Geomagnetic storms occur because of solar-terrestrial events. These geomagnetic storms can affect human-made technologies, such as satellites, which rely on radio communication through the ionosphere. Global navigation satellite systems (GNSS) are the most important satellites being affected. These are responsible for transmitting signals to GNSS receivers on the ground. Forecasting geomagnetic storms is important for satellite operators to solve radio signal disturbances during this event. This study aims to enhance the geomagnetic storm prediction model. We present the solar and geomagnetic indices analysis for one geomagnetic storm that occurred during the solar cycle 25 to measure the impact of the event on Earth and interpret how they impact the total electron content (TEC) over the Arabian Peninsula region. The Slant TEC was calculated using the collected data from the GNSS receiver at the Sharjah Academy for Astronomy, Space Sciences and Technology (SAASST). Using the gathered data, a prediction model has been developed to predict the disturbance storm time ( \({D}_{st}\) ) index using different machine-learning techniques. The analysis has been performed with the help of data taken from the National Aeronautics and Space Administration-Omniweb data documentation. The best machine-learning technique for training the data was the Gaussian process regression (GPR), and the best model results were for the Rational Quadratic GPR. The best machine-learning technique for testing the data was the ensembles of trees, and the best model results were for boosted trees. Comparing training models and test model results, it is suggested that the boosted tree training model learned from the error and enhanced the test model results.