Only 鈥渓ow confidence鈥� should be placed in most of the risk analyses of terrorist threats conducted by the US Department of Homeland Security, concludes a by a National Research Council panel.
Created after the 9/11 attacks to cover national perils ranging from terrorism to natural disasters, the DHS got off to a rough start. Critics blasted its lack of attention to natural disasters after its bungled response to hurricane Katrina.
Yet the panel finds that DHS models for natural disaster risks are 鈥渘ear the state of the art鈥� and well suited to assess the effects of earthquakes, floods and hurricanes.
Advertisement
However, it warns that different techniques are needed to assess the risks posed by terrorism. 鈥淣atural hazard risks are quantifiable,鈥� says panel chairman of Duke University in Durham, North Carolina. 鈥淢any terrorism risks are not quantifiable because of uncertainties arising from having an intelligent adversary.鈥�
Changing tactics
Standard risk models, such as those the DHS uses, assume that threat, vulnerability and the consequences of risks are constants. But the panel notes that humans, unlike natural disasters, change their targets and tactics in response to protective measures that the authorities take against them, so the risk factors are no longer constant. The report urges the DHS to develop risk models that react dynamically to account for this.
It also says the DHS 鈥渁ppears to be at a very immature state with respect to characterising uncertainty鈥� in its risk assessments, an issue that is key to making sound decisions based on the risks.
Moreover, the report says it may not be possible to quantify all risks. The risks posed by the fear and social disruption caused by terrorists are much harder to quantify than the risk of a bridge being blown up, for example. 鈥淨ualitative comparisons can help illuminate the discussion of risks and thus aid decision makers,鈥� it says.
