3:30 PM - 4:30 PM
[G03-P-03] GNSS-based calibration of GRACE accelerometers
In this study, two approaches for the determination of total non-gravitational accelerations acting on GRACE were investigated. To investigate two approaches, GRACE Level-1B data including accelerometer sensor data, GNSS based positions and star camera data was used. In the first approach, usual surface accelerations acting on the satellite were taken into account. In a second approach, the total acceleration, consisting of a gravitational and non-gravitational contribution, was first determined from the GNSS-based precise orbit data by means of smoothing polynomial and differentiation. Then the gravitational contribution determined by state of art models was removed to determine the non-gravitational part.
The results of these two methods were used to calibrate the linear acceleration measurements on-board GRACE. Daily constraint-free biases and scale factors were determined for two time periods including three month data with a different solar activity. The quality of the results, especially of the second approach is highly dependent on the solar activity and therefore the magnitude of the non-gravitational reference acceleration. The results show that calibration parameters for along track direction can be determined more precise without any information about satellite properties and atmospheric conditions. Constraints have to be applied to get more realistic calibration parameters for axes with a smaller contribution.
The results of these two methods were used to calibrate the linear acceleration measurements on-board GRACE. Daily constraint-free biases and scale factors were determined for two time periods including three month data with a different solar activity. The quality of the results, especially of the second approach is highly dependent on the solar activity and therefore the magnitude of the non-gravitational reference acceleration. The results show that calibration parameters for along track direction can be determined more precise without any information about satellite properties and atmospheric conditions. Constraints have to be applied to get more realistic calibration parameters for axes with a smaller contribution.