Estimates of boundary layer profiles by means of ensemble-filter assimilation of near surface observations in a parameterized PBL

In-situ surface-layer observations constitute a valuable data source if utilized effectively in NWP applications. If properly assimilated, data from existing mesonets could improve short range forecast within the PBL. A parameterized 1-D PBL model and an ensemble Kalman filter (EnKF) data assimilation approach are used to test the effect of surface observations assimilation on the estimate of PBL atmospheric profiles. Surface observations measured over Oklahoma during the Bow Echo and Mesoscale Convective Vortex Experiment (BAMEX) period were assimilated into the system described above. Initial conditions, large-scale forcing, and surface radiation are imposed by climatologic profiles that were calculated from WRF real-time forecasts run during the same period. Therefore, our experiments test the impact of the surface observations on the atmospheric profile estimation when no profile data other than climatologic is available.

Comparison of our results to WRF analyzed and forecasted profiles (at a horizontal resolution of 4 km) shows that the EnKF assimilation of surface observations into the 1-D PBL model has a positive impact up to a height of 1000 m in certain regimes. As part of our ongoing work we are investigating the use of observed or mesoscale model forecasted profiles to initialize and force the 1-D PBL model, i.e, we provide information about the initial atmospheric profile and large scale forcing closer to the true state of the atmosphere. The results of our work encourage the use of the present method as nowcasting and forecasting tool to estimate high vertical resolution PBL profiles where surface observations are available.