Influence of diabatic potential vorticity anomalies upon warm conveyor belt flow. Part I: 14-15 February 2003

Researchers have shown that diabatic potential vorticity (PV) anomalies can have significant impacts on the wind and mass field throughout the troposphere. Both convective and stratiform precipitation, if persistent, can produce diabatic PV anomalies in the mid-troposphere which can impact the location of heavy precipitation downstream. Researchers have used PV inversions to examine the role of PV anomalies in the mid-troposphere on the wind and mass fields throughout the atmosphere. One example is a study of PV anomalies produced by front-parallel convection in the vicinity of a cold front which has been shown to have a positive impact on precipitation downstream by increasing the strength of the low-level jet within the warm conveyor belt and increasing the northward moisture transport.

Numerical models used in operational weather prediction can simulate mesoscale processes including frontal bands and upscale growth of convection. Consequently, they can develop diabatic PV anomalies, advect them downstream, alter the wind and mass field, and initiate precipitation. However, the operational forecaster has no way to determine what impact a diabatic PV anomaly may have on the subsequent position of fronts or precipitation. Therefore, forecasters must rely on hypotheses of the impact of convection that cannot be tested through examination of model output. One common example, which we examine, is the development of heavy precipitation and/or convection in an east-to-west band in the vicinity of a surface warm front. This precipitation can be perpendicular to the low-level jet. Forecasters have hypothesized the presence of this heavy rain has a detrimental effect on precipitation amounts farther north by “cutting off” moisture within the warm conveyor belt.

We examine one case, 14-15 February 2003, where heavy rain develops in the southern plains of the United States and snow occurs across portions of the central Plains. Two sets of model data are examined – the North American Regional Reanalysis (NARR) output available from the National Climate Data Center and a simulation of the case using the Weather Research and Forecasting – Advanced Research WRF (WRF-ARW) model. PV inversions are done on both data sets with two goals. The first goal is to examine the impact of heavy rain that develops across Oklahoma and Arkansas approximately 12 h prior to the development of snow over the central and northern Plains. We find that south of the area of rain, the low-level jet was diverted to the east and it was enhanced to the north of the precipitation increasing moisture transport into the eastern Plains. Second, we use PV inversions to compare two different model solutions. A poor forecast of convection 6 h into the forecast by the WRF-ARW led to the placement of the mid-level front, and associated snow, 100 km too far south. This resulted in a significant underforecast (overforecast) of precipitation across southeast South Dakota, northeast Nebraska and northern Iowa (Kansas, southern Iowa, and northern Missouri). By using PV inversions, we found that the forecast error was a direct result of the diabatic PV anomaly produced by the poorly forecast convection. This illustrated that PV inversions can be used in an operational environment to help determine what is causing model differences and help forecasters to determine if diabatic PV anomalies are having an impact on their forecast. It provides forecasters with a scientific reason to favor one model solution over another and also alert them to areas of precipitation that will need to be monitored because of the large impact on the forecast.