The UN Framework Convention on Climate Change[45] (UNFCCC) signed at Rio in 1992 makes staying within ecological limits its "ultimate objective"[46] with greenhouse gas concentrations to be stabilized "at a level that would prevent dangerous anthropogenic [human made] interference with the climate system". This is objective is to be achieved "within a timeframe sufficient to allow ecosystems to adapt naturally to climate change, to ensure that food production is not threatened and to enable economic development to proceed in a sustainable manner". Further, the climate convention also requires Parties to "take precautionary measures to anticipate, prevent or minimize the causes of climate change and mitigate its adverse effects" and that "Where there are threats of serious or irreversible damage, lack of full scientific certainty should not be used as a reason for postponing such measures, taking into account that policies and measures to deal with climate change should be cost-effective so as to ensure global benefits at the lowest possible cost."[47]
These elements of the Convention are the essential policy context for an examination of ecological limits that should guide international policy on climate change. Both the rate and the magnitude of climate change need to be addressed if the objectives of the climate convention are to be met. This point is made clear by two of the key findings of the IPCC Second Assessment Report:
"In all cases the average rate of warming would probably be greater than any seen in the last 10,000 years..." [48]
"Most systems are sensitive to climate change. Natural ecological systems, socio-economic systems, and human health are all sensitive to both the magnitude and the rate of climate change."[49]
It is clear from a review of the entire IPCC Second Assessment Report that climate change poses a significant threat to sustainable development , particularly for developing countries who are likely to be much more adversely affected than developed countries by the climate change projected for an equivalent doubling of CO2[50].
Whilst the IPCC found that "existing studies show ... global agricultural production could be maintained relative to baseline production in the face of climate change projected under doubled equivalent CO2 equilibrium conditions", it also warned that:
"there may be increased risk of hunger and famine in some locations; many of the world's poorest people - particularly those living in subtropical and tropical areas and dependent on isolated agricultural systems in semi-arid and arid regions - are most at risk of increased hunger"[51]
Regions particularly at risk included sub-Saharan Africa; South, East, and Southeast Asia; and tropical areas of Latin America, as well as some Pacific island nations[52]. Rosenzweig and Parry calculated, under highly optimistic assumptions, that 60-350 million more people could be at risk of hunger as a consequence of climate change and that these would be predominantly in developing countries[53]. It is clear, on the basis of the evidence reviewed in the IPCC Second Assessment Report, that there is a dangerous threat to food production in a number of developing countries.
In this section we review the background behind the issue of ecological limits and advances in scientific understanding of how these limits may be applied to international climate policy.
In 1990, on the basis of scientific knowledge available before the IPCC First Assessment Report was concluded, the WMO/ICSU/UNEP Advisory Group on Greenhouse Gases (AGGG) produced an analysis of "targets and indicators" for climate change[54]. This work focused on developing quantitative targets for long term risk management which could be used as the basis for short term emission targets[55]. The purposes of these indicators was to set limits to rates and total amounts of temperature rise and sea-level rise, on the basis of known behaviour of ecosystems as a guide to policy in order to protect both human and natural ecosystems.
The indicators of climate change identified by this group were sea-level rise, committed (or equilibrium) global mean temperature and CO2 concentrations. Targets were identified for the climate change indicators that incorporated different levels of risk.
Temperature and sea-level rise targets for the lowest level of risk were:
- Maximum 1.0o increase above pre-industrial levels. Increases beyond this "may elicit rapid, unpredictable and non-linear responses that could lead to extensive ecosystem damage"[56].
* Maximum rate of warming of 0.1oC/decade. The rate of warming needs to be below this to ensure that most ecosystems can adapt. This would lead to some damage, however higher levels would lead to rapidly rising risk.
* Maximum rate of sea-level rise of 20mm/decade. This "would permit the vast majority of vulnerable ecosystems, such as natural wetlands and coral reefs to adapt with rates beyond this leading to rapidly rising ecosystem damage".
Targets for the higher levels of risk were:
* 2.0oC increase above pre-industrial levels. This is "an upper limit beyond which the risks of grave damage to ecosystems, and of non-linear responses, are expected to increase rapidly".
* Maximum 50 cm sea-level increase above 1990 global mean sea-level. This could "prevent the complete destruction of island nations, but would entail large increases in the societal and ecological damage caused by storms".
At this high level of risk Vellinga and Swart argue that there will be large impacts on many regions and that there is a high risk of climate instabilities and of strong feedbacks:
"We must expect that in many places in the world there will be a crisis in the world food supply and ecosystems and the corresponding disruption of socio-economic systems and a loss of several islands"[57].
Several important points need to be borne in mind when considering these targets and the different levels of risk associated with them:
(i) Long term climate commitments. The maximum sea-level and global mean temperature commitments are not limited to a specific time horizon i.e. 2100. This indicates that climate policy should be set with long term changes in mind. Once a change in sea-level rise is actually observed it may be irreversible for practical purposes and will almost certainly be associated with much larger change in the longer term.
(ii) Risk of large, local impacts and increase in the frequency of extreme events. The targets are for global mean changes. Regional changes of temperature and sea-level may be quite different. Mountain ecosystems and boreal forests, for example, are likely to experience more rapid change than the global average. In higher latitudes projected temperature changes are likely to be much higher than the global mean average. Global mean averages do not capture the effect of changes in the frequency or character of extreme events (i.e. storm, droughts, floods) or in seasonality patterns. Thus global mean averages are therefore only crude surrogates for indicators of damage in the most vulnerable places.
(iii) Precautionary principle and equity. Policy needs to be based on the precautionary principle and on equity. A 20 cm sea-level rise, for example, may not be problem for some countries but may be disastrous for others. Targets therefore need to be chosen that can guide policies to prevent dangerous climate change in the most vulnerable places.
Scientific work published since 1990 tends to support the lowest risk targets identified by the WMO/ICSU/UNEP Advisory Group. Whilst it is not possible here to fully review this work some examples are given below.
In 1995 the UK Meteorological Office Hadley Centre noted:
"The global mean rate of change is predicted to be a little above 0.2.C/decade in the early part of the next century; approximately twice the rate of change that many of the more sensitive ecosystems are thought to be capable of surviving"[58]
It further pointed out that such rates:
"are likely to exceed the adaptive capacity of many ecosystems. Indeed the IPCC concluded that 0.1oC/decade was probably the maximum that many ecosystems could tolerate."
The IPCC has confirmed that ecosystems and species are vulnerable to both the rate and extent of climate change and that rapid climate change is likely to lead to loss of biodiversity:
"Ecosystems contain the Earth's entire reservoir of genetic and species diversity and provide many goods and services critical to individuals and societies"
"These systems and the functions they provide are sensitive to the rate and extent of changes in climate."....
"there will likely be reductions in biodiversity and in the goods and service that ecosystems provide society"[59]
Quantitatively, the IPCC also found major changes in the earth's forests are projected for only a 1oC increase in global mean temperature leading to very large changes and the possible disappearance of entire forest types:
"Models project that a sustained increase of 1oC in global mean temperature is sufficient to cause changes in regional climates that will affect the growth and regeneration capacity of forests in many regions"
"A substantial fraction (a global average of one-third, varying by region from one-seventh to two-thirds) of the existing forested area of the world will undergo major changes in broad vegetation types with the greatest changes occurring in high latitudes and the least in the tropics."
"Climate change is expected to occur at a rapid rate relative to the speed at which forest species grow, reproduce, and reestablish themselves. ... Therefore, the species composition of forests is likely to change; entire forest types may disappear, while new assemblages of species and hence new ecosystems may be established."[60]
A sea-level rise of the order of 50 cm would lead to major impacts on many small islands. Other impacts include a dramatic increase in the number of people at risk of flooding and significant impacts on rice production in Asia:
"The present number of people at risk will double if sea level rises 50 cm (92 million people/year) and almost triple if it rises 1 meter (118 million people/year).
"Approximately 85% of the world's rice production takes place in South, Southeast, and East Asia. About 10% of this production is located in areas that are considered to be vulnerable to sea-level rise, thereby endangering the food supply of more than 200 million people."[61]
In relation to the potential for rapid climate change to lead to climate instabilities recent work by Stocker and Schmittner[62] has shown that the rate of increase of greenhouse gas concentrations could have a major impact on the thermohaline circulation system of the North Atlantic ocean. They have found that the thermohaline circulation is sensitive not only to the final concentration of CO2 in the atmosphere but also to its rate of change. Using a climate model with a climate sensitivity of 3.7oC it is found that an increase in CO2 of 1% per year to over 700 ppmv equivalent could lead to a permanent shut down of the thermohaline circulation. A slower increase to the same level slows down the thermohaline circulation.
A 1% per year increase in CO2 concentration approximates the rate increase of CO2 equivalent greenhouse gas concentration projected over the next century and would produce a rate of increase in temperature of about 0.2oC/decade[63].
The existence and strength of the thermohaline system contributes to the mild climates of north-west Europe. It also plays a significant role in the global carbon cycle - a strong thermohaline circulation carries large amounts of CO2 to the deep oceans. A weakening of the thermohaline circulation would lead to CO2 concentrations increasing faster and would lead to some very significant regional climate changes. (See discussion below in section 3.5.5). A shutdown in the thermohaline circulation could have quite dramatic and adverse effects on the climate of Europe in particular.
Overall assessments of the impacts of climate change point towards a high level of vulnerability for many natural and some human systems to rapid climate change. The most vulnerable systems are likely to be irreversibly damaged by sustained rates of temperature increase at or above 0.1oC/decade. Further there is a significant risk of feedbacks amplifying the changes and the potential for major climate instabilities.
In considering the application of indicators and targets for climate change the question of how these compare with the projected effects of the IPCC emission scenarios arises.
Using the IPCC's central estimate of emissions (IS92a) over the next century, assuming the IPCC's `best-estimate' value of climate sensitivity and including the effects of future increases in aerosol emissions(see below), global mean surface temperature relative to 1990 is projected to increase 2.0oC by 2100. With aerosol concentration held constant at 1990 levels the best estimate is 2.4oC by 2100. In other words, by 2100 the IPCC best-estimate is for a global mean temperature increase of 2.5-2.9oC above pre-industrial levels. Further, temperatures would continue to increase for some time even if atmospheric CO2 levels were stabilized in 2100. Rates of temperature increase over the next century would be in the range 0.2-0.3oC/decade.
The IPCC's `best-estimate' of sea-level rise from 1990 to 2100 based on the IS92a scenario with constant aerosol emissions (see below) over this period is 55 cm (with increasing aerosol emissions it is 49 cm). The full range of uncertainty in the IPCC estimates for constant aerosol emissions is 23-96 cm. Sea-level would not stop rising in 2100, even if concentrations were stabilized. Owing to the inertia of the climate system and oceans in particular, it would "continue at a scarcely unabated rate for many centuries after concentration stabilization."[64]
Using the same emission assumptions but with a climate sensitivity of 3.5 oC the projected mean global temperature increase would be in the range of 3.1-3.6 oC above the pre-industrial global mean temperature and the sea-level rise would be 54-60 cm above 1990 levels.
Table 11 shows the long term equilibrium warming commitment that would result for the IS92 scenarios ranges from 3.3oC to 8.4oC depending on the assumptions made.
In terms of the greenhouse gas concentration over the next century the IS92a scenario projects a doubling in CO2 equivalent terms above pre-industrial levels in the decade 2030-2040. If aerosol effects are accounted for (see section 4.5.3 below) then this CO2 equivalent doubling occurs in the decade 2050-2060, depending on the assumptions made. Actual CO2 levels would double around 2060. From a policy perspective based on the precautionary principle, the effective CO2 doubling, not counting aerosol effects, is the most salient point (see section 4.5.3 below).
Projections of the transient effects over the next century of an emission scenario similar to the IS92a scenario using the IMAGE integrated assessment model, which has a climate sensitivity of 2.4oC for CO2 doubling, show large damages[65]. The IMAGE baseline scenario emits 1691 GtC in the period 1990-2100, resulting in a CO2 concentration in 2100 of 737 ppmv more than double the 1990 level. Some of the major results include:
* The rate of increase of impacts on vegetation and agriculture could be larger in the first half of the next century than in the second half.
* By 2100 the global average surface temperature would increase by around 2.8oC from 1990 levels, an increase above pre-industrial levels of around 3.3oC. Temperature increases would be higher in higher latitudes at around 4oC.
* 32% of the area currently used for maize production is projected to experience decreasing yield (Note that 15% of this area is projected to experience increasing yield).
* The area of natural vegetation under threat is very large. By 2100 climate change will threaten terrestrial vegetation type over 41% of the land surface area.
* Sea-level rise would be around 42 cm across the same time period - and would still be rising exponentially at the end of the century.
It is clear from these results that both the rate and magnitude of temperature and sea-level rise will exceed even the highest indicators described above. If the indicators are accurate then the projected emissions of greenhouse gases over the next century risk causing grave damage to ecosystems and forcing non-linear climate responses. In other words the environmental consequences of the IS92a scenario would be enormous. Table 11 Effects of IPCC IS92 emission scenarios
Zero aerosol emission in 2100 |
High aerosol emissions and effects in 2100 |
Sea-level and temperature in 2100 | ||||||
Scenario |
CO2 equiv. conc. ppmv |
Total Radiative Forcing to 1765-2100 W/m2 |
Equilibrium warming commitment oC |
CO2 equiv. conc. ppmv |
Total Radiative Forcing to 1765-2100 W/m2 |
Equilibrium warming commitment oC |
Sea-level rise 1990-2100 cm |
Global mean temperature increase in 2100 oC |
IS92a |
1051 |
8.4 |
6.7 |
770 |
6.4 |
5.1 |
64 |
2.5 |
IS92b |
1017 |
8.2 |
6.5 |
751 |
6.2 |
5.0 |
63 |
2.4 |
IS92c |
632 |
5.2 |
4.1 |
534 |
4.1 |
3.3 |
48 |
1.3 |
IS92d |
729 |
6.1 |
4.9 |
604 |
4.9 |
3.9 |
52 |
1.6 |
IS92e |
1477 |
10.5 |
8.4 |
980 |
7.9 |
6.4 |
74 |
2.2 |
IS92f |
1294 |
9.7 |
7.8 |
910 |
7.5 |
6.0 |
70 |
2.0 |
The equilibrium warming commitment is calculated using a climate sensitivity of 3.5oC. The zero aerosol emissions case shows the radiative forcing if all aerosol emissions ceased. The high aerosol emissions case is the high indirect sulphate case of the IPCC SAR WGI which was used to calculate best-estimate projections of future temperature and sea-level rise, but with the radiative forcing computed from 1765. The sea-level and temperature projections for 2100 are for a climate sensitivity of 2.5oC and for zero aerosol emission changes from 1990 (i.e. constant aerosol emissions). See Table A.4 and A.5 of Raper et al (1996)[66].
As part of the 1994 IPCC Special Report a carbon cycle model intercomparison process was conducted using standardized atmospheric CO2 stabilization scenarios. Based on standard concentration profiles over time (Figure 8) the carbon cycle models were used to calculate backwards (inverse modelling) to arrive at emission profiles that corresponded to the atmospheric stabilization profile[67]. Five levels of CO2 stabilization were chosen - 350, 450, 550, 650 and 750 ppmv with the year of stabilization varying for each scenario ranging from 2100 for 450 ppmv to 2250 for 750 ppmv. The carbon cycle calculations were reviewed in the IPCC's Second Assessment Report in 1995, with the addition of further level at 1,000 ppmv.
Partly as a consequence of the IPCC exercise, and in the context of the climate convention negotiations, some countries have raised the idea of a long term concentration target rather than ecological targets as described above. Governments such as France and the USA have talked, formally or informally, of 550 ppmv for example as a long term target. Similarly many economic modelling exercises have focussed on 550 ppmv of CO2.
There is a sense then in which the IPCC stabilization levels may have been perceived as bracketing an acceptable range with those in the middle, 550 or 650 ppmv, becoming by default the compromise[68]. This tendency has been reinforced to some degree by the omission of the 350 ppmv scenario from the IPCC WGI SAR and the inclusion of a 1,000 ppmv scenario[69]. This is not however what the IPCC intended at all.
In this context then it is useful to review the efficacy of concentration targets as tools for climate policy and in particular the meeting of ecological objectives.
In the absence of policy action CO2 levels would approach 650-750 ppmv by 2100, with equivalent CO2 levels being much higher, over 1,000 ppmv[70]
Table 1[1]
. Table 12 shows the calculated long term implications of these CO2 stabilization scenarios for global average temperature and sea-level rise. It should be noted that the sea-level continues to rise well after the point at which atmospheric CO2 is stabilized. Figure 14 shows the corresponding emissions calculated by the mid-range of the carbon cycle models for each of the concentration stabilization profiles for the same time period. For comparative purposes the IS92a scenario is shown. Figure 8 IPCC CO2 concentration stabilization scenarios
This figure shows the prescribed CO2 concentration profile assumed in order to calculate the emissions profiles and hence the `carbon budget' calculations over the period to 2300 for the IPCC stabilization scenarios. The S350 scenario stabilized CO2 at 350 ppmv in 2150, S450 at 450 ppmv in 2100, S550 at 550 ppmv in 2150, S650 at 650 ppmv in 2200 and S750 at 750 ppmv in 2250. Source: CSIRO
Figure 9 shows the calculated rates of global average temperature change for CO2 stabilization scenarios ranging from 350 to 550 ppmv. This graph demonstrates that the emissions corresponding to the achievement of the lower concentration stabilization level lower the warming rate the most rapidly.
Alcamo and Kreileman[71] have calculated, using the IMAGE model, the effects of stabilizing at 350, 450, 550, and 650 ppmv of CO2. Their calculations include other greenhouse gases, which means that the equivalent CO2 level is higher than the nominal CO2 stabilization levels. They find that stabilization above 450 ppmv will have large impacts. Below this level impacts will be significantly lower, however there will still be some residual effects. For stabilization at 450 ppmv or higher global temperatures are steadily increasing, as are impacts on natural vegetation and crop production, up to at least 2100. Sea-level rise continues to increase after 2100 in all cases. For stabilization below 450 ppmv (i.e. 350 ppmv) the impacts on natural vegetation and crop production stabilize before the middle of the century, although sea-level continues to rise. In summary the results of these scenarios are:
* Under the 350 ppmv scenario sea-level rise is 24 cm in 2100 and the temperature increase is 0.7oC (1.2oC above pre-industrial levels). The area of current maize production with decreasing yield is 16% in 2100 and area of natural vegetation threatened by climate change is 15%.
* For the 450 ppmv scenario the temperature increase is 1.7oC above pre-industrial levels, sea-level rise is 29 cm, the threat to natural vegetation 23% and area of current maize production with decreasing yield is 21%.
* For the 550 ppmv scenario the temperature increase is 2.2oC above pre-industrial levels, sea-level rise is 33 cm, the threat to natural vegetation 28% and area of current maize production with decreasing yield is 25%.
* For the 650 ppmv scenario the temperature increase is 2.5oC above pre-industrial levels, sea-level rise is 36 cm, the threat to natural vegetation 31% and area of current maize production with decreasing yield is 26%.
The long term sea-level rise in each case would be some 2-3 times the increase to 2100.
One of the key results from this work is the finding that with each of the stabilization scenarios there is a much more rapid increase in temperature and of some climate impacts in the first half of the next century than before or after[72]. This implies that many of the projected impacts of future emissions are only avoidable of action is taken early. From a policy perspective this reinforces the urgency of early emission reductions to slow the rate of warming.
The work of Alcamo and Kreileman also sheds light on the question of the efficacy of concentration targets to meet the ultimate objective of the climate convention. Overall, they find that "stabilizing greenhouse gases in the atmosphere does not necessarily provide a high level of climate protection". This conclusion is linked in part to the fact that the concentration targets do not explicitly include an objective of limiting the rate of change.
Table 12 Temperature and Sea-level rise implications of IPCC CO2 Stabilization Scenarios.
IPCC CO2 scenario |
Year of CO2 stabilization |
Temperature increase above 1990[73] |
oC
Sea-Level Rise[74] |
( cm)
Above 1990 levels
This table is a compilation of results and shows the effects of the CO2 stabilization levels on global average temperature and sea-level in the longer term. The temperature increase incorporates the effects of other greenhouse gases by assuming that their post 1990 radiative forcing increase is 23% (see below) of that due to CO2. It includes the offsetting effects of the pre-1990 aerosol forcing. The sea-level rise calculations are from a more recent source and correspond to the "best- estimate" climate and sea-level rise parameters in the 1995 IPCC Second Assessment Report. It is to be noted that unlike the temperature estimates above they do not include post-1990 changes in non-CO2 greenhouse gas forcing. Despite this the sea-level rise estimates reported here are higher than those previously calculated using earlier (1990, 1992) IPCC best-estimate assumptions for sea-level rise. These former estimates are included in brackets.
The timing and level of emission reductions needs to be driven by the precautionary principle and scientific considerations. Because of the complexity of the climate system and its inertia (slowness to react to increases or reductions in greenhouse gas emissions) emission reduction pathways need to take account simultaneously of several climate protection goals, which together have a chance of avoiding extensive ecosystem damage. This idea is reflected, for example in the proposal of the Alliance of Small Island States that the "guiding objective" of the Kyoto Protocol shall be to:
"ensure that global mean sea-level rise resulting from climate change does not exceed 20 centimetres and that the global average temperature does not exceed 2 degrees Celsius above the pre-industrial level"[75].
Apart from minimizing the rate of climate change, the design of emission reduction pathways needs to avoid imposing very large emission reduction rates on future generations (which would be the result if action to reduce emissions is not begun immediately). Also, the risk of "surprises" and catastrophes should be reduced by taking action early.
The implications of a set of multiple climate constraints can be calculated using the "safe emissions corridor" concept developed by the Dutch IMAGE Climate Modeling team[76]. In essence this approach attempts to provide a tool for answering questions directly relevant to the climate negotiations:
* Which short term emission limits (to 2010) would be needed for the world to stay within both short and long term ecological limits of climate change throughout the next century provided no unforeseen catastrophes occurred?
* What are the allowable emissions from Annex I countries between now and 2010 still enable future generations to meet the defined ecological and economic limits, without imposing very large rates of economic adjustment on future generations?
The size of the safe "emissions corridor" in the period to 2010 depends on the assumed climate constraints. The more stringent the constraints the lower is the top of the corridor. The bottom of the corridor is determined by the assumed maximum rate of emission reductions.
Of quite fundamental concern to policy is the implications of the location in the emission corridor in the period to 2010 for allowed emissions after 2010. If emissions are at the top of a corridor in 2010, then it is likely that the band of allowable emissions after 2010 will be very small. In other words, high emissions in the period to 2010 reduce the flexibility for policy post-2010. This fact would tend to place emphasis on emissions being no higher than around the middle of a corridor in order to protect options for future generations.
Four key limits were chosen - three ecological and one economic. The ecological limits are similar to those defined by the WMO/ICSU/UNEP AGGG - total temperature change to 2100, the decadal average rate of global temperature change and sea-level rise to 2100[77]. The economic limit chosen was a maximum annual rate of reduction of CO2, reflecting possible economic constraints.
In addition, the assumption was made that developing country emissions would not be constrained by international climate policy over the period to 2010, by which time their total emissions would be in the range 5.5-7.0 GtC/year (CO2 equivalent). This is consistent with the Berlin Mandate under which negotiations are occurring aimed at securing legally binding emission reductions for industrialized countries at the Third Conference of the Parties to the Climate Convention (COP-3)[78]. For comparison, indusrtialised greenhouse gas emissions in 1990 totalled around 6 GtC (CO2 equivalent). Figure 9 Rates of temperature change for CO2 scenarios
The calculations were made with a climate sensitivity of 3.5oC[79]. In these calculations the radiative forcing of the non-CO2 gases (taking into account aerosols) in 2100 was 47% for the 350 ppmv scenario, 20% for the 450 ppmv scenario, 13% for the 550 ppmv scenario and 14% for the IS92a scenario. For the 350, 450 and 550 scenarios CH4 and N2O emissions were assumed to be approximately constant at 1990 levels. Fluorocarbon emissions were phased out in the 350 and 450 ppmv scenarios and continued at 500kt/yr. in 550 case. Sulphur emissions were correlated with fossil carbon emissions. The IS92a scenario was modified by lowering its sulphur intensity and modifying the halocarbon emissions to account for controls on CFCs and HCFCs. See section 5.2.2 for a discussion of the `carbon budget' scenario.
The most stringent ecological limits studied were for a maximum rate of temperature increase of 0.1oC per decade, a maximum temperature increase and sea-level rise by 2100 of 1oC and 20 cm (respectively) above 1990 levels. Such limits may enable many, but not all, ecosystems to adapt, and may limit to some degree the danger from sea-level rise over the next century. In this scenario the allowed maximum rate of emission reductions after 2010 was assumed to be 2% per year. In this case, to keep emissions below the top of the safe emissions corridor, Annex I[80] emissions must be cut by 48-62% from their 1990 levels by 2010[81].
Table 13 below shows the implications for industrialized country (Annex I countries) greenhouse gas emissions relative to 1990 levels of these constraints for two developing country emission growth scenarios, one medium high and the other relatively low. Emissions in 2005 need to be some 23-27% below 1990 levels in 2005 and some 45-54% below in 2010. It should be noted that a sea-level rise to 2100 of 20 cm implies a longer term sea-level rise of some 2-3 times this level.
In one of the least stringent cases studied by the IMAGE team a maximum rate of temperature increase of 0.2oC per decade, a maximum temperature increase and sea-level rise by 2100 of 2oC and 40 cm respectively, were applied to the climate system. Breaching these limits would have a high risk of leading to irreversible ecological damage and major problems from sea-level rise. The allowed maximum rate of emission reductions after 2010 was assumed to be 4% per year. To reach the middle of this corridor, Annex I emissions must be cut by 19-46% from their 1990 levels by 2010.
The maximum permissible emissions limits over the period to 2010 that would allow for the possibility of meeting strong climate protection targets in the future whilst ensuring CO2 emissions do not have to be reduced at more than 2% per year, amount to a reduction by Annex 1 countries of at least 37% from 1990 levels by the year 2010. Emissions in this range would be at the top of the strong target and in the middle of the weakest climate target studied. Being at the top of the "safe emissions corridor" however would provide relatively few options for future generations.
If such reductions are not met, then it may still be possible to meet ecological limits however this would require much more rapid global reductions of emissions. Whilst being more difficult to achieve it could create major problems for developing countries in the next century. Such problems would be a direct consequence of insufficient early action by Annex 1 countries
The "safe emissions corridor" analysis has significant implications for climate policy. Parties will have to agree on emission reductions of this magnitude if they are to avoid both exceeding ecological limits of climate change and imposing very large emission reduction rates on future generations.
A simplified analysis of this can be seen in Figure 10 which shows the allowed Annex 1 emissions to 2010 assuming that non-Annex 1 ( developing countries) emissions grow without restraint during this period[82] for CO2 stabilization scenarios. Superimposed on this is the emission profile corresponding to the top of the safe emissions corridor for the ecological limits calculated by the IMAGE model.
It is clear from the IMAGE analyses that if strong ecological goals are to be met the minimum and urgent first step required to move the world toward protecting the climate is a cut in industrialized country CO2 emissions of at least 20 per cent below 1990 levels by the year 2005. The European Environment Agency has found, based on the IMAGE safe emissions corridor analysis, that, depending on action to stop deforestation globally and on the transfer of clean and renewable technologies to developing countries, a reduction in emission of around 30-55 per cent by 2010 for industrialized countries will be needed to avoid ecologically dangerous climate changes. Table 13 IMAGE "Safe Emissions Corridor" - Annex I emission reductions for 20 cm sea-level rise target[83]
Annex I emission reductions relative to 1990 levels for the top of a "safe emissions corridor" for a 20 cm sea-level rise target. |
2005 |
2010 |
Medium to high growth in developing countries (IPCC scenario IS92a) |
27% |
54% |
Low growth in developing countries (IPCC scenario IS92d) |
23% |
45% |
The fundamental environmental constraint for this corridor was the 20 cm maximum sea-level rise by 2100. Such a sea-level rise limit implies a longer term rise over the next several centuries of 40-60 cm given the inertia of the oceans and the time it will take for warming to spread throughout the entire world ocean. Temperature limits up to 2oC by 2100 above the pre-industrial global mean temperature produced the same corridor in this case as the 1oC limit i.e. the sea-level rise constraint is the most significant. The significance of the low growth scenario for developing countries (IS92d) is that it shows how this would effect "allowed" emissions in Annex I countries. A 10% reduction in the growth of non-Annex I (i.e. developing countries) emissions still requires major reductions in Annex I emissions by 2005. Figure 10 Annex 1 CO2 Emissions and CO2 Stabilization Scenarios
Annex
1 refers to the total emissions of industrialized countries included in Annex 1
of the climate convention.
A range of uncertainties lie behind any attempt to evaluate and set global ecological limits (and calculate corresponding carbon budgets) for an issue as broad and complex as climate change. These uncertainties include, but are not limited to:
(i) How sensitive the climate system is to human interference;
(ii) What kinds and levels of damage a given global temperature rise will cause;
(iii) What level of damage is acceptable to society i.e. what is dangerous climate change.
(iv) The future rate and cumulative volume of emissions.
The last uncertainty has been characterised in the preceding sections covering projections of emissions and their effects and will not be further discussed here.
The first type of uncertainty is explored below in terms of the temperature rise that a given increase in carbon dioxide levels is expected to cause (climate sensitivity); the risk that temperature rises could be much greater due to positive feedbacks in the climate system; and the risk inherent in the large uncertainty in sea-level rise estimates. The role eof aerosols (small particles) inreducing or offsetting the effects of greenhouse gas increases is discussed from the point of view of precautionary climate policy.
The second type of uncertainty raises the question of how good an indicator of damages any global mean indicator can be. Actual impacts may well be driven by rates of change of temperature, local differences in warming and the resulting changes in weather, changes in the frequency of extreme events, shifts in regional climate systems (such as the monsoon or North Atlantic storm tracks) and whether or not major catastrophic events are triggered. These issues are not discussed in any detail here, but they should be borne in mind when ecological limits are considered.
Ultimately, since many of the above effects are difficult, if not impossible, to quantify, decisions must at this stage be based on broad indicators of the likely level of damage. Limits on the level and rate of global mean temperature change and sea-level rise must be set on the basis of the precautionary principle so as to have the best chance of avoiding dangerous climate changes and impacts at the regional level.
The determination of what is dangerous change is, up to a point, inevitably subjective and dependent on one's view point. Global equity demands that international policy view the question of what is dangerous climate change from the point of view of the most vulnerable systems. It would be inappropriate to define dangerous change simply from the point of view of the most powerful countries. Whilst a sea-level rise of 20 cm may not be perceived as dangerous by the U.S.A., for example, it would be extremely dangerous for many small island states and countries with large low-lying coastal zones such as Bangladesh.
For these reasons Greenpeace believes that a precautionary approach must be taken to uncertainties in our understanding of the climate system, and to what constitutes dangerous change.
The term "climate sensitivity" refers to the global temperature increase that would occur if atmospheric concentrations of CO2 were doubled and the climate allowed to stabilize (or reach equilibrium). Since 1990 the IPCC's estimated range for climate sensitivity is between a 1.5 and 4.5oC increase in global temperature with a `best-estimate' of 2.5oC. A higher sensitivity of 3.5oC may better fit observations and recent advances in the understanding of the climate system.
The 1995 IPCC report provides empirical evidence that the 1990 `best-estimate' of the climate sensitivity is too low. Taking "best-estimates" of aerosol, ozone depletion and solar irradiance effects into account, Chapter 8 of the IPCC 1995 Science Assessment[84] argues that a "best fit" analysis of the recent climate record indicates that the climate sensitivity is more likely to be between 3°C and 4°C. If the best-estimate of climate sensitivity were increased from 2.5 to 3.5°C (or 4.5°C) this would increase the projected warming by 40% (or 80%). The most advanced climate models reviewed in the IPCC Second Assessment Report have climate sensitivities in the range 2.1-4.6oC with the median of the models being around 3.7oC[85].
Recent modelling of the role of the tropical ocean and of the operation of the thermohaline circulation system (see section 4.5.5) during the last glaciation, along with new geochemical evidence from fossil corals, groundwater and ice-records, and of cooler tropical oceans during the last ice age, implies that the climate sensitivity may be around 4oC[86]. Webb et. al. argue that their results imply that the climate sensitivity is higher than the IPCC `best-estimate. Whilst these results are controversial, scientific commentators believe that from a risk-averse or precautionary policy perspective these results indicated that climate sensitivities higher than the IPCC `best-estimate' still "have to be seriously considered."[87]
A higher climate sensitivity magnifies the risk created by an increase in greenhouse gas concentrations and also reduces the `carbon budget' for any given set of global climate targets. From a precautionary policy perspective it would be prudent to base climate policy on a higher climate sensitivity than that adopted by the IPCC as its `best-estimate'. As this work is oriented at policy based on the precautionary principle, a climate sensitivity of 3.5 oC will be generally used for the analysis. For comparative purposes the results corresponding to the IPCC `best-estimate' of 2.5oC sensitivity is included in parentheses where appropriate.
Error! Reference source not found.Figure 11 shows the relationship between warming limits, equivalent CO2 concentration (i.e. the concentration taking into account other greenhouse gases - see Section 7 Appendix) and the climate sensitivity. From this graph it can be seen that the higher the climate sensitivity the lower is the equivalent CO2 concentration corresponding to a given warming limit. A similar relationship holds for the `carbon budget', as is shown in the Appendix. Figure 11 Equivalent CO2 stabilization levels for temperature targets vs. climate sensitivity
Page: 29
Table 14 illustrates the effect of the uncertainties in the climate sensitivity parameter, on the equivalent and actual CO2 concentrations corresponding to two different ultimate temperature targets - a maximum increase of 2.0oC above pre-industrial levels and a 1.0oC maximum increase above pre-industrial levels. The latter can be considered a threshold for significant ecosystem damage whereas reaching an increase of 2.0oC would most likely bring about major damage.
The limit of 2.0oC requires that equivalent CO2 be below 484 ppmv for the IPCC best estimate of climate sensitivity and below 413 ppmv for a sensitivity of 3.5oC. The actual CO2 concentration (assuming that the forcing of other gases is equivalent to around 23 per cent of the effect of the CO2 increase alone) is 436 ppmv and 378 ppmv in respectively.
A 1.0oC limit requires, in the long term, that equivalent CO2 be below 367 ppmv for the IPCC `best-estimate' of climate sensitivity and below 339 ppmv for a sensitivity of 3.5oC. The actual CO2 concentration (assuming that the forcing of other gases is equivalent to around 23 per cent of the effect of the CO2 increase alone) is 348 ppmv and 327 ppmv in respectively. Such levels could be reached in the 22 second century, but only as consequence of policies adopted early in the 21st. Table 14 Temperature and CO2 concentration targets vs. climate sensitivity
Climate Sensitivity for CO2 doubling |
CO2 concentration ( ppmv) for 1.0oC maximum increase above pre-industrial global mean temperature |
CO2 concentration ( ppmv) for 2.0oC maximum increase above pre-industrial global mean temperature | |
2.5 oC |
Equivalent |
367 |
484 |
IPCC "Best -estimate |
Actual |
348 |
436 |
3.5 oC |
Equivalent |
339 |
413 |
best fit to observations |
Actual |
327 |
384 |
4.5 oC |
Equivalent |
324 |
378 |
Upper end of IPCC range |
Actual |
315 |
357 |
Other gases are assumed to contribute a further 23% of the radiative forcing of CO2 only (see below for discussion).
The IPCC has consistently warned of the potential for human induced climate warming to amplify itself. The existence of powerful positive feedbacks, which act to amplify an initial warming (or climate forcing), is one of the most worrying aspects of the climate system, when considering rising levels of greenhouse gases. In 1990 the IPCC found, in relation to climate feedback processes "it seems likely that, overall, they will act to increase, rather than decrease, greenhouse gas concentrations in a warmer world"[88].
Four years later the IPCC found that ice-core records over the past 220,000 years imply the existence of a significant positive feedback:
"Additional insights into climatic feedbacks come from ice core records going back over many thousands of years (known as palaeo-records). A clear correlation between atmospheric CO2 concentration and global temperature (especially during warming periods) is evident in much of the palaeo-record over long time-scales, with increases of about 80 ppmv occurring during deglaciations. This relationship between CO2 concentration and temperature may carry forward into the future, possibly causing a significant positive climate feedback on CO2 fluxes."[89]
Similar relationships for methane are observed from the palaeorecords, with the IPCC finding that that the positive relationship between CH4 concentration and temperature, as for CO2, may carry forward into the future[90].
The role of the terrestrial biosphere as a potential source of CO2 and CH4 emissions in response to rapid climate change is quite fundamental. The role of oceans is also important and is discussed in section 3.5.5 below.
The IPCC Second Assessment report has added to concerns over the role of the terrestrial biosphere indicating that climate warming could lead to it releasing major amounts of CO2 and other greenhouse gases. Sustained, rapid climate change could lead to forest dieback altering the terrestrial uptake and release of carbon. The annual volume of carbon that could be released should the terrestrial biosphere release carbon in response to climate change is significant relative to human emissions. The IPCC found with medium confidence that:
"Large amounts of carbon may be released transiently into the atmosphere as forests change in response to changing climate and before new forests replace the former vegetation. The loss of above ground carbon alone has been estimated to be 01.-3.4 GtCyr-1 or a total of 10-240 GtC."[91]
Such large feedbacks would make it very difficult to control the problem of climate change once started. This issue is particularly pressing as these feedbacks were estimated in response to the equilibrium climate effects of CO2 doubling. Doubling of equivalent CO2 concentration ( above pre-industrial levels) is likely to occur by 2030 or 2040, unless action is taken. Significant emissions from the biosphere could overwhelm attempts to stabilize CO2 concentrations and meet ecological targets.
Other terrestrial feedbacks identified by the IPCC include[92]:
* Release of carbon from drying out of high latitude wetlands. Whilst the rate of release is uncertain the ultimate volume of carbon may be quite large as there are 450 GtC stored in these systems.
* Increased release of nitrous oxide from warmer, wetter soils.
* Effects of land-surface changes. Albedo changes from the replacement of tundra by forests in northern high latitudes with warming may amplify the initial greenhouse gas forcing.
* Plant physiological effects of CO2 on climate. Reduction in stomatal conductance as CO2 concentration increases could significantly enhance the surface warming over terrestrial areas as a consequence of reductions in evapotranspiration and increases in soil moisture.
- Carbon fertilization effect. Increases in CO2 can enhance plant productivity, which is assumed to continue at a rate linked to the CO2 concentration, and is a negative feedback. However the 1994 IPCC report concludes that "when the availability of water and nutrients is taken into account the fertilization effect is likely to be reduced; several model results suggest reduction by around a half"[93].
Overall, the effects of climate feedbacks on the biosphere, induced by rising CO2 level's could lead to CO2, and other greenhouse gas levels, rising even faster, and potentially counteracting the effect of emission controls. If deforestation continues then the capacity of the forests and other natural vegetation to absorb CO2 will be diminished and may even become negative, exacerbating this problem.
At a global average level aerosols are calculated to have offset some of the effects of the increase in greenhouse gas concentrations to date.
In 1990 the CO2 concentration of the atmosphere was 355 ppmv. Other greenhouse gases further increased the enhanced greenhouse effect to a level equivalent to 421 ppmv CO2 (discounting the effects of sulphur aerosols) [94]
[Delta][Delta]
. Such an increase corresponds to a long term warming commitment of 2.1oC (1.5oC). Inclusion of the globally averaged radiative effect of aerosols would reduce the calculated warming commitment to about 1.1oC (0.9oC), which would correspond to an equivalent CO2 level of 343 ppmv. In terms of radiative forcing, the IPCC has estimated the radiative forcing increase over the period 1765-1990, including full aerosol effects, to be 1.32 W/m2. Without aerosol effects it is estimated at 2.62 W/m2, with the CO2 radiative forcing contributing 1.52W/m2 . In other words, at a global level, aerosols reduce the net radiative forcing of all other greenhouse gases[95].
This situation has led some to argue for a trade-off between aerosol emissions and greenhouse gases. From a policy perspective, however, there several aspects of the role of aerosols that mitigate strongly against such an approach.
Aerosol cooling effects are limited in time and to particular regions, so are unlikely to significantly reduce actual damages from CO2 emissions. The UK's Hadley Centre argues that "although, at first sight, smaller globally-averaged temperature changes might be assumed to imply smaller impacts, this is not necessarily the case."[96] Cutting aerosol emissions in future, for example after allowing significant increase, would suddenly reveal a large underlying warming commitment that the high level of aerosol emissions had masked. Aerosol concentration drops quickly (i.e. within weeks) once emissions are reduced, whereas CO2 concentration takes much longer (i.e. many decades).
In considering scenarios where CO2 emissions are decreasing, sulphur aerosol emissions will also be declining and at a faster rate than CO2 emissions. This means that the previous masking effect of aerosol emissions will be reduced, resulting in a positive increase in radiative forcing (relative to 1990) from this effect. However the overall effect on the rate of warming is very small. ( Figure 10 ) shows the rate of warming for various emission scenarios. Emission scenarios lower than 450 ppmv produce a calculated warming rate of a few hundreths of a degree per decade more than the IS92a Scenario in the period to 2005. Beyond 2005, or 2010 at the latest, the rate of warming drops well below the business as usual levels.
The IS92a scenario used to calculate the results in Figure 10 differs from the IPCC in that the sulphur aerosol emissions have been modified to account for the second Sulphur Protocol and US Clean Air Act Amendments. In addition it was assumed that sulphur controls similar to those adopted by the OECD would be used by Asian countries in the future. As a consequence sulphur emissions in 2100 are about the same level as in 1990.
The direct effects of sulphate aerosols, which lead to acid rain effects, on the environment and agricultural systems are very large. As consequence it is quite unlikely that the emissions of aerosols projected in the IPCC scenarios will come about[97]. In other words "relying" on aerosols to "hide" some of the warming from CO2 emissions would lead to greater risk in the future.
The regional effects of climate change are what are actually important for many climate induced damages and aerosols could have quite significant effects in modifying regional climates. The regional patchiness and short lifetimes of aerosols could increase the range of climatic extremes[98].
From a precautionary perspective it is the long term warming commitment of greenhouse gas emissions that is of paramount concern. For this reason the analysis for evaluating the `carbon budget' in this work will focus on the long term effects of greenhouse gas emissions and will not include aerosol effects.
Profound uncertainties surround the projection of future sea-level rise arising from human induced climate change. In terms of improved certainty for policy makers the scientific assessment of this issue appears to have deteriorated significantly over the past decade.
The IPCC's 1990 `best-estimate' of sea-level rise calculated from all sources (thermal expansion of the ocean, small glaciers, the Greenland Ice Sheet and the Antarctic Ice Sheet) was 10.5 cm with range from -0.5 to 22 cm. At that time the `best-estimate' for observed sea-level rise over the past 100 years (to 1990) was 15 cm (with a range from +10 to +20 cm), leaving some 4.5 cm unexplained between the `best-estimate' of calculated and observed changes.
Rather than reduce the uncertainties in explaining past sea-level rise the IPCC Second Assessment has indicated that there is a growing gap between the observed sea-level rise and that calculated to have occurred. The gap between the total best-estimate of calculated contributions to sea-level rise (8 cm) and the `best-estimate' of observed sea-level rise (18 cm) has doubled. The range of uncertainty in the calculated sea-level rise has increased significantly and is now from minus 19 to plus 37 cm. At the same time the contribution of the Greenland Ice Sheet has been reduced to zero with a range of -4 to +4 cm, down from 2.5 +/-1.5 cm in the 1990 IPCC Assessment.
This situation should be sounding alarm bells to policy makers. Without improving certainty in explaining past sea-level rise their can be little confidence in the estimates of future sea-level rise. The state of understanding in relation to the contribution of the Greenland and Antarctic ice sheets to sea-level rise has not improved between the First and Second IPCC Assessment reports. The IPCC has however warned clearly of the implications of uncertainty in this area as small changes in these ice sheets could have large impacts on sea-level rise:
"of all the terms that enter the sea level rise equation, the largest uncertainties pertain to the Earth's major ice sheets"
and
"relatively small changes in these ice sheets could have major effects of global sea level, yet we are not even certain of the sign of their present contribution"[99] (emphasis added).
The future behaviour of the Antarctic ice sheet in response to greenhouse warming is one of the central issues of concern for the predictions of future rate and long term extent of sea-level rise. Profound scientific uncertainty plagues the assessment of the contribution of the Antarctic ice sheet both to sea-level rise over the past 100 years and to projections of future sea-level rise as a consequence of climate change. In the last six years scientific understanding of the quantitative contribution of the Antarctic ice sheet to past sea-level rise has, if anything, deteriorated. At the same time the IPCC `best-estimate' of the projected Antarctic ice sheet contribution to future sea-level to 2100 has remained slightly negative[100]. If this turns out to be incorrect much larger sea-level rise than currently projected would eventuate in the longer term and possibly much larger rates of change in the shorter term. In addition, the potential instability of the West Antarctic Ice Sheet, whose collapse would raise sea-level by 5-8 metres over several hundred years is of major concern in the assessment of sea-level rise risk from global warming.
The contribution of Antarctica to past sea-level rise is highly uncertain.
The 1990 IPCC `best-estimate' made for the Antarctic ice sheet contribution to sea-level rise over the past 100 years was zero cm (with a range of +/- 5 cm. Warrick and Oerlemans point out in summing up the evidence that:
"the `zero' entries [for the Antarctic ice sheet contribution to sea level rise] should be interpreted as a reflection o f the current poor state of knowledge, rather than as an estimate of the current state of balance."
and further that
"a large positive mass balance of both ice sheets would seem unlikely, as it would have led to a substantial sea level lowering and would therefore be highly inconsistent with the observed sea level rise."[101]
In the 1995 IPCC Assessment it was found that disagreement over the contribution of Antarctica had widened:
"the paucity of data does not allow a meaningful judgement of the current state of balance of the Greenland and Antarctic ice sheet s. Different workers claim changes with even different sign..."[102]
Whilst the `best-estimate' of the calculated Antarctic ice sheet contribution to past sea-level rise remained zero the range was increased significantly to plus or minus 14 cm[103]. This uncertainty range is 160% of the `best-estimate' for observed sea-level rise of 18 cm (range +10 to +25 cm).
In relation to future sea-level rise both the 1990 and 1995 IPCC Assessments found that the Antarctic ice sheet should contribute negatively to future sea-level rise as rising temperatures should lead to more precipitation over Antarctica (leading to a greater net accumulation of ice)[104]. However well based this judgement is it is fundamentally undermined by the fact the sea-level rise of the last century cannot be explained and where the sign of the role played by Antarctica (and to a lesser extent the Greenland Ice sheet) remain unknown.
The bottom line from a risk assessment perspective is that the large unexplained gap between `best-estimate's of the calculated and observed sea-level rise could easily be explained by a negative mass balance for the Antarctic ice sheet that is:
* Well within the range of uncertainty in the IPCC estimates
- Consistent with estimates of Antarctic Ice Sheet mass loss not reported in the IPCC Second Assessment but reported in the peer reviewed literature[105].
At present global sea-level appears to be rising by about 1 to 2.5 mm per year. Although both thermal expansion of the ocean and melting of small glaciers are accounted for in the estimates, the major source of water (25 per cent) for the current level rise is unknown[106]. It is possible that part of this "missing water" comes from meltwater escaping unnoticed for years from the polar glaciers[107]. On the basis of recent estimates of basal melting (melting from the bottom of floating ice shelves) Jacobs et.al.[108] have suggested that "the Antarctic ice sheet is currently losing mass to the ocean."
- Consistent with numerical modelling of the Antarctic ice sheet reported by the IPCC in 1995[109].
These factors point towards a serious concern that the IPCC assessment is not conveying sufficient information to policy maker on the grave risks posed in terms of large, long term, irreversible sea-level rise by the effects of greenhouse warming on the Antarctic Ice Sheet and to a lesser extent the Greenland Ice Sheet.
Warming of the Southern Ocean (and surrounding oceans south of 30oS latitude) in response to rising CO2 levels could play a significant role in determining the ultimate
levels of CO2 in the atmosphere. The oceans play a significant role in the global carbon cycle and hence in controlling the level of CO2 in the atmosphere. CO2 is taken from the atmosphere by the oceans and stored in the deep ocean, with the average ocean uptake over the 1980's being around 2 GtC/yr. offsetting some of the average 5.5 GtC/yr. emissions from fossil fuels over this period.
In calculating future atmospheric CO2 levels the IPCC assumed that ocean currents and temperature would not change[110], however it is well established that global warming would lead to significant changes in the ocean circulation. Feedbacks from ocean circulation changes induced by climate change to the carbon cycle could play a significant role in determining the future levels of atmospheric CO2.
There are three main processes which determine the future role of the ocean in terms of its capacity to take up a fraction of human emissions of CO2:
* Sea-surface temperature feedback.
Warmer sea surface temperature lowers the solubility of CO2 in the oceans. In 1994 the IPCC estimated that there would be a weak positive feedback between a global increase in sea surface temperature and atmospheric CO2 estimated at 10 ppmv of CO2 for each 1 degree rise in temperature.
* Changes in the oceanic circulation
Coupled Ocean-Atmosphere General Circulation Models (AOGCMs) indicate significant changes in the ocean circulation and in particular a weakening of the thermohaline system in response to rising greenhouse gas concentrations[111]. This would result in less transport of CO2 enriched surface water to depth, reducing the capacity of the oceans to take up carbon.
* Changes in the marine biological carbon pump
An important component of the carbon cycle is the so-called marine biological pump which "exports" excess carbon (dead organic matter) to the ocean depths. Modelling indicates that the marine biota plays a major role in regulating CO2. In the atmosphere - without marine biota it has been calculated that the pre-industrial CO2 levels would have been 450 ppmv rather than 280 ppmv.
Scientific uncertainty in relation to competing processes in the response to global warming of the biological carbon pump and of the ocean circulation system, as well as a lack of data and adequate models, mean that it is difficult to ascertain the implications of rising CO2 levels for this major component of the carbon cycle.
Since the conclusion of the IPCC Second Assessment Report Sarmiento and Le Quere[112] have published the first coupled AOGCM calculations of the effects of rising CO2 levels on the oceanic uptake of carbon. Their model includes a simple representation of marine biological processes and has been used to estimate the effects of a doubling and a quadrupling of CO2 concentration. The results show that the most of the oceanic uptake of carbon occurs in the Southern Ocean and that this ocean has the largest impact on the response of oceanic CO2 uptake to global warming.
Several conclusions can be drawn from Sarmiento and LeQuere's calculations of the oceanic scenarios involving a doubling and a quadrupling of CO2.
The weakening or collapse of the thermohaline circulation leads to a major reduction in the oceanic uptake of CO2. This significantly increases the rate of growth of future atmospheric CO2 concentrations. There is a 140 GtC reduction in CO2 uptake between Sarmiento and Le Quere's baseline scenario (without ocean circulation changes) and the scenario including these changes over 100 years. This would represent a strong feedback to the climate system adding over 60 ppmv CO2 to the atmosphere.
Changes to the marine biota are likely to offset some but not all of the effects of oceanic circulation changes[113]. The largest increase in biological CO2 uptake is in the southern ocean.
The overall implications from a policy perspective of this work appears to that:
* The anthropogenic CO2 emissions corresponding to concentration stabilization levels are likely to be lower than the IPCC has previously estimated when the effects of oceanic feedbacks are taken into account.
* Atmospheric CO2 levels for a given set of emissions are likely to be greater and hence the global warming will be higher than the IPCC best-estimate calculations for emission scenarios.
Objectives for global ecological targets for climate policy need to be based on the precautionary principle, which is incorporated into the Climate Convention through its ultimate objective to prevent dangerous interference in the climate system. Ecological targets need, in effect, to be surrogate measures of the risk of climate change and hence linked to both the rate and the magnitude of changes that occur as a consequence of increases in greenhouse gas concentrations.
Two kinds of global climate targets have been put forward, one based on the concentration of greenhouse gases and the other on surrogate measures of impacts - the rate and magnitude of global mean temperature and sea-level rise.
One of the key problems with concentration based goals is that dangerous rates of climate change are significantly determined by the trajectory (or time path) of emissions. Impacts of climate change are not caused by the increases in greenhouse gas concentrations but by the consequential changes in the climate. Concentration targets do not help deal with the question of the timing of emission reductions in order to minimise damage nor are they a good surrogate for impacts.
Uncertainties in the climate sensitivity make the use of long term concentration goals as a surrogate for the impacts of climate change misleading at best and at worst potentially quite dangerous. Whilst uncertainties in the climate sensitivity are relevant to any global ecological targets this presents peculiar problems for concentration goals. The key danger which arises from the use of a concentration target is that it does not focus policy attention on the inertia of the climate system . One of the most crucial and pragmatic questions of climate policy, that of what needs to be done in the short term in order that options are protected for the future, is answered in a potentially dangerous manner. In policy terms a concentration goal implies that the magnitude of emissions in the short term do not matter to the prevention of dangerous climate change. This overlooks the need to limit the rate of change.
If policy assumes that emissions in the short term do not matter, based on a concentration goal, and subsequently it is found that the concentration needs to be lowered considerably there is a significant risk that the second target may not be achievable. On the other hand policy based on the ecological targets described above, using an approach such as the "safe emissions corridor" methodology is capable of responding much more safely and dynamically to changing science and objectives
Efforts to set global ecological goals have tended to adopt, in a sometimes confusing manner, both a climate and a concentration target. The European Union Environment Council, for example, has proposed that the increase in global average temperatures should not be allowed to exceed 2°C above pre-industrial levels. However in doing so it also nominated a maximum CO2 level[114]:
"Given the serious risk of such an increase and particularly the very high rate of change the Council believes that global average temperatures should not exceed 2 degrees (Celsius) above pre-industrial level and that therefore concentration levels lower than 550 (parts per million of) CO2 should guide global limitation and reduction efforts. This means that the concentrations of all greenhouse gases should also be stabilised. This is likely to require a reduction of emissions of greenhouse gases other than CO2, in particular CH4 and N20."[115]
The European Union's goal of not exceeding 2oC can be translated into a CO2 equivalent concentration range of 380-560 ppmv depending on the climate sensitivity assumed for a CO2 doubling. Figure 14 shows the large difference in CO2 emissions corresponding to this range ranging from immediate reductions to approximately constant emissions at above current levels over the next century. With the IPCC best-estimate of climate sensitivity the EU target would correspond to 484 ppmv CO2 equivalent. If the climate sensitivity is 3.5oC, as assumed in this report, then the equivalent CO2 concentration would be 71 ppmv lower i.e. 413 ppmv. Whilst the EU Council Decision refers to atmospheric CO2 concentrations being kept below 550 ppmv in order to not exceed the 2oC limit it is clear from both the wording of the decision and from the science that this is an extreme upper bound. In any event the European Union's 2oC limit itself carries with it the risk of large, irreversible and dangerous changes.
From the scientific domain, Azar and Rodhe argue that international climate policy should aim to ensure that that global mean changes should not "substantially" exceed the natural fluctuations of the last thousand years of around 1oC:
"The burden of proof must lie on those who argue that it is safe and acceptable to cause changes in the global climate system that substantially exceed the natural fluctuations during the past millennium. Given that this fluctuation in global average surface temperature is around 1°C (or less), a temperature increase by 2°C may be seen as such a critical level. Until it has been proven that a temperature increase above 2°C is safe or that the climate sensitivity is lower than the central estimate, ... the global community should initiate policies that make stabilization in the range of 350 to 400 ppmv possible." [116]
With the radiative forcing assumptions made by Azar and Rodhe[117] this corresponds to an equivalent concentration range of 410-468 ppmv and an equilibrium warming of 1.4-1.9oC for the IPCC best-estimate climate sensitivity and 2.0-2.6oC for the climate sensitivity of 3.5oC used in this work.
From the point of view of the precautionary principle and taking into account knowledge of the impacts of climate change on species and ecosystems, Azar and Rodhe's target appears to high. A more robust approach would be to ensure that climate policies are aimed at making it possible to limit the long term warming to within the natural variability observed in the past few thousand years over periods of decades to centuries i.e. below 1oC. Given the large warming commitment built into the climate system as consequence of historic emissions this may mean that the observed warming over the next several decades exceeds this limit. Nevertheless, limiting the period of this exceedance maybe the only way of minimizing or avoiding dangerous impacts.
As can be seen from the foregoing discussion concentration goals such as 450 and 550 ppmv are far too high to be adopted as global ecological targets, quite apart from the general problems with adopting concentration goals as the basis for long term ecological targets.
As a consequence of these and other concerns Greenpeace believes that international climate policy on greenhouse gas emissions should aim to meet a set of global ecological targets:
(i) Limit the long term increase of temperature to less than 1°C above pre-industrial levels.
(ii) Bring the rate of climate change to below 0.1°C/decade as fast as possible, within a few decades at the most. Warming rates over the next century are projected to be in the range 0.2-0.3°C/decade.
(iii) Limit long term global average sea-level rise to less than 20 cm.
A sea-level rise of this extent would still lead to some damage for low lying islands and coastal areas, however higher levels would lead to rapidly rising risk. A warming limit of 20 cm by 2100 would entail an ultimate sea-level rise of 40-60 cm if there are no surprises in, for example, the behaviour of the large Greenland and West Antarctic Ice sheets. There is a high degree of inertia in relation to sea-level rise. It seems likely, for example, that around 10 cm of sea-level rise are already commited over the next century as a consequence of historic greenhouse gas emissions.
(iv) Bring the rate of sea-level rise to below 20mm/decade. This would permit the vast majority of vulnerable ecosystems, such as natural wetlands and coral reefs to adapt.
These four key global ecological targets need to be met simultaneously. In meeting these limits the emission pathway (i.e. the timing of emissions cuts) is very important. Given the uncertainties involved and the need to apply the precautionary principle, greenhouse gas emission policies aimed at meeting these targets should use a climate sensitivity of 3.5oC rather than the `best-estimate' of the IPCC.
The focus in the next section however is on overall limits to the emissions of CO2. Whilst methodologicall this necessarily must focus on overall warming limits, rather than on the rate of change, the .
[45] United Nations Framework Convention on Climate Change (UNFCCC), Adopted by the Intergovernmental Negotiating Committee 9 May 1992, Opened for Signature at Rio de Janeiro 4 June 1992. Entered into force 21 March 1995. U. N. Doc. A/AC.237/18 (Part II) (Add 1).
[46] Article 2 of the UNFCCC.
[47] Article 3.3 of the UNFCCC
[48] IPCC SAR WGI op.cit. Summary for Policy Makers, p. 6.
[49] IPCC SAR WGII op.cit. Summary for Policy Makers, p. 9.
[50] It is important to note the difference between CO2 equivalent and actual CO2 concentrations. Equivalent CO2 refers to the greenhouse effect of both the actual CO2 concentration and the other greenhouse gases combined and converted to a CO2 equivalent concentration. Stabilizing actual CO2 concentrations at for, example, 450 ppmv means equivalent CO2 concentrations of around 530 ppmv owing to the effects of the other gases. The climate difference between stabilizing at 450 ppmv actual CO2 and 450 ppmv equivalent is significant i.e. 0.6oC - 0.8oC depending on the climate sensitivity and the relative conribution of other greenhouse gases.
[51] IPCC Synthesis Report, op.cit. Par. 3.13
[52] Watson, R.T, M.C. Zinyowera and R. H. Moss (1996) IPCC WGII Technical Summary: Impacts, Adaptation and Mitigation Options in IPCC SAR WGII op.cit.
[53] Rosenzweig C. and M. Parry, "Potential impact of climate change on world food supply", Nature, v.367, p.133-138, 13 January 1994
[54 Rijsberman, F.J and R.J. Swart (eds.) (1990), ]Targets and Indicators of Climate Change, Stockholm Environment Institute. The World Meteorological Organization/International Council of Scientific Unions/United Nations Environment Programme (WMO/ICSU/UNEP) Advisory Group on Greenhouse Gases (AGGG) set up three working groups in 1988 one of which was to examine targets and indicators of climate change. This group was under the Chairmanship of P. Vellinga and P.H. Gleick.
[55] Rijsberman, F.J and R.J. Swart (eds.) (1990) op.cit. p.iv
[56] Rijsberman, F.J and R.J. Swart (eds.) (1990) op.cit. p.viii
[57] Vellinga, P. and R. Swart (1990), "The Greenhouse Marathon: Proposal for a Global Strategy", pp. 129-134 in J. Jager and H.L.Ferguson(Ed's) (1990), Climate Change: Science, Impacts and Policy", Proceedings of the Second World Climate Conference, World Meteorological Organisation. Cambridge University Press.
[58 ]Hadley Centre (1995), "Modelling Climate Change 1860 - 2050", UK Meteorological Office.
[59] IPCC SAR WGII op.cit., Summary for Policy Makers p. 5.
[60] IPCC SAR WGII op.cit., Summary for Policy Makers p. 5,6.
[61] IPCC SAR WGII op.cit, Chapter 9, p. 311.
[62] Stocker, T.F. and A. Schmittner (1997) "Influence of CO2 emission rates on the stability of the thermohaline circulation", Nature Vol. 388, pp. 862-865
[63] The rate of increase of CO2 alone in the 1990's is approximately 0.4-0.5%/yr, with the effects of other greenhouse gas emissions bringing the total rate of increase to around 0.6-0.7%/yr. in CO2 equivalent terms. The IS92a scenario projects an increase in equivalent CO2 concentration of 0.7-0.8%/yr. compound over the next century.
[64] IPCC SAR WGI Chapter 7 p. 388: Warrick R. A., C. Le Provost, M.F. Meier, J. Oerlemans, P.Ll. Woodworth. (1996) Changes in sea level Chapter 7 pp. 259-405 of IPCC SAR WGI op.cit.
[65] Alcamo, Joseph and Eric Kreileman (1996) "Emission scenarios and global climate protection", Global Environmental Change; Human Policy and Dimensions, Vol.6, Number 4, September 1996; pp. 305-334.
[66] Raper, S.C.B.; T.M.L. Wigley and R.A. Warrick (1996) "Global Sea-level Rise: Past and Future", Chapter 1 in John D, Milliman, Bilal U. Haq (eds.) Sea-Level Rise and Coastal Subsidence: Causes, Consequences, and Strategies, Dordrecht, Boston, London, Kluwer Academic Publishers, 1996
[67] Enting, I.G.; T.M.L. Wigley and M. Heimann (1994), Future Emissions and Concentrations of carbon dioxide, Technical Paper No. 31, CSIRO Division of Atmospheric Research, Mordialloc, Australia.
[68] Christian Azar and Henning Rodhe, "Targets for Stabilization of Atmospheric CO2" in Science, vol. 276, 20 June 1997; pp. 1818.
[69] This scenario was included to show the emission consequences of higher stabilization levels. The authors note however that whilst the environmental consequences have not been assessed "they are certain to be very large". IPCC SAR Chapter 1, op.cit p. 83.
[70] See [. This refers to the radiative forcing without consideration of aerosol effects. Inclusion of full aerosol effects would mean the equivalent CO]2 range is 750-1,000 ppmv.
[71] Alcamo and Kreileman (1997) op.cit. p. 315-314
[72] Alcamo and Kreileman (1997) op.cit. p. 317
[73 ]Wigley, T.M.L (1995), "Global Mean Temperature and Sea-Level Consequence of Greenhouse Gas Concentration Stabilization", Geophysical Research Letters, 22(1), 45-48.
[74 ]Raper et al (1996) op.cit.
[75 ]FCCC/AGBM/1997/MISC.1/Add.2
[76] Alcamo and Kreileman (1996) op.cit.
[77] Alcamo and Kreileman (1996) op.cit p. .317. The temperature limits were set with respect to 1990 global average surface temperature, rather than pre-industrial levels.
[78] The Berlin Mandate (Decision 1/CP.1) paragraph 2(b) specifically states that it will "not introduce any new commitments for Parties not included in Annex I (i.e. developing countries), but reaffirm existing commitments in Article 4.1 and continue to advance the implementation of these commitments in order to achieve sustainable development". FCCC/CP/1995/7/Add.1 24 May 1995
[79] The MAGICC model was used for these calculations. See Wigley, T.M.L. (1994) MAGICC Model for the Assessment of Greenhouse Gas Induced Climate Change, Users Guide and Scientific Reference Manual, Climate Research Unit, University of East Anglia and NCAR, Boulder Colorado, October 1994
[80] This refers to Annex 1 of the Climate Convention which is the list of industrialized countries, including central and eastern Europe and countries of the former Soviet Union which are subject to legally binding controls on greenhouse gas emissions.
[81] Alcamo and Kreileman (1996) op.cit. pp. 327-328. The smaller reduction refers to the top of the corridor and the larger to the middle.
[82] Approximately the IPCC business as usual (IS92a) scenario for non-Annex I countries.
[83 ]These results were computed by M. Berk at RIVM, The Netherlands.
[84] IPCC SAR WGI op.cit. Chapter 8, p. 424: Santer B. D., T.M.L Wigley, T.P Barnett, E.Anyamba. (1996) Detection of climate change and attribution of causes, Chapter 8 pp. 407-443.
[85] IPCC SAR WGI op.cit. Chapter 6, Table 6.3, pp. 298-299. Kattenberg A., F. Giorgi, H. Grassl, G. A.Meehl, J.F.B. Mitchel, R. J. Stouffer, T. Tokioka, A. J Weaver, T.M. L. Wigley (1996), Climate models - Projections of future climate.
[86] Webb, R.S., D.H. Rind, S.J. Lehmann, R.J. Healy and D. Sigman (1997), "Influence of ocean heat trasnport on the climate of the Last Glacial Maximum", Nature, vol. 385, pp.695-699.
[87] Harvey, L.D. of University of Toronto quoted in Global Environmental Change Report, vol. IX, no. 4, 28 February 1997, p. 2
[88] IPCC First Assessment Report, 1990, p. xviii: J.T. Houghton, G J Jenkins, and J.J. Ephraums (ed.'s) (1990), Climate Change - The IPCC Scientific Assessment. Report prepared for IPCC by Working Group I. Cambridge University Press.
[89] IPCC 1994, op.cit., p25.
[90] IPCC 1994, op.cit., p. 27.
[91 ]IPCC SAR WGII op.cit., p. 97: M.U.F Kirschbaum, A. Fischlin (1996) Climate change impacts on Forests, Part II Chapter 1 pp. 95-129.
[92] IPCC SAR WGI op.cit., Chapter 9: J. M Melillo, I.C Prentice, G.D. Farquhar, E.-D. Shultze, O.E Sala. (1996) Terrestrial biotic responses to environmental change and feedbacks to climate Chapter 9 pp. 445-481.
[93] IPCC 1994 op.cit., p. 18.
[94 ]Equivalent CO2 concentrations are defined by Cequiv=278*exp(Q/6.3)where Q is the radiative forcing due to the increase in greenhouse gas concentrations above pre-industrial levels. See Sect.6.3.2 of IPCC SAR WGI: Kattenberg A., F. Giorgi, H. Grassl, G. A.Meehl, J.F.B. Mitchel, R. J. Stouffer, T. Tokioka, A. J Weaver, T.M. L. Wigley (1996) Climate Models - Projections of future climate, Chapter 6 pp. 284-357.
[95] Raper et al (1996) op.cit., Appendix A
[96 ]Hadley Centre op.cit.
[97] See for example, Global Energy Perspectives to 2050 and Beyond WEC World Energy Council/IIASA International Institute for Applied Systems Analysis (1995).
[98 ]Hadley Centre op.cit.
[99] IPCC SAR WGI Chapter 7, op.cit. p. 396
[100] IPCC SAR WGI Chapter 7, op.cit. p. 364
[101] Warrick. R.A. and H. Oerlemans (1990) "Sea Level Rise", Chapter 9 in IPCC SAR WGI op.cit.
[102] IPCC SAR WGI Chapter 7, op.cit. p. 376-377
[103]IPCC SAR WGI Chapter 7, op.cit. p. 380
[104] IPCC SAR WGI Chapter 7, op.cit. p. 363.
[105] Jacobs, S.S. (1992), "Is the Antarctic ice sheet growing?" Nature, Vol. 360 pp. 29-33.,
Jacobs, S.S., H.H. Hellmer, C.S.M. Doake, A. Jenkins and R.M. Frolich, (1992) "Melting of ice shelves and the mass balance of Antarctica" , J. Glaciology, Vol. 38 No. 130 pp. 375-386 and Jacobs, S.S., H Heller, A Jenkins (1996) "Antarctic ice sheet melting in the Southeast Pacific", Geophysical Research Letters, Vol. 23 No. 9 pp. 957-960.
[106] Zwally, H.J. (1994) "Detection of change in Antarctica", in: Hempel, G (ed.), Antarctic Science: Global Concerns, Springer-Verlag, Berlin, pp. 126-143.
[107] Jacobs et.al.(1992), op.cit.
[108] Jacobs et.al.(1996), op.cit. Floating ice shelves do not contribute do sea-level rise if they melt, however large melting rates may indicate a net movement of ice on the land (ice shelves and glaciers) to the sea, which would raise sea-level.
[109] IPCC SAR WGI Chapter 7, op.cit.
[110] Schimel et. al. (1996) Chapter 1 in IPCC SAR WGI op.cit. and Sarmiento, J.L. and C. Le Quere (1996) "Oceanic Carbon Dioxide Uptake in a Model of Century-Scale Global Warming", Science Vol. 274 pp. 1346-1350.
[111] See for example Stocker, T.F. and A. Schmittner (1997) op.cit.
[112] Sarmiento and Le Quere (1996) op.cit.
[113] See Table 3 of Sarmiento and Le Quere (1996) op.cit.
[114] European Community (1996) Climate Change - Council Conclusions 8518/96 (Presse 188-G) 25/26.VI.96
[115] The Danish government has gone further in announcing that it is directing its energy and climate policy in manner that would be consistent with a global target of keeping CO2 levels below 450 ppmv, with 20% reduction on 2005 and a greater than 50% reduction by 2030.
[116] Azar, C and H. Rodhe, "Targets for Stabilization of Atmospheric CO2" in Science, vol. 276, 20 June 1997; p. 1818-1819.
[117] Azar and Rodhe assume a radiative forcing of 1 W/m2 for non greenhouse gases.