Ensemble Atmospheric GCM Simulation of Climate Interannual Variability from 1979 to 1994
Open Access
- 1 April 1999
- journal article
- Published by American Meteorological Society in Journal of Climate
- Vol. 12 (4) , 986-1001
- https://doi.org/10.1175/1520-0442(1999)012<0986:eagsoc>2.0.co;2
Abstract
The climate interannual variability is examined using the general circulation model (GCM) developed at the Laboratoire de Météorologie Dynamique. The model is forced by the observed sea surface temperature for the period 1979–94. An ensemble of eight simulations is realized with different initial conditions. The variability of the Southern Oscillation is studied. The simulated sea level pressure anomalies at both Tahiti and Darwin are realistic compared to observations. It is revealed, however, that the simulated convection activity response to the warm episode of El Niño is too weak over the eastern part of the tropical Pacific. This explains why the simulated Pacific–North American pattern is shifted westward. A global El Niño pattern index is defined and calculated for both the simulation and the National Centers for Environmental Prediction (NCEP) reanalysis data. This serves as a quantitative measure of El Niño’s global impact. A singular value decomposition analysis performed with the tropical Pacific sea surface temperature and the Northern Hemisphere 500-hPa geopotential height shows that the model’s teleconnection between the Tropics and high latitudes is similar to that of the NCEP reanalysis data. In an exploratory manner, the model’s internal variability versus the external forced variability is studied. It is shown that, except for the equatorial strip, the internal model variability is larger than the external variability. An ensemble mean is thus necessary in order to focus on the model’s response to external sea surface temperature anomalies. An attempt is also made to evaluate statistically the influence of the ensemble’s size on the model’s reproducibility. It is shown that, with this particular GCM, at least five realizations are necessary to correctly assess the teleconnection between the Tropics and the Northern Hemisphere extratropics. This dependency on the number of realizations is less for the tropical circulation.Keywords
This publication has 0 references indexed in Scilit: