Regional Gravity Field Modelling From GRACE Satellite Gravity Mission Data: A Case Study in Ethiopia
No Thumbnail Available
Date
2022-02
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Addis Ababa University
Abstract
The Gravity Recovery and Climate Experiment (GRACE) mission data-derived gravity field
models have provided us with remarkable accuracy in gravity field determination. Despite this, the
predicted GRACE baseline accuracy has yet to be met. One reason, among many, could be
insufficient satellite data modeling using a global representation by Spherical Harmonics (SH). It
appears appropriate to improve global solutions through regional recovery strategies to fully
recover the signal information included in satellite and sensor data. Varied geographical regions
have different gravity field characteristics, especially in the higher frequency part of the spectrum.
As a result, the recovery process should be tailored to the specific characteristics of the area. This
regional refinement strategy was employed in Ethiopia. Global gravity reference field, which is
represented by SH expansion is then refined using regionally tailored refinements that are specified
as space localizing basis functions using splines. The shape coefficients are obtained using the
global gravity model's degree error variances truncated up to a maximum degree of 120. This
means that the spline kernels must be adjusted to the spectral range of the gravity field features to
be calculated. In addition, the raw accelerometer data measured by GRACE Satellites using
onboard accelerometers was calibrated. Every short arc over the study area is used, with a 10° strip
added to the recovery area. After being synchronized, these short arcs created a partial system of
normal equations using an integral equation of the Fredholm type. These normal equations are
accumulated to account for varied accuracies of short arcs by estimating variance factors for each
arc using variance component estimation (VCE). For only one month of data collected in January
2016, the normal equations are solved using the least squares regularized approach. In the case of
regularization, the regularization parameter was calculated using the same VCE method. Finally,
the study area's regional refinement was determined a using geographic grid of 0.5° ��0.5° at sea
level in Ethiopia. In terms of gravity anomaly and geoid height, the regional solution improves the
Global gravity reference field in root mean square (������) from 5.95�������� to 4.33�������� and
52.90���� to 46.13���� respectively.
Description
Keywords
GRACE, Regional Gravity Field Recovery, Space Localizing Basis Functions, VCE