The possibility of earthquake forecasting with ionospheric total electron content and radon-222 concentrations using Monte Carlo simulations

This study was handled and discussed in two different dimensions in order to make future earthquake forecasting. First, the role of Radon-222 inert gas as an earthquake prediction precursor is discussed, and the factors affecting soil Rn concentration are examined in a multidisciplinary context in physics, artificial intelligence, geology, and geophysical frameworks. Determining the seismo-ionospheric changes of soil Rn-222 gas requires complex physical and mathematical calculations and forecasting. Secondly, the Total Electron Content (TEC), one of the ionospheric precursors of earthquakes, was discussed. Our aim here was to obtain solid physical evidence for earthquake-based gas and ion exchanges between the earth and the atmosphere; therefore, radon, atmospheric pressure, atmospheric temperature, soil temperature and their interactions were examined at various soil depths to see the physical changes in the journey of the above-mentioned variables from the underground to the Earth's atmosphere (Ionosphere). For all variables, three Autoregressive Integrated Moving Average (ARIMA) models were proposed. Then, the main statistical characteristics of the data were determined and future values of the variables were predicted using Monte-Carlo Simulations (MCSs). In addition to all these, Multiple Linear Regression and Multiple Artificial Neural Network models were obtained for Radon-222 using certain meteorological parameters. While obtaining all these models, as many variables as possible were used to increase the reliability of the calculations and the results were widely discussed within the framework of the relevant field.