The OCCAM papers page
This directory contains details of papers and OCCAM documentation. Additional
information is also available in the form of posters. Portable Document
Format (PDF) versions of many of the posters presented during the project
are available from our posters page.
Humphries, U.W. and Webb, D.J.,2008:
On the Indonesian Throughflow in the OCCAM 1/4 degree ocean model.
Ocean Science, 4, 183-198.
Webb, D.J. and de Cuevas, B. A., 2007:
On the fast response of the Southern Ocean to changes in the zonal wind,
Ocean Science, 3, 417-427.
Jia, Y., Coward, A.C., de Cuevas, B.A., Webb, D.J. and Drijfhout, S.S., 2007:
A model analysis of the behavior of the Mediterranean Water in the North Atlantic.
Journal of Physical Oceanography, 37, (3), 764-786. (doi: 10.1175/JPO3020.1)
Levine, R.C. and Webb, D.J., 2007:
On available energy in the ocean and its application to the Barents Sea,
Ocean Science Discussions, 4, 897-931.
Webb, D.J., and de Cuevas, B.A. 2003:
The region of large sea surface height variability in the southeast Pacific Ocean.
- Journal of Physical Oceanography, 33(5), 1044-1056.
Webb, D.J., and de Cuevas, B.A. 2002
An ocean resonance in the Indian sector of the Southern Ocean.
- Geophysical Research Letters, 29(14), 9.1-9.4, doi:10.1029/2002GL015270.
- Lee, Mei-Man, A.C. Coward and A.J. George Nurser, 2002:
- Spurious diapycnal mixing of the deep waters in an
eddy resolving global ocean model.
- Journal of Physical Oceanography, 32(5), 1522-1535.
PDF version or just the abstract
- Webb, D.J., 2000:
- Evidence of shallow zonal jets in the South Equatorial Current region of the Southwest Pacific.
- Journal of Physical Oceanography, 30 (4), 706-720. 2000
- Fox, A.D., K. Haines and B.A. de Cuevas 1999:
- Satellite altimetry data assimilation in the OCCAM global ocean model.
- Phys. Chem. Earth (A), Vol. 24, No. 4, 375-380.
- D. J. Webb et al (1998): The first main run of the OCCAM Global Ocean Model
- Southampton Oceanography Centre, Internal Document No. 34.
This report contains background information on the OCCAM global ocean model
and the first main run of the model. It includes a discussion of the underlying
model equations and information on the model grid, the representation of
advection and diffusion, the timestepping scheme and the free surface scheme
used by the model.
The report also contains information on how the model was initialised for
the first main run, how it was forced, the changes that were introduced
during the model run and the format of the archived data sets. The report
is available as a
Portable Document Format or
- P. Saunders et al (1999):
Circulation of the Pacific Ocean seen in a global ocean model (OCCAM)
- J. Geophys. Res., 104, C8, 18281-18299. (1999).
This paper compares observations made in the Pacific Ocean and
the 8-12 year climatology of the OCCAM global model.
Meridional mass and heat transports and their divergences are
calculated, and zonal averages compared with the air-sea
exchange values of Doney et al.  with mixed success.
The large scale distribution of sea surface height and
near-surface currents are shown to be well reproduced by the
model, but their variance is underestimated, severely so in
areas of weak variability. Synoptic wind forcing enhances the
latter, though only slightly, while higher model resolution
improves agreement in areas of strong variability. Model mean
transports in the East Australian Current, in the Vitiaz
Strait and in the Kuroshio exceed measurements by 10, 0, 18
Sv, respectively, and variability is underestimated by factors
between 2 and 4. Near 1000 m and in strong currents the flow
is overestimated, but over most of the ocean, model output
does not match either the magnitude or the spatial variability
of the observations. In contrast, the inflow of bottom water
and its spreading path throughout both South and North Pacific
is well described. Good agreement is found with measured
transport through the Samoan Passage, but at 32�S, east of the
Tonga-Kermadec Ridge, the model yields only 60% of the observed
value. The Sverdrup balance is shown to hold widely in
the model interior. Utilizing the complete depth-integrated
vorticity equation it is found that bottom pressure torques
balance the advection of planetary vorticity in both surface
and bottom intensified western boundary currents.
The report is available as a
Portable Document Format or
gzipped PostScript file.
de Cuevas, B.A., Webb, D.J., Coward,A.C., Richmond, C.S. & E. Rourke, 1999.
- Lee, Mei-Man, A.C. Coward and A.J. George Nurser, 2001:
Spurious diapycnal mixing of the deep waters in an
eddy resolving global ocean model.
Recent idealized studies have shown that the advection scheme in high
resolution z coordinate models may drive unrealistically high rates of
diapycnal mixing. Our aim here is to see whether the diapycnal mixing
associated with the advection scheme in a global eddy resolving (1/4°
by 1/4° ) z level model is sufficiently strong to corrupt the thermohaline
circulation. We diagnose the diapycnal fluxes by using the ideas of water mass
transformation. In the Southern Ocean, the model deep and bottom waters drift
rapidly away from the Levitus climatology, with dense isopycnals moving
downwards at rates of up to 35 m/year ; . The strong upward flux (up to 50
Sv) through the dense isopycnals cannot be explained by the incorrect surface
forcing (as a result of poor surface fluxes and no ice model) as most of the
anomalous diapycnal fluxes are occurring in the deep ocean far from surface
forcing. Hence, the excessive diapycnal flux is driven by diffusion in the
model, both explicit and implicit. The `effective' diapycnic diffusivity driven
by the numerical diffusion (associated with the horizontal advection scheme) is
found to be the same order, 1-10 cm²/s , as that driven by the
explicit horizontal diffusion. For strong vertical velocities (about 20 m/day
) as in models forced by high frequency winds, the vertical advection
scheme also gives similar effective diffusivities. These effective
diffusivities are considerably greater than suggested by observations. To
alleviate these problems, we suggest that eddy resolving z level climate models
will require (1) less diffusive horizontal advection schemes and (2) better
vertical resolution throughout much of the water column.
The UK Ocean Circulation and Advanced Modelling Project (OCCAM)
pp 325-335 in: High Performance Computing, Proceedings of HPCI Conference 1998, Manchester 12-14 January 1998, R.J. Allan, M.F. Guest, A.D. Simpson, D.S.
Henty, D.A. Nicole (Eds) Plenum Press.
Fox, A.D., Haines, K., de Cuevas, B.A. & Coward, A.C., 1999.
Altimeter data assimilation in the OCCAM global model,
pp 365-369 in: Proceedings of HPCI Conference1998, Manchester 12-14 January 1998, R.J.
Allan, M.F. Guest, A.D. Simpson, D.S. Henty, D.A. Nicole (Eds). Plenum
- Gwilliam, C.S. (1995): The OCCAM Global Ocean Model
- In: Coming of Age (The Proceedings of the Sixth ECMWF Workshop on the use
of Parallel Processors in Meteorology), World Scientific, pp: 446-454
Over the past two decades, numerical models have been widely used to investigate
the circulation, chemistry and biology of the world's oceans, as well as
their effect on climate. However, accurate models need very powerful computers
and, for this reason, it is expected that a lot more use will be made of
array processor computing. In this paper, work being undertaken by the Ocean
Circulation and Climate Advanced Modelling (OCCAM) Project is presented.
The conversion from the sequential GFDL Modular Ocean Model (MOM) to an
efficient array processor, global ocean code is described.
- Gwilliam, C.S, A. C. Coward, B. A. de Cuevas, D. J. Webb, E. Rourke, S. R. Thompson, K. Döös: The OCCAM Global Ocean Model
- A paper presented at the 2nd UNAM-Cray Supercomputing Conference on Numerical
Simulations in the Envionmental and Earth Sciences (June 1995). Accepted
for publication in the refereed proceedings.
Climate change is not affected by the atmosphere alone - the world's oceans
also play an important role, hence the need for global ocean models. In
the past ocean modelling has been restricted to coarse resolutions and to
limited areas, for example the Antarctic. The arrival of powerful array
processors is allowing global studies at resolutions high enough to resolve
eddies in the ocean. This paper describes one such project, OCCAM, and some
of the scientific results and problems from an initial three year run.
- D. J. Webb et al.(1992):
of state algorithms used by the FRAM model
- Institute of Oceanographic Sciences, Internal Document No. 313.
This report contains background information on the equation of state
used by both the FRAM and
The Eckart equation of state that had been used previously was found to produce
significant errors at high latitudes. The later FRAM runs use
the full EOS80 internation equation of state for sea water. The
OCCAM model uses a polynomial approximation to EOS80 similar to that
distributed with the MOM ocean model code. The report is available as a
Document Format or
- A.C. Coward (1993): The equation
of state algorithms used by the OCCAM model
- Institute of Oceanographic Sciences, Internal Document No. 323.
This report contains further details on the equation of state
used by the OCCAM model and the method used to calculate the
polynomial coefficients. The report is available as a
Document Format or
- Gwilliam, C.S.(1995): Modelling the Global Ocean Circulation on the T3D
- In Parallel Computational Fluid Dynamics: Implementations and Results using
Parallel Computers, Elsevier, pp:33-40.
The Ocean Circulation and Climate Advanced Modelling Project is developing
a global ocean model suitable for climate studies, especially the investigation
of global ocean characteristics. The model will also act as a test bed for
improved schemes which model ocean physics. The large memory and time requirements
for such a study has led to the use of array processors. In this paper an
outline of the model and the parallel code is given, along with some initial
- Richmond, C.S.(1996): Changes to the OCCAM code to allow for different sizes of difference stencils.
- An unpublished document detailing changes made to the OCCAM code to allow
different advection schemes, with different sizes of difference stencils
to be used.
The original OCCAM code uses a centred difference advection scheme in the
tracer and momentum equations. This is known to cause under and overshooting
of the tracer values where large tracer gradients and/or large velocities
occur. Improvements to this scheme include QUICK and variations on it. These
schemes are variants of an upwind method and the stencil is five points
wide in each of the advected planes. In a parallel context, this means that
each processor requires information from 2 rows of another processor i.e.
a processor needs 2 halo rows and not the 1 required by the centred difference
scheme. To allow for different schemes to be used, and the possibility of
higher order schemes as well, the OCCAM code has been rewritten to allow
for any number of halo rows. The details of these changes are documented
- Richmond, C.S.(1996): Optimisation of the OCCAM code.
- An unpublished document detailing optimisation changes made to the OCCAM
The OCCAM code has been developed for array-based, parallel processing
platforms. The code is run in production mode on 256 processors of the 512
processor Cray T3D at Edinburgh. During the day tests are also carried out
on 128 processors. For efficient running of the model on the T3D it is necessary
to look at how the processors are used. This includes investigating the
use of the small, 1MW cache, and the message passing system. This report
details changes to the code that have improved its performance. The base
code is the OCCAM code with `split Quick' in the tracer and momentum equations.
- Richmond, C.S.(1996): Grid Partitioning Code.
- This tar file contains codes and a README postscript file for interactively
developing partitions for the OCCAM grids.
- Richmond, C.S.(1997): Changes to the OCCAM code for the MPI implementation.
- An unpublished document detailing the alterations made to the OCCAM code
to allow the Message Passing Interface (MPI) to be used. OCCAM now permits
PVM, SHMEM and MPI.
- Webb, D.J.(1978): Non-linear transfers between sea waves.
- Hasslemann's equation describing the non-linear interactions between sea waves is studied numberically for a Pierson-Moskowitz spectrum. The apparent creation of order, due to the non-linear enhancement of the peak of the spectrum occurs as a by-produce of a large amount of disorder created at high wavenumbers.
It is also found that the physical process dominating the non-linear transfer is the scattering of waves near the peak of the spectrum by short waves in the tail of the spectrum. In this process the long waves become longer and the short waves shorter.
File last amended: March 2009