Between the dates of 28 June and 31 July 2003, the Joint Urban 2003 Field Experiment (JU2003) was held in Oklahoma City to study atmospheric dispersion through the urban environment. During the experiment, a tracer (sulfur hexafluoride) was released during 10 intensive operational periods (IOP). It was the task of the operational support team to forecast conditions favorable for IOPs and monitor weather conditions during each IOP. One key aspect of the planning and timing of the experiment were the persistent southerly winds typically observed during late June and July. Because of such conditions, portable and fixed instrumentation was strategically placed in and around Oklahoma City to account for a general northward movement of the tracer plume. However, it was necessary to provide a period of 24 hours advanced notice to properly prepare and place the portable samplers used to detect the tracer. Thus, very specific forecasts of wind speed and direction were requested for a 24-36 hour window encompassed by IOP operations.

The operational support team faced many challenges in providing accurate forecasts to the JU2003 experiment team. For example, while a number of numerical weather prediction models were used at grid scales ranging in excess of 50 km to 0.5 km, significant errors in wind speed and direction during the 24-36 hour forecast cycle were observed during JU2003. As such, many forecasts relied heavily upon local climatology and forecaster intuition. Furthermore, once an IOP was underway, nowcasters were forced to identify and deal with numerous atmospheric phenomena which threatened IOP operations including mesoscale convective systems, air mass thunderstorms, meso-high and meso-low pressure systems, persistent surface boundaries and fronts, the low-level jet, and local confluence zones. Each phenomenon threatened to impact surface winds within Oklahoma City and to advect the tracer off of the observational grid.