Most weather extremes have relevance for the financial sector, as shown in Table 8-1. Column 6 summarizes the impacts of extremes on the main sectors of activity considered by TAR WGII. The ways in which these impacts affect the insurance industry are shown in Column 7.
Hot Temperature Extremes. Hot summers are likely to become more common as a result of global warming. The nonlinear effect of global warming on extreme events (see Figure 8-2) can be clearly illustrated by the example of temperature. Hulme (1997) estimates for the UK that the change in mean annual temperature in 2035 relative to the 1961-1990 mean will be approximately 1°C. Yet as a result, temperature conditions similar to those in the exceptional (1-in-300 years) summer of 1995 should occur once every 10 years on average between 2021 and 2050. Insurance claims could rise because of land subsidence, business interruption, and crop failure. Although heat waves have been shown to lead to an increase in daily mortality and morbidity (see Section 9.4.1)an impact that may be compounded by poor air qualitythe effect is likely to be too small to noticeably affect the financial services sector.
Table 8-1: Extreme climate-related phenomena and their effects on the insurance industry: observed changes and projected changes during the 21st century [after Table 3-10; Munich Re, 1999b (p. 106)]. | ||||||
Changes in Extreme Climate Phenomena |
Observed Changes
|
Projected Changes
|
Type of Event Relevant to Insurance Sector
|
Time Scale
|
Sensitive Sectors/Activities
|
Sensitive Insurance Branches
|
Likelihood |
||||||
Temperature Extremes | ||||||
Higher maximum temperatures, more hot days and heat wavesb over nearly all land areas |
Likelya (mixed trends for
heat waves in several regions)
|
Very likelya
|
Heat wave | Daily-weekly maximum | Electric reliability, human settlements | Health, life, property, business interruption |
Heat wave, droughts | Monthly-seasonal maximum | Forests (tree health), natural resources, agriculture, water resources, electricity demand and reliability, industry, health, tourism | Health, crop, business interruption | |||
Higher (increasing) minimum temperatures, fewer cold days, frost days, and cold wavesb over nearly all land areas |
Very likelya (cold waves not treated
by WGI)
|
Very likelya
|
Frost, frost heave | Daily-monthly minimum | Agriculture, energy demand, health, transport, human settlements | Health, crop, property, business interruption, vehicle |
Rainfall/Precipitation Extremes | ||||||
cold wavesb over nearly all land areas More intense precipitation events |
Likelya over many Northern Hemisphere
mid- to high-latitude land areas
|
Very likelya over many areas
|
Flash flood | Hourly-daily maximum | Human settlements | Property, flood, vehicle, business interruption, life, health |
Flood, inundation, mudslide | Weekly-monthly maximum | Agriculture, forests, transport, water quality, human settlements, tourism | Property, flood, crop, marine, business interruption | |||
Increased summer drying and associated risk of drought |
Likelya in a few areas
|
Likelya over most mid-latitude continental
interiors (lack of consistent projections in other areas)
|
Summer drought, land subsidence, wildfire | Monthly-seasonal minimum | Forests (tree health), natural resources, agriculture, water resources, (hydro)energy supply, human settlements | Crop, property, health |
Increased intensity of mid-latitude stormsc |
Medium likelihooda of increase in
Northern Hemisphere, decrease in Southern Hemisphere
|
Little agreement among current models
|
Snowstorm, ice storm, avalanche | Hourly-weekly | Forests, agriculture, energy distribution and reliability, human settlements, mortality, tourism | Property, crop, vehicle, aviation, life, business interruption |
Hailstorm | Hourly | Agriculture, property | Crop, vehicle, property, aviation | |||
Intensified droughts and floods associated with El Niño events in many different regions (see also droughts and extreme precipitation events) |
Inconclusive information
|
Likelya
|
Drought and floods | Various | Forests (tree health), natural resources, agriculture, water resources, (hydro)energy supply, human settlements | Property, flood, vehicle, crop, marine, business interruption, life, health |
Wind Extremes | ||||||
Increased intensity of mid-latitude stormsb |
No compelling evidence for change
|
Little agreement among current models
|
Mid-latitude windstorm | Hourly-daily | Forests, electricity distribution and reliability, human settlements | Property, vehicle, aviation, marine, business interruption, life |
Tornadoes | Hourly | Forests, electricity distribution and reliability, human settlements | Property, vehicle, aviation, marine, business interruption | |||
Increase in tropical cyclone peak wind intensities, mean and peak precipitation intensitiesc |
Wind extremes not observed in the few analyses available;
insufficient data for precipitation
|
Likelya over some areas
|
Tropical storms, including cyclones, hurricanes, and typhoons | Hourly-weekly | Forests, electricity distribution and reliability, human settlements, agriculture | Property, vehicle, aviation, marine, business interruption, life |
Other Extremes | ||||||
Refer to entries above for higher temperatures, increased tropical and mid-latitude storms |
Refer to relevant entries above
|
Refer to relevant entries above
|
Lightning | Instant-aneous | Electricity distribution and reliability, human settlements, wildfire | Life, property, vehicle, aviation, marine, business interruption |
Refer to entries above for increased tropical cyclones, Asian summer monsoon, and intensity of mid-latitude storms |
Refer to relevant entries above
|
Refer to relevant entries above
|
Tidal surge (associated with onshore gales), coastal inundation | Daily | Coastal zone infrastructure, agriculture and industry, tourism | Life, marine, property, crop |
Increased Asian summer monsoon precipitation variability |
Not treated by WGI
|
Likelya
|
Flood and drought | Seasonal | Agriculture, human settlements | Crop, property, health, life |
a
Likelihood refers to judgmental estimates of confidence used by Working
Group I: very likely (90-99% chance); likely (66-90%
chance). Unless otherwise stated, information on climate phenomena is taken
from Working Group I's Summary
for Policymakers and Technical Summary. These
likelihoods refer to observed and projected changes in extreme climate phenomena
and likelihood shown in first three columns of table. b Information from Working Group I, Technical Summary, Section F.5. c Changes in regional distribution of tropical cyclones are possible but have not been established. |
Cold Temperature Extremes. As a result of global warming, cold extremes
of winter weather are likely to become rarer. In temperate latitudes, this development
generally would be beneficial for business activities in, for example, the construction
and transport sectors, with concomitant reductions in claims for business interruption.
Although cold conditions should become rarer, a more active hydrological cycle
might lead to more episodes of heavy snowfall, provided that temperatures remain
below freezing. Regional shifts in the occurrence of phenomena such as ice storms
may be expected. Ice storms occur when precipitation falls as rain but freezes
on contact with a solid surface. Air temperatures close to freezing are ideal
for ice storm occurrence. Thus, in colder regions where the weather currently
is well below freezing in the winter, ice storms may become more common as a
result of global warming, although they could become less frequent in areas
where they occur at present (Francis and Hengeveld, 1998). An ice storm that
occurred 7-10 January 1998, in the northeastern United States and eastern Canada,
led to insured damage estimated at US$1.2 billion (Lecomte et al., 1998).
Heavy Rainfall and Flooding. TAR WGI Chapter 9 indicates that "many models" now project that conditions in the tropical Pacific may become more El Niño-like, with associated changes in precipitation patterns (Meehl et al., 2000b). This would lead to more frequent patterns of El Niño-like floods and drought conditions in areas where teleconnections to the El Niño-Southern Oscillation (ENSO) exist. Observational studies assessed in TAR WGI Chapter 2 suggest that there has been a widespread increase in heavy and extreme precipitation events in regions where total precipitation has increased (i.e., the middle and high latitudes of the Northern Hemisphere). Flooding is responsible for 40% of total economic losses and 10% of weather-related insurance losses globally.
Tropical hurricanes can lead to landslides. Hurricane Mitch probably is the most well-known event in recent years. This system, the strongest ever October tropical storm in the Atlantic Basin, stalled over Central America and produced more than 600 mm of rainfall in 48 hours. Resulting landslides and mudslides led to an estimated 9,000 deaths and insured losses of US$513 million (Swiss Re, 2000b). In disasters of this magnitude, preparedness and planning can make a huge difference in loss of life and the amount of damage sustained.
Large river basin floods develop over huge areas following weeks of unusually high rainfall. In July and August 1997, flooding in central Europe caused 54 fatalities in Poland and required the evacuation of 162,000 people (Kundzewicz et al., 1999). The value of the economic losses throughout central Europe amounted to approximately US$5 billion, with insured losses of US$940 million. The intensity of such flood events is driven not only by climatology but also by human management of the watershed.
Low RainfallDrought, Land Subsidence, and Wildfire. Drought is important for the financial sector through impacts on commercial agriculture, building foundations, and wildfire occurrence. Figure 8-3 shows the cost of subsidence claims to the industry from 1975 to 1997 in England and Wales. There is a clear relationship with rainfall (with some lag effects). Similar effects are seen in France (Radevsky, 1999). Where insurance is used as the mechanism to finance repairs to building foundations, as in the UK and France, costs for domestic properties can be higher than where the damage is not insured, as in Australia. Adaptive responses such as stronger foundations in new buildings and repairs to older housing capital should reduce the problem.
The worst drought of recent decades has occurred (indeed, it persists) in theSahelian region of West Africa, where since 1968 rainfall has been below the long-term average in almost every year (Nicholson et al., 2000). The strength and persistence of this deficit is unparalleled in recent times. Despite the drought's severity, it has had minimal impact on the commercial financial sector because of the low penetration of insurance in the region. However, the drought's role in the development of the region has been significant.
Wildfire is an increasingly important insurance issue, as illustrated by the US$140 million economic losses sustained in the Los Alamos fire of 2000 (Hofmann, 2000b). Outdoor fire occurrence is likely to increase in a future warmer climate, particularly along the increasingly popular urban-rural fringe (Swiss Re, 1992; Torn et al., 1998). Whereas in Europe most wildfires are of human origin (either deliberate or accidental), lightning (see below) is still the leading cause of forest fires in the western United States and Alaska (the regions of North America with the greatest number of wildfires).
Even if rainfall amounts are unchanged by global warming, higher temperatures will increase the level of risk associated with these hazards because of increased water loss through evaporation and transpiration.
Lightning Strikes. Model experiments are not able to tell us anything directly about changes in lightning occurrence as a result of global warming. Any increase in convective activity should lead to more frequent electrical storms and lightning discharges, and it seems likely that global warming will have such an effect in the tropics (Lal et al., 1998) and in extratropical latitudes (White and Etkin, 1997). Reeve and Toumi (1999) suggest that a 1°C increase in average wet-bulb temperature can be accompanied in mid-latitudes by a 40% increase in lightning. Of relevance to insurers, lightning is a cause of fires and damage to electrical equipment, with associated business interruption claims (Mills et al., 2001).
Tropical and Extratropical Windstorm. Experiments with climate models to date have not produced a consensus regarding the likely future occurrence of tropical and extratropical wind storms. Both have a very large capacity to cause damage. Hurricane Andrew, for example, occurred in 1992 in the Atlantic Basin and made landfall over the United States, causing US$21 billion (1999 US$) in insured damage. Hurricane Floyd, which caused US$2.2 billion in insured losses in 1999, required the evacuation of 2 million people and imposed huge stress on infrastructure, resources, and ultimately health. The most damaging extratropical windstorm was Daria in 1990, which caused US$6.8 billion in insured losses in northwestern Europe. In December 1999, windstorms Martin and Lothar tracked south of the normal route, affecting France, northern Spain, and central Europe. Together they caused 140 fatalities and US$8.4 billion in insured damage.
Sea-Level Rise. Increases in sea level pose a major potential risk to coastal zones (TAR WGI Chapter 6), especially if they are associated with an increase in storminess. The mid-range increase in sea level by the year 2100 as a result of anthropogenic climate change is 49 cm, taking into account atmospheric aerosol concentrations, with estimates ranging from 26 to 72 cm (TAR WGI Chapter 11). The main risk to the financial sector is in the effect that this change in mean sea level may imply for the occurrence of tidal surges, which already cause enormous damage and loss of life, especially in the developing world (see Box 8-4). One of Europe's greatest natural disasters in terms of loss of life was the 1953 storm surge in the North Sea, which led to almost 2,000 fatalities in The Netherlands and the UK
Other reports in this collection |