This model overlays the dissolved inorganic carbon biogeochemistry model
(pkg/dic) over a 2.8o global physical model. The
physical model has 15 levels, and is forced with a climatological annual
cycle of surface wind stresses (Trenberth et al. 1989 [TOL89],
surface heat and freshwater fluxes (Jiang et al. 1999 [JSMR99]) with
additional relaxation toward climatological sea surface temperature and
salinity (Levitus and Boyer (1994a,b) [LB94a, LB94b]). It uses the Gent and
McWilliams (1990) [GM90] eddy parameterization scheme,
has an implicit free-surface, implicit vertical diffusion and uses the
convective adjustment scheme.
The biogeochemical model considers the coupled cycles of carbon, oxygen,
phosphorus and alkalinity. A simplified parameterization of biological
production is used, limited by the availability of light and phosphate.
A fraction of this productivity enters the dissolved organic pool pool,
which has an e-folding timescale for remineralization of 6 months
(following Yamanaka and Tajika 1997 [YT97]). The remaining fraction of this
productivity is instantaneously exported as particulate to depth
(Yamanaka and Tajika 1997 [YT97]) where it is remineralized according to the
empirical power law relationship determined by Martin et al. (1987]) [MKKB87]. The
fate of carbon is linked to that of phosphorus by the Redfield ratio.
Carbonate chemistry is explicitly solved (see Follow et al. 2006)
[FID06]) and the air-sea exchange of
CO2 is parameterized with a uniform gas transfer coefficient
following Wanninkhof (1992) [Wan92]. Oxygen is also linked to
phosphorus by the Redfield ratio, and oxygen air-sea exchange also
follows Wanninkhof (1992) [Wan92]. For more details see
Dutkiewicz et al. (2005) [DSSaPS05].
The example setup described here shows the physical model after 5900
years of spin-up and the biogeochemistry after 2900 years of spin-up.
The biogeochemistry is at a pre-industrial steady-state (atmospheric
ppmv is kept at 278). Five tracers are resolved: dissolved inorganic
carbon (\(DIC\)), alkalinity (\(ALK\)), phosphate (\(PO4\)),
dissolved organic phosphorus (\(DOP\)) and dissolved oxygen
(\(O2\)).
Figure 4.46 Modeled annual mean air-sea CO2 flux (mol C m-2 y-1) for pre-industrial steady-state. Positive indicates flux of CO2 from ocean to the atmosphere (out-gassing), contour interval is 1 mol C m-2 y-1.¶
The physical ocean model velocity and diffusivities are used to
redistribute the 5 tracers within the ocean. Additional redistribution
comes from chemical and biological sources and sinks. For any tracer
\(A\):
where \(\vec{u^{*}}\) is the transformed Eulerian mean circulation
(which includes Eulerian and eddy-induced advection), \(\mathbf{K}\)
is the mixing tensor, and \(S_A\) are the sources and sinks due to
biological and chemical processes.
\(F_{CO_2}\) is the flux of CO2 from the ocean to the
atmosphere
\(V_{CO_2}\) is “virtual flux” due to changes in \(DIC\) due
to the surface freshwater fluxes
\(r_{C:P}\) is Redfield ratio of carbon to phosphorus
\(J_{Ca}\) includes carbon removed from surface due to calcium
carbonate formation and subsequent cumulation of the downward flux of
CaCO\(_3\)
\(V_{ALK}\) is “virtual flux” due to changes in alkalinity due to
the surface freshwater fluxes
\(r_{N:P}\) Redfield ratio is nitrogen to phosphorus
\(f_{DOP}\) is fraction of productivity that remains suspended in
the water column as dissolved organic phosphorus
\(J_{prod}\) is the net community productivity
\(\frac{\partial F_P}{\partial z}\) is the accumulation of
remineralized phosphorus with depth
\(\kappa_{remin}\) is rate with which \(DOP\) remineralizes
back to \(PO_4\)
\(F_{O_2}\) is air-sea flux of oxygen
\(r_{O:P}\) is Redfield ratio of oxygen to phosphorus
\(O_{2crit}\) is a critical level below which oxygen consumption
if halted
These terms (for the first four tracers) are described more in
Dutkiewicz et al. (2005) [DSSaPS05] and by
McKinley et al. (2004) [MFM04] for the terms relating to oxygen.
code/packages.conf: which dictates which packages will be compiled in
this version of the model - among the many that are used for the
physical part of the model, this also includes pkg/ptracers, pkg/gchem,
and pkg/dic which allow the biogeochemical part of this setup to
function.
input/data: specifies the main parameters for the experiment. Some
parameters that may be useful to know: nTimeSteps number timesteps
model will run, change to 720 to run for a year taveFreq frequency
with which time averages are done, change to 31104000 for annual
averages.
input/data.pkg: set true or false for various packages to be used
input/data.ptracers: details of the tracers to be used, including
makes, diffusivity information and (if needed) initial files. Of
particular importance is the PTRACERS_numInUse which states how
many tracers are used, and PTRACERS_Iter0 which states at which
timestep the biogeochemistry model tracers were initialized.
bathy.bin: bathymetry data file
input/eedata: This file uses standard default values and does not
contain customizations for this experiment.
fice.bin: ice data file, needed for the biogeochemistry
lev_monthly_salt.bin: SSS values which model relaxes toward
lev_monthly_temp.bin: SST values which model relaxes toward
pickup.0005184000.data: variable and tendency values need to
restart the physical part of the model
pickup_cd.0005184000.data: variable and tendency values need to
restart the cd pkg
pickup_ptracers.0005184000.data: variable and tendency values
need to restart the the biogeochemistry part of the model
shi_empmr_year.bin: freshwater forcing data file
shi_qnet.bin: heat flux forcing data file
sillev1.bin: silica data file, need for the biogeochemistry
tren_speed.bin: wind speed data file, needed for the
biogeochemistry
As the model is set up to run in the verification experiment, it only
runs for 4 timesteps (2 days) and outputs data at the end of this short
run. For a more informative run, you will need to run longer. As set up,
this model starts from a pre-spun up state and initializes physical
fields and the biogeochemical tracers from the pickup files.
Physical data (e.g., S,T, velocities etc) will be output as for any
regular ocean run. The biogeochemical output are:
tracer snapshots: look in input/data.ptracers to see which
number matches which type of tracer (e.g., ptracer01 is DIC).
tracer time averages
specific DIC diagnostics: these are averaged over taveFreq (set in
input/data) and are specific to pkg/dic (currently are only
available in binary format):
DIC_Biotave: 3-D biological community productivity (mol P
m-3 s-1)
DIC_Cartave: 3-D tendencies due to calcium carbonate cycle
(mol C m-3 s-1)
DIC_fluxCO2ave: 2-D air-sea flux of CO2 (mol C
m-2 s-1)
DIC_pCO2tave: 2-D partial pressure of CO2 in
surface layer
DIC_pHtave: 2-D pH in surface layer
DIC_SurOtave: 2-D tendency due to air-sea flux of
O2 (mol O m-3 s-1)
DIC_Surtave: 2-D surface tendency of DIC due to air-sea flux
and virtual flux (mol C m-3 s-1)
