This research programme into climate intervention must include considerations of low probability / high impact scenarios as well as the most probable. Whilst easily overlooked, they are essential for a more complete risk response. There are complex interactions, both positive and negative, between solar climate intervention (SCI) and the potentially small but nonetheless significant risk of ruinous damage to human wellbeing. It will be essential to assess both how each SCI technology increases this risk and could reduce it, and how further research could be done to enhance the latter without adding to the former. These must be some of the goals of the proposed scientific assessment.
The definition of goals in relevant areas of scientific research- overall
Addressing how SCI interacts with Global Catastrophic Risk (GCR) should be a goal of scientific research, both by assessment of this interaction and by development of SCI in such a way that it reduces GCR.
GCR refers to the risk of serious, ruinous or near-existential damage to human well-being on a global scale. Even if one believes that the probability of GCR is low it is prudent to take precautions because they are so high impact. This is a common aspect of public safety, such as in relation to natural disasters or technological failures, but has not habitually been part of the framing when thinking about climate risk. To do a more complete risk assessment, “betting on the best case” or a median case is insufficient, although it is prevalent in the literature on both climate change and SCI (Kemp et al 2022, Tang & Kemp 2021). Including GCR in a research programme requires considering plausible low probability / high impact outcomes of both climate change and SCI, taking into account complex risk assessments including compound hazards. Any consideration of these risks by this research programme would go a long way to improving the assessment and management of SCI risk.
Other topics relevant to climate intervention research - nurturing the development of different aspects of SCI research
When investigating GCR, some of our ordinary research methods have significant limitations. GCR involves extreme and unprecedented events, so conventional modelling, analysis, and observation, whilst useful, are often insufficient. Generation and exploration of extreme scenarios is key, as well as use of foresight techniques and horizon scanning. Once these scenarios have been generated, they can then be used to pose useful research questions where more conventional inquiry can explore key empirical uncertainties within them. This allows us to work to reduce the likelihood of these scenarios coming to fruition. The use of such GCR-specific methods may generate important empirical questions and goals of research which have not been generated thus far.
Scenarios, modelling and physical and engineering science research should contribute to understanding how SCI may interact with GCR and how research into SCI could reduce GCR. This may have impacts on a variety of sub-disciplines of SCI research, including but not limited to the following.
The definition of goals in relevant areas of scientific research- specific sub-disciplines
Engineering Science: Assess the vulnerabilities of proposed deployment infrastructure to low probability / high impact events, such as a large solar flare or volcanic super-eruption. Attempt to design more resilient deployment systems, including designing systems with redundancies to reduce the probability of termination shock (Irvine & Parker 2018) in such cases. Design systems that can be sustained for months with little exogenous input to make intermittency less likely in the aftermath of a different catastrophic event, such as pandemic.
Aerosol research: Investigate aerosol coagulation to investigate the upper limits of cooling possible from stratospheric aerosol injection (SAI), providing evidence for both how useful SAI could be at reducing climate risk in extreme warming scenarios and how severe termination shock could be. Investigate aerosol coagulation where there is a mix of aerosols, for example soot and sulfate aerosols, to investigate how SAI might be affected by aerosol injecting events such as nuclear winter or catastrophic wildfires. Investigate the reactions of proposed aerosols with ozone taking into account the possibility of less likely events such as space weather events. Investigate the limits of cooling that could occur from Marine Cloud Brightening in both a local and global context.
Climate modelling: Investigate the relationship between SCI and climate tipping points, both near term tipping points and those where near term climate risk may have long term effects. Observations would also be necessary to constrain these. Investigate targeted interventions into reducing tipping points, such as local MCB, but also model the impact of intermittency of these, their effectiveness in extreme scenarios, and analyse the vulnerability of such a system. Model termination shock when it occurs alongside sunlight blocking events, e.g. supervolcanoes, nuclear winter etc. Model SCI being used in more extreme warming scenarios, such as scenarios of high ECS/TCR, or in response to a rapid change in forcing through methane release or volcanic eruption, and how to make deployment schemes responsive to this.
Broader goals: Assessment of SCI should account for its impact on the risks to critical systems, such as the food system, and its impact on hazards, exposures and vulnerabilities to the overall connected system of global society. The investigation required to account for these impacts will be from a variety of fields. Many of these overlap with key questions from a non-GCR perspective, although the focus on plausible worse-case scenarios will also invoke different questions.
Summary of key recommendations
- Consideration of GCR should be a goal of any attempt to assess or combat climate risk, and thus must be present in this research programme
- Consideration of low probability / high impact scenarios should be a goal of climate modelling involving SCI
- Consideration of these scenarios will likely affect non-modelling investigation, both natural science and engineering
- These scenarios should be considered in terms of assessing the relationship between SCI and GCR and in trying to develop SCI in such a way that reduces GCR as much as possible
This comment is provided by Gideon Futerman (University of Oxford) and SJ Beard (University of Cambridge)