Characteristics And Variability Of Storm Tracks In The North Pacific, Bering Sea And Alaska

Abstract

Storm activity in the North Pacific, Bering Sea and Alaska regions is investigated using various automated storm tracking and parameter extraction algorithms. Specific, novel details of storm activity throughout the year are presented. The influence of major climatic drivers is considered, including the Pacific/North American Index and sea ice variability. Details of synoptic-scale forcing on a specific, severe storm event are considered in the context of how different tracking algorithms are able to depict the event. New storm climatology results show that the inter-seasonal variability is not as large during spring and autumn as it is in winter. Most storm variables exhibited a maxima pattern that was oriented along a zonal axis. From season to season this axis underwent a north-south shift and, in some cases, a rotation to the northeast. Barotropic processes have an influence in shaping the downstream end of storm tracks and, together with the blocking influence of the coastal orography of northwest North America, result in high lysis concentrations, effectively making the Gulf of Alaska the "graveyard" of Pacific storms. Summer storms tended to be longest in duration. Temporal trends tended to be weak over the study area. Sea surface temperature did not emerge as a major cyclogenesis control in the Gulf of Alaska. Positive sea-ice anomalies in the Sea of Okhotsk were found to decrease secondary cyclogenesis, shift cyclolysis locations westward, and alter the North Pacific subtropical jet. In the Atlantic, a negative North-Atlantic-Oscillation-like pattern is observed; these results were confirmed by experiments on the ECHAM5 Atmospheric Global Circulation Model driven with sea-ice anomalies in the Sea of Okhotsk. The destructive west Alaska storm of autumn 1992, which flooded Nome, was investigated using two storm tracking algorithms: NOAA's (National Oceanic and Atmospheric Administration) current operational algorithm and the Melbourne algorithm. Manual tracking was performed as a control. The main storm location features were captured by both algorithms, but differed in the genesis and lysis location. The NOAA algorithm broke the event into two. This storm was shown to have been affected by a blocking high that influenced how the tracking algorithms handled the event.