Abstract: In order to develop a fusion dataset of global multi-source daily sea ice concentration, the following scheme is adopted to combine observations from various sources and model background field data. The observational data sources include the daily sea ice concentration analysis from OSI SAF (Ocean Sea Ice Application Center) EUMETSAT (European Meteorological Satellite Application Organization), the global sea ice concentration data of MWRI (Micro Wave Radiation Imager) and VIRR (Visible and Infrared Radiometer) from NSMC (National Satellite Meteorological Center) of China, the NISE (Near-Real-Time SSM/I EASE Grid Daily Global Ice Concentration and Snow Extent）sea ice concentration data of U.S. NSIDC (National Snow and Ice Data Center), the IMS (Interactive Multisensor Snow and Ice Mapping System) Northern Hemisphere sea ice data of U.S. NIC (National Ice Center). The model background field is the ERA-Interim sea ice concentration data. First, complete quality control is performed on all the datasets from various sources. Next, the sea ice concentration data of the OSI SAF is used as a reference, and the probability density function (PDF) matching method is used to correct systematic errors in other satellite data. The super observation is then generated based on adjusted error. Finally, the super observation and ERA-Interim sea ice concentration data (the background field) are combined using the STMAS (Space-Time Multiscale Analysis System) method to generate 1-year merged global daily sea ice concentration data with the resolution of 0.25°. The quality of the merged product is evaluated by comparing with the sea ice concentration data widely used internationally, e.g., the OISST (Optimum Interpolation Sea Surface Temperature) and OSTIA (Operational Sea Surface Temperature and Sea Ice Analysis). The results show that the PDF method can make the probability density distribution of non-reference data consistent with that of reference data by adjusting the probability density distribution of non-reference data, and the adjusted systematic errors of the three sea ice concentration satellite datasets are significantly reduced in the merged solution. The STMAS method can effectively combine the super observation and the background to generate a merged product. The use of FY（FengYun ） Satellite data improves the autonomous and controllable ability of the fusion data. In addition, the timeliness of satellite data sources, the stability of acquisition and other factors are considered in the fusion scheme. The spatial distribution of the merged product is highly consistent with the OISST and OSTIA sea ice concentration data in the Arctic and Antarctic, and the correlation coefficients between them exceed 0.985. The deviations of the merged product from the OISST and OSTIA are −1.170% and 0.276%, respectively. The overall bias of the fusion experiment product is between those of the OISST and OSTIA, which reflects the good quality of the fusion product that shows small systematic errors. It can be seen that the fusion scheme can meet the operational need and the fusion experiment product has a high quality.