Climate change is upon us. But what are the emerging trends and what can the insurance industry do to mitigate the effects? Dr Steve Smith investigates.

One of the most widely discussed topics of the last ten to 20 years has been the environment and humanity's impact upon it. Global climate change is a catch-all term to describe changes in the global climate system due to anthropogenic sources. It is often colloquially called "global warming" or "greenhouse warming" but this is misleading since some aspects of humanity's impact upon the climate does not produce a warming effect. A case in point is depletion of ozone in the Antarctic.

However, it is becoming clear that the activities of humans since the beginning of the industrial age have warmed the earth's climate. The changes humanity has wrought are irreversible and will have serious long-term consequences for life on earth. This article addresses what global climate change is, what evidence exists and what implications this will have for the insurance and reinsurance industry.

The earth's temperature is controlled by a balance between the incoming radiation from the sun and the amount of radiation that the earth emits back into space. Simply put, any factor that alters either the radiation incoming from the sun, or the amount emitted by the earth has the potential to affect the climate.

Finely balanced

Since the beginning of the industrial age, the primary source of energy for humanity has been fossil fuels (coal or oil-derivatives). Burning hydrocarbons, the main component of fossil fuels, releases carbon dioxide and methane into the atmosphere. These, so-called, greenhouse gases are both strong absorbers of infrared radiation. They trap more of the outgoing radiation, altering the balance between the heat received from the sun and that re-radiated into space. The more greenhouse gases we emit into the atmosphere, the more the balance will change and the hotter the earth will become.

The evidence for climate change is broad and compelling. Globally, the 1990s was the warmest decade on record (Houghton et al, 2001) with 1998 being the warmest year on record. Snow cover and ice cover have both declined while the global average sea level has risen.

It is clear that the sheer abruptness of the temperature change over the last 100 years, as well as the magnitude, is unprecedented. The global average surface temperature increased by 0.6 deg C over the course of the 20th Century and while there have been instances in the earth's climate history where there have been warmings, none of them have been as severe or as abrupt as that seen in the modern era. The climate does change over time, but natural variations in the climate occur more gradually and over a longer timescale than that being currently experienced.

Given a number of approximations and assumptions, we can make an attempt to predict how the climate may respond to increases in greenhouses gases in the future. The Intergovernmental Panel on Climate Change (IPCC) has developed a set of future emission scenarios which quantify the effects of potential greenhouse gas policy (eg if carbon dioxide emission levels stabilise at the 2010 level). These scenarios are then used as the input to computer simulations of the atmosphere. There are over 30 emissions scenarios specified by the IPCC and seven different models. This range of scenarios and models allows a robust set of statistics to be developed for the model results.

It is clear that all the models present a consistent picture of the global mean temperature increasing, irrespective of the emission scenario. What that implies is, even if emissions are held at current levels, the earth will continue to warm. The globally averaged surface temperature is projected to increase by 1.4 to 5.8 deg C over the period 1990 to 2100, although the temperature change in any particular geographic region could be very different (some areas may get markedly hotter while others might even cool slightly). Also, the projected rate of warming will be much larger than that observed during the 20th Century and will be without precedent in the last 10,000 years.

Like the global mean temperature, even if emissions are held at current levels the earth will experience a sea level rise of significant proportions over the next century. The sea level is projected to rise by 0.09 to 0.88m over the period from 1990 to 2100. Coastal inlets and low-lying coastal areas are going to be most affected by the sea level rise. For the US, the places most at risk are the Gulf coast, especially New Orleans, and the Carolinas' coastline.

It's stormy outside

Given the range in uncertainty in basic measurements like the global mean temperature, quantifying the tropical cyclone response to climate change in a statistically significant way is going to be difficult. The IPCC addressed this in their 2001 report noting that increases in tropical cyclone peak wind intensities and mean and peak precipitation intensities were "likely, over some areas" in the 21st Century.

Investigation into the effect on winter storms in Europe (or extra-tropical cyclones in general) of global climate change is in its infancy. None of the three IPCC assessments of climate change have yet mentioned extra-tropical cyclones in any depth. However, recently some detailed studies have been undertaken and it is assumed that, in the forthcoming fourth assessment, the IPCC will make some statement on extra-tropical cyclones.

However, on some points a consensus is beginning to emerge from the climate modelling studies. Studies such as those by Hanson et al (2004) and Leckebusch & Ulbrich (2004) have found that:

- There is likely to be an increase in the frequency of more intense extra-tropical cyclones (those with a central pressure below 970mb); and

- The more intense storms could likely take a more southerly track over the southern UK and into Western and Central Europe.

The picture is not entirely clear for severe thunderstorms either but a series of papers by Changnon & Changnon have shown that the risk from hailstorms (and by proxy other severe thunderstorm risks) has changed across the US. Most interestingly, Changnon & Changnon show that, while hail activity has fallen across much of the US, hail activity has increased in the areas most prone to severe thunderstorms. The implication is that in areas where severe thunderstorms pose a great risk, the risk is going to increase.

Global climate change also has the potential to impact the frequency and severity of wildfires. One reason for this is obvious - higher temperatures, combined with longer periods of drought conditions, could lead to more ignition of plant material. The second reason is not so obvious. One indicator of climate change is the change in habitat for animals and insects, in particular, more widespread beetle infestations are likely. Some beetles burrow into trees and kill them, turning the trees into dead wood, which is excellent source of fuel for wildfires. This potential for increasing fuel loads, combined with warmer, drier conditions could well lead to increased wildfire activity.

Insuring an adequate response

The one central point of all this can be simply put - due to global climate change, the nature of property catastrophe risk will change. A recent study by the Association of British Insurers (ABI) attempted to examine the financial impacts of climate change on the insurance industry. Taking as a starting point the potential response of tropical cyclones and extra-tropical cyclones mentioned above, the AIR Worldwide cat models were used to examine the likely changes in industry loss.

It is clear that, if we accept the climate change effects to the underlying perils, by the middle of this century insured losses are going to increase significantly.

Insurers should also be aware of societal factors. In the US especially, society has seen a move towards the coastal regions. This is unlikely to abate even as the risk through flooding along the coastline increases as the sea levels rise. This may constitute a double hazard to insurers - an increasing exposure in areas of increasing risk. However, there may be a mitigating response as society adapts to a changing climate by, for example, increasing the strength of building codes.

The timeframe for climate change seems, in the context of yearly renewal cycles, to be a long way off. However, as noted above, we are already potentially seeing the first impacts of climate change in the reports of noticeable increases in tropical cyclone strength. However, the most significant climate change effects are unlikely to become measurably significant until around 2050. This will give the industry some 20 to 50 years to prepare for the change by, for example, lobbying for increased strength and enforcement of building codes.

The catastrophe modellers, too, will need to react to global climate change by incorporating changes in the intensity and frequency of catastrophic events. It should be noted that this assumes that the current system for managing catastrophic property insurance and reinsurance remains the same as today. This is by no means certain, or even likely. It is entirely feasible that changes in business practice or technological advances will make the current property catastrophe paradigm of model-based pricing obsolete in 20 to 50 years.

However, we have to remember we are in new territory and nature constantly surprises us. There appears to be a window of opportunity to prepare but in reality nature has no script; it makes sense to plan for the potential sooner rather than later.