ADJOINT MODEL APPLICATIONS IN DYNAMIC METEOROLOGY

Adjoint models are powerful, but still underutilized, tools for efficiently estimating sensitivities of aspects of model output with respect to small but significant perturbations of model input. Their potential applications include all problems for which reliable estimates of sensitivity would be useful. In dynamic meteorology, these applications include data assimilation, stability analysis, ensemble forecasting, observation targeting, synoptic studies, and parameter estimation, among others. In order to encourage their use, teach their proper interpretation and development, and provide a public forum for evaluating their works, a series of "adjoint workshops" devoted to these and …