Ongoing oil development in the Beaufort/Chukchi Seas is accompanied by the potential threat of oil spills. In the event of a spill, time is of the essence in directing mitigation, cleanup, and recovery efforts, and thus improving the predictability of oil spill transport is of great importance. As the surface wind field is the primary factor in driving ocean currents, and thus the dispersal of any accompanying oil, accurate modeling of the surface winds is essential in enhancing the prediction of oil spill transport. As such, a study of the mesoscale meteorology of the Beaufort/Chukchi region has been commissioned by the U.S. Bureau of Ocean Energy Management, Regulation and Enforcement (BOEMRE) in an effort to ensure the accurate simulation of near-surface winds, which will thereby lead to improved prediction of oil spill dispersal. The final goal of the project is to produce a high-quality, long-term, high-resolution reanalysis of the region's mesoscale meteorology that will be used to drive oil spill transport models.

The Beaufort/Chukchi region represents a highly complex geographical environment. It comprises highly varying topography, bounded in the south by the Brooks Range, and a constantly changing sea ice presence in the ocean. All of these present a great challenge to the accurate modeling of the Beaufort/Chukchi regional meteorology and the associated surface winds. In addition, due to its remote nature, observations are sparse throughout the area, further complicating efforts to accurately simulate atmospheric conditions in the region, and making it all the more important to fully utilize any available observations through data assimilation. In this study, the Weather Research and Forecasting (WRF) model and its variational data assimilation system were used to conduct numerical simulations of the region's mesoscale meteorology at a grid spacing of 10 km. As part of the effort to optimize the model configuration for application over the study domain, various assimilation approaches have been attempted, using different assimilation packages, background error computations, and combinations of observation types, including both in situ and satellite, with varying degrees of success. In addition to observational assimilation, the use of various sea ice, snowcover, and SST datasets to modify the lower boundary conditions has been explored. The experimental results of these sensitivity tests are presented, demonstrating the successes and failures of the different approaches in the study region, and documenting the process undertaken to finalize the model configuration. After testing was concluded, a 31-year (1979–2009) reanalysis with the optimized model and assimilation configuration was produced. The final reanalysis demonstrates significant improvements in representing surface conditions, particularly surface winds, relative to those given in the ERA-Interim reanalysis, to this point the most accurate reanalysis available over the study domain.