Evaluation of an operational ice-ocean analysis and forecasting system for the Gulf of St Lawrence

Our ability to simulate and understand oceanic conditions of the Gulf of St Lawrence (GSL) has significantly increased in the last decade with the development of regional ocean models combining scientific knowledge, oceanographic data, and computing techniques. This has created opportunities for their integration with meteorological forecast models where the importance of sea iceocean dynamics in airsea exchanges was recently demonstrated. In order to produce accurate initial conditions for coupled weather forecasts, a GSL iceocean pseudo' analysis (IOPA) system has been developed and implemented operationally at the Canadian Meteorological Centre (CMC). Based on prognostic iceocean model solutions the IOPA system has sufficient skill to produce realistic daily fields without flux correction or data assimilation of temperature and salinity. Using only a simple assimilation method for sea-ice data, daily oceanic fields are obtained and can be used to initialize iceocean or coupled atmospheric forecasting systems. A 3-year retrospective evaluation of this operational method is presented focusing on sea-surface temperature (SST) and sea-ice conditions. Model estimates are compared with in situ and satellite-derived data. Results show that the IOPA system has an accuracy equivalent to or better than the current operational SST and ice analyses produced at the CMC. Together with preceding modelling efforts, this work lays the foundation for the first state-of-the-art GSL ocean analysis and forecast systems combining ocean modelling and real-time three-dimensional data assimilation. Coupled atmosphereiceocean forecasts produced with the same iceocean model and initialized with the IOPA system are presented in a companion paper. Copyright (c) 2012 Crown in the right of Canada. Published by John Wiley & Sons Ltd.