The Background Error Analysis Testbed (BEATBOX) is a new data assimilation framework for box models. Based on the BOX Model eXtension (BOXMOX) to the Kinetic Pre-Processor (KPP), this framework allows to conduct performance evaluations of data assimilation experiments, sensitivity analyses and detailed chemical scheme diagnostics from an Observation Simulation System Experiment (OSSE) point of view. The BEATBOX framework incorporates an observation simulator and a data assimilation system with

more »

... he possibility of choosing ensemble, adjoint or combined sensitivities. A user-friendly, python-based interface allows tuning of many parameters for atmospheric chemistry and data assimilation research as well as for educational purposes, e.g. observations error, model covariances, ensemble size, perturbation distribution on initial conditions, and so on. In this work, the testbed is described and two case studies are presented to illustrate: the design of a typical OSSE experiment, data assimilation experiments, a sensitivity analysis and a method for diagnosing model errors. BEATBOX is released as an open source tool for the atmospheric chemistry and data assimilation communities.