Climate Modelling Group
School of Earth and Ocean Sciences

Climate Research Network

Collaborative Research Agreement at the University of Victoria on Behalf of the Canadian Institute for Climate Studies and Environment Canada
(#7 CICS-Variability)

Progress Report:

October 1, 1996

1. Principal Investigator

Andrew Weaver

2. Institution

School of Earth & Ocean Sciences
University of Victoria
PO Box 1700
Victoria, BC, V8W 2Y2

3. Progress

The #7 CICS -- Variability Grant is used to undertake research on climate variability on the seasonal-centennial timescale. Research grant funding is to provide full support for one PhD student (E. Wiebe), partial support for two research associates (S. Zhang, S. Valcke), partial support for a PhD student (A. Fanning) and minor operating expenses.

A copy of my progress reports and the original grant proposal are available on the world wide web at:

This progress report highlights the work conducted during the 1996 fiscal year. Some reference to earlier funded CICS Variability research is also made.

3.1 Decadal Variability in a coupled Energy-Moisture Balance Model (EMBM) -Ocean General Circulation Model (OGCM) - Thermodynamic Ice Model (TIM)

Ed Wiebe, an MSc. student, is currently utilizing a version of our coupled EMBM-TIM-OGCM. The model employs the same horizontal resolution, geometry, and forcing as that described by Fanning and Weaver (1996) although it is now coupled to the GFDL MOM2 model (Pacanowski, 1995). The only appreciable difference between the models is the implementation of the flux corrected tracer algorithm (Gerdes et al., 1991; Weaver and Eby, 1996) and the explicit convection scheme of Rahmstorf (see Pacanowski, 1995). Of particular importance is the generation of spontaneous decadal-scale variability (period 26 years) centered in the North Atlantic. Oscillations in the meridional overturning streamfunction span about 10 Sv in magnitude with accompanying temperature anomalies of almost 5oC.

We are currently analysing the mechanism for the decadal oscillation, and continuing the model integration time to ascertain whether centennial scale variability (centred in the Southern Ocean) is a robust feature of the coupled climate system, or merely a transient phenomena.

In addition to this analysis Ed Wiebe has also made extensive modifications to a collection of existing IDL routines used to visualize the output from this coupled model. Extensions include the capability to calculate mean fields and plot anomalies. In addition, movies of time-dependent phenomena can be created and saved in a compact form for later viewing. This software is extensible and adaptable and may be of use to other researchers in the field of ocean modelling.

3.2 Decadal Variability in the GFDL Coupled Model

The numerical simulations of Delworth et al. (1993), using the fully coupled ocean-atmosphere model developed by the Geophysical Fluid Dynamics Laboratory (GFDL) in Princeton, NJ, showed interdecadal variability of the thermohaline circulation in the North Atlantic. However, it is still unclear if this variability is a coupled ocean-atmosphere or an ocean-only phenomenon.

In order to clarify this problem, the GFDL coupled model, previously run on a Cray at GFDL, has been adapted to our IBM machines. The oceanic part of this model has been spun-up by Dr. S. Valcke (an NSERC postdoctoral fellow) to equilibrium in the same configuration as in the run of Delworth et al. (1993). This ocean model is now being run under fixed-flux boundary conditions, the fluxes being the sum of the atmospheric fluxes (diagnosed from the atmospheric model alone at equilibrium) and the flux adjustment terms (artificial terms used in coupled models to correct the fluxes going into the ocean in order to remove systematic climate drifts). If the same decadal variability as in the fully coupled experiments is observed, or if it appears when a stochastic forcing is added to the fixed-flux boundary conditions, we will conclude that the variability is due to the internal ocean dynamics.

Dr. S. Zhang is also working with the GFDL model. He has spent a good deal of time trying to make the atmospheric component of the model run more efficiently on our local workstation cluster. Dr. Zhang is attempting to obtain a version of the GFDL coupled model which does not require flux adjustments. He has identified a number of problems and is currently seeking methods to overcome them. These problems include: a) an ocean model surface flux which is significantly weaker than that produced by the atmospheric model. This is responsible for a large part of the flux adjustment; b) a very strong local salinity adjustment which is related to the melting of ice; c) unphysically strong restoring in the ocean with same strength in temperature and salinity during the oceanic spin up. If all of the problems are solved, then the oceanic component of the coupled model will probably not be the source of climate drift.

In the atmospheric component of the coupled model Dr. Zhang has reduced the frequency of synoptic eddies while retaining all the model physics. He implemented a number of other acceleration techniques in an attempt to speed the atmospheric model up. Specifically, this is achieved by using the original time step in the dynamical code and using a much longer time step for all other processes. This effectively assumes that the response time of the atmosphere is much shorter than the ocean, and that only the mean of synoptic system is important for the energy balance and that its variance only generates noise, at least on timescales longer than a decade.

3.3 Decadal variability in OGCMs with various subgrid-scale boundary layer parameterizations

Thierry Huck, a PhD student, is using locally-developed planetary geostrophic ocean models under various sub-grid-scale boundary layer parameterizations to study the mechanism of decadal variability found in ocean only models (Greatbatch and Zhang 1995). Under flux boundary condition on the surface density, a parameter sensitivity analysis has been carried out. The horizontal tracer diffusivity has a critical damping effect, while the vertical diffusivity (which determines the strength of the thermohaline circulation) enhances the oscillatory behaviour. The inclusion of a parameterization of convection (excluding the effect of vertical velocities) and the beta-effect are found not to be necessary in sustaining the variability, and so exclude the role of Rossby waves in the mechanism (Winton 1996). The influence of the lateral boundaries along which convection takes place (weakening the stratification so that Kelvin waves may propagate with decadal time scales -- Greatbatch and Peterson, 1996) has been rejected in two ways: 1) By moving northward the polar boundary (by several tens of degrees), with no atmospheric forcing in the extended area (a buffer zone where the stratification remains strong). In this case the oscillatory behaviour is weakly modified. 2) With a symmetric forcing in an f-plane model (that is twice as wide as in the control experiments), so that a 'tropical' thermocline is present along both zonal boundaries of the basin. In this case the oscillation is more profoundly affected but even stronger. None of these major changes weaken the variability, which is maximum in the region of Gulf Stream separation and its eastward extension.

Comparisons between the primitive equation GFDL-MOM code and the planetary geostrophic models at various horizontal resolution suggest that the higher the resolution, the more likely the occurrence of decadal variability (even without changing the horizontal diffusivity along with the resolution). In addition, below 100 km resolution, the non-linear terms and the time-derivative in the horizontal momentum equations play a driving role in producing the variability. In the vertical, a discretization as crude as two levels can sustain the oscillation.

Since the oscillations have a strong signature in the zonally-averaged fields, a two-dimensional mechanism is being investigated. This would be consistent with the findings of decadal variability within the idealized coupled ocean-atmosphere of Sarvanan and McWilliams (1995). With non-steady surface forcing in a zonally-averaged ocean model, we expect the ocean to generate decadal time-scales associated with the overturning period.

We are presently writing up this work in Huck et al. (1996)

3.4 Climate Variability as a function of mean climatic state

Over the next two years I hope to continue improving our understanding of the mechanisms of decadal-interdecadal climate variability through the development of increasingly more sophisticated coupled models. The coupled EMBM-TIM-OGCM represents the simplest form of our coupled modelling studies. We shall continue to use it to explore simple thermodynamic feedbacks and gain insight into what results we might expect and which experiments we should undertake with the more complicated GFDL and CCC coupled models. In addition, the GFDL and EMBM-OGCM-TIM coupled model will be used to investigate questions concerning the existence of variability in the coupled climate system and how it varies as the mean climatic state changes (i.e, does the decadal-interdecadal climate variability found in the coupled model change as CO2 is increased in the atmosphere. Since the GFDL coupled model is far more computationally efficient than the CCC coupled model, it is hoped that the insight we gain from it will allow us to better streamline the future experiments that will be performed with the more sophisticated CCC coupled model.

3.5 References

Delworth, T., S. Manabe, and R.J. Stouffer, 1993: Interdecadal variations of the thermohaline circulation in a coupled ocean-atmosphere model. J. Climate, 6, 1993-2011.

Fanning, A.F. and A.J. Weaver, 1996: An atmospheric energy moisture-balance model: climatology, interpentadal climate change and coupling to an OGCM. J. Geophys. Res., 101, 15111-15128.

Gerdes, R., C. Koberle, J. Willebrand, 1991: The influence of numerical advection schemes on the results of ocean general circulation models, Clim. Dynamics, 5, 211-226

Greatbatch, R. J., and S. Zhang, 1995: An interdecadal oscillation in an idealized ocean basin forced by constant heat flux. J. Climate, 8, 81-91.

Greatbatch, R.J. and K.A. Peterson, 1996: Interdecadal variability and oceanic thermohaline adjustment. J. Phys. Oceanogr., in press.

Huck, T., A. J. Weaver, and A. Colin de Verdiere, 1996: Decadal variability in simplified models of the thermohaline circulation. in preparation

Pacanowski, R.C., 1995: MOM 2 Documentation User's Guide and Reference Manual, Version 1.0, GFDL Technical Report #3, 232 pps.

Saravanan, R., and J. C. McWilliams, 1995: Multiple equilibria, natural variability, and climate transitions in an idealized ocean-atmosphere model. J. Climate, 8, 2296-2323.

Weaver, A.J. and M. Eby, 1996: On the numerical implementation of advection schemes for use in conjunction with various mixing schemes in the GFDL ocean model. J. Phys. Oceanogr., in press.

Winton, M., 1996: The role of horizontal boundaries in parameter sensitivity and decadal-scale variability of coarse-resolution ocean general circulation models. J. Phys. Oceanogr., 26, 2.
Return to the Climate Modelling Group WWW page
This page is maintained by
Last updated: