A simple method for integrating a complex model into an ensemble data assimilation system using MPI

5 This paper details a strategy for modifying the source code of a complex model so that the 6 model may be used in a data assimilation context, and gives the standards for implementing a data 7 assimilation code to use such a model. The strategy relies on keeping the model separate from any 8 data assimilation code, and coupling the two through the use of Message Passing Interface (MPI) 9 functionality. This strategy limits the changes necessary to the model and as such is rapid to program, 10 at the expense of ultimate performance. The implementation technique is applied in different models 11 with state dimension up to 2.7 × 10. The overheads added by using this implementation strategy 12 in a coupled ocean-atmosphere climate model are shown to be an order of magnitude smaller than 13 the addition of correlated stochastic random errors necessary for some nonlinear data assimilation 14 techniques. 15