NASA/GMAO Coupled General Circulation Model Dynamical Model
NASA Goddard Space Flight Center (GSFC)
Greenbelt, Maryland, U.S.
The NASA/GMAO (Global Modeling and Assimilation Office), which now includes the former NSIPP project, produces various seasonal forecasts each month. These include so-called Tier-I ensemble forecasts, run for 12 months with a fully coupled global ocean-atmosphere-land model using an 18 member ensemble, and Tier-II, atmosphere-only forecasts, using specified estimates of future SSTs.
For the Tier 1 forecasts, the ocean is initialized with states from the GMAO's ocean assimilation system, which analyzes in situ observations of ocean temperature profiles using an optimal interpolation (OI) scheme. Atmospheric states for the ensemble are a combination of states obtained from an ensemble of AMIP-style (Atmospheric Model Intercomparison Project) runs of the atmospheric model and states from NCEP's Reanalysis (CDAS) product. The land surface states are a combination of states from an LSM offline with GLDAS (Global Land Data Assimilation System; browse here) forcing and states from the AMIP run.
The ocean model used in the Tier 1 forecasts is Poseidon V4 (Schopf and Loughe, 1995) and uses a 1/3-degree meridional resolution, 5/8-degree zonal resolution, and 27 layers. The atmospheric model is the NSIPP1 AGCM (Bacmeister et al., 2000) and has 2 X 2.5 X 34L resolution. The land surface model is the Mosaic LSM (Koster and Suarez, 1992).
Beginning in September 2002, 18-member ensemble 12-month forecasts were conducted. The ensembles are generated by perturbing the initial states of either the ocean or atmosphere. Six ensemble members use ocean-only perturbations with the AMIP atmosphere and the GLDAS land; six use ocean-only perturbations with the CDAS atmosphere and the GLDAS land; five use the base GMAO ocean analysis with perturbed atmospheres from CDAS and land from GLDAS land. One uses the AMIP atmosphere with the GMAO ocean analysis and AMIP land. Forecasts prior to September 2002 use only 6-member ensembles - 3 ocean-only perturbations, 2 atmosphere-only perturbations, and 1 with no perturbations; for these all atmospheres are based on AMIP runs.
The perturbations for the ocean are generated from differences between analysis states randomly chosen within 15 days of the forecasts' initialization time. The perturbations for the atmosphere are generated by choosing a random pair of ensemble members of an existing up-to-date AMIP run. A fraction of the difference of this pair is then added to a "base member".
To date 12-month forecasts or hindcasts have been conducted from the 1st of each month from 1993 to the present. Forecast anomalies are produced by subtracting the ensemble mean climatological forecast drift as a function of lead time. This drift correction is estimated separately for each starting month from hindcasts done over the 1993-2005 period.
GMAO also runs a 12-member ensemble of its coupled land-atmosphere model with prescribed sea surface temperatures. These forecasts are referred to as the Tier 2 forecasts and are run for seven months. The ensemble is made by using four initial states (taken from the AMIP ensemble), together with the three forecast SST scenarios produced by IRI as a lower boundary condition.
View the model SST forecast plume.
Contact: Dr. Michele Rienecker: Michele.Rienecker@nasa.gov and Dr. Max Suarez: Max.J.Suarez@nasa.gov
References:
Bacmeister, J. T., P. J. Pegion, S. D. Schubert, and M. J. Suarez, 2000: Atlas of seasonal means simulated by the NSIPP1 atmospheric GCM, NASA Tech. Memo-2000-104606, Vol. 17, 194pp.
Koster, R., and M. Suarez, 1992: Modeling the land surface boundary in climate models as a composite of independent vegetation stands. J. Geophys. Res., 97, 2697-2715.
Pegion, P., S. Schubert, and M. J. Suarez, 2000: An Assessment of the Predictability of Northern Winter Seasonal Means with the NSIPP1 AGCM. NASA Tech. Memo-2000-104606, Vol. 18, 100pp.
Schubert, S., M. J. Suarez, P. Pegion and M. Kistler, and A. Kumar, 2002: Predictability of Zonal Means During Boreal Summer, J. Climate, 15, 420-434.
Schopf, P., and A. Loughe, 1995: A reduced-gravity isopycnic ocean model: Hindcasts of El Nino. Mon. Wea. Rev., 123, 2839-2863.
Troccoli, A., M. M. Rienecker, C. L. Keppenne, and G. C. Johnson, 2003: Temperature data assimilation with salinity corrections: Validation for the NSIPP Ocean Data Assimilation System in the tropical Pacific Ocean, 1993-1998, NASA Tech. Memo-2003-1046-6, Vol. 24, 23pp.