|
Early warning systems mitigate risk by reducing consequences, giving sufficient lead time to implement actions to protect persons and/or property. Monitoring and early warning systems are more than an implementation of a technological solution. The human factors, social elements and information communication are essential parts.
An early warning system is also decision making authorities (e.g. government agencies, local authorities, police/fire), the development of the form and content of warnings issued (how are these understood and relayed in the protected populations), social aspects of how a population responds, the implementation of emergency plans and services to assist the population when it responds, and plans for reconstruction/recovery when the emergency is over. Installing the technology part of an early warning system without considering the social aspects, in particular the response to the system, results in an incomplete system, which may simply create a new type of emergency. Although geologists and engineers are not directly involved in the ''human factor'' part of the system development, they cannot ignore the human aspect.
Early warning systems are by definition critical systems, and as such it is absolutely necessary that the system remains in operation at all times. The system requires a chain of functional components: monitoring of key parameters - raw data transmission - data processing, interpretation and modelling - decision making - issuing of warnings - reaction of protected population and emergency services. A breakdown at any point in this chain renders the system out of service.
In general, good measurement solutions in critical systems employ both alternative measurement technologies (e.g. to different ways to measure a quantity) as well as redundancy in the sensors themselves. To issue an alarm, one must make a prediction of what is about to happen. It is also necessary that sufficient lead time is available to be able to take an effective action. Making a prediction requires an understanding of the physical process being evaluated (a quantitative understanding allowing modelling) or sufficient experience to make correlations (qualitative understanding allowing statistical methods).
The main challenge is designing the threshold values for an early warning system is to balance the conflicting requirements of capturing the critical events with a sufficient margin while avoiding false alarms. A population subjected to false alarms reacts the first time, the second time, and maybe a third time; but quickly becomes tired of the alarm and the value of the alarm loses credibility. Initially an implicit trust exists between the affected population and the "experts", or at least an acknowledgment of the system can exist, which can eventually lead to trust. However, that initial trust is quickly destroyed by system failure (false alarms), and the trust will be difficult (or impossible) to win completely back.
|
