The estuarine geochemical reactivity of Zn isotopes and its relevance for the biomonitoring of anthropogenic Zn and Cd contaminations from metallurgical activities: Example of the Gironde fluvial-estuarine system, France

Zinc stable isotopes measurements by MC-ICP-MS, validated by laboratory intercalibrations, were performed on wild oysters, suspended particles and filtered river/estuarine water samples to provide new constraints for the use of Zn isotopes as environmental tracers. The samples selected were representative of the long range (400 km) transport of metal (Zn, Cd, etc.) contamination from former Zn-refining activities at Decazeville (i.e. δ66Zn > 1 ‰) and its phasing out, recorded during 30 years in wild oysters from the Gironde Estuary mouth (RNO/ROCCH sample bank). The study also addresses additional anthropogenic sources (urban and viticulture) and focuses on geochemical reactivity of Zn in the turbidity gradient and the maximum turbidity zone (MTZ) of the fluvial Gironde Estuary. In this area, dissolved Zn showed a strong removal onto suspended particulate matter (SPM) and progressive enrichment in heavy isotopes with increasing SPM concentrations varying from δ66Zn = -0.02 ‰ at 2 mg/L to +0.90 ‰ at 1310 mg/L. These signatures were attributed to kinetically driven adsorption due to strongly increasing sorption sites in the turbidity gradient and MTZ of the estuary. Oysters from the estuary mouth, contaminated sediments from the Lot River and SPM entering the estuary showed parallel historical evolutions (1979-2010) for Zn/Cd ratios but not for δ66Zn values. Oysters had signatures varying from δ66Zn = 1.43 ‰ in 1983 to 1.18 ‰ in 2010 and were offset by δ66Zn = 0.6 - 0.7 ‰ compared to past (1988) and present SPM from the salinity gradient. Isotopic signatures in river-borne particles entering the Gironde Estuary under contrasting freshwater discharge regimes during 2003-2011 showed similar values (δ66Zn ≈ 0.35 ± 0.03 ‰; 1SD, n=15), i.e. they were neither related to former metal refining activities at least for the past decade nor clearly affected by other anthropogenic sources. Therefore, the Zn isotopic signatures in Gironde oysters reflect the geochemical reactivity of Zn in the estuary rather than signatures of past metallurgical contaminations in the watershed as recorded in contaminated river sediments. The study also shows that the isotopic composition of Zn is strongly fractionated by its geochemical reactivity in the Gironde Estuary, representative of meso-macrotidal estuarine systems.

Influence of the sea state on Mediterranean heavy precipitation: a case-study from HyMeX SOP1

Sea state can influence the turbulent air–sea exchanges, especially the momentum flux, by modifying the sea-surface roughness. The high-resolution non-hydrostatic convection-permitting model MESO-NH is used here to investigate the impact of a more realistic representation of the waves on heavy precipitation during the Intense Observation Period (IOP) 16a of the first HyMeX Special Observation Period (SOP1). Several quasi-stationary mesoscale convective systems developed over the western Mediterranean region, two of them over the sea, and resulted in heavy precipitation on the French and Italian coasts on 26 October 2012. Three different bulk parametrizations are tested in this study: a reference case (NOWAV) without any wave effect, a parametrization taking into account theoretical wave effects (WAV) and a last one with realistic wave characteristics from the MFWAM analyses (WAM). Using a realistic wave representation in WAM significantly increases the roughness length and the friction velocity with respect to NOWAV and WAV. The three MESO-NH sensitivity experiments of the IOP16a show that this surface-roughness increase in WAM generates higher momentum fluxes and directly impacts the low-level dynamics of the atmosphere, with a slowdown of the 10 m wind, when and where the wind speed exceeds 10 m s−1 and the sea state differs from the idealized one. The turbulent heat fluxes are not significantly influenced by the waves, these fluxes being controlled by the moisture content rather than by the wind speed in the simulations. Although the convective activity is globally well reproduced by all the simulations, the difference in the low-level dynamics of the atmosphere influences the localization of the simulated heavy precipitation. Objective evaluation of the daily rainfall amount and of the 10 m wind speed against the observations confirms the positive impact of the realistic wave representation on this simulation of heavy precipitation.