Appendix 2

The objective of this Appendix is to quantify the French (FR) and Netherlands (NL) emissions limitation proposals in order to put them into a perspective relative to the proposals put forward by other countries (see Table 3 in the main text) and the IPCC IS92 scenarios. We begin by considering the proposals themselves in terms of Annex I country emissions. We then extend these results to the global scale by incorporating non-Annex I country emissions. As noted in the main text, non-Annex I country emissions are based on the assumption of no emissions limitations for these countries. Note also that, although the proposals refer to reductions of greenhouse gas emissions, we take these to apply to fossil CO2 emissions alone.

The FR and NL proposals provide a useful comparison because they are based on quite different emissions criteria, per capita emissions in the former and absolute emissions in the latter. The use of per capita emissions introduces an additional element of uncertainty, future population, which makes the calculation of absolute emissions quite complex. While per capita emissions can provide a useful unifying concept for comparisons between different countries (and between Annex I and non-Annex I countries in particular), the complexity of the calculations given below would make any emissions limitations proposal worded in this way more difficult to implement.

Annex I Country Emissions

For the NL proposal, delineation of future fossil CO2 emissions for Annex I countries is straightforward. For these countries, IPCC92 (Leggett, et al., 1992) and Pepper, et al. (1992) may be used to give a value of 4.59 GtC/yr for fossil CO2 emissions in 1990 (further information is given in Appendix 3). We assume (see main text) that emissions of Annex I countries remain at 1990 levels to 2000. Therefore, the suggested decline by 1-2 per cent (compound) per year (main text, Table 1) must be assumed to begin in 2000 (as given in Table 2 in the main text). We consider two separate cases: a decline from 2000 at 1 per cent compound per year (NL-1%) and a decline from 2000 at 2 per cent compound per year (NL-2%).

The FR proposal is more complicated. Again, our analysis of this proposal includes the assumption of constant 1990-level Annex I country emissions over 1990-2000. Over 2000-2010, emissions changes are specified as a drop of 7-10 per cent in emissions per capita. To calculate the implied absolute emissions values, we need to know the projected value of emissions per capita in the year 2000. IPCC documents may be used to estimate this value, but as a range of possibilities

dependent on the scenario for future population (see Table 4 in the main text).

For the IS92 scenarios, three population projections were employed, the UN Medium-Low case (used in IS92c and d), a central World Bank case (used in IS92a, b and e), and the UN Medium-High case (used in IS92). Estimated population values for 2000 in these projections for Annex I countries are 1.286 bn, 1.313 bn and 1.354 bn, linearly interpolated from Table 4. These values, together with a total emissions value of 4.59 GtC/yr, give per capita annual emissions of 3.57 tC/yr, 3.50 tc/yr and 3.39 tC/yr, respectively in 2000, compared with a 1990 value of 3.63 tC/yr. Since the range is relatively small, and the 2000 value is, in any event, uncertain, we take the average of the three values (3.485 tC/yr if the figures are not rounded) as a single, representative value for annual per capita emissions in 2000 (some 4 per cent below the 1990 level).

Applying the specified 7-10 per cent reduction in per capita emissions over 2000-2010 to this value for the year 2000 gives an annual emissions per capita range in the year 2010 of 3.14-3.24 tC/yr. To convert this range to actual emissions we multiply by the population value. In doing so we must again account for uncertainties in population growth: the IPCC range for 2010 population linearly interpolated from Table 4 is 1.307-1.444 bn for the UN scenarios and 1.363 bn for the World Bank scenario. The extreme-high emissions value for 2010 (high population, high per capita emissions) is, therefore, 4.68 GtC/yr. The extreme-low value (low population, low per capita emissions) is 4.10 GtC/yr. Central values range from 4.27 GtC/yr (mid population, low per capita emissions) to 4.41 GtC/yr (mid population, high per capita emissions), with a mean of 4.34 GtC/yr. This set of 2010 emissions values is illustrated in Figure B1. Numerical details are given in Table B1.

For 2100, the FR proposal gives an absolute range of emissions per capita values for Annex I countries of 1.6-2.2 tC/yr. To convert these to total emissions, we must continue to account for population growth uncertainties, as specified by IPCC and given in Table 4 of the main text. To give a high range of emissions values in 2100, we use the high emissions per capita value (2.2 tC/yr); for low, mid and high population levels this gives 1.85, 3.12 and 4.87 GtC/yr (see Table B1). Similarly, a low range, using 1.6 tC/yr per capita emissions, is 1.34, 2.27 and 3.54 GtC/yr (see Table B1).

The complete Annex I emissions limitation scenarios are obtained by combining the set of emissions values in 2010 with those for 2100, which we do simply by assuming a linear

46-495-28

1800 2000 2010 2020 2030 2040 2050 2000 2070 2000 2000 2100 Year

Figure B1. Fossil CO2 emissions (GtC/yr) for Annex I countries under the French (FR) emissions limitation proposal. The words high, mid and low refer to the UN Medium-Low, World Bank and UN Medium-High population projections used, while the appended designators "HIGH", "LOW" or no designator refer to the use of high, low or mid values of per capita emissions. Changes over 2000-2010 correspond to 7-10 per cent reductions in per capita emissions accounting for uncertainties in future populations. Values in 2100 correspond to per capita emissions of 1.6 tC/yr or 2.2 tC/yr again accounting for population projection uncertainties.

high population estimates. Thus, when used in conjunction with emissions for non-Annex I countries under the IS92 scenarios, only scenarios with the same population assumptions should be considered together. In the calculations in the main text, we consider only the IS92a, c and e scenarios for non-Annex I countries, which span the full range of emissions possibilities considered by IPCC. The only fully consistent combinations involving the FR emissions limitation proposals for Annex I countries and IS92a, c or e emissions for non-Annex I countries are therefore FR-Low with IS92c, FR-Central with IS92a, and FR-Central with IS92e.

The high population scenario is used only in IS92f, which, because of other assumptions made in developing this scenario (see IPCC92-Leggett, et al., 1992) has lower emissions than the mid-population-based IS92e case. Since we need here to span the full range of possibilities, we combine the FR-High and IS92e

1

0.5

For the NL emissions limitation proposals, the Annex I country emissions cases (NL-1% and NL-2%) are independent of and may, therefore, be combined with non-Annex I country emissions for each of IS92a, c and e. The full set of global emissions scenarios is therefore (non-Annex I case first): IS92a with FR-Central, NL-1% and NL-2% (Figure B3a), IS92c with FR-Low, NL-1% and NL2% (Figure B3b); and IS92e with FR-High, NL-1% and NL-2% (Figure B3c). In the main text, these are illustrated together in Figure 4. Here we give the a, c and e results separately for greater clarity.

Fosall CO, emissions (C/y)

(b)

2000 2010 2020 2030 2040 2060 2080 2070 2000 2000 2100

Figure B3. (a) Global fossil CO2 emissions (GtC/yr) for the IS92a scenario compared with emissions where Annex I countries follow emissions limitation proposals NL-1%. NL-2% or FR-Central and non-Annex I countries follow IS92a. (b) Global fossil CO2 emissions (GtC/yr) for the IS92c scenario compared with emissions where Annex I countries follow emissions limitation proposals NL-1%, NL-2% or FR-Low and non-Annex I countries follow IS92c. (c) Global fossil CO2 emissions (GtC/yr) for the IS92e scenario compared with emissions where Annex I countries follow emissions limitation proposals NL-1%, NL-2% or FR-High and non-Annex I countries follow IS92e.

33

Appendix 3

The emissions values used in this Paper, and their breakdown into Annex I and non-Annex I country emissions, are based primarily on information given in or prepared for the IPCC92 (Leggett, et al., 1992; Pepper, et al., 1992). In this Appendix we assess how sensitive the results are to the precise emissions values used in the main text. In particular, we consider the effect of uncertainties in the baseline value for Annex I country emissions, which, here, is the 1990 level for these emissions. We begin by considering the global emissions level in 1990. For the IS92a scenario, the global CO2 emissions value for 1990 given in the IPCC92 (Leggett, et al., 1992, Table A3.6) is 7.4 GtC/yr. In Table A3.11 of that report, this is broken down into 6.0 GtC/yr from energy, 1.3 GtC/yr from deforestation (land-use changes), and 0.2 GtC/yr from cement production. The inconsistency here, of 0.1 GtC/yr, is a rounding error. Further details are given in Pepper, et al. (1992, Table 3.6.1); here, the cement production contribution is 0.15 GtC/yr. It has been noted previously (Enting, et al., 1994, pp. 69, 70) that these values are insufficiently precise for carbon cycle model calculations, that rounding errors cannot be admitted into any credible analyses, and that the IPCC92 (Leggett, et al., 1992) and Pepper, et al. (1992) emissions values in 1990 do not agree with more recent estimates. Accordingly, in IPCC work subsequent to IPCC92 the 1990 emissions value used has been 6.10 GtC/yr for emissions from energy and cement production (from Marland and Boden, 1991; but consistent with the later analysis of Marland, et al., 1994 — for information, the most recently determined value is 6.11 GtC/yr). This global total includes emissions from all sources: fossil fuels including bunker fuels (fuels used for international travel) and gas flaring: cement production; and non-fuel CO2 production.

For the breakdown into Annex I and non-Annex I countries, we rely primarily — as in TPI (Acosta Moreno, et al., 1996, Table A3, p. 78) — on IPCC92 (Leggett, et al., 1992) and Pepper, et al. (1992). IPCC92 Table A3.7, gives 4.5 GtC/yr for the sum of OECD, USSR and Eastern Europe emissions. Although not clearly stated, this is the energy component only (as can be ascertained from Pepper, et al., 1992, Tables 3.1.16-21). The cement production term adds an additional 0.09 GtC/yr (Pepper, et al., Table 3.6.1). The total is therefore 4.59 GtC/yr, the main value used in this Paper. Although it is not stated specifically in IPCC92 (Leggett, et al., 1992) or Pepper, et al. (1992), it is likely that this value does not include emissions from combustion of bunker fuels27. There is good reason to believe that this value is too high, partly because the IPCC92 analyses are now out of date, and partly because the above country grouping is not identical with the grouping that defines

Annex I countries. In particular, the former USSR group contains a number of countries that are not in Annex I.

More recent work, noted in SAR WGII (Nakicenovic, et al., 1996; Kashiwagi, et al, 1996), also suggests that the value 4.59 GtC/yr may be too high. This report uses information from di Primio (1993) and Marland, et al. (1994), which is probably of higher quality than that used by IPCC92. Data in Marland, et al. (1994) indicate that a more correct value is around 4.0 GtC/yr. Since we are only concerned with estimating the sensitivity of results presented in the main text to this 1990 reference number, we do not need to know it precisely. We therefore carry out a set of analyses using 4.09 GtC/yr for 1990 Annex I country emissions, 0.5 GtC/yr below the value used for our primary calculations. How sensitive are the concentration projections given elsewhere in this report to the assumed value of Annex I country emissions in 1990? To answer this question, we need to determine how both components of future global emissions, Annex I emissions and the residual (non-Annex I plus bunker fuels28) emissions, would be influenced by a change in the 1990 “reference" emissions value for Annex I countries. To be specific, we consider the NL-1% and NL-2% emissions limitation proposals.

For Annex I countries, the recalculated emissions are lower at all times because future emissions are tied to a lower 1990 reference value of 4.09 GtC/yr. For the residual, the situation is less clear and different choices are possible. The 1990 value must be higher (viz. 2.01 GtC/yr instead of 1.51 GtC/yr), because all cases have the same 1990 value for global emissions (6.10 GtC/yr). In the year 2000 and subsequently, we could either retain the IS92 values used previously (viz. 2.00 GtC/yr in 2000 for the IS92a case, etc. — see Figure 2 in the main text), or we could inflate all values by the 1990 "error" of 0.5 GtC/yr (i.e., retain the IS92 changes rather than their absolute values). The first method produces global emissions

27 Note that it is not clear how bunker fuels will be considered in any legal instrument regarding emissions limitations, nor how or if the global bunker fuel component of emissions could be apportioned between Annex I and non-Annex I countries. We have therefore assumed throughout that the Annex I component of emissions does not include this term.

28 Note that there is no information on future emissions from combustion of bunker fuels, so these are not accounted for. Since this small component would be common to both the original and modified emissions scenarios, neglecting it will not affect the results in any noticeable way.