Lower-tropospheric potential vorticity (PV) maxima in the vicinity of cold-frontal rain bands have been shown to contribute substantially to the strength of the low-level jet (LLJ) and warm-sector moisture transport in extratropical cyclones. A case from February 1997 was selected for detailed analysis based on the presence of a strong LLJ, and a prominent cold-frontal rain band. Errors in operational Eta model forecasts from this case are consistent with a negative bias in the prognosis of a lower-tropospheric, cold-frontal PV maximum. Potential vorticity budget results confirm that latent heat release was a prominent factor in the development and propagation of the lower PV maximum, and potential vorticity inversion demonstrates that this feature contributed approximately 25% to the strength of the LLJ in that case. These results support the hypothesis of a positive feedback involving diabatic PV redistribution, the LLJ, and warm-sector moisture transport. The errors associated with the PV forecast are consistent with errors in forecasts of sea level pressure, geopotential height, and the strength of the LLJ. The cold-frontal rain band in this case was characterized by organized convection, with trailing stratiform precipitation; Eta model forecasts exhibited both parameterized convective and grid scale precipitation in the vicinity of this feature. This case study illustrates the linkage between model precipitation forecasts and predictions of other dynamical variables, and underscores the importance of proper representation of lower PV maxima in heavy precipitation events.

It is hypothesized that model representation of the diabatic cold-frontal PV maximum in cases such as that discussed above is sensitive to the choice of convective parameterization scheme (CPS) and perhaps also to details of the grid scale precipitation scheme. Towards the ultimate goal of determining which existing CPS best represents diabatic redistribution of PV in organized convective systems, this study examines the sensitivity of the lower PV maximum to the choice of CPS and precipitation scheme. The Penn State/NCAR MM5 model was used for the sensitivity experiments. The control run utilized the Kain-Fritsch CPS and Reisner mixed-phase precipitation scheme. Experiments included runs with no latent heat and "simple ice"; choices of CPS included the Grell, Betts-Miller, and Arakawa-Schubert schemes, and an explicit run with a horizontal grid spacing of 1 km. The results confirm that the representation of this PV maximum is sensitive to CPS, and again highlight the prominent role of diabatic processes in the development of the LLJ. Additional research will be necessary in order to identify a CPS that optimally redistributes PV in convective rain bands.