Tuesday, 12 August 2008: 2:30 PM
Rainbow Theatre (Telus Whistler Conference Centre)
Counter to dynamical intuition, storminess in the Pacific Ocean minimizes in the middle of winter, when meridional temperature gradients and jet stream winds are at their peak. It's been speculated, but never demonstrated, that a lull in wintertime seed disturbances over Asia could explain this puzzling, yet pronounced feature of Northern hemisphere climate. The dominant source of initial disturbances in the mid-latitudes is lee cyclones, and activity over Mongolia is notably frequent, with an average of ~6 events per month between 40-55N, 90-120E. We thoroughly investigate the hypothesis that lee cyclogenesis over the Mongolian Altai mountains plays a dominant role in the midwinter suppression of the Pacific storm track. By applying feature tracking algorithms to 23 years of reanalysis data we have built a comprehensive database of storms, allowing us to investigate the connection definitively. Furthermore, feature tracking offers a new storm-centered Lagrangian perspective, as Eulerian variance techniques have been used almost exclusively to describe the midwinter minimum in the past. Our results have revealed that the Japan Sea is one of two major cyclogenesis regions for the Pacific storm track, and approximately half of the storms that enter this area are lee cyclones from Mongolia. In addition, there are ~20% fewer Mongolian lee cyclogenesis events in winter than in fall and spring, a reduction similar in magnitude to that found over the Pacific Ocean. The decrease in Mongolia is likely related to a southward shift of the jet stream relative to the mountains during winter, an idea we are testing through consideration of the dynamics controlling individual storms in our database. These are powerful suggestions that, via seeding, mountains are a control on storminess over the Pacific Ocean and an important component to atmospheric circulation in general.
- Indicates paper has been withdrawn from meeting
- Indicates an Award Winner