Ensemble Forecasting of Return Flow over the Gulf of Mexico, Value of a Single Upper Air Observation

A case study of return flow over the Gulf of Mexico has been conducted with the intention of determining the value of a single upper-air observation. This case study makes use of a set of upper-air observations collected by the U. S. Coast Guard ship Salvia that followed the trajectory of return-flow air parcels—essentially collecting observations in a Lagrangian frame of reference. A mixed layer model is used to make an ensemble forecast during the outflow phase of the phenomenon—that period when surface-driven buoyancy is the dynamical mechanism that transports moisture and heat into the atmospheric boundary layer. With this low-order nonlinear model, the contributions to forecast uncertainty that stem from initial conditions, boundary conditions, and physical/empirical parameters can be determined separately. These uncertainties serve as input to a three-dimensional variational (3D-Var) data assimilation scheme. Results indicate that the uncertainty in initial conditions dominates the full complement of uncertainties. An experiment is conducted to evaluate the value of a single set of upper-air observations over the Gulf (at a point near the onset of return flow). It is clear from this experiment that observations near the initial onset point significantly improves the forecast of mixing ratio and offers hope for improvement in operational forecasting with a modest increase in resources.