[We the authors decided to put this paper up on the EA Forum as we believe it is highly relevant to global catastrophic risk and global priorities research. This a 'preprint' of the accepted for publication version. 

The journal Futures has provided 50 days' free access to the final, published version of the article. Anyone clicking on this link before March 24, 2021 will be taken directly to the article on Futures, which they are welcome to read or download. No sign up, registration or fees are required.  After March 24th, this is a permalink to the paper.]

Assessing Climate Change’s Contribution to Global Catastrophic Risk

Simon Beard,^[1],[2] Lauren Holt,¹ Shahar Avin,¹ Asaf Tzachor,^1,[3] Luke Kemp,^1,[4] Phil Torres,[5] and Haydn Belfield^1,[6]

Many have claimed that climate change is an imminent threat to humanity, but there is no way to verify such claims. This is concerning, especially given the prominence of some of these claims and the fact that they are confused with other well verified and settled aspects of climate science. This paper seeks to build an analytical framework to help explore climate change’s contribution to Global Catastrophic Risk (GCR), including the role of its indirect and systemic impacts. In doing so it evaluates the current state of knowledge about catastrophic climate change and integrates this with a suite of conceptual and evaluative tools that have recently been developed by scholars of GCR and Existential Risk. These tools connect GCR to planetary boundaries, classify its key features, and place it in a global policy context. While the goal of this paper is limited to producing a framework for assessment; we argue that applying this framework can yield new insights into how climate change could cause global catastrophes and how to manage this risk. We illustrate this by using our framework to describe the novel concept of possible 'global systems death spirals,’ involving reinforcing feedback between collapsing sociotechnological and ecological systems. 

Climate Change; Global Catastrophic Risk; Planetary Boundaries; Food Security; Conflict

“Understanding the long-term consequences of nuclear war is not a problem amenable to experimental verification – at least not more than once" Carl Sagan (1983)

With these words, Carl Sagan opened one of the most influential papers ever written on the possibility of a global catastrophe. “Nuclear war and climatic catastrophe: Some policy implications” set out a clear and credible mechanism by which nuclear war might lead to global catastrophe by triggering a nuclear winter. While concerns about global threats from nuclear explosions had been raised since the Manhattan Project; they had previously been either vague (Russell 1961) or misplaced (Teller et al. 1946). Thus, even though Sagan’s thesis, and the models behind it, remain the subject of considerable scientific controversy; they transformed the way nuclear risk was understood and addressed. 

Today, global catastrophic climate risk research mirrors the condition of nuclear risk research in the early 1980s. Many climate scientists, global catastrophic risk scholars, and environmental groups are seeking to raise awareness of the severe risk climate change poses to humanity. However, despite the well-developed consensus on the science of climate change and wide-ranging discussions about the threats it poses to individuals, communities, and nations, there remains no clear and credible mechanism for how a changing climate could cause global catastrophes. 

This paper aims to improve our understanding of climate risk by building an analytical framework to assess its contribution to global catastrophic risk. It begins by reviewing the current state of global catastrophic climate risk research and its limitations (section 1) and then describes a new set of conceptual and evaluative tools developed for improved assessment of Global Catastrophic Risk (section 2). Finally, it highlights some initial insights that can be gained from combining these tools and suggests questions for future research, most notably concerning the possibility of positive feedback loops between collapsing sociotechnological and ecological systems, which we refer to as ‘global systems death spirals,’ (section 3) and the obstacles and opportunities to mitigating this risk (section 4). 

1 – Existing Assessments of Climate Change’s Contribution to Global Catastrophic Risk

Global Catastrophic Risk (GCR) refers to risks with “the potential to inflict serious damage to human well-being on a global scale” up to and including human extinction (Bostrom and Cirkovic 2008). Such risks are characterized both by their direct impacts, e.g. taking the lives of a significant portion of the human population, and their indirect impacts, such as leaving survivors at heightened risk by undermining global resilience systems (Avin et al. 2018).[7] There are at least three overlapping scenarios that would constitute such a catastrophe. First, a large and sudden reduction in the global population (Cotton-Barratt et al. 2016 use a 10% threshold for this); second, a collapse of the global systems that facilitate social complexity and co-operation (henceforth “human civilization”); third, a permanent reduction in humanity’s potential for technological, scientific, moral, and cultural development.

Given the significant amount of scientific attention that has been given to anthropogenic climate change and its effects, we know surprisingly little about its contribution to GCR. Recent studies have tended to be both vague in their assertions and inconsistent in their conclusions: Some, such as David Wallace-Well’s (2019) book The Uninhabitable Earth, acknowledge this possibility but shy away from considering it since doing so involves too much speculation; while others, such as Jem Bendall’s widely read (2018) paper, “Deep Adaptation: A Map for Navigating Climate Tragedy”, go too far in the opposite direction, speculating wildly about disaster scenarios without credible evidence. These flaws are especially troubling given that commentators often conflate unfounded assertions about the catastrophic risk of climate change with the settled corpus of climate science.

Some scholars have avoided this problem by confining themselves to what we know most about and focus on widely accepted limits to humanity’s ability to survive the direct effects of climate change, such as heat stress (e.g. Halstead 2018). These could become serious issues for certain regions after about 7°C of global warming from pre-industrial levels, and for most of the world’s population after 11°C (Sherwood and Huber 2010). Such scenarios are not impossible, as one recent assessment noted: “[o]n the highest emissions pathway ([Representative Concentration Pathway] 8.5), a rise of 7°C is a very low probability at the end of this century, but appears to become more likely than not during the course of the 22nd century. A rise of more than 10°C over the next few centuries cannot be ruled out." (King et al. 2015) However, this would require either that high levels of greenhouse gas (GHG) emissions continue far into the future or that natural feedback mechanisms are far stronger than expected. Other impacts of similar severity could be triggered by far smaller temperature increases.

One notable study published in the Proceedings of the National Academies of Science suggests that a global temperature rise of more than 3°C would be “catastrophic” while a rise of more than 5°C, which is above anything seen during the past 20 million years, would “pose existential threats to a majority of the population” from deadly heat and sea-level rise. The authors’ models suggest that within the next eight decades, there is a 5% chance of exceeding 5 degrees of warming (Xu and Ramanathan 2017).

Similarly, the 2015 book Climate Shock analyzes the uncertainty in standard climate models and finds a 3% chance of passing 6°C under an ambitious “low-medium emissions pathway” and an 11% chance of passing it under a more realistic “medium-high emissions pathway”. While stating that we cannot know the full implications of such a temperature rise, the authors describe it as an “indisputable global catastrophe” (Wagner and Weitzman 2015).

Finally, a more alarmist, and non-peer-reviewed, report by the National Centre for Climate Restoration considered a range of future scenarios and argued that even global temperature increases of 3-4 °C “will drive increasingly severe humanitarian crises, forced migration, political instability and conflict” and “may result in ‘outright chaos’ and an end to human civilization as we know it”. Citing the findings of Reilly et al. (2015), it estimates a 50% chance of crossing this threshold, even if commitments under the Paris Agreement are met (Dunlop and Spratt 2017).^[8]

These studies differ in their assessments of both the likelihood and impact of different levels of warming, and they fail to address the questions of why or how this could produce a global catastrophe or how we should respond. This largely reflects the difficulty of these questions, which relate to complex interacting global systems. However, it also highlights some limitations of current climate risk analysis.

First, studies tend to assess climate change primarily in terms of global mean temperature change, which is only one, albeit very important, aspect of this risk. Human emissions of GHGs also affect many other aspects of the global system, including rainfall patterns, ocean acidity, extreme weather, and the balance of energy between the upper and lower atmosphere and the oceans, while local effects can differ markedly from the global average. To assess climate change’s impact on humanity we must think holistically about all its impacts (Briggs, Kennel and Victor 2015).

Secondly, these assessments all build on existing climate and integrated assessment models. These are sophisticated scientific tools that represent our best, albeit limited, understanding of how climate change could affect human societies. However, they are poorly suited to studying GCRs as they may underestimate the upper limit for the damage that climate change could cause. One problem is that existing models are best-calibrated for scenarios close to the status quo and are widely acknowledged to perform poorly when applied to more extreme scenarios (Pindyck 2013, Weaver et al. 2017). More importantly, these models focus on the direct effects of a changing climate, while the greatest risks from climate change may be due to its indirect effects. 

Finally, these assessments are unidirectional: moving from assumed emissions pathways, to climate projections, to impact assessments. They thus pay insufficient attention to how the impacts of climate change will be shaped by human responses to it, which may worsen or alleviate the risks. For instance, societies may respond with renewed cooperation and accelerated emissions reductions, with unilateral geoengineering projects and technical fixes or with draconian measures to secure control of dwindling resources and habitable land. At best, climate risk researchers focus on the limits to adaptation (Dow et al. 2013), but this needs to be coupled with a comprehension of social and environmental changes in a dynamic state of co-evolution.

Addressing these deficiencies requires anchoring impact assessment to the expected social and ecological responses to climate change (Travis 2010). This is particularly pertinent when considering the level of GCR. Social and ecological collapse and other elements of a global catastrophe involve non-linear systemic shifts. Understanding these requires an assessment of their complex causes and this demands that we think about future trajectories in more than just climatic terms. Such an assessment cannot be carried out without grasping how and when social and environmental responses might translate into a global catastrophe.

When scientists began studying nuclear winter, their first step was to demonstrate that this was a real and catastrophic possibility. Only then did they attempt to calculate the necessary number and location of nuclear explosions that could trigger it; climate risk researchers should emulate this strategy. The assessments we reviewed did both too little, in failing to specify what a climate catastrophe would be or how it might unfold, and too much, in seeking to link this risk solely and directly to projected future emissions. Instead, we should consider how a climate catastrophe might occur and work backwards to understand all the conditions that could induce one. The next section describes one way to perform such an analysis.[9]

2 –Studying Climate Change from the Perspective of Global Catastrophic Risk

We should not underestimate humanity’s resilience and adaptability. Humanity survives, and even thrives, in most environments; including, for short periods, at the bottom of the ocean or the vacuum of space. To present a credible risk of producing a global catastrophe, any change would need to be at least one of the following:

• So profound that adaptation is impossible;
• So rapid that adaptation is unfeasible;
• So complex that the level of coordination needed to adapt to it is unachievable;
• Able to work against our efforts to adapt to it, or to adapt itself to us.

Historically, assessments of GCR have revolved around identifying changes that are sufficiently profound and/or rapid to pose a direct threat to humanity, such as asteroid impacts or nuclear war; neither is likely to be the case for GHG-induced climate change. However, recent years have seen the development of new conceptual and evaluative tools for assessing slow-moving disasters, civilizational vulnerabilities and systems collapse that open new pathways for the study and management of complex and dynamic catastrophes like climate change. In this section, we set out a framework that brings together these tools into a three-stage model for risk assessment. In the first step, the broader context of climate change is explored by moving away from reductive measures of ‘global warming’ to more holistic assessments of the state of the earth system and how these might impact on humanity, via the Boundary Risk for Humanity in Nature framework. The second stage then focuses on the human impacts of climate change by combining two broad classificatory schemes that seek to explore the entire risk space it occupies. Finally, the model places these risks into a policy context, by understanding that global risks exist because we fail to mitigate them, and setting out some of the obstacles that might prevent us from doing so. 

2.1 Linking Planetary Boundaries to Global Catastrophic Risk

One suite of conceptual tools builds on the well-known ‘planetary boundaries’ framework for assessing threats to the stability of the earth system (Steffen et al. 2015). This concept provides a useful tool for studying the broader systemic effects of climate change and humanity's other environmental impacts. Its empirical basis lies in nine observed ’boundaries’ of relative environmental stability provided by the Holocene epoch, in which human civilization has arisen, and how our planet is moving beyond these boundaries. While it provides some indication of where environmental threats to human civilization are likely to emerge, on its own, it says little about their impact on societies. ^[10]

For this reason, Seth Baum and Itsuki Handoh have sought to expand the planetary boundaries concept by creating a framework called “Boundary Risk for Humanity and Nature” (BRIHN). This assesses the risk that crossing planetary boundaries will incur an irreversible loss for humanity (Baum and Handoh 2014). Their framework is based around the twin concepts of “resilience” (humanity's ability to adapt to changes in the global systems that surround us) and its “probabilistic threshold” (the degree of change over which the risk of our resilience being insufficient to avoid an irreversible loss moves from a near impossibility to a near certainty). At present, this framework remains underdeveloped and only informally applied.

A similar, more evaluative, approach was taken by researchers at the UCLA Institute for Environment and Sustainability (Kareiva and Carranza 2018). They argue that crossing a planetary boundary is most dangerous when it is likely to produce reinforcing environmental feedback loops and multiplicative stresses (for instance when there are multiple spread mechanisms and obstacles to mitigation, see below). Yet, through a study of popular media, they find that most individuals expect global catastrophes to occur through scenarios involving simple and direct threats to life.

2.2 Classifying the Features of Global Catastrophic Risks

A second set of tools provides a classification of key features of risks for further analysis, applying lessons from risk analysis and disaster studies to global catastrophes. Researchers at the Centre for the Study of Existential Risk (CSER) have developed a scheme that classifies GCRs into three key components: “(i) a critical system (or systems) whose safety boundaries are breached by a potential threat, (ii) the mechanisms by which this threat might spread globally and affect the majority of the human population, and (iii) the manner in which we might fail to prevent or mitigate both (i) and (ii)” (Avin et al. 2018).

A critical system is one whose ordinary operation plays a crucial role in supporting humanity’s ability to survive in its current form. The safety boundaries of these systems are the limit or scale of disturbance that could disrupt their normal functioning and trigger a significant reduction in the support they provide (thus generalizing the principles behind planetary boundaries to cover all kinds of systems). Avin et al. (2018) distinguish seven levels of critical systems and order these according to their dependence on, and emergence from, one another (See Figure 1). 

Figure 1: The hierarchy of critical global systems (adapted from Avin et al. (2018))

When the safety boundaries of critical systems are breached, and the systems enter an abnormal or failed state, this can have cascading effects with the potential to spread disruption to other critical systems. The mechanisms of this spread include: “natural global scale” mechanisms, such as changes to biogeochemical stocks and flows; “anthropogenic network” mechanisms, such as the global energy distribution network; and mechanisms involving biological or informational replication, such as the spread of pests or ideologies. 

Classifying risks by their critical systems and spread mechanisms provides an analytical tool for studying systemic risks without reducing their inherent complexity and helps identify policy levers, and other opportunities, for mitigating them. However, GCRs also present obstacles to designing and implementing such strategies. These form the third pillar of this scheme, which identifies many kinds of obstacle to mitigation, ranging across four levels: ‘individual,’ ‘interpersonal,’ ‘institutional,’ and ‘beyond-institutional.’

Hin-Yan Liu and colleagues (2018) built on this framework by incorporating additional insights from the field of disaster studies. They argue that GCRs consist of three distinct components: hazards/threats (the “external source of peril” i.e. what harms humanity), vulnerabilities (“propensities or weaknesses inherent within human social, political, economic or legal systems, that increase the likelihood of humanity succumbing to pressures or challenges that threaten existential outcomes” i.e. how humanity could come to harm), and exposures (“the number, scope and nature of the interface between the hazard and the vulnerability” i.e. why humanity is in harm’s way). Individual vulnerabilities and exposures can be categorized along the following lines:

• Humanity is ontologically exposed or vulnerable because humans live on earth and can only exist within certain parameters (e.g. Wet Bulb Temperatures below 35^OC).
• Humanity is intentionally exposed or vulnerable because humans actively seek out ways to do large amounts of damage (e.g. nuclear weapons).
• Humanity can be either actively or passively vulnerable, depending on whether this is ‘because’ or ‘in spite’ of our existing social structures.
• And humanity can be either directly or indirectly exposed to harm, depending on the complexity of the causal connection involved.

The authors draw two key lessons from their approach. Firstly, the study of GCRs has focused too heavily on the importance of ‘existential hazards’ (such as AI, biotechnology, or indeed climate change) and now also needs to assess existential vulnerabilities and exposures. Secondly, the mitigation of GCS’s has tended to focus on technical solutions; however, since vulnerabilities and exposures are social phenomena, global catastrophic risk must also be amenable to legal and governance-based solutions, including legal institutions such as rights, responsibility and societal relations.

2.3 Integrating the Assessment of Governance and Global Catastrophic Risk

Thus, we see that taking the broader classificatory approach to understanding global catastrophic risk leads us to the need to understand these risks within a policy context. The third set of tools that we want to include in our framework meets this need by combining risk assessment with insights from ethics and policy evaluation to assess humanity’s ability to respond to potential global catastrophes. Indeed, both theoretical models and historical precedents show that people can either elevate or reduce the emergence and spread of catastrophes depending on how they respond to them. 

Karin Kuhlemann (2019) has argued that climate change is one of a group of “unsexy” risks characterized by 1) declining access to resources that 2) results from collective action and 3) can be expected to pass a point at which societies will become unable to satisfy the minimum requirements of well-being within the life expectancy of the youngest members of society (what she calls a “threshold of significance”). She argues that these risks are, by their nature, less compelling than “sexy” risks, which are “neat, quick and techy,” such as nuclear war or biotechnological threats - echoing the conclusions of Kareiva and Carranza (2018). Unsexy risks also tend to be harder to predict and monitor and are more challenging to address because they relate to basic human functions such as consumption and reproduction.

Nick Bostrom (2018) also proposes a model of GCRs based on governance failures, this time related to ‘civilizational vulnerabilities’ that are inherent in our “semi-anarchic default condition” (a world order characterized by a limited capacity for preventive policing, a limited capacity for global governance, and diverse motivations among state and non-state actors). Under such conditions, Bostrom argues that humanity faces an existential vulnerability to two classes of threat (each of which can be split into two further sub-classes):

1. Technologies that make it too easy for individuals or small groups with the appropriate motivation to cause mass destruction, so that it is either:
   1. extremely easy to cause a moderate amount of harm, or
   2. moderately easy to cause an extreme amount of harm (these vulnerabilities were first extensively described in Torres 2018a, 2018b).
2. A level of technology that strongly incentivizes actors to use their powers to cause mass destruction, so that either:
   1. powerful actors can produce civilization‐devastating harms and face incentives to use that ability, or
   2. a great many actors face incentives to take some slightly damaging action such that the combined effect of those actions is civilizational devastation (see also Torres 2018c). ^[11]

Bostrom does not suggest that climate change in its current form is a significant contributor to GCR because he does not believe it will cause enough damage. However, he concedes that global warming does present actors with the kind of incentive structure described under 2b, and suggests possible scenarios under which it could produce a global catastrophe. 

It is worth noting that when both Kuhlemann and Bostrom discuss climate change they focus on the challenges of coordinating very many actors who each have incentives to do small amounts of harm. However, there is another side to the problem that may be even harder to resolve, namely, that a small number of actors (such as fossil fuel companies) currently face strong economic and political incentives to do a great deal of harm.

3 – Applying These Tools

Together, these tools provide a framework for better analyzing different kinds of complex, and or slow-moving, contributors to GCR. In the following two sections we illustrate how such a framework might be applied to combine diverse evidence sources of evidence for assessing risk. This section will use on the first two stages of our framework to describe one plausible catastrophe scenario (which we label the Global System Death Spiral), while the next section uses the final stage to analyse opportunities and obstacles to mitigating this risk.

3.1 Climate Change and Planetary Boundaries

While most of the impacts of climate change so far have fallen within the range of what was experienced during the Holocene, the rate of change is faster than in the Holocene and we are now beginning to see climate change push beyond these boundaries. In the latest edition of the planetary boundaries’ framework, climate change is placed in the zone of increasing risk, implying that while this boundary has been breached, there remains some potential for normal functioning and recovery (Steffen et al. 2015). It thus lies between what the authors identify as the ‘safe zone’ and other ‘high risk’ transgressions, such as disruption to the biochemical flows of nitrogen and phosphorus and loss of biosphere integrity.

As part of their discussion of BRIHN Baum and Handoh (2014) note that climate change is the planetary boundary for which the risk to humanity has received most meaningful consideration and they suggest that this attention is deserved. Yet little research attention has been paid to climate change’s extreme or catastrophic effects. Kareiva and Carranza (2018) argue that, despite currently falling outside of the area of high risk, climate change has the clear potential to push humanity across a threshold of irreversible loss by “changing major ocean circulation patterns, causing massive sea-level rise, and increasing the frequency and severity of extreme events… that displace people, and ruin economies.” Even if humanity was resilient to each of these individual impacts, a global catastrophe could occur if these impacts were to occur rapidly and simultaneously. 

One scenario that has received comparatively more attention is that of the global climate crossing a tipping point that would trigger environmental feedback loops (such as declining albedo from melting ice or the release of methane from clathrates) and cascading effects (such a shifting rainfall patterns that trigger desertification and soil erosion). After this point, anthropogenic activity may cease to be the main driver of climate change, making it accelerate and become harder to stop (King et al. 2015).

Other scenarios can be discerned from the numerous historical cases in which the modest, usually regional, climatic changes experienced during the Holocene have been implicated in the collapse of previous societies, including the Anasazi, the Tiwanaku, the Akkadians, the Western Roman Empire, the lowland Maya, and dozens of others (Diamond 2005, Fagan 2008). These provide a precedent for how a changing climate can trigger or contribute to societal breakdown. At present, our understanding of this phenomena is limited, and the IPCC has labelled its findings as “low confidence” due to a lack of understanding of cause and effect and restrictions in historical data (Klein et al. 2014). Further study and cooperation between archaeologists, historians, climate scientists and global catastrophic risk scholars could overcome some of these limitations by identifying how the impacts of climate change translate into social transformation and collapse, and hence what the impacts of more rapid and extreme climatic changes might be. There is also the potential for larger studies into how global climate variations have coincided with collapse and violence at the regional level (Zhang 2005; 2006). However, these need to be interpreted and generalized with care given the differences between pre-industrial and modern societies.

Societies also have a long history of adapting to, and recovering from, climate change induced collapses (McAnay and Yoffee 2009). However, there are two reasons to be sceptical that such resilience can be easily extrapolated into the future. First, the relatively stable context of the Holocene, with well-functioning, resilient ecosystems, has greatly assisted recovery, while anthropogenic climate change is more rapid, pervasive, global, and severe. Large-scale states did not emerge until the onset of the Holocene (Richerson et al, 2001), and societies have since remained in a surprisingly narrow climatic niche of roughly 15 mean annual average temperature (Xu et al, 2020). A return to agrarian or hunter-gatherer lifestyles could thus have more devastating and long-lasting effects in a world of rapid climate change and ecological disruption (Gowder 2020).^[12] Second, modern human societies may have developed hidden fragilities that amplify the shocks posed by climate change (Mannheim 2020) and the complex, tightly-coupled and interdependent nature of our socio-economic systems makes it more likely that the failure of a few key states or industries due to climate change could cascade into a global collapse (Kemp 2019b).

A third set of plausible scenarios stem from climate change’s broader environmental impacts. Apart from being a planetary boundary of its own, Steffen et al. (2015) point out that climate change is intimately connected with other planetary boundaries (see Table 1). Climate change is thus identified by the authors as one of two ‘core' boundaries with the potential “to drive the Earth system into a new state should they be substantially and persistently transgressed.” This transformative potential was elaborated on in subsequent work exploring how the world could be pushed towards a ‘Hothouse Earth’ state, even with anthropogenic temperature rises as low as 2°C (Steffen et al. 2018).

Table 1: Relationship of climate change to other planetary boundaries (after table S3 in Steffen et al. 2015)

The connection between climate change and biosphere integrity (the survival of complex adaptive ecosystems supporting diverse forms of life) is particularly strong. The IPCC is highly confident that climate change is adversely impacting terrestrial ecosystems, contributing to desertification and land degradation in many areas and changing the range, abundance and seasonality of many plant and animal species (Arneth et al. 2019). Similarly, the Intergovernmental Science-Policy Platform on Biodiversity and Ecosystem Services (IPBES) has reported that climate change is restricting the range of nearly half the world’s threatened mammal species and a quarter of threatened birds, with marine, coastal, and arctic ecosystems worst affected (Diaz et al. 2019). According to one estimate, climate change could cause 15-37% of all species to become ‘committed to extinction’ by mid-century (Thomas et al. 2004). 

Disruption to biosphere integrity can have profound economic and social repercussions, ranging from loss of ecosystem services and natural resources to the destruction of traditional knowledge and livelihoods. For instance, desertification, which threatens a quarter of Earth’s land area and a fifth of the population, is already estimated to cost developing nations 4-8% of their GDP (United Nations 2011). Many other rapid regime shifts involving loss of biosphere integrity have been observed, including shifts in arid vegetation, freshwater eutrophication, and the collapse of fish populations (Amano et al. 2020). There is a theoretical possibility of still more profound regime shifts at the global level (Rocha et al. 2018). However, the contribution of loss of biosphere integrity to GCR is yet to be assessed. Kareiva and Carranza (2018) argue that it is unlikely to threaten human civilization, due both to a lack of plausible mechanisms for this threat and the fact that “local and regional biodiversity is often staying the same because species from elsewhere replace local losses.” However, in their classification of GCRs, Avin et al. (2018) suggest the potential for ecological collapse to threaten the safety boundaries of multiple critical systems with diverse spread mechanisms at a range of scales, from the biogeochemical and anatomical to the ecological and sociotechnological. Note that both these studies were conducted for largely conceptual purposes and should not be taken as rigorous analyses of this risk, this topic warrants further investigation.

3.2 Classifying Climate Change’s Contributions to Global Catastrophic Risk

Climate change's contribution to GCR goes well beyond its impact on the earth system. Taking Avin et al.’s list of critical systems, we note that previous studies have mostly focused on the effects of climate change on physical and biogeochemical systems (e.g. global temperature and sea-level rise) or the lower-level critical systems that are most directly related to human health and survival (e.g. Heath Stress). However, these represent a very limited assessment of risk as it only accounts for climate change as a direct hazard/threat and our "ontological" vulnerabilities to it. A more comprehensive risk assessment must consider the higher-order critical systems threatened by climate change passively (through a lack of alternatives) and actively (through intentional design).

The probability of a global catastrophe is higher when sociotechnological and environmental systems are tightly coupled, creating a potential for reinforcing feedback loops. If environmental change produces social changes that perpetuate further environmental change, then this could actively work against our efforts at adaptation. When this change has the potential to produce significant harm, via human vulnerabilities and exposure, we describe such loops as ‘global systems death spirals.’ These spirals could produce self-perpetuating catastrophes, whereby the energy and resources required to reverse or adapt to collapse are beyond the means of dwindling human societies. Feedback loops like this could thus create tipping points beyond which returning to anything like present conditions would become extremely difficult. Global systems would shift to very different states in which the prospects for humanity would likely be bleaker. 

In the rest of this section, we explore just one potential spiral, between an ecological system (the biosphere) and two sociotechnological systems (the human food and global political systems). We explore each system and its interactions. Figure 2 illustrates our model of this spiral.

Figure 2: A global systems death spiral

The human food system

Climate change’s impact on biosphere integrity (discussed in the previous section) could harm the human food system due to loss of ecosystem services, disruption of the cycles of water, nitrogen and phosphates, and changes in the dynamics of plant and animal health (Bélanger and Pilling 2019). Crossing this planetary boundary is already having severe implications for global food security, including loss of soil fertility and insect-mediated pollination (Diaz et al. 2019).

Systems for the production and allocation of food are already enduring significant stress. The sources of stress include climate change, soil erosion, water scarcity, and phosphorus depletion. The natural resource base, arable land and freshwater upon which food production rely are being degraded. While global food productivity and production has increased dramatically over the past century to meet rising demand from an expanding global population and rising standard of living, these constraints and risks are increasing the vulnerability of our global food supply to rapid and global disruptions that could constitute global catastrophes (Baum et al. 2015). 

Climate change will further reduce food security in at least three interconnected ways. First, it will affect growing conditions, including direct threats to agricultural yields from heat, humidity, and precipitation in many regions; although initially improving conditions in some (Lott, Christidis and Stott 2013). Second, it will increase the range of agricultural pests and diseases (Harvell 2002). Third, it will increase the occurrence of extreme weather events that impair the integrity of food production and distribution networks, from production to harvest, post-harvest, transport, storage, and distribution, thereby increasing our vulnerability and exposure to supply shocks (Bailey et al. 2015). The IPCC estimates, with medium confidence, that at around 2°C of global warming the risk from permafrost degradation and food supply instabilities will be ‘very high’, while at around 3°C of global warming the risk from vegetation loss, wildfire damage, and dryland water scarcity will also be very high (Arneth et al. 2019). Very few studies have considered the impacts of 4°C of global warming or more; however, the IPCC highlighted one study finding that any potential agricultural gains from climate change will be lost by this point and there could be a decrease of 19% in maize yields and 68% in bean yields in Africa, an 8% reduction in yields in South Asia, and a substantial negative impact on fisheries by 2050 (Porter et al. 2014). Furthermore, multiple extreme weather events could disrupt food distribution networks (Baily and Wellesley 2017).

While there are opportunities to adapt, disruption to the entire global food system cannot be resolved via food aid alone. Indeed, there is the potential for isolationist or heavy-handed responses that would do more harm than good. Given the high degree of interconnectivity and feedback within the global food system, our initial research suggests that any one of these climate change effects could trigger scenarios that would critically undermine the global food system’s ability to meet the minimum nutrition for well-being; making food security for all an unachievable goal, let alone rise to the challenge of continuing to grow (Tzachor 2019a; 2020); this would constitute what Kuhlemann (2019) terms a ‘threshold of significance.’

The global political system

Disrupting the global food system can create and exacerbate conflict and state failure (Brinkman and Hendrix 2011). However, once again, this needs to be seen against the backdrop of a global political system under stress, with climate change as a significant contributing factor. Climate change influences political systems in many ways, from being a locus of activism and a stimulus for reform to driving rising inequality and population displacement (Arneth et al. 2019, Diffenbaugh and Burke 2019). This is not a new phenomenon, changes in the climate are believed to have contributed to conflict between people and states throughout human history, driven by resource scarcity, population displacement, and inequality (Lee 2009, Mach et al. 2019). As part of a comprehensive risk assessment of climate change, David King and colleagues (2015) conducted an extensive literature review on climate change and conflict and used this to inform a series of international wargaming exercises. These found that climate change is expected to increase international conflict while highlighting the role that population displacement, state failure, and water and food insecurity would play in this (see also Natalini, Jones and Bravo 2015, Mach et al. 2019).

Quantitative studies of the impact of climate change on violence and conflict have provided more mixed results. A survey of empirical studies by Detges (2017) found that there may be multiple differing trends: extreme weather events appear to have more significant effects on violence than do long-term climate trends, while levels of small-scale conflict and interpersonal violence appear to be more affected than large-scale conflicts and international war. Empirical studies also highlight how climate change’s impact on conflict is predominantly as a risk multiplier and intensifier. Thus, climate change may contribute more by increasing our vulnerability to other conflict-inducing factors, such as loss of livelihood, forced migration, environmental change, and food insecurity, than by acting as a direct cause of conflict (Schubert et al. 2008, Hsiang et al. 2013, Abel et al. 2019).[13]

Of particular relevance to GCR is the effect of climate change on the risk of nuclear war (Parthemore, Femia and Werrell 2018). However, to our knowledge, this has never been rigorously assessed, although the potential is certainly there. One recent model of the risk of nuclear war highlighted how varied, and common, incidents with the potential to trigger a nuclear exchange are (Baum, de Neufville and Barrett 2018). It outlined 14 different causal pathways to an exchange, including the escalation of conventional wars and international crises, human error, and the emergence of new non-state actors. For all but two of these, they identify historical examples of potentially precipitating incidents, with 60 incidents in total (i.e. a little less than one a year). This suggests that the absence of nuclear war was less due to a lack of potential causes, tan the global political system’s ability to defuse them. Thus, the real significance of climate change may be its capacity to undermine this system: the combination of social, political, and environmental disruption, a lingering sense of global injustice, and rising food, water, and energy insecurity could increase the probability that crises escalate or that false alarms are mistaken for genuine emergencies. This topic needs further research.

3.3 The emergence of a global systems death spiral

Yet, we should not conclude that a nuclear exchange is the only, or even most likely, scenario in which political instability might produce a global catastrophe. Conflict and political instability, even of moderate severity, are themselves two of the most significant drivers of biodiversity loss due to breakdowns in monitoring, governance, and (public and private) property rights (Baynham-Herd et al. 2018). This closes a potentially reinforcing feedback loop between loss of biosphere integrity, food insecurity and political breakdown. 

The mechanisms by which these cascading failures might spread include many of the natural, anthropogenic, and replicator effects identified by Avin et al. (2018), making them harder to contain. At the natural level, climate change involves changes to the global atmospheric and biogeochemical systems and poses other naturally spreading harms, like global ecological collapse. At the anthropogenic level, the global interconnectedness of sociotechnological systems means that while small shocks are easier to recover from, larger shocks can be harder to contain and control. Finally, biological and informational replication can also spread the negative impacts of climate change, from vector-borne diseases and invasive species to climate fatalism and dangerous geoengineering technologies.

Given these numerous spread mechanisms, critical system failures could precipitate global catastrophes. Furthermore, the spiral we have explored is unlikely to be the only set of interlinked systemic disruptions that climate change could initiate (other death spirals could involve bio-insecurity and disease), nor are these the only causal connections between these three systems. Until we understand the nature of such death spirals better, we must act cautiously. We now turn to consider what this would mean.

4 – Opportunities and Obstacles for Reducing Climate Change’s Contribution to Global Catastrophic Risk

Many of the opportunities for managing GCR are common to different drivers of risk, including climate change and other environmental problems, dual-use applications of bio- cyber- and nuclear technology, and naturally occurring disasters like asteroid impacts and volcanic super-eruptions. They can be split into three levels (Cotton-Barratt, Daniel, and Sandberg 2020): 

1. Preventing events that could precipitate global catastrophes (by identifying hazards, understanding their dynamics, and fostering cooperation on matters of safety through dedicated institutions or beneficial customs); 
2. Responding to such events so that they do not precipitate global catastrophes (by detecting them early, reducing the lag between detection and response, ensuring that responses aren’t stymied by cascading impacts, and identifying leverage points to maximize impact); and 
3. Developing resilience so that the worst effects of global catastrophes are avoided (by enhancing diversity and redundancy in critical systems, preparing large scale evacuation and recovery infrastructure, and planning late-stage responses to deploy under worst-case scenarios.

However, there are many obstacles to realizing these opportunities, which range across the individual, interpersonal, and supra-institutional scales. Avin et al. (2018) identified many such obstacles, which we summarize in Table 2.

Table 2: A Classification of obstacles to mitigating GCR (after figure 3 in Avin et al. 2018) 

Building on these opportunities and obstacles we can identify a range of risk management policies that can help to reduce the level of GCR across the board: 

• Overcoming institutional barriers to risk prevention. Some recent proposals for achieving this have included: improving governments understanding of GCRs by building national risk registers, improving information flows between scientists and policymakers, and training civil servants in futures analysis (Sepasspour 2019); representing future generation’s in policymaking in ways that avoid capricious party politics (Jones et al. 2018); and harnessing international diplomacy to build trust and collaboration on emerging issues of common interest (Farquhar et al. 2017).
• Preparing for a rapid institutional response to global systems death spirals and other non-linear events by introducing balancing feedback and adaptive management systems. For instance, alongside existing efforts at climate diplomacy, states could negotiate ‘tail risk treaties’ that pre-commit them to robust responses should circumstances turn out to be worse than expected (Kemp and Rhodes 2020).
• Enhancing individual, interpersonal, and institutional resilience by supporting heterogeneity, modularity and redundancy across ecosystems, cultures, and economies. (Kareiva and Carranza 2018) This encourages experimentation and adaptation while reducing dependency on conditions being ‘just right’ for a small number of client species or economic institutions. Some pathways to achieve this include well-worn proposals made by critics of contemporary political economy, such as the need to de-emphasize economic growth as an objective of societies or to promote viability rather than optimality of solutions to global challenges. More generally they involve supplementing (but not necessarily replacing) existing global governance mechanisms with principles of adaptive design. (Kreienkamp and Pegram 2020)

Such policies present further opportunities for research as they are currently under-explored, at least among academics and policymakers.[14]However, alongside these general considerations, our framework can also be used to identify some obstacles and opportunities that are relatively unique to climate change’s particular contribution to GCR. 

The first of these arises from the ability of classification schemes like the two we set out above to disaggregate the space of mitigation opportunities as these relate to reducing hazards, vulnerabilities and exposures around many critical systems and global spread mechanisms. Considering only the three critical systems that  make up the global death spiral we describe in the previous section it is immediately clear that we have many opportunities to mitigate this risk:

Table 3: Some proposals for reducing climate changes contribution to GCR

One clear take away from this table is that not only are the policy options for mitigating climate change’s contribution to GCR potentially numerous but, as Liu and colleagues (2018) note, they are often “solutions that will be of more general benefit to humanity as tangential effects of efforts taken to reducing our collective vulnerability and exposure.” However, this should hardly surprise us as many, if not most, of the policies for mitigating climate risk are such ‘no-regrets’ actions, which would likely be worth pursuing even in the absence of climate change. This is especially true of energy efficiency measures and vehicle electrification (GCEC, 2018). There is also a mounting case that emissions reductions are generally a net economic benefit once ‘co-benefits,’ such as improvements in health, productivity, and employment, are accounted for, for instance, in the United States coal-fired generation creates health costs that are 0.8-5.6 times the added economic value (Muller et al, 2015, West et al, 2013). Decarbonization and a coal phase-out thus appear to be prudent, no-regrets options even without a consideration of climate change; especially given the already low projected costs of meeting net-zero targets due to the plummeting cost of renewable energy (Jotzo and Kemp, 2015).[15] The problem of cutting emissions is not a prisoner’s dilemma; it is a problem of framing. Of ensuring that co-benefits are included in economic analysis and public debate. 

This can make it challenging to identify and address those aspects of global risk that correspond to climate change’s unique contribution to GCR. However, a reductive approach to assessing and managing climate risk in terms of individual problems and solutions can miss important systemic interactions and effects, such as the potential for global system death spirals we have identified here. This is a non-trivial issue for risk classification and communication, but one that is already being addressed in areas such as extreme weather (Mann, Lloyd and Oreskes 2017). We believe that further work on bringing together the diverse indirect and systemic impacts of climate change to a unified assessment framework (as we have started to do here) is the best way to address this problem

The second way of harnessing the tools in our framework to maximize the opportunities (and overcome the obstacles) of mitigating climate risk relates to its assessment of the global policy context. In particular, the slow emergence and global nature of climate change’s impacts make it hard to decide when actions need to be taken to prevent possible future catastrophes.[16] It may be that at least some of the impacts we discussed above will be overcome by future technologies, such as genetic engineering and carbon capture and storage, so that we could potentially  ‘overshoot’ with our efforts to avoid them, imposing costly policies that stifle the very process of technological development that could otherwise have saved us. However, our assessment of the global policy context suggests that this is unlikely. As Karin Kuhlemann notes, slowly emerging risks are invariably easier to ignore than more explosive ones, encouraging people to question whether they should be the generation to respond until it is too late to avoid disaster. Shifting baselines for what is normal also make it harder to identify when a threshold of significance has been crossed. Similarly, as Nick Bostrom argues, the semi-anarchic state of global governance offers poor incentives for the development and implementation of risk mitigation technologies and supports a culture of risk denial and inaction. This also means that much of the cost of climate risk mitigation is likely to be borne by entrenched interests that are least able to provide the kind of creativity that we need to overcome them. Thus, while technology can and will play a crucial role in addressing climate change’s contribution to GCR, it is extremely unlikely that we will do more harm than good in encouraging more aggressive climate mitigation policies.

5 – Conclusion 

There is much we do not know about climate change’s contribution to GCR, including future impacts at temperatures beyond 3°C and humanity’s resilience to climatic disruption. The dynamics and tipping points of global ecological and social systems also remain highly uncertain. This is especially true given the lack of attention being paid to creating regenerative and resilient global systems, while the drive for economic growth and efficiency is making these systems more complex, tightly coupled, and vulnerable. However, we can, and should, do more to assess and manage climate change’s contribution to GCR, and as our framework shows the indirect and systemic impacts of climate change are no less important to this assessment than its direct effects.

Climate change should therefore be studied alongside other contributors to GCR, such as nuclear war and biological threats, and further research is needed both to understand the scale of its contribution and how to manage it. New conceptual and evaluative tools that are being developed by global catastrophic risk scholars point the way towards a fuller research agenda. As the framework we have developed here shows, these tools can contribute to the proper assessment of risk, whether by GCR scholars or experts in climate change more generally, and in future work, we hope to contribute to this assessment. The impacts of climate change are global and complex; a meaningful understanding of the plausible worst-case scenarios will require the concerted efforts of many researchers across the GCR, climate science and policy communities.

References 

Abel, Guy J., Michael Brottrager, Jesus Crespo Cuaresma, and Raya Muttarak. "Climate, conflict and forced migration." Global Environmental Change 54 (2019): 239-249.

Amabile, Teresa M., and Steven J. Kramer. "The power of small wins." Harvard Business Review 89, no. 5 (2011): 70-80.

Arneth, A., H. Barbosa, T. Benton, K. Calvin, E. Calvo, S. Connors, A. Cowie, E. Davin, F. Denton, and R. van Diemen. IPCC special report on climate change, desertification, land degradation, sustainable land management, food security, and greenhouse gas fluxes in terrestrial ecosystems. IPCC (2019).

Avin, Shahar, Bonnie C. Wintle, Julius Weitzdörfer, Seán S. Ó hÉigeartaigh, William J. Sutherland, and Martin J. Rees. "Classifying global catastrophic risks." Futures 102 (2018): 20-26.

Bailey, R., T. G. Benton, A. Challinor, J. Elliott, D. Gustafson, B. Hiller, A. Jones et al. Extreme weather and resilience of the global food system: Final project report from the UK-US Taskforce on extreme weather and global food system resilience. UK: The Global Food Security Programme (2015).

Baum, Seth D. "Winter-safe deterrence: The risk of nuclear winter and its challenge to deterrence." Contemporary Security Policy 36, no. 1 (2015): 123-148.

Baum, Seth, Robert de Neufville, and Anthony Barrett. "A model for the probability of nuclear war." Global Catastrophic Risk Institute Working Paper (2018).

Baum, Seth D., David C. Denkenberger, Joshua M. Pearce, Alan Robock, and Richelle Winkler. "Resilience to global food supply catastrophes." Environment Systems and Decisions 35, no. 2 (2015): 301-313.

Baum, Seth D., and Itsuki C. Handoh. "Integrating the planetary boundaries and global catastrophic risk paradigms." Ecological Economics 107 (2014): 13-21.

Baum, Seth D., Timothy M. Maher, and Jacob Haqq-Misra. "Double catastrophe: Intermittent stratospheric geoengineering induced by societal collapse." Environment Systems & Decisions 33, no. 1 (2013): 168-180.

Baynham-Herd, Zachary, Tatsuya Amano, William J. Sutherland, and Paul F. Donald. "Governance explains variation in national responses to the biodiversity crisis." Environmental Conservation 45, no. 4 (2018): 407-418.

Bélanger, J., and D. Pilling. The State of the World’s Biodiversity for Food and Agriculture. FAO Commission on Genetic Resources for Food and Agriculture Assessments: Rome, Italy (2019): 572.

Bostrom, Nick. "The vulnerable world hypothesis." Global Policy 10, no. 4 (2019): 455-476.

Bostrom, Nick, and Milan M. Cirkovic, eds. Global Catastrophic Risks. OUP, (2008).

Briggs, Stephen, Charles F. Kennel, and David G. Victor. "Planetary vital signs." Nature Climate Change 5, no. 11, (2015): 969. 

Brinkman, Henk-Jan, and Cullen S. Hendrix. Food Insecurity and Violent Conflict: Causes. Consequences, and Addressing the Challenges, World Food Programme (2011).

brown, adrienne marie. Emergent Strategy. AK Press, (2017). 

Cotton‐Barratt, Owen, Max Daniel, and Anders Sandberg. "Defence in Depth Against Human Extinction: Prevention, Response, Resilience, and Why They All Matter." Global Policy 11, no. 3 (2020): 271-282.

Cotton-Barratt, Owen, Sebastian Farquhar, John Halstead, Stefan Schubert, and Andrew Snyder-Beattie. Global Catastrophic Risks, 2016. The Global Challenges Foundation, Stockholm (2016).

Detges, Adrien. Climate and conflict: reviewing the statistical evidence. A summary for policy-makers. Adelphi, Berlin (2017).

Denkenberger, David, and Joshua M. Pearce. Feeding everyone no matter what: Managing food security after global catastrophe. Academic Press, 2014.

Diaz, S., H. T. Ngo, M. Guèze, J. Agard, A. Arneth, P. Balvanera, K. Braumpan, S. Butchart, K. Chan, L. Garibaldi, K. Ichii, J. Liu, S. M. M. Subramanian, G. Midgley, P. Milosloavich, Z. Molnár, D. Obura, A. Pfaff, S. Polasky, A. Purvis, J. Rzzaque, B. Reyers, R. R. Chowdhury, Y. Shin, I. Visseren-Hamakers, K. Willis, C. Zayas. Summary for policymakers of the global assessment report on biodiversity and ecosystem services of the Intergovernmental Science-Policy Platform on Biodiversity and Ecosystem Services. IPBES (2019).

Diffenbaugh, Noah S., and Marshall Burke. "Global warming has increased global economic inequality." Proceedings of the National Academy of Sciences 116, no. 20 (2019): 9808-9813.

Dow, Kirstin, Frans Berkhout, Benjamin L. Preston, Richard JT Klein, Guy Midgley, and M. Rebecca Shaw. "Limits to adaptation." Nature Climate Change 3, no. 4 (2013): 305.

Dunlop, Ian, and David Spratt. Disaster Alley: Climate Change Conflict & Risk. Breakthrough National Centre for Climate Restoration, Melbourne. (2017).

Fagan, Brian. The great warming: Climate change and the rise and fall of civilizations. Bloomsbury, London (2008).

Farquhar, Sebastian, John Halstead, Owen Cotton-Barratt, Stefan Schubert, Haydn Belfield, and Andrew Snyder-Beattie. Existential risk: Diplomacy and governance. Global Priorities Project, Oxford (2017).

Gjerde, Jon, Sverre Grepperud, and Snorre Kverndokk. "Optimal climate policy under the possibility of a catastrophe." Resource and energy economics 21, no. 3-4 (1999): 289-317.

GCEC. 2018. Global Commission on the Economy and Climate. New Climate Economy Report 2018: Unlocking the Inclusive Growth Story of the 21st Century: Accelerating Climate Action in Urgent Times. Global Commission on Economy and Climate

Gowdy, John. "Our hunter-gatherer future: Climate change, agriculture and uncivilization." Futures 115 (2020): 102488.

Halstead, John. "Stratospheric aerosol injection research and existential risk." Futures 102 (2018): 63-77.

Head, Brian W. "Wicked problems in public policy." Public policy 3, no. 2 (2008): 101.

Hsiang, Solomon M., Marshall Burke, and Edward Miguel. "Quantifying the influence of climate on human conflict." Science 341, no. 6151 (2013): 1235367.

Jones, Natalie, Mark O’Brien, and Thomas Ryan. "Representation of future generations in United Kingdom policy-making." Futures 102 (2018): 153-163.

Jotzo, Frank and Luke Kemp. 2015. “Australia Can Cut Emissions and the Cost is Low”. World Wildlife Fund Australia, Canberra. 

Kareiva, Peter, and Valerie Carranza. "Existential risk due to ecosystem collapse: Nature strikes back." Futures 102 (2018): 39-50.

Kemp, L. “Civilisational collapse has a bright past – but a dark future.” Aeon (2019): https://aeon.co/ideas/civilisational-collapse-has-a-bright-past-but-a-dark-future.

Kemp, L, and Rhodes, C. "The Cartography of Global Catastrophic Governance." Global Challenges Foundation (2020)

Khagram, S., Clark, W., & Raad, D. F. (2003). From the environment and human security to sustainable security and development. Journal of Human Development, 4(2), 289-313.

Kunreuther, Howard, Geoffrey Heal, Myles Allen, Ottmar Edenhofer, Christopher B. Field, and Gary Yohe. "Risk management and climate change." Nature Climate Change 3, no. 5 (2013): 447-450.

King, D, Daniel Schrag, Zhou Dadi, Qi Ye, and Arunabha Ghosh. Climate change: A risk assessment: Report commissioned by the Foreign and Commonwealth Office. Centre for Science and Policy, Cambridge (2015).

Klein, Richard TJ, G. F. Midgley, B. L. Preston, M. Alam, F. G. H. Berkhout, K. Dow, and M. R. Shaw, et al. "Adaptation opportunities, constraints and limits. Impacts, Adaptation and Vulnerability." In Climate Change 2014, Cambridge University Press, (2014): 899-943.

Kreienkamp, Julia, and Tom Pegram. "Governing Complexity: Design Principles for the Governance of Complex Global Catastrophic Risks." International Studies Review (2020).   

Kuhlemann, Karin. "Complexity, creeping normalcy and conceit: sexy and unsexy catastrophic risks." foresight 21, no. 1 (2019): 35-52.

Lee, L. “A brief history of climate change and conflict.” Bulletin of Atomic Scientists Online (2009): https://thebulletin.org/2009/08/a-brief-history-of-climate-change-and-conflict/.

Liu, Hin-Yan, Kristian Cedervall Lauta, and Matthijs Michiel Maas. "Governing Boring Apocalypses: A new typology of existential vulnerabilities and exposures for existential risk research." Futures 102 (2018): 6-19. 

Lott, Fraser C., Nikolaos Christidis, and Peter A. Stott. "Can the 2011 East African drought be attributed to human‐induced climate change?" Geophysical Research Letters 40, no. 6 (2013): 1177-1181.

Lovelock, James. Novacene: The coming age of hyperintelligence. MIT Press, 2019.

Mach, Katharine J., Caroline M. Kraan, W. Neil Adger, Halvard Buhaug, Marshall Burke, James D. Fearon, Christopher B. Field et al. "Climate as a risk factor for armed conflict." Nature, (2019).

Manheim, David. "The Fragile World Hypothesis: Complexity, Fragility, and Systemic Existential Risk." Futures (2020).

Mann, Michael E., Elisabeth A. Lloyd, and Naomi Oreskes. "Assessing climate change impacts on extreme weather events: the case for an alternative (Bayesian) approach." Climatic change 144, no. 2 (2017): 131-142.

McAnany, Patricia A., and Norman Yoffee, eds. Questioning collapse: human resilience, ecological vulnerability, and the aftermath of empire. Cambridge University Press, Cambridge (2009).

Meehl, Gerald A., Warren M. Washington, William D. Collins, Julie M. Arblaster, Aixue Hu, Lawrence E. Buja, Warren G. Strand, and Haiyan Teng. "How much more global warming and sea level rise?." science 307, no. 5716 (2005): 1769-1772.

Muller, Nicholas Z, Robert Mendelsohn, and William Nordhaus. 2011. “Environmental Accounting for Pollution in the United States Economy”. American Economic Review 101(5): 1649-1675. 

Mitchell, Audra, and Aadita Chaudhury. "Worlding beyond ‘the end’ of ‘the world’: white apocalyptic visions and BIPOC futurisms." International Relations (2020): 0047117820948936

Natalini, Davide, Aled Jones, and Giangiacomo Bravo. "Quantitative assessment of political fragility indices and food prices as indicators of food riots in countries." Sustainability 7, no. 4 (2015): 4360-4385.

Ng, Yew‐Kwang. "The importance of global extinction in climate change policy." Global Policy 7, no. 3 (2016): 315-322.

Pindyck, Robert S. "Climate change policy: what do the models tell us?" Journal of Economic Literature 51, no. 3 (2013): 860-72.

Parthemore, Christine, Francesco Femia, and Caitlin Werrell. "The global responsibility to prepare for intersecting climate and nuclear risks." Bulletin of the Atomic Scientists 74, no. 6 (2018): 374-378.

Porter, John Roy, Liyong Xie, Andrew J. Challinor, Kevern Cochrane, S. Mark Howden, Muhammed Mohsin Iqbal, David B. Lobell, and Maria Isabel Travasso. "Food security and food production systems." In Climate Change 2014, Cambridge University Press, (2014): 485-533.

Reilly, John, S. Paltsev, E. Monier, H. Chen, A. Sokolov, J. Huang, Q. Ejaz, J. Scott, J. Morris, and A. Schlosser. Energy and climate outlook: perspectives from 2015. MIT Joint Program on the Science and Policy of Global Change. (2015).

Richerson, Peter J., Robert Boyd, and Robert L. Bettinger. "Was agriculture impossible during the Pleistocene but mandatory during the Holocene? A climate change hypothesis." American Antiquity 66, no. 3 (2001): 387-411.

Rocha, Juan C., Garry Peterson, Örjan Bodin, and Simon Levin. "Cascading regime shifts within and across scales." Science 362, no. 6421 (2018): 1379-1383.

Russell, Bertrand. Has man a future?. George Allen & Unwin, London (1961).

Sagan, Carl. "Nuclear war and climatic catastrophe: Some policy implications." Foreign Affairs 62, no. 2 (1983): 257-292.

Schubert, R., H. J. Schellnhuber, N. Buchmann, A. Epiney, R. Grießhammer, M. Kulessa, D. Messner, S. Rahmstorf, and J. Schmid. Climate Change as a Security Risk. German Advisory Council on Global Change (2008).

Sears, Nathan Alexander. "Existential Security: Towards a Security Framework for the Survival of Humanity." Global Policy 11, no. 2 (2020): 255-266.

Sepasspour, Rumtin. Managing global catastrophic risks Part 1: Understand. Centre for the Study of Existential Risk, Cambridge (2019).

Shafer, Glenn. A mathematical theory of evidence. Princeton university press, 1976.

Sherwood, Steven C., and Matthew Huber. "An adaptability limit to climate change due to heat stress." Proceedings of the National Academy of Sciences 107, no. 21 (2010): 9552-9555.

Steffen, Will, Katherine Richardson, Johan Rockström, Sarah E. Cornell, Ingo Fetzer, Elena M. Bennett, Reinette Biggs et al. "Planetary boundaries: Guiding human development on a changing planet." Science 347, no. 6223 (2015): 1259855.

Steffen, Will, Johan Rockström, Katherine Richardson, Timothy M. Lenton, Carl Folke, Diana Liverman, Colin P. Summerhayes et al. "Trajectories of the Earth System in the Anthropocene." Proceedings of the National Academy of Sciences 115, no. 33 (2018): 8252-8259.

Teller, E. E. J. Konopinski, and C. Marvin. Ignition of the atmosphere with nuclear bombs. Report LA-602. Los Alamos Laboratory, Los Alamos, NM (1946).

Thomas, Chris D., Alison Cameron, Rhys E. Green, Michel Bakkenes, Linda J. Beaumont, Yvonne C. Collingham, Barend FN Erasmus et al. "Extinction risk from climate change." Nature 427, no. 6970 (2004): 145

Torres, Phil. “Agential risks and information hazards: An unavoidable but dangerous topic?.” Future 95, (2018a): 86-97.

Torres, Phil. “Who would destroy the world? Omnicidal agents and related phenomena.” Aggression and Violent Behavior 39, (2018b): 129-138.

Torres, Phil. "Superintelligence and the Future of Governance: On Prioritizing the Control Problem at the End of History." In Roman Yampolskiy ed. Artificial Intelligence Safety and Security, pp. 357-374. Chapman and Hall (2018c).

Travis, William R. "Going to extremes: propositions on the social response to severe climate change." Climatic Change 98, no. 1-2 (2010).

Tzachor, A. “Down the Hunger Spiral: Pathways to the Disintegration of the Global Food System.” Council on Foreign Relations (2019a): https://www.cfr.org/blog/down-hunger-spiral-pathways-disintegration-global-food-system.

Tzachor, Asaf. "The future of feed: Integrating technologies to decouple feed production from environmental impacts." Industrial Biotechnology 15, no. 2 (2019b): 52-62.

Tzachor, A. Famine Dynamics. Global Relations Forum (2020). 

United Nations. Global drylands: a UN system-wide response. UN Environment Management Group. (2011).

Von Braun, Joachim. Food and financial crises: Implications for agriculture and the poor. Vol. 20. Intl Food Policy Research Institute, (2008).

Wagner, Gernot, and Martin L. Weitzman. Climate shock: The economic consequences of a hotter planet. Princeton University Press, (2015).

Wallace-Wells, David. The uninhabitable earth: Life after warming. Tim Duggan Books, (2019).

Weaver, C. P., Richard H. Moss, Kristie L. Ebi, Peter H. Gleick, Paul C. Stern, Claudia Tebaldi, Robyn S. Wilson, and J. L. Arvai. "Reframing climate change assessments around risk: recommendations for the US National Climate Assessment." Environmental Research Letters 12, no. 8 (2017).

West, Jason, et al. 2013. “Co-benefits of Global Greenhouse Gas Mitigation for Future Air Quality and Human Health.” Nature Climate Change 3(10): 885–889.

Weitzman, Martin L. "On modelling and interpreting the economics of catastrophic climate change." The Review of Economics and Statistics 91, no. 1 (2009): 1-19.

Wilson, Edward O. Half-earth: our planet's fight for life. WW Norton & Company, 2016.

Xu, Chi, Timothy A. Kohler, Timothy M. Lenton, Jens-Christian Svenning, and Marten Scheffer. "Future of the human climate niche." Proceedings of the National Academy of Sciences 117, no. 21 (2020): 11350-11355.

Xu, Yangyang, and Veerabhadran Ramanathan. "Well below 2 C: Mitigation strategies for avoiding dangerous to catastrophic climate changes." Proceedings of the National Academy of Sciences 114, no. 39 (2017): 10315-10323.

Yunkaporta, Tyson. Sand talk: How Indigenous thinking can save the world. Text Publishing, (2019).

Zhang, Dian, Chiyung, Jim., Chusheng, Lin., Yuanqing, He, and Fung, Lee. “Climate change, social unrest and dynastic transition in ancient China.” Chinese Science Bulletin 50(2) (2005): 137-144

Zhang, David., Jim, C., George, C.S., Yuan-quing, He., Wang, James J., and Lee, Harry, F. “Climatic change, wars and dynastic cycles in China over the last millennium.” Climatic Change 76 (2006): 459–477.

Footnotes

^[1] Centre for the Study of Existential Risk, University of Cambridge

^[2] Institutet för framtidsstudier

^[3] Global Food Security Research Centre, University of Cambridge

^[4] Fenner School of Environment and Society and Crawford School of Public Policy, Australia National University

^[5] Institute of Philosophy, Leibniz University Hannover

^[6] Although a co-authored work, in recognition of the extensive help and support that a wide range of scholars gave to preparing it, the overall argument of this paper about where climate change's greatest contribution to global catastrophic risk comes from and how to assess and manage this is primarily the work of the principal author, who is solely responsible for its shortcomings and defects. The authors are extremely grateful to Seth Baum, Peter Irvine, Peter Watson, Kristian Strommen, Alex Wong and two anonymous reviewers for their feedback and suggestions.

[7] It is important to note that these impacts may be distributed highly unequally. For instance, it is likely that the greatest number of victims from a climate induced global catastrophe, in terms of loss of life, health, and livelihood, will be among poorer people in marginalized communities; however, the global resilience systems that such a catastrophe would undermine currently benefit predominantly wealthy people in privileged societies. Such inequalities can have profound impacts on the way that potential catastrophes are perceived, studied, and managed and may contribute greatly to the current level of risk (Mitchell and Chaudhury 2020).

^[8] Other works also engage with the question of climate change and GCR, some of which are cited later in this paper. One literature concerns the economics of global catastrophic climate risk, though usually this involves mere assumptions of its likelihood (see Gjerde et al. 1999, Weitzman 2009, Ng 2016). Another concerns GCR and geoengineering (Baum et al. 2013, Halstead 2018)

[9] Note that what we offer here is merely a qualitative framework for evidence synthesis. We believe that such frameworks are useful aids for reasoning through risk assessment and management; However, they should not be confused with complete risk assessment solutions. To produce such a solution, it would be necessary to combine our framework with proper tools for evidential reasoning, such as that of Glenn Shafer (1976).

^[10] Researchers at CSER are exploring the potential of a companion ‘civilization boundaries’ framework, based on observations of primary drivers of historical social and ecological collapses.

^[11] Bostrom also defines a further, class-0, vulnerability, to technologies that carry a hidden risk such that the default outcome when they are discovered is inadvertent civilizational devastation. However, this stems from our epistemic default position, rather than the semi-anarchic condition of human civilization. Furthermore, David Mannheim (2020) has proposed another class of vulnerability that is also relevant to the risk from climate change, to a level of technology where systemic fragility increases to a point where the probabilistic threshold that a shock will lead to civilizational devastation approaches certainty.

^[12] Note that, even if anthropogenic GHG emissions were to cease immediately, the climate would continue changing for decades, if not centuries, to come (Meehl et al. 2005).

[13] Indeed, these factors may themselves not be directly conflict-inducing either. For instance, migration serves the purpose of enhancing diversity and resilience to environmental change in many species and is not intrinsically a source of conflict. However, bad policy responses and systemic inequality and racism have substantially reduced the benefits of migration in humans, while directly causing many of its harmful effects.

[14] On the other hand, intergenerational decision making, planning for worst case scenarios, and promoting heterogeneity are common practise across a wide range of indigenous communities and traditional societies.

[15] Other no-regrets policies include small wins and approaches to risk management that draw on locally-generated knowledge; work from the bottom up to identify appropriate risk management solutions and building consensus for their implementation; and seek resilience by mobilizing existing structures for adaptive work (Head 2008, Amabile and Kramer 2011). One place to find such approaches is in the diverse and growing fields of Afro-, Indigenous- and Diaspora-futurisms (brown 2017, Yunkaporta 2019, Mitchell and Chaudhury 2020). However, the value of this approach to risk management should in no way be seen as counting against the implementation of top-down policies where these are aimed at ensuring action from large-scale economic and political actors.

[16] Although at least some of this uncertainty could be overcome by applying robust decision-making tools for risk management under uncertainty to global scale tipping points across critical systems (Kenreuther et al. 2013)