Tropical Pacific SST Forecasts Utilizing Multiple Coupled
GCMs
Contributed
by David G. DeWitt1, Edwin K. Schneider2,3 and Zeng-Zhen
Hu3
1International Research
Institute for Climate Prediction, Columbia University
2George Mason University 3Center for Ocean-Land-Atmosphere
Studies
We describe forecasts for
tropical Pacific sea surface temperature (SST) made by combining the forecasts
produced by multiple independent coupled atmosphere ocean general circulation
models (CGCMs). In this initial
forecast, we are using two CGCMs; however, in the future we hope to add
additional models. The two CGCMs
utilize the same ocean component model and are initialized using the same ocean
data assimilation product. This work is an extension of the combined CGCM
forecasts described in Schneider et al. (2003). Differences between that work and the current forecasts include
use of an extended period for developing prior statistics, use of fewer models,
and employment of a different technique for combining the SST forecasts from
the different models.
Atmospheric Models
The first coupled model is
run at the International Research Institute for Climate Prediction (IRI) of
Columbia University and employs the ECHAM4.5 atmospheric general circulation
model (AGCM) from the Max Planck Institute for Meteorology. This is a spectral model with triangular
truncation at wavenumber 42 (T42). The
model is discretized in the vertical on 19 unevenly spaced hybrid
sigma-pressure layers. Longwave radiative transfer is modeled following Morcrette
et al. (1986), while shortwave radiation uses the scheme of Fouquart and Bonnel
(1980). Cloud water is a prognostic
quantity, and cloud properties are specified as Rockel et al. (1991) and
Roeckner (1995). Cumulus convection is parameterized using the mass flux scheme
of Tiedtke (1989) as modified by Nordeng (1994). The turbulent surface fluxes are calculated from Monin-Obukhov
similarity theory (Louis, 1979) and a higher-order closure scheme (Brinkop and Roeckner,
1995) is used to compute the vertical diffusion of momentum, heat, moisture,
and cloud water. The drag associated
with orographic gravity waves is simulated following Miller et al. (1989). A complete description of the model can be
found in Roeckner et al. (1996).
The
second coupled model is run at the Center for Ocean-Land-Atmosphere (COLA)
studies using version 2 of the COLA AGCM.
The model is discretized in the vertical on 18 unevenly spaced sigma
layers. The dynamic core used is from
the National Center for Atmospheric Research (NCAR) Community Climate Model
version 3 (CCM3) which uses a spherical harmonic expansion in the horizontal
which is truncated at wavenumber 42.
The parameterization of shortwave radiation is the Lacis and Hansen
(1974) scheme updated by Davies (1982), and the parameterization of longwave
radiation follows Harshvardhan et al. (1987).
The deep convection is an implementation of the Relaxed Arakawa-Schubert
scheme of Moorthi and Suarez (192) described by DeWitt (1996), and shallow
convection is the scheme of Tiedtke (1984).
The convective cloud fraction follows the scheme used in the CCM (Kiehl
et al., 1994; see DeWitt and Schneider, 1996 for additional details). There is a turbulent closure scheme for the
subgrid scale exchange of heat, momentum, and moisture (Mellor and Yamada,
1982; level 2.0). There is also a
parameterization of gravity wave drag (Kirtman et al., 1992). This model is described in detail in
Schneider (2002).
The
ocean component used in both of the coupled models is version 3 of the
Geophysical Fluid Dynamics Laboratory (GFDL) modular ocean model (MOM)
(Pacanowski and Griffes, 1998). This
model is a finite difference treatment of the primitive equations of motion
using the Boussinesq and hydrostatic approximations in spherical
coordinates. The domain is that of the
global ocean between 74°S and 65°N.
The coastline and bottom topography are realistic except that ocean
depths less than 100 meters are set to 100 meters and the maximum depth is set
to 6,000 meters. The artificial
high-latitude meridional boundaries are impermeable and insulating. The zonal resolution is 1.5°
everywhere. The meridional grid
spacing is 0.5° between 10°S and 10°N, gradually increasing to
1.5° at 30°S and 30°N and fixed at 1.5° in the extratropics. There are 25 layers in the vertical with 17
layers in the upper 450 meters. The
vertical mixing scheme is the non-local K-profile parameterization (KPP) scheme
of Large et al. (1994). The horizontal
mixing of tracers and momentum is Laplacian. The momentum mixing uses the space-time dependent scheme of
Smagorinsky and the tracer mixing uses Redi (1982) along with Gent and
McWilliams (1990) quasi-adiabatic stirring.
In the two respective coupled
systems, the AGCMs provide heat, momentum, fresh water, and surface solar flux
to the OGCM. The OGCM provides SST to
the AGCM. Information is exchanged
between the AGCM and OGCM once daily.
No empirical corrections were applied to either the fluxes or the SST,
i.e. the models are directly coupled.
The coupled system using the COLA AGCM uses the coupling software
provided with MOM3 for interpolation between the atmosphere and ocean model
grids. The coupled model employing the
ECHAM AGCM uses the Ocean Atmosphere Sea Ice Soil (OASIS) coupling software
(Terray et al., 1999) which is produced by the European Center for Research and
Advanced Training in Scientific Computation (CERFACS).
The ocean initial conditions are
taken from an ocean data assimilation system produced at GFDL using a
variational optimal interpolation (Derber and Rosati, 1989). The ODA for the period 1980 to 1999 was
produced at GFDL (A. Rosati and M. Harrison, personal communication), while the
period from January 2000 to present is
being
run at COLA (B. Kirtman and D. Min, personal communication). The ODA was run with a higher resolution version of the ocean model described
earlier but with identical physics and parameter settings. In these experiments, the ODA data is
interpolated to the lower resolution version of the ocean.
Atmospheric initial conditions
were taken from long simulations made with the respective AGCMs forced by the
observed time-dependent SST. The upper
ocean heat content is thought to contain the memory of the coupled system, at
least in the equatorial ocean (e.g. Schneider et al., 1995), which provides a
justification for using an analysis of the ocean, together with a “balanced”
atmosphere as the initial states for the forecasts.
Retrospective forecasts using 7
ensemble members for the period October 1980 to present have been made using
the coupled system with ECHAM AGCM component.
Similar
retrospective forecasts using 5 ensemble members have been made using the
coupled system with COLA AGCM component.
The variance of SST anomalies differs between the two models and is also
not the same as found in nature. The
retrospective forecasts from both of the coupled systems have been normalized
by the observed variance. At each point
a bootstrap cross validation procedure is applied to bring the forecast
variance closer to observed. In this calculation, the forecast for a particular
year is not used to compute the mean or the standard deviation for either the
model or the observations. The forecast
anomaly for that year is taken as the difference from the model climatology
made without using that year multiplied by the ratio of observed to model
standard deviation also calculated without the data from the year of the
forecast. The final SST forecast is the
simple arithmetic mean of the normalized forecasts from the two coupled models.
An
estimate of the skill of the combined coupled models is provided by the
retrospective forecasts for the period 1980 to 2002. Here the forecasts for October 1
initial conditions are shown. It is known that the skill for tropical
Pacific SST forecasts is a strong function of season (Balmaseda et al., 1995)
so it is important to examine forecast skill as a function of initial condition
date. As a comparison forecast we use
persistence of the SST anomaly at the start of the forecast. The persistence forecast has been processed
using the same bootstrapping methodology to bring the variance closer to
observed as described for the coupled forecast above. Figure 1 shows the anomaly correlation
coefficient and root mean square error (RMSE) for the combined coupled model
forecast and persistence for SST averaged over two regions. The regions shown are Nino3.4 (170°W
to 120°W, 5°S to 5°N) and Nino3 (150°W to 90°W). For all
lead times and both Nino regions the combined coupled model forecast has a
higher correlation coefficient and a smaller RMSE indicating that the model is
more skillful then the persistence forecast.
Also shown in Figure 1 are the skill scores for
the individual coupled model systems.
It can be seen that the multi-model combination always has equal or
higher anomaly correlation and equal or lower RMSE compared with the two
component models.
The
latitude-longitude distribution of anomaly correlation and RMSE at lead times
of 3 and 6 months are shown in Figures 2 and 3 respectively. For
the near equatorial region (within 7 degrees of the equator) from the western
boundary of Nino3.4 (located at 170°W) east to the South American coast the combined
coupled model forecast has higher anomaly correlation coefficient and lower
RMSE then the persistence forecast.
Further the level of skill of the coupled forecasts in the near equatorial
region is relatively high with rather a large region of correlation greater
then 0.6.
The current forecast was made
using initial conditions from October 1, 2003. The horizontal distribution of
the forecast is given in Figure 4 for two months,
December 2003 and March 2004. The forecast for December 2003 is for weak warm
anomalies in the central and eastern tropical Pacific. At this time the warm anomalies are forecast
to be stronger in the east (Nino3) then the central (Nino3.4) Pacific as seen
in Figure 5.
In
March 2004, the forecast is for weak warm anomalies with magnitude very similar
in the eastern and central Pacific.
These forecasts were made
possible due to help from several institutions. Matt Harrison and Tony Rosati of GFDL developed the ODA system
and ran the 1980 to 1999 period. Ben Kirtman and Duhong Min of COLA have ported
the ODA to their system and run the 2000 to present period. Bohua Huang of COLA implemented the lower
resolution of the OGCM used here. Max
Planck has kindly provided the ECHAM4.5 AGCM to the IRI. David DeWitt was supported by a grant from
the National Oceanic an Atmospheric Administration: (NA07-GP0213). Support for
Schneider and Hu was provided by the National Science Foundation (ATM
98-14295,ATM01-22859), the National Oceanic and Atmospheric Administration (NA
96-GP0056), and the National Aeronautics and Space Administration (NAG 5-8202).
Figure Captions
Figure 1. Anomaly correlation of retrospective forecasts
for the combined coupled model, ECHAM4.5 coupled model, COLA coupled model, and
persistence for October 1 initial conditions during period 1980 to 2002. (a) Nino3.4, (c) Nino3. Panels (b) and (d)
are same as panels (a) and (c) but for root mean square error (RMSE).
Figure 2. Anomaly correlation for the combined coupled
forecast and persistence for October 1 initial conditions during period 1980 to
2002. (a) Combined coupled forecast at
3 month lead. (b) Persistence at 3 month
lead. (c) Same as panel (a) but for 6 month lead. (d) same as panel (b) but for
6 month lead.
Figure 3. Root mean
square error (RMSE) for the combined coupled forecast and persistence for
October 1 initial conditions during period 1980 to 2002. (a) Combined coupled model forecast at 3
month lead. (b) Persistence at 3 month
lead. (c) Same as panel (a) but for 6
month lead. (d) Same as panel (b) but for 6 month lead.
Figure 4. Combined coupled model forecast for the tropical
Pacific. (a) Forecast for December 2003. (b) Forecast for March 2004.
Figure 5. Combined coupled model forecast for the Nino3 and
Nino3.4 regions.
References
Balmaseda, M. A., M. K.
Davey, and D. L. T. Anderson, 1995:
Decadal and seasonal dependence of ENSO prediction skill. J. Climate, 8, 2705-2715.
Brinkop, S. and E. Roeckner,
1995: Sensitivity of a general
circulation model to parameterization of cloud-turbulence interactions in the
atmospheric boundary layer. Tellus,
47A, 197-220.
Davies, R., 1982:
Documentation of the solar radiation parameterization in the GLAS climate
model. NASA Technical Memorandum 83961,
57pp. [Available from NASA Center for
Aerospace Information (CASI), 7121 Standard Dr., Hanover, MD 21076-1320.]
Derber, J. and A. Rosati, 1989: A global oceanic data assimilation
system. J. Phys. Oceanogr., 19, 1333-1347.
DeWitt, D. G., 1996: The effect of the cumulus convection scheme on the
climate of the COLA general circulation model.
COLA Rep. 27, 58 pp. [Available from COLA, 4041 Powder Mill Rd, Suite
302, Calverton, MD 20705.]
DeWitt, D. G., and E. K. Schneider, 1996: The Earth radiation budget as
simulated by the COLA GCM. COLA Rep.
35, 39pp. [Available from COLA, 4041 Powder Mill Rd., Suite 302, Calverton, MD
20705.]
Fouquart, Y., and B. Bonnel,
1980: Computations of solar heating of
the Earth’s atmosphere: A new parameterization. Beitr. Phys. Atmos., 53, 35-62.
Gent, P. R., and J. C. McWilliams, 1990: Isopycnal
mixing in ocean circulation models. J.
Phys. Oceanogr., 25, 150-155.
Harshvardhan,
R. Davies, D. A. Randall, and T. G. Corsetti, 1987: A fast radiation parameterization for atmospheric circulation
models. J. Geophys. Res, 92(D1),
1009-1016..
Kiehl, J. T., J. J. Hack,
and B. P. Briegleb, 1994: The simulated
Earth radiation budget of the National Center for Atmospheric Research
community climate model CCM2 and comparisons with the Earth Radiation Budget
Experiment (ERBE). J. Geophys. Res.,
99, 20815-20827.
Kirtman, B. P., A. Vernekar,
D. DeWitt, and J. Zhou, 1992: Impact of orographic gravity wave drag on
extended range forecasts with the COLA GCM.
Atmosfera, 6 3-24.
Lacis, A. A., and J. E.
Hansen, 1974: A parameterization for
the absorption of solar radiation in the earth’s atmosphere. J. Atmos. Sci., 31, 118-133.
Large, W. G., J. C.
McWilliams, and S. C. Doney, 1994: Oceanic vertical mixing: A review and a
model with a nonlocal boundary layer parameterization. Rev.
of Geophys., 32, 363-403.
Louis, J. F., 1979: A parametric model of vertical eddy fluxes
in the atmosphere. Bound. Layer
Meteor., 17, 187-202.
Mellor, G.L. and T. Yamada,
1982: Development of a turbulence
closure model for geophysical fluid processes.
Rev. Geophys. Space Phys., 20, 851-875.
Moorthi, S. and M. J.
Suarez, 1992: Relaxed Arakawa-Schubert: A parameterization of moist convection
for general circulation models. Mon.
Wea. Rev., 120, 978-1002.
Morcrette, J.-J., L. Smith,
and Y. Fouquart, 1986: Pressure and
temperature dependence of the absorption in longwave radiation
parameterizations. Beitr. Phys.
Atmos., 59, 455-469.
Nordeng, T. E., 1994:
Extended versions of the convective parameterization scheme at ECMWF and their
impact on the mean and transient activity of the model in the Tropics. ECMWF Research Department Tech. Memo No.
206, European Centre for Medium-Range Weather Forecasts, 41pp.[Available from
ECMWF, Shinfield Park, Reading, Berkshire, RG29AZ, United Kingdom.]
Pacanowski, R. C., and S. M.
Griffes, 1998: MOM 3.0 Manual,
NOAA/Geophysical Fluid Dynamics Laboratory, Princeton, USA 08542.
Redi, M. H., 1982: Oceanic isopycnal mixing by coordinate
rotation. J. of Phys. Oceanogr.,
12, 1155-1158.
Rockel, B. E. Raschke, and
B. Weyres, 1991: A parameterization of
broad band radiative transfer properties of water, ice and mixed clouds. Beitr. Phys. Atmos., 64, 1-12.
Roeckner, E., 1995:
Parameterization of cloud radiative properties in the ECHAM4 model. Proc. WCRP Workshop on Cloud Microphysics
Parameterizations in Global Atmospheric Circulation Models, Kananaskis, AB,
Canada, WCRP, 105-116.
Roeckner, E., and Coauthors,
1996: The atmospheric general
circulation model ECHAM4: Model
description and simulation of present day climate. Max-Planck-Institut fur Meteorologie, Rep. 218, 90pp. [Available
from MPI fur Meteorlogie, Bundesstr. 55, 20146 Hamburg, Germany.]
Schneider, E. K., B. Huang,
and J. Shukla, 1995: Ocean wave dynamics and El Nino. J. Climate, 8, 2415-2439.
Schneider, E. K., 2002: Understanding the differences between the
equatorial Pacific as simulated by two coupled GCMs. J. Climate, 15,
449-469.
Schneider, E. K., D. G.
DeWitt, A. Rosati, B. P. Kirtman, L. Ji, and J. J. Tribbia, 2003: Retrospective ENSO Forecasts: Sensitivity to atmospheric model and ocean
resolution. Mon. Wea. Rev., 131,
3038-3060.
Smagorinsky, J. 1963: General circulation experiments with the
primitive equations: I. The basic experiment.
Mon. Wea. Rev., 91, 99-164.
Terray, L., A. Piacentini,
and S. Valcke, 1999: OASIS2.3, Ocean Atmosphere Sea Ice Soil Users Guide,
CERFACS Technical Report TR/CMGC/99/37, CERFACS, Toulouse, France. [Available
on the web at http://www.cerfacs.fr/globc/publication.html]
Tiedtke, M., 1984: The effect of penetrative cumulus convection
on the large-scale flow in a general circulation model. Beitr. Phys. Atmos., 57
216-239.
Tiedtke, M., 1989: A
comprehensive mass flux scheme for cumulus parameterization in large-scale
models. Mon. Wea. Rev., 117,
1779-1800.
