percentage abstraction of total available water resources, which includes internally generated water and inputs from neighbouring countries via transboundary rivers (see Figure 5.4).

The calculated water use intensity indicators cover a very wide spectrum, from less than 0.1 per cent in Iceland to over 70 per cent in Belgium, with a general average of around 15 per cent. If this indicator were calculated by relating abstractions exclusively to internal resources (Table 5.2), its values would be appreciably higher in those countries where total resources are, to a considerable extent, accounted for by external resources.

This is particularly true for Belgium, Bulgaria, Hungary, Moldova and The Netherlands, where water abstraction approximately equals total internal resources, or even exceeds it, as in the case of The Netherlands. These countries, as well as Romania, Lithuania, Portugal and Ukraine, are therefore particularly sensitive to upstream impacts affecting the quantity and quality of the water they receive through transboundary rivers.

GROUNDWATER

Characteristics and distribution

Groundwater is an important element in the Earth's hydrological cycle. It remains one of the least-studied and most difficult water resources to determine. Natural groundwater resources are stored in aquifers, which are permeable rock formations or unconsolidated deposits, chiefly gravels, sands and silts.

The main characteristics of groundwater systems are:

In spite of its non-visibility, groundwater has very important functions, including economic, ecological and those relating to public health ­ which are not always fully recognised. For example, groundwater is an important source for drinking water. Human activity, however, can have a great effect on quantity and quality of the available groundwater resources. Due to the aforementioned characteristics, groundwater systems are normally very stable, in both quantity and quality. However, the effects of pollution and overexploitation will accumulate over time. In general, the periods of recovery will be centuries and decades, respectively.

The availability of groundwater (as a natural resource) is limited by three factors:

  1. the total amount of recharge (renewal of groundwater), resulting from precipitation, evapotranspiration, infiltration and seepage from rivers and lakes;
  2. the quality of the recharged water; and
  3. the properties of the soil and aquifer (permeability, porosity, etc).

As with surface water, there is an uneven distribution of natural groundwater resources within Europe. The geographical distribution of different types of aquifers is shown in Map 5.4. A note on how to interpret the maps appears in Box 5B.

Recharge and loads

The renewal of groundwater resources occurs through natural and artificial recharge. The natural recharge is the amount of water available for percolation into the aquifer as a result of excess precipitation and runoff. The precipitation excess is the amount of precipitation less evapotranspiration.

Artificial recharge is the result of excess irrigation or of human-induced recharge (forced feeding) of aquifer systems. Although irrigation is common in large parts of Europe, systematic data on water sources and quantities used are not available. As a result, it was not possible to quantify the effects of irrigation on the groundwater systems. The essential objective of forced feeding recharge is to transform surface water with a periodically unreliable quality into a safe source for water supply. It can be done by applying natural gravity forces, for example, by infiltration canals, or by damming rivers, or via injection wells. The importance of forced feeding could not be assessed due to the lack of systematic data, but projects are in general limited to a local scale. Forced feeding of aquifers is currently being used in a number of European countries (eg, river bank infiltration in Germany and The Netherlands, and surface water infiltration in the Uppsala area of Sweden, and in dunes in The Netherlands). In Germany, for example, these techniques together account for 15.7 per cent of public water supply (Umweltbundesamt, 1992).

All groundwater contains natural chemical constituents in solution. During its passage underground, the water dissolves and deposits various substances, while other solutes are being transformed or degraded. Most of these changes tend to be slow but long-lasting. Pollution of the percolating water due to human activities sometimes has a severe impact on the constituent load reaching the groundwater.

In many places, the natural quality of groundwater is degraded because of human activities. Basically, two different sources of groundwater pollution can be recognised: non-point (diffuse) sources and point sources. The diffuse sources are mainly from agriculture (eg, irrigation excess, animal wastes, fertilisers, pesticide residues), urban areas (storm drainage) and contaminant deposition from the atmosphere. The point sources comprise municipal and industrial activities, such as surface disposal of liquid and solid waste, improper storage of materials used in manufacture, sewer, tank and pipeline leakages, and activities related to mining (especially mine tailings) and oil exploitation.

Groundwater quantity

During recent centuries, human-made changes in the hydrological cycle have had enormous impacts on groundwater levels and flows. These have changed for many reasons, including pumping of groundwater for domestic, industrial and agricultural water supply, provision of cooling water, mine pit drainage, intensified and deeper drainage of agricultural lands, changing natural land into agricultural and urban areas, and regulation of surface waters.

These changes have important economic and ecological effects. Economic effects can include crop and industrial production as well as the production of drinking water. Groundwater-related terrestrial ecosystems are notably affected by changing groundwater levels, while at the same time wetlands are becoming more scarce. Moreover, groundwater is the last remaining water source for rivers and small surface waters during dry periods. If the base flow in the rivers is not maintained, severe damage can be caused to aquatic ecosystems, and the economic functions of the rivers ­ such as navigation and water supply ­ will also suffer.

Abstractions

In many areas groundwater abstractions exceed the recharge, and the aquifer becomes overexploited (Map 5.5), leading to a systematic and in many cases ongoing lowering of the groundwater levels. This results in a wide range of problems: damage to wetlands, drying up of springs and upper river reaches, reduction of river flows, salt water intrusion (the flow of salt water into an aquifer) and settling phenomena and damage to buildings.

Reported cases of overexploitation include abstraction for drinking and industrial water supply, irrigation water and drainage of mines. Drainage of waterlogged agricultural soils (see Chapters 7 and 9) and some river engineering works also lead to lowering of the groundwater level. Many cases of lowering of the groundwater table up to several dozens of metres have been reported, and in a number of cases the lowerings are in the order of hundreds of metres, affecting the groundwater level in an area up to 500 km away from the overexploited site. Major lowerings of the groundwater table caused by mining activities have, for example, been reported for the area around Kharkov (Ukraine), Lille (France) and in the Ruhr basin (Germany).

However, due to their requirement for water, most overexploited sites are found in or near large urban and industrial centres. Thus, about 60 per cent of the European cities with more than 100 000 inhabitants (or a total of approximately 140 million people) are located in or near areas with groundwater overexploitation. The consequences of this can be far-reaching, as shown by the recent severe problems in Spain and Greece, where cities have been given restrictions on the use of water. Rivers and marshes also have dried out, leading to loss of wetlands. An estimated 6 per cent of the area with aquifers suitable for abstractions is presently overexploited. This was estimated by overlaying the aquifer typology map (Map 5.4) with the overexploitation map (Map 5.5).

Salt water intrusion in aquifers may result from groundwater overexploitation along the coast. As urban, tourist and industrial centres are commonly located in this zone, intrusion of salt water is a problem of many coastal regions, especially along the Mediterranean, Baltic and Black Sea coasts.

A systematic European inventory of human-induced lowerings of groundwater levels and other changes in the groundwater systems is not available. Therefore, the information presented in Map 5.5 is based largely on problem cases. The risk of underestimating the extent of the problem is thus considerable.

Potential wetland damage

Wetlands, or wet ecosystems, are considered very important nature reserves, with a high ecological value (see Chapter 9). Many wetlands are located in areas prone to flooding such as river floodplains and the systems are highly dependent on the shallow depth of the water table, making them very sensitive to minor changes in the groundwater level. A tragic example is the Tables de Daimiel nature reserve in Spain (Custodio, 1991). The two main rivers feeding the area are fully dependent on groundwater but the groundwater resources in the area have been severely overexploited for irrigation, and at the beginning of 1993 the nature reserve had almost completely dried out.

An estimate of the major European wetlands being threatened by lowering of the groundwater level is shown in Map 5.6, obtained by combining a map of these areas with the inventory of overexploited areas.

Map 5.6 is based on data for major bogs, mires and waterbird sanctuaries in Europe (Goodwillie, 1980; Grimmett and Jones, 1989), and forms the basis of the inventory of the most important wetlands (total area 55 000 km2). As can be seen, about 25 per cent of the wetlands are potentially endangered.

Raising of groundwater levels

The water table has been rising due to reduced groundwater abstractions by industry in some cities and industrial areas (eg, Paris, London, Birmingham, Liverpool). Rises in groundwater level can cause problems like flooding of tunnels and basements, chemical attack on structural materials and increased humidity in houses (UK DoE, 1992). Also, increased volumes in sewers may decrease their effective capacity. Similar effects may result from large hydro-technical constructions changing the groundwater level in the drainage basin (eg, in the Russian Federation and Ukraine) and from artificial recharge and irrigation. Another example can be found in the Russian Federation around the Caspian Sea, where the rise in groundwater level was caused by a sea level rise of a few metres. In Ukraine raising groundwater level, in an area of approximately 25 000 km2 in the northern and central parts of the country, has caused problems in 2000 towns and settlements. The groundwater rise, which was caused by excess irrigation, produces unfavourable geological processes like displacements, landslides and soil salinisation.

Groundwater pollution

The main pollution threats to groundwater are:

Point sources

Soil pollution on urban, industrial, mining, military and landfill areas can affect the quality of groundwater, either directly or after a delay in time. In Table 5.4 potential contaminants are presented with respect to point sources. Severe problems of groundwater pollution have already occurred (Brömssen, 1986).

Actual data on polluted groundwater sites in Europe are very scarce. The area of groundwater potentially polluted by industry, mining, military activities and landfills (Table 5.5) can be estimated using various assumptions and by extrapolating available data (Franken, forthcoming). It is thus estimated that potential groundwater pollution by point sources occurs in less than 1 per cent of the European territory. Although this seems to be a rather insignificant portion, it should be borne in mind that most groundwater abstraction takes place in the vicinity of the areas potentially contaminated, posing a threat to the public water supply.

Nitrate leaching

The use of manure and fertilisers can lead to leaching of nitrate, ammonium, sulphate, potassium and, to a lesser extent, phosphorus into the groundwater, and hence into surface water. Because nitrate in water above certain concentrations can be a danger to human health and adversely affects the stability of aquatic ecosystems, this report focuses only on this contaminant. Whether or not nitrate will leach to the groundwater depends primarily on the time and rate of application, crop uptake, the soil type and climatological conditions. In Europe, there is a lack of nitrate monitoring data for groundwater. Therefore, a model was used to generate a European overview of nitrate leaching.

Taking into account the present rate and spatial distribution of nitrogen application (see Chapter 22), together with the soil type and climate, a model (based on Drecht, 1991) was used to predict the nitrate leaching from the root zone of agricultural land in Europe. Agricultural land is taken as the total of arable land, grassland and land used for permanent crops. The results of the calculations are presented in terms of the values 10, 25 and 50 mg NO3/l. The latter two values are specified in the Council Directive (80/778/EEC) relating to the quality of water intended for human consumption as guide level (25 mg NO3/l) and maximum admissible concentration (50 mg NO3/l), respectively.

Map 5.7 shows the areas with computed nitrate concentrations exceeding the guide level (GL) and the maximum admissible concentration (MAC) at one metre below the soil surface where concentrations would be at their maximum. Model computations indicate that over 85 per cent of the agricultural area in Europe has nitrate levels above the GL, and also that the MAC is exceeded below approximately 20 per cent of the agricultural area. The red spots on the Kola Peninsula are due to lacking information on regionally distributed nitrogen input to agricultural land in that area.

Several processes influence the fate of nitrate as it percolates from the root zone to the groundwater. Denitrification is particularly important in this respect, leading to a decrease of nitrate concentrations between the root zone and groundwater level (see Chapter 14 on emissions to water) as well as in the deeper groundwater, provided that (bio)chemical conditions are favourable.

The physical processes affecting the fate of nitrate involve mixing of 'old' resident groundwater with percolating 'young' water from upper soil layers. As a result of mixing of two types of water with different nitrate concentrations (lower in groundwater than in the percolating water), the nitrate concentrations in abstraction wells and seepage areas supporting wetlands, rivers and lakes are generally well below the leachate concentrations and/or their increase is retarded in time.

In order to obtain an overview in terms of a 'hot spots' map of the areas in Europe where groundwater is expected to be particularly vulnerable to nitrate contamination arising from agricultural activities, the nitrate leaching map (Map 5.7) was superimposed on an 'aquifer-risk map' prepared for, but not shown in, this report. The rationale behind the calculation of the risk map is explained in Box 5D. This analysis indicates that 'hot spots' for nitrate pollution (Map 5.8) can be expected primarily in areas where intensive cultivation of sandy soils takes place on top of unconsolidated, unconfined aquifers. For instance, this is the case for extended areas of several north-western European countries, the Czech Republic, the Slovak Republic, Hungary, Ukraine and Belarus.

Monitoring data on nitrate in groundwater supplied by countries were very heterogeneous. About 60 per cent of the countries replying to the questionnaires indicated that they had serious nitrate problems and the majority of these countries gave some regional details showing the geographical distribution of the problem areas and actual concentration measurements. Some countries (eg, Poland) reported quality data for the upper groundwater, while for others (eg, Spain) concentration values for deep groundwater have been used. In other cases, no specifications at all were provided about the monitoring depth, and quality data for both upper and deeper groundwater may therefore have been mixed. For large parts of Europe there is a complete lack of monitoring data on groundwater pollution by nitrate. A standardised groundwater monitoring network across Europe (based on unified and agreed principles for network design, site selection, sampling frequency, choice of determinants, etc) would have enabled a much clearer picture of the problems to be presented.

A comparison of the nitrate 'hot spots' map based on model calculations (Map 5.8) with the data on nitrate pollution leads to the following conclusions:

Therefore, it is concluded for nitrate that the 'hot spots' map (Map 5.8) is in reasonable agreement with the available information on groundwater quality.

Further comparison of the nitrate 'hot spots' map with the nitrate problem data suggests that:

In general it may be concluded that, despite the generalisations and assumptions attached to the preparation of the nitrate 'hot spots' map, the application of nitrogen (manure and fertilisers) in agriculture does lead to nitrate contamination (occasionally severe contamination) of European groundwater.

Pesticides

Approximately 600 different pesticides (as broadly defined to include herbicides as well as fungicides and insecticides) are applied in European agriculture, silviculture and horticulture (see Chapter 17). On their passage through the subsurface environment, these 600 active ingredients are transformed into an (often) unknown number of degradable products (residues). The effects of active ingredients and their residues on non-target terrestrial organisms (side effects), their fate in the soil and their undesired effects in groundwater and surface waters are far from known with certainty for every substance.

The extent to which pesticides in groundwater have reached concentrations limiting the use of groundwater for drinking water abstraction is essentially unknown. Only a few pesticide measurements in groundwater are available for a restricted number of constituents. The information about pesticide use per country (amounts of pesticide) available for this assessment was also limited. The modelled applications of pesticides to agricultural soils are shown in Chapter 22. Taking into account the available data on the present rate and spatial distribution of pesticide application, the organic matter content of the soil, and climate, a model (Boesten and van der Linden, 1991) was used to predict the pesticide leaching from the root zone of agricultural soils. Based on this model, the computed area with pesticide levels exceeding the EU drinking water standard (maximum admissible concentration of 0.5 µg/l for the total amount of pesticides and their residues as specified in Council Directive (80/778/EEC)) at one metre below the soil surface is shown in Map 5.9.

Based on the model, the EU standard is exceeded in the leachate under 75 per cent and 60 per cent of the total arable and permanent crop land in EU/EFTA and Central and Eastern European countries, respectively. The use of pesticides on grasslands was not taken into account because (a) application on grassland does not in general occur and (b) if it does occur the application rate is low.

The pesticide 'hot spots' map for groundwater (Map 5.10) was produced following the procedure as described above for nitrate, that is, using an overlay of the pesticides leaching map (Map 5.9) and the 'aquifer-risk map' (not shown). The expected 'hot spots' are in areas where intensive cultivation of coarse textured soils, low in organic matter, occurs on top of unconsolidated, unconfined aquifers. This is the case in (parts of) Denmark, northern France, The Netherlands, Lithuania and Belarus. The most extensive risk of high pesticide contamination is estimated to occur in Belarus.

There is an almost complete lack of systematic monitoring data for pesticides in Europe. Where measurements are available, they are mostly incidental (not part of a systematic monitoring network) and the density of monitoring points is very low. Often no specifications are given about the exact location of the measurement site and the monitoring depth. In general, only a few compounds (mostly fewer than ten) have been included in monitoring programmes.

Comparison of the calculated pesticide 'hot spots' map (Map 5.10) and the areas with 'observed' pesticide problems leads to the following considerations. For the Western European countries, the calculated 'hot spots' (eg, in France, The Netherlands, Denmark, the UK, Germany and Italy) show a reasonable fit with observations. Moldova and Ukraine have reported serious pesticide problems, but neither of these countries has indicated the specific areas where the problems occur. The 'hot spots' map shows only a small part of Ukraine having a high aquifer risk for pesticide pollution. One obvious explanation for the apparent mismatch is the lack of regionally distributed data on observed pesticide problems within Moldova and Ukraine. Another explanation is that in these countries very mobile pesticides have been used, which may lead to groundwater pollution not accounted for by the model used to generate the pesticide 'hot spots' map, which is based on the average mobility of pesticides. Finally, Map 5.10 shows that a large 'hot spot' area was calculated for Belarus. Unfortunately, no observations were available for Belarus for verification of the calculated 'hot spot' area. Taking into account the above considerations, it is assumed that the pesticides 'hot spots' map (Map 5.10) is in reasonable agreement with the available information on groundwater quality with respect to pesticides.

In summary, it may be concluded that, despite the generalisations and assumptions attached to the preparation of the pesticide 'hot spots' map, the application of pesticides leads to pesticide contamination of groundwater in many European countries.

As previously indicated, very few measurements of pesticides are available for groundwater. The main reason is the high cost of chemical analyses, combined with the high number of chemical compounds. Out of approximately 600 pesticides, only about 30 have been monitored. These 30 pesticides are neither the most mobile, nor representative of all pesticides. The availability of suitable analytical methods is also an important limiting factor (see also Chapters 7 and 17).

Acidification

Acidification of the soil is a well documented and serious problem in large parts of Europe (see Chapters 7 and 31). In addition to natural acidification, sandy and poorly buffered (low alkalinity) soils are subject to enhanced acidification caused by atmospheric deposition of sulphur and nitrogen compounds, fertiliser application and land drainage. Because water must filter through the soil before it reaches the aquifer it is likely that groundwater below those areas may also have become acidified. Unlike surface waters (treated in detail later in this chapter) there are very little data available to show the present extent of groundwater acidification and its trends in Europe and only a few examples of documented groundwater acidification can be presented.

Acidification of forest soils is, in particular, a well described and generally recognised problem in Northern and Central Europe (Last and Watling, 1991). It is caused primarily by atmospheric deposition of acidifying compounds emitted from human activities (see also Chapters 4, 7 and 31). As a result, increased concentrations of aluminium, sulphate, hydrogen ions, and sometimes nitrate are reported in the upper groundwater layer below sandy soils in Norway, Sweden, Denmark, Finland, the Czech Republic, the Slovak Republic, Germany and The Netherlands.

In Sweden, for example, a national groundwater survey showed severe influence of acid inputs on wells in forested areas (Bernes and Grundsten, 1992). Thus more than 50 per cent of surveyed wells in 14 municipalities (in total 52 municipalities were surveyed) showed very strong or strong acidification effects in the mid-1980s.

Acidification of deeper groundwater under agricultural land has also been reported (Targbill, 1986; RIVM/RIZA, 1991; Brömssen, 1986; Overgaard, 1986), but in these cases acidification cannot be explained solely by atmospheric deposition. Other processes (such as oxidation of ammonia to nitrate by nitrification) can also contribute to acidification (RIVM, 1992a; Rebsdorf et al, 1991), as can denitrification if nitrate is reduced by ferrous sulphide (see Chapter 14).

In Denmark, for example, an analysis of a 30-year time series (period 1952­82) of pH, alkalinity and nitrate in groundwater from unconfined aquifers below predominantly agricultural land in the western, sandy part of the country showed that pH and alkalinity had declined steadily over time, whereas nitrate had increased. The average yearly decrease was 0.023 units for pH and 33 µmol/l for alkalinity. For nitrate the increase was 14 µmol/l (Rebsdorf et al, 1991). If alkalinity continues to decrease at the present rate the groundwater in that part of Denmark would completely lose its buffering capacity within 20 to 25 years. This would make the groundwater unsuitable for human consumption and almost all other purposes, unless expensive water treatment processes are introduced.

The most serious consequences of acidification of groundwater are the increased mobilisation of trace elements, especially aluminium, and the increased solubility of some metals in water distribution systems, both resulting from a lowering of the pH (Chapter 7).

RIVERS, RESERVOIRS AND LAKES

Introduction

This section focuses on European inland surface waters, that is, rivers, lakes and reservoirs, and assesses their environmental state and the environmental problems caused by human activity. It is one of the first attempts at an all-European assessment aimed at providing a general overview as a basis for implementing measures to improve the environmental state of rivers and lakes, and to identify areas with environmental problems.

Considerable environmental information on rivers and lakes is currently collected and reported by various regional or national authorities. However, as the European continent covers about 10 million km2 and there are several million kilometres of flowing waters and more than a million lakes, this information is very heterogeneous, and therefore difficult to collate on a pan-European basis. Nevertheless, a first attempt at doing so has been made in this section. The primary focus is on frequently measured water quality parameters (eg, organic matter in rivers, nutrients in rivers and lakes, and acidification of rivers and lakes) since the wide geographical coverage makes these variables well suited to illustrate the general environmental state of European inland surface waters. Some environmental problems related to more rarely monitored variables are also described.

Characteristics of European rivers, reservoirs and lakes

A river is a system comprising both the main course and all the tributaries that feed into it; the area that the river system drains is known as the catchment. The main characteristic of rivers is their continuous one-way flow in response to gravity. In addition, because of changes in physical conditions such as slope and bedrock geology, rivers are dynamic and may change nature several times during their course (eg, from a fast-flowing mountain stream to a wide, deep, slowly flowing lowland river).

When assessing river characteristics and water quality it is important to bear in mind that a river comprises not only the main course, but also a vast number of tributaries. Thus although the main course of Europe's largest river, the Volga, is 3500 km long, it receives water from ten tributaries each longer than 500 km, and more than 151 000 tributaries each longer than 10 km (Fortunatov, 1979).

Rivers are greatly influenced by the characteristics of the catchment area (Figure 5.9). The climatic conditions influence the water flow, as does bedrock geology and soil type. The latter also affects the mineral content of the river water. Human activity affects river systems in numerous ways, for example, through afforestation or deforestation, urbanisation, agricultural development, land drainage, pollutant discharge, and flow regulation (dams, channelisation, etc). The lakes, reservoirs and wetlands in a river system attenuate the natural fluctuation in discharge and serve as settling tanks for material transported by the rivers. For example, whereas the water of the Rhine is very muddy and turbid when entering the Bodensee, it is clear and transparent when leaving. Water flow and water quality are therefore the net result of the various characteristics of the catchment.

Lakes are bodies of standing water that is usually fresh, but which may also be brackish. Although lakes may be characterised by physical features of the lake basin, such as lake area and water depth, the characteristics of the catchment are important when describing the lake environment. Nutrient loading of a lake is determined not only by the bedrock geology and soil type in the catchment, but also by the human activity.

Reservoirs are human-made lakes created to serve one or more purposes. As their water residence time is generally relatively short, and as the water level fluctuates much more widely and frequently than in natural lakes, they can be regarded as hybrids between rivers and lakes.

European rivers

On average, European rivers discharge a total of 3100 km3 of freshwater to the sea each year, about 8 per cent of total world discharge . Because Europe has a temperate humid climate and a high percentage of limestone in the surface rock, the weathering rate is the highest of all the continents; as a result, 12.6 per cent of all dissolved solids discharged to the oceans are derived from Europe (Kempe et al, 1991). That Europe is relatively densely populated and has a high proportion of agricultural areas also affects the concentration of dissolved substances in river water; thus the median nitrate level is 1.8 mg N/l in European rivers as compared with only 0.25 mg N/l in non-European rivers (Meybeck et al, 1989).

Major European river catchments

In proportion to its land area, Europe has the longest coastline of all continents. As it is a relatively young and structured continent, geologically, river catchments are numerous but relatively small and rivers are short (Map 5.11). About 70 European rivers have a catchment area exceeding 10 000 km2, and only rivers arising deep inside the continent are relatively large. The three largest rivers in Europe, the Volga, the Danube and the Dnepr, drain one quarter of the continent, but are only small by world standards, their catchments ranking 14th, 29th and 48th, respectively (Showers, 1989).

The 31 largest European rivers, all of which have catchments exceeding 50 000 km2, drain approximately two thirds of the continent. More than half of these rivers have their catchment area in the European part of the former USSR. The major rivers flowing north into the Barents Sea and the White Sea are the Severnaya (Northern) Dvina and the Pechora. The Volga and the Ural which flow south and the Kura which flows east drain into the Caspian Sea while the Dnepr and the Don drain south into the Black Sea. The largest river to discharge into the Black Sea is the Danube, which has its catchments in 16 countries of Central Europe and the Balkans. The main rivers to discharge into the Baltic Sea are the Neva, the Wisla, the Oder and the Neman. Ten rivers with catchments larger than 50 000 km2 drain into the Atlantic and the North Sea, with the Rhine, the Elbe, the Loire and the Douro being the largest. The European rivers that drain into the Mediterranean are relatively small, the Rhone, the Ebro and the Po being the largest. Nevertheless, since the damming of the Nile, the Rhone has become the Mediterranean's most important freshwater source (Kempe et al, 1991).

Major rivers in European countries

Countries whose coastline is long in relation to their area, for example Iceland, the UK, Ireland, Norway, Sweden, Denmark, Italy and Greece, are usually characterised by having a large number of relatively small river catchments and short rivers, the three to four largest of which drain only 15 to 35 per cent of their area (Table 5.9). The population tends to congregate in towns along the coastline and wastewater is consequently discharged directly into coastal areas rather than into the river systems.

Many European countries are drained by only a few river catchments; thus the Wisla and Oder drain more than 95 per cent of Poland, and the Danube drains most of Hungary, Romania and Slovenia (Table 5.9).

European lakes and reservoirs

Many natural European lakes appeared 10 to 15 thousand years ago, being formed or reshaped by the last glaciation period, the Weichsel. The ice sheet covered all of Northern Europe, but in Central and Southern Europe it was restricted to the mountain ranges. As a rule the regions that have many natural lakes are those that were affected by the Weichsel ice. Norway, Sweden, Finland and the Karelo-Kola part of the Russian Federation have numerous lakes that account for approximately 5 to 10 per cent of national surface area. Large numbers of lakes were also created in the other countries around the Baltic Sea, as well as in Iceland, Ireland and the northern and western parts of the UK. In Central Europe most natural lakes lie in mountain regions, those at high altitudes being relatively small and those in the valleys being the largest, for example Lac Léman, Bodensee, Lago di Garda, Lago di Como and Lago Maggiore in the Alps and Lake Prespa and Lake Ohrid in the Dinarian Alps. Exceptions are the two large lakes, Lake Balaton and Neusiedler See, that lie on the Hungarian Plain.

In contrast to glaciation, processes such as tectonic and volcanic activity have played only a minor role in the formation of European lakes. Numerous lakes have been created by natural damming of rivers and coastal areas, however.

Countries that were little affected by the glaciation period, such as Portugal, Spain, France, Belgium, southern England, central Germany, the Czech Republic, the Slovak Republic and the Central European part of the Russian Federation, have few natural lakes. In these areas human-made lakes such as reservoirs and ponds are often more frequent than natural lakes. Many river valleys have been dammed to create reservoirs, and a large number have been built in mountain ranges for use by the hydroelectric industry. In several countries, for example The Netherlands, Germany, France and the former Czechoslovakia, numerous artificial small lakes have been created by other human activities such as peat and sand quarrying, and for use as fish ponds.

Natural lakes in Europe

There are more than 500 000 natural lakes larger than 0.01 km2 (1 ha) in Europe; of these about 80 to 90 per cent are small, with a surface area between 0.01 and 0.1 km2, and only about 16 000 have a surface area exceeding 1 km2 (Table 5.10). Three quarters of the lakes are located in Norway, Sweden, Finland and the Karelo-Kola part of the Russian Federation.

The approximate number and size distribution of natural lakes is shown for each country in Table 5.11; however, the number of small lakes is somewhat uncertain, and the figures given are generally minimum estimates.

Human-made lakes

Reservoirs are the most important human-made lakes in Europe, there being more than 10 000 major reservoirs covering a total surface area of more than 100 000 km2. The numbers of relatively large reservoirs are greatest in the Russian Federation (ca 1250), Spain (ca 1000), Norway (ca 810) and the UK (ca 570). Other countries with a large number of reservoirs are Hungary (ca 300), Italy (ca 270), France (ca 240) and Sweden (ca 225). Many European countries have numerous smaller human-made lakes, for example Latvia, Bulgaria and Estonia, which have about 800, 500 and 60, respectively.

Large lakes and reservoirs in Europe

There are 24 natural lakes in Europe that have a surface area larger than 400 km2, the largest being Lake Ladoga, which covers an area of 17 670 km2 (Map 5.12). The latter is located in the northwestern part of the Russian Federation, together with Lake Onega, the second largest lake in Europe. Both are considerably larger than other European lakes and reservoirs, but nevertheless rank only 18th and 22nd in world order (Herdendorf, 1982). The third largest European freshwater body is the 6450 km2 Kuybyshevskoye reservoir on the Volga. Another 19 natural lakes larger than 400 km2 are found in Sweden, Finland, Estonia and the northwestern part of the Russian Federation, and three in Central Europe ­ Lake Balaton, Lac Léman and Bodensee, the surface areas of which are 596, 584, and 540 km2, respectively (Map 5.12).

The six largest reservoirs are located in the Volga river system in the Russian Federation, the two largest being the 6450 km2 Kuybyshevskoye and the 4450 km2 Rybinskoye reservoirs. Of the 13 European reservoirs with an area exceeding 1000 km2, only the Dutch reservoir Ijsselmeer lies outside the Russian Federation and Ukraine.

Deep and shallow lakes and reservoirs

Lake water depth is an important parameter with which to characterise the lake environment. It is determined largely by the surrounding topography, lakes in mountainous regions generally being deeper than those in lowland areas. In two lowland countries, Finland and Poland, most lakes have a mean depth of 3 to 10 m; lakes with a mean depth greater than 10 m are rare. In Austria and Switzerland, in contrast, large shallow lakes are virtually absent, and most lakes have a mean depth greater than 25 m. As with natural lakes, the deepest reservoirs are located in mountainous regions of countries such as Norway, Spain, France, Scotland and Greece. Examples are the 190 m deep Spanish reservoir Almendra, the 132 m deep Greek reservoir Kremasta, and the 125 m deep Norwegian reservoir Blåsjø (these are maximum depths).

Regulation of European rivers

River regulation is a general term describing the physical changes that people impose on watercourses. Various human activities that physically influence rivers are listed in Box 5F.

Many of the rivers in Europe have now been regulated; in some countries there are very few unregulated rivers (Brookes, 1987; Petts, 1988; Garcia De Jalon, 1987). River regulation has been undertaken to the greatest extent in Western and Southern Europe. Thus in countries such as Belgium, England, Wales and Denmark, the percentage of river reaches that are still in a natural state is low, typically 0 to 20 per cent. By contrast, in countries such as Poland, Estonia and Norway, many rivers still have 70 to 100 per cent of their reaches in a natural state.

River regulation often causes major changes in river processes, primarily the flow regime and the transport of dissolved and particulate matter. The effects are seen not just locally, but may be extensive, with downstream reaches nearly always being affected, and upstream reaches and the surrounding areas often being affected as well. Some types of river regulation, for example land drainage, may affect much if not all of a catchment area, but that which has the most widespread and marked effect is the construction of reservoirs. Nevertheless, many river systems have also been changed by channelisation, especially those in lowland areas.

Reservoirs

Because reservoirs usually have a relatively short water residence time ­ often less than a year and sometimes just a few days ­ they can be regarded as a hybrid between a river and a lake that is internally divided in three zones: a river-like zone at the inflow end, a lake-like zone in the outflow end, and a transitional zone in between. Another prominent feature of many reservoirs is that the water level fluctuates, so that the littoral zone is biologically poorly developed where the fluctuations exceed that of the natural lakes in the region.

Reservoirs have been constructed in Europe for thousands of years; the earliest were relatively small, and used mainly for domestic water supply and crop irrigation. During the last two centuries there has been a marked increase in both reservoir size and number, with large storage capacity reservoirs constructed in many countries, especially the former USSR. Thus there are currently about 3900 large reservoirs with dams higher than 15 m in Europe (not including the territory of the former USSR), half of which have been built since 1961 (Boon, 1992). To this must be added the many large reservoirs in the European part of the former USSR, and the thousands of smaller reservoirs and ponds spread throughout Europe.

The development of reservoir construction in Europe can be illustrated using the UK and Spain as examples (Figure 5.10). In the UK the number of reservoirs grew rapidly during the second half of the 19th century from about 50 to about 200; from then until the mid-1970s, the rate was about six new reservoirs per year (Boon, 1992). In Spain the number of reservoirs grew at the rate of about two per year between 1900 and World War 2, but at a rate of about 20 per year in the post-war period from 1950 to 1980 (Garcia De Jalon, 1987; Riera et al, 1992). Reservoir construction in Europe has now fallen off and growth in total reservoir area seems to be stagnant. This is mainly because of the lack of suitable sites (Williams and Musco, 1992).

Reservoir usage

Reservoirs are usually built to serve several purposes, the primary uses typically being the generation of hydroelectric power, irrigation, flood control, and domestic and industrial water supply. Other uses are commercial fishery and various recreational activities.

Environmental problems related to reservoirs

Reservoir construction leads to a number of environmental problems, both during the building phase and following completion. As the water level in the reservoir rises upon the closing of the dam, major changes often take place in the area to be inundated; farmland can be lost, settlements flooded, and the groundwater table elevated. Once the reservoir has been established, the environmental problems can be divided in two groups: those that render the reservoir unsuitable for its purpose, for example algae and toxic substances in reservoirs used for drinking water, and those that induce ecological deterioration of the river system, especially downstream of the reservoir (Box 5G).

The water quality of a reservoir, as reflected by the content of pathogens, toxic chemicals and poisonous algae, is of primary concern when the reservoir is used for drinking water, commercial fishery, industrial processes, and recreational activities such as bathing and water sports. At present it is not possible to give a general overview of the extent to which European reservoirs are contaminated with toxic substances. However, contamination of reservoirs with oil, organic solvents, heavy metals and radionuclides has been reported to be a problem in the Russian Federation and in Ukraine (Mnatsakanian, 1992; Gavrilov et al, 1989). Many of the environmental problems to which reservoirs are subject, such as eutrophication and heavy metal, organic and thermal pollution, also occur in natural lakes and rivers.

Since reservoirs interrupt the natural continuity of a river, the ecological consequences can be manifold. Access to spawning sites for migratory fish is prevented, the problem being especially acute for fish such as salmon, trout and sturgeon. As reservoirs trap the suspended matter flowing into them, they reduce the suspended matter load to downstream reaches; in contrast, because of their high biological productivity, the organic load to the downstream reaches may increase. Since reservoirs regulate the water flow, sudden flow fluctuations may occur downstream of reservoirs used for hydroelectric power generation. In other cases reservoirs may have a stabilising effect on the downstream flow regime, thereby benefiting downstream floral and faunal communities. Still other reservoirs warm up the river water, thereby elevating the downstream water temperature. Finally, the water flowing from reservoirs with bottom water outlets comes from the deeper layers and is consequently colder and, due to poor mixing and decomposition of organic matter, occasionally low in oxygen content. The changes in flow regime and water temperature detrimentally affect the downstream aquatic community. To quote Casado et al (1989), these are: 'a reduction of macrophytes, a reduction in faunistic richness both of fish and invertebrates, and a reduction of fish biomass, density and growth'.

During the last 10 to 15 years there has been growing public opinion against the construction of dams and reservoirs. For example, in Norway, the Lappish population protested against the damming of the river Alta, and, in Spain, a 101 m high dam built on the river Esla in 1970 to form the Riano reservoir remains unused because the inhabitants of the valley to be expropriated have refused to move (Garcia De Jalon, 1987). A more recent example is the construction of the Gabcikovo hydroelectric power plant on the Danube; although both Hungary and the former Czechoslovakia were jointly involved in the original project, environmental groups in Hungary forced the Hungarian Government to withdraw. The Hungarians are now calling attention to an environmental catastrophe in the area.

River channelisation

The objective when modifying the course of a river is to improve certain features, for example flood control, drainage of the surrounding land, navigation and erosion prevention. River channelisation comprises a number of physical measures, each of which is related to hydrological parameters; hence straightening changes the slope, dredging changes the depth and width, and dredging and weed cutting change the roughness. Other more radical methods of river channelisation are culverting, lining, and piping.

In many countries where there is intensive agricultural production, many of the rivers have been regulated. In Denmark, for example, 85 to 98 per cent of the total river network has been straightened (Brookes, 1987; Iversen et al, 1993).

Physical and biological effects

Channelisation has great impact on a river because it disrupts the existing physical equilibrium of the watercourse; to compensate for the alteration in one or more of the hydraulic parameters, and to establish a new, stable equilibrium, other parameters will change. Because straightening of a river increases its slope, the energy in the moving water has to be dispersed over a smaller surface; as a result the water is able to move larger particles and sediment discharge increases through bank erosion. If the river is not repeatedly manipulated or stabilised by culverting, lining etc, this will eventually lead to widening of the river channel and to a subsequent reduction in water velocity. River channelisation generally changes a heterogeneous system into a homogeneous one such that flow becomes uniform, pools are lost and the substrate becomes uniform throughout the channel (Figure 5.11).

Channelisation can also have great impact on riparian vegetation; trees are often logged to allow channel maintenance (eg, machine dredging) and scrubs are cut to ensure sufficient drainage. This increases solar radiation at the stream surface, thereby increasing the water temperature, reducing the concentration of dissolved oxygen, and increasing the in-stream primary production. In nutrient-rich watercourses this results in enhanced growth of benthic algae and macroalgae.

Another effect of the channelisation of rivers and drainage of wetlands may be increased nutrient and organic matter loading of rivers and the marine environment. The reason is that while the annual nitrogen removal capacity of wetlands and natural rivers can be as much as several hundred kg/ha, that of channelised rivers and drained wetlands may be negligible. Naturally meandering riparian zones alongside rivers may therefore play an important role in balancing intensive agricultural and ecological interests.

The velocity of river water is one of the major factors regulating the structure of riverine plant and animal communities (Brookes, 1988; Westlake, 1973). The uniform and often unstable sediment found in channelised watercourses is suitable for few, if any, plant species. Furthermore, as the uniform water flow precludes areas with little or no flow, resting sites for fish and invertebrates are virtually absent. The general effect of channelisation is therefore a reduction in habitat number and diversity and a consequent reduction in species number and diversity. The latter may be further reduced by the above-mentioned decrease in oxygen concentration. Hence, the biomass of organisms such as fish and invertebrates is usually lower in channelised watercourses, as illustrated in Figure 5.12.

It is not only animals and plants living within the watercourse that are affected by channelisation, however. Thus animal species which depend on the bank for foraging and/or breeding decline in number, with the consequence that species diversity on the river banks also decreases (Figure 5.13). In addition, a number of plant species that are confined to the more or less water saturated soil adjacent to the river are also affected.

Organic pollution of rivers

Organic matter derived from diverse human activities is a major source of pollutant discharge to rivers. The decomposition and breakdown of this organic matter is mediated by microorganisms and takes place mainly at the surface of the sediment and vegetation in smaller rivers, and in the water column in larger rivers. As the process requires the consumption of oxygen, severe organic pollution may lead to rapid deoxygenation of the river water and hence to the disappearance of fish and aquatic invertebrates. The habitat then becomes uniform with only a few robust species able to tolerate the low oxygen concentration. Decomposition of organic waste also results in the release of ammonium, which, although not in itself toxic, may, depending on the pH and temperature of the water, be converted to ammonia, which is poisonous to fish.

The most important sources of the organic waste feeding into rivers are domestic and industrial sewage. Immediately downstream of a sewage effluent, organic matter decomposition reduces the oxygen content of the water and results in the release of ammonium (Figure 5.14). Further downstream the concentration of organic matter decreases as a result of dilution and continuing decomposition. As the distance from the effluent increases, bacteria oxidise the ammonium to nitrate, and oxygen enters the water via the water surface, thereby increasing its oxygen content. Eventually the levels of organic matter, oxygen and ammonium reach those present immediately upstream of the sewage effluent; this process of recovery is called self-purification. An example is the Danube, which is already polluted by organic matter when it enters Hungary. As it winds its way through the country the river receives large amounts of organic matter from tributaries and cities, especially from Budapest. However, by the time it leaves the country and enters Croatia, an amount of organic matter equal to that discharged in Hungary has been decomposed (Varga et al, 1990; Benedek and Major, 1992). Nevertheless, this does not imply that rivers can take up an unlimited amount of organic matter without suffering as a result; the pollution may be so severe, widespread and long lasting that self-purification is insufficient. Thus the Danube is still polluted when it leaves Hungary, and the Rhine was polluted with such excessive amounts of organic matter between World War 2 and the early 1970s that there was very serious oxygen depletion in its middle and lower course and the river virtually died (Friedrich and Müller, 1984).

Organic matter content in European rivers

Because decomposition of organic matter requires oxygen, the amount of organic matter in a river can be measured in terms of the biochemical oxygen demand (BOD) or the chemical oxygen demand (COD), the units of which are mg O2/l. River reaches little affected by human activities generally have a BOD below 2 mg O2/l whereas a BOD exceeding 5 mg O2/l generally indicates pollution. Measurement of BOD is the most widespread method in Europe, but many countries also measure the COD, and some use only the COD. Although both BOD and COD indicate the potential oxygen demand of the organic matter in the water, there is not necessarily a correlation between the two measurements. In large rivers suffering from severe eutrophication, elevated BOD values can occur due to decomposition of phytoplankton, and in these cases high BOD values are not necessarily indicative of organic pollution.

BOD, COD and oxygen content data from a large number of river stations in 33 European countries are collated in Table 5.12. Median BOD, COD and oxygen are 2.8, 14.5, and 9.7 mg O2/l, respectively. As can be seen, annual mean BOD was below 5 mg O2/l and the oxygen content above 8 mg O2/l at more than 75 per cent of the river stations. Extremely high BOD is generally seen only in smaller rivers polluted with raw sewage or animal slurry, a BOD exceeding 500 mg O2/l then being possible.

The information obtained from the 30 European countries concerning the percentage of rivers with BOD or COD levels in specified classes is summarised in Map 5.13, and the BOD and COD concentrations of a large number of European rivers is presented in Map 5.14.

In Iceland, Norway, Sweden and Finland organic matter content is measured only as COD. In these countries discharge into rivers of organic waste derived from human activity is negligible and COD levels therefore are generally low (Map 5.13). In terms of BOD, rivers in Ireland, Georgia, Estonia, Latvia, Austria, Switzerland, The Netherlands, the UK, Denmark and Croatia are least affected, with less than 25 per cent of the rivers having a BOD exceeding 3.5 mg O2/l. In Hungary, Lithuania, Portugal, France, Ukraine, Germany, Slovenia and Italy the rivers are moderately affected, with less than 25 per cent having a BOD exceeding 5 mg O2/l. More affected rivers are found in Albania, Poland, the Czech Republic, Moldova, the Russian Federation and Spain, where more than 25 per cent of the rivers have a BOD exceeding 5 mg O2/l. BOD is highest in Bulgaria, Belgian Flanders and Romania, exceeding 5 mg O2/l in 60 per cent, 69 per cent and 80 per cent of the rivers, respectively.

Human activities and organic matter

The organic matter naturally occurring in rivers originates from soil erosion and from dead plants and animals and is normally relatively insoluble and only slowly decomposed. In contrast, organic matter derived from human activities is generally soluble, finely divided and rapidly decomposed, the result being a marked and abrupt increase in oxygen consumption in the river.

As population density in catchments increases, the level of organic matter in the rivers generally increases, and the oxygen content decreases (Figure 5.15). Thus whereas BOD concentration is lower than 2 mg O2/l in catchments with fewer than 15 inhabitants/km2, it generally exceeds 5 mg O2/l in catchments with more than 100 inhabitants/km2. The great variation found in extensively populated areas is attributable mainly to variation in the extent of wastewater treatment ­ well-functioning treatment plants can decompose up to 90 per cent of the organic matter in the wastewater.

Assessment of river quality

In most European countries the environmental state of the rivers has been monitored for many years. However, as pollution has both physico-chemical and biological effects on the receiving water, the quality of the water can be assessed in many different ways, and there are numerous methods in use throughout Europe. The definition of river quality used in this chapter is given in Box 5H.

The quality of European rivers is summarised in Table 5.13. About a quarter of the river reaches are classified as having poor or bad water quality. Most of the countries classify 50 per cent or more of their river reaches as having good or fair quality. Iceland, Scotland, Northern Ireland and Ireland have the highest proportion of rivers classified as having good water quality, while the Russian Federation, Finland, England and Wales, West Germany, Austria, Croatia, Lithuania and Latvia classify more than 75 per cent of their rivers as having good or fair water quality. More than 25 per cent of the rivers have poor or bad water quality in Bulgaria, Romania, the Czech Republic, Poland, Denmark, The Netherlands, Luxembourg, Slovenia and Italy. Furthermore, the percentage of river reaches classified as having bad water quality is highest in the Czech Republic and Poland (28 per cent) and in Belgian Flanders (37 per cent).

The greater the amount of organic matter present in river water, the lower the oxygen concentration, and the higher the ammonium concentration. These two relationships are illustrated in Figure 5.16 using data from 482 and 365 river stations, respectively. A low oxygen concentration influences the river fauna. To maintain a salmon or trout population in a river, a minimum oxygen concentration of 6 mg O2/l is necessary (Council Directive 78/659/EEC) (Figure 5.16a). To reduce the risk of fish kill, the ammonium concentration should not exceed 1 mg N/l (Figure 5.16b). Although ammonium is not in itself toxic to fish, it becomes toxic when converted to ammonia. Thus if the ammonia concentration exceeds 0.025 mg/l, trout growth is prevented, and if it exceeds 0.25 mg/l, the trout die. Rivers with large sewage discharges exceed these limits.

Trends in organic matter discharge to rivers

After World War 2 riverine discharge of organic waste increased in many European countries with resultant severe oxygen depletion. During the last 15 to 20 years, however, biological treatment of domestic and industrial wastewaters has intensified, and organic matter loading of rivers has consequently decreased in many parts of Europe, the result being that many rivers are now fairly well oxygenated.

An example is the river Rhine. Rebuilding of industry after World War 2 led to high loads of poorly or untreated sewage being discharged into the river. This caused oxygen contents of the water to fall considerably (Figure 5.17), and long stretches became biologically devastated. In the 1960s the deterioration in river quality became so apparent that countermeasures such as increasing the number of sewage treatment plants were implemented. In the eight-year period 1973­81, the percentage of sewage water that was treated before being discharged into the Rhine rose from about 30 per cent to 80 to 90 per cent (Dijkzeul, 1982 (in Wolff, 1987)). The consequent marked reduction in organic pollution led to a clear improvement in water oxygen content and the return of several of the animal species that had disappeared. Thus, although the general pollution of the Rhine is still serious, there are signs that conditions are improving.

A comparison of BODlevels at 223 river stations throughout Europe reveals signs of improving conditions. From the period 1977­82 to the period 1988­90 the organic matter concentration decreased at almost 75 per cent of the river stations (Figure 5.18), the reduction being greater than 25 per cent at 41 per cent of the river stations. An increase of more than 25 per cent was seen at 8 per cent of the river stations, however.

The improvement was greatest in Western Europe, where about half of the river stations show a decrease in organic matter concentration of more than 25 per cent, primarily as a result of intensified wastewater treatment. The improvement was less pronounced in Eastern Europe, where about 30 per cent of the river stations show a decrease of more than 25 per cent and about 15 per cent show an increase of more than 25 per cent (Figure 5.18).

Organic pollution is still a serious problem in many European rivers, and will continue to be so for as long as large amounts of sewage water are discharged into the rivers without being treated. Countries presently experiencing industrial depression should be especially attentive to this problem and develop their sewage plant capacity, as their industrial production can be expected to increase in the future.

Nutrients in European rivers and lakes

Precipitation that falls on the surface of the land dissolves and absorbs minerals as it drains through the soil. In pristine areas, the chemical composition of river and lake water is determined mainly by that of the soil and underlying bedrock, nutrient levels in such waters generally being low.

Human settlement and associated clearance of forest, agricultural development and urbanisation greatly accelerate the runoff of materials and nutrients into rivers and lakes. This stimulates the growth of phytoplankton and other aquatic plants and, in turn, that of organisms higher up the aquatic food chain. The process is usually known as 'cultural eutrophication'. Enhanced biological production and other associated effects of eutrophication are generally more apparent in lakes, reservoirs, coastal areas, and large, slowly flowing rivers, than in small rivers. Although phosphorus tends to be the nutrient that most limits plant growth in lakes and reservoirs, increased nitrogen levels can also lead to higher biological production (OECD, 1982), especially in marine areas, where it is usually the nutrient that most limits plant growth.

The ecological consequences of cultural eutrophication can be significant. High nutrient input to standing waters may provoke a shift in the biological structure; clearwater shallow lakes with submerged plants become dominated by phytoplankton, and the water therefore becomes so turbid that the submerged plants disappear. In addition, excessive growth of blue-green algae may sometimes occur, leading to the formation of surface scum and the production of toxins potentially poisonous to fish, cattle, dogs and humans. The fish community also changes and becomes dominated by species more tolerant to the turbid environment. When the phytoplankton sink to the lake bottom their decomposition may reduce the oxygen concentration in the water to levels too low to support fish and benthic invertebrates, the result being fish kill and the disappearance of benthic invertebrates. Low oxygen levels may also enhance the release of phosphorus from the lake sediment, thereby further enhancing phytoplankton production.

Excessive phytoplankton growth is easy to detect and it significantly affects both the use and the aesthetic quality of the lake. In cases where lake water is used for domestic water supply, eutrophication leads to taste and odour problems, and necessitates treatment and filtration of the water before use; this is an expensive and time-consuming process, especially in the case of very eutrophic water.

Nutrient concentrations in European rivers

Phosphorus in rivers

Phosphorus is measured both as total phosphorus and soluble reactive phosphate (dissolved orthophosphate). In 321 European rivers dissolved orthophosphate was found to average 59 per cent of total phosphorus. In some countries total phosphorus is not measured and data are available only for dissolved orthophosphate. In such cases the data have not been converted to total phosphorus, but are presented as dissolved phosphate and the fact noted in the figures and tables.

Information concerning annual mean phosphorus concentrations at 546 river stations in 34 countries is collated in Table 5.14. Median annual mean total phosphorus and dissolved phosphate were found to be 173 and 126 µg P/l, respectively, annual mean phosphorus levels being below 50 µg P/l at only 25 per cent of the stations. In catchments with little or no human activity phosphorus levels in rivers are generally lower than 25 µg P/l. Phosphorus levels exceeding 50 µg P/l indicate an anthropogenic influence, for example sewage effluent and agricultural runoff. In rivers heavily polluted by sewage effluent, phosphorus levels may exceed 500 to 1000 µg P/l.

The frequency distribution of phosphorus levels in rivers has been compiled for 34 European countries, each country having supplied information about the percentage of rivers with annual mean phosphorus levels in specified classes: below 25 µg P/l, 25 to 50 µg P/l, 50 to 125 µg P/l, etc (Map 5.15). This provides an overview of the percentage of rivers with low and high phosphorus levels. The phosphorus level at a large number of specific European river stations is presented in Map 5.16 as well.

The lowest phosphorus levels are found in the rivers in Norway and Iceland. Although many rivers in Sweden, Finland, Ireland, Austria, France, the Russian Federation, Slovenia, Albania, Georgia and Bulgaria also have a low annual mean phosphorus concentration (Map 5.15), the concentration exceeds 50 µg P/l in 10 to 40 per cent of the rivers in Sweden, Finland and Ireland, and exceeds 125 µg P/l in 10 to 50 per cent of the rivers in France, the Russian Federation, Slovenia and Bulgaria.

In many other countries only 10 to 20 per cent of the rivers have a phosphorus concentration below 50 µg P/l. These relatively unpolluted rivers are generally situated in catchments in mountainous and forested regions where the population density is low. In Estonia, Latvia, Lithuania, Switzerland, Austria, Croatia, Italy and Portugal, more than 40 per cent of the rivers have a phosphorus level below 125 µg P/l; however, many of the rivers in these countries have a high phosphorus level.

The highest phosphorus levels are found in a band stretching from southern England, across the central part of Europe through Romania and Moldova to Ukraine; in these countries more than 80 per cent of the rivers usually have a phosphorus concentration exceeding 125 µg P/l. In Poland, Belgium (Flanders), Luxembourg and the UK, about 50 per cent or more of the rivers even have a phosphorus level exceeding 500 µg P/l (Maps 5.15 and 5.16).

With the exception of large rivers in the Russian Federation and the Nordic countries, the phosphorus levels generally exceed 100 µg P/l in the downstream reaches in all of the largest rivers in Europe (Map 5.16).

Bearing in mind the regional differences described above, Europe can be divided into four main regions, the river phosphorus levels of which are shown in Figure 5.19. The phosphorus levels are lowest in the Nordic countries, medium in both the Southern and Eastern European countries (although slightly lower in Southern European rivers), and highest in the Western European countries.

Nitrogen in rivers

Dissolved inorganic nitrogen, particularly nitrate and ammonium, constitutes most of the total nitrogen in river water; thus inorganic nitrogen was found to constitute 88 per cent (nitrate 78 per cent and ammonium 4 per cent) of the total nitrogen in 240 to 420 European rivers. In some countries data are available only for nitrate and occasionally ammonium. The results presented in this chapter are therefore either total nitrogen, inorganic nitrogen, or only nitrate nitrogen, as indicated in the figures and tables.

Annual mean nitrogen concentration at 654 river stations in 34 countries is summarised in Table 5.15. Whereas ammonium levels are generally below 0.5 mg N/l, nitrate and total nitrogen concentrations exceed 1 mg N/l in most of the rivers. The average level of ammonium and nitrate in pristine rivers is reported to be 0.015 mg N/l, and 0.1 mg N/l, respectively (Meybeck, 1982). Although rivers in a strictly pristine state are rarely found in Europe because of high atmospheric nitrogen deposition, the levels of nitrogen in relatively unpolluted streams ranged from 0.1 to 0.5 mg N/l (Table 5.15). Nitrogen levels exceeding 1 mg N/l indicate an anthropogenic influence, for example agricultural runoff and sewage effluent. The ammonium level normally rises and the oxygen level falls when rivers receive sewage effluent or effluent from animal husbandry farms; in heavily polluted rivers the ammonium level may rise to as much as 1 to 5 mg N/l, which, when converted to ammonia, may be toxic to fish and other river fauna.

The nitrogen concentration is generally lowest in rivers in Iceland, Norway, Sweden, Finland and Albania, being below 0.75 mg N/l in 60 to 100 per cent of the rivers (Map 5.17). In southern Sweden and Finland, and in Latvia, Lithuania, the Russian Federation, Ireland, Austria, Switzerland, Slovenia, France, Portugal and Spain, nitrogen concentrations are higher, ranging from 1 to 3 mg N/l in the majority of the rivers (Maps 5.17 and 5.18).

The highest nitrogen levels are found in Eastern and Western European rivers, particularly in the Czech Republic, Denmark, Estonia, Germany, Luxembourg, Moldova, The Netherlands and Romania, the concentration exceeding 2.5 mg N/l in more than two thirds of the rivers (Map 5.17).

Whereas the median nitrate concentration is only 0.18 mg N/l in Nordic rivers, it is 3.5 mg N/l in the Western European countries, where rivers with nitrogen levels below 1 mg N/l are rare (Figure 5.20). The nitrate levels in the rivers of Southern and Eastern European countries are basically the same, although nitrate levels in the Southern European countries are rather homogeneous whereas there are large regional differences in Eastern Europe; thus the concentration is high in Central European countries like Romania, the Czech Republic and Poland, but low in the new Baltic states and large parts of the Russian Federation.

In Chapter 14 the relationship between human activities and nutrient concentrations in rivers is described.

Nutrient concentrations in European lakes

The nutrient concentration of the inflowing water generally determines the lake water nutrient concentration (Vollenweider, 1976; OECD, 1982). However, nutrients may be partly lost from the lake water, either to the sediment, especially in the case of phosphorus, or to the atmosphere through denitrification, in the case of nitrogen. The proportion of nutrients lost is directly related to water residence time. Nutrient levels in lakes are therefore generally lower than in their tributaries. If external phosphorus loading of a lake is high for a period, a large amount of phosphorus will accumulate in the lake sediment; when external loading is subsequently reduced, some of the accumulated phosphorus may be released to the water and thereby delay lake recovery (Sas, 1989; Cullen and Forsberg, 1988; Marsden, 1989).

In relatively unpolluted areas such as remote mountain regions and national parks, lake nutrient levels are usually low, the total phosphorus concentration being below 10 to 20 µg P/l and the nitrogen level below 0.5 mg N/l (Table 5.16). Higher concentrations generally imply an anthropogenic influence on the lake catchment.

The frequency distribution of lake phosphorus and nitrogen concentrations gives an indication of lake water quality in each country. However, although the presence of a high percentage of lakes with low nutrient levels suggests that the water quality is generally good, the frequency distribution may be dominated by clearwater lakes in sparsely populated areas and as a consequence may not be fully representative of the relatively more important lakes in the most inhabited areas.

Phosphorus in lakes

The European lakes with the lowest phosphorus concentration are found in Norway, the concentration of total phosphorus being less than 10 µg P/l in more than 70 per cent of lakes (Map 5.19) and only exceeding 25 µg P/l in approximately 10 per cent of lakes. In nearby Sweden and Finland, as well as in Latvia and Austria, lake phosphorus levels are also relatively low, being below 25 µg P/l in 75 to 90 per cent of lakes. In Switzerland, Italy and Germany, the lake phosphorus levels are higher, approximately 50 per cent of the lakes having a concentration below 25 µg P/l, and the other 50 per cent having a concentration between 25 and 125 µg P/l. German lakes with low phosphorus concentrations tend to be located in the Alpenvorland, while those with high concentrations are located in the northern part of the country. In Italy there are also many lakes with low phosphorus levels in the Alpine regions. In Estonia and Northern Ireland, phosphorus levels in the majority of the lakes are below 50 and 125 µg P/l, respectively.

In Spain, England and Wales, Romania, Denmark, Poland, The Netherlands and Moldova, lake phosphorus levels are generally high, exceeding 50 and 125 µg P/l in more than 80 per cent and 45 per cent of lakes, respectively.

Nitrogen in lakes

The frequency distribution of total nitrogen in lakes shows the same tendency as for phosphorus. The percentage of lakes with low nitrogen levels is highest in Norway, where 50 per cent and 90 per cent of the lakes have nitrogen levels below 0.3 and 0.75 mg N/l, respectively. In Sweden, Finland, Latvia and Estonia, nitrogen levels are lower than 0.75 mg N/l in the majority of the lakes. The picture is the opposite in Denmark, Poland and Moldova, where 85 to 90 per cent of the lakes have nitrogen levels higher than 0.75 mg N/l.

Phosphorus concentrations in large lakes

Extreme nutrient levels are rarely seen in the large European lakes, total phosphorus and total nitrogen generally being below 125 µg P/l and 1.0 mg N/l, respectively.

In the two largest European lakes, Lake Ladoga and Lake Onega in the Russian Federation, total phosphorus concentrations are currently 26 and 10 µg P/l, respectively (Map 5.20), levels three to five times higher than in the 1950s (Gutelmacher and Petrova, 1982; Petrova 1987). Although information about nutrient levels in many of the other large European lakes and reservoirs that lie in the Russian Federation is limited, the phosphorus concentrations in Lake Peipus and Lake Il'men' are 40 and 70 µg P/l, respectively. In 13 large reservoirs on the Volga and the Dnepr, the total phosphorus concentration ranges from 49 to 201 µg P/l, the mean value being 87 µg P/l (Datsenko, 1990).

In Sweden, Finland and Norway the large lakes usually have a phosphorus concentration below 15 µg P/l. Nevertheless, a few lakes located in more densely populated areas have higher phosphorus levels, examples being the Swedish lakes Mälaren and Hjälmaren, the concentrations of which are around 30 and 50 µg P/l, respectively. The largest lake in Ireland and the UK, Lough Neagh in Northern Ireland, has a phosphorus level of approximately 100 µg P/l, while other large lakes in Ireland and the large lochs in Scotland generally have low phosphorus concentrations. The majority of large lakes in The Netherlands, northern Germany, Denmark and Poland have phosphorus levels around or exceeding 100 µg P/l.

Large lakes in the Alps have phosphorus concentrations ranging from 10 to 60 µg P/l, with concentrations in the largest approximately as follows: 60 µg P/l in Lac Léman, 50 µg P/l in Bodensee, 10 µg P/l in Lago di Garda, 30 µg P/l in Lake Neuchâtel, and 18 µg P/l in Lago Maggiore. The two large lakes at the Hungarian Plain, Lake Balaton and Neusiedler See, have phosphorus levels of 40 to 100 µg P/l, while the lakes in the mountains between Albania, Greece and the former Yugoslavia generally have low phosphorus levels.

Impact of nutrients on lake water quality

Phosphorus is usually the nutrient that most limits plant growth in lakes and reservoirs. Many studies have shown that the amount of phytoplankton (generally measured as the level of the photosynthesis pigment, chlorophyll) increases in lakes with increasing phosphorus levels (eg, OECD, 1982; Canfield and Bachman, 1981; Berge, 1987; Jeppesen et al, 1991). The increase in phytoplankton concentration reduces water transparency and at high nutrient levels the lake water becomes very turbid.

In lakes with summer mean total phosphorus levels lower than 10 µg P/l, chlorophyll concentrations are low and the water is clear, that is, water transparency is high (Figure 5.21). With increasing phosphorus concentrations chlorophyll levels rise and water transparency declines. Thus in lakes with phosphorus levels between 50 and 125 µg P/l, the chlorophyll level ranges from 10 to 50 µg/l, and water transparency is reduced to 1 to 2 m. At phosphorus concentrations exceeding 125 µg P/l the chlorophyll concentration may reach very high levels, and water transparency becomes reduced to less than 1 m. However, in lakes with extremely high phosphorus concentrations ­ above 500 µg P/l ­ the increase in chlorophyll levels out, the biological structure becomes very unstable, and fish kill may occur (Søndergaard et al, 1990).

Blue-green algae and phosphorus

In lakes with high phosphorus levels excessive growth of blue-green algae may lead to surface scum and the production of toxins potentially poisonous to fish, cattle, dogs and humans. Data from 300 shallow Danish lakes indicate that blue-green algae are of only minor importance in the summer at phosphorus levels below 50 µg P/l, but become increasingly important at higher phosphorus levels (Figure 5.22), and become the dominant phytoplankton group at phosphorus concentrations from 200 to 1000 µg P/l. In deeper lakes, however, blue-green algae tend to dominate at even lower phosphorus levels, and the concentration generally has to be below 20 µg P/l for blue-green algae not to become dominant in the summer (Sas, 1989).

Trends in nutrient concentrations

Time series data for the period 1977­90 concerning nutrient concentrations in more than 250 European rivers have been collated for this report. Most of the time series cover the whole period but, especially in the case of Southern European rivers, some of the time series cover only the last five to eight years.

Phosphorus trends

The increase in phosphorus levels that occurred in many lakes during the 1960s and 1970s caused severe eutrophication problems and led several countries to take measures to reduce phosphorus discharge to rivers and lakes. This was achieved primarily by reducing discharge from point sources, especially wastewater, by improving wastewater treatment. In some countries part of the reduction was achieved by reducing or banning the phosphorus content of detergents.

The reduction in phosphorus loading is reflected by a marked decrease in the phosphorus levels of many European rivers (Figure 5.23); the phosphorus concentration in the majority of rivers (64 per cent) having decreased between the periods 1977­82 and 1988­90. In one third of the rivers, the reduction was greater than 25 per cent.

While the phosphorus concentration decreased in the majority of rivers in the Nordic countries and in Western and Southern Europe, it rose in many Eastern European rivers, largely because of very limited construction in the Eastern European countries of sewage plants incorporating phosphorus removal. The reduction in phosphorus levels was most marked in Western Europe, where a reduction of more than 25 per cent was observed in half of the rivers. Nevertheless, the highest phosphorus levels are still to be found in the rivers of Western Europe (Figure 5.19).

A reduction in phosphorus loading and concentration has also been observed in several European lakes during the last 10 to 20 years (Table 5.17). During the 1960s and early 1970s an increase in phosphorus concentration caused environmental deterioration in many of the large lakes in the Alps (Figure 5.24). Action plans were consequently implemented in the 1970s with the objective of reducing phosphorus loading. As a result there was a marked reduction in phosphorus levels during the 1980s. In the two largest alpine lakes, Lac Léman and Bodensee, the phosphorus level has been reduced from about 90 µg P/l to 60 and 50 µg P/l, respectively (Stabel, 1991; Lang, 1991). However, these concentrations are still high, and little improvement has been observed in the water quality (Tilzer et al, 1991; Lang, 1991).

In Denmark there has been a general reduction in the phosphorus level of many lakes (Figure 5.25a), but without any significant improvement in water transparency (Figure 5.25b); this discrepancy may be explained by the fact that the phosphorus level still exceeds 125 µg P/l in 70 per cent of the lakes, and so water transparency is therefore generally less than 1 m (Figure 5.22).

Nitrogen trends

Nitrate levels have tended to increase in many European rivers over the last 20 to 40 years (Figure 5.26), and the same applies to European lakes (Barbieri and Mosello, 1992; Tilzer et al, 1991; Mosello, 1989; Henriksen et al, 1988). The increase is attributable mainly to a corresponding increase in the use of nitrogen fertilisers in most European countries (see Chapter 22). However, the increase in lake nitrate levels in areas with relatively low human activity, for example Norway, is attributable mainly to increased atmospheric deposition of nitrogen caused by the combustion of fossil fuels and agricultural activity (Henriksen et al, 1988). A decrease in the ammonium level in European rivers is expected as a consequence of general improvement in wastewater treatment and a growing awareness of the problem of agricultural pollution with manure and silage.

Nitrate levels increased between 1977­82 and 1988­90 in more than two thirds of European rivers, the median concentration increase being 0.14 mg N/l, or 13 per cent (Figure 5.27). The percentage of rivers in which the concentration increased was nearly identical in the four regions of Europe, but the median concentration increase varied, being 0.013 mg N/l in Nordic rivers and 0.18, 0.49, and 0.35 mg N/l, in the rivers of Western, Eastern and Southern European countries, respectively. The increase in nitrate concentration was most marked in Eastern and Southern European rivers, because of the marked increase in the use of nitrogen fertilisers during the same period. In many Western European countries, in contrast, the increase in the use of nitrogen fertilisers peaked in the late 1970s and remained at that high level during the 1980s (see Chapter 22).

The ammonium level in 70 per cent of 230 European rivers decreased markedly between 1977­82 and 1988­90 (Figure 5.28). The decline was most significant in rivers in which the ammonium level was between 0.5 and 1 mg N/l in the period 1977­82, the ammonium level falling to below 0.5 mg N/l in most of these rivers. The decreasing trend in ammonium levels is in concert with the decline in organic matter content of the water (Figure 5.17), and can be explained by improved wastewater treatment and the consequent reduction in organic matter discharge. The reduction in ammonium levels was most significant in the Western European and Nordic countries, while the concentration tended to increase in many Eastern European rivers.

Acidification of surface waters

Surface water acidification has been of public concern since the early 1970s, when awareness of the problem was aroused by episodes of severe fish kill in rivers and lakes in the southernmost part of Norway, and along the west coast of Sweden (Overrein et al, 1980; Monitor 12, 1991). These episodes of fish kill were attributed to 'acid rain', that is, precipitation acidified by the combustion products of fossil fuels. Norwegian and Swedish observations have indicated that the emission of acidifying gaseous by-products such as sulphur dioxide and nitrogen oxides was probably the main cause of acidic precipitation in areas situated up to hundreds of kilometres from the source.

Studies were initiated to evaluate the extent, the cause/effect relationships, and the chemical and biological impact of surface water acidification in Norway and Sweden (Overrein et al, 1980; National Swedish Environment Protection Board, 1983). It was found that, of a total of 5000 lakes in a 28 000 km2 region of southern Norway, 35 per cent had lost their fish populations. In southern and central Sweden, damage to fish communities was observed in 2500 lakes; however, on the basis of lake pH, and from the knowledge that many fish species are unable to tolerate pH levels below 5.5, it was estimated that the fish populations of about 18 000 lakes were affected.

Surface water acidification and damage to freshwater life forms have since been documented in numerous studies (eg Muniz, 1991). Over the last two decades, surface water acidification as a consequence of the atmospheric emission of sulphur dioxide and nitrogen oxides has been recognised as a serious environmental problem in most other European countries, as well as in North America (Merilehto et al, 1988; Howells, 1990; Skjelkvale and Wright, 1990). In the mid-1980s a further acidifying air pollutant was identified. Laboratory studies in The Netherlands documented that the deposition of ammonia volatilised from agricultural areas with large concentrations of livestock could be a major cause of acidification in softwater lakes (Schuurkes et al, 1986; van Dam, 1987; van Breemen and van Dijk, 1988).

The acidifying effect of atmospheric deposition has been documented from studies in areas such as nature reserves and remote mountain lakes (Mosello et al, 1992b), where the only possible influence of human activity is through atmospheric deposition. When considering areas with industrial and agricultural activity, several factors that can positively or negatively affect surface water acidity have to be taken into account: for example mining, afforestation, clear cutting, draining, crop removal, fertilisation and liming. Internal processes in lakes (eg, reduction and oxidation processes) may also influence their acid/base status. These various factors are described in standard textbooks, for example Cresser and Edwards (1987); Howells (1990); Monitor 12 (1991).

Acidification of surface waters in Europe

Surface water acidification can be expected in areas where acidic deposition is high and the catchment soil or bedrock is poor in lime and other easily weatherable minerals that buffer against acid precipitation. Small high altitude lakes and streams are generally affected more severely than larger lowland surface waters, the latter usually being acidified only in well leached, sandy areas covered by heath or forest. Dune lakes and pools may also be acidified. Even naturally acidic bog lakes (eg, in the Campine region of Belgium) may be further acidified by acid precipitation, thereby causing a change in their ecology.

Geological characteristics render large areas of Europe sensitive to surface water acidification if exposed to acidic atmospheric deposition (Map 5.21). The areas of Europe where surface water acidification has actually been observed are shown in Map 5.22 and the current state of surface water acidification in Europe is briefly summarised below. A more comprehensive summary can be found in Merilehto et al (1988), Howells (1990), and Skjelkvale and Wright (1990).

In southern Finland, Sweden and Norway, lake and river acidification increased slowly from 1915 to 1950 and rapidly from 1950 to 1980. Since then conditions have remained relatively unchanged, possibly as a result of a reduction in acidic deposition (Brodin and Kuylenstierna, 1992). In southern Norway lake sulphate concentration tended to decrease between 1974 and 1986 in line with the decrease in sulphur emission in Western Europe. Surprisingly, however, this was not accompanied by a decrease in lake acidity (Norwegian State Pollution Control Authority, 1987). The reason could be the two-fold increase in nitrate concentration that occurred during the same period as a result of an increase in the emission of nitrogen oxides. According to a Swedish survey undertaken in the winter of 1990, 45 per cent of the 85 000 lakes in Sweden had a winter pH below 6.0, and 7 per cent had a pH below 5.0 (Monitor 12, 1991). Most of the acidic lakes were small, however: 23 per cent of lakes smaller than 0.1 km2, but only 2 per cent of lakes larger than 1 km2, had a pH below 5.5 (Monitor 12, 1991). In Finland about 10 per cent of 1000 lakes were found to be seriously affected (Merilehto et al, 1988). Lake acidification has also been found in the Russian Federation in Karelia and in the Kola Peninsula. As surface waters in Karelia have a high humus content, mean pH is 6 to 7 (Merilehto et al, 1988).

Most surface waters in Western and Central Europe are not as seriously affected by acidification as those in Finland, Sweden and Norway, despite the fact that acidic deposition is greater; this is because the soil of Western and Central Europe is generally well buffered, and therefore able to neutralise the acid deposited. In the UK acidified surface waters are found in various regions, but primarily in areas dominated by acidic soils, especially Scotland, northern England and Wales (Howells, 1990). In parts of Scotland (Battarbee, 1989) and Wales (Jenkins et al, 1990) afforestation of areas heavily loaded by sulphur oxides has exacerbated river acidification, while in more pristine areas, such as the northernmost part of Scotland, acidification has not occurred despite an increase in seasonal input through canopy interception. In Ireland there is little evidence of surface water acidification attributable to acidic precipitation. However, rivers draining evergreen afforested catchments of soil poor in buffering capacity were found to be more acidic when the catchment was covered with evergreen forest than when not, the increased acidity being attributable to the scavenging of acidifying air pollutants by the canopies of the evergreen trees, and therefore affected by acid rain (Bowman, 1991). In The Netherlands 50 to 60 per cent of about 5000 lakes are acidic (Leuven et al, 1986). Furthermore, of 187 randomly selected softwaters, mainly small isolated moorland and dune pools and lakes, 35 per cent were extremely acid with pH below 4.0. Although the main cause of this acidification is believed to be deposition of acidifying air pollutants, oxidation of sulphides is a contributing factor under certain geological conditions.

Surface water acidification has also been recorded in Belgium, Denmark, Germany, Poland, the Czech Republic, the Slovak Republic, Austria, France, Switzerland and Italy (Merilehto et al, 1988). Acidification of high altitude lakes has been reported at a workshop in 1991 on remote mountain lakes (Mosello et al, 1992b). Studies of such lakes are important because many of them are acid-sensitive and, since they are situated far from human settlements, atmospheric deposition is the main source of pollution (Wathne, 1992). Acidification has been recorded in mountain lakes in Italy (Mosello et al, 1992a; Schmidt and Psenner, 1992), the Slovak Republic (Fott et al, 1992) and the Czech Republic (Vesely and Majer, 1992). In contrast, no acidified lakes have been recorded in the Pyrenees (Catalan and Camarero, 1992) and in the northern Apennines (Viaroli et al, 1992). Acidification of lakes and rivers in forested areas has been recorded in the Vosges Mountains, the Schwarzwald, the Bavarian Forest, Thüringerwald and Erzgebirge. Acidification of small, isolated seepage lakes has been recorded in lowland areas of The Netherlands, Belgium, Denmark and northern Germany, mainly in regions dominated by sandy soils of low buffering capacity (Merilehto et al, 1988).

Impact of acidification

Acidification affects aquatic ecosystems at all levels and has a profound impact on both plant and animal communities. Aquatic organisms are influenced both directly, because of the resulting toxic conditions, and indirectly, because of the loss of suitable, acid-sensitive prey. One of the first recorded effects of acidification caused by human activity was a decrease in fish diversity in the Nordic countries (Overrein et al, 1980; Drabløs and Tollan, 1980).

Aquatic organisms have extremely different tolerance levels towards acidification, as illustrated for various species/groups of benthic invertebrates and fish in Figure 5.29. The toxicity of acid freshwaters towards fish is only partly attributable to the low pH. In most cases, the crucial toxic factor is the content of inorganic aluminium ions that have been leached from minerals in the catchment by acid deposition and subsequently transported to the lakes and streams (Drabløs and Tollan, 1980; Muniz, 1983).

That acidification can severely affect aquatic ecosystems is especially well documented in Sweden and Norway. For example, during 1940­75, 1750 of 5000 lakes in southern Norway became completely devoid of fish as a result of acidification, with another 900 lakes being seriously affected (Overrein et al, 1980). Recent Norwegian results (T. Hesthagen, NINA, personal communication) indicate serious and ongoing damage. Of 13 600 surveyed fish stocks in lakes in southern Norway, 2600 had become extinct and 3000 had reduced their number of individuals. In rivers in southern Norway, Atlantic salmon are badly affected. In 25 rivers, the salmon is now virtually extinct due to acidification (Hesthagen and Hansen, 1991). The total annual number of adult salmon lost was estimated to be between 92 000 and 305 000 individuals weighing some 345 to 1150 tonnes. Fish kill on such a scale is detrimental to both commercial fishery and recreation and quality of life.

Acidification trends

There is good evidence that many lakes have become acidified during the past 100 years; this has been documented both chemically and by palaeolimnological studies of pH-indicator species such as diatoms in lake sediment cores (Charles et al, 1990; Howells, 1990). Such studies indicate, for example, that whereas the pH of lake Gårdsjön in Sweden fell slowly from pH 7 to 6 over the 12 500 years from the end of the last ice age until 1960 by natural acidification processes, it has subsequently fallen rapidly from pH 6 to 4.5 (Figure 5.30). The pH values determined in this way agree reasonably well with lake water chemistry, the measured pH of lake Gårdsjön having been 6.3 in 1948, and 4.7 in 1977 (Wright, 1977). A trend towards acidification has also been recorded in freshwaters in Finland, Sweden, Norway, the UK, The Netherlands, Belgium, Denmark and Germany (Charles et al, 1990; Monitor 12, 1991).

Critical loads

As a result of international cooperation on reduction of emissions of sulphur and nitrogen oxides (see Chapter 31) the concept of 'critical load' emerged in order to establish reduction strategies for emissions of these substances, which, in contrast to the preceding strategy, should be based on scientific knowledge of the capacity of different ecosystems and sites to withstand various types of acid deposition.

Since the early 1980s both 'target load' (ie, a political goal for loading reduction) and 'critical load' have been extensively explored. Since 1988 the most widely used definition of the critical load for surface waters has been: 'A quantitative estimate of the loading of one or more pollutants below which significant harmful effects on specified sensitive elements of the environment are not likely to occur according to present knowledge' (Kämäri et al, 1992).

Operational criteria for assessing the effects of acid deposition on surface waters (rivers and lakes) have, for example, been developed in Scandinavia. These criteria are based upon selected organisms and populations (fish and invertebrates) which are sensitive to chemical changes in the water as a result of changes in atmospheric inputs of acidifying substances (Figure 5.31).

The emission reduction needed to reach critical loads for nitrogen and sulphur is estimated to be as high as 70 to 80 per cent for southern Sweden and southern Norway, and even higher for other European countries (Brodin and Kuylenstierna, 1992).

The critical load concept appears to be a rational method for quantifying how much acidic deposition and hence emission must be reduced in order to protect sensitive sites or ecosystems. Furthermore, the concept has now found acceptance in political negotiations concerning reduction strategies. European critical load maps have been prepared by the Coordination Centre for Effects (Hettelingh et al, 1991), and Kämäri et al (1992) have described in more detail how the critical load concept should be employed.

Heavy metals, organic micropollutants, radioactivity and other hazards

Human activity leads to chemical compounds being discharged into the aquatic environment in various ways, for example in domestic sewage water and industrial effluent, through mining and through atmospheric deposition. While some of these compounds are known to be toxic, the environmental impact of many others remains to be elucidated. Nevertheless, these contaminants together represent a threat both to the aquatic ecosystem per se, and to human health ­ a threat that is intensified by the vast number of new synthetic organic compounds being produced and released into the environment.

Heavy metals

Production and use of heavy metals has increased markedly in Europe during the 19th and 20th centuries. As a result heavy metal contamination of inland surface waters has also increased, the main sources being mining and industrial activities. Other important sources are sewage discharge, runoff from the land and atmospheric deposition. In the 1970s surface water concentrations of heavy metal levels reached alarming proportions and national and international regulations (eg, Council Directive 76/464/EEC and daughter directives on pollution caused by certain dangerous substances discharged into the aquatic environment of the Community) were therefore implemented to control heavy metal release at the source. Although this has led to reduced levels of harmful metals in many Western European rivers during the last decade (RIVM, 1992b; UK DoE, 1992; Umweltbundesamt, 1992), the level is still high in some European rivers.

General assessment of the state of heavy metal pollution of European rivers and lakes is difficult, primarily because measurement of metals is rarely included in monitoring programmes, but also because concentration levels are usually so low that problems arise with sample preparation and methodological precision. Until recently metals have generally been analysed either as total water concentration or on the dissolved fraction. Recent evidence, however, suggests that anthropogenic inputs generally can be better evaluated from particulate associated metals (Meybeck et al, 1989). Comparison and assessment of the state of heavy metals in European rivers is therefore even more difficult than for most of the other water quality variables. The following evaluation of heavy metals in European rivers must be looked at with these reservations in mind. It has been possible in this report to collate data on total heavy metal concentrations in more than 200 European rivers (Table 5.18). In general, the concentrations are well below standards for drinking water; only cadmium and mercury exceed drinking water standards in some rivers. It should be remembered, however, that the aquatic biota are usually affected by much lower concentrations: thus while the drinking water standard for copper is 100 to 3000 µg Cu/l, salmonoid fish are affected at 10 to 50 µg Cu/l (Hodson et al, 1979). Statistical descriptive variables ­ the median and 10 and 90 percentiles ­ of heavy metals in European rivers were found to be nearly identical to those for a river dataset covering the whole world (Meybeck et al, 1989). Heavy metal levels were generally lowest in the rivers of the sparsely populated Nordic countries, while those in the rivers of Western, Eastern, and Southern European countries were roughly the same.

High concentrations of heavy metals were recorded at some of the river stations, in particular in the upstream reaches of the Wisla, Oder and Elbe (where the majority of the Polish and Czech heavy industry is located), the Danube and many of its tributaries, and some of the rivers of Ukraine. These heavy metal 'hot spots', which are generally located near mining areas and industries using large quantities of metals, are only examples: a more extensive survey would undoubtedly reveal many more locations with high levels of many metals.

The increase in acid deposition described in the previous section has led to increased mobilisation of metals from the soil (see Chapter 7), and hence to elevated levels in the aquatic environment; thus increased levels of aluminium in surface waters and elevated levels of mercury in fish have been found in areas of Europe known to be acidified (Henriksen et al, 1989; Andersson et al, 1987).

The problems associated with heavy metal contamination of inland surface waters are illustrated below using cadmium, copper and zinc as examples.

Cadmium

Cadmium ranks among the most hazardous metal pollutants, cadmium-induced Itai-Itai disease in the Japanese Jinzu River basin being a frightening reminder of its toxicity (Nogawa, 1981). The cadmium concentration in natural waters unaffected by humans is generally less than 1 µg Cd/l (Meybeck et al, 1989). Although cadmium levels in European rivers rarely exceed 1 µg Cd/l, there are a few rivers with high cadmium concentrations, especially the Iskar in Bulgaria, the rivers Guadiana and Tajo in Portugal, and the Danube in Romania. The cadmium that is discharged into inland surface waters accumulates in the sediment; as a result, the sediment of several European rivers and lakes is contaminated (Forstner, 1980).

A study of the cadmium cycle in the Rhine basin showed that the most important sources of cadmium pollution during the period 1983­87 were industrial point sources and runoff from the land (RIVM/GLOBE, 1992). Emission reduction was implemented in the 1980s, especially at point sources, and as a result cadmium levels are presently declining; however, the levels are still high, and additional measures may be needed to reduce cadmium concentrations further.

Copper

Because of its widespread use, copper is both an actual and a potential pollutant of the aquatic environment. In catchments with no human activity the river copper concentration is generally lower than 2 to 5 µg Cu/l (Nriagu, 1979; Hodson et al, 1979), a level well below the drinking water standard of 100 to 3000 µg Cu/l. Such a high concentration is rarely found, and then primarily in connection with mining. It is well known that copper is highly toxic to aquatic biota ­ copper compounds are often used as insecticides, fungicides, algicides and molluscides. Its toxicity is generally higher in water with a low mineral content than in water with high alkalinity and hardness. Although the Council Directive (78/659/EEC) on the quality of freshwaters needing protection or improvement in order to support fish life recommends that the copper level be below 40 µg Cu/l in water with a hardness of 100 mg CaCO3/l, fish and invertebrates have been reported to be affected by copper levels as low as 10 to 20 µg Cu/l (Hodson et al, 1979). Since about 20 per cent of the river stations analysed in this report have copper levels exceeding 10 µg Cu/l, the aquatic biota may be affected. However, the copper levels rarely exceed 40 µg Cu/l.

Zinc

Like copper, zinc is also widely used. It is produced in 12 European countries, with what is now the Russian Federation being the single most important producer (Cammarota, 1980). There are also, for instance, at least 140 zinc mining sites in Belgium, England, Wales, France, Ireland and northern Spain (Whitton, 1980). In addition, mining of many other metals also results in zinc being released into the environment. It can thus be assumed that several European rivers are being polluted by zinc.

In natural water unaffected by human activity the zinc concentration is generally below 5 µg Zn/l. Although human tolerance to zinc is generally high, zinc may be rather poisonous to the aquatic biota. The Council Directive (78/659/EEC) concerning water standards for fish recommends that zinc levels be below 300 µg Zn/l in water with a hardness of 100 mg CaCO3/l but below 30 µg Zn/l in water with a hardness of 10 mg CaCO3/l. The annual mean zinc concentration exceeded 35 µg Zn/l at 25 per cent of the river stations analysed in this report.

Organic micropollutants

Some organic micropollutants, for example the pesticide DDT and polychlorinated biphenyls (PCBs), are well known, as are their toxic effects. Others are suspected of having effects of various types, including carcinogenic effects, while the environmental effects of others remain to be elucidated (see Chapter 17).

Traditional monitoring for organic micropollutants in the aquatic environment has focused on organochlorine compounds (eg, DDT, PCBs), and polycyclic aromatic hydrocarbons (PAHs). However, organic micropollutants have rarely been measured on a large-scale basis in national monitoring programmes. In recent years there has been a growing awareness of the problem of aquatic pollution by other dangerous compounds, such as non-organochlorine pesticides. A number of surveys aimed at defining the magnitude of the problem have therefore been undertaken. Nevertheless, it is extremely difficult to quantify the risk presented by organic micropollutants, firstly because the biological effects of most of them are poorly known, and secondly, most of them occur at levels too low to be analytically determined. In addition, the behaviour of many of the compounds in freshwater ­ their adsorption, degradation, temporal and spatial variability, as well as their bioaccumulation and combined biological effects ­ are virtually unknown.

Although some of the very toxic organic micropollutants have been either restricted or banned in several European countries during the last 20 to 30 years, they remain in use in other parts of Europe. In the late 1960s and early 1970s several European countries banned the use of DDT; as a result, there has been a subsequent marked reduction in DDT levels in their surface waters (Olsson and Reutergardh, 1986; RIVM, 1992b). Similarly, since the use of PCBs was restricted in several European countries about 15 years ago, PCB levels in their surface waters have also declined (Olsson and Reutergardh, 1986; RIVM 1992b). High levels of PCBs are still found in some rivers, however (Chevreuil et al, 1987).

Micropollutants in inland surface waters originate primarily from industrial and urban activity and from agriculture. Only a few European studies of the pollution sources, their flow pattern and their environmental impact exist. Galassi et al (1992) studied the toxic effect of organic micropollutants on the zooplankton Daphnia magna in the large river Po, in Italy, and found that toxicity was highest in May, this being related to the use of pesticides in the river catchment area. The European aquatic environment is currently exposed to growing numbers and quantities of pesticides as a consequence of the marked increase in the use of pesticides during the last three decades (see Chapter 22). The level of pesticide is particularly high in rivers and lakes located in intensively cultivated areas with crops requiring large quantities of water soluble pesticides together with high surface runoff during the pesticide spraying season. Assessment of the environmental risk of pesticides must be based on a combination of studies of their impact on the aquatic biota and model calculations of potential pesticide runoff.

In the EU a long-term strategy has been adopted (Council Regulation 93/793/EEC) to control the environmental risk from chemicals (see Chapter 38). This, among others, involves testing the ecotoxicity of a list of priority chemicals, including some pesticides.

Accidents and leakage from waste disposal sites

Of the many threats posed by organic micropollutants, industrial accidents and leakage from industrial or agricultural storehouses also need to be considered. For example, the well publicised fire in 1986 at a chemical warehouse in Basle, Switzerland, resulted in a significant discharge of organic micropollutants into the Rhine; the spill caused fish kill in more than 250 km of the river, and macroinvertebrates were affected more than 600 km downstream (Stumm, 1992).

The leakage of substances from the numerous European waste sites, for example landfills and military sites, is another potential source of organic micropollution of inland surface waters.

Radioactivity

Although most radiation stems from natural sources (background radiation), various human activities have increased the potential for contaminating the aquatic environment with radionuclides (see Chapter 16). Natural radionuclides in river and lake waters include: tritium (H-3) (0.02 to 0.1 Bq/1), potassium-40 (K-40) (0.04 to 2 Bq/l), radium, radon and their short-lived decay products (<0.4 to 2 Bq/1).

The effect of human activities can be to enhance these levels (such as through the production of phosphate fertilisers), or to introduce into inland waters artificial radionuclides (such as caesium-137 (Cs-137) and strontium-90 (Sr-90)) by the testing of nuclear weapons, the production of nuclear power (including the operation of the full nuclear fuel cycle) as well as through scientific and medical uses of radionuclides.

Nuclear power production and radioactive waste

At present there are more than 200 nuclear power plants in operation in Europe (see Chapter 19). They are usually built close to a major water source, either inland or coastal, the water being needed for cooling purposes. Besides a reduction in flow and thermal pollution of the water, authorised discharges of low-active effluents from the nuclear power plant will lead to low-level contamination by radionuclides.

The radioactivity levels (total alpha and residual beta ­ ie total beta less K-40 activity) at 35 locations in 11 large EU rivers which have been monitored since 1984 lie mainly between 0.04 and 0.4 Bq/l (CEC, 1989; in press). Sparse information about radioactivity in surface waters was obtained from the countries questioned for the present report. However, the information available indicates levels ranging from 0.01 to 0.8 Bq/l, the highest levels being found in rivers in Ukraine and the Czech Republic, and the lowest levels generally in northern Finland. In some rivers, eg, the Rhone, increased levels of radioactivity have been recorded in the lower reaches (Figure 5.32). This is attributable to discharge from nuclear installations. However, the radiation levels observed are well below those considered of any significance from the point of view of human health (CEC, 1989; in press). While it may be theoretically possible that aquatic biota could be affected in some areas, in practice there is no evidence of such effects.

Nuclear waste disposal sites are another potential source of contamination of inland surface waters. There have been several reports from Eastern European countries of careless handling of radioactive waste products leading to contamination of lakes and rivers.

In the southern Urals, an area of severe contamination is found between Chelyabinsk and Sverdlovsk resulting from the operation of the Mayak nuclear facility and the Kyshtym accident (see Chapters 16 and 18). Whilst operating this plant between 1949 and 1956, high levels of liquid waste were released into the Techa river (about 1017 - Bq in total) causing severe contamination and radiation exposure (Burkart and Kellerer in WHO (in press)). The river is still largely fenced off, although the remaining radioactive contamination is substantially reduced.

Accidents

The Chernobyl accident in 1986 was the most severe nuclear accident to affect almost the whole of Europe (see Chapter 18). It resulted in a large amount of radioactive material being released into the atmosphere. Fall-out spread to many countries, with Ukraine, Belarus, the Russian Federation, Finland, Sweden, Norway and Alpine regions being affected most seriously. The year after the accident the highest effective radiation levels recorded in nearby countries were up to 30 per cent greater than background levels (RIVM/GLOBE, 1992).

In the areas affected, part of the radioactive material was washed out in the rivers and lakes. In the rivers situated near the nuclear plant, the Pripyat and the Dnepr, very high radiation levels (approximately 3500 Bq/l) were observed in the first days following the accident. Within a month the levels decreased by a factor of 100, but high radiation levels can still be observed (Voitsekhovitch et al, 1993). In Finland, the southern part was most heavily affected by the nuclear fall-out and high radiation levels were observed in the two large rivers draining the area, Kokemäenjoki and Kymijoki (Figure 5.33a). In the following years the level decreased, but it is still much higher than before the accident. In areas not so severely affected directly by fall-out, for example northern Finland and around St Petersburg, rivers nevertheless showed appreciable effects of runoff within their drainage areas (Figure 5.33b).

The effect of this fall-out on the biota of Swedish rivers and lakes was studied shortly after the accident (Petersen et al, 1986), but no immediate damage to aquatic organisms was found. Nevertheless, the possibility of a long-term biological effect cannot be ruled out because the effects of the increase in background radiation and the bioaccumulation of radioactivity through the aquatic food chain remain unclear.

Other hazards

Pathogens

Inland surface water polluted by faecal discharge from humans and animals may transport a variety of pathogens (bacteria and viruses). Pathogens in rivers and lakes are generally observed in the most densely populated areas where domestic and animal excreta are not adequately treated. In Europe increasing awareness of this pollution problem during the last century resulted in more efficient wastewater collection and treatment. Additionally waterbodies have been divided into those used for waste removal and those used for drinking water. Today, pathogens in European rivers and lakes generally do not pose a problem for public water supply. However, pathogens may restrict the use of a waterbody for bathing. Council Directive (76/160/EEC) sets standards for the microbiological quality of bathing waters. For the 1992 bathing season the microbiological quality of 5266 freshwater bodies in the EU was assessed, of which 63 per cent complied with standards for total and faecal coliforms (CEC, 1993).

Salinisation

In many dry parts of the world rivers have become saline because of high concentrations of dissolved salts, which may impair the use of water for human drinking and livestock water, just as it affects water use for irrigation and even industry. Salinisation may impoverish the ecological state of the rivers as well.

The downstream areas of the river Volga and Don catchments in the southern part of the Russian Federation are very dry and affected by salinisation, which has also become a problem in other dry parts of Southern Europe. In these places salinisation is aggravated by the increasing use of water for irrigation, which allows rapid surface water evaporation (see Chapter 7).

Mining activity, particularly salt, potash and iron mines as well as ore fields, releases brines, which contain high concentrations of dissolved salts, into inland surface waters. In Poland, the daily load to the upstream reaches of river Wisla amounts to 7000 tonnes of salt. Several other European rivers, such as the Rhine and the Elbe, have also been seriously affected by mining for decades. Chloride impacted rivers are also found in the southern Harz potash mining area in East Germany (LAWA, 1990) where, for example, concentrations reach values above 30 000 mg Cl/l in tributaries to the river Saale.

As an example, Figure 5.34 shows the development of the chloride concentration in the river Rhine. The pollution first became a problem in the 19th century. It then increased gradually until the Second World War, after which it rose steeply (Wolff, 1987). The high salt content of the Rhine water has negatively affected the crops grown by the market gardeners of the Dutch Westland district, and they have fought many juridical battles with the potash mining industry to seek compensation for their economic losses (Wolff, 1987).

CONCLUSIONS

Recent geological evolution in Europe is responsible for creating the complex pattern of aquifers and the multitude of short rivers and small lakes that form the European freshwater system. Thus, only three rivers (the Volga, the Danube and the Dnepr) and two lakes (the Russian lakes Ladoga and Onega) rank among the 30 largest rivers and lakes of the world. There are more than 500 000 natural lakes with a surface area larger than one hectare, and more than 10 000 artificial lakes.

The water resource

Overall, there is no water shortage problem in Europe. However, the amount of water available for sustained consumption is very unevenly distributed across the continent. Total renewable water resources, including water generated internally in a country from precipitation, as well as external contributions from transboundary rivers, show extreme variations that range from less than 100 m3/capita per year in Malta to more than 630 000 m3/capita per year in Iceland.

Internal water resources ranging from under 25 mm to over 4500 mm, taken as average annual river runoff, are determined by rainfall and topography. The highest runoff is in Northwestern Europe and in mountain ranges, decreasing towards the east and south. The annual average runoff for Europe is estimated to be approximately 3100 km3 or 4560 m3/capita per year for a population of 680 million.

Many countries are heavily dependent on external contributions of water through transboundary rivers to meet their demands. Ten countries receive more than 50 per cent of their total water resources from neighbouring countries. Countries located downstream on large rivers (Moldova, Romania, Hungary, Luxembourg and The Netherlands) receive more than 75 per cent of their water in river flows from other countries. This means that there is a source of possible dispute over transboundary water pollution, or disposal of water resources.

Limited analysis of long flow records from throughout Europe indicates that:

On the average, the total annual European abstraction of water amounts to approximately 480 km3 (or 700 m3/capita per year). This is about 15 per cent of the total renewable resources, ranging from 0.1 per cent in Iceland to over 70 per cent in Belgium.

For a number of countries, water abstraction is approximately equal to, or even in excess of, the amount of water generated within the country. This leaves those countries dependent on upstream impacts which can affect the quantity and quality of the water they receive from transboundary rivers.

Groundwater

Overexploitation of aquifers takes place in about 60 per cent of the industrial and urban centres in Europe. About 140 million inhabitants of major cities are supplied with water from overexploited sources, sometimes leading to restricted supply. About 6 per cent of the area of aquifers suited for abstractions suffers from overexploitation, and this portion is increasing. Consequently, severe drawdown of the groundwater level in most of these aquifers has occurred and is probably continuing.

In many European coastal areas, salt - water intrusion due to overexploitation seriously affects soils and the drinking water, or will affect them in the near future. This may threaten the development of urban and tourist areas. This problem is most prominent along the coasts of the Mediterranean, the Baltic and the Black Sea.

The amount of abstractions for agricultural irrigation, the irrigation excess and related effects on the groundwater budget are uncertain due to an almost complete lack of data.

Economic consequences of lowering groundwater tables, such as reduced crop production, deterioration of well-water quality or pumping wells falling dry, are already considered to be important problems. However, ecological consequences of groundwater level drawdown (impoverishment of groundwater-related vegetation and deterioration of aquatic ecosystems due to diminished groundwater supply to rivers) are still not emphasised enough. The potential loss of wetlands is considerable; about 25 per cent of major wetlands are suffering from overexploitation of the groundwater.

Groundwater quality is threatened by high concentrations of nitrate due to excessive use of manure and artificial fertilisers on agricultural soils. Model computations of nitrate leaching from agricultural soils indicate that 87 and 22 per cent of the agricultural area in Europe has nitrate levels above the EU target value (25 mg NO3/l) and the drinking water standard (50 mg NO3/l), respectively. Typical 'hot spots' are found where a high load of nitrogen is applied to sensitive soils overlying sensitive aquifers, as in, for example, parts of Germany, The Netherlands, the UK and Denmark.

Using model calculations and the sparse monitoring data available it has been demonstrated that groundwater quality is threatened by pesticides in almost all European countries despite the fact that there is great spatial variability in both pesticide loads (from 2 to more than 10 kg/hectare for active ingredients) and sensitivity to leaching.

The EU standard of 0.5 &micro;g/l for the total amount of pesticides in drinking water is (according to model calculations) exceeded on 75 and 60 per cent of the total arable and permanent crop land in EU/EFTA and Eastern Europe, respectively. The expected hot spots are in areas with intensive agriculture on top of unconsolidated, unconfined aquifers, covered by soil with low organic matter content ­ for example, in parts of Denmark, northern France, The Netherlands, Lithuania and Belarus.

In the longer term, if no future action is taken to reduce the leaching of nitrate and pesticides to groundwater, it will be necessary to carry out additional purification measures to produce good quality drinking water, and the functioning of groundwater-fed ecosystems might become threatened.

At present, 10 000 to 20 000 km2 is thought to be seriously polluted by point sources. Since much industrial activity takes place in urbanised areas, near which large amounts of groundwater are abstracted, the probability of polluted groundwater being abstracted is relatively high. It has been estimated that, if no remedial action is taken, 100 000 to 200 000 km2 (1 to 2 per cent of the total European area) will be potentially polluted by industry, mining, military activities and landfills within a period of 50 years.

Rivers and lakes in Europe

Monitoring of nutrient and organic concentrations in rivers and lakes is included in most national programmes for surveying the state of the environment. Although the selection of determinants and methods of measurement differ somewhat between countries, it has been possible (by carefully evaluating the available data) to assess the present state of pollution by these substances on a European scale. However, at this stage, it has not been possible to include a general European assessment of water pollution by heavy metals, organic micropollutants and radionuclides, mainly due to lack of comparable data.

Excessive discharges of organic matter from human activities to waterbodies can lead to undesirable changes in the composition of aquatic biota. In addition, the water can become unsuitable for human consumption. Generally, the lowest levels of organic matter are found in rivers in the Nordic countries, and the highest in Central and Southern Europe. There is an evident relationship between the concentration of organic matter in a river and the population density of its catchment.

The concentration of organic matter declined markedly in many European rivers from the end of the 1970s to the end of the 1980s, with the concentration level decreasing at 75 per cent of monitored river sites. The improvement is greatest in Western European rivers, primarily because of intensified wastewater treatment.

Excessive inputs of nutrients (nitrogen and phosphorus) to waterbodies can result in a series of adverse effects known as cultural eutrophication. This can cause significant ecological changes, and negatively affect the quality of water for human consumption and other uses. Phosphorus is usually the primary nutrient responsible for freshwater eutrophication, whereas nitrogen is the main nutrient causing eutrophication of coastal areas and seas.

In general, nutrient levels are lowest in the rivers of the sparsely populated Nordic countries, and highest in those in a band stretching from the southern part of the UK to the Balkan area and Ukraine. There is a close relationship between phosphorus concentrations and the catchment population density and between nitrogen levels and the percentage of the catchment that is agricultural land. Most of the phosphorus loading of inland surface waters is attributable to discharge from point sources, especially municipal sewage water and industrial effluent, while the nitrogen loading is derived primarily from agricultural activity, especially the use of nitrogen fertilisers and manure.

Many European lakes also have high nutrient concentrations, especially those lakes in densely populated areas. Because of the high nutrient levels, many of the lakes have high concentrations of phytoplankton and therefore very turbid water.

The phosphorus concentration in most European rivers and many lakes has decreased during the last 10 to 15 years as a consequence of improved wastewater treatment and substitution for phosphorus in detergents. The ammonium level in many European rivers has also declined, mainly because of improved wastewater treatment and increased awareness of the importance of correct handling and storage of manure and silage. The improvement has been most marked in Western European rivers; phosphorus levels have increased in many Eastern European rivers. In contrast to the phosphorus and ammonium levels, the nitrate level in most European rivers has increased during the last 10 to 15 years, mainly as a result of the increasing use of nitrogen fertilisers.

The nutrient levels in many areas of Europe are still too high, and unless drastic efforts are made to reduce inputs of nutrients, eutrophication is likely to continue to be an important European environmental issue. In many cases, it is essential to remove phosphate in wastewater treatment plants and to reduce the phosphorus content of detergents, as well as to reduce the nitrogen (in particular) and phosphorus loadings from agricultural areas.

Their geological characteristics make rivers and lakes in large parts of Europe sensitive to acidification, especially the Nordic countries, parts of Central Europe and the northwestern part of the Russian Federation. Atmospheric deposition of acidifying substances in these areas has led to an impoverishment of flora and fauna in many rivers and lakes, resulting in decreased biodiversity and fish kills (see Chapter 31).

In general, the concentration of heavy metals in European rivers is well below drinking water standards. However, aquatic biota are usually affected at much lower concentrations. Relatively high concentrations of heavy metals are observed in rivers receiving mine effluents and discharges from metal processing industries.

The use of toxic substances (eg, DDT, PCBs) has been either restricted or banned in several European countries during the last 25 years. As a result, there has been a marked reduction in their levels in surface waters. However, other organic micropollutants suspected of having detrimental environmental effects, and many with unknown effects, are still being discharged into freshwater bodies. Nuclear accidents, due to low safety standards at nuclear power plants and careless handling of nuclear waste, are another potential source of water contamination.

A river with water of good quality does not necessarily support a diverse flora and fauna. For rivers to be of high ecological quality it is also necessary that they do not fall dry and that their natural physical conditions (such as meandering course, pool-riffle sequences) are maintained (where not already destroyed) or re-established. River regulation works, in particular reservoir construction and channelisation, have in almost all European countries severely degraded the physical conditions of rivers, leading to a reduction of their self-purification capacity and to hindrance of migration of important fish species like trout, salmon and sturgeon to their spawning sites in upstream river reaches. The percentage of river reaches still in a natural state is in many countries very low, ranging from below 20 per cent in densely populated countries with intensive agriculture like Belgium and Denmark, to above 70 per cent in Norway, Estonia and Poland, with lower population densities or less intensive agriculture.

To improve the physical conditions, and hence the ecological quality of watercourses, river restoration projects, covering a broad spectrum of river sizes and restoration works, are currently being initiated in a number of countries (including Sweden, Denmark, the UK, Germany and France).

Endpiece

At present, European freshwater resources, although plentiful and of good quality in some regions, are still under threat from a multitude of human impacts. This can reduce the amount and quality of the water available for human consumption and other intentional uses. Because of these problems, better integrated water management is urgently needed to halt and reverse deterioration of water, taking into account the acuteness of the problems, their transboundary nature, their linkage with human activities in the catchments, and their cross-media importance. With this background, the condition of European freshwater resources and their management has been identified as a prominent environmental problem, to be treated in detail in Chapter 33.