6 Results - Benefits
6.1 Pollutant exposures
Table 15 highlights change in exposure to photo-oxidants and PM10 (in the form of aerosols generated chemically in the atmosphere following emission of SO2, NOx and NH3). The change in PM10 is expressed as a percentage against total PM10 (i.e. including primary particulates as well as the secondary aerosols) across the UK.
These results are particularly important to UK policy because imports of both classes of pollutant (and their precursors) create difficulty in meeting objectives laid down in the NAQS. The third column of the table includes benefits from UK abatement as well as the other UNECE Member States. Accordingly, the difference between columns 2 and 3 represents the contribution to UK benefits from abatement in countries outside of the UK.
AOT40 and AOT60 provide information on the extent to which peak ozone concentrations are reduced. For example, AOT60 is calculated by subtracting 60ppb from hourly ozone concentrations over the year. Results, where positive (i.e. for which measured or modelled ozone is in excess of 60 ppb) are then summed over the year to give total exposure in excess of 60 ppb. The combination of time and concentration gives an indication of annual dose in excess of possible thresholds.
AOT40 is of interest because research on crops, particularly wheat, indicates a threshold around 40 ppb. AOT60 is selected as an indicator for health impacts (WHO, 1997) though evidence for a threshold for ozone effects on health is far from conclusive, and is indeed contradicted by the results of a number of epidemiological studies. Taken together, AOT40 and AOT60 provide some indication of progress to meeting the NAQS objective of 50 ppb as a running 8-hour mean. Effects on peak concentrations do not provide good guidance on the consequences for mean exposures, which were not modelled in this study.
The results in Table 15 show a relatively small decline in mean particle exposure, but a significant (>10%) fall in metrics of peak ozone exposure for both scenarios J1 and H1. The importance of the PM10 reductions associated with each scenario is possibly downplayed by expressing them in terms of % of mean exposure levels. A substantial fraction (estimated at around a third to a half) of particle exposure arises from natural and other sources which are largely beyond human control. On this basis, the reductions in particulate exposure shown in the Table should be roughly doubled, to show the effect on the component of concentration that can potentially be controlled. Overall, the benefits to be gained under WGS31c from abatement within the UK are roughly half those of H1 and J1.
Table 15. Fall in PM10 exposures(1) as a consequence of moving to scenarios of increasing emissions abatement. Figures for 1990 and UKREF show predicted total PM10 exposure metrics in 1990 and 2010 respectively, figures for other scenarios show % change compared to UKREF.
|
|
Benefit to UK from UK abatement |
Benefit to UK from UNECE abatement
|
|
Population weighted mean particulate exposure, µg/m3
|
|
1990 |
25.51) |
25.51)
|
|
UKREF1) |
19.11) |
19.11)
|
|
WGS31c |
-0.8% |
-0.9%
|
|
J1 |
-1.8% |
-2.9%
|
|
H1 |
-1.8% |
-2.5%
|
|
|
|
|
|
1990 |
204 |
204
|
|
UKREF |
148 |
148
|
|
WGS31c |
-7% |
-13%
|
|
J1 |
-13% |
-25%
|
|
H1 |
-22% |
-35%
|
|
|
|
|
|
AOT60 million person ppm.hours2)
|
|
1990
|
125 |
125
|
|
UKREF |
75 |
75
|
|
WGS31c |
-5% |
-16%
|
|
J1 |
-12% |
-35%
|
|
H1 |
-20% |
-40%
|
|
|
|
|
Note: nq = not quantified.
1) Baseline particles data based on Stedman et al (1997 and 1999; for 1995 and 2005 respectively) increased by a factor of 1.3 to account for under-estimation of certain species through the use of TEOMs in the monitoring network.
2) Ozone data are taken from estimates generated using the web version of the RAINS model.
6.2 Critical loads exceedence
The concept of critical loads for deposition of acidifying and nutrifying pollutants has developed since the mid-1980s, and forms the basis of much of the analysis carried out for both UNECE and the European Commission in developing emission targets. The critical load defines a threshold for pollutant deposition, beyond which ecological change is likely to occur. In some cases the changes linked to exceedence of critical loads can be dramatic, as in the loss of trout and salmon from acidified freshwater systems. In others the changes might be more subtle. The subject is reviewed in depth in the reports of the Review Group on Acid Rain and INDITE (Impacts of Nitrogen Deposition in Terrestrial Ecosystems), produced for DETR.
Critical loads work for the UK up until 1994 (the date of the Second Sulphur Protocol) was based on a map of 1x1km acidity critical loads for soils. The critical load values were assigned on the basis of the mineralogy of the dominant soil type in each 1km square. At that point in time acid deposition was considered in terms of sulphur deposition only, ignoring any additional acidification from nitrogen deposition or any ameliorating effects of base cation (calcium & magnesium) deposition. Exceedances of the critical load (ie the level of deposition above the critical load) were, therefore, calculated using sulphur deposition only. The exceedance maps generally showed high exceedance values in the north and west of Britain, where critical loads tend to be lower (due to thinner, base-poor soils) and sulphur deposition higher. This sulphur deposition was generally highest in central Britain (the area around the major power station sources) and in the north and west where the amount of rainfall is also high.
The multi-pollutant multi-effect Protocol (signed in 1999) was developed for the control of sulphur and nitrogen pollutant emissions and addressed the problems of acidification and eutrophication. For this work, acidification included the effects of both sulphur and nitrogen deposition, so new calculations had to be performed to give estimates of critical loads and their exceedances. The methods developed and used in the UK are the same as those agreed internationally by the Task Force on Mapping, under the UNECE Convention on Long Range Transboundary Air Pollution (LRTAP). Each country under the Convention calculates critical loads for ecosystems they consider to be sensitive to acidification and/or eutrophication.
In the UK(2) critical loads are calculated for five soil-vegetation ecosystems ie acid grassland, calcareous grassland, heathland, coniferous woodland and deciduous woodland. These critical loads also take into account nitrogen and base cation uptake and other removal processes (ie, nitrogen immobilisation) within the ecosystems. In addition, critical loads are calculated for 1445 lake or headwater streams throughout Great Britain; these are generally high altitude sites with small catchment areas.
To calculate exceedances of these acidity critical loads, non-marine sulphur, nitrogen (both oxidised and reduced) and non-marine base cation deposition are taken into account. Deposition maps for 1992-94 show the highest non-marine sulphur in central England, the mountains of Wales and south-west Scotland. Nitrogen deposition is generally lower across Scotland compared to England and Wales. Non-marine base cation deposition is high across the western half of the UK but with the highest values in the north west of Scotland. Inputs of sulphur and nitrogen deposition acidify, whereas non-marine base cation deposition “buffers” the incoming acidity and can ameliorate the effects of acidification. So, the net incoming acid deposition is the sum of the non-marine sulphur and nitrogen deposition less the amount of non-marine base cation deposition. Figure 3 clearly shows high net acid deposition across much of the UK, but with lower values in the far north of Scotland, the west coast of Scotland and the west of Northern Ireland. When this net deposition is compared with the critical load values, it gives rise to the small exceedance values, or even areas of non-exceedance in these parts of Scotland and Northern Ireland. Similar patterns of exceedance are also obtained when estimates of acid deposition for 2010 are used.
In practice, non-marine base cation deposition is included in the calculation of critical loads, rather than as part of the net acid deposition, because:
- The EMEP model used to provide estimates of sulphur and nitrogen deposition at the European scale, does not provide values for non-marine base cation deposition. The EMEP model deposition data are used in the calculation of critical load exceedances for Europe, and in particular for looking at future emission and deposition scenarios for 2010. Individual countries under the LRTAP Convention use their national estimates of base cation deposition in their calculations of critical loads;
- The HARM model used in the UK to look at emission and deposition scenarios for the future also excludes base cation deposition.
However, the way in which base cation deposition is incorporated in the calculation of critical loads gives the same results as including it in the sum of net acid deposition.
The change in critical loads exceedence through the different scenarios in England, Northern Ireland, Scotland, Wales and the UK as a whole is shown in Table 16 and Table 17. Table 18 and Table 19 show improvements compared to the UKREF scenario. The data shown in these Tables relate to the area over which exceedence is predicted. Further information is given in Appendix 7. Table 16 and Table 17 have different total ecosystem areas for two reasons:
- critical loads for nutrient nitrogen are not calculated for UK freshwater ecosystems;
- the number of grid squares (and therefore ecosystem areas) for which nutrient nitrogen critical loads are available is slightly smaller than the number for which acidity data are available for the UK.
In general ecosystems at high elevation in the UK tend to be at the greatest risk. Given the uneven distribution of ecosystems across the UK, it follows that exceedence of critical loads would not affect equally all species, types of ecosystem, etc. within each group shown in the Tables. Even slight exceedence of critical loads (using % at risk, as reported here) could in theory have a significant and long-term impact, for example affecting the viability of a species.
Figure 3 Acid deposition in the UK, 1992-1994.
6.2.1 England and Wales
The two countries are treated together as the trends shown are very similar. Under the UKREF scenario the most significant exceedence of critical loads for both acidification and eutrophication would appear to affect acid grassland and heaths. Even under the most restrictive (J1) scenario it is estimated that 25% of acidic grasslands (by area) would show exceedence of the critical load for acidification. However, there is appreciable exceedence also for deciduous and coniferous forests and freshwaters.
6.2.2 Northern Ireland
There is only very slight exceedence of the critical load for acidification on acidic grassland in Northern Ireland. No other impacts are expected.
6.2.3 Scotland
Problems of acidification in Scotland are less pronounced than in England and Wales, but greater than in Northern Ireland. The ecosystems most affected are predicted to be acidic grasslands and coniferous forests. Our results indicate that eutrification is unlikely to be a problem in Scotland under any scenario.
6.2.4 United Kingdom
The general patterns seen for England and Wales are repeated for the UK as a whole. They are, however, less pronounced, as a consequence of the limited exceedence of critical loads in Scotland.
In addition to the results presented above, maps of total areas exceeded were prepared for each scenario. Examples are shown in Figure 4. The maps demonstrate the uneven distribution of exceedence noted above, this reflecting variation in both the sensitivity of ecosystems and in deposition. Of particular note are problems for acidic grasslands in Wales, the Pennines and Southern Scotland.
Table 16. Exceedence of critical loads for acidification across the UK under different scenarios. Ecosystem area in hectares, exceedences by % of area.
|
England |
Ecosystem area (ha) |
UKREF |
H1 |
J1
|
|
Acid grassland |
1126445 |
37 |
25 |
23
|
|
Calcareous grassland
|
855030 |
0.0 |
0.0 |
0.0
|
|
Heath
|
139459 |
24 |
4.9 |
3.8
|
|
Coniferous woodland
|
181039 |
6.9 |
6.2 |
4.9
|
|
Deciduous woodland
|
647548 |
5.6 |
2.1 |
1.7
|
|
Freshwaters
|
142860 |
6.6 |
5.5 |
4.8
|
|
All ecosystems
|
3092381 |
17 |
10 |
9.5
|
|
|
|
|
|
|
|
Northern Ireland
|
Ecosystem area (ha) |
UKREF |
H1 |
J1
|
|
Acid grassland |
132145 |
0.20 |
0.10 |
0.05
|
|
Calcareous grassland
|
146958 |
0.00 |
0.00 |
0.00
|
|
Heath
|
31176 |
0.00 |
0.00 |
0.00
|
|
Coniferous woodland
|
50629 |
0.00 |
0.00 |
0.00
|
|
Deciduous woodland
|
26839 |
0.00 |
0.00 |
0.00
|
|
Freshwaters
|
0 |
0.00 |
0.00 |
0.00
|
|
All ecosystems
|
387747 |
0.07 |
0.02 |
0.02
|
|
|
|
|
|
|
|
Scotland
|
Ecosystem area (ha) |
UKREF |
H1 |
J1
|
|
Acid grassland |
3736432 |
6.2 |
2.8 |
2.2
|
|
Calcareous grassland
|
0 |
0.0 |
0.0 |
0.0
|
|
Heath
|
709889 |
0.3 |
0.1 |
0.1
|
|
Coniferous woodland
|
429652 |
4.5 |
3.5 |
3.1
|
|
Deciduous woodland
|
122452 |
1.4 |
0.4 |
0.2
|
|
Freshwaters
|
171155 |
0.3 |
0.2 |
0.2
|
|
All ecosystems
|
5169580 |
5 |
2.4 |
1.8
|
|
|
|
|
|
|
|
Wales
|
Ecosystem area (ha) |
UKREF |
H1 |
J1
|
|
Acid grassland |
462379 |
54 |
29 |
25
|
|
Calcareous grassland
|
14400 |
0.0 |
0.0 |
0.0
|
|
Heath
|
111726 |
22 |
11 |
8.4
|
|
Coniferous woodland
|
76801 |
4.2 |
3.4 |
3.0
|
|
Deciduous woodland
|
236779 |
3.5 |
2.5 |
2.1
|
|
Freshwaters
|
26438 |
2.6 |
1.3 |
1.2
|
|
All ecosystems
|
928523 |
31 |
17 |
14
|
|
|
|
|
|
|
|
United Kingdom
|
Ecosystem area (ha) |
UKREF |
H1 |
J1
|
|
Acid grassland |
5457401 |
16 |
9.5 |
8.4
|
|
Calcareous grassland
|
1016388 |
0.0 |
0.0 |
0.0
|
|
Heath
|
992250 |
6.1 |
2.0 |
1.6
|
|
Coniferous woodland
|
738121 |
4.5 |
3.9 |
3.3
|
|
Deciduous woodland
|
1033618 |
4.1 |
1.9 |
1.6
|
|
Freshwaters
|
340453 |
2.9 |
2.5 |
2.2
|
|
All ecosystems
|
9578231 |
11 |
6.2 |
5.4
|
Table 17. Exceedence of critical loads for eutrophication across the UK under different scenarios. Ecosystem area in hectares, exceedences by % of area.
|
England |
Ecosystem area (ha) |
UKREF |
H1 |
J1
|
|
Acid grassland |
1126445 |
9.0 |
4.5 |
4.2
|
|
Calcareous grassland
|
855030 |
0.0 |
0.0 |
0.0
|
|
Heath
|
139459 |
11 |
5.1 |
4.9
|
|
Coniferous woodland
|
181039 |
0.6 |
0.1 |
0.1
|
|
Deciduous woodland
|
647548 |
1.3 |
0.4 |
0.4
|
|
All ecosystems
|
2949521 |
4.2 |
2.0 |
2.0
|
|
|
|
|
|
|
|
Northern Ireland
|
Ecosystem area (ha) |
UKREF |
H1 |
J1
|
|
Acid grassland |
132145 |
0 |
0 |
0
|
|
Calcareous grassland
|
146958 |
0 |
0 |
0
|
|
Heath
|
31176 |
0 |
0 |
0
|
|
Coniferous woodland
|
50629 |
0 |
0 |
0
|
|
Deciduous woodland
|
26839 |
0 |
0 |
0
|
|
All ecosystems
|
387747 |
0 |
0 |
0
|
|
|
|
|
|
|
|
Scotland
|
Ecosystem area (ha) |
UKREF |
H1 |
J1
|
|
Acid grassland |
3736432 |
0 |
0 |
0
|
|
Calcareous grassland
|
0 |
0 |
0 |
0
|
|
Heath
|
709889 |
0 |
0 |
0
|
|
Coniferous woodland
|
429652 |
0 |
0 |
0
|
|
Deciduous woodland
|
122452 |
0 |
0 |
0
|
|
All ecosystems
|
4998425 |
0 |
0 |
0
|
|
|
|
|
|
|
|
Wales
|
Ecosystem area (ha) |
UKREF |
H1 |
J1
|
|
Acid grassland |
462379 |
7.7 |
5.5 |
4.7
|
|
Calcareous grassland
|
14400 |
0.0 |
0.0 |
0.0
|
|
Heath
|
111726 |
11 |
9.3 |
7.6
|
|
Coniferous woodland
|
76801 |
2.8 |
1.2 |
1.1
|
|
Deciduous woodland
|
236779 |
3.6 |
1.1 |
0.7
|
|
All ecosystems
|
902085 |
6.5 |
4.4 |
3.6
|
|
|
|
|
|
|
|
United Kingdom
|
Ecosystem area (ha) |
UKREF |
H1 |
J1
|
|
Acid grassland |
5457401 |
2.5 |
1.4 |
1.3
|
|
Calcareous grassland
|
1016388 |
0.0 |
0.0 |
0.0
|
|
Heath
|
992250 |
2.8 |
1.8 |
1.5
|
|
Coniferous woodland
|
738121 |
0.4 |
0.1 |
0.1
|
|
Deciduous woodland
|
1033618 |
1.6 |
0.5 |
0.4
|
|
All ecosystems
|
9237778 |
2.0 |
1.1 |
1.0
|
Table 18. Exceedence of critical loads for acidification across the UK under different scenarios: Improvement compared to the UKREF scenario.
|
|
|
Critical loads exceedence by area |
Reduction in exceedence against UKREF (ha)
|
|
England
|
Total area (ha) |
ha |
% of total |
ha |
ha
|
|
Acid grassland
|
1126445 |
424663 |
38% |
163476 |
140025
|
|
Calcareous grassland
|
855031 |
0 |
0.0% |
0 |
0
|
|
Heath
|
139460 |
33949 |
24% |
28683 |
27132
|
|
Coniferous woodland
|
181039 |
12535 |
6.9% |
3729 |
1374
|
|
Deciduous woodland
|
647548 |
35947 |
5.6% |
25248 |
22044
|
|
Freshwater
|
142861 |
9495 |
6.6% |
2654 |
1599
|
|
All Ecosystems
|
3092383 |
516588 |
17% |
223791 |
192174
|
|
|
|
|
|
|
|
|
Northern Ireland
|
Total area (ha) |
ha |
% of total |
ha |
ha
|
|
Acid grassland |
132145 |
260 |
0.2% |
199 |
183
|
|
Calcareous grassland
|
146958 |
0 |
0.0% |
0 |
0
|
|
Heath
|
31176 |
0 |
0.0% |
0 |
0
|
|
Coniferous woodland
|
50629 |
0 |
0.0% |
0 |
0
|
|
Deciduous woodland
|
26839 |
0 |
0.0% |
0 |
0
|
|
Freshwater
|
0 |
0 |
0.0% |
0 |
0
|
|
All Ecosystems
|
387747 |
260 |
0.1% |
199 |
183
|
|
|
|
|
|
|
|
|
Scotland
|
Total area (ha) |
ha |
% of total |
ha |
ha
|
|
Acid grassland |
3736432 |
232950 |
6.2% |
152343 |
127628
|
|
Calcareous grassland
|
0 |
0 |
0.0% |
0 |
0
|
|
Heath
|
709889 |
2387 |
0.3% |
1836 |
1729
|
|
Coniferous woodland
|
429653 |
19192 |
4.5% |
5874 |
4199
|
|
Deciduous woodland
|
122453 |
1705 |
1.4% |
1513 |
1239
|
|
Freshwater
|
171155 |
453 |
0.3% |
123 |
102
|
|
All Ecosystems
|
5169582 |
256687 |
5.0% |
161688 |
134897
|
|
|
|
|
|
|
|
|
Wales
|
Total area (ha) |
ha |
% of total |
ha |
ha
|
|
Acid grassland |
462380 |
251480 |
54% |
134942 |
115231
|
|
Calcareous grassland
|
14401 |
0 |
0.0% |
0 |
0
|
|
Heath
|
111727 |
24328 |
22% |
14939 |
12502
|
|
Coniferous woodland
|
76802 |
3228 |
4.2% |
914 |
634
|
|
Deciduous woodland
|
236780 |
8340 |
3.5% |
3413 |
2350
|
|
Freshwater
|
26439 |
700 |
2.6% |
391 |
368
|
|
All Ecosystems
|
928528 |
288077 |
31% |
154599 |
131086
|
Table 19. Exceedence of critical loads for eutrophication across the UK under different scenarios: Improvement compared to the UKREF scenario.
|
|
|
Critical loads exceedence by area |
Reduction in exceedence against UKREF (ha)
|
|
England
|
Total area (ha) |
ha |
% of total |
ha |
ha
|
|
Acid grassland
|
1126445 |
101715 |
9.0% |
53803 |
51440
|
|
Calcareous grassland
|
855031 |
0 |
0.0% |
0 |
0
|
|
Heath
|
139460 |
14981 |
11% |
8180 |
7867
|
|
Coniferous woodland
|
181039 |
1020 |
0.6% |
762 |
752
|
|
Deciduous woodland
|
647103 |
8591 |
1.3% |
5963 |
5948
|
|
All Ecosystems
|
2949078 |
126308 |
4.3% |
68708 |
66005
|
|
|
|
|
|
|
|
|
Northern Ireland
|
Total area (ha) |
ha |
% of total |
ha |
ha
|
|
Acid grassland |
132064 |
0 |
0.0% |
0 |
0
|
|
Calcareous grassland
|
146958 |
0 |
0.0% |
0 |
0
|
|
Heath
|
31176 |
0 |
0.0% |
0 |
0
|
|
Coniferous woodland
|
50629 |
0 |
0.0% |
1 |
0
|
|
Deciduous woodland
|
26839 |
0 |
0.0% |
0 |
0
|
|
All Ecosystems
|
387666 |
0 |
0.0% |
0 |
0
|
|
|
|
|
|
|
|
|
Scotland
|
Total area (ha) |
ha |
% of total |
ha |
ha
|
|
Acid grassland |
3736432 |
0 |
0.0% |
0 |
0
|
|
Calcareous grassland
|
0 |
0 |
0.0% |
0 |
0
|
|
Heath
|
709889 |
0 |
0.0% |
0 |
0
|
|
Coniferous woodland
|
429653 |
0 |
0.0% |
0 |
0
|
|
Deciduous woodland
|
122453 |
0 |
0.0% |
0 |
0
|
|
All Ecosystems
|
4998427 |
0 |
0.0% |
0 |
0
|
|
|
|
|
|
|
|
|
Wales
|
Total area (ha) |
ha |
% of total |
ha |
ha
|
|
Acid grassland |
462380 |
35540 |
7.7% |
13749 |
10002
|
|
Calcareous grassland
|
14401 |
0 |
0.0% |
0 |
0
|
|
Heath
|
111727 |
12046 |
11% |
3535 |
1654
|
|
Coniferous woodland
|
76632 |
2181 |
2.8% |
1348 |
1283
|
|
Deciduous woodland
|
236728 |
8584 |
3.6% |
6860 |
5906
|
|
All Ecosystems
|
901867 |
58352 |
6.5% |
25492 |
18845
|
Figure 4. Specimen maps showing the percentage change in area under critical loads exceedence for acidification between the UKREF and J1 scenarios.
(1) The population weighted mean particulate exposure is calculated by summing the product of (population*concentration) over all grid cells for the country, and then divided by total population. AOT40 is calculated by summing concentrations of ozone in excess of 40 ppb over the growing season for plants. Hence exposure to 50 ppb for 8 hours would lead to an AOT40 of (8*(50-40)) = 80 ppb.hours. The accumulation of ppb.hours over the growing season leads to the use of the unit ppm.hours (parts per million.hours). This is multiplied by the area affected to give the units shown in the Table. AOT60 is calculated similarly over the summer months, and multiplied by the number of people exposed to give the units of person.ppm.hours.
(2) Further details on the derivation and calculation of critical loads in the UK can be found in the following report, also on the ITE Monks Wood web site (http://www.nmw.ac.uk/ite/monk/critical_loads/nclmp.html):
Hall, J., Bull, K., Bradley, I., Curtis, C, Freer-Smith, P., Hornung, M., Howard, D., Langan, S., Loveland, P., Reynolds, B., Ullyett, J., Warr. T. 1998. Status of UK Critical Loads and Exceedances January 1998. Part 1 – Critical Loads and Critical Load Maps. Report to Department of Environment, Transport & the Regions. NERC/DETR Contract EPG1/3/116.
6.3 Results for impacts on crops, materials, mortality and respiratory hospital admissions
Results for the set of impacts quantified by the IGCB in its review of the NAQS are shown in Table 20. Results disaggregated by country are shown in Table 21. There are some apparent inconsistencies in the Table with respect to crops and materials damage, comparing, for example, damages in the UK caused by UK and full-UNECE emissions respectively. These arise through the use of different models of pollutant chemistry and dispersion for analysis at different scales, and (in the case of the analysis of impacts on crops) as a result of the balance between positive and negative effects of reducing emissions of SO2 and nitrogenous pollutants. They do not, however, greatly affect the outcome of the comparison of costs and benefits. Results for the health impacts are based on best estimates of exposure-response functions. Results are given in Appendix 8 for the range in functions provided by COMEAP/EAHEAP.
Table 20. Reductions in pollution damages to crops and materials (in terms of monetary benefit), and from acute exposures on premature mortality and respiratory hospital admissions (RHAs) through movement from UKREF to each scenario. Units shown in left hand column.
|
Effect |
Scenario |
UK to UK |
UK to |
UNECE (including UK)
|
|
|
|
|
UNECE |
to UK
|
|
Crop damage |
WGS31c |
6 |
10 |
nq
|
|
(£ million)
|
J1 |
14 |
20 |
10
|
|
|
H1 |
25 |
40 |
17
|
|
|
|
|
|
|
|
Materials damage
|
WGS31c |
2 |
3 |
5
|
|
(£ million)
|
J1 |
4 |
4 |
15
|
|
|
H1 |
4 |
4 |
15
|
|
|
|
|
|
|
|
Premature
|
WGS31c |
160 |
350 |
160
|
|
mortality(cases)
|
J1 |
330 |
700 |
540
|
|
|
H1 |
380 |
880 |
530
|
|
|
|
|
|
|
|
RHAs,
|
WGS31c |
90 |
240 |
110
|
|
additional/
|
J1 |
190 |
530 |
350
|
|
brought forward
|
H1 |
240 |
760
|
377
|
The results for crop damages shown in Table 20 imply a lower level of benefit to crops in the UK from abatement across the UNECE than from abatement in the UK alone. This is a function of the complexity of the assessment of crop damage, which brings together effects that are both damaging to crops (i.e. direct effects of ozone, acidification) and effects that are potentially beneficial (fertilisation with sulphur and nitrogen). For other receptors the issue of potential benefit from increased exposure does not arise.
Results for the health effects are broken down by pollutant (including the most important secondary species) for each region of the UK in Table 22 for the J1 scenario. For effects on crops, materials, and, for health for SO2 and the secondary aerosols, the split by country follows the detailed mapped analysis. The split for ozone effects on health has been made by extrapolation from the results for crops. Benefits are concentrated in England, partly through the population distribution and partly through proximity to mainland Europe. Benefits are spread fairly evenly across the pollutants listed, though ammonium aerosols play a lesser role than the other pollutants.
Table 21. Regional breakdown of reductions in pollution damages to crops and materials (in terms of monetary benefit), and from acute exposures on premature mortality and respiratory hospital admissions (RHAs) through movement from UKREF to each scenario, for abatement across UNECE. Units shown in left hand column. Results were not available for scenario WGS31c.
|
Effect |
Scenario |
England |
Northern Ireland |
Scotland |
Wales
|
|
Crop damage |
J1 |
10 |
0 |
-0.9 |
1.1
|
|
(£ million) |
H1 |
16 |
0 |
-0.7 |
1.5
|
|
|
|
|
|
|
|
|
Materials damage |
J1 |
13 |
0.5 |
0.7 |
0.8
|
|
(£ million) |
H1 |
13 |
0.5 |
0.7 |
0.8
|
|
|
|
|
|
|
|
|
Premature |
J1 |
490 |
6 |
18 |
23
|
|
mortality(cases) |
H1 |
480 |
5 |
14 |
23
|
|
|
|
|
|
|
|
|
RHAs, |
J1 |
330 |
2 |
8 |
14
|
|
additional/ |
H1 |
350 |
2 |
7 |
19
|
|
brought forward |
|
|
|
|
|
Some discussion of the health effects is needed to aid interpretation. Firstly, the extent to which short term (acute) exposures affect mortality: clinical judgement suggests that those at risk are already ill, probably seriously, and as such are likely to have only a very limited life expectancy. For respiratory hospital admissions it is uncertain to what extent the results show additional cases, and to what extent they are simply cases that would have occurred anyway within a limited period. Again, clinical judgement suggests that those affected are not in the prime of health. These factors clearly complicate valuation of health effects.
A further factor relates to uncertainty over the mechanism of pollutant action on health. Knowledge of mechanisms would undoubtedly improve the quality of assessment for all of the health-damaging pollutants for which effects are quantified here. Relative to mechanism, the following caveats should be drawn against the health assessment in this study:
10 are assumed to be equally potent;
ozone effects are quantified only for the case where a threshold of 50 ppb is assumed. The figure of 50 ppb follows the analysis carried out for Department of Health (1998), rather than the 60 ppb preferred by WHO. A lack of data on average ozone levels (rather than peaks) prevented assessment of effects assuming that there is no threshold.
Potential effects of VOCs and NO
2 have also been omitted from the analysis. For VOCs the inventory would need to be speciated before any assessment of health effects could be undertaken. Even then, exposure-response data tends to be lacking. For NO2 the evidence for health effects from epidemiological studies is regarded here to be too inconsistent. It could also generate a risk of double counting if exposure-response functions are not adequately characterised and the perceived effects of NO2 are, in reality, attributable to other pollutants.
On the other hand, there is evidence for a number of additional effects on health from exposure to air pollution. These effects are explored below, and include hospital admissions for heart disease and effects on asthmatics.
Table 22. % split between pollutants with respect to health effects. The reference point for this assessment is the J1 scenario, mid estimate of acute effects on mortality. Ozone effects calculated with threshold only.
|
Pollutant
|
England |
N. Ireland
|
Scotland |
Wales |
UK
|
|
NO3 |
17% |
0.2% |
0.7% |
0.7% |
18%
|
|
NH4 |
5% |
0.0% |
0.3% |
0.2% |
6%
|
|
SO4 |
17% |
0.4% |
1.3% |
0.9% |
1%
|
|
SO2 |
28% |
0.3% |
1.0% |
1.1% |
30%
|
|
Ozone |
23% |
0.0% |
0.1% |
1% |
26%
|
|
Overall |
90% |
1% |
3% |
4% |
81%
|
6.4 Application of additional exposure-response functions identified by COMEAP
The COMEAP and EAHEAP reports included a number of exposure-response functions for effects of fine particles, additional to those reported above. However, quantification of these effects was considered to be less certain on grounds of:
- evidence being based on a relatively small number of studies;
- lack of data specific to the UK;
- consistency in outcomes of different studies; and
- availability of appropriate baseline data for incidence of effects.
However, given that the functions identified were based on the results of well-conducted studies, it is appropriate, in the context of this analysis, to accept them (whilst recognising the uncertainties that are present) in order to achieve a holistic overview of the effects of emissions abatement.
Results are given in
Table 23 and show the number of cases either additional or brought forward, rather than the additional number of people affected. Again, results are based on the use of best estimates of exposure-response functions.
Results in the Table for cardiovascular disease are not additional to those for ischaemic heart disease and congestive heart failure. However, it is notable that both sets of functions indicate broadly similar totals. Results for cerebrovascular hospital admissions (stroke) are, however, additional to both groups.
The following effects were quantified for asthmatics:
- bronchodilator usage
- cough
- wheeze
Of these effects the most objective are undoubtedly changes in the frequency of use of bronchodilators. There is some possibility that adding together all three effects could lead to double counting (e.g. use of a bronchodilator to relieve other symptoms). Accordingly, for a lower estimate of the number of symptoms, bronchodilator usage only is taken, whilst the upper estimate adds the number of all three types of effect together. The difference is within a factor of 2 for all scenarios. The results raise questions on which effects might be considered to be the most serious: effects on heart disease and stroke, because of severity at the level of the individual sufferer, or effects on asthmatics as a consequence of the much larger number of people affected.
Table 23. Effects of secondary aerosols on heart disease, cardiovascular disease and asthmatics.
|
Effect |
Scenario |
UK to UK |
UK to |
UNECE (including UK)
|
|
|
|
|
UNECE |
to UK
|
|
Cardiovascular |
WGS31c |
|
|
nq
|
|
disease in the |
J1 |
85 |
150 |
140
|
|
elderly (hospital |
H1 |
86 |
150 |
120
|
|
admissions) |
|
|
|
|
|
|
|
|
|
|
|
Congestive heart |
WGS31c |
|
|
nq
|
|
failure (hospital |
J1 |
50 |
88 |
80
|
|
admissions) in |
H1 |
50 |
89 |
67
|
|
the elderly |
|
|
|
|
|
|
|
|
|
|
|
Ischaemic heart |
WGS31c |
|
|
nq
|
|
disease (hospital |
J1 |
47 |
83 |
75
|
|
admissions) in |
H1 |
47 |
84 |
63
|
|
the elderly |
|
|
|
|
|
|
|
|
|
|
|
Cerebrovascular |
WGS31c |
|
|
|
|
hospital |
J1 |
50 |
88 |
80
|
|
admissions |
H1 |
50 |
89 |
67
|
|
(all ages) |
|
|
|
|
|
|
|
|
|
|
|
Bronchodilator |
WGS31c |
|
|
nq
|
|
usage in children |
J1 |
155,000 |
280,000 |
250,000
|
|
and adults |
H1 |
160,000 |
280,000 |
210,000
|
|
(person days) |
|
|
|
|
|
|
|
|
|
|
|
Cough in |
WGS31c |
|
|
nq
|
|
asthmatic adults |
J1 |
180,000 |
320,000 |
290,000
|
|
and children |
H1 |
180,000 |
320,000 |
240,000
|
|
(person days) |
|
|
|
|
|
Wheeze in |
WGS31c |
|
|
nq
|
|
asthmatic adults |
J1 |
86,000 |
150,000 |
140,000
|
|
and children |
H1 |
86,000 |
150,000 |
120,000
|
|
(person days) |
|
|
|
|
6.5 Application of EAHEAP Valuations
Assessment from this point goes beyond that accepted by the Department of Health. However, it can be justified on the grounds of seeing how total benefits may compare to costs. Valuation of acute effects on mortality and respiratory hospital admissions is presented in Table 24. The analysis combines the following data to provide ranges:
- low estimate of deaths brought forward through acute effects on mortality with the lower EAHEAP valuation (£2,600/case);
- mid estimate of deaths brought forward with the intermediate EAHEAP valuation (£110,000/case);
- high estimate of deaths brought forward with the upper EAHEAP valuation (£1.4 million/case).
All values used in the analysis were expressed in 1990£ rather than the 1998£ used by EAHEAP and cited here, to enable comparison with the costs of abatement.
For the lower bound the benefits of reduced mortality from acute exposure are minimal, less than 1% of costs. Using the upper bound leads to benefits of reducing this one type of impact to exceed costs in all scenarios. Using the intermediate position, benefits from reducing acute effects on mortality account for between 5 and 53% of costs, depending on scenario and the range of analysis.
Results for the intermediate valuation of mortality cases should not be regarded as a best estimate simply because they are intermediate. Based on the EAHEAP report, there is no reason for preferring any one of the three positions on valuation there presented to the other two.
For valuation of respiratory hospital admissions ranges were derived as follows:
- low estimate of the number of hospital admissions combined with a valuation of £0/case, on the assumption that all cases are simply brought forward by a limited amount of time;
- high estimate of the number of hospital admissions combined with the upper EAHEAP valuation for RHAs of £3235/case (adding together the upper figures given by EAHEAP for willingness to pay [£735] and for NHS cost savings [£2,500]). This upper estimate is based on the opposite assumption to the lower estimate: that all cases are additional, rather than simply ‘brought forward’.
Results show that the benefits from reducing RHAs are low compared to the costs of pollution abatement, making at most a 1 to 2% contribution for the scenarios studied here.
Sensitivity analysis has also been applied to test how large the value of statistical life (VOSL) would need to be for benefits to match costs, considering only benefits data for materials and crops from Table 20 and premature cases of mortality. Given the exclusion of numerous other effects this sets a maximum for the required VOSL, which can then be compared against the EAHEAP data. Analysis of so-called ‘switching values’ is indeed recommended in DETR guidance on conducting regulatory impact assessment. The analysis has looked at both extremes. First the low estimate of benefits was combined with the upper estimate of costs to give a maximum value. Then the high estimate of benefits was combined with low costs to give a minimum value (subject to the constraint that only effects on crops, materials and acute effects on mortality are included: on this basis the value derived is clearly not a true minimum). Results are given in Table 25. In almost all cases the result lies between the intermediate and upper estimate from EAHEAP.
Table 24. Valuation of acute effects on mortality and respiratory hospital admissions.
|
Effect |
Scenario |
UK to UK |
UK to |
UNECE (including UK)
|
|
|
|
|
UNECE |
to UK
|
|
Acute effects on |
WGS31c |
|
|
0.21
|
|
mortality) |
J1 |
0.46 |
0.89 |
0.72
|
|
(£ million) |
H1 |
0.49 |
0.99 |
0.66
|
|
Lower bound |
|
|
|
|
|
|
|
|
|
|
|
Acute effects on |
WGS31c |
|
|
12
|
|
mortality |
J1 |
26 |
54 |
42
|
|
(£ million) |
H1 |
29 |
69 |
41
|
|
Intermediate |
|
|
|
|
|
|
|
|
|
|
|
Acute effects on |
WGS31c |
|
|
190
|
|
mortality |
J1 |
390 |
840 |
640
|
|
(£ million) |
H1 |
450 |
1,100 |
630
|
|
Upper bound |
|
|
|
|
|
|
|
|
|
|
|
RHAs (£million) |
WGS31c |
0 |
0 |
0
|
|
Lower bound |
J1 |
0 |
0 |
0
|
|
|
H1 |
0 |
0 |
0
|
|
|
|
|
|
|
|
RHAs (£million) |
WGS31c |
|
|
0.4
|
|
Upper bound |
J1 |
0.70 |
2.0 |
1.3
|
|
|
H1
|
0.91 |
2.9 |
1.4
|
Table 25. Determination of ‘switching values’ for the VOSL, including only acute effects on mortality and effects on agriculture and materials. All values in 1990£.
|
|
|
UK to UK |
UK to UNECE |
UNECE to UK
|
|
|
PART 1: Upper estimates
|
|
|
|
|
|
Scenario: J1
|
|
|
|
|
I |
Acute deaths (cases, lower bound)
|
253
|
484 |
394
|
|
II |
Crop + material damage (£M) |
16 |
25 |
25
|
|
III |
Costs of abatement for UK (£M) |
161 |
161 |
161
|
|
IV |
Residual cost (III – II) (£M) |
145 |
136 |
136
|
|
V |
VOSL – switching value/case (IV/I) (£) |
570,000 |
280,000 |
350,000
|
|
|
Scenario: H1
|
|
|
|
|
I |
Acute deaths (cases, lower bound)
|
266
|
541 |
357
|
|
II |
Crop + material damage (£M) |
28 |
45 |
33
|
|
III |
Costs of abatement for UK (£M) |
567 |
567 |
567
|
|
IV |
Residual cost (III – II) (£M) |
539 |
522 |
534
|
|
V |
VOSL – switching value/case (IV/I) (£) |
2,000,000 |
960,000 |
1,500,000
|
|
|
PART 2: Lower estimates
|
|
|
|
|
|
Scenario: J1
|
|
|
|
|
I |
Acute deaths (cases, upper bound)
|
393
|
848 |
644
|
|
II |
Crop + material damage (£M) |
16 |
25 |
25
|
|
III |
Costs of abatement for UK (£M) |
148 |
148 |
148
|
|
IV |
Residual cost (III – II) (£M) |
132 |
123 |
123
|
|
V |
VOSL – switching value/case (IV/I) (£) |
340,000 |
150,000 |
190,000
|
|
|
Scenario: H1
|
|
|
|
|
I |
Acute deaths (cases, upper bound)
|
449
|
1097 |
637
|
|
II |
Crop + material damage (£M) |
28 |
45 |
33
|
|
III |
Costs of abatement for UK (£M) |
554 |
554 |
554
|
|
IV |
Residual cost (III – II) (£M) |
526 |
509 |
521
|
|
V |
VOSL – switching value/case (IV/I) (£) |
1,200,000 |
460,000 |
820,000
|
6.6 Valuation of additional effects for which COMEAP provided functions
The results from Section 6.4 were valued using data from the European Commission ExternE Project. Results made only a minor contribution to total benefits (between £1 million and £3 million in total, depending on scenario and range considered). Because the contribution compared to costs was small the results are not presented separately here, but are given in Appendix 8.
6.7 Chronic effects on mortality
The COMEAP report (paragraphs 3.49 to 3.50) notes that quantification of chronic effects on health is prone to a high degree of uncertainty. However, they also note that, if available data are reliable, then the overall impacts on health are likely to be substantially greater than estimates that ignore chronic effects on mortality. Analysis in the Netherlands suggests a reduction in life expectancy amongst men of about one year on average, as a consequence of exposure to particle levels that are typical of the UK (Brunekreef, 1997). Here quantification of chronic exposure effects on mortality is based on the results of earlier work in the UK by Hurley et al, under the EC GARP II Project (Markandya et al, 1999). The assumptions underlying this part of the assessment are described in Appendix 6.
The outputs of the quantification of these chronic effects are estimates of the reduction in longevity (life years lost) spread across the population, as a result of long-term pollution exposure. Results are, therefore, not in the same units as those for acute effects on mortality, which relate solely to change in the number of cases of death brought forward in each scenario. It would, in theory, be possible to express results in terms of the number of deaths brought forward for a specified time period, though this would require additional analysis beyond the scope of this study.
EAHEAP did not specifically consider valuation of chronic effects on mortality. However, some of their results did, in effect, quantify Willingness to Pay (WTP) against change in life expectancy related to air pollution exposure. Their lower valuation (£2,600) was based on the loss of one month of life amongst the elderly, for those with a much reduced quality of life, whilst the intermediate valuation (£110,000) was based on the loss of a year of life for those in reasonable health. If one accepts the EAHEAP approach as being broadly correct, it is likely that one would wish to introduce additional factors to generate a VOSL or range specific to chronic effects on mortality. For the upper estimate £110,000 per life year lost was used. For the low estimate the figure of £2,600 was multiplied by 12 to give £31,200 per year. Given that this part of the analysis seeks to value an effect that will happen after a long period of exposure, it may be appropriate to discount effects over a number of years. Accordingly, the £31,200 figure has been scaled back in accordance with earlier analysis under the ExternE Project (European Commission, 1999) to give a figure of £19,000 (1990£). Doubtless, there would be debate about whether this is applicable or not, and whether additional factors should be introduced. However, this analysis is not about generating specific data but, instead (given a lack of confidence in much of the data that are available) about testing alternative assumptions and investigating ranges. Given the broad range applied here, and the outcome of the analysis, any final figure would not be so different as to significantly affect the results. It could be argued that the upper estimate given here is too conservative – that the real figure could well be higher. However, this is of limited (though still some) relevance to the present analysis, given the extent by which benefits exceed costs (see below) when the upper estimate is taken.
Results are shown in Table 26. They are illustrative for a given set of assumptions and alternative, and equally plausible, assumptions are possible. The ranges shown are therefore unlikely to represent the full range of possible answers, though broadening the range would have only a limited effect on the outcome of the comparison of costs and benefits.
Table 26. Chronic effects on mortality.
|
|
Scenario |
UK to UK |
UK to UNECE |
UNECE (incl. UK) to UK
|
|
Life years lost |
WGS31c |
|
|
1,800-2,300
|
|
|
J1 |
3,200-4,200 |
5,700-7,500 |
5,300-6,900
|
|
|
H1 |
3,200-4,200 |
5,800-7,600 |
4,400-5,800
|
|
|
|
|
|
|
|
£ million |
WGS31c |
|
|
37-270
|
|
|
J1 |
43-330 |
77-580 |
110-800
|
|
|
H1 |
44-330 |
78-590 |
89-670
|
|
|
|
|
|
|
Clearly, these results would make a major difference to the analysis if accepted. The upper end of the ranges shown for economic benefit would be sufficient on their own to more than match costs for all scenarios.
6.8 Application of other Functions and valuations used in the studies for UNECE and the EC
A series of other effects are quantified and monetised in Appendix 8, including:
- ozone damage to forests;
- asthma attacks induced by ozone exposure;
- incidence of chronic bronchitis in adults;
- incidence of bronchitis in children;
- incidence of chronic cough;
- restricted activity days linked to exposure to secondary particles;
- minor restricted activity days linked to ozone exposure.
Quantification of these effects can be done with very limited confidence. However, where possible, broad ranges have been applied to give an indication of the possible scale of benefits. These ranges are used in the comparison of costs and benefits that follows.
6.9 Comparison of Costs and Benefits
This chapter and Appendix 8 are structured to show readers what effects can be considered quantified with the highest confidence, and which with the least. Appendix 8 includes running totals for monetised effects, enabling identification of the point at which benefits outweigh costs (if at all). These results are summarised here (Table 27). The lower bound contains conservative estimates of the benefits to be derived from reducing emissions. This includes taking the lower bound for exposure-response functions and valuations, and elimination of any effects that might be double counted, though this transpires to be of very little importance. The upper bound is naturally based on upper estimates for exposure-response functions and valuations, and also a slightly less cautious approach on the potential for double counting. Again, this last point is of little importance. For effects where double counting is probable precaution is applied to both the upper and lower bounds in the running totals. Only a few relatively minor effects of those quantified are in the grey area where it is not clear whether or not their inclusion would lead to double counting.
Table 27. Cost-benefit analysis: effects required under each scenario for benefits to outweigh costs.
|
|
UK to UK |
UK to UNECE |
UNECE (incl. UK) to UK
|
|
WGS31c |
|
|
|
|
Lower bound |
Not quantified |
Not quantified |
Not quantified
|
|
Upper bound |
Not quantified |
Not quantified |
Not quantified
|
|
J1 |
|
|
|
|
Lower bound |
Costs exceed sum of all quantified benefits |
Costs exceed sum of all quantified benefits |
Costs exceed sum of all quantified benefits
|
|
Upper bound |
Crops
Materials
Acute mortality (NO3, SO4, NH4 only) |
Crops
Materials
Acute mortality (NO3, SO4only) |
Crops
Materials
Acute mortality (NO3, SO4only)
|
|
H1 |
|
|
|
|
Lower bound
|
Costs exceed sum of all quantified benefits
|
Costs exceed sum of all quantified benefits |
Costs exceed sum of all quantified benefits
|
|
Upper bound |
Crops
Materials
Acute mortality
Acute RHAs
Heart disease
Asthma
Chronic mortality (SO4 only) |
Crops
Materials
Acute mortality (NO3, SO4, NH4, SO2 only)
|
Crops
Materials
Acute mortality (NO3, SO4, NH4, SO2 only)
|
Note: Health effects are quantified against exposure-response functions for PM10. There is no direct evidence of effects of nitrate aerosols from epidemiological studies, though some studies have looked specifically at associations with sulphates.
Table 27 shows that costs lie somewhere in the range calculated for the total benefits. At first sight this does not appear to be particularly useful. After all, the analysis started from the position of wanting to know for any scenario whether or not costs were likely to be bigger than benefits! However, the ranges selected for each variable were very broad, seeking to quantify the full potential range in benefits. The next stage of the assessment investigates benefit-cost ratios.
Table 28. Benefit-cost ratios, with all monetised effects taken into account (excludes damage to cultural heritage and natural ecosystems). Positive numbers denote benefits in excess of costs by the factors quoted. Negative numbers denote costs in excess of quantified benefits by the factors quoted. Columns identify the source of emission and the receptor for benefits of abatement (in both cases this is either UK or full UNECE). Costs applied are the costs to the UK in all cases.
|
|
UK to UK |
UK to UNECE |
UNECE (incl. UK) to UK
|
|
J1 (cost £161M) |
|
|
|
|
Lower bound |
-2.38 |
-1.35 |
-1.11
|
|
Upper bound |
4.01 |
8.04 |
8.20
|
|
H1 (cost £567M) |
|
|
|
|
Lower bound |
-7.14 |
-3.85 |
-4.17
|
|
Upper bound |
1.26 |
2.77 |
2.15
|
|
WGS31c (cost £61M) |
|
|
|
|
Lower bound |
Not quantified |
Not quantified |
Not quantified
|
|
Upper bound |
Not quantified |
Not quantified |
Not quantified
|
The results shown in Table 28 demonstrate where estimated costs lie between the upper and lower bounds for benefits. Costs for J1 tend to be towards the lower end of the range, in other words, there appears a reasonable likelihood that the real benefit would exceed the estimated costs. This is not the case for H1, however, where the bias is the other way round, with costs towards the upper end of the benefits range.
Chapter 5
Chapter 7
Report and site prepared by the National
Environmental Technology Centre, part of AEA Technology, on behalf of the UK
Department of the Environment, Transport and the Regions