The Impact of Assimilating Dense Smartphone Pressure Observations on a Hailstorm Simulation

Recent studies have shown that smartphone barometric pressure observations (SPO) can capture mesoscale surface pressure features associated with convective storms; however, the potential of these data to enhance weather forecasting skill remains uncertain. This study investigates the impact of assimilating SPO data and traditional weather station (TWS) pressure data on the 30 June 2021 Beijing hailstorm using an hourly updated 3DVAR (three-dimensional variational data assimilation) system in the WRF (Weather Research and Forecasting) model. Three assimilation experiments were performed: a TWS-based experiment (STA); an SPO-based experiment with constant observation error variance (PHO-C); and an SPO-based experiment with real-time observation error variance (PHO-R). Compared to the control run without data assimilation (CNTL), all assimilation experiments yielded a more accurate representation of the hail spatial distribution and improved the simulation of the cold pool and gust front. The fractions skill scores for radar-derived MESH (maximum estimated size of hail) increased by 14%–17% in SPO runs, which was larger than the improvement obtained with TWS assimilation. Moreover, the PHO runs outperformed the STA run in correcting the cold pool deviation in the CNTL run. Among all simulations, the gust front position in the PHO-R run was closest to the observations. Based on this specific case, these findings suggest that high-resolution SPO data have the potential to improve the forecasting of mesoscale convective systems, achieving results comparable or even superior to those obtained with TWS data.