WHAT IS AN ADJOINT MODEL

Adjoint models are powerful tools for many studies that require an est imate of sensitivity of model output (e.g., a forecast) with respect t o input. Actual fields of sensitivity are produced directly and effici ently, which can then be used in a variety of applications, including data assimilation, parameter estimation, stability analysis, and synop tic studies. The use of adjoint models as tools for sensitivity analys is is described here using some simple mathematics. An example of sens itivity fields is presented along with a short description of adjoint applications. Limitations of the applications are discussed and some s peculations about the future of adjoint models are offered.