Sedimentation Processes on Intertidal Areas of the Lagoon of Venice: Identification of Exceptional Flood Events (Acqua Alta) Using Radionuclides

Particular atmospheric conditions produce frequent storm surges in the Lagoon of Venice, locally called "acqua alta": the highest event of this kind was registered in 1966. The process became of some importance in the last 100 years, when man-made subsidence caused a gradual sinking of the town and lagoon’s bed. Four cores were collected in a range of intertidal environments of the Northern lagoon to undertake radionuclide studies using profiles of natural 210Pb (using the Constant Rate of Supply and Constant Initial Concentration models) and anthropogenic 137Cs. The best agreement between the three dating methods was found at San Giacomo, an eroding saltmarsh at the edge of a navigation channel, with a sedimentation rate ranging from 0.22 to 0.29 cm yr -1. This site recorded the flood of 1966 as a characteristic break in the Pbex profile, as dated by the CRS model. For the Cona tidal flat, both the CRS model and the position of the Cs peak-marker gave similar accretion rates, 0.16 and 0.18 cm yr-1. However, two different CIC accumulation rates were calculated, 0.29 cm yr-1 for the deepest section of the core and 0.17 cm yr-1 for the uppermost part. The break in the 210Pbex profile, again corresponds to the flood of 1966. The effects of subsidence were recorded as an increase in accumulation rate between 1910 and 1931, when there were up to 15 floods per year (1926). Higher sedimentation took place during the period 1958-1973, when years with over than 50 flood events were frequent. The maximum deposition rate (0.43 cm yr-1) occurred again around 1967, consistent with the record of the exceptional flood, if the accuracy of the dating is taken into account. The sedimentation rates calculated for the two other mudflats, Rosa and Saline, were more problematic to interpret because of downcore mixing and/or the occurrence of reducing conditions. ADDITIONALINDEXWORDS: Sedimentation rates, radionuclides, floods, intertidal areas Journal of Coastal Research SI 36 139-147 (ICS 2002 Proceedings) Northern Ireland ISSN 0749-0208 Lagoon of Venice 140 Journal of Coastal Research, Special Issue 36, 2002 STUDY SITE AND AIMS OF THE RESEARCH The lagoon of Venice is the largest lagoon in the Mediterranean and is located in the northern part of the Adriatic Sea (Figure 1). The lagoon has a drainage basin of 1850 km2 which provides a mean yearly freshwater input of 35.5 m3s-1 (ZULIANI et al., 2001), of which the main contributors are Silone River (23.1%), the Dese River (21.1%), the Naviglio-Brenta (14.3 %) and the TaglioNuovissimo (13.2%) partially channelised systems. It has an extension of 550 km2 and a microtidal regime (average range at Spring Tides in the order of 0.8 m). Particular atmospheric conditions like strong southerly winds (Scirocco) and low atmospheric pressures produce frequent storm surges, which can increase the maximum tidal range (PIRAZZOLI, 1991; CANESTRELLI et al., 2001). These Scirocco winds typically have maximum speeds in the order of 50-60 km h -1, they generally occur in spring and autumn, producing extensive flooding of the historical city centre (PIRAZZOLI, 1991). A range of man-made changes has affected the lagoon over the period spanning from the 15th to the 19th century. G AT TO and CARBOGNIN (1981) present a succinct summary of these anthropogenic activities that included diverting river outflows outside the lagoon, opening and widening the tidal inlets, the creation of water ways for navigation towards the inner coastal area. The last century experienced a boom of commercial activities in the lagoon, leading to exploitation of ground water resources that generated high subsidence rates (GAT TO and CARBOGNIN, 1981; CARBOGNIN and TARONI, 1996). The last factor caused a Relative Sea Level Rise, calculated for the period 1930-1970, vary between 3.8 mm yr1 (CARBOGNIN and TARONI, 1996) and 6 mm yr1 (SESTINI, 1992, BONDESAN et al., 1995). Following the end of groundwater withdrawal in the early 1970s, land subsidence slowed down, to the extent that it is no longer a problem and some areas have even experienced a rebound (CARBOGNIN and TOSI, 1995; CARBOGNIN and TARONI, 1996). Figure 1. Index map with the location of the sampled sites. Ciavola, Organo, Vintró and Mitchell 141 Journal of Coastal Research, Special Issue 36, 2002 content and the dry bulk density were calculated from the samples after drying, while the porosity was calculated using the formula of RAVICHANDRAN et al. (1995). The sediments were analysed for 2 1 0Pb and 1 3 7Cs by γ s p e c t r o m e t r y, from the gamma emission at 46.52 and 661.66 keV, respectively. Three detectors were used in this study, all of them supplied by EG&G OrtecTM, one a low energy photon spectrometer (LO-AX or LEPS), and two high-resolution, low-background _ detectors (high purity ntype germanium detectors LB-GMX). The eff i c i e n c y calibration was directly performed using a standard solution (SRM-4276, supplied by the U.S. National Institute for Standards and Technology) containing 125Sb+125mTe, 154Eu and 155Eu, thus providing 13 calibration peaks between 27 and 1596 keV. A separate mono-energetic 210Pb standard (RBZ24, supplied by Amersham International plc) was used to calibrate at 46.52 keV. Counting times were typically ca. 48 hours, no correction was made for salt content in the sediments. RESULTS AND DISCUSSION Sedimentary environments, dating and sediment accretion rates The determination of recent sedimentation rates is based on two fundamental assumptions underlying 2 1 0P b e x accumulation: (1) the "constant initial concentration" (CIC) model, which assumes that the initial Pbex concentration in the deposited sediment is always the same, regardless of changes in the sediment accumulation rate; (2) the "constant rate of supply" (CRS) model, which assumes that the supply of 2 1 0P be x to the accreting material is constant when averaged over a timescale of 100-200 years (APPLEBYand OLDFIELD, 1978). In the case of sediment of unvarying composition deposited at a constant rate, the CRS and CIC models are equivalent. The 137Cs method is fundamentally different from the 210Pb method in that it provides date "markers". The depth at which the maximum peak activity is observed can be used to estimate the depth of the 1963 and 1986 horizons, while the depth at which significant 137Cs activities are first recorded can be used to estimate the position of the surface in the mid-1950s (WALLING and HE, 1997). While the 210Pb dating method gives an average accumulation rate for the past 100-200 years, 137Cs is only applicable for the period beginning from 1954 until present. Palude di Cona (Cona 30) This core was collected on a tidal flat colonised by Zostera noltii, with the surface sediments covered by a microalgal mat. Grain-size analyses indicate a poorly sorted population with a net predominance of the silt fraction (5 to 8 phi) throughout the core, with a very fine sand component In the central part of the lagoon, the subsidence lead to a progressive drowning of tidal flats and marshes at the side of canals, with a deepening of areas that once became exposed at Low Spring Tides (velme), generating vast flat shoals with an average water depth in the order of 1 m below Mean Sea Level (MSL). Secondary effects on the morphology of the lagoon bed were the disappearance of tributary channels branching away from the main ones and a progressive deepening of the main canals, together with a scouring of the bed in the inner part of tidal inlets. Feed-back processes between estuarine circulation, transport of sediments and phytobenthos were the subject of the Project F-ECTS (EU-MAST III Programme) that supported collection of four push cores, in August 1998, on the salt marshes and mud flats of the northern basin of the lagoon. Radionuclide measurements were undertaken to establish the down-core profile variations of excess-210Pb and 137Cs to obtain the sediment accumulation history over the past century.