Estimating error covariance and correlation region in UV irradiance data fusion by combining TOMS-OMI and UVMRP ground observations

Surface ultraviolet (UV) observations can be obtained from satellite or ground observations. This paper uses one data fusion technique (similar to Kalman filter) to combine the advantages from both sources of observations, aiming at achieving a better estimate of surface UV. In this paper, new mathematical methods and algorithms were developed to estimate the error covariance and correlation region, which are the most important components in this data fusion technique. This technique was applied to the satellite data from the Total Ozone Mapping Spectrometer (TOMS)-Ozone Monitoring Instrument (OMI) combined with ground measurements from UV-B Monitoring and Research Program (UVMRP) within the region of continental U.S. from 2005 to 2015. Numerical experiments showed that the technique is effective, and TOMS-OMI data were improved by combining UVMRP data. In addition, the innovative ensemble-based method is generic and can be applied to other fields for data fusion/assimilation.