Uncertainty of the peak flow reconstruction of the 1907 flood in the Ebro River in Xerta (NE Iberian Peninsula)

Thumbnail Image
Ruiz Bellet, Josep Lluís
Castelltort Aiguabella, F. Xavier
Balasch Solanes, J. Carles (Josep Carles)Balasch Solanes, J. Carles (Josep Carles) - ORCID ID
Tuset Mestre, Jordi
Other authors
cc-by-nc-nd (c) Elsevier, 2017
Journal Title
Journal ISSN
Volume Title
There is no clear, unified and accepted method to estimate the uncertainty of hydraulic modelling results. In historical floods reconstruction, due to the lower precision of input data, the magnitude of this uncertainty could reach a high value. With the objectives of giving an estimate of the peak flow error of a typical historical flood reconstruction with the model HEC-RAS and of providing a quick, simple uncertainty assessment that an end user could easily apply, the uncertainty of the reconstructed peak flow of a major flood in the Ebro River (NE Iberian Peninsula) was calculated with a set of local sensitivity analyses on six input variables. The peak flow total error was estimated at ±31% and water height was found to be the most influential variable on peak flow, followed by Manning’s n. However, the latter, due to its large uncertainty, was the greatest contributor to peak flow total error. Besides, the HEC-RAS resulting peak flow was compared to the ones obtained with the 2D model Iber and with Manning’s equation; all three methods gave similar peak flows. Manning’s equation gave almost the same result than HEC-RAS. The main conclusion is that, to ensure the lowest peak flow error, the reliability and precision of the flood mark should be thoroughly assessed.
Journal or Serie
Journal of Hydrology, 2017, vol. 545, p. 339–354