Combining ocean models and observations provides an analysis of the time-varying ocean state that must necessarily be more complete and more accurate than the information from either of them alone. The goal of this project is to combine the dynamics of an eddy-resolving configuration of the MIT general circulation ocean model (MITgcm) and all available observations in the Red Sea to estimate as accurately as possible the circulation and the variability of this region. We are using two four-dimensional assimilation systems: a 4DVAR system based on the MITgcm and its adjoint as it has been developed by the Estimating the Circulation and the Climate of the Ocean (ECCO) consortium, and an ensemble Kalman filtering system based on the MITgcm and the Data Assimilation Research Testbed (DART). Outputs from these systems will be used to study and analyze the variability of the Red Sea in a way that has not been possible before. We are interested in studying water mass exchange with adjacent water bodies and deep circulation, seasonal and interannual overturning circulation, eddy variability, and better understanding air-sea interactions in this region. The two assimilation systems will also provide an advanced setting to develop an operational system for the Red Sea circulation.