Investigation of the gridded flash flood Guidance in a Peri-Urban basin in greater Athens area, Greece

In this research work, an implementation of the gridded Flash Flood Guidance (FFG) method is conducted for the prediction and evaluation of flash floods in Greece. The FFG system is a well-established Early Warning System (EWS) for the estimation of flash flood threats, especially in small ungauged basins, where flash floods are the most dominant form of flooding. First, an overview of the gridded FFG system is conducted, where all computations are performed at grid level. The methodology applied consists of the derivation of the threshold rainfall, referred to as the FFG. The catastrophic flash flood event that occurred on the 15th of November 2017 and caused the loss of 24 human lives and extensive economic damage within the Mandra settlement, in the Attica region in Greece, is used to assess the lead time provided by the system. Finally, a sensitivity analysis is performed regarding the two main aspects of the system, i.e. the threshold runoff and the estimation of the soil moisture conditions, and their impact on the generated lead times. Results show that the most crucial aspect of the system is the soil moisture conditions, since in wet conditions a 10% deviation on a scale from zero to one, can result in a complete time-step loss of the possibilities provided lead time, while in dry conditions the deviation is much larger. Finally, concerning the studied event, results show that if the system was operational, and the forecasted rainfall matched the values of the actual rainfall, at least a five-hour lead time warning could have been issued based on the results of the system.

Automated summary

This research work implements the gridded Flash Flood Guidance (FFG) method to predict and evaluate flash floods in Greece. The FFG system serves as an Early Warning System (EWS) for flash flood threats, especially in small ungauged basins. The study examines the system's lead time using the catastrophic flash flood event that occurred in Greece in November 2017. The sensitivity analysis shows that soil moisture conditions are the most crucial aspect of the system and can significantly affect the lead time. If the system had been operational during the studied event, a five-hour lead time warning could have been issued based on the forecasted rainfall.