Modelling the Impact of Climate Change
Climate models such as the ESM used by CoreLogic are computational simulations of the Earth’s climate, capturing physical processes of the atmosphere, ocean, land, and ice. They can be run to model both the past and current climate, as well as to predict the impact of emission scenarios from the present to the end of the 21st century, and beyond.
Coupled Model Intercomparison Project (CMIP)
The Coupled Model Intercomparison Project (CMIP) coordinates climate runs produced by models developed by teams from around the world. Models are constantly updated in terms of spatial resolution and inclusion/refinement of physical processes, leading to greater predictive certainty.
By combining the output from these models, to form a model ensemble, it enables a probabilistic prediction of the impacts of climate change, which is more robust than the results of any single model taken in isolation. This is the key strength of the ensemble.
Individual windstorms can be identified in each simulation and tracked through their lifetimes using objective algorithms. Hence, it is possible to build a catalogue of simulated storm tracks and associated winds at the surface. A similar procedure can be applied to historical reanalysis runs (such as ECMWF’s ERA dataset), leading to a current climate baseline against which climate change can be measured.
IPCC Assessment Reports & Climate Scenarios
The Intergovernmental Panel on Climate Change (IPCC) is the United Nations body responsible for assessing the science related to climate change.
The IPCC assessment reports consolidate the latest scientific consensus on climate change . The 4th Assessment Report (AR4), published in 2007, made use of a set of scenarios introduced in the IPCC Special Report on Emissions Scenarios (SRES). CoreLogic’s pioneering study of climate change impact on European windstorm risk (Activity Of Catastrophic Windstorm Events In Europe In The 21st Century) is based on the A1B and A2 SRES scenarios within that report. According to the AR4 report, the multi-model mean surface air temperature warming and associated uncertainty ranges for the period 2090 to 2099 relative to the historical period of 1980 to 1999 are as follows:
- A1B: +2.8°C (1.7°C to 4.4°C)
- A2: +3.4°C (2.0°C to 5.4°C)
This provides an end of century perspective on the accumulated temperature increase.
The next generations of IPCC Assessment Reports used improved methodologies to assess the impact of climate change. The 5th Assessment Report (AR5) from 2014, makes use of an updated set of scenarios known as Representative Concentration Pathways (RCPs). The latest report, AR6, published in 2021, uses the Shared Socioeconomic Pathways (SSPs).
The SSPs are projections of socioeconomic global changes up to the year 2100. Each SSP includes a specification of atmospheric CO₂ concentration as a function of time (Figure 2) where increasing concentrations leads to increasing mean global surface temperature.
Figure 2 CO2 concentrations (ppm) as a function of year under the Shared Socioeconomic Pathways (SSPs). Source: DKRZ
The best estimate of the temperature changes caused by each SSP are shown in Figure 3, reproduced from the AR6 report. For example, under SSP2-4.5 on a 20-year time horizon centred on 2050, the best estimate is a 2° C warming above the average global surface temperature of the period 1850-1900. Under SSP5-8.5 for the same time horizon the warming best estimate increases to 2.4°. By the end of the century (a horizon centred on 2090), SSP2-4.5 leads to a warming estimate of 2.7°, but SSP5-8.5 causes an increase of 4.4° C owing to the continued increased emissions under this scenario.
Figure 3 Source: IPCC, 2021: IPCC, 2021: Summary for Policymakers Climate Change 2021: The Physical Science Basis. Contribution of Working Group I to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change
Of particular importance to the insurance industry is the assessment of possible climate change on a maximum time horizon of around 30 years, which corresponds to the Mid-term 2041-2060 SSP horizon. Some scientific studies concentrate on the climate change impact on windstorm activity at the end of the century (i.e., the 2081-2100 horizon), rather than the preferred 2041-2060 SSP horizon. As an approximation, the impact of SSP2-4.5 at 2081-2100 can be treated as an upper approximation for the impact of SSP5-8.5 at 2041-2060 owing to the relatively small difference of 0.3° between the respective expected warmings (which are, in turn, a consequence of the similar C02 concentrations shown in Figure 2).
Results from the CMIP model runs are fed into the IPCC assessment reports, which consolidate the latest scientific consensus on climate change, including impacts on ETCs. The 6th Assessment Report (AR6) is the latest of these, published in 2021 and utilises the 6th phase of CMIP models, CMIP6. The runs included are based on SSP scenarios. The previous Assessment Report (AR5) using CMIP5 model run data according to RCPs presented impacts on ETCs. Broadly speaking the AR6 results are similar to AR5, however the CMIP6 models are more accurate.
For a given SSP, each CMIP model leads to its own temperature response, and in turn, produces a catalogue of simulated ETCs.
Having introduced climate models and climate change scenarios, in the next section we move on to consider the predicted impact of different scenarios on European windstorm hazard.