Search

The neodymium isotopic composition (εNd) of seawater is one of the most important geochemical tracers to investigate water mass provenance, which can also serve as a proxy to reconstruct past variations in ocean circulation. Nd isotopes have recently also been used to reconstruct past circulation changes in theMediterranean Sea on different time scales. However, the modern seawater ε Nd dataset for the Mediterranean Sea, which these reconstructions are based on, is limited and up to now only 160 isotopic measurements are available for the entire basin. The lack of present-day data also limits our understanding of the processes controlling the Nd cycle and Nd isotopic distribution in this semi-enclosed basin. Here we present new εNd data from 24 depth profiles covering all Mediterranean sub-basins, which significantly increases the available dataset in the Mediterranean Sea. The main goal of our study is to better characterize the relationship between the dissolved Nd isotope distributions and major water masses in the Mediterranean Sea and to investigate the impact and relative importance of local non-conservative modifications, which include input of riverine particles and waters, aeolian-derived material and exchange with the sediments at continental margins . This comprehensive εNd dataset reveals a clear εNd – salinity correlation and a zonal and depth gradient with εNd systematically increasing from the western to the eastern Mediterranean basin (average εNd = −8.8 ± 0.8 and −6.7 ± 1 for the entire water column, respectively), reflecting the large-scale basin circulation. We have evaluated the conservative εNd behaviour in theMediterranean Sea and quantified the non-conservative components of the ε Nd signatures by applying an Optimum Multiparameter (OMP) analysis and results from the Parametric Optimum Multiparameter (POMP) analysis of Jullion et al. (2017). The results of the present study combined with previously published Nd isotope values indicate that dissolved εNd behaves overall conservatively in the open Mediterranean Sea and show that its water masses are clearly distinguishable by their Nd isotope signature. However, misfits between measured and OMP- and POMP-derived εNd values exist in almost all sub-basins, especially in the eastern Levantine Basin and Alboran Sea at intermediate-deep depths, which can be explained by the influence of detrital lithogenic εNd signatures through interaction with highly radiogenic Nile sourced volcanic fractions and unradiogenic sediments, respectively. The radiogenic signature acquired in the eastern Levantine Basin is carried by the Levantine Intermediate Water and transferred conservatively to the entire Mediterranean at intermediate depths. Our measured εNd values and OMP- and POMP-derived results indicate that non-conservative contributions originating from sediment sources are then propagated by water mass circulation (with distinct preformed εNd) along the Mediterranean Sea through advection and conservative mixing. Mediterranean εNd effectively traces the mixing between the different water masses in this semi-enclosed basin and is a suitable water mass tracer.

International audience; The aim of this study was to develop a conceptual model of saline intrusion within coastal karst aquifers by analyzing Rainfall–Discharge–Salinity data and to assess freshwater resources using a lumped modeling approach. In a first step, we analyzed 4 years of data (rainfall, discharge and salinity times series) ofthe Port-Miou brackish submarine spring in South France (400 km2). A conceptual model of the aquifer was then designed to differentiate a deep brackish reservoir and a shallower fresh one. Salinity variations at the spring are assumed to be controlled mainly by dilution originating from the fresh water in the shallower reservoir. In a second step, a lumped modeling approach was developed based on the conceptual model to simulate discharge as well as salinity over time. We proposed a reservoir-model to take into account slow and fast components in the shallower part of the aquifer and a saline intrusion in the deeper one. This Rainfall–Discharge–Salinity model was calibrated and validated for two periods of 1.5 years at a daily time step and was also tested to reproduce a multi-annual evolution of the available discharge and salinity time series. Good simulation results were obtained to reproduce water and mass budgets as well as discharge and salinity dynamics during several hydrological cycles. The simultaneous modeling of hydrodynamics and quality data showed the robustness of the model in addition to its easy implementation. Our results led us to propose a new type of seawater mixing mechanism for brackish springs: the dilution type, in addition to the well-known Ventury suction and Head balance types. The application of the lumped model on the Port-Miou brackish spring validated the hydrogeological processes deduced from experimental data, given an initial quantification of the freshwater resources available in such complex brackish karstic aquifers.