Four-dimensional variational data assimilation applications to the field experiments

The intensive field experiments provide a large amount of observations and thus make the ideal testbeds for the application of four-dimensional variational data assimilation technique (4D-Var) with Chemical Transport Models (CTMs). In the current application, The STEM-2K1 (Sulfur Transport Eulerian Model, version 2K1) and its adjoint model were applied to assimilate observations from different platforms during ICARTT (International Consortium for Atmospheric Research on Transport) and TRACE-P (TRansport And Chemical Evolution over the Pacific) field experiments. We show here that extensive sensitivity analyses can be obtained effectively through the adjoint model. Reanalysis data after assimilating various observations were generated. If evaluated by the independent measurements that were not used in the assimilations, the reanalysis data demonstrate a better match to the observations. A real-time surface ozone forecast framework was tested during the ICARTT field campaign. It shows better forecast results after using the 4D-Var reconstructed initial ozone concentrations. In addition, top-down estimations of emission inventories using the 4D-Var technique are presented and discussed.