Correlation between Speed of Rain Storms and Temporal Properties of Precipitation: Applications to the Synthetic Storm Technique

We show that many properties of rain rate time series of a 10-year data bank, such as duration, length, average rainfall, total accumulation of water, variance, maximum rainfall rate, time and space decorrelation are statistically uncorrelated with the speed of rain storm, measured at a site 193 km from the raingauge, located at the McGill Observatory in Montreal (Canada). A given rain event can thus move with a different speed than that measured, extracted randomly (Monte Carlo simulation) from the probability distribution of speed, modelled as log-normal. The simulation, conducted with the synthetic storm Technique to transform rain rate time series into rain attenuation time series at 18.7 GHz in a 37.7deg slant path, shows that 3 years of rain rate data (not necessarily consecutive) can estimate long-term (10-years) rain attenuation (dB) with an overall RMS less than 7%.