GENIE:Biogeochem:FAQs:1

Do biogeochem tracer concentrations vary with salinity, and what happens when freshwater is added to the surface ocean?
A complication in the implementation of biogeochemical cycling arises because of the rigid lid surface boundary condition employed in many ocean circulation models such as GOLDSTEIN. Net evaporation or precipitation at the ocean surface is implemented as a virtual salinity flux rather than an actual loss or gain of freshwater, because grid cell volumes are not allowed to change. This decouples the concentration of the biogeochemical tracers (DIC, ALK, PO4, etc) from salinity, whereas in the real world they should respond in a similar way to net evaporation/precipitation. A common solution is to apply virtual fluxes of the biogeochemical tracers along with salinity to the ocean surface (e.g., see the OCMIP-2 protocol [Najjar and Orr, 1999]). We take an alternative approach here in which ocean circulation transforms a biogeochemical tracer concentration field that has been salinity-normalized [Marchal et al., 1998; Müller et al., in press]. The salinity-normalized field is converted back to real tracer concentrations before the effects of biological uptake and remineralization, and air-sea gas exchange are calculated.

For biogeochemical tracer fields that are decoupled from salinity (i.e., where biological uptake at the surface and reminerilization reorganized the global distribution compared to that salinity) it turns out that the process of normalizing and then de-normalizing (with the circulation step in beteen) is not conservative overall. Biogeochemical tracer concentrations are therefore re-scaled at each time-step to ensure that there is no spurious gain or loss from the ocean as a whole.

With an addition of freshwater to the ocean surface such as through a hosing experiment, the salinity-normalized ocean transport will drive biogeochemical tracer concentrations lower until they are consistent with the applied change in salinity. You dont get an instant adjustment of biogeochemical tracer concentrations by this process, but they will adjust appropriately to the salinity change. Perhaps on a time-scale of a few years(?) If someone would like to analyse the system perhaps they might like to add the facts here!

In extreme glacial situations such as the so-called snowball Earth, you could easily end up with >100m of sea-ice thickness everywhere and a considerably more saline ocean. The biogeochemical tracer concentrations are re-scaled according to the global salinity change to account for the fact that the ocean volume should have reduced, although the rigid lid in GOLDSTEIN prevents the effect of this being represented explicitly.