Flood design recipes vs. reality: Can predictions for ungauged basins be trusted? – A perspective from Greece

As a result of its highly fragmented geomorphology, Greece comprises hundreds of small- and medium-scale steep hydrological basins of usually ephemeral regime. Typically, their drainage area does not exceed few hundreds of km2, while the vast majority of them lacks of measuring infrastructures. For this reason, and despite the great scientific and technological advances in flood hydrology, the everyday engineering practices still follow simplistic rules-of-thumb and semi-empirical approaches, which are feasible and easy to implement in ungauged areas. In general, these “recipes” have been developed many decades ago, based on field data from few experimental catchments abroad. However, none of them has ever been validated against the peculiarities of the hydroclimatic regime and the geomorphological conditions of Greece. This has an obvious impact on the quality and reliability of hydrological studies, and, consequently, the safety and cost of the related flood-protection works. In order to provide a consistent design framework and ensure realistic predictions of the flood risk in ungauged basins (which is key issue of the 2007/60/EU Directive), it is imperative to revise the rather outdated engineering practices, by incorporating methodologies that are adapted to local peculiarities. In particular, the collection of reliable hydrological data is essential for evaluating and verifying the existing “recipes” and updating the design criteria. In this context, we are elaborating a research program titled “Deukalion”, in which we already have developed a fully-equipped monitoring network, extending over four pilot river basins. Preliminary outcomes, based on historical flood data from Cyprus and Greece, indicate that a substantial revision is required within multiple aspects of the flood modeling procedure.