ABSTRACT：A long-term ocean gridded observational dataset (with complete ocean coverage) is crucial for a wide range of applications, including climate change, oceanography research and operational applications. Based on the domestic high-performance computing environment, the team designed and implemented the whole process construction framework of ocean observation gridded dataset, realized the reconstruction of data from in-situ data to gridded data, and applied it in the construction of IAPv4 ocean temperature and heat content gridded dataset. This framework has the following advantages: (1) The framework has been deployed based on the domestic high performance computing environment, which can realize the (near) real-time calculation and update of datasets; (2) The framework can integrate a variety of data processing schemes (such as quality control schemes, deviation correction schemes, vertical interpolation schemes, climate states and telephone schemes, etc.), which can accurately analyze the uncertainty of observation data and optimize the deployment of observation systems; (3) Beyond ocean temperature and heat content, the framework can also be used to construct grid datasets for temperature, dissolved oxygen, and other observational data; (4) The framework can integrate ocean observation data from different data sources (WOD, GTSPP, or private data) to build high-quality grid data sets with global coverage and improve the utilization of existing observation data.
 

ocean gridded dataset, framwork, ocean temperature, high performance computing