Multitemporal geomorphological analysis to predict flash flood impacts: its contribution to inform flood risk management

The 2016 implementation of the EU Flood Directive in Spain defines within the flood-prone zones the Preferential Flow Zone (Zona de Flujo Preferente, ZFP). This zone includes a) broadly, the area where the floods flow is concentrated; b) for the 100 years return period flood, the intensive drainage waterway and the zone dangerous to persons. The ZFP is usually defined for the 100 years flood applying hydraulic modelling. However, the calculation of the 100 years flood poses multiple limitations. For instance, different probability distributions produce different results for the same data series, or for rainfall and discharge data, depending on the time interval considered in the calculation, the results are also different. Regarding rainfall, the meteorological radar data are still too new to extrapolate to 100 years. The destruction of meteorological and gauging stations during storms and floods is not rare; hence, a lack of data on major events in the data series can deeply affect the calculations. Furthermore, similar rainfall can produce different discharges due to differences in the antecedent conditions or to land use changes. All the above and the climate change, question the hypothesis of stationarity at the base of the floods return period concept¹ and, thus, its calculation reliability.

Since the middle of the 20^th century, significant socio-economic and land use changes occurred in the western Mediterranean region, resulting in changes in the morphology of rivers (e.g., reduced channel section, entrenchment). The record of these morphological changes, including the effects of major floods, can provide insights to define the high-energy flow zone or ZFP. This work contributes to determine the flash flood effects and, therefore, to define the ZPF, through multitemporal geomorphological analysis applied to a case study of the upper basin of the Francolí river in Catalonia, Spain. It was affected by several major floods in 1874, 1930, 1994 and 2019, where the first and the last events were the largest and of quite similar, centenial magnitude. Different reaches of the river are studied and compared to validate the analysis: reaches where 1994 and 2019 flood were similar and reaches where these floods were of very different magnitude; reaches where all the basic dataset is available (1946, 1956, 1995 post flood, pre and post 2019 orthophotos; 2003 detailed DTM; stereo photographs, post 2019 flood field data and GNSS-RTK data of river cross sections) and reaches with lack of some data (especially of the 1995 post flood image). Historical information (water levels attained by the past floods and the calculated discharges) are also used to complement and validate the geomorphological analysis results.

With this work we test whether the main geomorphic effects of the 2019 flood could have been predicted using the multitemporal geomorphological analysis. The ZFP can be reasonably determined for major floods in this Mediterranean river. This multitemporal geomorphological analysis appears as a good complementary tool to inform flood risk.

¹ Sofia, G., E. I. Nikolopoulos, L. Slater (2020), It’s time to revise estimates of river flood hazards, Eos, 101, https://doi.org/10.1029/2020EO141499. 16 March 2020.