A semi-vectorized and relationally-operated algorithm for fast geoid computation using Stokes’s integration
Document Type
Article
Publication Date
9-1-2022
Abstract
We propose an optimization algorithm to speed up the computation of a high-frequency geoid based on the surface integral of Stokes’s formula. Stokes’s integral computes high-frequency geoid undulations from the terrestrial gravity data. The global geopotential model (GGM) provides the respective low-frequency component of the geoid solution. The combination of high and low-frequency geoid solutions with the associated corrections represents the final geoid model. However, the practical implementation of Stokes’s integration is time-consuming over dense gravity grids and needs to be optimized efficiently. Therefore, the vectorization techniques and relational operators have been utilized in Matlab®; and Octave to the loops and conditional statements. The efficiency of the presented algorithm is evaluated and tested on different gravity grids and capsize values before and after the optimization. The total run time (TRT) of the original and modified Stokes’s integral is employed to assess the efficiency of the algorithm. Finally, our algorithm shows high performance in all tested scenarios. It reduces the TRT by 7 times on 1 arc-min resolution grid with varying capsize values from 0.25∘ to 1.5∘.
Publication Source (Journal or Book title)
Earth Science Informatics
First Page
2017
Last Page
2029
Recommended Citation
Abdalla, A., & Ferreira, V. (2022). A semi-vectorized and relationally-operated algorithm for fast geoid computation using Stokes’s integration. Earth Science Informatics, 15 (3), 2017-2029. https://doi.org/10.1007/s12145-022-00822-7