This study investigates the calibration of accelerometer data for the Next Generation Gravity Mission (NGGM), proposed by the European Space Agency. With improved precision, NGGM aims to continue gravity field observations beyond the Gravity Recovery and Climate Experiment Follow-On (GRACE-FO) mission. The mission consists of a satellite pair measuring Earth's gravity field using an onboard laser tracking instrument. To isolate the gravity field signature in these observations, each satellite carries accelerometers to estimate non-gravitational accelerations. This thesis supports the accelerometer calibration process by applying lessons from previous gravity field missions.

A historical review of gravity missions highlights the evolution of scientific and hardware requirements. The study examines accelerometer principles, sources of instrumental imperfections, and existing data calibration techniques. NGGM’s preliminary design includes multiple accelerometers placed away from the satellite’s center of mass, allowing the use of shaking manoeuvres—first introduced in the GOCE mission—for calibration.

A comprehensive model is developed that can generate shaking manoeuvres with varying thrust magnitudes, shaking durations, and shaking frequencies to excite the satellite. This model is used in conjunction with various accelerometer units (two, three, and four accelerometer layouts are considered) and their placement in the satellite's body frame to evaluate the calibration quality against the scientific requirements posed for the mission.

Results indicate that along-track accelerometer placement minimizes non-gravitational acceleration measurement errors due to enhanced centrifugal acceleration from the satellite’s pitch rate during calibration. Furthermore, the along-track placement performs better than radial placement, even though it has the same centrifugal acceleration boost. The suspected cause is the electrode layout of the accelerometer, which boosts the acceleration signal due to the projection of the angular acceleration about the y-axis onto the z component of the linear acceleration. The radial placement of the accelerometers provides no additional signal to the x component of the linear acceleration due to a lack of projection. Lower shaking frequencies improve calibration by accumulating higher angular rates over time. However, due to volume constraints imposed by the laser tracking instrument, cross-track placement may be more favourable. This configuration requires higher thrust levels, as the absence of a pitch rate signal on the cross-track axis worsens the signal-to-noise ratio of the observations, which warrants a revision of the thruster requirements and accelerometer performance. Moreover, more than two accelerometers reduce measurement errors by providing redundancy in the observations. Even with three accelerometers placed in the along-track direction, at least 24 hours of shaking at maximum thrust, as stated by the thruster requirement, is required for effective calibration. Lower thrust or shorter shaking durations would necessitate four accelerometers—two on the x-axis and two on the y-axis. Finally, the accelerometer pair’s arm length is treated as a free variable, as it has minimal impact on calibration performance.

This report provides foundational insight for future gravity missions. Smart accelerometer placement and shaking manoeuvre parameters can improve the measurement quality of the non-gravitational forces and subsequently improve gravity field recovery, which is crucial for tackling the climate crisis.