External safety risks: the risk of an accident (introduction)
The government has a policy for tackling the risks in the vicinity of a company or of hazardous substances in transit. This policy is covered by the term 'external safety'. What is 'external safety'? What does the policy consist of? And what does the Environmental Data Compendium have to say about external safety?
What is external safety?
External safety deals with the risks affecting local residents caused by air traffic and by the production, the use, the storage and the transport of hazardous substances. It does not include accidents in hotels or catering establishments. On the other hand, the firework disaster in Enschede, that involved major damage to the immediate vicinity of the firework factory, is included under external safety.
The government policy for external safety therefore assigns a central role to the vicinity of a company or of hazardous substances in transit. Working conditions policy deals with the safety of people on company premises.
Sorts of companies
The types of companies covered by external safety policy include: airports, railway marshalling yards, LPG stations and large chemical companies. This is not just a question of the locations of the companies themselves but also of transportation to and from those locations. Companies that constitute a major risk because they use large quantities of hazardous substances such as chlorine are obliged to produce safety reports. They are usually large chemical companies. In a 'safety report', a company describes - in accordance with statutory regulations - itself, the measures taken to reduce and manage risks, and other items. A report of this kind has to be submitted to the government authorities - the province or the municipality - every five years or if there are major changes at the company (VROM, 1993 and 1999).
What factors determine risks?
The severity of a risk depends on various factors. To start with, there is the size of the company and the nature of the activities that take place there, including transport to and from the site. For example, it is generally more risky to produce substances than to store them. The level of compliance with regulations also plays an important role here. Finally, the risk is determined by the number of people living in the vicinity of the company. The firework disaster in Enschede is a clear example of a situation in which all three factors were involved: the size of the company where the fireworks were stored (or the amount of fireworks stored), poor compliance with the regulations and the fact that the company was located near to a residential area.
How severe is a risk and for whom?
The risk caused by a particular company is calculated, for example, on the basis of information about the structure and the operations of the company and on the basis of historical data about the nature, numbers and severity of accidents with fatal casualties in this type of company. This calculation results in figures which express the risk of fatal accidents.
The policy makes a distinction between two types of risk.
The external network safety policy makes a distinction between two types of risk: the local risk and the group risk (VROM, 2002). The local risk is the probability of an imaginary person being killed in a particular place as a result of an accident in a company. The group risk defines the probability of a disaster in which a proportion of the people present are killed. You will find both terms on the web pages dealing with external safety risks.
Local risk
We can calculate the risk of a fatal accident occurring at a particular location during a period of one year as the direct result of an incident in a company if someone is present at that location. It is therefore not important here to know whether a person is actually present. This risk, calculated for an imaginary person, is known as the local risk. People who are actually present at the location in question run that risk. For example, 440 people in the Netherlands who live in the vicinity of airports are subject to an annual risk (in other words: the risk for this year and for next year again) of at least 1 in 100,000 of being killed by an accident.
In the Netherlands, the government has determined that the local risk must not, in principle, exceed 1 in 1 million (in other words, 10-6). So the risk that an imaginary person permanently present at the location in question for a year (the location for which the risk has been calculated) will suffer a fatal accident as a result of a company or transportation accident must not be greater than once in a million years. This policy objective must have been achieved no later than 2010. There are a few exceptions to this rule. For example, if international obligations make it impossible. An example would be the free access of shipping via the Western Scheldt to Antwerp.
More than 37,000 people in the Netherlands are currently subjected to a higher risk than one in 1 million of a fatal accident caused by excessive company risks. These people therefore live in the vicinity of a possibly hazardous location.
Group risk
When calculating a risk, it is also possible to look - rather than at an imaginary person - at the population that is actually present and the distribution of that population around a company. Calculations are made to determine the size of the risk of one, ten or more simultaneous casualties in that population. This probability of a disaster of a certain size is known as the group risk.
The level of risk that is still considered acceptable depends on the size of the disaster. An accident with 100 deaths results in more disruption, suffering and grief than an accident with ten fatal casualties. The limit set for the probability of a disaster with 100 deaths is therefore one hundred times lower than for a disaster with ten deaths.
The probability of a group of at least ten people being killed in the vicinity of airports is, for example, once in 700 years. In the case of railway marshalling yards, the probability is once in 25,000 years. The probabilities for accidents involving 100 deaths or more are smaller. The probability of larger accidents is highest in the case of LPG stations: the probability of more than 100 people being killed at a single location in the case of LPG stations is once in 5,000 years. This probability is once in 10,000 years in the case of the transportation of hazardous goods, once in 33,000 years in the case of airports and once in 200,000 years in the case of companies required to produce safety reports.
Numbers are not everything.
These numbers fail to tell us what the public thinks about risks of this kind. For instance, for the average person, the probability of being involved in a traffic accident is much higher than the probability of being a casualty in a firework disaster. Nevertheless, few people will see this as a reason for being casual about the risks of a firework factory located in the vicinity. The degree to which the government takes steps to deal with a particular danger or risk is a question of public opinion and political considerations.
There is another problem with describing danger in terms of numbers: calculations about many risks are virtually impossible because there is simply not enough information. This is the case, for example, when it comes to the dangers involved in the use of a cell phone or the risks associated with genetic manipulation. Nevertheless, risks of this kind can be very important to the general public.
What risk statistics also fail to tell us is how many people fall ill as a result of an accident; they deal exclusively with fatal casualties as the direct consequence of an accident. In the case of nuclear power plants, they do take into consideration the number of deaths in the long term as a result of the intake of radioactive substances released during nuclear accidents. These consequences often only emerge after several years or decades. Although nuclear power plants are not covered by external safety policy, the Environmental Data Compendium does include two indicators for the risks of European power plants.
What is the aim of the policy?
The government's primary objective is to reduce the local risk. The government wants people who live in the Netherlands in locations where there is an unaccepted risk to live more safely. In principle, nobody should live in a location where the risk exceeds 1 in 1 million (10-6).
Which companies cause excessive risks at present? Of a total of approximately 4,000 risk companies, a number fail to meet the standards at present. In these cases, the risk is therefore excessive, for example because a municipality has, in the past, approved building in the 'risk contour'. There may also be problems with overdue maintenance or with a permit. The companies that do not meet the standards include: 500 to 600 LPG stations, 40 ammonia cooling plants, approximately 75 storage locations for pesticides and chemicals, and approximately five railway marshalling yards.
In concrete terms, the government wishes to 'remedy' these shortcomings. That will be done in accordance with the fourth National Environmental Policy Plan (NEPP4) (Ministry of Housing, Spatial Planning and the Environment, 2001) in three stages:
- 1. The registration of high-risk activities and informing the public and government authorities about them. This process should start no later than 2002. See, for example, the risk map for the province of Friesland.
- 2. All the data will then be checked, as will the permits and compliance with the permits. Checks will also take place to ensure that adequate emergency plans are in place. This phase must be completed in 2003.
- 3. Finally, any shortcomings must be remedied. This can be done in a variety of ways: by means of additional measures within the company or by demolishing houses in the vicinity. If this is not possible, NEPP4 states that "termination of operations is an option". The measures to be taken within a company can vary widely: from better fire-fighting resources and the collection of leaking liquids to changes in the size and frequency of transports and the reduction of production and storage quantities.
