Stratospheric Network for the Assessment of Predictability
The draft experimental protocol for SNAP experiments is shown below. These protocols will be discussed and revised at the first SNAP workshop in Reading in April 2013. There are several interesting recent events which might form the basis of some of the SNAP experiments including the very large major stratospheric sudden warming in January 2009, the major warming in February 2010 and the anomalously strong vortex in March and April 2011.
| Tier 1 |
Best High - model with a high-top above the stratopause and with all stratospheric processes included, initialised with state from data assimilation of the high-top model. |
| Tier 2 |
Best Low - standard version of low-top model with parameter settings and physics appropriate for the low-top model, initialised with the same state as the best high run. Tropospheric levels should be identical to best high. |
| Tier 3 |
Degraded - model with a low-top (possibly around 10hPa), degraded from the high-top version with parameter settings identical to the best high run, initialised with the same state as the best high run. High common - best version of high-top model as above but initialised with a standard common analysis (most likely from ECMWF). This set of integrations will be used to quantify the forecast improvement resulting from the improved initial conditions from the high-top model. |