In order to derive boundary conditions, atmospheric and seismo-ionospheric models are coupled with generic assumptions. These boundary conditions are used to simulate possible interaction between earth’s atmospheric, geological processes and the Ionosphere. Rn is one of the major contributors to surface ionization, which is also believed to contribute to the geophysical and geochemical processes causing disturbance in the Ionosphere. Its relationship with metrological parameters in a given region gives an insight on the natural processes in the underground, as well as the relationship with atmospheric processes. It can enable models to establish realistic results rather than ideal theoretical consequences. Rn can be influenced by a number of physical variables, and therefore, this influence is sometimes very complicated to study, because of the Forcing researchers to make ideal assumptions. The relationship between Rn and some geological variables are studied, namely; soil temperature at 5 cm, 10 cm, 20 cm, and 50 cm, atmospheric pressure, and atmospheric temperature. A hybrid model is established based on the artificial neural network (ANN), which is referred to as multiANN model. This model is a combination of multi-regression and ANN models. Itenables Rn prediction to metrological parameters. To test the robustness of our model 50% training periods is employed with 50% testing periods. The model is able to forecast the remaining 50% effectively. With the aid of the Monte-Carlo method, it is possible to predict multiple future Rn variations with high precision. The regions with low performance of the multiANN are identified for possibly relationship to seismic events. The model could be a good candidate for predicting of Rn concentration from metrological parameters, especially in establishing the lower boundary conditions in seismo-ionospheric coupling models.