Impact of Ground Based Remote Sensing Boundary Layer Observations on Convective-Scale NWP

Due to the lack of high spatial and temporal resolution boundary layer observations, the rapid temporal and spatial changes in the near storm environment are not well represented in current convective-scale numerical weather prediction (NWP) models. Better representation of the near storm environment in the model initial conditions will likely further improve the forecast of severe convective weather. During the Plains Elevated Convection at Night (PECAN) field campaign from 1 June to 15 July 2015, two ground-based remote sensing instruments namely Atmospheric Emitted Radiance Interferometer (AERI) and Doppler Lidar were deployed at six fixed sites in central Great Plains. AERI measures the downwelling infrared spectral radiances from which the thermodynamic profiles in the boundary layer up to the cloud base can be retrieved, and Doppler Lidar measures the horizontal wind speed, direction and vertical velocities through the boundary layer. The combination of both instruments provides high-temporal resolution kinematic and thermodynamic profiles of the boundary layer. This study investigates the impact of assimilating AERI and Doppler Lidar retrievals on convective-scale NWP. Two sets of ensemble data assimilation and forecast experiments are conducted, one experiment assimilates routinely available metar, radiosonde, acars, satellite derive winds and marine platforms and another experiment assimilates Doppler Lidar and AERI retrivals in addition to the routinely available observations. Short-term 0-6 h ensemble forecasts are launched from the 3-km convective-scale ensemble. Results indicate that the positive impact of AERI and DL observations are more pronounced during the first ~3 hour of the forecasts and will likely improve convective initiation.