You don’t have to look far to find plenty of examples of extreme weather in January this year, from searing heat waves and wildfires in south-east Australia and Tasmania to record rainfall and floods in Indonesian capital Jakarta. In the UK and Europe, heavy snow has caused widespread travel disruption. According to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change (IPCC) these are just the kind of climate extremes we can expect more of in the future.
A challenge for insurers is how to assess the impact of increased weather extremes on their portfolios. In doing so, catastrophe modelling plays a vital role. Yet whether such extreme weather event trends are too long-term in nature to be captured by the models remains to be seen.
Record heat wave
Australia has started 2013 with a record-breaking heat wave lasting over two weeks. Temperatures have frequently exceeded 48°C causing intense wildfires in many parts of the country. The highest recorded maximum was 49.6°C at Moomba Airport in South Australia. The Australian average temperature rarely exceeds 39°C for more than two days running. A run of three days has only occurred on three occasions in the past and a run of four days just once, in 1972. The January 2013 heat wave saw average temperatures of over 39°C for seven days running and above 38°C for 11 days straight. It is not just Australia that has suffered from such heat waves. In the US last summer, the worst drought in 50 years hit America’s agricultural heartland, affecting 85% of the soybean crop and 87% of its maize crop. In Russia and the Ukraine grain yields were also dramatically reduced thanks to a summer-long drought. A more severe drought in 2010 led to wildfires and saw Russia’s grain harvest contracting by a third. Such droughts have driven up the global cost of food commodities such as wheat, soybeans and corn.
Monitoring extreme weather
A recent study by Munich Re shows that North America has been most affected by weather-related extreme events in recent decades. The study shows a nearly quintupled number of weather-related loss events in North America over the past three decades, compared with a factor of 4 in Asia, 2.5 in Africa, 2 in Europe and 1.5 in South America. Munich Re has made the rough estimate that 30% of the increase of losses from weather-related disasters is due to climate change. “In all likelihood, we have to regard this finding as an initial climate change footprint in our US loss data from the last four decades, ” says Professor Peter Höppe, the head of Munich Re’s Geo Risks Research unit. “Previously, there had not been such a strong chain of evidence. If the first effects of climate change are already perceptible, all alerts and measures against it have become even more pressing.” According to IPCC scientists some of the climate extremes we can expect to see more of in future will include more severe heat waves in southern Europe and the Mediterranean, an increased incidence of heavy precipitation in East Africa and significantly reduced rainfall in northeast Brazil. The increasing exposure of people and economic assets to extreme weather events has been a major cause of long-term increases in economic losses from climate-related disasters. Socio-economic factors will be among the main drivers of future increases in related losses, concludes the IPCC.
Modelling a changing climate
The insurance industry plays an important role helping societies rebuild after disasters hit. As insurance penetration increases in countries exposed to extreme weather perils, this role will grow in importance. While the influence of climate change on weather patterns is a long-term trend, the industry continues to invest in research and take part in collaborations in order to better understand the climate of the future. Although catastrophe models are shorter-term in their outlook, it is possible to adapt them to account for more extreme weather, rather than just using the historical record. This was particularly evident following the record-breaking hurricane seasons of 2004 and 2005, following which the catastrophe modelling companies released new storm models for the US.
“In 2006 we produced an explicit medium-term view of the risk that took into account the fact hurricane frequency was expected to be higher in that coming medium term than the average of history, ” says Claire Souch, vice president of model solutions at Risk Management Solutions, Inc. “This was a recognition we were moving off the baseline of the historical average of hurricane activity.”
“To date, there are not many perils for which there’s proof that the next few years will already see a consistent change compared to a long-term average, but as soon as that evidence becomes clear in any peril we would reflect that in our view of the risk and provide that adjusted view for comparison.”
One of the aims of new open source catastrophe modelling initiatives such as Oasis is to build a modelling community whereby different views of risk can be brought into the industry. Under its “plug-and-play” model the latest climatology can be accessed and used to inform day to day underwriting and pricing decisions.
(source: Lloyd’s website)