Current research on extending trajectory analysis of oil spills includes the analysis of best use of wind forecast data. For longer range trajectory support, it is necessary to blend the forecast and climatological wind data. Mobilization of oil recovery teams and equipment often requires lead times of several days, however, verification studies have shown little to no skill beyond day two in trajectory forecasts being forced by routine NWS or objective model forecast surface winds. This research focuses on identifying the most appropriate timescales for the application of both real-time forecasts and climatologically derived winds to achieve the most skillful trajectory forecasts. The approach includes vectorizing both forecast winds and the historical climatological wind data, normalizing and comparing those hourly vectors within some mean error. The sensitivity of this mean error in relation to the data will give a general scale of regional wind uncertainty. Understanding of this scale of uncertainty will provide information that can be used in better estimating confidence bounds for trajectory analysis products therefore providing more accurate predictions.