Impact of Climate Change on Natural Perils

Author: Christopher Allen

This is the third of six blogs within our Climate Explainer series; the aim of this series is to set a common basis of understanding of climate change for non-climate practitioners within the (re)insurance industry.

The key topics that will be addressed over the series include the impact of climate change, taxonomy of climate risk, the current regulatory landscape and misconceptions of climate risk. This blog looks at how climate change impacts natural perils.

Figure 1: Expected impact of climate change on different natural perils, showing the confidence the scientific community has in each case. Source: Gallagher Re.

The average annual natural catastrophe insured loss globally over the last 11 years (2011-2021) was US $71 billion¹. In most years, typically those without major earthquakes, climate perils (floods, tropical cyclones, windstorms, wildfires, severe convective storms and winter weather) are responsible for the vast majority of natural catastrophe insured loss. Understanding how climate change impacts climate perils is therefore of great interest to the insurance industry.

Different climate perils arise from different atmospheric conditions; therefore it is reasonable to expect that climate change will also impact each climate peril differently, which is indeed the case. In fact, research shows that climate change can also have a different impact on the same peril, depending on where in the world it occurs.

In addition to climate change having a different impact on different perils, confidence in our understanding also varies by peril because sometimes there are competing ‘drivers’ of change and we do not know which will become dominant, or simply because of climate model limitations in ability to simulate relevant processes (Figure 1). For instance, the impact of climate change on sea level rise is very well understood, while its impact on severe thunderstorms is much more uncertain. The blog will review the state of scientific understanding with respect to climate change for a selection of key insurance perils, starting with flood.

Inland Flooding

For every 1°C that the planet warms, the amount of moisture that the atmosphere can hold increases by about 7%. We can think of it becoming more ‘spongy’. Because of this physical law, the academic community has high confidence that climate change will increase the frequency and intensity of heavy precipitation over all continents – a bit like wringing out a wetter sponge. The temperature-moisture relationship means that the increase in the frequency of heavy precipitation will be larger for rarer events. For example, 10- and 50-year events will be approximately twice and three times as frequent, respectively, at the 4°C warming level than today². Figure 2 shows modelled precipitation change projections with different degrees of global warming.

Figure 2: Model projections of precipitation change with global warming from the IPCC Sixth Assessment Report²

There is a close correspondence between heavy precipitation and pluvial flooding (i.e. rainfall-driven surface flooding not associated with a water body such as a river). However, there is not a one-to-one relationship between increasing rainfall and increased inland flooding in general, because floods are affected by many additional factors, including but not limited to soil moisture and antecedent conditions, snowmelt, land use changes, drainage basin morphology, changes in evapotranspiration, and flood defenses (human and natural). Nonetheless, global hydrological models project with medium confidence that a larger fraction of land areas will be affected by an increase in river floods than by a decrease in river floods in the future². The picture is more complicated at a regional scale; for example, around the Mediterranean, research suggests that a decrease in the magnitude of the 100-year flood will occur³.

Coastal Flooding

One of the clearest signals in climate change projections is sea level rise. Indeed, global mean sea level increased by 0.2 metres between 1901 and 2018⁴. Depending on the emissions scenario, global mean sea level could rise by a further 0.3-1.0 metres by 2100 due to thermal expansion of the ocean and ice loss on land⁴. As a result, there is high confidence that coastal flooding will increase, especially in low-lying areas. A special case of coastal flooding is storm surge, which is mainly driven by high onshore winds at coastlines where water can be channeled onto land easily. With higher mean sea levels, storm surge risk increases, all else being equal. However, storm surge changes depend not only on sea level rise but also on changes to the frequency and intensity of storms themselves.

Tropical Cyclones

Figure 3: Left: Wind footprint (m/s) for Hurricane Irma (2017). Right: NCAR/Gallagher Re modelled wind footprint (m/s) for Hurricane Irma under a 2°C global warming scenario with 10% increase in maximum wind speed along the track.

Although sea surface temperatures – a vital ingredient for the formation of tropical cyclones – are increasing in the majority of ocean regions as a result of climate change, most models that do a reasonable job of resolving the dynamics of the tropical cyclone simulate a reduction or no change to their numbers globally as the earth warms². However, peak tropical cyclone wind speeds are expected to increase², as is the proportion of tropical cyclones that reach the highest intensities (Category 4 and 5 storms), by about 13% for 2°C of global warming⁵. Rainfall generated by tropical cyclones is also expected to increase, by about 12% for 2°C of global warming⁵, and by more in some regions due to increased low-level moisture convergence. In other words, from a global perspective, while tropical cyclone frequency may not change, or even decrease, tropical cyclone hazards are expected to become more severe (Figure 3 and Figure 4). Based on evidence from observations and model simulations, the scientific community has moderate to high confidence in these results.

Figure 4: Summary of changes in storm activity with global warming from the IPCC Sixth Assessment Report²

Windstorms (Extra-tropical cyclones)

While the scientific community is relatively confident in how climate change impacts tropical cyclones, there is more uncertainty when it comes to extra-tropical cyclones. Some studies⁶ find that windstorm wind speeds could show little change, others⁷ find that extreme wind speeds may increase, and others find that the frequency of extreme windstorms could decrease (Figure 4). The results are heavily region-dependent, and future migration of storm tracks poleward could lead to directionally opposing wind risk changes over short geographic distances. Finally, research results differ not just depending on the region analyzed but also on the climate models used for the study, with a high degree of spread in model results, even for the historical period⁸.

Other perils

Figure 5: The Bobcat Fire in Angeles National Forest in California on September 6, 2020. Source: https://www.noaa.gov/media-advisory/wildfire-season-and-fire-weather-resource-guide-for-reporters-and-media

In a short blog it is not possible to cover the impact of climate change on every natural peril, but it is worth briefly mentioning three more: heat waves, drought and wildfires. There is high confidence that heat waves and drought will worsen with climate change (Figure 1), with consequences including a rise in heat-related deaths, strain on water resources, crops, livestock, energy and transportation⁹. Aside from the impact on livelihoods, these changes also affect the insurance industry, in particular crop insurance, life and health insurance and in some cases business interruption coverages. With regard to wildfires, which have made particular headlines in the U.S. in recent years, the IPCC has medium confidence that fire weather conditions will increase by 2050 under moderate or high emissions scenarios in several regions of Africa, Australia, several regions of South America, Mediterranean Europe, and North America¹⁰. As air warms, it exerts an increased evaporative demand on the surface, drying it and making fuel more available for fires. If ignitions happen in these regions, particularly during windy, hot and dry weather, wildfires can grow very rapidly.

Moving forward

The global natural catastrophe protection gap in 2020 was 76% - in other words, only 24% of natural catastrophe risk globally was covered by insurance¹¹. As we have seen, climate change is increasing the frequency or severity of many natural perils, but the science is complex and, in some cases, lacks consensus. To reduce the protection gap, and improve coverage in exposed regions, we need to be aware of both how risk is changing and what confidence can be attributed to expected changes. Changing natural perils in the face of climate change certainly present the insurance industry with a challenge, but the industry is not new to dealing with uncertainty or tail risk after all.