An attempt to incorporate growth and decay into MAPLE nowcasts /

The term nowcast is used to emphasize that a forecast is being produced for very short lead times, typically 0 to 6 h. Most nowcasting systems use weather radar as the primary tool for forecasting. Modern techniques use computer algorithms to compute these short-term forecasts, and have been shown to have more skill than human forecasters or numerical models alone. Applications of nowcasting span across many industries, with aviation being of particular importance.The essence of nowcasting can be summarized in two steps: (1) obtain a motion estimate for an existing storm, and (2) advect the current storm using the derived motion estimate. Since the inception of weather radar, nowcasting methods have noticeably improved, particularly with respect to estimating storm motion. However, nowcasting is still largely based on Lagrangian persistence, where the forecast field (i.e. reflectivity) is held constant in the Lagrangian frame. This has been done principally because of the difficulties associated with forecasting storm growth and decay.In this paper, an attempt is made to incorporate storm growth and decay into the McGill Algorithm for Precipitation Nowcasting by Lagrangian Extrapolation (MAPLE), by using mesoscale parameters to physically constrain the evolution storm systems. The analysis is conducted on a continental scale, over the contiguous United States. The parameters selected for study were equivalent potential temperature (thetae) and convective available potential energy (CAPE), because they have been shown by theory and observations to be directly related to the intensity and duration of storm systems (Zawadzki and Ro 1978 and Zawadzki et al. 1981, 1994). While past studies were based on observations on a local scale, valid on a day-to-day basis, the current work is applied to a continental scale, on an hour-by-hour basis.Results of the study show little promise for application to nowcasting. No correlations were found between various measures of storm growth/decay and the mesoscale parameters. Several explanations for the results are proposed, which include poor data quality, insufficient sample size and the possibility of a land-surface feedback