Implementing the Latent Heat Nudging (LHN) Algorithm in the Canadian Regional Deterministic Prediction System (RDPS)

This study reports on an implementation of the Latent Heat Nudging (LHN) algorithm in the Canadian Regional Deterministic Prediction System (RDPS). This system operates at a resolution of 10 km over a domain than encompasses the North-American continent. For this work, we assimilate reflectivity composites constructed from the volume-scans measured by all the radars in Canada, the continental US and Alaska.

With expected increases in the spatial and temporal resolution of current assimilation systems, the assimilation of radar data is envisioned to play a role in improving the quality of short-term precipitation forecasts. This research project is also motivated by the upcoming replacement of the actual Canadian C-band radars by S-band radars.

It is the first time that radar data are being assimilated in a Canadian operational forecasting system. For this first step, the LHN technique has the advantage of being well known and relatively easy to implement. The system developed here is intended to provide a benchmark against which more advanced assimilation techniques are to be compared in the future.

Examples of forecasts with and without LHN will be given. Through these examples, we will illustrate the inner workings of the LHN algorithm during the nudging period. The lifetime of the improvements at different scales will be discussed along with a comparison with precipitation nowcasts obtained with the MAPLE algorithm.