Climate Extremes
Earthquakes
Floods
Climate Change
Volcanoes
Tsunamis
Indian
Ocean Tsunami
Landslides
Disaster Studies & Management
Rapid Environmental Impact
Assessment in Disaster Response
Intranet
| Home Contact us Gallery Site Map Resource Centre Search | ||
 |
||
| About us People News Publications Education & Training Events | ||
| Publications | Education and Training | Current & Recent Projects | Events | Research Opportunities |
When the top blows: why volcanic eruption has huge damage potential
BILL MCGUIRE, professor of Geophysical Hazards and director at the Benfield UCL Hazard Research Centre, University College London, looks at the volcanic threat EVEN a cursory examination of annual event tables will reveal that both economic and insured losses due to natural catastrophes are dominated by windstorm, quake and flood.Volcanic eruptions generally only appear on tables based upon casualty numbers rather then value of loss, largely because most recent volcanic events have occurred in developing world countries where wealth concentrations are negligible.
This situation cannot last, however, and there are a number of hotspots across the planet where the potential for big losses due to a volcanic eruption is escalating.
Somewhere between 500 and 600 volcanoes have erupted during historic time, and every year around 50 will erupt. This is not the whole story, however, and there may be up to 3,000 active or potentially active volcanoes, many located in developed countries and within range of large urban centres.
Some that immediately spring to mind are Mount Rainier, close to Seattle and Tacoma, Popocatapetl, which dominates sprawling Mexico City, and, of course, Vesuvius, in the shadow of which around 1m Neapolitans now live. Other major cities, such as Reykjavik and Auckland, also face a direct threat from volcanic action.
During the 20th century, volcanic eruptions took around 80,000 lives, the majority in two great catastrophes on the Caribbean island of Martinique in 1902 and at Columbia’s Nevado del Ruiz volcano in 1985. In absolute terms, however, economic and insured losses have been small.
To date, the costliest volcanic eruption was the 1980 blast of Mount St Helens (Washington State, US), which resulted in economic losses on the order of $1bn.
More recently, the property damage during the continuing eruption of the Soufrière Hills volcano on the Caribbean island of Montserrat is estimated to have cost insurers around $95m.
These figures are peanuts when compared, for example, to economic losses arising from the 1995 Kobe earthquake ($200bn) or the cost to the insurance industry of Hurricane Andrew in 1992 ($15.5bn).
Consequently, volcanic risk assessment has attracted little interest within the insurance industry and no generic volcanic loss model is yet available to aid in estimating the impact of volcanic action on a portfolio.
The low profile that volcanic eruptions have kept in recent decades have also led to a generally poor awareness within the insurance industry of the hazardous phenomena associated with volcanic action.
Lava continues to be perceived as the major threat, which was certainly the case during the recent spectacular eruption of the Nyirogongo volcano in the Congo, and it constitutes probably the main volcanic threat to Reykjavik and Auckland.
There are other volcanic phenomena, however, that are faster moving, farther-reaching and most importantly much more damaging. These include pyroclastic flows, the hurricane blasts of super-heated gas, incandescent ash and blocks as big as houses, which obliterated the Montserration capital, Plymouth, in 1997.
Equally potentially damaging are the volcanic mudflows, known as lahars, that buried the Colombian town of Armero and 23,000 of its inhabitants during a relatively small eruption through the snow-covered Nevado Ruiz volcano in 1985.
Neither catastrophe resulted in swingeing insured losses but such destructive phenomena are also characteristic of volcanoes that threaten urban centres with high insurance penetration.
The next eruption of Washington State’s glacier-covered Mount Rainier volcano is likely to generate copious lahars along the valleys draining the flanks and threaten the city of Tacoma and its environs.
Similarly, pyroclastic flows are a common outcome of eruptions of Vesuvius, as the population of the Roman cities of Pompeii and Herculaneum learnt to their cost in 79 AD. Vesuvius has now been ominously quiet for 57 years, a period over which largely unhindered and unmanaged construction has enormously increased building density on its flanks.
Arguments continue to rage between groups of scientists and the Italian government about just how much warning there will be before Vesuvius next goes bang, and whether or not there will be sufficient time to evacuate the 800,000 or so inhabitants who currently live in the potential danger zone.
What is certain, however, is that even assuming a moderate eruption property damage will be severe, with resulting economic losses likely to rise into the billions at the very least. Insured losses are likely to be proportionally high.
Whenever and wherever the next big volcanic loss occurs, there is a strong chance that as at Montserrat it will result in argument and animosity between insurers and reinsurers. This is partly due to unhelpful definitions and partly to the applications of an hours clause that may be suited to quake or windstorm but is far from appropriate to volcanic action.
The heart of the definition problem lies in the fact that there is no agreed meaning for the term volcanic eruption. At Montserrat, for example, magma has been constantly intruding into a growing dome of lava since the autumn of 1995.
Property damage and consequent insurance losses occurred, however, in discrete bursts involving collapse of the lava dome and the generation of pyroclastic flows. Did each dome-collapse constitute an event for insurance purposes, or should the entire eruption cycle which is still going on be classed as a single event?
Even given agreement that each dome collapse did constitute a discrete event, disagreement between the allocation of losses to these events critical in terms of triggering of the reinsurance cover remained.
This proved to be particularly difficult to ascertain when dome collapses occurred repeatedly over a period of a few weeks, as they did in the summer of 1997, and where the establishment of an exclusion zone around the volcano prevented reliable verification of damage timing.
The problem lay with the 72 hours clause that is generally applied to volcanic action but which is often unsuitable for this purpose. In order to take better account of the volcanic hazard characteristics, and improve the loss adjustment situation post-event, Alistair Milward, of Wellington Underwriting, and I proposed that the definition of loss occurrence for volcanic action should be extended to 672 hours. Although many eruptions continue for longer than this, the climactic phases of most eruptions, during which time most damage occurs, would be contained within this period.
Application of this clause would negate the requirement for often-unverifiable loss allocation to individual damaging episodes, and ensure a more harmonious relationship between insurer and reinsurer.
This definition of loss occurrence is now available (NMA2842) but appears to have been little utilised to date. With future volcanic action now threatening at a number of new locations, including at Mammoth Mountain ski resort (California) and on the Caribbean island of Dominica, this would appear to be a good time to start.
While NMA2842 will undoubtedly help to clarify post-event loss issues, for many volcanically active regions a great deal of work remains to be undertaken before pre-event portfolio exposure can be fully and accurately assessed.
To this end, the Benfield UCL Hazard Research Centre and Benfield Group are working closely to better understand eruption-related risks and to provide more informed advice for insurers exposed to the volcanic threat.
The lesson of Montserrat also provides strong justification for academics and insurers working together to protect local markets from the post-loss artificial tightening of insurance capacity and alongside local government in designing social and infrastructure projects better able to withstand future volcanic peril.
* Bill Mcguire’s new book, A Guide to the End of the World: Everything You Never Wanted to Know is published by Oxford University Press at £11.99
| > | Alert 22 |
| > | UK Greenhouse Gas Emissions - Are We On Target? |
| > | Workshop: UK Flood: Recent Lessons; Future Prospects |
| > | Hazard & Risk Science Review 2007 |
| > | Tropical Cyclones |
| > | UK & European Weather |
| > | North Atlantic Oscillation (NAO) |
| > | Natural Hazards for Insurers Certificate Course |
| > | Masters in Geophysical Hazards |
| > | BUHRC Workshops |
