Heavy storms and flash floods in the Mediterranean basins: new observations and concepts
The main characteristics of flash floods in the Mediterranean basin may be described as follows:
1. The occurrence of heavy rainfall in autumn and winter may produce flash floods in catchments and streams, which remain dry throughout much of the year. These flash floods are of short duration (from a few minutes to a few hours) and have high specific flood peaks (up to 20 m3/(km2 s)). The ephemeral character of most rivers and the small size of the catchments make more acute the observation problem of these events, which are poorly sampled by conventional hydrometeorological networks.
2. During flash floods, soil erosion and sediment transport are important and may lead to the failure of flood-defensive engineering structures (reservoirs, spillways, gates).
3. In karst areas, which make up more than half of the Mediterranean drainage basin, flash floods may be more intense and unpredictable than elsewhere. Excessive flooding occurs in these areas after the karstic cavities are filled by a huge amount of rainfall water.
4. Heavy concentrations of population in urban and residential areas around the centres of historic cities have, in many cases, resulted in the occupation of the beds and floodways of ephemeral streams. This phenomenon mainly occurs near the coastal areas, where tourist activity has dramatically increased in the past decades.
Flash floods do pose a continuing threat to life and property in the Mediterranean basins, and flood warning can potentially save lives. Due to the short lead times associated with flash floods, however, there is limited potential for saving property. Past experience shows that flash flood warning should include a hydrologic forecast model. Without a model, it can be difficult to forecast the flood potential of storms that have complex space-time texture, particularly when the storm is near the flood/no flood threshold. The fundamental problems for flash flood forecasting can be summarised as follows:
i) the downscaling problem due to the incoherent space and time scales between atmospheric models and the flash-flood triggering processes,
ii) the ungauged basin problem due to the fact that the small basins prone to flash-floods are seldom gauged and must be modelled without calibration and
iii) the question of the limits of the soil retention control of runoff under the considered range of rain accumulations.
The strong coupling between atmospheric and hydrologic generating mechanisms, the wide and continuous range of space and time scales concerned, the low local occurrence of these events and their high observational uncertainties makes it necessary to break many traditional fences between disciplines. Advances in understanding flash-floods can only be gained through interdisciplinary studies based on observation and modelling methods shared over wide areas like the Mediterranean basin.
I advocate that a central role in the research agenda is plaid by the development of a Mediterranean long-term Hydrometeorological Observatory (HO) for flash flood monitoring. The common objective of the HO is to observe flash flood by combining: conventional hydrometeorological monitoring; weather radar observations; complementary information acquired from field surveys executed during the days following the event.
Post event field surveys mainly focus on river flood discharges at small scales, compensating the usual lack of runoff data (due to sparsity of the hydrometric network and to its fragility during extreme flood events).
The main motivation for the development of a Mediterranean HO is to observe these locally rare events wherever they occur in the region. The presentation will provide a summary of early results obtained from the implementation of this flash flood observation strategy during two European Research projects.
HyMeX – Hydrological cycle in the Mediterranean Experiment 2010-2020