No abstract

Abstract

No abstract

Rivers in historical cities, such as the Arno River in Florence, are typically characterised by unique complex-shaped hydraulic structures (such as bridges and weirs). The flow interaction with these structures can lead to a fully 3D flow field which cannot be properly investigated with commonly employed 1D, and even 2D, hydraulic models. Nowadays, 3D computational fluid dynamics (CFD) tools can be successfully used in river management context. Florence is characterised by a high risk of flooding and the disastrous consequences of such events being greatly increased due to its inestimable artistic heritage. The main cause of flooding is the limited hydraulic conveyance capacity of the Arno River in Florence due to several complex hydraulic structures along the reach. The present work represents the first 3D hydraulic model of the Arno River in the urban reach of Florence. The geometric model was created using the 3D bed topography of the river surveyed in 2015. The hydraulic model was calibrated and validated using discharge and water stage field data measured in 2016. The 3D model can be used as a more realistic tool for exploring mitigation solutions for the reduction of hydraulic risk. K E Y W O R D S flood damages, flood mitigation, hydraulic modelling, hydraulic structures

In agro-urban environments, the water resource conveyed by rural channels is susceptible to a gradual impoverishment due to the continuous combined sewer overflow release, constituting a pending and urgent issue for water management companies and the entire community. Reliable one-dimensional longitudinal dispersion coefficients D are required to model and study the hydrodynamics and water quality patterns at the scale of rural channel networks. Empirical formulas are usually adopted to estimate D but the accuracy in the prediction could be questionable. In order to identify which are the most suitable formulas to determine D in rural channels, field tracer measurements were carried out in three rural channels with typical geometry and configuration. The obtained D values were then compared with the most commonly used predicting formulas that the literature provides. The accuracy of the predictors was further checked by simulating different flow rates inside the tested channels by using a one-dimensional hydraulic model. Starting from the obtained results, indications and guidelines to choose the most suitable formulas to predict D in rural channels were provided. These indications should be followed when developing realistic quality models in the agro-urban environments, especially in those cases where direct measurements of the longitudinal dispersion coefficient D are not available.