Bias estimation for near-surface observations in ensemble Kalman filter with WRF model over complex terrain

Near-surface temperature, moisture, and wind impact air quality by modifying gas-phase chemistry, transport, removal, and natural emissions. Numerical weather prediction (NWP) models are essential in political strategies aimed at improving air quality, but they require confidence in model simulations. The accuracy of NWP is limited by both model inadequacies and effective use of surface observations in the data assimilation process. A relevant but often overlooked source of deterioration in NWP is the assimilation of biased observations, which cause an under- or over-estimation of a temperature, moisture, or wind posterior state.

The aim of this work is to use data assimilation to estimate the bias of individual in-situ observations, with a methodology that has been tested before in low-order (simple) models. The approach is based on parameter estimation using an augmented state in an ensemble adjustment Kalman filter. This methodology is applied by coupling the Weather Research and Forecasting (WRF) NWP model with the Data Assimilation Research Testbed (DART). The domain encompasses a region with complex terrain over the western U.S. where the MATERHORN field campaign was carried out during fall 2012 (25 September to 21 October). The data used for assimilation include radiosonde, aircraft, satellite winds, METARs, and surface mesonets. Observations are assimilated every 3 hours using a 96-member ensemble on a 30-km domain. Biases are estimated for mesonet observations of wind components at 10 m and temperature at 2 m. Results show that this methodology is able to estimate observation bias with some limitations. These systematic errors are more relevant in areas of complex terrain, mainly due to representativeness errors.