How Can We Use Risk Models?
Risk models which have a physical basis inherently allow improved forecasting of events and their impacts. This is particularly relevant for events in which the total risk might be affected by changes in the natural environment, such as through climate change, or by changes to the built environment as a result of new building codes and construction, or by changes to the social environment such as through increased public awareness.
Risk models can be used to perform cost-benefit analysis for various forms of mitigation involving short term solutions, such as early warning and response, along with long term resolution, such as land-use planning and improvements to building codes and infrastructure. Although some residual risk is inevitable, it can be moderated through insurance when it is available. Risk models can also be used to develop disaster scenarios for emergency response and recovery, to improve community awareness and to evaluate risk acceptance thresholds for a wider range of stakeholders.
The effective development of risk models requires comprehensive data which embraces hazard, exposure and vulnerability. Model development also needs to be carried out by experts from a range of physical sciences including the earth sciences, meteorology and hydrology as well as from engineering in areas such as structural engineering, the environment, software and computational methods. It also requires experts from the social sciences, including sociology, economics and emergency management.
Most models capture risk in a rather limited context and are usually confined to the direct damage or cost of a future disaster. Research is needed to extend these estimates to include indirect effects such as loss of income and quality of life as well as other social, political, and economic factors which invariably play a role in decisions about risk. Advances in risk modelling also can be used to develop scenarios for natural disaster response and urban planning, to educate the community and to evaluate risk acceptance thresholds for a wide range of stakeholders.
For future models to be effective, broader input and more comprehensive data will be required from a range of stakeholders and end users, such as governments, emergency managers, planners, insurance companies, utility managers and operators. Models and databases also need to be tested and validated using a variety of sources, including data collected from past disasters. Tools will need to be developed to translate the complex analysis into user friendly information to support decision-making on risk treatment options.