For this research, a simple Eulerian measure of precipitation intermittency was developed and applied to Portland, Oregon, National Weather Service WSR-88D observations for the cool season of years 2002 to 2008 (time resolution of approximately 6 min). We define precipitation intermittency as the ratio (PIR) of the number of precipitation transitions occurring to the number of opportunities for transition. The highest PIR indicates the most intermittent precipitation. Precipitation transitions are identified by a well-established and calibrated convective/stratiform identification algorithm, which identifies convective, stratiform, and weak echo regions. When calculated over short time intervals (1-3 hours), a spectrum of PIR can be seen and intra-storm and inter-storm precipitation intermittency comparisons are easily made. Results are dependent on how well the convective/stratiform algorithm is tuned; therefore, a range of tuning parameters is applied to the convective/stratiform algorithm to account for algorithm uncertainty in final results.
Preliminary results indicate that the PIR changes with time and varies among storms. Convection embedded in stratiform precipitation is a primary source of intermittency. Although stratiform precipitation typically spans a larger geographic area during storms, the fraction of precipitation contributed by embedded cellular convection is often higher domain-wide compared to the fraction contributed by stratiform precipitation. Many convective cells originate over the Pacific Ocean, pass over the coastal mountains and dissipate over the Cascades. Varying levels of precipitation intermittency across uneven terrains suggest that orography impacts precipitation intermittency in complex ways. Ultimately, precipitation intermittency is related to the vertical structure and microphysics of overhead clouds; therefore, understanding precipitation intermittency will aid in identifying numerical weather prediction deficiencies.