Contributed by Iracema F. A. Cavalcanti, Luciano P. Pezzi, Paulo Nobre, Gilvan Sampaio, Helio Camargo Jr.

Monthly to seasonal dynamic atmospheric predictions at CPTEC have been made since January 1995. The model used for these predictions is the CPTEC version of the COLA AGCM which was derived from the NCEP model (Kinter et al., 1988) and includes a sophisticated biosphere model (Xue et al., 1991). The horizontal resolution of the AGCM is T62 and there are 28 levels in the vertical (L28; Cavalcanti et al., 1995). Predictions for the rainy season of Nordeste (Northeast of Brazil) in 1995 were discussed in Nobre et al. (1995), and those for 1996, in Nobre and Cavalcanti (1996). Results from simulations (with a T42L18 version of the model) of the Nordeste rainy season of 1993 and 1994 using observed monthly sea surface temperature (SST) as boundary conditions compared well with observed values of precipitation, and with NCEP reanalysis data are described in Cavalcanti et al., 1996. Precipitation anomalies in Nordeste (area averaged, 2⁰S-12⁰S; 45⁰W-35⁰W) from an 11 year simulation (1986 to 1996) using the T42L18 model is shown in Fig. 1 along with the observed precipitation anomaly. With some exceptions, the model reproduces much of the interannual rainfall variability in the Noredeste region.

Three sets of predictions are presented here. The first set of predictions were initialized in October 1997 and the model was integrated through June 1998. In this set of predictions, 25 different initial conditions (i.e., October 1-25) were used. The SST includes the climatological annual cycle plus persisted SST anomalies from November 1997. The second set of predictions used 5 initial conditions (February 14-18) and were integrated through June 1998. In this second set of predictions, the SSTA in both the Atlantic and Pacific are based on real predictions. The Pacific SSTA is provided by the NCEP coupled model (Ji et al., 1996) initialized on February 1. The Atlantic SSTA is predicted using statistical techniques (see Pezzi et al., in this Bulletin). In all the other ocean basins persisted SSTA are used. In the third set of predictions, the same 5 February initial conditions were used. The SSTA in the Pacific came from the NCEP coupled model and in all other basins persisted SSTA was used.

In order to define predicted anomalies, an AGCM climatology must be defined. The AGCM climatology is obtained from a separate ensemble of 23 integrations using climatological SST.

In all three sets of predictions the spatial structure of the March-May 1998 (MAM98) rainfall anomalies are quite similar. However, the amplitude of the anomalies differs considerable among the three sets of predictions. The MAM98 rainfall anomalies in Noredeste and central-west Brazil, for example, are largest in the case of predicted SSTA in the tropical Pacific only. All three sets of predictions call for below average rainfall in Nordeste, central-west and southwest Amazonia, and above average rainfall in the southwest of south Brazil.

The predicted rainfall from all 25 members of the first set of prediction with persisted SSTA is shown in Figs. 2a-e. In the Nordeste region (Fig. 2a) all 25 predictions are similar and call for below average rainfall from January to June 1998. In the Southeast Region (Fig. 2b) there is a large spread among the ensemble members which significantly reduces the confidence in the ensemble mean forecast. Although there is considerable variability among the ensemble members, the south region (Fig. 2c) shows fairly consistent positive anomalies for March to June 1998. In the central west region (Fig. 2d) near normal conditions are predicted for May-June 1998. In the north region (Fig. 2e), consistent negative anomalies are predicted with the ensemble average minimum in April 1998.

Figures 3a and 3b show the predicted MAM98 rainfall anomalies from the second and third set of predictions, respectively. Both sets of predictions indicate below average precipitation over much of northeast Brazil. Predicted anomalies in the Atlantic (Fig. 3a) gives stronger negative anomalies in northern portion of northeast Brazil and somewhat weaker anomalies over central northeast. With persisted anomalies in the Atlantic (Fig. 3b), the forecast calls for positive anomalies in the northwest of Nordeste, whereas with predicted SSTA in the Atlantic, the positive anomalies are shifted to the north over the ocean. In Amazonia, both sets of predictions call for below average precipitation. Over south Brazil, positive anomalies are more intense in the persisted SST case. These results are similar to those obtained in December 1997 and January 1998 which showed precipitation below average for MAM, using persisted SST anomalies of December 1997 and January 1998 (not shown) and predicted SST of Pacific Ocean (NCEP) and Atlantic Ocean (CPTEC). The negative values of anomalies over northeast and Amazonia are related to the above average SST in the tropical east Pacific and to the unfavorable conditions for rain over northeast in the Atlantic Ocean (warmer than normal in the North Atlantic and colder than normal in the South Atlantic).

Acknowledgments: To NEC (Japan), Jonas Tamaoki and Mariano Pereira Silva. 

Cavalcanti, I. F. A; P.Nobre; M.L.Abreu; M.Quadro; L.P.Pezzi, 1995. Vertical and horizontal resolution comparisons of CPTEC/COLA GCM. Proceedings of the twentieth annual climate diagnostics workshop., Seattle, Washington, Oct. 23-27, 1995. pp 73-76.

Cavalcanti, I.F.A., P. Nobre; I. Trosnikov. Simulação de verão e outono de 92/93 e 93/94 com o GCM CPTEC/COLA. IX Congresso Brasileiro de Meteorologia, Campos do Jordão, Nov.1996, pp. 807-811.

Ji. M., A. Leetmaa and V. E. Kousky, 1996: Coupled model forecasts of ENSO during the 1980s and 1990s at the National Meteorological Center. J. Climate, 9, 3105-3120.

Kinter, J.L.; J. Shukla; L. Marx; E.K. Schneider, 1988. A simulation of winter and summer circulation with the NMC global spectral model. J. Atmos. Sci., 45, 2486-2522.

Moura, A. D., J. Shukla, 1981: On the dynamics of drought in Northeast Brazil: observations, theory and numerical experiments with a general circulation model. J. Atmos. Sci., 38, 2653-2675.

Nobre, P., I. F. A. Cavalcanti, 1996: Previsão Climática Sazonal no CPTEC-A estação chuvosa de 1995 e 1996 no Nordeste do Brasil. Congresso de Meteorologia Argentino (Congremet VII e Climet VII), Buenos Aires, 2-6 setembro, 1996, pp.351-352.

Nobre, P., M. L. Abreu, I. F. A. Cavalcanti; M.Quadro; L. P. Pezzi, 1995. Climate ensemble forecasting at CPTEC. Proceedings of the twentieth annual climate diagnostics workshop., Seattle, Washington, Oct. 23-27, 1995. pp 417-420.

Xue,Y.; P. Sellers. J. L. Kinter; J. Shukla, 1991. A simple biosphere model for global climate studies. J. Climate, 4, 345-364.

Fig. 1: Averaged precipitation anomalies in Nordeste (2° S-12° S; 45° W-35° W) for MAM, 1986 to 1996, from the model CPTEC/COLA results (T42L18), (dashed line) and observed data (solid line).

Fig. 2: Time evolution of area averaged precipitation anomalies for all 25 cases with persisted November 1997 SSTA. The five regions are (a) Northeast (3^oS-15^oS, 45^oW-35⁰W), (b) Southeast (15^oS-25^oS, 40^oW-52⁰W), (c) South (21^oS-34^oS, 48^oW-58⁰W), (d) Central west (10^oS-23^oS, 50^oW-62⁰W) and (e) North (1^oS-10^oS, 69^oW-45⁰W)

Fig. 3: Predicted MAM98 precipitation anomalies for second and third set of predictions (see text). (a) Shows the AGCM ensemble mean prediction using predicted SSTA in the Pacific and Atlantic and (b) shows the ensemble prediction using predicted SSTA in the Pacific only.