GENIE:Biogeochem:HOWTOs:1

HOW TO ... manipulate ocean circulation by salinity forcing (e.g. collapse Atlantic meridional overturning)
Basically, you can prescribe a salinity flux to the ocean surface. Make this flux negative (removal of salinity == addition of freshwater) and hey-presto - you have a freshwater flux.

You will need to;


 * Take off the built-in 'saftey catch' that prevents you modifying climate and ocean circulation. For instance, you might just want to modify salinity and/or temperature to changes surface ocean CO2 solubility without changing ocean circulation. You can find this flag in the configuration file biogem_config.par. Set the flag; 'enable temperature and/or salinity forcing of climate' to 't'.


 * Set a salinity (tracer name: 'sal') flux forcing ('t') in gem_config_ocn.par (COLUMN #10). The flux forcing should not be scaled - i.e., set the flag in COLUMN #11 to 'f'. Ensure that you do not also have a restoring forcing simultaneously set :o)


 * Include the necessary flux forcing files in the biogem configuration directory. An example set of files suitable for the 36x36x8 configuration of goldstein that will completely shut down the Atlantic meridional overturning circulation (MOC) can be found here. The *_sig.dat file has been set up to stop the MOC at year 1000, and then allow circulation to recover from year 2000 onwards). In the *_II.dat file, there are some minus one values around the N Atlantic in the surface layer - this is where the fresh water is applied (remember; negative salinity flux == freshwater addition). All other wet points in the grid (i.e., ocean) have a value of +0.0119, which is chosen to give a net global surface ocean salinity flux of zero - there are eleven cells assigned minus one, so the sum of all the other wet cells must make a total of 11.0, which is where the unlikely looking number 0.0119 comes from. In effect, this is redistributing freshwater rather than adding new freshwater and making the entire ocean progressively less salty. If you want to modify the number of cells in which the flux is applied and keep the zero net ocean change balance, then all the 0.0119 numbers must be changed accordingly. Note that all wet grid points in the *_I.dat are zero, because the applied 3-D forcing = I + (II - I) x sig.

By modifying the *.sig file you can scale the magnitude of the freshwater flux perturbation and as well as having complete control over the shape of the flux function (i.e., you can create ramps, sharp transitions, and a on-off-on-off-on-of behavior like some sort of Dansgaard-Oeschger Yo-Yo).

Note that the example files are provided as a (rather crude) example only, and do not necessarily bear any resemblence with paleoclimate reality. In particular, the choice of North Atlantic cells could be made more appropriately. As could what the total freshwater flux actually is (see FAQ for converting units of freshwater forcing from psu*mass per year to Sverdrups).

What happens to the concentration of biogeochemical tracers (DIC, ALK, PO4, etc) in the surface ocean when freshwater is added is described in a FAQ.