Monitoring Sediment Transport During Floods in Tuscany

  • Simona Francalanci
  • Enio Paris
  • Luca Solari
  • Lorenzo Minatti
  • Giorgio Valentino Federici
Conference paper


This work describes the monitoring activity of flow and sediment discharges carried out in four river stations on four different rivers in Tuscany: Arno, Serchio, Magra and Versilia characterized by a large difference in the drainage areas ranging from about 100 km2 to about 4,100 km2. From December 2006 to November 2012, 43 floods events were monitored and measured in these stations, with the aim of deriving flow-sediment rating curves. Field measurements include more than 150 samples of bed load and suspended load and flow discharges up to about 700 m3/s. Laboratory analyses of collected samples show that bed load was mostly composed by gravel in the case of the Versilia and Magra rivers, while Arno and Serchio showed a dominant sand transport. The measured data constitute a precious and unique dataset at the regional scale for calibrating sediment budget model and morphodynamic model, and for predicting the sand and gravel volume fractions annually conveyed to the sea.


Sediment transport Field measurements Rating curve Gravel-bed river 

70.1 Introduction

The knowledge of sediment load carried by a river is a fundamental, yet difficult, task.

Estimates of sediment transport rates are required in the field of stream restoration, river engineering, and hydro-geological risk (Ferguson 1986). Problems such as sediment budgets and morphodynamic evolution of the shoreline, relys on verified estimates of sediment fluxes. An example of these problems is given by the shoreline of Tuscany region, located in the central part of Italy, which nowadays is, in some areas, experiencing large erosion phenomena, of the order of several meters per year, and the primary source for summer tourism is disappearing.

In order to have a better understanding of this type of problems, a reliable estimate of the volume of sediments annually delivered by the rivers to the coastal zone is needed. This estimate can be obtained through the development, at given monitoring stations, of accurate flow-sediment rating curves, i.e. relationships between flow and sediment discharge (Jansson 1996; Asselman 2000; Wilcock 2001).

In the framework of the present monitoring project, the rating curves were derived using an integrated approach combining several methods: field measurements, laboratory analyses, numerical modelling, analysis of the accuracy of the estimated flow discharges. A detailed description of the approach is reported in Francalanci et al. (2013), and applied to the case of the Versilia river: it appears to be suitable for the formulation of physically-based flow-sediment rating curves in gravel-bed rivers characterized by rapid and intense flood events. Moreover, it overcomes the limitations related to either purely field investigations, which suffer from a limited amount of measurements, or theoretical studies, which require some field calibration.

In this work, we present the whole set of collected measurements, which were further used to calibrate the sediment rating curves. Despite the large difference in the drainage areas and sediment size, results show that rating curves (total sediment discharge vs. flow discharge) display a similar behaviour thus suggesting the existence of common scaling laws.

70.2 Field work

An extensive monitoring activity of flow and sediment discharge was started in four monitoring stations in Tuscany: “Nave di Rosano” on Arno River upstream of Florence, “Ripafratta” on Serchio River close to the sea mouth, “Barbarasco” on Magra River close to Aulla, and “Ponte Tavole” on Versilia River close to the sea mouth. The location of the monitoring stations is shown in Fig. 70.1.
Fig. 70.1

Position of the four monitored stations in Tuscany (small circles and triangles indicate the position of the hydrometers for stage monitoring by the Regional Hydrological Agency)

From December 2006 to November 2012 43 floods events were monitored and measured in these catchments. A summary of the monitored events and of the characteristics of the basin are reported in Table 70.1.
Table 70.1

Summary of monitored events and drainage areas

Monitored station





Drainage area at monitored station (km2)

Basin area (km2)

D50 (mm) at monitored station

























For each flood event, the monitored cross-section was divided into a given number of vertical transects, properly chosen to minimize the errors due to flow unsteadiness and to optimize the description of the lateral variation of flow and sediment discharge. In each vertical, the following measurements were collected: water depth; flow velocity at several water depths by means of an USGS type AA current meter; sampling of suspended load and bed load were collected using, respectively, the depth-integrating reel type US D-74 and classical Helley-Smith bed load sampler with a 7.6 cm square intake opening and a sample bag with a 0.25 mm mesh of polyester.

70.3 Analyses and Results

For each event we analyzed the flow discharge, following the “competence area” method (Turnipseed and Sauer 2010), and the sediment discharges which were analyzed in term of bed load, suspended load and wash load; the total sediment load was assumed to be the sum of bed load and suspended load. The whole set of data is reported in Fig. 70.2 in terms of flow discharge and total load: Arno and Serchio stations show the highest flow discharges which correspond to sediment discharges of the same order of Magra with lower flow discharges; the two highest discharges for Serchio correspond to a very intense and extraordinary flood event. Versilia station shows the smallest flow and sediment discharges; this is presumably related to the different drainage areas of these basins, and to the hydraulic characteristic of the reaches.
Fig. 70.2

Flow and sediment discharges in the monitored events

Additionally the bed load samples gave an estimate of the relative content of sand and gravel fractions which are transported as bed load (Fig. 70.3): gravel fraction is predominant in Versilia and Magra stations, while sand fraction prevails on Arno and Serchio stations, which correspond to larger drainage areas. Finally Fig. 70.4 gives details of the characteristic diameters of the bed load: Versilia, Magra and Arno stations show the presence of medium and coarse gravel as transported fractions, while Serchio shows finer fractions; this is due to the fact that the monitored station of Ripafratta on Serchio is closer to the sea mouth, if compared to the drainage area.
Fig. 70.3

Averaged gravel and sand percentages, Pg and Ps, in the bed load samples in the monitored stations

Fig. 70.4

Averaged characteristic diameters (D16, D50, D84) in the bed load samples in the monitored stations

70.4 Conclusions

The paper summarizes field sampling in unwadeable and flashy flood events occurred in Tuscany from 2006 to 2012 in four river stations. These data are precious because event based gravel-sand mixed transport data in rapidly changing conditions are largely missing, in particular for gravel-bed rivers in small catchments.

The collected measurements were used to calibrate reliable flow-sediment rating curves for each station, to be useful for river management and restoration practices and for morphodynamic models devoted to predicting the evolution of the river-coastal system.



This work has been funded by Regione Toscana within the projects “Realizzazione del primo stralcio della rete di monitoraggio su tre corsi d’acqua della Regione Toscana”, “Realizzazione del secondo stralcio del monitoraggio del trasporto solido nei corsi d’acqua della Toscana”, “Monitoraggio delle portate liquide e solide sul Fiume magra”. CERAFRI-LAV is also acknowledged for sustaining the publication of this work.


  1. Asselman NEM (2000) Fitting and interpretation of sediment rating curves. J Hydrol 234:228–248CrossRefGoogle Scholar
  2. Ferguson RI (1986) River loads underestimated by rating curves. Water Resour Res 22(1):74–76. doi:10.1029/WR022i001p00074 CrossRefGoogle Scholar
  3. Francalanci S, Paris E, Solari L (2013) A combined field sampling-modeling approach for computing sediment transport during flash floods in a gravel-bed stream. Water Resour Res 49:6642–6655. doi:10.1002/wrcr.20544 CrossRefGoogle Scholar
  4. Jansson MB (1996) Estimating a sediment rating curve of the Revezanton river at Palomo using logged mean loads with discharge classes. J Hydrol 183:227–241CrossRefGoogle Scholar
  5. Turnipseed DP, Sauer VB (2010) Discharge measurements at gaging stations: U.S. geological survey techniques and methods book 3, Chap. A8, 87 p. (Also available at
  6. Wilcock PR (2001) Toward a practical method for estimating sediment-transport rates in gravel-bed rivers. Earth Surf Process Landforms 26:1395–1408CrossRefGoogle Scholar

Copyright information

© Springer International Publishing Switzerland 2015

Authors and Affiliations

  • Simona Francalanci
    • 1
  • Enio Paris
    • 2
  • Luca Solari
    • 2
  • Lorenzo Minatti
    • 1
  • Giorgio Valentino Federici
    • 2
  1. 1.CERAFRI—Center of Research and Advanced Education for Hydrogeological Risk PreventionRetignano di StazzemaItaly
  2. 2.Department of Civil and Environmental EngineeringUniversity of FlorenceFlorenceItaly

Personalised recommendations