A Local Geoid for Egypt's Mediterranean Coast: A Model Based on Artificial Neural Networks

Abstract

The principal goals of this study are to assess five global geoid models, namely EGM2008, EIGEN-6C4, GECO, SGG-UGM-1, and XGM2019e_2159, as well as to create a centimeter-accurate local geoid model for Egypt's Mediterranean coast. To generate the geoid model for the Mediterranean coast, GNSS/leveling points were integrated with the best global geoid model in the region. This process is a cost-effective replacement for the more expensive conventional leveling technique. The Artificial Neural Network (ANN) was created and is now being utilized to do interpolation and statistical geoid height prediction. The findings show that XGM2019e_2159 is the best global model for modeling the geoid surface on the Mediterranean coast, with a Standard Deviation of 14 cm. The findings showed that using this model constructed with the aid of the ANN, it is feasible to determine the geoid height with a Standard Deviation of 3 cm in the Mediterranean coastal areas.