Unless emissions are dramatically reduced, many of the direct impacts of climate change will likely be locked in by 2040, and could become so severe they go beyond the limits of what many countries can adapt to.
3.1 Approach: how to understand the impacts presented
This chapter presents impact indicators of the most significant global and regional direct risks of climate change, under the RCP4.5 GHG concentration trajectory, which approximately maps to the emissions trajectory identified in Chapter 2. The quantitative analysis of each direct risk and associated impact (Figures 3–18) is entirely sourced from Nigel Arnell and colleagues (2019). Whereas they present the major climate hazards as a function of time, with the associated climate impacts illustrated at discrete time horizons (2050 and 2100), in this paper we draw on their decadal data to present – for the first time – these climate risk impacts as function of time.
The approach of presenting the major climate risk impacts as a function of time partially follows the approach laid out by Simon Sharpe (2019):
The difference between Sharpe’s approach and the one used in this paper pertains to the thresholds applied to a given climate impact. Defining ‘what it is that we most wish to avoid’ requires geographically specific thresholds of concern/impact, as societies across different regions do not necessarily have equivalent vulnerabilities to a given climate hazard. For instance, an equivalent temperature in two regions could be defined as a severe heatwave in one while having little impact in another. As such, thresholds of concern/impact need to be defined by stakeholders in each region to enable an assessment of what it is we wish to avoid, and subsequently its likelihood over time. This bottom-up approach to assessing climate risks and impacts requires significant stakeholder engagement across all regions of the world, to then work backwards and identify the climatic conditions that would bring them about.
For this paper, a top-down approach to assessing direct climate impacts is followed, with standardized thresholds applied across all regions. For instance, the threshold for defining a major heatwave requires the temperature of a given region to exceed the 99th percentile of the reference period for four or more consecutive days in one year. As such, the direct risks and associated impacts in this chapter should be treated as indicators of impact. As highlighted by Sharpe (2019), it is integral to define what we wish to avoid in assessing the risks of climate change. The thresholds of impact for each indicator are described in the coming sections, as each impact is presented, but for fuller detail see the work of Arnell and colleagues (2019).
The risks of climate change can be understood more clearly when research starts by identifying what it is that we most wish to avoid and then assesses its likelihood as a function of time (Sharpe, 2019).
There is a clear need for subsequent research to engage stakeholders in all regions of the globe to define geographically specific thresholds of concern/impact. However, the top-down approach used here does enable impact indicators to be assessed under a common emissions scenario (as described in Chapter 2).
Climate hazards do not translate neatly into impacts. Impacts require exposure and vulnerability to be defined in order to quantify the impact of any given hazard. Shifts in population, innovation, advances in healthcare, and infrastructure will all alter the vulnerability and exposure of societies to a given climate hazard. Exposure is represented by shared socio-economic pathway 2 (SSP2). While five SSPs are commonly used, SSP2 has been selected on the basis that: (a) it represents a similar trend in decarbonization action that the emissions trajectory indicates; (b) it represents a middle ground along the spectrum of challenges for mitigation and adaptation. Just as with the selection of RCP4.5, using SSP2 to quantify climate impacts does not represent a prediction, but rather a plausible projection. SSP2 is characterized in full by O’Neill and colleagues (2017), but is summarized as:
While direct risk impacts are graphically represented (Figures 3–18) at a given indicator threshold of impact, plotting their likelihood as a function of time, the 2040–50 time horizon is generally used to highlight the impact within the text. This allows readers to make a comparative assessment of risks between geographies and impact types.
The uncertainty associated with the following impacts of direct climate risks (under RCP4.5) set out in this chapter has two principal sources: (1) the change in global mean temperature, which in turn is a function of equilibrium climate sensitivity, ocean diffusivity and carbon cycle feedback; and (2) the spatial pattern of change in temperature and precipitation. The total uncertainty is represented by the shaded areas in Figures 3–18, with the lower and upper bounds of that area indicating the 10th and 90th percentiles of the distribution of impacts in each year. These bounds can be regarded as the low and high estimates of the given impact, with the solid line representing the median, or central, estimate. As discussed in Box 1, the high or upper estimate represents the plausible worst-case scenario (under RCP4.5). However, these do not capture the behaviour of the tails of the probability distribution – e.g. what is the maximum plausible number of days of heatwaves – and so almost certainly under-represent the plausible worst case.
Whereas a wide range of hazards and impacts were assessed by Arnell and colleagues (2019), both at the continental and regional levels, this paper presents the impacts to which the greatest number of people or cropland are exposed. It also takes into consideration impacts with the greatest increases relative to historic baselines, and where the avoidance of climate change significantly reduces a given impact.