| Abstract: | On 18–19 February 1979, an intense cyclone developed along the east coast of the United States and produced heavy snowfall accumulations from Virginia to southeast New York. A series of forecast experiments was conducted to assess the accuracy of the GLA model's prediction of this storm and the importance of oceanic heat and moisture fluxes and initial data to the cyclogenesis. The GLA model forecast from the GLA NOSAT analysis at 0000 GMT 18 February correctly predicted that intense coastal cyclogenesis and heavy precipitation would occur, even though important subsynoptic details of the development were underestimated or not forecast. A repetition of this forecast with surface heat and moisture fluxes eliminated failed to predict any cyclogenesis while a similar forecast with only the surface moisture flux excluded showed only very weak cyclonic development. An extended-range forecast from 0000 GMT 16 February as well as forecasts from the GLA FGGE analysis or the NMC analysis at 0000 GMT 18 February interpolated to the GLA grid predicted weaker coastal low development than the forecast from the NOSAT analysis.Detailed examination of these forecasts shows that diabatic heating resulting from oceanic fluxes increased low-level baroclinicity, decreased static stability and significantly contributed both to the generation of low-level cyclonic vorticity, and to the intensification and slow rate of movement of an upper-level ridge over the western Atlantic. As an upper-level short-wave trough approached this ridge, the diabatic heating associated with the release of latent heat intensified and the gradient of vorticity, vorticity advection and upper-level divergence in advance of the trough were increased, which provided strong forcing for the surface cyclogenesis.An examination of the NMC and GLA analyses indicated that a weaker representation of the upper-level trough in the interpolated NMC analysis was primarily responsible for the resulting forecast differences. Comparison of the GLA FGGE and NOSAT initial analyses showed that the FGGE analysis of cloud-track wind data probably underestimated the maximum wind speeds associated with an upper-level jet streak near the east coast. This diminished the effect of the oceanic fluxes in the forecast from the FGGE analysis and resulted in weaker cyclogenesis. |
