Numerical weather prediction models often fail to correctly forecast convection initiation (CI) at night in the Great Plains. To improve our understanding of such events, researchers collected unique observations from thermodynamic and kinematic profilers as part of the Plains Elevated Convection at Night (PECAN) experiment. The assimilation of these observations can aid in analyses of key features that are not easily observed by conventional datasets This talk presents forecasts of a nocturnal CI event from PECAN in which assimilating the PECAN dataset improves the timing, location, and orientation of CI.
Day-ahead (20 – 22 hour) 3-km grid spacing convection-allowing model forecasts are performed for a severe hail event on 8 May 2017 using six different multi-moment microphysics (MP) schemes including: the Milbrandt and Yau double-moment (MY2), Thompson (THO), NSSL double-moment (NSSL), Morrison double-moment graupel (MOR-G) and hail (MOR-H), and Predicted Particle Properties (P3) schemes.
In a previous study, Chen and Snyder (2007) showed that a large location error in the background forecast can result in a poor performance of ensemble Kalman filter (EnKF) data assimilation (DA) due to the violation of the Gaussian assumptions. One way to alleviate this issue is to apply vortex relocation (VR) before the ensemble-based DA.
Machine learning (ML) has become a widely used tool in meteorological research and often makes better predictions than thecompeting methods. Despite this, many meteorologists are reluctant to adopt ML in day-to-day operations, due to the per- ception thatthe models are “black boxes” (cannot explain their decisions to the user). My recent work in ML interpretation […]