MAPLE McGill Algorithm for Precipitation Nowcasting by Lagrangian Extrapolation is an algorithm developed at the JS Marshall Radar Observatory for very short term forecasting (nowcast) of precipitation. Here, we apply this algorithm to composite radar maps over the Southern Quebec and the Northeastern US to obtain precipitation forecasts for the next three hours.
Because precipitation varies dramatically in space and time it presents a major challenge in weather forecasting. Unlike Numerical Weather Prediction models, which use a variety of complicated equations to forecast precipitation, MAPLE simply uses the motion derived from prior radar observations to estimate the future location of precipitation. Because they rely solely on past data, MAPLE forecasts become less accurate with time.
MAPLE defines three precipitation types – rain, snow and the transition from rain to snow – based on how temperature changes with height. The menu on the right allows you to control the animation and to display additional information such as maps of motion and temperature.
The MAPLE algorithm developed by Germann and Zawadzki (2002) uses the Variational Echo Tracking technique (VET, Laroche and Zawadzki 1995) to estimate the motion field of precipitation and a modified semi-Lagrangian backward scheme for advection. For more information see Predictability of Precipitation from Continental Radar Images in the Publications section.