Proposal of New Hazard Metrics to Mainstream Flood Risks into Urban Planning. Application to the Juan Diaz River Basin in Panama City

It is generally acknowledged that urban planning must take into account urban risk factors. The standard approach is based on the definition of the spatial hazard as some function of the local depths and/or velocities: H=f(h,v). While useful, this approach is very sensitive to the choice of the design event, and cannot take into account the shape of the hazard distribution, failing to provide a more nuanced information about the suitability of the territory for urban development. This paper approaches this problem by proposing three new hazard metrics which reflect the local conditions for urban development:

• Mean hazard index (MHI). It is the mean value of the probability density function (pdf) of the hazard. By averaging the hazard for all the return periods, it encompasses information about all the potential events. This is equivalent to the calculation of the expected annual losses (pure premium), but using the hazard instead of the economic losses.
• Apparent hazard index (AHI). It is the hazard that people can recognize, based on their own experience and memory of recent flood events. It will be estimated, using a Bayesian framework, as the mean value of a posterior pdf of the hazard, where the prior is the standard distribution; the likelihoods are estimated by a memory factor, directly related to the frequency of the events (people only assign credibility to what they have experienced recently).
• Latent hazard index (LHI). It is defined as the complementary of the apparent hazard. It is also obtained as a posterior distribution of the hazard, using as likelihoods the complementaries (1-X) of the ones used to obtain the AHI.

While MHI is a global indicator of the hazard, AHI is a proxy for the social perception of it, reflecting the shape of the lower tail of the hazard pdf; it represents the amount of hazard that people will often face during a life time, and should potentially be avoided by standard risk reduction measures. On the other hand, LHI informs about the shape of the right-tail of the hazard distribution, and indicates the amount of hazard that will be difficult to mitigate. The usefulness of these indices will be demonstrated for the case study of the Juan Díaz Basin in Panama City.