Stable isotopes in surface waters of the Atlantic Ocean: Indicators of ocean-atmosphere water fluxes and oceanic mixing processes

Publisher's version (útgefin grein)The surface ocean hydrological cycle is explored based on ∼300 new δ18O and δD measurements from surface waters of the Atlantic Ocean and the Mediterranean Sea over the period 2010–2016. Our approach combines these surface observations with salinity (S) and stable isotope measurements of atmospheric water vapor. The distinct regional S‐δ distributions are used to identify different surface water masses and their horizontal advection. Moreover, based on assumptions on the δ‐S characteristics of seawater sources and the isotope composition of the evaporative (δe) and meteoric water (δMW) fluxes, the δ‐S distribution is used to indicate the relative importance of evaporation (E) and meteoric water inputs (MW). Here δe is estimated from the Craig and Gordon's equation using 120 days of measurements of the ambient air above the Atlantic Ocean collected during three cruises. To provide quantitative estimates of the E:MW ratio, we use the box model from Craig and Gordon (1965). This identifies the subtropical gyre as a region where E:MW ∼2 and the tropical ocean as a region were MW:E ∼2. Finally, we show that the δ18O‐δD distribution is better represented by a linear fit than the δ‐S relationship, even in basins governed by different hydrological processes. We interpret the δ18O‐δD distribution considering the kinetic fractionation processes associated with evaporation. In the tropical region where MW exceeds E, the δ18O‐δD distribution identifies the MW inputs from their kinetic signature, whereas in regions where E exceeds MW, the δ18O‐δD distribution traces the humidity at the sea surface.The Picarro equipment was purchased with support from different French institutions, in particular by IPSL, LOCEAN, LMD, and LATMOS. Work during Strasse on R/V Thalassa and Pirata FR24 on R/V Suroit was supported by three LEFE/IMAGO INSU grants (Strasse, Strasse/SPURS, and PIRATA), with additional support for equipment from IPSL and from OSU Ecce Terra. The authors gratefully acknowledge the association “Les Amis du Jeudi et du Dimanche” for the measurements aboard the RARA AVIS. The authors thank the TOSCA‐SMOS program whose founding was used for installing the SBE45 onboard the RaRa Avis Vessel. The Suratlant Project, as well as the data collection from the Toucan, Colibri and Cap San Lorenzo, are supported by SO SSS in France. Data collection onboard Ovide‐2010 (doi:10.3334/CDIAC/OTG.CLIVAR_OVIDE_2010) and Ovide‐2012 (doi:10.3334/CDIAC/OTG.CLIVAR_OVIDE_2012) cruises was supported by LEFE‐INSU. We also acknowledge Philipe Poupon and Cedric Courson for the sampling aboard the sailing boat Fleur Austral, Tara Expéditions for the sampling aboard the sailing boat Tara, CSIC for the sampling aboard the Sarmiento de Gamboa during the Midas Cruise and MOOSE program (ALLENVI‐INSU) for the sampling onboard R/V L'ATALANTE during the MOOSE‐GE2016 cruise (doi:10.17600/16000700). The authors thank the National Power Company of Iceland Landsvirkjun for their contribution to this research. The data are shared with the free Global Seawater Oxygen‐18 database [Schmidt et al., 1999].Peer reviewe