University of Oklahoma

100 E. Boyd, Rm. 1110

Norman, OK 73019

403-325-0453

djahn@ou.edu

Research Interest:Ê Numerical Prediction of High-Impact, Localized Weather

In order to identify the regions that would be affected by the release of a toxic, air-borne agent into the atmosphere, it is paramount that accurate forecasts of wind and precipitation are available at a scale commensurate with the detail of a highly-specified region.Ê Weather can vary significantly over a relatively short distance, such as tens of kilometers, especially in the region of frontal boundaries and as the result of complex storm systems.Ê Numerical weather models over the past several decades have provided information as to the future evolution of the atmosphere over general sub-regions of the country, but are not capable of forecasting weather for specific localities in consideration of highly localized, fine-scale effects.Ê

The Center for Analysis and Prediction of Storms (CAPS) at the University of Oklahoma has developed over the past decade a numerical atmospheric model that is appropriate to predict weather conditions down to a sub-county scale.ÊÊ The Advanced Regional Prediction System (ARPS) is capable of producing in only one hour a 10-hour weather forecast, which provides detail down to a spatial resolution of 2 km.Ê The resulting datafiles can be provided in a format appropriate for use by a dispersion model, such as the EPAâs Models3.

Figure 1.Ê Streamlines showing direction of the wind regime as analyzed by the ARPS over the Southern Plains at 9:00 a.m. on 10/6/02.

The accuracy of the ARPS to forecast the weather at relatively fine scales is highly linked to unique capabilities that have been developed at CAPS for the use of high-resolution observations in model initialization.Ê In particular, CAPS has exploited the data from the NEXRAD radar network and has formulated techniques to numerically retrieve unobserved variables, such as the winds tangential to the radar beam, in order to acquire a highly-detailed representation of the full 3D wind component.Ê The resulting high-resolution analysis of the current wind regime over a highly specified region is in itself very valuable information in the assessment of particulate dispersion (Fig. 1).Ê Such information has also been shown to be paramount in the numerical prediction of severe local weather.

Real-time ARPS analyses and forecasts are available on-line at http://caps.ou.edu/wx.