Assimilation of the ceilometer observed planetary boundary layer height into WRF-Chem/DART

Abstract:

The planetary boundary layer height (PBLH) is vital in the air pollution study to determine the vertical mixture of air pollutants. Nevertheless, the current operational air quality is poor and the inaccurate PBLH forecast is one of the uncertainty factors. A ceilometer network is being built over the continental U.S. supported by NASA. With the ceilometer retrieved PBLH, this study evaluates the different model performance on forecasting the PBLH from the most advanced PBL parameterization schemes. Then data assimilation techniques are employed to assimilate observed or retrieved PBLH into the model. The objectives of this study are to connect and evaluate the PBLH calculation in model with observation, and to test a data assimilation scheme with observed PBLH into a forecast model. We employed the Weather Research and Forecasting model coupled with Chemistry (WRF-Chem) to conduct the numerical experiments. First, we used the ceilometer retrieved PBLH to evaluate PBLH forecasting from WRF-Chem. After evaluating the PBL parameterization schemes performance, we initiated two approaches for the assimilation of the observed PBLH into WRF-Chem model to assess improvements of the model performance on forecasting PBLH. The assimilation tool employed in this study is the Data Assimilation Research Testbed (DART), developed by NCAR. Since PBLH is not a prognostic variable, it could not be assimilated directly into WRF-Chem using WRF-Chem/DART. We considered two proposed assimilation methods. One, to find the relationship between observed PBLH and other related prognostic variables and construct a covariance matrix between them. The second method is based on the assimilation of the ceilometer observed aerosol backscattering profile into the model aerosol backscattering derived parameterization, then using the machine learning method (edge detection) to find the assimilated PBLH. We will present results of both works.