The Nowcasting sub-system of SCRIBE (Weather Forecast Product Expert System) has been recently implemented operationally in all forecast offices across Canada. The sub-system ingests the latest observations and nowcasting model data and allows forecasters to update interactively, in real time, the Scribe weather elements. This sub-system has been developed to minimize these necessary manual adjustments done by the forecaster to merge the current weather conditions with the forecast.

Version 1.0 of the Scribe Nowcasting is currently using surface observations, North American radar mosaic data and the lighting data from the Canadian Lighting Detection Network. These observations are used to feed three different nowcasting models. The statistical model "PubTools" (CMC - P. Bourgouin) uses surface observations to predict weather elements. The radar reflectivities are forecasted in the next 6 hours with an algorithm developed by McGill University. Finally, and algorithm has been designed at CMC (M. Ouellet) to predict the future position of the lightning clusters. All these observed and forecasted data are processed into a rules base system to determine the most likely sequence of weather elements. Thus, the first 6 to 9 hours of the regular Scribe weather elements will be influenced by the nowcasting data.

To assess whether these changes to the regular Scribe Weather elements contribute to improve the first hours of the forecast or not, objective verifications were performed on all elements for the first 6 hours of the forecast. In general, the Probability of Detection for precipitation occurrence and types increases and the related False Alarm Ratio decreases. Other verification scores for winds, temperatures and clouds also indicate a significant improvement. The addition of the Nowcasting sub-system to Scribe greatly improves the first hours of the forecast.