In 2012, the CBD Secretariat published Technical Series No. 66: Geoengineering in Relation to the Convention on Biological Diversity: Technical and Regulatory Matters comprising two studies: one on the impacts of climate-related geoengineering on biodiversity, the other on the regulatory framework for climate-related geoengineering relevant to the Convention.

This update report was prepared for peer review in July 2015, taking account of relevant publications available then. If you want to download the full document click here.

Key messages from the document:

1. Climate-related geoengineering is here defined as a deliberate intervention in the planetary environment of a nature and scale intended to counteract anthropogenic climate change and its impacts. This definition is the same as used in CBD (2012)1, and is used here without prejudice to any definition that may subsequently be agreed under the Convention. “Climate engineering” and “climate intervention” may be considered as equivalent to “climate-related geoengineering”, hereafter geoengineering. Generally, climate-related geoengineering is divided into two main groups at the technique level: i) techniques involving greenhouse gas removal (GGR), also known as “negative emission techniques”; most existing and proposed techniques fall under the term “carbon dioxide removal” (CDR); and ii) techniques known as sunlight reflection methods (SRM; alternatively “solar radiation management” or “albedo management”). In addition there are other proposed techniques, that could directly increase heat loss, or redistribute energy within the Earth system. Key features of the definition are that the interventions are deliberate, and are on a scale large enough to significantly counter-act the warming effect of greenhouse gases. They are thus distinct from actions to reduce emissions. However, some of the techniques involving greenhouse gas removal, such as afforestation, reforestation, techniques for managing soils to increase carbon sequestration, and the use of bioenergy combined with carbon capture and storage, are also considered climate mitigation techniques. Not all of the latter techniques are considered by all stakeholders to be geoengineering. In any case, interventions (both GGR and SRM) that are carried out at a small scale (e.g. local tree planting projects; roof whitening) are not normally considered as geoengineering. In line with decision X/33, the definition also excludes carbon capture at source from fossil fuels (CCS; i.e. preventing the release of CO2 into the atmosphere), while recognizing that the carbon storage components of that process may also be shared by other techniques that are considered as geoengineering.

2. Assessment of the impacts of geoengineering on biodiversity is not straightforward and is subject to many uncertainties. Relatively little research has directly addressed the issue of ‘impacts on biodiversity’, nor even broader environmental implications: instead effort by natural scientists has mostly focussed on climatic (physico-chemical) issues or impacts on agricultural systems, while social scientists have addressed governance, framing and ethical considerations. This report considers the impacts of geoengineering on the drivers of biodiversity loss, including the potential decrease in the climate change driver from effective geoengineering techniques; changes in other drivers, including land use change, that are inevitably associated with some geoengineering approaches; as well as the other positive and negative side effects of specific techniques. Consequences for biodiversity are therefore mostly discussed in terms of climatic effectiveness, land use change or other indirect impacts; e.g. fertilizer application or water extraction. It is important to note that both decreased and increased productivity tend to be undesirable from a natural ecosystem perspective, although the latter is likely to be beneficial in agricultural systems.

Climate Change

3. Climate change, including ocean acidification, is already impacting biodiversity and further impacts are inevitable. It may still be possible that deep and very rapid decarbonization by all countries might allow climate change to be kept within a 2°C limit by emission reduction alone. However, any such window of opportunity is rapidly closing. Even so, climate change associated with 2°C warming will have serious impacts on biodiversity. Emissions under current trajectories, broadly consistent with RCP 8.5 (the highest of the four main scenarios used in the IPCC AR5) would lead to an extremely large loss of biodiversity. Current commitments made by Parties to the UNFCCC would significantly reduce climate change and its impacts (probably to a pathway between RCP 6.0 and RCP 4.5) but are insufficient to keep warming within 2°C. Geoengineering techniques, if viable and effective, would be expected to reduce climate change impacts on biodiversity. However some techniques would lead to biodiversity loss through other drivers such as land use change.

Carbon dioxide removal (Greenhouse gas removal)

4. Scenarios of future climate change to 2100 that are likely to keep global average temperature increases within a limit of 2°C above pre-industrial levels mostly rely on technologies for carbon dioxide removal (CDR) as well as emission reductions, with pathways that feature net negative emissions in the second half of the century. However, the potential to deploy CDR at this scale is highly uncertain. The deployment of CDR envisaged by scenarios reported in the IPCC Fifth Assessment Report in the period 2050-2100 would allow additional anthropogenic greenhouse gas emissions in the period up to 2050, extending the period of fossil fuel use and potentially reducing the cost of their phase-out. For RCP 2.6, ~90% of the pathways considered in the IPCC AR5 assume the deployment of CDR technologies. Bioenergy with carbon capture and storage (BECCS) and/or afforestation/reforestation (AR) are seen as the most economically viable ways to provide such net negative emissions. The land and water use requirements of BECCS and AR are limiting factors, but those requirements, and their implications, are not well factored into existing models. For BECCS, CO2 storage capacity may also be limiting.

5. The removal of a given quantity of a greenhouse gas would not fully compensate for an earlier “overshoot” of emissions. The occurrence of an overshoot in most RCP 2.6 scenarios allows for current emissions to be offset by future negative emissions. The assumption is made that CDR will be achievable at the scale required, without such actions themselves having significant undesirable consequences; this assumption seems unlikely to be valid. In particular, not all the climatic and environmental consequences of the overshoot will be directly cancelled by future CO2 removal. The net effect of adding and subsequently subtracting a given quantity of CO2 only equals zero when there is no significant time difference between the addition and subtraction processes; a delay of ~50 years would lead to significant and potentially irreversible consequences for biodiversity and the Earth system. For those reasons, the evaluation of the potential role of CDR techniques should focus on their effectiveness in helping to reduce net emissions to zero on a shorter timescale than envisaged in most current scenarios, complementing stringent emission reductions.

6. The large-scale deployment of bioenergy with carbon capture and storage (BECCS) seems likely to have significant negative impacts on biodiversity through land use change. If BECCS were deployed to a scale assumed in most RCP 2.6 scenarios, substantial areas of land (several hundred million hectares), water (potentially doubling agricultural water demand) and fertilizer would be needed to sustain bioenergy crops. Limiting irrigation to reduce water use, or not replacing nutrients, would increase land requirements. Even under optimistic scenarios, less than half of the requirements for negative emissions could likely be met from abandoned agricultural land. Land use change envisaged in the central RCP 2.6 scenario would lead to large losses of terrestrial biodiversity.

7. Ecosystem restoration including reforestation and appropriate afforestation can contribute to removing carbon dioxide and provide substantive biodiversity co-benefits. However, these activities on their own would be insufficient to remove carbon at the scale required in most current scenarios. Avoiding deforestation, and the loss of other high-carbon natural vegetation, is more efficient than restoration or afforestation in contributing to climate mitigation and has greater biodiversity co-benefits. Afforestation of ecosystems currently under non-forest native vegetation would result in the loss of the biodiversity unique to such habitats, and from an ecological perspective, should be avoided2. Furthermore, the greenhouse effects of N2O arising from nitrogen fertilizers may outweigh the CO2 gains; afforestation of boreal areas and desert areas would increase global warming though albedo effects; and future climate change may jeopardize forest carbon sinks, through increased frequency of fire, pests and diseases and extreme weather events.

8. Biochar may potentially contribute to carbon dioxide removal under certain circumstances, and the technique applied to agricultural soils may offer productivity co-benefits. The application of biochar (charcoal) to soils may have positive or negative impacts on soil biodiversity and productivity, but there is greater evidence of positive impacts, especially in acidic soils. In addition, biochar application to soils may also decrease soil carbon emissions. A quantitative understanding of the factors affecting the permanence of biochar carbon sequestration is being developed. However, until the use of coal and other high-emission fossil fuels are phased out, the alternative use of charcoal as fuel may have greater potential in climate mitigation. Assessments of the climatic benefits, co-benefits and costs of different biochar processes and products are needed to fully evaluate the potential of this technique. Current scenarios envisage the production of biochar from crop residues and food wastes. Nevertheless, deployment of this technique on a large scale would have significant direct and indirect impacts on the use of land, water and fertilizers to generate the biomass required.

9. The viability of alternative negative emission techniques such as direct air capture (DAC), enhanced weathering and ocean fertilization remains unproven. There has been significant research work since CBD (2012), yet conclusions remain broadly the same. Likely costs and energy requirements of DAC for CO2 are still very high, albeit considerably lower than those reported in CBD (2012). Since there may be further potential for cost reductions, additional research on DAC techniques for CO2, as well as methane, warrants attention. The potential contribution of enhanced weathering, on land or in the ocean, to negative emissions is unclear but logistical factors seem likely to limit deployment at large scales. Local marine application might be effective in slowing or reducing ocean acidification, with consequent benefits for marine biodiversity, though there might also be negative effects; e.g. from sedimentation. Enhancing ocean productivity, by stimulating phytoplankton growth in the open ocean and through nutrient addition (“ocean fertilization”) or modification of upwelling, is only likely to sequester relatively modest amounts of CO2, and the environmental risks and uncertainties associated with large-scale deployment remain high.

10. Carbon dioxide (or other greenhouse gases) captured from the atmosphere must be stored in some form. Options include vegetation, soils, charcoal, or carbon dioxide in geological formations. Vegetation, soils and charcoal demonstrate varying levels of (im)permanence. Technical considerations relating to safe carbon storage in geological formations, mostly expected to be beneath the seafloor, have recently been reviewed. The main effects of marine leakage would be local ocean acidification with experimental studies indicating that (at least for slow release rates) environmental impacts would be relatively localized. The extensive literature on ocean acidification, including the biodiversity changes observed at natural CO2 vents, is relevant here. However, relatively few experimental studies on the impacts of high CO2 on marine organisms cover the full range of values that might occur under leakage conditions. Other forms of storage in the ocean are considered to have unacceptable risks, and are not allowed under the London Convention/London Protocol.

Sunlight reflection methods / Solar radiation management

11. Recent studies and assessments have confirmed that SRM techniques, in theory, could slow, stop or reverse global temperature increases. Thus, if effective, they may reduce the impacts on biodiversity from warming, but there are high levels of uncertainty about the impacts of SRM techniques, which could present significant new risks to biodiversity. Modelling work consistently demonstrates that reduction in average global temperature (or prevention of further increase) and, to some extent, associated precipitation changes, would be possible, but would not fully restore future climatic conditions to their present day status. The regional distribution of temperature and precipitation effects are also different for different SRM techniques; these have been modelled, but many uncertainties remain. Even if, on average, the resulting disruptions to regional climates under SRM are less than those under climate change in the absence of SRM, this cannot be known with certainty: the possibility that some regions would benefit while others might suffer even greater losses, would have complex implications for governance. The implications for biodiversity have not been examined in most models. However, if SRM were to be started, but subsequently halted abruptly, termination effects (involving very rapid climate changes) would likely lead to serious losses of biodiversity. The use of CDR in addition to ‘moderate’ SRM could reduce such risks, and there is increasing emphasis in the scientific literature on the potential complementarity of the two approaches.

12. Models suggest SRM could slow the loss of Arctic sea ice. However, preventing the loss of Arctic sea ice through SRM is unlikely to be achievable without unacceptable climatic impacts elsewhere. Models suggest that even if SRM were globally deployed at a scale that returned average global temperatures to pre-industrial levels, Arctic sea ice loss would continue, albeit at a slower rate. Further loss of Arctic sea ice might be prevented by locally-strong SRM (using asymmetric application of stratospheric aerosols) but this would be associated with extremely negative impacts elsewhere due to major shifts in atmospheric and oceanic circulations. Cirrus cloud thinning may, in theory, be able to stabilize Arctic sea ice, but many uncertainties remain regarding that technique.

13. SRM may benefit coral reefs by decreasing temperature-induced bleaching, but, under high CO2 conditions, it may also increase, indirectly, the impacts of ocean acidification. Notwithstanding uncertainties over regional distribution, lowered average global temperatures under SRM would be likely to reduce the future incidence of bleaching of warm-water corals (compared to RCP 4.5, 6.0 or 8.0 conditions). The interactions between ocean acidification, temperature and impacts on corals (and other marine organisms) are complex, and much will depend on the scale of additional measures taken to reduce the increase in atmospheric CO2. If warming is prevented by SRM, there will be less additional CO2 emissions from biogeochemical feedbacks; however, relative cooling would reduce carbonate saturation state, that may reduce calcification or even dissolve existing structures (for cold-water corals) if CO2 emissions are not constrained.

14. The use of sulphur aerosols for SRM would be associated with a risk of stratospheric ozone loss; there would also be more generic side effects involved in stratospheric aerosol injection (SAI). While ozone depletion effects may be avoidable if alternative aerosols are used, their suitability and safety have yet to be demonstrated. All SAI techniques would, if effective, change the quality and quantity of light reaching the Earth’s surface; the net effects on productivity are expected to be small, but there could be impacts on biodiversity (community structure and composition).

15. The climatic effectiveness of marine cloud brightening depends on assumptions made regarding microphysics and cloud behaviour. Many associated issues are still highly uncertain. The potential for regional-scale applications has been identified; their environmental implications, that include salt damage to terrestrial vegetation, have not been investigated in any detail.

16. Large scale changes in land and ocean surface albedo do not seem to be viable or cost-effective. It is very unlikely that crop albedo can be altered at a climatically-significant scale. Changing the albedo of grassland or desert over sufficiently large areas would be very resource-demanding, damaging to biodiversity and ecosystems, and likely cause regional-scale perturbations in temperature and precipitation. Changes in ocean albedo (through longlasting foams) could, in theory, be climatically effective, but would be also accompanied by many biogeochemical and environmental changes, likely to have unacceptably large ecological and socioeconomic impacts.

Techniques aimed at increasing heat loss

17. Cirrus cloud thinning may have potential to counteract climate change, but the feasibility and potential impacts of the technique have received little attention. This technique would allow more heat (long-wave radiation) to leave the Earth, in contrast to SRM (which aims to reflect incoming short-wave energy).

Socioeconomic and cultural considerations

18. Recent social science literature has focussed on framing, governance and ethical issues relating to atmospheric SRM. Research has also covered international relations, national and international law, and economics, with most papers by US and European authors. While the socioeconomics of large-scale, land-based CO2 removal techniques has, to some degree, been covered by discussion on biofuels and their implications for food security, there are major gaps regarding the commercial viability of CDR techniques, such as BECCS, their associated institutional frameworks relating to carbon trading or tax incentives, and evaluations of environmental impacts (incontext of ecosystem services) and implications for indigenous and local communities. For SRM, many different frames have been considered, with those based on ‘climate emergencies’ or ‘tipping points’ attracting particular interest and criticism. There is an increasing trend towards multidisciplinary and transdisciplinary programmes on climate geoengineering, and these are now beginning to deliver more integrated analyses, with a collaborative role for social scientists.

19. Where surveyed, the public acceptability of geoengineering is generally low, particularly for SRM. Nevertheless, studies in a range of countries have found broad approval for research into both CDR and SRM techniques, provided that the safety of such research can be demonstrated. Regulatory framework

20. An amendment to the London Protocol to regulate the placement of matter for ocean fertilization and other marine geoengineering activities has been adopted by the Contracting Parties to the London Protocol. This relates to the 1996 Protocol to the Convention on the Prevention of Marine Pollution by Dumping of Wastes and Other Matter, 1972 administered by the International Maritime Organization. The amendment, adopted in 2013, is structured to allow other marine geoengineering activities to be considered and listed in a new annex in the future if they fall within the scope of the Protocol and have the potential to harm the marine environment. The amendment will enter into force following ratification by two thirds of the Contracting Parties to the London Protocol. This amendment, once entered into force, will strengthen the regulatory framework for ocean fertilization activities and provide a framework for the further regulation of other marine geoengineering activities. The CBD COP, in decision XII/20, took note of Resolution LP.4(8) and invited parties to the London Protocol to ratify this amendment and other Governments to apply appropriate measures in line with this amendment, as appropriate.

21. The 2007 amendment to the OSPAR Convention which allows storage of carbon dioxide in geological formations under the seabed of the North-East Atlantic entered into force in July 2011 and is currently in force for 11 of the 16 OSPAR parties.

22. As noted in the original report, the need for science-based global, transparent and effective control and regulatory mechanisms may be most relevant for those geoengineering techniques that have a potential to cause significant adverse transboundary effects, and those deployed in areas beyond national jurisdiction and in the atmosphere. These would comprise a subset of the techniques included in the broad definition of climate geoengineering (para 1, above). Many ocean-based potential geoengineering approaches are already covered under the LC/LP, as noted above. However, the large-scale BECCS and afforestation proposed in many IPCC AR5 scenarios may raise new regulatory issues at the international level regarding the associated scale of land use change. The potential international governance implications of such large-scale BECCS have so far not been specifically addressed by the international regulatory framework or literature.

23. The lack of regulatory mechanisms for SRM remains a major gap. With regard to SRM, IPCC AR5 notes that “the governance implications…are particularly challenging”, specifically in respect of the political implications of potential unilateral action. The spatial and temporal redistribution of risks raises additional issues of intragenerational and inter-generational justice, which has implications for the design of international regulatory and control mechanisms. The ethical and political questions raised by SRM would require public engagement and international cooperation in order to be addressed adequately. Other approaches that involve modifications to the atmospheric environment include cirrus cloud thinning are also not covered. 24. A recurring question is how research activities (as opposed to potential deployment) should and could be addressed by a regulatory framework. However, once the modelling and laboratory stage has been left behind, the distinction between research and development could become difficult to draw for regulatory purposes. It has been argued that governance can have an enabling function for “safe and useful” research; the London Protocol’s concept of “legitimate scientific research” underlying the 2013 amendment can be seen in this context.

24. A recurring question is how research activities (as opposed to potential deployment) should and could be addressed by a regulatory framework. However, once the modelling and laboratory stage has been left behind, the distinction between research and development could become difficult to draw for regulatory purposes. It has been argued that governance can have an enabling function for “safe and useful” research; the London Protocol’s concept of “legitimate scientific research” underlying the 2013 amendment can be seen in this context.

25. These developments have not changed the validity of the key messages from Part II of the 2012 report, including that “the current regulatory mechanisms that could apply to climate-related geoengineering relevant to the Convention do not constitute a framework for geoengineering as a whole that meets the criteria of being science-based, global, transparent and effective” and that “with the possible exceptions of ocean fertilization experiments and CO2 storage in geological formations, the existing legal and regulatory framework is currently not commensurate with the potential scale and scope of climate related geoengineering, including transboundary effects.”

Conclusions

26. Biodiversity is affected by a number of drivers of change that will themselves be impacted by proposed CDR and SRM geoengineering techniques. If effective, geoengineering would reduce the impacts of climate change on biodiversity at the global level. However, in the case of SRM under conditions of high CO2 this would not necessarily be the case at local levels, due to an inherently unpredictable distribution of temperature and precipitation effects. On the other hand, the benefits for biodiversity of reducing climate change impacts through large-scale biomass-based CO2 removal seem likely to be offset, at least in part, and possibly outweighed, by land use change. Changes in ocean productivity through large-scale fertilization would necessarily involve major changes to marine ecosystems, with associated risks to biodiversity. In general, technique-specific side effects that may be detrimental for biodiversity are not well understood.

27. Assessment of the direct and indirect impacts (each of which may be positive or negative) of climate geoengineering is not straightforward. Such considerations necessarily involve uncertainties regarding technical feasibility and effectiveness; scale dependencies; and complex comparisons with non-geoengineered conditions as well as value judgements and ethical considerations. Technique-specific considerations important for the evaluation of climate geoengineering techniques include effectiveness, safety and risks; co-benefits; readiness; governance and ethics; and cost and affordability. Many of these factors cannot yet be reliably quantified, and it is important that ‘cost’ includes both market and non-market values. Further research, with appropriate safeguards, could help to reduce some of these knowledge gaps and uncertainties.