Exploring the Range of Uncertainties of Modeled and Observed PBL Heights During the PECAN Field Campaign

The Planetary Boundary Layer (PBL) height (PBLH) plays a crucial role in weather and air quality studies. This study examines the impact of three PBL schemes [Mellor-Yamada-Janjic (MYJ), Yonsei University (YSU), and Mellor-Yamada-Nakanishi-Niino (MYNN)] of the Weather Research and Forecast (WRF) model on the estimation of PBLH. We run WRF simulation during Plain Elevated Convection at Night (PECAN) campaign periods (June to mid July 2015) with two different reanalysis datasets, Global Forecast System (GFS) and ECMWF Reanalysis V5 (ERA5) and compare the results with radiosondes and lidar measurements at four PECAN sites (FP1, FP2, FP3, and FP6). We analyze in particular the uncertainties of ensemble WRF simulations during daytime (convective PBL) and nighttime (stable PBL) separately relative to PECAN estimates. Our results show that the GFS input with MYJ scheme yields the closest PBL height compared with radiosonde measurements but underestimate the daytime PBL height when compared with lidar measurements during daytime at FP1 site. While during nighttime, MYJ with GFS still have the smallest error compared with radiosonde at all sites. We also explore the error covariances of WRF state variables (e.g., θ, Q, wind) on PBLH. Our goal is to quantify the range of uncertainties PBLH estimates and improve the accuracy of PBL simulation through ensemble-based data assimilation.