Horizontal spatial correlation of the ionospheric TEC derived from GPS global ionospheric maps

The spatial correlations of ionospheric day-to-day variability are investigated by statistical analysis on TEC production that provided by the Jet Propulsion Laboratory (JPL). First, we use the monthly-averaged TEC over the world to calculate the deviation of the TEC and then the spatial correlation coefficient matrix of the deviation is also derived. According to the definition of correlation distance in statistics, the spatial characteristic scale is retrieved in the Zonal and meridian directions, and their variation of solar activity levels, geomagnetic field configuration conditions and seasonal conditions are investigated in details. Our study indicates strong correlations between geomagnetic conjugate points, which are large at low-latitudes than at mid-latitudes. Furthermore, in the equatorial anomaly region, due to the process of forming double crest structure, the electron distribution at low-latitudes will be affected by the upwelling of the plasma, thus the anomaly of correlation scale is notable. The correlation scale, which usually increases in LT 6:00, peaks at LT 12:00 and LT 14:00 and then decreases. It is overall larger in the daytime and smaller in the nighttime. These spatial correlations are important for understanding the physical mechanisms that cause ionospheric weather variability. For the potential application in development of ionospheric data assimilation model, the spatial correlations are fitted by elliptic Gaussian formula scale. In general, the correlation scale is regressed better in the Zonal direction than in the Meridional direction, better in low-latitude than in middle-latitude, and better in spring and autumn than in winter and summer. These results can help us to accurately construct the background field error covariance matrix in ionospheric data assimilation.