Royal Commission on Environmental Pollution

The views expressed in the paper are those of the authors and do not necessarily represent the thinking of the Royal Commission. Any queries about the paper should be directed to the author indicated * above.

Whilst every reasonable effort has been made to ensure accurate transposition of the written reports onto the website, the Royal Commission cannot be held responsible for any accidental errors which might have been introduced during the transcription.

TABLE OF CONTENTS

Executive Summary

1. Introduction

2. Emissions and Burdens of Energy Technologies

  2.1 Fossil Fuels
    2.1.1 Greenhouse Gases
      2.1.1.1 Carbon Dioxide
      2.1.1.2 Methane
      2.1.1.3 Nitrous Oxide
    2.1.2 Acid Emissions and Ozone Precursors
      2.1.2.1 Sulphur Dioxide
      2.1.2.2 Oxides of Nitrogen (NOx)
      2.1.2.3 Volatile Organic Compounds (VOC)
    2.1.3 Particulates
    2.1.4 Other Emissions

  2.2 Nuclear Energy
    2.2.1 Routine Emissions
    2.2.2 Nuclear Accidents
    2.2.3 Nuclear Wastes
    2.2.4 Nuclear Decommissioning
    2.2.5 Other Emissions

  2.3 Renewable Energy
    2.3.1 Direct Emissions
    2.3.2 Indirect Emissions
    2.3.3 Other Environmental Burdens

3. Environmental Impacts
  3.1 Impacts of Climate Change
    3.1.1 Impacts on Human Health
    3.1.2 Impacts on Agriculture
    3.1.3 Impacts on Water Supply
    3.1.4 Impacts of Sea Level Rise
    3.1.5 Impacts of Extreme Weather
    3.1.6 Impacts on Ecosystems
    3.1.7 Other Impacts

  3.2 Impacts of Acidity, Ozone and Particulates
    3.2.1 Impacts on Human Health
    3.2.2 Impacts on Agriculture
    3.2.3 Impacts on Forests and Other Ecosystems
    3.2.4 Impacts on Buildings and Materials

  3.3 Impacts of Radiation
    3.3.1 Impacts on Human Health
    3.3.2 Other Impacts of Radioactivity and Ionising Radiation
    3.3.2 Health Impacts of Non-Ionising Radiation

  3.4 Amenity and Aesthetic Impacts
    3.4.1 Noise
    3.4.2 Landscape Impacts
    3.4.3 Atmospheric Visibility

4. Impact Evaluation
  4.1 Monetary Valuation
  4.2 Sustainability indicators
  4.3 Alternatives and Intermediate Approaches

5. Conclusions

6. References

7. Glossary

ENVIRONMENTAL IMPACTS OF ENERGY

Executive Summary

This report is to provide background material on environmental impacts of energy supply and use for the Royal Commission on Environmental Pollution's study on 'Energy and the Environment'. It has three broad objectives:

• to assess the environmental burdens of different energy sources,
• to review the type and magnitude of the environmental impacts of those burdens, and
• to identify and discuss the main approaches to evaluating those impacts.

The key findings are as follows:

Fossil fuel use is largely responsible for emissions of greenhouse gases, of which carbon dioxide is the most important. All the main fossil fuels contribute significantly; and energy use in industry, homes and transport are all important. In general, gas is the least polluting fuel and coal the most. Energy efficiency is a critical approach to impact reduction.

Greenhouse gases are responsible for climate change, which is potentially the most severe environmental impact of energy use, with a global range and long timescales. Potentially serious effects are expected on human health, agriculture, water supply, ecosystems and coastal zones. Impacts are very uncertain, but potentially disastrous in some developing countries. Major cuts in emissions are required to stabilise the climate.

Fossil fuels are also the main sources of sulphur dioxide and NO_x, which contribute to acid deposition and acid aerosols over Europe. NO_x also contributes to the formation of low-level ozone. Emissions of sulphur dioxide are largest from power generation. For NO_x, the transport sector is the largest emitter. Particulate pollution results both from direct combustion emissions and from acid aerosols. Coal combustion and diesel engines are the largest sources.

Particulates are believed to have the largest health impacts of fossil fuel related pollution, with both acute and chronic effects, including premature mortality. Acidity and ozone also have impacts on materials, crops, forests, fisheries and other ecosystems. Impacts on materials are dominated by the effects of urban sulphur dioxide. In contrast, ozone has the largest impact on crops. Forests, fresh waters and other ecosystems are affected by the gaseous pollutants and by deposition of acidity and nitrogen. Impacts of acidity on sensitive ecosystems have been severe, but are declining. Deposited nitrogen and ozone will become relatively more important.

For nuclear power, environmental concerns are dominated by radiological health impacts. Routine emissions have impacts through the marine food chain and longer term effects due to global dispersion of radioactive pollution. The risks of major accidents are controversial. New nuclear power stations are unlikely in the UK. The use of nuclear reprocessing is challenged on economic and environmental grounds. There are no effective plans in place for final disposal of either high or intermediate level wastes.

Renewable energy sources, in general, offer major benefits by displacing more polluting energy sources. Life cycle emissions are generally small and well planned projects have no serious health or ecological impacts. The main environmental concerns relate to land use and amenity, which could constrain deployment.

There is no agreed method for evaluating environmental impacts of energy. Economic and environmental science approaches are both used. Monetary valuation offers a common metric of evaluation and recent results are reasonably reliable. However, for long term impacts, such as climate change and nuclear wastes, results are extremely uncertain. Other metrics such as "eco-point" ratings have been developed, but are highly subjective. Some approaches focus on the analysis of "safe" pollution levels, but there is increasing evidence that these do not exist for many pollutants.

The quality of data on emissions and other burdens is reasonably good. On the other hand, many analyses of impacts, especially those of climate change, have very high uncertainty. Many topics requiring research are identified and listed in the Conclusions section.

A number of scientific, technical and policy issues are identified which the Royal Commission may wish to examine in more detail. These are brought out in bold type in the body of the report and listed by theme in the Conclusions section.

Despite all the difficult issues and uncertainties in the analysis of the environmental impacts of energy supply and use, it can be concluded that current patterns of energy use in the UK are not environmentally sustainable.

ENVIRONMENTAL IMPACTS OF ENERGY

1. Introduction

Production and use of energy have always had an impact on the environment. Widespread effects followed the Industrial Revolution. Smoke and sulphur dioxide, largely from coal burning, were produced in large amounts in many cities. In some weather conditions there were obvious effects on health, most notably in the "great London smog" of 1952. The Clean Air Acts alleviated the worst problems by the 1960s. Concerns about urban air pollution now focus on fine particulates, nitrogen dioxide, ozone and other toxins, largely related to vehicle emissions.

Concerns about environmental pollution outside urban areas are more recent. For the energy sector, initial concerns focused on acute impacts close to power stations, mines and other production facilities. By the early 1970s, long range atmospheric transport of certain pollutants was evident - notably the impacts of acid deposition on forests and fresh waters. It was recognised that impacts in acid sensitive areas many hundreds of kilometres away from large pollution sources may be much larger than impacts closer to the emitters.

Widespread recognition of the enhanced greenhouse effect is even more recent. The basic science is well understood, and there is no doubt that greenhouse gases are accumulating in the atmosphere and that some warming will result. The actual level of warming is subject to some uncertainty. Large regional variations in temperature rise and rainfall are predicted, but existing models are insufficiently detailed to forecast these with great confidence. Many potential impacts have been documented, notably by the Intergovernmental Panel on Climate Change, but big uncertainties and research needs remain (Watson et al, 1996).

The main alternative to fossil fuels in the UK has been nuclear power. This does not directly emit conventional pollutants. However, it has its own environmental problems, largely due to the health effects of ionising radiation. The main risks come from the potential consequences of accidents and the storage of radioactive materials, particularly wastes.

Renewable energy sources are generally considered to be more environmentally benign. Whilst current production in the UK is rather small, the technologies are developing rapidly and many have great potential. With the exception of biomass fuels, there are no significant direct emissions. However, the renewables are not impact free. They have implications for land use and there are particular amenity issues, such as the visual impacts of wind turbines.

This report has three broad objectives:

• to assess the environmental burdens of different energy sources,
• to review the type and magnitude of the environmental impacts of those burdens, and
• to identify and discuss the main approaches to evaluating those impacts.

There are suggestions for issues on which the Royal Commission might wish to focus.

The report covers all the major energy sources:

• fossil fuels (coal, oil and natural gas),
• nuclear energy and,
• the major renewable energy sources - hydropower, wind, biomass and photovoltaics.

Different end-use technologies are distinguished where this affects the environmental impacts. The technologies considered are those identified as important in the medium term in other background papers for the Royal Commission (Jackson, 1998; Smith, 1998). Attention is concentrated on use of energy in electric power generation, domestic, industrial and commercial uses. Less attention is given to transport, in view of the Royal Commission's extensive recent work on its impacts (Royal Commission, 1994; Royal Commission, 1997).

Impacts covered include:

• the main impacts of climate change, including effects on human health, agriculture, water supply, ecosystems and the impacts of sea level rise and extreme weather events,
• impacts of regional scale air pollution (particulates, acid gases and tropospheric ozone) on human health, materials, crops, forests and ecosystems,
• human health impacts of radiation, and
• impacts on amenity, for example, noise, visual intrusion and impaired visibility.

They are reported in the four categories shown. This enables a division of impacts by scale - global, regional and local. The use of a separate category of radiation impacts allows a clear distinction of the impacts of nuclear energy, which occur at all scales, from those of other sources. With some exceptions, impacts of renewable energy are local and affect amenity, and therefore the amenity category allows some focus on renewable energy impacts. Relevant timescales are distinguished throughout.

The approach used in this report draws upon a "bottom-up", impact pathway methodology developed in the European Commission's comprehensive ExternE Project (CEC, 1995a). The report is accordingly divided into three stages, corresponding to the three objectives:

• assessment of burdens (emissions and other "bads") resulting from each energy source,
• quantification of important impacts to health and the environment, and
• impact evaluation methods and results.

Assessment of burdens is undertaken separately for fossil fuels, nuclear energy and renewable sources, as there is limited overlap between their important burdens. For fossil fuels, priority is given to impacts mediated via long range air pollution, as most earlier analyses and policy concerns indicate these are the most important. The emissions responsible are greenhouse gases, fine particulates and the precursors of acid deposition and tropospheric ozone. For nuclear power, the emphasis is on releases of radioactivity - both routine and accidental. For renewable energy sources, no single category of effects is dominant and more eclectic approach is taken.

The analysis uses a "full fuel cycle" approach, addressing burdens due to all stages from fuel extraction through to end use. The whole life cycle of each technology is considered (construction, operation and decommissioning), but, within this framework, attention is concentrated on the activities which are the largest polluters.

The impact assessment uses reports from peer-reviewed studies by expert groups, including those of the Intergovernmental Panel on Climate Change (IPCC) for climate change, the United Nations Scientific Committee on the Effects of Atomic Radiation (UNSCEAR) for ionising radiation and various review groups of UK Government departments. The consensus of expert opinion is reported, including an assessment of the uncertainty. Where there is no consensus, this is reported with a description of the underlying controversies and the research required or in progress to resolve them. Results of unpublished work are reported where they advance the state of knowledge.

There is no consensus on the best methodology for evaluation of environmental impacts. Not only are there significant uncertainties in data, but also subjective differences in evaluation approach. Two broad schools of thought may be identified. One is based in cost-benefit analysis, in which the monetary value of environmental impacts is compared with the costs of abating pollution. The other is based on the achievement of environmentally acceptable goals with economic analysis limited to assessment of cost effective approaches for delivering environmental targets. The strengths and weaknesses of both approaches are assessed.

The conclusions summarise what is known with reasonable confidence about the impacts of energy use and what is broadly agreed about their evaluation. Issues which the Royal Commission may wish to pursue are highlighted.

2. Emissions and Burdens of Energy Technologies

2.1 Fossil Fuels

Greenhouse gases are those gases which contribute, directly or indirectly, to global climate change. The Kyoto Protocol to the United Nations Framework Convention on Climate Change (FCCC) requires the UK and other EU countries to reduce annual average emissions by 8% from 1990 levels by the period 2008-2012 (FCCC Secretariat, 1997). Some reductions may be achieved by emissions trading with other developed countries and joint implementation with developing countries. The detailed mechanics of these schemes may critically affect the effectiveness of the Protocol and remain to be clarified at the next meeting of the parties later in December 1998. The Royal Commission may therefore wish to give early consideration to the role of emissions trading and Joint Implementation in international greenhouse gas control agreements.

At Kyoto, the EU argued for a 15% reduction in greenhouse gas emissions from developed countries by 2010 (CEC, 1997). The UK has a more stringent unilateral target of reducing carbon dioxide emissions by 20% from 1990 levels (Labour, 1997) and the UK Government is urging the EU to adopt its Kyoto negotiating position as a unilateral target.

Six gases are controlled by the Kyoto Protocol - carbon dioxide, methane, nitrous oxide, hydrofluorocarbons (HFCs), perfluorocarbons (PFCs) and sulphur hexafluoride (SF₆). The last three are used mainly in specific applications - HFCs in refrigeration, PFCs in aluminium manufacture and SF₆ in high voltage equipment. They make minor contributions to global warming and, except to the extent SF₆ is used in electricity transmission, are not energy related, and therefore not further considered here.

Other gases also contribute to climate change (Schimel et al, 1996). Chlorofluorocarbons (CFCs) and hydrochlorofluorocarbons (HCFCs) are greenhouse gases, but also destroy stratospheric ozone which is itself a greenhouse gas. The net effect on global warming is not certain. Emissions of CFCs and HCFCs are controlled under the Montreal Protocol. They are not generally energy related and therefore are not considered here.

Emissions of oxides of nitrogen (NO_x) and volatile organic compounds (VOC) are precursors of tropospheric ozone, and emissions of sulphur dioxide contribute to aerosol formation. Both ozone and aerosols are relatively short-lived compared to the direct greenhouse gases. They affect radiative forcing in opposite directions (ozone positively, aerosols negatively), but with a net effect which is small compared to that of the direct greenhouses gases. They have other important impacts and the relevant emissions are therefore considered in Section 2.1.2

The relative contribution of different gases to climate change is complex. Each has a different infra-red absorptive capacity (forcing factor), but also a different lifetime, so that relative contributions over time vary. The most common indicator is the 100 year global warming potential (GWP). This is defined as the contribution of unit mass of a gas to radiative forcing over 100 years relative to unit mass of carbon dioxide (Derwent, 1990).

The use of GWPs has been criticised on the grounds that the time horizon is arbitrary and that they do not accurately reflect the scale of impacts and damages. Alternative measures which incorporate discounting have been proposed (Lashof and Ajuha, 1990; Reilly and Richards, 1993; Hammit et al, 1996). However, the Kyoto Protocol measures of aggregate emissions will be based on 100-year GWP, so this is likely to remain the metric of aggregation despite its problems. Conventional GWPs based on a pulse of emissions may under-estimate the effect of a sustained (step change) in emissions of ozone precursors (Derwent, 1994; Fuglestvedt et al, 1996). Some modification of the GWP methodology may therefore be needed if short lived gases are to be added to emissions control constraint regimes.

Emissions of the major direct greenhouses gases are listed in Table 1.

Table 1

Greenhouse Gases and their Global Emissions

1. 100 year Global Warming Potential (Schimel et al, 1996)
2. 1980-1990 average (Houghton et al, 1996)

The uncertainties in global emissions are estimated to be ± 15% for carbon dioxide, ± 24% for methane and ± 45% for nitrous oxide (Houghton et al, 1996). The uncertainties in GWPs are estimated to be ± 35% (Schimel et al).

2.1.1.1 Carbon Dioxide

Carbon dioxide emissions have been responsible for most anthropogenic global warming over the last century, and they will remain the most important contributor. Approximately 80% of carbon dioxide emissions are from the combustion of fossil fuels.

Combustion of all fossil fuels produces carbon dioxide, but the emission factors (emissions per unit of energy output) are not equal. The UK National Atmospheric Emissions Inventory is based on emissions factors developed by the National Environmental Technology Centre (Salway et al, 1997; Eggleston and McInnes, 1987)). For carbon dioxide, these are based on recent industry estimates of carbon content and are broadly consistent with other sources (e.g. Grubb, 1989; Bates, 1995). Average values for the main fossil fuels are shown in Figure 1. Carbon contents are highest for coal and lowest for natural gas. Abatement of carbon dioxide at source, by capture and disposal is technically possible, but not currently economically feasible (IEAGHG, 1994; ETSU, 1994a). In the absence of viable "end-of-pipe" technology, emissions depend only on the type and quality of fuel burnt.

Figure 1
Emissions of Carbon Dioxide from Different Fuels

Source: DTI, 1997

Given the limited potential for emissions abatement by end-of-pipe processes and switching between fossil fuels, more efficient energy use is widely accepted to be the most cost effective approach to reducing emissions. Energy efficiency is the subject of another background paper (Fisher, 1998), and therefore is not considered here, but that should not be taken to detract from the importance of energy efficiency for reducing emissions.

Carbon dioxide is also emitted from other stages of fuel cycles, notably electricity generation. Emissions factors are as shown in Figure 1, but the emissions due to use of a unit of electricity depend upon the efficiency of power generation, which can range from 30% for an old coal or oil powered plant, to 50% for a new combined cycle gas turbine and 80% or more for a combined heat and power plant (CHP). The emissions from fossil fuel power generation can therefore be reduced significantly by using the most efficient technologies, especially CHP. The Government effectively acts as planning authority for power plants under Section 36 of the 1989 Electricity Act. The Royal Commission may therefore wish to consider the merits of a strong presumption in favour of CHP in land use planning.

Emissions from other fuel cycle stages are typically very much lower than from combustion for end use and power generation. For UK fuels, electricity use in coal mining contributes 7 g/kWh (Bates, 1995), fuel use in oil production (including flaring) 12 g/kWh and in gas production and pressurisation 6 g/kWh (Gover et al, 1996). For oil products, carbon dioxide emissions due to refineries depend on the methodology for apportioning emissions between products. They are estimated to be 11 g/kWh for diesel, but 22 g/kWh for petrol (Gover et al, 1996). Fuel transport contributes little to carbon dioxide emissions unless the distances are very large. Even transporting coal from Australia only adds 4% to total fuel cycle emissions (IEAGHG, 1994). Most fossil fuels used in the UK are of domestic origin, and therefore fuel transportation is a minor emission source.

Fuel use in manufacture and decommissioning of combustion equipment is relatively small. For non-fossil fuels, these fuel cycle stages are the main sources of carbon dioxide (see Section 2.3.2). However, for fossil fuel power stations, these emissions are only about 1% of operating emissions over the plant lifetime (Meridian, 1989; ORNL, 1994a). In transport, emissions from vehicle manufacture are about 14% of lifetime exhaust emissions (IEA, 1993).

Total UK carbon dioxide emissions by fuel and by end use sector are shown in Figures 2 and 3 respectively. Emissions are estimated to be accurate to within ± 5% (Salway et al, 1997a).

Figure 2
UK Carbon Dioxide Emissions by Fuel (1995)

Source: DETR, 1997a

Figure 3
UK Carbon Dioxide Emissions by End Use Sector (1995)

Source: DETR, 1997a (corrected for error in service sector)

2.1.1.2 Methane

Methane emissions from combustion are typically more than 10,000 times less than those of carbon dioxide (Eyre and Michaelis, 1991; Salway et al, 1997), but upstream emissions in the fuel cycle are larger. Release during mining is the biggest source from coal. The methane content of coal is a strong function of the depth of the coal seam. Deep mined coal typically releases a weight of methane approximately equal to 1% of the coal extracted (Williams, 1994; IEAGHG, 1994; CEC, 1995b). This gives fuel cycle emissions of approximately 1.4 g/kWh. Opencast coal extraction produces at least ten times less (Williams, 1994).

In the gas fuel cycle, annual upstream losses from the UK Continental Shelf are estimated to be about 123,000 tonnes (DETR, 1997a), which is less than 0.1% of production. Onshore losses are more controversial. Some early estimates indicated that emissions might be several percent (Mitchell et al, 1993; Wallis, 1991). Based on subsequent empirical work by British Gas, estimates are now lower - around 1% - mainly from the old cast iron low pressure network (Williams, 1994). But emissions from major producers outside the OECD are not so well documented, and may well be higher, notably in Russia (Makarov and Bashmakov, 1990), so that connection of the UK to the European grid using Russian gas may have important implications for methane emissions.

UK methane emissions are summarised in Table 2. Uncertainties differ by source, but average ± 30-40% (Salway et al, 1997a). Even when weighted by GWP, emissions of methane are small compared to those of carbon dioxide.

Table 2
UK Methane Emissions (1995)

2.1.1.3 Nitrous Oxide

Nitrous oxide is also emitted from combustion. UK combustion emissions are estimated to be 20,700 tonnes annually, which is only 22% of the UK total (DETR, 1997a). Emission estimates have relatively high uncertainty. Combustion emissions from coal fired power plants are typically 0.01 g/kWh (Laird and Sloan, 1993) giving a radiative forcing negligible in comparison to carbon dioxide from the same source. However, there is significant technological variation. Fluidised bed combustion emissions are typically 0.2 g/kWh (Dahlberg et al, 1988; Fenger et al, 1990) due to lower combustion temperatures. In the transport sector, catalytic converters also raise emissions to 0.06 g/kWh (Royal Commission, 1994; Wade et al, 1994). But radiative forcing of nitrous oxide from all combustion sources is very small compared to carbon dioxide emissions.

2.1.2 Acid Emissions and Ozone Precursors

Acid deposition is caused by emissions of sulphur dioxide (SO₂) and oxides of nitrogen (NO_x). Tropospheric ozone is related to emissions of NO_x and volatile organic compounds (VOC). In both cases, there is complex atmospheric chemistry, involving a range of other pollutants including ammonia. However, sulphur dioxide, NO_x and VOC are the only species involved which are significantly related to energy use. Both sulphur dioxide and NO_x also contribute to the formation of fine particulates.

These pollutants are more reactive than the major greenhouse gases, and therefore have shorter lifetimes in the atmosphere. Their impacts are therefore largely, but not entirely, confined to the regional (i.e., European) scale. The major emitters are again combustion processes, in power generation and at the point of energy use. The level of emissions depends critically upon the fuel type, and the combustion and emissions abatement technology.

2.1.2.1 Sulphur Dioxide

Sulphur dioxide emissions originate from the combustion of sulphur in the fuel. Both coal and heavy fuel oil contain significant quantities of sulphur, light oils less, and natural gas usually only a negligible quantity.

Common coals have a variety of sulphur contents ranging from under one percent to several percent. UK coal has a typical value of 1.6% (Bates, 1995). It is possible to remove sulphur in fuel processing. For coal sulphur removal is unusual and the only sulphur abatement process in use in the UK is flue gas desulphurisation (FGD) which removes 90% of the pollutant at two power stations (Drax and Ratcliffe-on-Soar). These power stations are not fully utilised because FGD raises operating costs and the electricity "Pool" market fails to account for environmental benefits. The Royal Commission may which to pursue this apparently perverse operation of the Pool in ignoring environmental issues.

Newer clean coal technologies, such as integrated gasification combined cycle (IGCC), achieve higher levels of abatement. These are not currently used in the UK, but are under consideration. For details, see Background Paper 2 (Smith et al, 1998).

Some sulphur is removed from oil in refining, largely to meet the increasingly stringent requirements for transport fuels. Sulphur content limits are currently 0.05% for both diesel and petrol respectively, with further reductions proposed under the EU Auto-Oil Directive to 0.035% for diesel and 0.015% for petrol by 2000 and 0.005% for both in 2005. Heavier products have higher sulphur contents. Proposals under the EU acidification strategy seek to limit these to 0.2% for gas oils and 1% for heavy fuel oil.

Figure 4 shows the relative sulphur dioxide emissions per unit of energy for both fuels and power generation sources. It makes clear the relative importance of coal fired power stations and the "dirty" traditional fuels, and also illustrates the potential of fuel switching and technological change.

Full fuel cycle analysis adds little to this overall picture. In some cases, natural gas has a significant hydrogen sulphide content, but this is usually removed as elemental sulphur. The gas fuel cycle only releases sulphur dioxide where "sour gas" is flared and this is very limited in the UK (Eyre, 1990; Bates, 1995). The only oil fuel cycle stage other than end use and power generation where large quantities of energy are used is refining. Sulphur dioxide emissions here are typically 0.1 to 0.4 g/kWh of product (Gover et al, 1996). Emissions in fuel transportation are only significant where a high sulphur fuel is used, such as marine fuel oils, which currently have very high sulphur contents compared to other oil fuels. International negotiations to control marine transport sulphur dioxide emissions are ongoing under the auspices of the International Maritime Organisation (UKNAQS, 1997).

Figure 4
Sulphur Dioxide Emissions from UK Fuels

Basis: Fuels - g/kWh of fuel input; Electricity - g/kWh of electricity output
Source: Author's calculations assuming
1. UK coal with 1.6% sulphur content.
2. Oil fuels of quality proposed for year 2000.
3. Emissions abatement of 90% using FGD and 99% using clean coal technology.

Total UK emissions of sulphur dioxide in 1995 were 2.365 Mt. The breakdown is given in Figure 5, which shows the dominance of power stations as a source. The 1995 total is only 53% of the 1980 figure and, as the use of coal in power generation falls, further reductions can be expected. The uncertainty in emissions is estimated to be ± 10% (Salway et al, 1997).

In 1990 UK emissions were 23% of the EU total and 10% of the European total in 1990 (Amann et al, 1996). Under the Second Sulphur Protocol of the United Nations Economic Commission for Europe (UNECE), UK emissions have to be further reduced to 1.47 Mt (i.e. by 38% of 1995 levels) by 2005 and 0.98 Mt (59%) by 2010. Emissions from existing large plant (>50 MW thermal) are controlled under the EU Large Combustion Plant Directive and have to be reduced to 1.553 Mt by 2003. This is not an onerous goal, but the Directive is being redrafted and the current proposal from the European Commission is that UK emissions should be reduced to only 0.075 Mt (i.e. by 97% from 1995 levels) by 2010. The Royal Commission may wish to consider whether the European Commission's proposals for separate national targets for acid emissions from large plant are feasible or necessary.

The Environment Agency is also consulting on emissions from power stations in England and Wales. It has proposed that the current target of reducing emissions by 85% from 1991 levels should be brought forward by 4 years from 2005 to 2001, because of the rapid penetration of gas into power generation (Environment Agency, 1998). Whilst these limits would be less drastic than those proposed by the European Commission, they would still increase the economic pressures on the older coal- and oil-fired power stations.

Figure 5
UK Sulphur Dioxide Emissions by Source (1995)

Source: DETR, 1997a

2.1.2.2 Oxides of Nitrogen (NO_x)

Emissions of nitric oxide (NO) and nitrogen dioxide (NO₂) are referred to collectively as NO_x. They result from both the oxidation of nitrogen in the fuel and the reaction of atmospheric oxygen and nitrogen. Emissions are not a simple function of fuel composition, but also depend on combustion conditions. For large combustion plant, like power stations, "low NO_x" burner designs to reduce flame temperatures can reduce emissions by 30-40% (Sloan and Laird, 1991). Typical values for NO_x emissions are shown in Figure 6. Power generation and transport sources have relatively high emissions.

"End-of-pipe" treatments like selective catalytic reduction, which chemically reduce NO_x using ammonia can achieve 75% reductions in power generation (Hjalmarsson, 1990; CEC, 1995b). They are used in some other countries, notably Japan and Germany, but not in the UK.

NO_x emissions from stages of fuel cycles other than end use and power generation are generally much smaller. In petroleum production, emissions depend on the extent of flaring, but average 0.03 g/kWh for oil products and 0.01 g/kWh for natural gas. Refinery emissions are typically 0.02-0.05 g/kWh of product (Gover et al, 1996). Emissions in transport are only high in extreme cases, e.g. for long haul transport of coal by sea, the emissions may be to 0.7 g/kWh (IEAGHG, 1994).

UK total emissions were 2.295 Mt in 1995. The breakdown is shown in Figure 7. The critical role of transport can be clearly seen. The uncertainty in emissions is estimated to be ± 30% (Salway et al, 1997).

Figure 6
NO_x Emission Factors for Typical UK Sources

Basis: Fuels - g/kWh of fuel input; Electricity - g/kWh of electricity output
Source: Author's calculations based on Salway et al,1997; CEC, 1995b,c; Gover et al, 1996

Figure 7
UK NO_x Emissions by Source (1995)

Source: DETR, 1997a

UK emissions were 18% of EU total emissions and 12% of the European total in 1990 (Amann et al, 1996). The UK is committed under the EU Large Combustion Plant Directive to reducing emissions from existing large plant (>50 MW thermal) to 0.75 Mt by 1999, but the Directive is being revised and European Commission proposals are for a UK limit of 0.06 Mt in 2010. These proposals have implications for the UK power sector (see Section 2.1.2.1). In addition, new UK commitments under the UNECE Convention on Long Range Air Pollution will begin to be negotiated in 1998.

2.1.2.3 Volatile Organic Compounds (VOC)

VOC is not a single substance but a mix of different compounds. To some extent, each has a different environmental impact. Some, for example benzene and 1-3 butadiene, are carcinogens (QUARG, 1993a; EPAQS, 1994a,b); others, like formaldehyde, are also of health concern (QUARG, 1993b). It is useful to consider them as a class of compounds, because all contribute to ozone formation. But they have different potential to form ozone, so accurate modelling requires identification of Photochemical Ozone Creation Potentials (POCP) (Derwent et al, 1995) and the production of speciated VOC inventories (PORG, 1993).

VOC emissions (excluding methane) come from a number of sources, as shown in Figure 8. Total UK emissions in 1995 were 2.337 Mt. About half is energy related. The largest energy-related contribution is from road vehicles. This has been considered in earlier reports of the Royal Commission (Royal Commission, 1994; Royal Commission, 1996). The uncertainty in emissions is estimated to be ± 50% (Salway et al, 1997).

Figure 8
UK Emissions of VOC (1995)

Source: DETR, 1997a

The UK is committed under the UNECE Protocol on VOC to reducing emissions to 1.905 Mt by 1999.

2.1.3 Particulates

Emissions of particulates are of concern largely because of their effects on human health. But particulates are also responsible for some other impacts, in particular reduction of visibility and soiling of buildings. The impact considered affects the appropriate measure of particulate pollution. For health effects there is general agreement that it is fine particulates which are of concern, but less agreement whether this should be measured as PM₁₀ (particles smaller than 10m in diameter) or a smaller size fraction such as PM_2.5. For soiling of materials, the preferred measure is "black smoke" which includes particles smaller than about 4 m with source specific multiplicative factors to allow for differential soiling capability (Hamilton and Mansfield, 1992). A similar measure is preferred for visibility effects (Landrieu, 1997). Because, health impacts are of the greatest concern, the focus here is on fine particulates.

UK emissions of PM₁₀ in 1995 were 0.232 Mt (DETR, 1997a). A breakdown by source is given in Figure 9. Transport, industry and the domestic sector are all significant contributors.

Figure 9
UK PM₁₀ Emissions by Source (1995)

Source: DETR, 1997a

The inventory uncertainties for black smoke are ± 20-25% (Salway et al, 1997). For the PM₁₀ fraction they are higher (QUARG, 1996) and the government has recently announced the formation of an Airborne Particles Expert Group to review the inventory (ENDS, 1997a). Estimates of emissions factors are given in Figure 10, based on work for UK Government (Gover et al, 1996, QUARG, 1996) and the European Commission (CEC, 1995b,c).

Transport is of greater importance for health impacts than Figures 9 and 10 might indicate, because a high proportion of emissions are at ground level in areas of high population density. Diesel vehicles are the main emitters. Household coal combustion is an important source where it is used extensively.

Not all PM₁₀ in the atmosphere is due to direct anthropogenic emissions. There are also important natural sources, such as dust. In addition, the primary pollutants sulphur dioxide, NO_x and ammonia contribute to PM₁₀ concentrations via the formation of secondary aerosols. A large fraction of sulphur and nitrogen is present in as secondary aerosols (RGAR, 1997). These tend to be small, <2m, and contribute about 30% by weight of urban particulates (QUARG, 1993a), which has implications for the attribution of particulate impacts.

Figure 10
Estimates of PM₁₀ Emission Factors

Source: Author's calculations based on Gover et al, 1996; QUARG, 1996; CEC, 1995b,c

A semi-quantitative guide to the relative contributions of the different fuels and sectors to emissions of the main air pollutants from the UK (based on 1995 data) is given in Table 3.

Table 3
Contributions to Air Pollution in the UK

2.1.4 Other Emissions

Coal contains radioactive isotopes which are emitted in combustion. Most activity is from radon-222, but most human dose comes from actinides emitted in fly ash and then inhaled (Wan and Wrixon, 1988). The uncertainty is high because of the variability of trace element content in coals. The collective dose to the UK population in 1988 was 10 person.Sv (Hughes et al, 1989), implying a collective dose of about 0.05 person.Sv/TWh - three orders of magnitude lower than the estimates for the nuclear fuel cycle (see Section 3.3.1).

Fossil fuel burning also has a positive impact, by emitting carbon dioxide depleted in carbon-14, thereby reducing the radioactivity of biomass carbon and hence the dose received from internal sources (Suess, 1965). The collective dose reduction over the lifetime of carbon dioxide in the atmosphere is small, but likely to outweigh the effects of ash emissions.

Emissions of toxic trace elements to air are very limited. Coal and heavy oil contain trace elements, but, with the exception of mercury, they mainly end up in the ash which is largely captured (CEC, 1995b). Nevertheless, combustion is a major source of air emissions of lead, copper, nickel, chromium, arsenic and selenium (Salway et al, 1997). Emissions in cooling water are also small, but emissions to water from the FGD process are potentially more serious. Heavy metals are captured in the FGD system and may enter the liquid effluent when the gypsum output of the FGD plant is purified. This effluent is treated to remove most of the toxic components as a sludge, but inevitably some remains in solution. Concentrations of trace metals in waste water are typically in the range 1-50 ppb (CEC, 1995b; IEAGHG, 1994), which are comparable with environmental quality standards (EQS) for inland waters, but the emissions are rapidly diluted, so there is little chance of standards being breached.

Cumulative effects in the North Sea, into which most UK coal-fired power plants drain, may need to be considered. A single new 2000 MW power plant might add the following to total UK emissions:- 2.5% for mercury, 0.5% for cadmium, 0.1% for lead, 0.08% for copper and 0.03% for zinc (based on CEGB, 1988 and DETR, 1997a). The higher figures might be judged some cause for concern, and therefore warrant investigation of improvements in abatement technology since the data on which they are based. The Environment Agency's revised Chemical Release Inventory (available from February 1998) should provide better data on most trace element releases. Cumulative effects in the sediments of the Humber Estuary, which drains the Central Coalfield have not been investigated (CEC, 1995b). As the UK Government has announced its intention to reduce emissions of hazardous substances to the marine environment to "close to zero" by 2020, and the Royal Commission may want to examine heavy metal emissions from power generation to water in the light of international treaty obligations.

FGD has other potentially problematic burdens. It uses limestone (calcium carbonate) to react with sulphur dioxide, producing a solid product of gypsum (calcium sulphate). Limestone sources close to the main UK coalfield are concentrated in the South Pennines. As a matter of policy, actual sources are outside the Peak District National Park (National Power, 1998), but may be close enough to be easily visible and audible from it. A typical medium sulphur coal produces 60 kg of gypsum for every tonne burnt. This is used for plasterboard manufacture (thereby reducing gypsum mining), but it is not clear that all the product of an expanded programme could be utilised.

Large amounts of mine spoil are produced in coal production. The general amenity impacts are considered in Section 3.4.2. Ground water from coal mines is often acidic and may be contaminated with heavy metals. Control is possible, but problems continue after mine closure and the contraction of the deep mined coal industry may therefore generate new problems.

For a typical coal, each tonne burnt in power stations produces 150 kg of ash. About 10% of this accumulates in the boiler as "bottom ash"; the remaining 90% is entrained in the flue gas as "fly ash" and most is captured in the flue gas treatment. Both fly ash and bottom ash can be used in construction. In Germany, federal regulations require the use of combustion residues wherever possible and 90% is reused. In the UK, most bottom ash is used in road construction, but only 30-40% of fly ash is recycled (National Power, 1998). It poses disposal problems because of the toxic metal components. These can leach from ash disposal sites (Chadwick et al, 1987; Cope and Dacey, 1984; Jones, 1995). Modern landfill operations, in which leachate is controlled through the use of impervious linings, reduce problems. Wastes from some clean coal technologies are a more stable slag and therefore likely to be less problematic (Clarke, 1991). The overall impacts of trace elements in coal can be controlled to acceptable levels with good practice (Swain and Goodarzi, 1995). Nevertheless, landfill is far from an ideal solution, so the Royal Commission may wish to investigate the scope for increasing ash recycling rates.

There are oil emissions direct to the sea from offshore oil exploration and extraction processes. Drill cuttings and produced water are the main contributors, generating about 10,000t annually (DETR, 1997a). Water based drilling fluids are now replacing more damaging oil-based muds. Oil spills from rigs are relatively small, but tanker accidents can be much larger, e.g. the Braer disaster alone resulted in spillage of 84,000t. The impacts of a major spill are very dependent on weather conditions and their effect on oil dispersal. A variety of modelling approaches have been developed for both individual spills and bigger areas (see e.g. Venkatesh, 1990; Elliot, et al, 1992). Impacts on coastal and marine species are complex and many studies on individual species have been published.

Decommissioning of offshore oil rigs is a controversial issue, especially following the Brent Spar incident. International Maritime Organisation guidelines require removal of smaller rigs from shallower waters (CEC, 1995c) and this is now being undertaken in the North Sea (DTI, 1997). The environmental merits of different disposal options are under investigation, but the UK Government has announced to partners in the Oslo and Paris Commissions (OSPAR) that on land disposal is now planned wherever "safe and practicable" (ENDS, 1997b).

There are other environmental burdens of the activities involved in using fossil fuels. These range from the heat in cooling tower plumes to the wastes of construction activities. Most are of mainly local importance and, properly managed, they are relatively unimportant.

2.2 Nuclear Energy

Nuclear energy is released by the fission of uranium and plutonium nuclei, rather than by chemical combustion, and therefore there are no significant emissions of the pollutants most associated with fossil fuel use. The important impacts of nuclear energy arise due to its use and production of radioactive materials. Routine emissions, accidental emissions, radioactive wastes and decommissioning are treated separately in the following sections.

2.2.1 Routine Emissions

The fuel cycle for UK nuclear power consists of the following stages:

• ore extraction and processing - mining and milling,
• fuel manufacture - conversion, enrichment and fabrication,
• power generation,
• fuel reprocessing, and
• radioactive waste disposal.

In each case facilities have to be decommissioned at the end of their operational lives.

Uranium mining is undertaken entirely outside the UK. Emissions of radon from mill tailings last for tens of thousands of years. The collective health impacts are generally the largest of the nuclear fuel cycle (UNSCEAR, 1993; ExternE, 1998d), but are mainly confined to the mining region (see Section 3.3.1). Emissions of some important radioisotopes from other fuel cycle stages in the UK are shown in Table 4.

Table 4
UK Radioactivity Emissions from the Nuclear Industry

Carbon-14, krypton-85, iodine-129 and tritium have lifetimes in excess of 10 years and are global distributed (CEC, 1995d). They are therefore of concern with respect to long term collective dose impact. Emissions have risen in recent years. Health impacts are discussed in Section 3.3.1.

Emissions of caesium-137, technetium-99, plutonium-241 and alpha emitters are largely to the marine environment. UK emissions are overwhelmingly from the Sellafield reprocessing plant in Cumbria. Although all have sufficiently long half lives to become very widely distributed, impacts are largest in the Irish Sea. Emissions of technetium-99 have increased dramatically in recent years, as a result of which it has been found in lobsters in the Irish Sea in concentrations which exceed intervention levels in the aftermath of a nuclear accident (MAFF, 1996).

The Environment Agency is currently consulting on proposals to alter emissions authorisations for Sellafield, including increased authorisations for emissions to air of carbon-14, iodine-129 and ruthenium-106. Some reductions in Tc-99 emissions to sea are proposed, but the Government has made a commitment to OSPAR partners to reduce emissions to close to zero in the long term. It is not clear that proposals are consistent with this.

Reprocessing extracts much of the plutonium and depleted uranium from spent fuel. In the absence of strong markets for these products, reprocessing is now widely considered to be less economic than long term storage of spent fuel (Berkhout and Walker, 1990; ETSU, 1994a; Beck, 1994; Walker, 1997). Reprocessing also contributes significantly to the economic liabilities of the industry (MacKerron and Sadnicki, 1995) to nuclear proliferation problems (Royal Society, 1998) and is the dominant source of intermediate level waste (see Section 2.2.3). The Royal Commission may wish to explore how the Government and BNFL intend to meet international commitments to reducing emissions of radioactivity to the marine environment, and whether reprocessing of spent fuel should be phased out.

2.2.2 Nuclear Accidents

Serious accidents from world-wide nuclear operations have been recorded at:

• Windscale (now Sellafield) in 1956, due to a graphite fire in a reactor, leading to an estimated collective dose of 300 person.Sv,
• Kyshtym in the Soviet Union in 1957, due to a cooling failure in a high level waste tank, leading to an estimated collective dose of 1200 person.Sv,
• Three Mile Island in the USA in 1979, due to reactor fuel overheating, leading to an estimated collective dose of 40 person.Sv, and
• Chernobyl in the Soviet Union in 1986, due to a reactor explosion and fire, leading to an estimated collective dose of 600,000 person.Sv (UNSCEAR, 1993).

It can be seen that the emissions of the Chernobyl accident dwarf all the others. For that reason, attention is usually focused on the most severe accident scenario - "the maximum credible accident" - rather than other, more probable but non-catastrophic, events.

Severe accidents can release more than 10% of the radioactive inventory of a nuclear reactor. Under these conditions, the total radioactivity released to atmosphere is typically greater than 10 EBq - about ten million times more than annual routine emissions (CEC, 1995d). The potential consequences of such accidents have been modelled. The potential collective dose from a severe reactor accident in the UK is comparable with that of Chernobyl (NRPB, 1988a) - i.e. tens of thousands of fatal cancers. The consequences of the "maximum credible accident" are therefore very severe.

By UK standards, the Chernobyl reactor was poorly designed and operated. It therefore provides limited information on accident probability in current UK conditions. Estimating the probability of a severe accident is difficult (see Box 1)

The risks of nuclear power accidents remain controversial and ideally some international consensus on evaluation approaches should be developed. However, given the paradigms of the proponents of different approaches, it is probably an irresoluble problem (Lee, 1997). As new nuclear power stations are not likely in the UK for foreseeable future, it may not be the highest priority issue for the Royal Commission.

2.2.3 Nuclear Wastes

Nuclear waste is classified into 3 categories: high level waste (HLW), intermediate level waste (ILW) and low level waste (LLW). LLW is waste which does not require any special precautions in handling. Over 1 million m³ has already been disposed of to landfill at Drigg in Cumbria. Current stocks are only 8,000 m³ (Nirex, 1996).

ILW requires more care in handling and requires "conditioning" by immobilisation in a cement matrix. Reprocessing is the major source of ILW and total output of conditioned waste from expected operations is 60,000m³ (Nirex, 1996). In April 1997, the former Secretary of State for the Environment upheld the decision of Cumbria County Council to refuse planning permission to Nirex for a rock characterisation facility (RCF) in West Cumbria to investigate the site's suitability for ILW disposal. Both Nirex and the Government are considering future plans, but, in the interim, there is no strategy for ILW disposal.

HLW is spent nuclear fuel and/or its reprocessing products. The latter is a solution of nitric acid which is cooled then vitrified for storage. HLW is produced in much smaller volumes than other categories. Current UK stocks are 1,600m³, largely of liquid waste, and final arisings of vitrified waste are estimated to be 2,300m³ (Nirex, 1996). HLW contains very high levels of radioactivity, and therefore poses major handling and storage problems. Final storage or disposal is problematic, because of the very long lifetimes of some actinide components. UK policy is to allow vitrified waste to cool for 50 years before placement in a final repository, which is yet to be designed.

The options for nuclear waste disposal have been explored extensively (e.g. Royal Society, 1994; DOE, 1995; POST, 1997). However, the decision on the RCF leaves the UK with no agreed final disposal policy for either ILW or HLW. This major deficiency of nuclear policy, identified by the Royal Commission more than 20 years ago (Royal Commission, 1976), has therefore not been rectified. The problems are complex - scientific, economic, social and institutional. In a sense it is not a major issue for energy policy, as new nuclear power stations are not planned. However, it is a major environmental problem. The Royal Commission may wish to investigate the policy vacuum on nuclear waste storage and disposal.

2.2.4 Nuclear Decommissioning

Decommissioning of nuclear reactors is problematic because of the large inventories of radioactive materials. There is successful experience in the UK and elsewhere, but largely confined to smaller experimental reactors. However, three Magnox reactors have now closed, so decommissioning is an increasingly urgent issue.

The proposed approach to decommissioning in the UK is a three stage process:

• Stage 1- removal of nuclear fuel to reprocessing and/or storage,
• Stage 2 - dismantling of plant outside the reactor containment, and
• Stage 3 - dismantling of the reactor.

The first two stages can be undertaken within a few years of closure. Current policy envisages that the final stage would be delayed for 100 years to allow a reduction in radioactivity, although earlier site clearance is possible. British Energy now proposes a "deferred safestore" strategy, in which Stage 2 is limited to non-radioactive buildings, with final dismantling delayed for 135 years. This has cost advantages, because deferred expenditure is discounted.

The final stage of decommissioning will generate large additional quantities of nuclear waste, including ILW. It is essential that satisfactory provision for waste is made before significant decommissioning activities proceed (ETSU, 1994).

British Energy has a segregated fund designed to fund post-closure liabilities, but Magnox Electric does not. Whether the current level of payments into the segregated fund is sufficient is an extremely complex and controversial question (MacKerron and Sadnicki, 1995). Whatever the financing arrangements, some interpretations of sustainable development would imply a preference for early site clearance rather than leaving decommissioning to future generations. The Royal Commission may wish to consider best option for nuclear decommissioning in the light of conflicting economic and environmental benefits.

2.2.5 Other Emissions

A full life cycle analysis of the nuclear fuel cycle identifies operations in which fossil fuels are used, and therefore from which carbon dioxide, sulphur dioxide and NO_x emissions occur. Energy use in mining and milling is significant, but occurs outside the UK. In principle it could be extremely large if very low grade ore were used and this constrains the grades of ore which are useful energy resources (Mortimer, 1980). Uranium enrichment can also be energy intensive if the gaseous diffusion process is used, but UK nuclear fuel production is by centrifuge enrichment (ETSU, 1994a) and uses 0.2% of the final output of UK nuclear power stations (DUKES, 1997). Fossil fuels are also used in the significant quantities in the manufacture of construction materials used in the nuclear fuel cycle, in particular steel and concrete.

For grades of ore currently used in the UK and elsewhere, detailed life cycle analysis indicates that the carbon dioxide emissions are about 4% of those of a coal fuel cycle with similar electricity output (Mortimer, 1991; Yasukawa et al, 1992). Greenhouse gas emissions from nuclear fuel cycles are therefore not a major issue.

2.3 Renewable Energy

There is a large number of renewable sources of energy. The ones considered here are those judged potentially important in the UK in the medium term (Jackson, 1998). They do not produce radioactive materials or, with the exception of biomass combustion, air pollution. Renewable energy sources are therefore generally considered more compatible with environmental protection and their extensive use is advocated as an important contribution to sustainable development in influential international reports (e.g. WCED, 1987; Watson et al, 1996). However, renewable energy source are not free of environmental impact. Developments often need to be located at sites where the resource is plentiful, which can be in environmentally sensitive areas, where changes to landscapes are an issue.

2.3.1 Direct Emissions

Direct emissions only result from combustion of biofuels. In recent years, many countries have considered the use of biofuels as substitutes for oil based liquid fuels, e.g. there is large scale bioethanol production in Brazil (Goldemberg et al, 1993) and research programmes continue in many countries (WEC, 1993). In the UK, the use of biofuels for heat and power is more economically attractive than conversion to liquid fuels (ETSU, 1994b). There is a range of potential fuels. Municipal solid waste (MSW) is already used, but not as extensively as in many other developed countries, where there have been more restrictive policies on the disposal of organic wastes to landfill. Forestry and agricultural wastes, such as cereal straw and chicken litter, are also now being used in the UK. In addition, coppicing of fast growing hard woods is beginning, with willow and poplar the most attractive species.

Biomass combustion emits carbon dioxide, but, provided the source is sustainably harvested, the same amount is fixed in photosynthesis. Any net flow of carbon to the atmosphere is generally accounted for as resulting from land use changes not combustion. There are only net emissions due to combustion where mixed wastes contain fossil carbon. For MSW about 20% of the carbon content is petroleum-based plastics (CSERGE, 1993). Its combustion for power generation, even at the low efficiency of most waste to energy plants, produces approximately 400 g/kWh (Bates, 1995), slightly less than a modern gas-fired plant. In most cases, incineration displaces landfill, thereby reducing methane emissions. The net effect of MSW combustion on UK greenhouse gas emissions is therefore negative (Bates, 1995), even neglecting the displaced emissions of fossil fuel power generation.

The sulphur content of most biofuels is low, and therefore sulphur dioxide emissions are small compared to coal and heavy oil. Emissions factors are typically 0.3-3 g/kWh for power generation (ETSU, 1994b).

The emissions of most concern from biofuels are NO_x and particulates. NO_x emissions are limited as flame temperatures are lower than for fossil fuels. Fuel ash contents are generally lower than for other solid fuels, so uncontrolled particulate emissions are not high, but final emissions depend most on abatement technology. In essence, there are conventional technologies in which the fuel is combusted for direct heat or steam generation, and more recent, cleaner, high efficiency technologies based on gasification and combined cycle power generation. There is a recent comprehensive analysis of emissions from biomass fuels used for heating (Kaltschmitt and Reinhardt, 1997). Emissions of NO_x are 0.2-0.8 g/kWh; emissions of particulates are 0.01-0.1 g/kWh. There is less reliable data on emissions from power generation, but European funded research work is ongoing to improve this (Groscurth, 1998). Table 5 lists emissions estimates from a range of biomass power generation plants.

Table 5
Typical Emissions from Biomass Power Generation

Particulate emissions from older technologies can be quite high, depending upon the fuel, but are reduced by good flue gas clean-up, or to low levels by gasification. NO_x emissions are of the same order of magnitude as for fossil fuels, but lower for new technologies. The relatively high figure for MSW combustion will be reduced, if proposals from the European Commission for an emission standard of 100 mg/m³ are adopted. In general, small biomass power plants have less restrictive emission standards than larger fossil fuel plants, but abatement of emissions of NO_x and particulates to similar levels is possible without excessive costs.

One other emissions issue is controversial in the case of MSW. Incomplete combustion of chlorinated polymers can lead to highly toxic combustion products - dioxins and furans. Current UK regulations require emissions to be less than 1 ng/m³ (equivalent to 10ng/kWh), which is more than 100 times lower than some older incinerators (Woodfield, 1987). Emissions can be reduced to extremely low levels in modern incinerators, and the European Commission has proposed a limit of 0.1 ng/m³. Despite the ongoing controversy, there seems little basis, on pollution grounds, for the Royal Commission to revise its view that, in the context of a policy to increase recycling and with strong pollution regulation, incineration is preferable to landfill (Royal Commission, 1993).

In principle, methane emissions can result from anaerobic combustion of vegetation flooded in large hydropower projects. This might be significant where large areas of forest are flooded (Moreira and Poole, 1994), but not for small projects of the type likely in the UK.

2.3.2 Indirect Emissions

Energy use in the construction of renewable energy power plants is generally small compared to the output (Mortimer, 1991). For hydropower plants, carbon dioxide emissions from production of construction materials are estimated to be approximately 1% of those of coal-fired generation (Meridian, 1989; Moreira and Poole). For wind power emissions are also about 1% of those of coal (CEC, 1995e). Only for photovoltaics (PV) do the manufacturing inputs appear significant. Carbon dioxide emissions from constructing a building mounted PV system are about 80 g/kWh (Oliver, 1997), i.e. 10% of those from a coal fired plant. In general, for any economic technology, the energy inputs are less than 10% of the total cost and therefore their carbon dioxide (and other energy related) emissions are small compared to those from fossil fuel cycles. In this field, detailed life cycle analysis has little added value.

More detailed analysis is required for biomass. For biodiesel from oil seed crops, emissions due to fertiliser manufacture, agricultural fuel use and processing are estimated to be 45% of those of conventional diesel combustion (Culshaw and Butler, 1992). However, solid biofuels have higher yields, lower inputs and little processing. Life cycle energy use is estimated to be 2-8% of energy output for heating fuels (Herendeen and Brown, 1987).

Emissions from biomass transportation are potentially significant. The fuel is less dense than fossil fuels, and therefore not only energy use but also other costs are higher per kilometre of transportation. In general costs will limit biomass transport to distances of 50km (ETSU, 1994b), so that transportation emissions will be relatively small.

2.3.3 Other Environmental Burdens

Because renewable energy technologies use extensive sources compared to the concentrated resources of fossil and nuclear fuels, the land use implications can be significant. Hydropower can be responsible for flooding large areas, affecting settlements, agriculture and ecosystems (Sims, 1993; Moreira and Poole, 1993). In the UK this type of project is now very unlikely. Attention is concentrated on small-scale "run-of-river" hydropower (ETSU, 1994b; Jackson, 1998) for which there is minimal inundation and therefore land use implications are negligible.

For small hydropower projects, the main potential impacts are on freshwater ecology. Provided that diversion leaves sufficient water in the original river course, most species will not be severely affected. Fish mortality in the turbines is potentially a problem (WEC, 1993), but can be controlled by simple physical measures. Migrating fish species are potentially the most affected, but impacts can be limited using linked pools with low barriers - "fish ladders" (ETSU, 1994b).

Wind farms extend over a relatively large area per unit of energy generated, but the land use implications are usually limited. Although a generating capacity of 1000 MW might extend over 100 km², only 0.1% of the surface is physically occupied (CEC, 1995e). Normal agricultural activities can continue on the remainder. The main land use constraint is that noise impacts would generally preclude settlement within 200-300m. Wind power may not be acceptable in some designated landscapes, but that is a constraint on its development rather than a land use impact see (Section 3.4.1).

The potential land requirements of PV cells are substantial, if large scale centralised generation is envisaged. A generating capacity of 1000 MWp requires about 7-8 km² of modules (Hill, 1993). In practice, the land use implication in the UK will not be large. The most cost effective approach to PV generation is to use roofs and walls (Taylor, 1990; ETSU, 1994b).

Photovoltaics generate specialised pollution burdens from the use of toxic materials in semiconductor manufacturing. Whilst silicon itself poses no major toxicity problem, the manufacturing process uses trichlorosilane, boron trichloride and phosphorus oxychloride. Higher efficiency PV may use other materials, notably cadmium telluride, copper indium diselenide and gallium arsenide. Many of these materials are toxic and arsenic, selenium, tellurium and copper have very low concentration limits in water (ENDS, 1992). Cadmium emissions in the refining process may be a particular problem (Taylor, 1990; Oliver, 1997). Life cycle emissions are estimated to be 6g/kWh (Oliver, 1997). This is the same order of magnitude as from a modern coal plant (CEC, 1995b). Disposal of waste from arrays containing these elements would be problematic, but environmental controls may ensure materials recycling.

Production of energy crops may involve use of fertilisers and pesticides with pollution to ground and surface waters. For oil crops, such as rape, inputs are similar to other arable crops. But for short rotation coppice (e.g. willow and poplar), fertiliser use is typically 20% of that for cereals. Energy crops may therefore be advantageous in regions like East Anglia where nitrate pollution is a particular problem (ETSU, 1994b).

3. Environmental Impacts

The emissions and burdens of energy supply and use described in Section 2 have a range of impacts on human health and the environment. These are considered in this section.

3.1 Impacts of Climate Change

The science of climate change has been reviewed by the IPCC (Houghton et al, 1996). Some of the infra-red radiation emitted by the surface of the Earth is absorbed by greenhouse gases in the atmosphere. A fraction is re-emitted downwards, thereby warming the atmosphere and the surface. The basic physics is straightforward, but there are significant uncertainties in the size of the effect due to feedback mechanisms at work in the atmosphere. Most importantly, the effects of warming on cloud cover, altitude and structure are complex. As clouds contribute to both warming (by absorption) and cooling (by reflection) effects, the overall change is difficult to calculate. This feedback contributes most of the uncertainty in the range of 1.5 to 4.5 K estimated to be the equilibrium warming at greenhouse gas concentrations equivalent to double the pre-industrial carbon dioxide level (Houghton et al, 1996).

Uncertainties in regional climate change are larger. These are calculated using complex models of the atmosphere, General Circulation Models (GCMs). The spatial resolution of the models and the description of atmospheric-ocean interactions remain inadequate to predict regional climate change with more than low confidence. Research is required to define changes in patterns of precipitation with the resolution required for reliable impact prediction.

A secondary effect of climate change is rising sea levels. The overall change in sea level will be due to a combination of thermal expansion and any changes in the mass of ice on land (the polar ice caps and alpine glaciers). The latter will be affected by both temperature and precipitation changes. Projections of sea level rise vary. IPCC reports a range of 0.15-0.95m, with a best estimate of 0.5m by 2100 (Houghton et al, 1996).

The impacts of climate change have also been reviewed by the IPCC (Watson et al, 1996). Impacts on the UK have been considered by the UK Climate Change Impacts Review group (CCIRG, 1996). The following sections briefly report the principal conclusions of these major reviews, along with those of important subsequent work. Impacts are most commonly reported with reference to a greenhouse gas concentration equal to double the pre-industrial level of carbon dioxide (2xCO₂). The effect on impacts of other levels of climate change and the rate of climate change needs further research. All the impacts are global in range and persist long into the future. Although some of the key features of likely impacts are known, the uncertainties in quantitative analyses are typically high.

3.1.1 Impacts on Human Health

Climate change will have impacts on human health both directly due to the effects of temperature and via other changes caused by climate change (McMichael, 1996; McMichael et al, 1996). A listing in given in Table 6.

Table 6
Climate Change Effects on Human Health

Heat stress and cold stress mortality impacts will be influenced in opposite directions. There is a growing literature on the climate change impacts of heat waves (e.g. Kalkstein and Tan, 1995). There has so far been less attention to the beneficial effects of warming on cold stress, except for a UK study (Langford and Bentham, 1993). Provisional estimates of global effects indicate that the two changes may be of roughly equal magnitude, but there are large uncertainties (ExternE, 1998a). Even for the UK, it is not known if the overall effect will be positive or negative (CCIRG, 1996). Further research is required. Work in this field is ongoing, but the extension to chronic impacts will be difficult (see Section 3.2.1) and therefore reliable results will not be achieved quickly.

The area amenable to parasitic and vector borne diseases will expand and impacts could be large. Malaria is the largest killer and best studied (Martens et al, 1995; Martin and Lefebvre, 1995; Matsuoka and Kai, 1995). Up to 80 million additional cases a year are possible due to climate change by 2100. The mortality rate is currently about 0.4%, but most fatalities are children (McMichael, 1996). Climate projections indicate that the UK will become susceptible to malaria, but public health measures should prevent serious problems. Other infective agents causing diarrhoeal and dysenteric infections are likely to be more problematic (CCIRG, 1996).

Impacts of extreme weather and sea level rise on health are considered in the relevant sections below. Conditions suitable for the formation of high concentrations of ozone are likely to be increased (Edgerton, 1991), and the implications for mortality may be of the same magnitude as those of direct temperature impacts (ExternE, 1998a). Other impacts are likely to be smaller.

The biggest, but very uncertain, potential impact is the effect of regional disasters (such as famine, drought or major inundation) on vulnerable communities in the developing world. This is considered in Box 2. Implications for policy may be profound. Further research is needed and the Royal Commission may wish to consider the failure of most economic analyses of climate change to take account of the most potentially severe impacts.

3.1.2 Impacts on Agriculture

Green plants will tend to benefit from the direct effects of higher levels of carbon dioxide on photosynthesis. However, temperature and precipitation constraints delineate areas suitable for cultivation. Soil moisture in the growing season is often a critical parameter, and this will be affected in a complex way by changes in temperature and rainfall. Pests and weed risks are likely to be higher in a warmer wetter climate (Reilly, 1996).

Optimum crops and practices will change in most places, and selective breeding can improve features such as drought resistance, so that impacts will depend on the extent, speed and quality of adaptation. Overall yield changes will be sensitive to changes in climate and will vary considerably by region. When carbon dioxide direct effects are included, many crops show net benefits from climate change in temperate regions. In the tropics, the effects are more likely to be negative.

Major problems exist in scaling up from regional assessments. Effects on global production are therefore uncertain and need further research. Effects of adaptation have been shown to be critical in determining the sign of global production (Fischer et al, 1996; Rosenzweig and Parry, 1994). Lower income countries are more vulnerable and have lower adaptive capacity, and therefore are likely to be more negatively affected. National production levels might be affected by ± 20%, although global changes of only ± 10% are expected (Reilly, 1996).

The link from changes in production to economic effects can be addressed through agricultural trade models. There have been various global studies (e.g. Kane et al, 1992; Reilly et al, 1996; Rosenzweig and Iglesias, 1994). However, they have some serious constraints when applied to the long term evolution of agriculture. Much depends on assumptions about consumer demand, investment and agricultural technology. Very little quantitative work exists on relationships between climate change and famine, desertification and migration from semi-arid regions, but these impacts could be the most significant (see Box 2).

UK impacts are likely to vary across the country. Arable production in the south and east may be damaged by reduced water availability in summer. Grassland and forest productivity in the north and west is likely to improve. Crop patterns will change in all areas (CCIRG, 1996).

3.1.3 Impacts on Water Supply

The hydrological links between climate change and water resources can be modelled. However, since most of humanity's water resources are for irrigation, water use efficiency of crops is also an important factor. Higher carbon dioxide levels will tend to improve this. Water basin studies show that changes in run-off, and therefore water resources, are likely to increase in some regions and decrease in others. Changes in some major river basins, notably the Nile, might be very large (Riebsame et al, 1995).

Changes in the availability of water for drinking and agriculture are potentially serious in regions where groundwater levels fall. Water systems are designed to supply reliable yields. A small shift in risk could imply a costly redesign of storage and delivery systems. Large infrastructure development may be required to redistribute water resources. In OECD countries, the main effect would be on water costs; in developing countries with capital scarcity, more water shortages are likely. Demand management and technology improvement could alleviate problems in some cases. Overall, most water supply systems are expected to be increasingly stressed (Kaczmarek, 1996).

The effect of sea level rise on coastal aquifers and water use in coastal basins has not been included in either global sea level rise studies or global water resource assessments. This is potentially a serious deficiency, requiring further research.

The only global modelling estimate of water resources (Darwin et al, 1995) considered water resources in the context of their impact on irrigated agriculture. Water resources for four climate scenarios were found to increase for the world as a whole (by 6-12%), but shortages were modelled in some regions. Global data on water supply, use, and prices are sparse. The links from changes in water resources to economic and social impacts needs further research. Migration and conflict arising from drought are possible outcomes (see Box 2).

Impacts in the UK may be beneficial in the north, but are likely to be adverse in the south, where climate change is expected to increase demand and reduce summer rainfall (CCIRG, 1996). The uncertainty is, however, high. A recent assessment showed increases in evapo-transpiration are less if all relevant climate variables are considered than temperature alone (Arnell et al, 1997).

3.1.4 Impacts of Sea Level Rise

Low lying coasts and islands face various threats from sea level rise - inundation, flooding, erosion and saline intrusion. Protection by sea defences is technically feasible and likely to be used to protect populated areas in developed countries. But for developing countries with large deltaic areas, e.g. Bangladesh, Egypt, China and Nigeria, and small low-lying islands protection may be very costly (Bijlsma, 1996; Nichols and Leatherman, 1995). The impacts of sea level rise are therefore conventionally divided into three types:

• capital costs of protection,
• loss of dry land, and
• loss of wetland.

The relative importance depends on decisions about what to protect.

There are many national case studies, but no detailed global study of all impacts (Bijlsma, 1996). Extrapolating national studies, it is estimated that, using existing technology, world-wide costs of protecting settled areas amount to several hundred million US dollars annually (Pearce et al, 1996). Numbers of people likely to be at risk of flooding are likely to double or treble over the next century, to around 100 million. Risks will also depend on changes to extreme weather events (see below). In most countries, only a fraction of a percent of the total land is at risk, but this rises to a high fraction in Bangladesh and some Pacific island states (Bijlsma, 1996). Migration may well be one of the most pronounced impacts of sea level rise, with potentially severe social and economic implications (see Box 2). Saline contamination of fresh water supplies is also a threat, but it has not been studied in any detail.

Coastal wetlands are also threatened in some areas. Tidal marshes may have the capacity to migrate, provided sedimentation is maintained, but are likely to be constrained from doing so where dry land protection is a priority. For saline forest ecosystems, such as mangroves, the impacts remain uncertain - some studies indicate resilience to sea level rise at expected rates, others project large reductions in ecosystem area. The implications for populations dependent upon the resources of these areas are therefore uncertain.

Impacts of sea level on the UK will be exacerbated by the falling land level in the south and east, but offset by rises in the north and west. There is a high concentration of population and economic activity in coastal zones, so protection is generally economic. However, managed retreat may be the best option in some areas where there is rapid coastal erosion. Impacts are only likely to be severe if sea defences fail (CCIRG, 1996).

3.1.5 Impacts of Extreme Weather

For many impacts of climate change it is weather (particularly extreme conditions) which is the principal driver, not average climatic conditions. Unfortunately, the frequency, distribution and severity of extreme conditions is not well understood (Houghton et al, 1996) and needs further research. The capacity of society to deal with extreme weather is also very varied. Cyclones and storm surges in Bangladesh in the 1970s killed hundreds of thousands of people; hurricanes in North America rarely kill more than a hundred. Future vulnerability is therefore dependent on social development and not known with any accuracy.

Cold spells are likely to decrease in number and severity (Houghton et al, 1996). This has potentially important health benefits (see above) and other modest advantages, e.g. fewer frozen pipes, less road ice and crop loss. Heat waves are likely to increase in frequency and severity. Apart from health effects, other impacts are relatively small. There may be some crop loss, but most problems associated with heat waves are due to drought.

Drought is already a severe problem in some areas of the world, notably the Sahel region of Africa. Climate change will affect different regions very differently, but the number of severe droughts will probably increase (Houghton et al, 1996). Impacts are likely to be largest in those developing countries where water shortage is already chronic. Very few people die of dehydration, but drought can be linked with food shortage, with potentially severe results if there are not appropriate response measures. In Europe, drought risks are likely to decrease in Northern Europe but increase in Southern Europe (Brignall et al, 1996).

Extreme wind speeds cannot be modelled with any certainty and different models generate contradictory results (Houghton et al, 1996). Average wind speeds are likely to increase, at least in the UK (CCIRG, 1996; Palutikof and Downing, 1994). There has been an increase in the number of very damaging wind storms in Europe in recent years and existing studies tend to a assume a correlation between average wind speeds and wind storm impacts. In mid-latitude countries, few lives are lost. The most obvious others impacts are damage to buildings and mature vegetation. A single storm can cause many billions of dollars worth of damage (Dorland et al, 1995). If windstorms became sufficiently frequent, building standards may need to change.

The impacts of climate change on the frequency and severity of tropical cyclones is also unknown. Sea surface temperature is an important factor, and is projected to increase. But, the synoptic causes of tropical storms are complex (Kattenberg et al, 1996). The potential consequences of cyclones are larger than all of the other weather hazards - thousands of lives lost in developing countries and billions of dollars of economic damage.

Impacts in the UK may not be too severe. On the basis of recent trends, damage to buildings via drought-induced clay shrinkage and wind storms may be the most important (CCIRG, 1996).

3.1.6 Impacts on Ecosystems

Impacts of climate change on natural ecosystems have not been researched comprehensively. Most attention has been given to modelling studies, although there are now some reliable observations of effects in the field (Briggs, 1997).

At a global scale, climate change will shift zones suitable for various classes of ecosystem - generally towards higher latitudes. At 2xCO₂ warming, nearly half of the land surface will change eco-climatic zone (Smith et al, 1995). And areas suitable for ecosystems currently found at high latitudes, e.g. boreal forests, will contract by more than 30% (Kirschbaum and Fischlin, 1996). The rate of change of climate may be at least as important as the level. In mid-latitudes, eco-climatic zones are moving at 5-8 km/year (CCIRG, 1996). This exceeds the rate at which some habitats can migrate (Kirschbaum and Fischlin, 1996).

Forests, in particular, can only adapt slowly. Although net primary productivity may increase with higher carbon dioxide levels and temperatures, this may not be reflected in forest mass. Some tropical and temperate forests may not be severely impacted, except where soil moisture levels fall. Boreal forests will be more affected as grassland invades from lower latitudes more rapidly than northern advance into the tundra. Changes between forest types may create new types of habitat, with uncertain impacts (Kirschbaum and Fischlin, 1996).

Range lands may be less affected than forests by the rate of climate change and are likely to expand in some regions. The tundra is an exception - it is likely to contract and to suffer reduced species richness. In lower latitudes, forage quality may decline due to nitrogen imbalances (Allen-Diaz et al, 1996). Deserts are likely to expand in area as most models project higher temperatures, without increased rainfall. Changes in frequency and intensity of rainfall may allow some invasion of species from other systems (Noble and Gitay, 1996).

Ecosystems which cannot easily migrate are potentially at risk from climate change: notably island, polar and montane systems. In mountainous areas, some ecosystems may be lost, for example alpine species currently confined to peaks (Beniston and Fox, 1996). Similar considerations apply in island systems, but many are also at risk from sea level rise (Bijlsma et al, 1996). The overall effect of climate change on biodiversity is poorly understood, and will depend on other pressures. Significant losses of species due to climate change are expected, and may be the most important impact of climate change. The impacts of climate change on biodiversity loss and its implications for human societies require further research.

Impacts on the UK are understood perhaps as well as anywhere in the world. Land use changes are likely to remain more important than those of climate change. Some montane and northern species may be lost, but new species will migrate from continental Europe. Overall a reduction in plant species and increase in animal species is expected (CCIRG, 1996).

3.1.7 Other Impacts

Patterns of energy demand will change as temperatures rise. The main impacts will be that demand for space heating will be reduced, and energy use in space cooling and refrigeration will increase. Some early influential studies concluded that overall impacts would increase energy use (e.g. Smith and Tirpak, 1989). Recent assessments indicate that the balance is uncertain, with the uptake and efficiency of air conditioning a critical factor (IEAGHG, 1994; Moreno and Skea, 1996; ExternE, 1998a). In the UK, energy demand is likely to be reduced (CCIRG, 1996). Other impacts in the energy sector will include changes to hydropower production. These could be positive or negative, depending on regional precipitation, and involve seasonal changes as ice melt starts earlier in spring (Moreno and Skea, 1996).

Tourism is likely to be affected in some fairly obvious ways. Marginal locations for winter sports will become unviable. Beaches and coastal facilities may become increasingly difficult and expensive to protect against sea level rise and storms. Holidays at higher latitudes may be more attractive as the climate warms. Net changes for the UK tourist sector may be positive (CCIRG, 1996).

There is increasing attention to the implications of climate change for financial services, particularly insurance. There have been increased claims for weather related disasters in recent years. Property insurance costs are likely to continue to rise, but the overall effects on the finance sector are unclear (Dlugolecki, 1996) and need further research. The UK has a large insurance industry and therefore is very open to these impacts (CCIRG, 1996).

Overall, the impacts of climate change will be complex and potentially very serious, but still poorly understood. UK researchers are already making major contributions in many fields. The Government has recently established a centre to co-ordinate work on climate change impacts, but the effort remains limited compared to that on climate science. Given the likely international focus on impacts, the Royal Commission may wish to consider the scale of research on impacts within the overall climate change programme.

Very large reductions in emissions of greenhouse gases, especially carbon dioxide, are required if atmospheric concentrations are to be stabilised (Houghton et al, 1996). Although the UK is widely acknowledged to have taken a leadership role with its commitment to reduce emissions by 20% by 2010, it is clear from the science that long term reductions will certainly need to be much larger. This is now widely recognised to be the biggest challenge facing environmental policy in the developed world.

3.2 Impacts of Acidity, Ozone and Particulates

Both sulphur dioxide and NO₂ undergo oxidation in the atmosphere to form strong acids in both the gas phase and as aerosols. The oxidation chemistry is complex and dependent on atmospheric moisture levels. Both the gas species and aerosols reach the ground via wet and dry deposition. Dry deposition velocities are low, and therefore pollutants can travel long distances. Ammonia contributes to aerosol formation, and therefore a high proportion of oxidised sulphur and nitrogen is in particulate form. A number of different models of long range acid pollution have been developed. The process are reviewed in detail elsewhere (e.g., RGAR, 1997).

The influence of emissions on the concentrations of pollutants and distribution of deposition of sulphur, nitrogen and acidity can be calculated with reasonable accuracy at a European scale. There is agreement between modelled and measured data to within a factor of two (Barrett and Berge, 1996).

NO_x also plays an important role in the chemistry of tropospheric ozone. Its concentrations are very variable. Ozone is a natural component of the troposphere, due largely to periodic invasions of ozone rich air across the tropopause. However, the average tropospheric concentration has increased substantially as a result of human activities. The reaction which creates ozone is the photolytic destruction of NO₂. At equilibrium, this is counterbalanced by the reduction of ozone by NO. In the absence of light, the latter reaction is dominant, so ozone levels are reduced to low levels at night. In the daytime, ozone concentrations are determined by the ratio of NO₂ to NO. This ratio is significantly increased by VOC, which, through a series of complex reactions, oxidise NO to NO₂. High levels of ozone at ground level are therefore observed in stable atmospheric conditions in summer in the presence of high levels of both NO_x and VOC (PORG, 1993; Simpson and Malik, 1996).

Fine particulate pollution, acid deposition and ozone pollution are usually conceived of and modelled as regional scale problems. But there may be some effects at the global level. The range of aerosols depends upon the particle size and some smaller particulates may therefore have a very long range (CEC, 1995b). Similarly, ozone which escapes the polluted boundary layer into the free troposphere may have a global range (Hough, 1989) and therefore aggregate impacts may be significant even outside Europe (Rabl and Eyre, 1998). As global pollution issues attract more attention these issues may need further research.

Acid species, particulates and ozone have impacts on many of the same receptors. Their impacts are therefore considered together in the following sections.

3.2.1 Impacts on Human Health

A cocktail of pollutants have been associated with cardio-respiratory illness - particulates, sulphur dioxide, ozone and nitrogen dioxide. Impacts identified in epidemiological studies range from minor symptoms to mortality.

It has been accepted for several years that the clearest relationship with mortality effects is with particulates (e.g. CEC, 1995a; COMEAP, 1995; QUARG, 1996; COMEAP, 1998). The Expert Panel on Air Quality Standards (EPAQS) recommends a 24-hour mean standard for PM₁₀ of 50µg/m³. This is exceeded frequently at most UK urban monitoring sites (UKNAQS, 1997), although annual averages are all below 30 µg/m³ (DE