The Potential of Smartphone Pressure Observations Evaluated in a Convective Case Study

Pressure observations obtained from smartphones could vastly increase the density of available surface weather observations and may offer substantial improvements to mesoscale forecasts.  Here, currently-available smartphone observations (and derived 1-hr smartphone pressure tendencies) are considered over a 3-day, convectively active period in the eastern United States.  A series of observing system experiments is performed, using a 50-member, mesoscale ensemble based on the High Resolution Rapid Refresh (HRRR) configuration.  Smartphone pressure observations and 1-hr tendency observations are assimilated hourly using an ensemble Kalman filter, and the performance of subsequent forecasts is evaluated.  The smartphone observations are found to have considerable noise despite gross quality control checks, and this limits the scales at which they provide usable information.  Nevertheless, some improvement is seen in short-term forecasts of surface pressure, 10-m winds, and mesoscale areas of precipitation.  Little improvement is seen on the convective scale.  However, the results suggest that with more rigorous quality control and by increasing density from additional smartphones whose observations are not yet available, smartphone pressure observations could capture valuable and timely mesoscale information to support rapid-update forecasts.