Ensemble-Based Observation Targeting for Improving Ozone Prediction in Houston and the Surrounding Area

This study examines the effectiveness of targeted meteorological observations for improving ozone prediction in Houston and the surrounding area based on perfect-model simulation experiments. Supplementary observations are targeted for the location that has the highest impact factor (maximum Kalman gain) estimated from an ensemble and is expected to minimize ozone forecast uncertainty at the verification time. It is found that the observational impact factor field varies with time and is sensitive to ensemble resolutions and physics parameterizations. The efficiency of observation targeting is further examined through assimilating observations in areas with different impact factors using an ensemble Kalman filter. It is found that the ensemble sensitivity analysis is capable of locating supplementary observations that may reduce meteorological and ozone forecast error, but not as effectively as expected.