Bivalves for salinity reconstruction: the case of Mytillus Edulis (the blue mussel)

Testing climate models for future and past climate change critically depend on our ability to quantitatively reconstruct past climate. This requires the proper deconvolution of temperature and salinity effects on proxy incorporation. Paleosalinity is the single most important oceanographic parameter which currently can still not be quantified from sedimentary records. During the PaleoSalt project focus was on the development of proxy-tools allowing past salinity reconstructions from carbonate archives. Our research team selected a marine bivalve which is tolerant to a wide range of salinities and is available also from archaeological collections. The fast growing and relatively short-lived blue mussel also offers the potential to investigate seasonal variability and amplitudes.
Our objectives were to refine existing proxy tools such as d18Oc (calibration and signal treatment) and to develop new proxies (dD, isotopes of Mg and trace elements; i.e. a multi-proxy approach) as well as mathematical tools to improve the robustness and sensitivity of reconstructions of environmental conditions. To that purpose we have used two parallel approaches: an in situ study along the salinity gradient of the Scheldt estuary and an ex situ growth experiment under controlled laboratory conditions. Though the d18O results for shells grown under controlled conditions of temperature, salinity and food supply validate the existing equations for temperature reconstruction based on d18O water (e.g. Wanamaker et al., 2006), the carbonate system of the growth medium stands up as an important additional control factor that needs to be taken into account. Other methodologies are being developed for the identification of lipid compounds and their dD signature in shell organic matter and for the assessment of dMg in shell and tissue material, as potential salinity proxies. We recently acquired a new Fast Excimer (193 nm) Laser - ICP-MS system to investigate trace element records in the mussel shells from the controlled growth experiments and Schelde sites and first results will possibly be schown. We also developed mathematical tools to correct for bias in proxy records due to variable growth rate induced time base distortion and to correct the averaging effect that can be induced when sampling the shell. Finally a multi-proxy model was developed for temperature and salinity reconstruction .