Implications for climate and sea level of revised IPCC emissions scenarios

T. M. L. Wigley & S. C. B. Raper

IN 1990, the results of an international assessment of the greenhouse problem were published under the auspices of the Intergovernmental Panel on Climate Change, referred to below as IPCC90¹. Scenarios were given for future emissions of the ma jor greenhouse gases, and these were interpreted in terms of future changes in greenhouse gas concentrations, global-mean temperature and global-mean sea level. Since then, there have been important scientific and methodological developments. To account for these, and as background material for the forthcoming United Nations Conference on Environment and Development, IPCC has produced an update of the 1990 assessment, IPCC92². This includes a new set of emissions scenarios^3,4 which are more comprehensive and, methodologically, more soundly based than the earlier scenarios.

The IPCC update, however, only briefly analyses the implications of these new scenarios for future greenhouse gas concentrations, radiative forcing and global-mean temperature changes⁵; sea-level implications are not considered. Here we aim to assess the implications in detail. In so doing, we investigate not only the direct consequences of the changes in the emissions scenarios, but also how the results are affected by uncertainties in carbon cycle modelling, the negative feedback on halocarb on radiative forcing through stratospheric ozone depletion^6,7, and the possible negative forcing due to anthropogenic SO₂ emissions and sulphate aerosol production^7-10.

The six new emissions scenarios, IPCC scenarios (IS) 92a-f, differ fundamentally from the four earlier ones (from IPCC90) in a number of ways. First, they account for new policies already implemented or proposed for controlling CO₂ emissions a nd halocarbon production, and allow for recent political changes. Second, and more important, they were not developed 'top down' to achieve a prescribed set of radiative forcing targets, but were based on a wide range of assumptions regarding the socioec onomic factors that influence the development of emissions in the absence of new multilateral or unilateral efforts to reduce emissions^3,4. The new scenarios thus represent a range of possible futures all corresponding to the continuation of ' Business as Usual'. They therefore replace the single such scenario given by IPCC90. The new scenarios differ from each other because they make different assumptions regarding population growth, economic growth, technological developments, resource limitations, fuel mixes, agricultural development, the implementation or not of already-proposed policies, and so on.

The most important background assumptions for the scenarios and their emissions values for the year 2100 are given in Table 1. Radiative forcing implications and comparisons with the IPCC90 scenarios are discussed later.

The overall range of emissions projections (Table 1) may seem large given that all scenarios are based on an 'existing policies' assumption, and that IPCC90 gave only a single such scenario. Nevertheless, the range realistically reffects the extreme uncertainties in predicting future population growth and the development of economic, agricultural and industrial activities. For CO₂ at least, as noted in ref. 4, the results are in accord with uncertainties expressed in earlier studies^11,12. Our main concern here, however, is not the emissions scenarios, but their implications for climate and sea level. They are taken here as the starting point for further analysis.

Concentrations and radiative forcing

To convert concentration changes to radiative forcing, we follow IPCC90 and use the relationships given in ref. 14 (Tables 2.2 and 2.4). In accord with IPCC^7,14, no attempt is made to include radiative forcing changes arising from possible cha nges in tropospheric ozone.

Carbon dioxide. The carbon cycle model we use employs a convolution integral representation of ocean uptake^15,16, which is based on the ocean general circulation carbon cycle model of Maier-Reimer and Hasselmann¹⁷. The ocean component is coupled to a four-box terrestrial carbon cycle¹⁸. The model has considerable flexibility in that its rate of ocean uptake can be varied in an internally consistent way by changing the decay times in the Green's function of the co nvolution integral^15,16, and the efficiency of various biological feedback processes can be altered by changing parameters in the terrestrial component of the model¹⁸.

The model is run in two modes. The first is an ocean-only mode with no feedbacks (the NFB mode (or model)). In NFB mode, the projections accurately reproduce all of the IPCC90 best guess projections¹³. In the second (FB) mode, the terrestria l biosphere is included. This introduces a variety of feedback processes into the model. The most important of these is the negative feedback due to CO₂ fertilization, the enhancement of biomass productivity caused by increasing CO₂ concentrations.

The following items apply to both cases. First, the model's mean flux of CO₂ into the ocean for the 1980s is 1.95 gigatonnes (Gt) C yr^-1, close to the IPCC90 best guess¹⁹ of 2 Gt C yr^-1. Second, as in IPCC90 ( and all other projections of future CO₂ concentration changes), the relatively small sources due to oxidation of fossil fuel-derived methane and carbon monoxide are ignored. Third, to initialize the model with 1990 conditions that are consiste nt with observed concentration changes before 1990, the model is first run in inverse mode from 1765 to 1990. This exposes a considerable problem that arises when using the ocean-only (NFB) form of the model.

For an inverse calculation, the model input is the observed concentration history and the output is the corresponding total emissions history, E(t). When the NFB model is used in inverse mode, E(t) does not agree with observationally based estimates of past emissions. For example, the implied value of E(1990) is 6.2 Gt C yr^-1, which is noticeably less than even the lowest estimate of the sum of industrial and land-use-change emissions¹⁹. It is also much less than the 1990 value in the IS92 scenarios (7.4 Gt C yr^-1). This discrepancy effectively means that the NFB model fails to balance the contemporary carbon budget. The most likely explanation for this is that the model is missing a significant sink ter m. As a consequence, projections using the model in NFB mode will probably overestimate future concentration levels. In IPCC90, it was the NFB model that was used (with an unbalanced 1990 emissions level of 6.2 Gt C yr^-1). For any given emis sions scenario, IPCC90 concentration projections are likely to be too high.

In FB mode, this problem is avoided by including CO₂ fertilization feedback as an additional sink term. The fertilization term is logarithmic in concentration, and its strength is determined by the CO₂ fertilization factor (ß factor). As noted by IPCC92²⁰, the global-scale or ecosystem value of ß is not known, although we do have some idea of its value from small-scale plant growth experiments. We can therefore leave ß as a free parameter in the model. The inverse calculation can then be done with E(1990) specified at any required level, and ß adjusted, within realistic limits, to match this level. Achieving a contemporary balance of the carbon budget in this way means that the FB projections s hould be better than the NFB projections, but there are still important uncertainties.

For our FB projections, we are assuming that the 'missing sink' is explained solely by the CO₂ fertilization effect. Although there is increasing evidence that the CO₂ fertilization effect is the main factor involved in explaining t he missing sink²¹, other factors could also be operating²⁰, such as enhanced plant growth due to nitrogen fertilization or increasing temperature. The strengths of other sinks need not grow at the same rate as the model-projected ch anges in the strength of the CO₂ fertilization sink; temperature feedback may even change sign. Balancing the contemporary carbon budget using these factors as contributors to the missing sink would lead to different CO₂ projections from those given here, although they would probably lie between our FB and NFB values.

Our FB projections are reasonably robust to other uncertainties. If both ocean flux and CO₂ fertilization are adjusted to achieve a contemporary balance (greater ocean flux requiring a smaller fertilization factor and vice versa), then, for an y given emissions scenario, similar concentration projections are obtained for a wide range of combinations of ocean flux and ß. In other words, the projections depend critically on whether or not the contemporary carbon budget is balanced, but the y are relatively insensitive to how this balance is achieved.

CO₂ concentration projections are shown in Fig. 1 for four of the IS92 scenarios. The effect of balancing the carbon budget is to reduce future concentration changes over 1990-2100 by 11-25%. Radiative forcing ch anges for concentration change from C₀ to C are computed using [Delta] Q = 6.3 ln (C/C₀) (ref. 14). Because of this logarithmic dependence, the percentage radiative forcing reductions due to the use of a balanced model are slightly smaller than the concentration changes, 7-22% (See Table 1). Highest percentage reductions occur for the lowest emission case (IS92c).

Methane. Methane concentration changes are calculated using a mass balance model

where b is a units conversion factor, [Tau] is the chemical lifetime and C/ [Tau]_S, is a soil sink term. The present soil sink is estimated to be ~30 Tg CH₄ yr^-l (ref. 22), corresponding to [Tau]_S ~ 150 yr. We assume that [Tau]_S is constant, although there is some evidence to the contrary²⁰. The chemical lifetime of CH₄ is certainly not a constant, changing as the concentration of the hydroxyl radical (OH) changes. These chan ges, in turn, depend mainly all how the concentrations of CH₄, CO and NO_x change. In the present model, [Tau] is assumed to be a function of CH₄ concentration and emission rates of CO, NO_x and VOCs (volatile or ganic compounds), the latter three quantities as given in the IS92 scenarios. The emissions of these non-greenhouse gases act as proxies for their concentrations because of their short lifetimes. The functional dependence for [Tau] involves empirical co nstants that have been evaluated using the results from a number of more detailed chemical modelling experiments, including those carried out as part of the IPCC90 exercise (see ref. 23 for details).

As with CO₂, IPCC90 gave only a single 'best-guess' projection for methane concentration changes under each of the IPCC90 emissions scenarios. This was obtained by averaging the results from a number of individual models. The different models used showed fairly wide departures from the average (shown in ref. 13), indicating the uncertainty that surrounds future concentration projections. The present model has been calibrated using the IPCC90 best-guess projections, and matches these projections well²³.

In calibrating the model against IPCC90 results, the best-ht value of [Tau]₀ was 10.1 yr, compared with the IPCC90 best estimate of 10 +/- 2 yr (ref. 19). Since IPCC90, the best estimate for [Tau]₀ has been revised upward to 11.3 yr (refs 24, 25). We therefore base our projections on this value for [Tau]₀. This increase in the chemical lifetime (by about 10%) means that the concentration projections presented here are higher than they would have been if calculated using the models in IPCC90.

Radiative forcing changes for methane were calculated as in IPCC90¹⁴, allowing for overlap between the CH₄ and N₂O absorption spectra. We also include the contribution to radiative forcing from stratospheric water vapour produced through the oxidation of methane calculated as in ref. 14. This contribution is uncertain, and may be less important than assumed in IPCC90 and here^7,26. It is, however, only a small term, amounting to a maximum of ~0.3 Wm^-2 by 2100. The forcing enhancement due to possible methane-derived ozone increases in the lower stratosphere²⁷ is not included. Radiative forcing information for methane is summarized in Table 1.

Nitrous oxide. Nitrous oxide concentration changes were calculated using a mass balance relationship similar to equation (1), but without a soil sink term. The N₂O case is simpler than for methane, because the lifetime can be taken as a constant (we used 132 yr, following ref. 24, slightly less than the IPCC90 recommended value of 150 yr). Radiative forcing changes were calculated using the formula given by Shine et al.. Changes over 1990-2100 are given in Table 1. N₂O accounts for, at most, 0.3 Wm^-2 of the total forcing to 2100.

Halocarbons. The IS92 halocarbon emissions scenarios are extremely detailed, covering 18 individual compounds. Eight of these are possible new hydrochlorofluorocarbon (HCFC) or hydrofluorocarbon (HFC) substitutes for gases controlled under the Mo ntreal Protocol, all of which currently have zero or near-zero emissions. In contrast, IPCC90 considered only three halocarbons, CFC-11, CFC-12, and HCFC-22, with the latter acting as a proxy for all substitutes. Other halocarbons were considered by sim ply scaling up the radiative forcing of CFC-11 plus CFC-12 by 43% (ref. 14). Now, specific scenarios are given for almost all of the important gases.

Five gases that currently add to the total radiative forcing from halocarbons are, however, not covered by the IS92 scenarios CFC-13, CFC-14, CFC-116, chloroform (CHCl₃) and methylene chloride (CH₂Cl₂). We have added thes e to the scenarios, assuming CFC-13 emissions to be proportional to those for CFC-12, CH₂Cl₂ emissions to be proportional to HCFC-22 (because the former is used in the manufacture of the latter), CHCl₃ emissions to be prop ortional to methyl chloroform (CH₃CCl₃) (because these have similar uses), and emissions of CFC-14 and CFC-116 to be constant.

The latter two compounds, both of which arise mainly through the aluminum manufacturing process, are potentially important. They are not likely to be controlled under the Montreal Protocol (or its extensions) as they do not deplete stratospheric ozone, b ut they are strong greenhouse gases with very long lifetimes (> 500 years). Furthermore, as they do not deplete ozone, they are not affected by ozone-depletion feedback (see below). With time, the radiative forcing importance of these two gases relat ive to most other halocarbons will increase, especially if their emissions increase.

Concentration projections for all halocarbons have been made using the same constant-lifetime mass-balance approach as for N₂O. In certain scenarios, the lifetimes of some of the gases are likely to increase noticeably between 1990 and 2100, s o assuming a constant lifetime will lead to an underestimate of future concentration levels. The net effect of lifetime variations on total halocarbon radiative forcing, however, will be small (< 0.1 Wm^-2). The lifetime values used here ar e the latest estimates, as adopted by IPCC92⁷ and based on refs 24 (mainly) and 28.

Although some of the new lifetimes differ markedly from those given in IPCC90¹⁴, the main differences from IPCC90 arise through use of the London Amendment to the Montreal Protocol as the reference policy background. Under the IPCC90 'Business as Usual' scenario (SA90), which was based on the original protocol, the direct halocarbon radiative forcing change over 1990-2100 was ~ 1.1 Wm^-2. Under the IS92 scenarios, this change reduces to ~ 0.4 Wm^-2. The precise net forcin g changes depend, however, on whether one accepts the reality of the mechanism for stratospheric ozone depletion feedback.

This feedback mechanism^6,7 is the other potentially important new factor in assessing the overall radiative forcing contribution from halocarbons. The hypothesized effect is this. Chlorine- and bromine- containing compounds have been strongly implicated in producing stratospheric ozone loss, especially at high latitudes²⁴. The lower stratospheric cooling resulting from this loss causes a negative radiative forcing of the surface-troposphere system, which acts to offset the positiv e forcing due to the direct greenhouse effects of these gases. Ramaswamy et al.⁶ have used observed ozone losses over 1979-90 to calculate the magnitude of this effect. For the eight gases considered (CFC-11, CFC-12, CFC-113, CFC-114, C FC-115, HCFC-22, methyl chloroform and carbon tetrachloride), 80% of their total, direct, global-mean greenhouse effect is offset.

This process is a potentially important negative feedback on the radiative forcing due to halocarbons. The implications for the IS92 scenarios depend on whether or not the observed ozone losses can be attributed fully to halocarbons. If only part of the ozone loss is due to halocarbons, the negative forcing over 1979-90 would still remain, but the part attributed to and therefore the strength of, the feedback mechanism would be reduced.

We only attempt to quantify this feedback effect in global-mean terms. The negative forcing due to ozone depletion, like ozone depletion itself, is not spatially uniform and is weighted heavily towards mid-to-high latitudes. It is clearly a gross simpli fication to consider only global means, but the results do give quantitative insights into the overall importance of the feedback mechanism.

First, we need to break the process down to an individual gas basis. Once again, this requires some gross simplifying assumptions: specifically, we assume that the global-mean negative forcing is linearly dependent on the global-mean stratospheric ozone loss, which in turn is assumed to be linearly dependent on the halocarbon concentration change and on the number of chlorine atoms in the molecule. (We assume that bromine atoms are five times more effective than chlorine atoms in this regard, qualitative ly in accord with the large ozone depletion potentials of halons compared with CFCs²⁴.) The (negative) forcing due to ozone depletion is then

where a is independent of the species, and N_i and [Delta] C_i, are the number of chlorine atoms in and the concentration change for species i. For the direct greenhouse effect (no feedback), we have

where [alpha] = [partial differential] Q/ [partial differential] C. Values used for [alpha] here are from ref. 14; for gases not covered by ref. 14 we have used ref. 28. For the eight species considered in ref. 6 with observed concentration chang es from ref. 24, and [Delta] Q_oz = 0.8 [Delta] Q_NFB over 1979-90, we obtain from equations (2) and (3) [alpha] = 0.0762 (Wm^-2 per chlorine atom per part per 10¹² (p.p.t.v.)). We therefore evaluate past and futu re halocarbon radiative forcings in two ways, either using equation (3), the standard no-feedback method, or using

to account for the feedback effect, with [Delta] Q_oz given by equation (2).

The overall effect of the feedback is largest in the IPCC90 'Business as Usual' case (SA90), as this has much larger future CFC concentrations than any of the IS92 scenarios. For SA90, over 1990-2100, [Delta] Q_FB] = 0.5 Wm^-2 compare d with [Delta] Q_NFB = 1.1 Wm^-2. The difference is also significant for the total halocarbon forcing up to 1990, [Delta] Q_FB = 0.03 Wm^-2 compared with [Delta] Q_NFB = 0.28 Wm^-2. For all IS9 2 scenarios, however, the effect of the feedback is small, largely because the emissions of chlorine-containing compounds are much reduced from earlier (original Montreal Protocol) projections. The results over 1900-2100 are summarized in Table 1. Note that there are only three distinct halocarbon scenarios, with a, c, f and d, e being identical. For scenarios d and e, [Delta] Q_NFB is noticeably less than [Delta] Q_FB. This is because these are changes from 1990; from a zero conce ntration initial state, [Delta] Q_NFB must always be greater than [Delta] Q_FB. To see how the present result arises, consider [Delta] Q split into contributions from gases with no ozone-depletion feedback (HFCs, CFC14 and CFC116) and those with a feedback effect. The former give a positive forcing change over 1990-2100 which is the same for the NFB and FB cases. For the other gases in IS92d and e most have concentration decreases over 1990-2100. In the NFB case, they therefore con tribute negatively to the forcing, whereas in the FB case (because ozone-depletion feedback counteracts the direct radiative forcing) they have a negligible overall forcing effect. The net result is that [Delta] Q_NFB is less than [Delta] Q_FB. For halocarbon radiative-forcing changes over 1990-2100, differences between scenarios become negligible when the feedback is included (see Table 1). Figure 2

These feedback results have interesting implications for 'comprehensive' or multi-gas approaches to the control of greenhouse gas emissions. In the absence of feedback, the strengthening of the Montreal Protocol implied a considerable reduction in future radiative forcing, by ~0.7 Wm^-2 over 1990-2100 if SA90 results are compared with the IS92 scenarios after allowing for lifetime revisions. In a comprehensive approach, this might be offset against the effects of other greenhouse gases, thereb y requiring weaker restrictions on their emissions. When ozone-depletion feedback is included, the strengthening of the protocol leads to a much smaller effect on 1990-2100 radiative forcing, a reduction of only ~0.1 Wm^-2. This implies less l atitude for easing restrictions on other greenhouse gas emissions if we are to achieve given goals of radiative forcing change.

Sulphate aerosols. The climatic importance of sulphate aerosols derived from man-made SO₂ emissions has only recently been considered seriously^8-10,29. There are three main effects¹⁰, a reduction in incoming short -wave radiation under clear sky conditions^8,30, a possible increase in cloud reffectivity due to sulphate aerosols acting as cloud condensation nuclei^31,33 and a possible increase in cloud lifetimes³³. The first of these (the 'direct effect') has been best quantified by Charlson et al.⁸ who obtain a negative forcing of 1.07 Wm^-2 in the Northern Hemisphere for global SO₂ emissions of 71 Tg S yr^-1 (~64 Tg S yr^-1 in the Northern Hemisphere). The stated uncertainty is a factor of two. This negative forcing is of comparable magnitude to the greenhouse-gas-related positive forcing, ~2 Wm^-2 over the period of significant anthropogenic SO₂ emiss ions, implying that aerosol changes may have noticeably offset the enhanced greenhouse effect in the Northern Hemisphere. The indirect effects through cloud albedo and lifetime changes are currently difficult to quantify reliably because of the uncertain link between changes in mass of sulphate aerosol and changes in number of cloud condensation nuclei³⁰, but they could add appreciably to the direct effect.

Because of the wide uncertainty in the aerosol forcing, we use a range of possible values. As a best-guess value for the relationship between radiative forcing and emissions under the direct effect we use the results of ref. 8 and assume that a Northern Hemisphere emission level of 64 Tg S yr^-1 leads to a corresponding hemispheric radiative forcing of -1.07 Wm^-2. We also assume that global emissions are always partitioned into 90% in the Northern and 10% in the Southern Hemisphere. As the estimate of the direct effect in ref. 8 assumes that emission sources are at present-day elevations, we need to account for changes in these elevations, the 'tall stack effect'. This change has increased the fraction of SO₂ that is ox idized to sulphate, currently ~0.5 (ref. 8), whereas before 1950, it was probably around one-third. We account for this effect by reducing the forcing by 30% before 1950, with the reduction factor decreasing to zero in 1970. For emissions after 1990, we assume forcing to be directly proportional to the anthropogenic component of SO₂ emissions, with a 1990 global emissions level of 75 Tg S yr^-1 (refs 4, 20; see Table 1).

For the indirect effect, we assume that it is real but small, and of current (1990) magnitude equal to 20% of the direct effect. To quantify the link between SO₂ emissions and indirect forcing more generally, we use a relationship suggested by Wigley⁹ which leads to

where E₀ is the initial (natural) emissions level of sulphate precursors (taken as 42 Tg S yr^-1, the midpoint of the range given by IPCC92²⁰), and E is the anthropogenic emissions level. This formulation attempts to accou nt for the nonlinear relationships between changes in initial emissions and final cloud condensation nuclei. For scenarios in which SO₂ emissions grow (a, b, e and f), it leads to a much slower rate of increase after 1990 for the indirect effe ct than for the direct effect. Because the indirect effect is assumed to be a minor component in 1990, and because it becomes relatively less important subsequently, the results for total radiative forcing are insensitive to uncertainties in the [Delta]Q_i formulation.

Finally, to account for uncertainties in the aerosol forcing we consider a range of possibilities, 50-150% of the combined aerosol forcing. As the indirect forcing is assumed to be substantially less than the direct forcing, the implied uncertainty in th e former is much greater than in the latter. For 1990, we deduce a global-mean aerosol forcing range of -0.75 +/- 0.38 Wm^-2. These values are much less than estimated in ref. 10, which gave best guesses of -1 Wm² for both direct an d indirect effects separately, and a best-guess range of -1 Wm^-2 to -2 Wm^-2 for their sum. As shown in the next section, however, and as noted by Wigley²⁹, aerosol forcing effects of this magnitude are almost certainly in compatible with observed hemispheric-mean temperature changes.

Total facing projections. A summary of the radiative forcing projections over 1990-2100 is shown in Fig. 2. This gives two total forcing possibilities for each scenario: a base case which follows the methodology of IPCC90 and ignores CO₂ fertilization feedback, ozone-depletion feedback and aerosol effects; and a 'best-guess' case, in which the two feedbacks are accounted for and in which the best-guess, aerosol forcing is included. Numerical values for the year 2100 are summarized in Table 1.

For comparison, the 2100 values for the IPCC90 scenarios are also shown in Fig. 2.

The most important points to note are the substantially reduced forcing in each best-guess case relative to the corresponding base case; the wide range of forcing possibilities, all corresponding to the same basic 'existing policies' assumption; and the r educed range of uncertainty in the best guess compared with the base case, largely because the high forcing cases have their larger CO₂ effects offset more by aerosol forcing. The reduced forcing compared with what would have been obtained usi ng the IPCC90 methodology arises primarily from our inclusion of CO₂ fertilization feedback and sulphate aerosol forcing. As can be seen from Table 1, the aerosol effect depends on the scenario: IS92b is similar to IS92a, scenarios e and f hav e larger negative aerosol forcing contributions, whereas c and d have small positive contributions. For the best-guess case, total forcing under the c and d scenarios is below the (policy-driven) IPCC90 B scenario. None of the IS92 scenarios, however, a pproaches the very low forcing projections of IPCC90 scenarios C and D.

Temperature changes to 1990

Climate models of the type used here incorporate very simplified representations of the physical processes involved. Nevertheless, they seem to simulate well the responses to external forcing produced by more complex models. They account for the lag eff ect of oceanic thermal inertia, and they allow one to assess the sensitivity of the results to model uncertainties, in particular to uncertainties in the climate sensitivity. The present model additionally differentiates between the hemispheres and so pe rmits the consideration of forcing changes specific to one hemisphere, as is necessary if we are to account for the effects of sulphate aerosol forcing.

The most important model parameter is the climate sensitivity which is specified by the equilibrium global-mean temperature change for a CO₂ doubling. Largely on the basis of general circulation model experiments, this temperature change, [Del ta] T_2x, is thought to lie in the range 1.5-4.5°C, with a 'best-guess' value of 2.5°C (ref. 1). The other parameters affecting the model's response are the mixed-layer depth (h), the oceanic vertical diffusivity (K), the upwelling ra te (w) and the temperature change of high-latitude sinking water relative to the global-mean change (¼). Sensitivity to these parameters is less than to [Delta] T_2x (ref. 35). We assume h = 90 m, K = 1 cm²s^-1, w = 4 m yr^-1 and ¼ = 0.2. Slightly different values were used in IPCC90. The set of values used here is justified more fully in ref. 36. The greatest difference from IPCC is in our choice of ¼ (IPCC90³⁷ used ¼ = 1). Our choice of a lower value is supported by Schlesinger and Jiang³⁸. Smaller ¼ produces larger surface warming and smaller oceanic thermal expansion. The net effect on sea level change is a small reduction.

Figure 3 shows the radiative forcing to 1990 and the corresponding modelled temperature changes, both for the global mean and for the Northern Hemisphere. Modelled temperature changes are shown only for [Delta] T_2x = 2.5°C. For comparison, observed temperature changes, as recently updated by IPCC³⁹, are also shown.

If ozone-depletion feedback and best-guess sulphate aerosol effects are included, then the total forcing change to 1990 is reduced markedly (by ~1 Wm^-2 or 40%) from the IPCC90 estimate¹⁴. This brings the predictions of temperature c hange using [Delta] T_2x = 2.5°C into closer agreement with observations. In the absence of these two factors, the modelled changes with [Delta] T_2x = 2.5°C are appreciably larger than the observed changes; the value of [De lta] T_2x giving the best fit to the 1880-1990 warming is only 1.5°C (compared with 1.4°C in ref. 36, which used slightly different temperature data). For the case with ozone-depletion feedback and best-guess aerosol forcing, the mod elled changes using [Delta] T_2x = 2.5°C are noticeably less than the observed changes. The required sensitivity to obtain the best fit to the observations is [Delta] T_2x = 3.4°C. If only best-guess aerosol forcing is incl uded (without ozone-depletion feedback), our best-guess empirical value for the sensitivity is 3.0°C (compare with ref. 40 for similar assumed forcing). The inclusion of the aerosol forcing flattens the model-projected warming trend over 1950-70, es pecially in the Northern Hemisphere, in qualitative agreement with observations.

Figure 3 also indicates the possible range of values for the negative aerosol forcing. Our largest negative values, which are at the smaller end of the range given by Charlson et al.¹⁰, lead to an overall negative forcing trend over 186 0-1980 in the Northern Hemisphere. As a consequence, the modelled temperature change for the Northern Hemisphere over this period is much less than observed, and, more important, there is a marked difference in warming between the Northern and Southern h emispheres which is not evident in the observations. It is possible that differential warming has been masked by natural variability, but this seems unlikely. On empirical grounds, therefore, and in accord with refs 29 and 40 and three-dimensional model ling results⁸, our choice for the range of likely aerosol forcing is more defensible than the values suggested in ref. 10 which were based on a simpler one-dimensional analysis.

Future temperature and sea level

The climate model also produces results for global-mean thermal expansion of the oceans, a principal component of future sea-level rise. To obtain total sea-level rise, meltwater contributions are added from separate, highly aggregated models representin g small glaciers, the Greenland ice sheet and Antarctica, each driven by the global-mean temperature projection. The ice-melt models are similar to those used by IPCC90 (which we developed in conjunction with Warrick and Oerlemans⁴¹), with imp rovements as described in refs 35 and 42. Ice-melt uncertainties are accounted for by using a range of values for ice-melt model parameters^35,42.

Detailed results for IS92a are given in Fig. 4. This is the scenario most nearly equivalent to SA90⁴. The upper panel shows the sequential reductions in forcing that arise from including a CO₂ fertili zation effect (I to II), accounting for ozone-depletion feedback (II to III) and accounting for the negative forcing due to sulphate aerosols (III to L, M or H). Including CO₂ fertilization to balance the contemporary carbon budget reduces the 1990-2100 forcing change by ~11%, (compared with a maximum of 22% for IS92c and a minimum of 7% for IS92e). The effect of including stratospheric-ozone-depletion feedback on post-1990 forcing is negligible (in all scenarios), whereas aerosols reduce the IS92a forcing by a further 6-19% relative to the base case. Overall, the best-guess forcing for IS92a is ~19% below the base case in the year 2100 (see Table 1 for results for other scenarios).

Low, middle and high temperature and sea-level projections for IS92a are shown in the lower panels of Fig. 4. For temperature, the warming over 1990-2100 is between 1.7°C and 3°C for the best-guess forcing case (compared with a warming range of 2.1°C to 5.0°C for base-case forcing). The corresponding sea-level rise lies between 15 and 90 cm for best-guess forcing and between 22 and 115 cm for base-case forcing. Temperature and sea-level change projections based on best-guess forcing are 20-24% and 21%-30%, respectively, below those for base-case forcing.

The most direct comparison that can be made with IPCC90 is between IS92a and the SA90 results. The latter results are shown (for the year 2100 only) on the right-hand side of Fig. 4. IS92a and SA90 are comparable only in a limited sense, and differences arise because of differing emissions projections, revised methods for calculating the implied concentration and radiative forcing changes, changes in the assumed best-guess climate model parameters, and improvements in the ice-melt models. For temperatu re changes over 1990-2100, the IPCC90 best guess for SA90 (based on [Delta] T_2x = 2.5°C) was 3.3°C. Our corresponding best guess for IS92a is 2.5°C. For sea level, the equivalent projected changes are: SA90, 66 cm; IS92a, 48 cm . These are substantial reductions, but the projected changes in global-mean temperature and sea level are still very large. For comparison, the warming corresponds to a rate roughly five times that observed over the past century, and the sea level rise is at a rate roughly four times that estimated for the past century. The uncertainty ranges for these projections are large, and they are higher in the present results than in IPCC90.

Figure 5 summarizes temperature and sea-level projections for all scenarios for the overall best-guess cases ([Delta] T_2x = 2.5°C, best-guess ice-melt model parameters, and best-guess forcing). IPCC90 best-g uess cases are also shown (2100 values). For the IS92 scenarios, the differences in temperature and sea level projections up to 2050 are fairly small, less than the forcing differences between scenarios (see Fig. 2). For the next 60 years or so, future temperature and sea-level changes are, therefore, relatively insensitive to the uncertainties in future emissions. This means that the IS92a results given in Fig. 4 can be taken as representative of all the emissions scenarios until about 2050.

These projections are subject to other large uncertainties, particularly uncertainties in the climate sensitivity (Fig. 4). Reducing the error bounds on the climate sensitivity should therefore have high priority in greenhouse research^36,43.

Conclusions

The following important results have emerged. First the effects so far of ozone-depletion feedback and sulphate aerosol forcing bring the best-guess empirical estimate of the climate sensitivity ([Delta] T_2x = 3.4°C) into closer agreement with the range of estimates based on general circulation climate model results. If both of these factors are ignored, the best-guess empirical sensitivity is only 1.5°C. The improved agreement here is encouraging, but it should not be over-interpr eted. If other uncertainties are taken into account³⁶, then one finds a very wide range of [Delta] T_2x values to be compatible with observed temperature changes.

This point reinforces our second result, namely that uncertainty in future greenhouse-gas-related temperature and sea-level changes is strongly affected by the climate sensitivity. Reducing uncertainties in this parameter should have high priority, altho ugh the best strategy for this is far from clear. The main reason for our conclusion here is that projections of temperature and sea-level changes out to 2050 have only a small range for any given [Delta] T_2x value, but differ widely according to the chosen value of [Delta] T_2x. The similarity between scenarios should not be interpreted as implying a general insensitivity to future emissions trajectories. Rather, it emphasizes the very long-term nature of the greenhouse problem. A more salutory result is the fact that, beyond the middle of next century, projections for the various scenarios diverge rapidly. Inertia in population growth and socio-economic systems would make it difficult to shift from one scenario to another.

Climate sensitivity is not the only uncertainty that needs to be reduced. In the scenarios in which SO₂ emissions increase markedly (a, b, e and f), an important additional uncertainty arises through not being able to quantify the effect of ae rosols to better than around +/- 50%. In scenario IS92a, for example, this translates to an uncertainty of +/- 6% in total forcing, +/- 8% in temperature change and +/- 9% in sea-level change (over 1990-2100). Uncertainties arising from our inadequate u nderstanding of the global carbon cycle, judged from a comparison of results with and without CO₂ fertilization feedback, are of comparable magnitude.

The projections presented here show only the anthropogenic component of future change, and natural variability will be superimposed on this. Natural variability includes both internally generated variability⁴⁴ and the effects of external forci ng agents⁴⁵, the latter including short-term factors like the recent volcanic eruption of Mount Pinatubo. The expected anthropogenic rates of change, however, are much greater than the corresponding natural rates or change over intervals of 30 years or longer, becoming increasingly so as the interval increases. The reduced rates of warming and sea-level rise that we have obtained compared witn IPCC90 are still four to five times those that have occurred over the past century. Such changes ar e certain to present a considerable challenge for humanity.

T. M. L. Wigley and S. C. B. Raper are at the Climatic Research Unit, School of Environmental Sciences, University of East Anglia, Norwich NR4 7TJ, UK.

ACKNOWLEDGEMENTS.

1. Houghton, J.T., Jenkins, G.J. & Ephraums, J.J. (eds) Climate Change . The IPCC Scientific Assessment (Cambridge Univ. Press, 1990).

2. Houghton, J.T., Callander, B.A. & Varney, S.K. (eds) Climate Change 1992. The Supplementary Report to the IPCC Scientific Assessment (Cambridge Univ. Press, 1992).

3. Pepper, W.J. et al. Emission Scenarios for the IPCC- an update: Background Documentation on Assumptions, Methodology, and Results. (US Environmental Protection Agency, Washington DC, 1992).

4. Leggett, J.A., Pepper, W.J. & Swart, R.J. in Climate Change 1992. The Supplementary Report to the IPCC Scientific Assessment (eds Houghton, J.T., Callander, B.A. & Varney, S.K.) 69-95 (Cambridge Univ. Press, 1992).

5. Mitchell, J.F.B. & Gregory, J.M. in Climate Change 1992. The Supplementary Report to the IPCC Scientific Assessment (eds Houghton, J.T., Callander, B.A. & Varney, S.K.) 171-175 (Cambridge Univ. Press, 1992).

6. Ramaswamy, V., Schwarzkopf, M.D. & Shine, K.P. Nature 355 810-812 (1992)

7. Isaksen, I., Ramaswamy, V., Rodhe, H. & Wigley, T.M.L. in Climate Change 1992. The Supplementary Report to the IPCC Scientific Assessment (eds Houghton, J.T., Callander, B.A. & Varney, S.K.) 47-67 (Cambridge Univ Press. 1992).

8. Charlson, R.J., Langner, J., Rodhe, H., Leovy, C.B. & Warren, S.G. Tellus AB 43, 152-163 (1991).

9. Wigley, T.M.L. Nature 349, 503-506 (1991).

10. Charlson, R.J. et al. Science 255, 423-430 (1992).

11. Nordhaus,W.D. & Yohe, G.W. in Changing Climate 87-153 (National Academy Press, Washington DC, 1983).

12. Edmonds, J.A., Reilly, J.M., Gardner, R.H. & Brenkert, A. Uncertainty in Future Global Energy Use and Fossil Fuel CO₂ Emissions 1975 to 2075, TRO36 (US Dept of Energy, Carbon Dioxide Research Division, Washington DC, 1986).

13. Wigley, T.M.L., Holt, T. & Raper, S.C.B. STUGE (an Interactive Greenhouse Model): User's Manual (Climatic Research Unit, Norwich, 1991).

14. Shine, K.P., Derwent, R.G., Wuebbles, D.J. & Morcrette, J.-J. in Climate Change 1992. The Supplementary Report to the IPCC Scientific Assessment (eds Houghton, J.T., Jenkins, G.J. & Ephraums, J.J.) 41-68 (Cambridge Univ. Press, 1990) .

15. Harvey, L.D.D. Clim. Change 15, 343-381 (1989).

16. Wigley, T.M.L. Global biogeochem. Cycles 5, 373-382 (1991).

17. Maier-Reimer, E. & Hasselmann, K. Clim. Dynam. 2, 63-90 (1987).

18. Wigley, T.M.L. Tellus (in the press).

19. Watson, R.T., Rodhe, H., Oeschger, H. & Siegenthaler, U. in Climate Change. The IPCC Scientific Assessment (eds Houghton, J.T., Jenkins, G.J. & Ephraums, J.J.) 1-40 (Cambridge Univ. Press, 1990).

20. Watson, R.T., Meira Filho, L.G., Sanhueza, E. & Janetos, A. in Climate Change 1992. The supplementary Report to the IPCC Scientfic Assessment (eds Houghton, J.T., Callendar, B.A. & Varney, S.K.) 25-46 (Cambridge Univ. Press, 1992).

21. Enting, I. & Mansbridge, J. Tellus B43, 156-170 (1991).

22. Born, M., Door, H. & Levin, I. Tellus B42, 1-8 (1990).

23. Osborn, T.J. & Wigley, T.M.L. Clim. Dynam. (submitted).

24. Watson, R.T. & Albritton, D.L. Scientific Assessment of Ozone Depletion: 1991 (World Meteorological Organization, Geneva, in the press).

25. Vahjiani, G. & Ravishankara, A. Nature 350, 406-409 (1991).

26. Lelieveld, J. & Crutzen, P. Nature 355, 339 -342 (1992).

27. Lal, M. & Holt, T. Clim. Res. 1, 85-95 (1991).

28. Ramanathan, V. Cicerone, R.J., Singh, H.B. & Kiehl, J.T. J. geophys. Res. 90, 5547-5566 (1985).

29. Wigley, T.M.L. Nature 339, 356-367 (1989).

30. Charlson, R.J., Langner, J. & Rodhe, H. Nature 348, 22 (1990).

31. Twomey, S.A., Peipgrass, M. & Wolfe, T.L. Tellus B 36, 356-366 (1984).

32. Charlson, R.J., Lovelock, J.E., Andraea, M.O. & Warren, S.G. Nature 326, 655-661 (1987).

33. Albrecht, B.A. Science 245, 1227-1230 (1989).

34. Wigley, T.M.L. & Raper, S.C.B. Nature 339, 127-131 (1987).

35. Wigley, T.M.L. & Raper, S.C.B. in Climate and Sea Level Change: Observations, Projections and Implications (eds Warrick, R.A., Barrow, E.M. & Wigley, T.M.L.) (Cambridge Univ. Press, in the press).

36. Wigley, T.M.L. & Raper, S.C.B. in Climate Change: Science, Impacts and Policy (eds Jäger, J. & Ferguson, H.L.) 231-242 (Cambridge Univ. Press, 1991).

37. Bretherton, F.P., Bryan, K. & Woods, J.D. in Climate Change, The IPCC Scientific Assessment (eds Houghton, J.T., Jenkins, G.J. & Ephraums, J.J.) 173-193 (Cambridge Univ. Press, 1990).

38. Schlesinger, M.E. & Jiang, X. Nature 353, 219-221 (1991).

39. Folland, C.K. et al. Climate Change 1992. The Supplementary Report to the IPCC Scientific Assessment (eds Houghton, J.T., Callendar, B.A. & Varney, S.K.) 135-170 (Cambridge Univ. Press, 1992).

40. Schlesinger, M.E., Jiang, X. & Charlson, R.J. in Proc. Conf. on Global Change: Its Mitigation through Improved Production and Use of Energy (eds Allred, J.C. & Nichols, A.S.) (American Institute of Physics, in the press).

41. Warrick, R.A. & Oerlemans, H. in Climate Change, The IPCC Scientific Assessment (eds Houghton, J.T., Jenkins, G.J. & Ephraums, J.J.) 257-281 (Cambridge Univ. Press, 1990).

42. Raper, S.C.B., Wigley, T.M.L. & Warrick, R.A. in Proc. SCOPE Workshop on Rising Sea Level and Subsiding Coastal Areas (ed. Milliman, J.D.) (Wiley, Chichester, in the press).

43. Hansen, J.E., Rossow, W. & Fung, I. Issues Sci. Technol. 7, 62-69 (1990).

44. Wigley, T.M.L. & Raper, S.C.B. Nature 344, 324-327 (1990).

45. Hansen, J.L. & Lacis, A.A. Nature 346, 713-719 (1990).

46. Möller, D. Atmos. Envir. 18, 19-27 (1984).

47. Smith, F.B. Q. JI R. met. Soc. 117, 657-683 (1991).