Climate Change 2001:
Working Group I: The Scientific Basis
Other reports in this collection

8.7.6 Pacific-North American (PNA) and Western Pacific (WP) Patterns

The Pacific-North American (PNA) and Western Pacific (WP) patterns are low-frequency teleconnection patterns (Wallace and Gutzler, 1981). Observations show that the PNA and WP are sensitive to the frequency distribution of the SST anomalies associated with ENSO. The HadAM3 model correctly reproduces the changes in frequency distribution of the PNA pattern between the El Niño years and the La Niña years (Renshaw et al., 1998). However, this model fails to reproduce the WP mode distribution. On the other hand, the JMA atmospheric model showed an ability to simulate the WP with reasonable intensity, responding to SST anomalies (Kobayashi et al., 2000). How extra-tropical air-sea interactions affect such weather regimes is not yet clear and an evaluation of the ability of coupled climate models to simulate these modes is yet to be undertaken.

8.7.7 Blocking

Blocking affects the large-scale flow and storm tracks and thus is important for mid-latitude climate. D’Andrea et al. (1998) evaluated the statistical behaviour of fifteen AMIP AGCMs in simulating Northern Hemisphere mid-latitude blocking. The AMIP models simulate reasonably well the seasonality and geographical location of blocking, but have a general tendency to underestimate both blocking frequency and the average duration of blocks. Using the ECMWF model, Brankovic and Molteni (1997) obtained a more realistic representation of Pacific blocking. This was due to reduced systematic error of zonal flow over the north-eastern Pacific. However, model deficiencies still remain in the Atlantic region. A link between the mean flow error in a model and blocking was also shown by Stratton (1999). In general, more recent atmospheric models show an improvement in ability to reproduce atmospheric blocking, but a corresponding evaluation of coupled climate models has not yet been undertaken.

8.7.8 Summary

Recent atmospheric models show improved performance in simulating many of the important phenomena, compared with those at the time of the SAR, by using better physical parametrizations and using higher resolutions both in the horizontal and in the vertical domain. A systematic evaluation of the ability of coupled climate models to simulate a full range of the phenomena referred in this section is yet to be undertaken. However, an intercomparison of El Niño simulations, one of the most important phenomena, has revealed the ability of coupled climate models to simulate the El Niño-like SST variability in the tropical Pacific and its associated changes in precipitation in the tropical monsoon regions, although the region of maximum SST variability is displaced further westward than in the observations.



Other reports in this collection