Ionospheric Vertical Correlation Analysis and Modeling Using COSMIC-2 Global Ionospheric Specification Data

The background error covariance matrix is essential in ionospheric data assimilation, enabling the integration of observational data with model predictions by defining spatial relationships between grid points. The vertical component plays a key role in enhancing model accuracy by capturing altitude-dependent spatial correlations, directly influencing electron density profile reconstruction. In this study, we used Constellation Observing System for Meteorology, Ionosphere, and Climate-2 Global Ionospheric Specification data to analyze vertical correlation distances in the ionosphere, identifying the altitude range where the correlation coefficient exceeds 0.75 in both upward and downward directions. The results revealed a clear pattern: vertical correlation distances initially increased with altitude, then the rate of increase slowed or decreased before rising again. Significant variations were observed with local time (LT), geomagnetic latitude, and seasonal effects. Winter anomalies, possibly due to changes in the O/N₂ ratio, amplified vertical correlations, particularly between LT 12:00 and 18:00. A layered Gaussian model was developed to parameterize the vertical correlation coefficients, and its accuracy and adaptability were validated across different seasonal conditions. The model's correlation with observational data was greater than 0.75 across all seasons, with relatively small errors at lower altitudes, typically within 50 km. At reference heights exceeding 500 km, the error range remained within approximately 100 km for most points. These findings contribute to the development of more accurate ionospheric models, which are essential for satellite communication, navigation, and space weather forecasting.