Seasonal Storminess in the North Pacific, Bering Sea, and Alaskan Regions

In all seasons of the year, storms represent high-impact weather events over Alaska and its adjacent seas. Alaska's extensive coastline and, in many cases, shallow onshore shelves makes the region especially vulnerable to coastal flooding and erosion, particularly where a protective sea ice buffer is absent. It is of interest to gain an assessment of possible storminess patterns at the medium range; that is, multi-week to seasonal. As with existing, similar products for precipitation and temperature, such information could aid stakeholders in planning and resource allocation. The NOAA's Climate Prediction Center (CPC) Storm Tracks website currently includes summaries of storm tracks and accumulated precipitation for the past 10-, 30- and 90-day periods. This project aims to explore probabilistic methods to enable a medium-range (out to 90 days) storminess prediction product to be added to the CPC portfolio. Given the limits of deterministic predictability, methods will be based on present and CPC-predicted states of ENSO, the Pacific Decadal Oscillation (PDO), the Pacific/North American Pattern (PNA) and the Arctic Oscillation (AO) -- large-scale modes of variability known to affect Alaska – to the extent that they control storminess in the Alaska region. In order to link these large-scale climatic modes of variability to storm probabilities for various Alaskan sub-regions, and to account for the fact that these modes influence each other, we will explore the use of composites and analog years. The goal of the initial stage of this effort is intended to identify spatial and temporal patterns in loci of storm frequency maxima and minima as determined by the CPC storm track database, compiled using the method of Serreze (1995). Highlighted will be potential links of storm frequency patterns to each climatic mode noted above using Empirical Orthogonal Function (EOF) analysis. Additionally, the seasonal storminess hindcasts used to generate the seasonal storminess outlook for the region are shown, with discussion and interpretation of the utility of individual teleconnection patterns in the generation of a seasonal storminess outlook product for the region.