home & introduction


projects & ideas



Island Vulnerability

Iceland Iceland's Flag

Island Vulnerability explores the challenges which isolated geographies face when dealing with risk and disasters by examining the processes which create, maintain, and could be used to reduce their vulnerability. This page provides information on vulnerability issues in Iceland:

Bird Cliff on Vestmannaeyjar

Bird Cliff on Vestmannaeyjar.
(Copyright Ilan Kelman 2001.)

Razorbills on Vestmannaeyjar

Razorbills on Vestmannaeyjar.
(Copyright Ilan Kelman 2001.)

Vulnerability and Disaster Management in Iceland

By Ilan Kelman.
Slightly updated from the original 2001 report.


Funded by the University of Cambridge's Scandinavian Studies Fund, I travelled to Iceland from 20 April 2001 to 20 May 2001 to investigate disaster management practices. This report describes my findings. I can claim no expertise, as evidenced by the gaps and vague statements in my report. I am interested in learning much more about Iceland, particularly in the context of Island Vulnerability and I have plenty yet to learn. Therefore, please contact me with feedback and suggestions, including omissions or errors. The views here are my own and do not reflect the Icelanders who kindly provided me with information, resources, and ideas and whose hospitality made me feel welcome in a beautiful and fascinating country.


(Copyright Ilan Kelman 2001.)


This grant was sought to travel to Iceland to introduce Icelandic ideas and past work into my PhD dissertation on the impacts on the built environment of the physical forces from flood water. In Iceland, physical forces from flood water are a concern due to jökulhlaups (glacial outbursts). Jökulhlaups occur principally when subglacial volcanic activity melts ice resulting in a sudden release of a large volume of water. Dam failure is another potential source for the sudden release of a large volume of water since 93.5% of Iceland’s energy comes from hydroelectric sources, many of which are constructed in earthquake-prone areas or near active volcanoes.

This trip has the objectives:
1. To gain an understanding of risk, risk perception, and risk management from an Icelandic perspective, particularly with respect to a culture continually experiencing multiple natural hazards.
2. To explore the impacts on Icelandic society of previous large floods, the coping mechanisms of Icelanders in these situations, and any steps taken to prevent or to prepare for future disasters due to floods such as jökulhlaups or dam failure.

The first objective would establish a framework for dealing with risk issues in an Icelandic environment. The second objective would apply these concepts in a practical context which could be compared and contrasted to other locales. If possible, strategies would be examined for introducing Icelandic disaster management concepts and experience-based strategies to the U.K.

Fumarole in a blizzard Snow and fog
Geysir at the beginning of an eruption Geysir erupting with tourists

Icelandic natural hazards: Snow, fog, fumaroles, hot springs, and geysers. But opportunities too such as energy, tourism, beauty, and inspiration.
(Copyright Ilan Kelman 2001.)


Two important results emerged:
1. Iceland has many disaster management and risk culture lessons to teach but also has many lessons to learn.
2. Little material exists with respect to extreme flood risk in the urban environment from jökulhlaups or dam break. The most likely reason is that such events do not threaten urban areas. Plenty material exists on storm surge and on forces imposed on structures by avalanches.

Iceland suffers from some main difficulties in risk management which afflict most countries, yet when a problem is tackled, the Icelanders quickly produce excellent results. The example of avalanches illustrate this situation. Because Iceland is a country with steep mountains and much snow, it is obviously prone to avalanches. Furthermore, many settlements are coastal, built at the bottom of slopes. The lack of strong vegetation to support snow on a slope creates an obvious avalanche hazard for these urban environments. Nonetheless, prior to 1995, little support for avalanche loss prevention existed. Those who spoke of or who studied avalanches met derision and were labelled scaremongers.

In 1995, Iceland was stunned by two separate avalanches in small, isolated communities killing 14 people in January and 20 people in October. Most of the fatalities occurred in zones labelled as safe. These disasters sparked an overhaul of Iceland’s approach to avalanches. Such a sequence is common in disaster management: little interest is shown until a disaster occurs and then the issue becomes prominent. In six years, however, Iceland managed, at a relatively low cost, to produce a world-class avalanche monitoring and evacuation programme; to develop and implement an innovative hazard mapping methodology; and to experiment with a wide variety of solutions aimed at eliminating all avalanche deaths. To achieve such dedication and results with relatively minimal resources is highly impressive and a lesson for all disaster managers.

Explaining how such an achievement is realised is challenging. The Icelanders had few thoughts on the issue except "That is simply what we do". I can identify two important cultural elements which I have not seen in flood management in England. The first is willingness to bring in foreign expertise. Iceland went to the Norwegians, the Swiss, and the Canadians amongst others to ask for help with their avalanche problem. In contrast, insularity has prevailed throughout England's flood crises with a strong reluctance to admit that other countries could help or even to examine other countries’ activities. This situation has improved over the last few years, particularly in terms of interaction with the Dutch. Nonetheless, I have seen several articles and news items where flood managers in England boost their own stature by suggesting that the Dutch come to England to learn about flood management. An exchange of knowledge and experience is far more realistic and a far more appropriate attitude for action and publicity.

Second, Icelanders are willing to admit mistakes and the public is willing to forgive mistakes, although an important proviso is that no avalanche fatalities have yet arisen due to mistakes. The positive results are building institutional memory and permitting those responsible to learn and to apply their learning. In contrast, the crises which the British government has recently faced (e.g. train track replacement, petrol blockade, floods, and foot-and-mouth disease) led to mainly a vociferous defence of all government actions as being correct and ideal. Meanwhile, the opposition and much of the public continually sought to place blame and to find scapegoats. The Environment Agency, for example, is often lambasted whenever a flood problem manifests even though many flood management issues are not within their mandate. Often, particularly for disasters, it is much more productive in the long-term to permit admission of, and forgiveness for, poor decisions as long as the decision-makers are willing to learn and to apply their learning. Iceland has shown how effective this approach can be if it is implemented properly.

Nevertheless, as in the avalanche case, an issue must be accepted as being a problem before it can be tackled. Iceland, as with other countries, has a record of not accepting natural hazards to be threatening until after a catastrophe has occurred, usually with serious casualties. Earthquakes, for example, would seem to be an obvious threat to Iceland, particularly in the relatively populated southwest. Despite excellent earthquake engineering work in Iceland, the Iceland Catastrophe Insurance (ICI) programme does not seem overly concerned about a major earthquake near Reykjavík according to their booklet "Renewal Information 2001" (English version). One of the principal arguments for their stance is that "There is no historical record (which date [sic] back to the founding of a monastery in Reykjavík in 1225) of such an event" (p. 3). Otherwise, they base their analysis mainly on return period and probable maximum loss calculations "but lack of specific study makes any assessment difficult" (p. 3). The methodologies articulated by ICI often have serious dangers, as most people in disaster and risk management know. If a "lack of specific study" exists, then an obvious solution is to provide funding to fill in the gap. ICI also seems uninfluenced by the fact that the two June 2000 earthquakes in the south were of moderate magnitude but were felt throughout Reykjavík.

Earthquake Fault in a Blizzard with Steam Earthquake Fault in the Fog and Rain

Icelandic Earthquake Faults.
(Copyright Ilan Kelman 2001.)

Evidence exists from the June 2000 earthquakes to illustrate that cheap, easy prevention measures, such as locking cupboards and placing heavy items lower on shelves, can reduce earthquake losses substantially. Anecdotal evidence suggests that implementing such measures is not the norm. As few injuries resulted from the June 2000 earthquakes, a blasé attitude may exist that earthquake vulnerability is low and that earthquakes are a bit of fun to experience once in a while. I was, however, told that building construction in Iceland is highly seismic resistant. One engineer mentioned that Iceland's earthquake-resistant buildings are a coincidence, resulting more from the climate and material availability than from a drive to protect against earthquakes. The minimal damage suffered by bridges during the June 2000 earthquakes cannot be explained with the same reasoning.

Such examples from my trip’s first result segue into the second result. A possibility is that the lack of information on extreme flood risk in the urban environment from jökulhlaups or dam break due to earthquakes may be because such disasters have never before occurred or been considered, rather than because negligible risk exists. Practically, it does appear that the locations of dams and of possible jökulhlaup sites are far from the main urban centres, but a study to establish this fact would be helpful if one does not already exist.

In contrast, Iceland has a long history of, and work related to, storm surges affecting urban areas (more below). Slysavarnafélagid Landsbjörg (ICE-SAR, the Icelandic Association for Search and Rescue) and Siglingastofnun Íslands (Icelandic Maritime Administration) are also highly sophisticated. Strong elements of vulnerability management enter in addition to the primary focus on emergency management. I learned immensely from Iceland’s excellent work in all these areas. Therefore, Iceland has plenty to teach in disaster management areas which they have examined but may be falling into the same trap as others of being uninterested in events of perceived low risk.

Rescue Vehicle Maritime and Aviation Monitoring Station

Rescue vehicle.
(Copyright Ilan Kelman 2001.)

Maritime and Aviation Monitoring Station.
(Copyright Ilan Kelman 2001.)

Regarding storm surge, following major losses due to storms at sea (mainly fishing vessels) and on land (significant structural damage), Iceland has created an excellent real-time information system for maritime data which is currently being expanded into all meteorological and oceanographic data. Furthermore, Icelanders have developed and implemented innovative approaches to the physical modelling of storm surge yet have stopped short of detailed damage prediction. Damage assessment, prevention, and mitigation, though, are prominent with respect to storm surge. Such techniques would be applicable elsewhere.

One aspect requiring further investigation is whether or not social vulnerability has been tackled in a similar manner. Issues include education and awareness on warning systems, reacting appropriately to warnings, simple measures individuals and communities can take to reduce their vulnerability, accepting vulnerability-reducing land use and planning measures, and identifying and tailoring programmes to the most vulnerable sectors of the population.

Sociological lessons which were observed are challenging to apply elsewhere. As mentioned with respect to avalanches, the question why Icelanders can focus so well on an identified problem and produce such excellent results in such a short time with so few resources remains unanswered. Their ability to accept outside critique could be similarly emulated. During my meetings, I often asked fairly pointed questions, of the nature which has caused consternation with British researchers, but the Icelanders responded thoughtfully and with equanimity, admitting to weaknesses and proposing potential solutions where possible. If cultural characteristics and attitudes could be transferred, then flood management in Britain would gain immensely by imitating these Icelandic characteristics. If not, then the best technical solutions backed up by plenty money--consider, for example, the more than £70 million pledged by Westminster for flood defences during the U.K. floods throughout Autumn 2000--will be unable to achieve much in alleviating risk from natural hazards over the long-term.

One final note is that much of the Icelandic work which I viewed is not well-disseminated outside Iceland. When I queried the possibilities for publicising their activities, the response was that time and resources are finite and their primary mandate is developing good programmes and working within Iceland to achieve proper risk management. Wider dissemination is unfortunately low priority, despite its importance. Therefore, it is vital to go to Iceland to learn, since the information gleaned would be difficult to obtain otherwise.

I learned an immense amount about Icelandic approaches to natural disaster management, yet much remains to be learned. The wider scope includes combinations of social and technical solutions, parallels amongst different disasters, and principles of disaster management. One gap in my itinerary, for example, was my failure to meet Avrik, Almannavarnir ríkisins (the National Civil Defence of Iceland), although their website contains useful information and ideas; for example, the green and red text for preventative and reactive measures (but are alternatives for colour-blind people available?). [Update from their website: "Recently the Civil Defence Act was amended and the National Civil Defence Agency was disbanded and it’s responsibilities were entrusted to the National Commissioner of the Icelandic Police."] As with all such trips, many more questions were raised than could be answered, but such an experience is part of the continual exploration in research and application in Island Vulnerability.

The volcano Eldfell Looking down the slope of Eldfell to Heimaey

The volcano Eldfell, born in 1973, on the Icelandic island of Vestmannaeyjar.
(Copyright Ilan Kelman 2001.)

Looking down the slope of Eldfell to the town of Heimaey which was partly destroyed during the 1973 eruption.
(Copyright Ilan Kelman 2001.)

Iceland Catastrophe Insurance

By Ilan Kelman.
As of 2001.

Some further notes on Iceland Catastrophe Insurance (ICI) may be of interest. Quoting from ICI's "Renewal Information 2001" (English version):

p. 17, Appendix II, Regulations on the Iceland Catastrophe Insurance, Article 1:
"The perils insured against by the catastrophe insurance, cf. Act No. 55/1992 on the Iceland Catastrophe Insurance are the following:
5. A waterflood means a flood which occurs when rivers or brooks abruptly overflood their banks or when floodwaves from the sea or lakes run ashore causing damages or destruction of insured valuables. Annual or regular floods from rivers, brooks, the sea or lakes are not to be construed as waterfloods herein. The same applies to usual melt-water or floods caused partly or wholly by man, e.g. when watertanks, dams or other structures break due to conditions other than catastrophes."

Clearly some floods are covered, but others are not. Regular or annual floods are not covered, thereby providing a remarkable disincentive for construction in vulnerable areas without appropriate mitigation measures (which would not necessarily be external or structural). This approach is intriguing because the insurance scheme literally becomes "Catastrophe Insurance"; that is, for the most extreme events.

On the other hand, many flood catastrophes are not covered. Floods caused by failure in human structures are covered for only some circumstances (why?). Jökulhlaups are not mentioned under either floods or volcanoes (why not?). If a waterflood causes rare sewage backup, is it covered? Groundwater rise? Rain storm runoff or rainwater ponding? ICI displays similar inconsistencies in their coverage of other natural perils.

I do not know enough to be able to criticise or commend them for their choices. The main issue is that insurance schemes have disparate definitions on what "flood cover" and "natural peril cover" mean in practice. Which approaches are most appropriate for a given peril situation, insurance culture, and public culture? Do we need further work in (a) comparing and appraising the different approaches we see in different countries and (b) developing a tool to determine the impact which national and private insurance policies have on long-term loss prevention and loss mitigation for natural perils?

The structure of ICI is of interest too. Iceland Catastrophe Insurance is an insurance company created by the Iceland Insurance Act, of which Article 2 states:

"Five persons shall sit on the Corporation's Board of Directors. Three shall be elected by the Althing; one nominated by those insurance companies which collect premiums...the Minister of Insurance Affairs shall appoint one member who shall be Chairman. Deputies shall be chosen in the same manner."

The Alþingi is Iceland's parliament. Therefore, ICI is a state scheme in practice, but is legally a company.

Article 19 of the Iceland Insurance Act states that "The Corporation is permitted to reinsure its risk in this country as well as abroad", so reinsurance of the scheme is not mandatory. Presumably ICI generally purchases reinsurance. If assets and reinsurance cannot cover claims made, the Board may take a loan and, again from Article 19, "The State Treasury shall guarantee such loans by means of surety", more evidence of the state's support for catastrophe insurance.

The Alþingi, Iceland's Parliament The Alþingi, Iceland's Parliament

The Alþingi, Iceland's Parliament.
(Copyright Ilan Kelman 2001.)

Contact Island Vulnerability.

The material on the Island Vulnerability website is provided as only an information source. Neither definitive advice nor recommendations are implied. Each person or organisation accessing the website is responsible for making their own assessment of the topics discussed and are strongly advised to verify all information. No liability will be accepted for loss or damage incurred as a result of using the material on this website. The appearance of external links on this website does not constitute endorsement of the organisations, information, products, or services contained on that external website.