Projection of future emissions

The development of emission projections typically requires assumptions about economic growth, population growth and the emission characteristics of new production technologies. These are building blocks for the development of more detailed sets of assumptions and parameters, such as energy use projections, livestock developments, production of goods, changes in environmental legislation leading to different emission factors over time, etc.

To consider changes in the spatial distribution of emissions, additional factors may need to be projected, e.g. the pace of economic and demographic development in the considered regions (also at subnational level). Important shifts in spatial distribution of emissions could occur, for example, if the scenario of opening up of the Arctic region for shipping would be realized. Likewise, emissions associated with oil and gas exploration and development in Siberia and Central Asia could alter the patterns of emissions of hydrocarbons and other species.

4.4.1 Driving forces

The most important factors determining future emission levels are activity, level of technology development and penetration of abatement measures. Activity changes are strongly linked to economic growth, population growth and energy growth, but they are also dependent on the geopolitical situation, trade agreements, level of subsidies, labour costs, etc. While production technology improvements (with respect to emission levels) are also related to economic growth, a far more important factor is environmental legislation. The latter can be a key factor in determining the penetration of abatement measures and consequently the apparent emission factors. Comparison of historical per-capita NOx emissions in the United States and Europe shows a strong relationship to per capita income, in which growth beyond about $5,000 leads to a strong increase in car ownership.

Recently, several developing countries reached such income levels, and a rapid increase in traffic-related emissions and worsening of air quality, especially in megacities, has occurred. Historically, however, societal acceptance of measures to improve local air quality has also grown with increasing economic wealth. Consequently, it is unlikely that global air pollutant emissions will grow substantially in the future.

As one example, figure 4.3 shows the varying stages of automobile emission restrictions in Asian countries, compared to the stage of application in the EU (Euro 1-4). Traditionally, national legislation drove the installation of control technology, but in some regions international (regional or global) agreements have become the key drivers. Examples include the Kyoto Protocol, the protocols to the UNECE Convention on Long-range Transboundary Air Pollution, and EU Directives. At the national level, the economic projections are frequently updated, as are as some key activity factors, e.g. population and energy use. Regional or global projections of drivers are updated less frequently,

and such work is often driven by policy needs, e.g. the global SRES (Special Report on Emission Scenarios), EU energy or agricultural projections and the work of international agencies such as the International Energy Agency (IEA), the Organisation for Economic Co-operation and Development (OECD), and the Food and Agriculture Organization of the United Nations (FAO).

Figure 4.4 demonstrates the estimated impact of already committed legislation on NOx

emission estimates for the SRES scenarios compared to the original SRES scenario results. Cofala et al. (2007) assessed the technically feasible emission reduction potential within the next 20 years.

Such potential varies for various pollutants and regions, but on a global scale emissions of NO_x or SO2 could be reduced by about 50–70 per cent compared to the current levels, while those of BC could probably be reduced by no more than 30–40 per cent. The major reason for the lower reduction potential for BC is the significantly different emission source structure; over 60 per cent of emissions originate from small scale domestic combustion, a sector where few add-on measures exist and the most efficient reduction strategy is to replace stoves and boilers with new ones. It is expected, however, that many of the traditional appliances will remain in use until 2030, especially in the developing world. One additional aspect of environmental legislation is consideration of the actual level of compliance. Typically, historical inventories take into account information on compliance both in terms of the timely introduction of respective laws (emission standards) and the actual performance of the installed control equipment. For projections, it is assumed that the technical abatement measures will be installed in a timely manner to comply with the law. As far as performance of the equipment is concerned, approaches vary between studies: some assume that emission factors equal emission standards, while others make explicit assumptions about the probability of failure, e.g. the percentage of “smokers” among the vehicle population. The latter assumptions most often rely on the experience with existing equipment that might not necessarily be representative for new and future technologies. For a good understanding of the projections, it is of utmost importance to state these assumptions explicitly.

Figure 4.5 demonstrates the importance of careful consideration of potentially rapid changes of emission factors, especially in developing countries. In China, the building of new, large coal-fired power plants to replace or augment the set of older, smaller plants has dramatically altered the mix of plants, and in just 10 years has cut the average emission factor of the ensemble of plants by about 16 per cent.

Last but not least, there is strong interdependence among different air pollutant species, such that many species can be mitigated at the same time by certain kinds of environmental policies, i.e.

ambitious CO2 reduction targets will also result in significant reduction of air pollutants. The IIASA GAINS model showed such results for a number of European scenarios (Amann et al., 2007; IIASA, 2004).

Country 95 96 97 98 99 00 01 02 03 04 05 06 07 08 09 10 11 12 13 14

d – Delhi, Chennai, Mumbai, Kolkata, Bangalore, Hydrabad, Agra, Surat, Pune, Kanpur, Ahmedabad, Sholapur, Lucknow; Other cities in India are in Euro 2

e – Beijing and Guangzhou (as of 01 September 2006) have adopted Euro 3 standards; Shanghai has requested the approval of the State Council for implementation of Euro 3 f – Euro 4 for gasoline vehicles and California ULEV standards for diesel vehicles

g – Gasoline vehicles under consideration

d – Delhi, Chennai, Mumbai, Kolkata, Bangalore, Hydrabad, Agra, Surat, Pune, Kanpur, Ahmedabad, Sholapur, Lucknow; Other cities in India are in Euro 2

e – Beijing and Guangzhou (as of 01 September 2006) have adopted Euro 3 standards; Shanghai has requested the approval of the State Council for implementation of Euro 3 f – Euro 4 for gasoline vehicles and California ULEV standards for diesel vehicles

g – Gasoline vehicles under consideration

2000 2010 2020 2030 2040 2050 2060 2070 2080 2090 2100

Million tons NOx

This work: range A2-B1 This work B2 Original IPCC SRES B2 Original IPCC SRES A2

Original SRES A2

Figure 4.4 NOx emissions in the SRES scenarios. Source: Cofala et al. (2006).

Figure 4.5 Illustration of how technology renewal in a rapidly industrializing country (China) can change the mix of plants and the net emission factor in a relatively short time.

4.4.2 Methods

There are two principal categories of approaches used to develop views of future emissions:

• Projections of activities that generate emissions (e.g. energy use, fertilizer use, livestock and production of goods) together with production technology development and penetration of abatement measures, based on existing and forthcoming legislation and autonomous improvement of technology over time, e.g. Streets et al. (2004).

• Projections of proxies, such as population or economic growth, to change emissions over time, assuming little or no change in unit emissions; an enhancement of this approach is to use elasticity against emissions to account for improvements in production technology or increased penetration of abatement measures, but this approach requires much historical data.

It is important to carefully consider consistency when compiling projections from different sets of data where the underlying methods may differ or the assumptions are not well known or documented.

4.4.3 Future emission inventories

There are a number of key studies and papers that provide important information on future emission levels, globally and in certain world regions and countries.

The IPCC SRES (Special Report on Emission Scenarios) scenarios (IPCC, 2000a) are well known and reflect a large, global, long-term effort, and so cannot be updated very often (the last scenarios were developed in the mid-1990s). Although the SRES scenarios assume improvements in production technology, they do not include some of the expected changes in the future penetration of abatement measures (the impacts of existing legislation); also, they do not include some of the aerosols and PM species and are available only for aggregated regions rather than countries.

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 6.50 7.00 7.50 8.00 8.50 9.00

g/kg

>300 GW 200-300 GW 100-200 GW

<100 GW EF

There are a number of global projections that have been published in the peer-reviewed literature. For example, Streets et al. (2004) developed a forecast of future BC and OC emissions, drawing on SRES activity data and incorporating the evolution of production and control technology, specifically for non-industrial sectors. Cofala et al. (2007) developed global projections for air pollutants (excluding NMVOCs, NH3, and PM) and methane up to 2030. The spatial resolution varies by continent. A longer-term projection (up to 2100) for BC and OC but also taking into account CO2

abatement options and policies was prepared by Rao et al. (2005); the activity data draw on the SRES scenarios. As part of its Clean Air Interstate Rule (CAIR), the U.S. EPA has developed near-term emission forecasts of SO2 and NOx (http://www.epa.gov/cair/index.html).

For Europe, the RAINS model includes projections of air pollutants and greenhouse gases up to 2030, developed in consultation with national experts (Amann et al., 2006a), and the EMEP database contains official projections (up to 2010) for several European countries. The EMEP database, however, often lacks the supporting data that would allow for reconstruction and verification of emissions.

For Asia, several studies looked at particular pollutants, while the RAINS-Asia model has been used to prepare a consistent set of projections drawing on national energy data and international studies for other drivers. The work on global projections (Cofala et al., 2007) includes updates for Asia (reflecting changes in legislation) and new projections for the Russian Federation.

Figure 4.6 presents examples of SO2 and NOx emission projections up to 2030 for OECD countries, Asia, and the rest of the world. Also shown are trends in three of the main driving forces of emissions: population, GDP, and energy use.

Figure 4.6 Examples of SO2 and NOx emission projections up to 2030 for OECD countries, Asia, and the rest of the world. Projections of emissions of SO2 (upper frames) and NOx (lower frames) and their major driving forces in three world regions (Amann et al., 2006b).