Modeling Thermal Behavior of Geothermal Systems at Mount Meager, Southwestern Canada, Using Artificial Neural Networks and Audio-Magnetotellurics

Precise modeling of subsurface temperatures is crucial for the comprehensive understanding and exploitation of geothermal reservoirs. An artificial neural network (ANN) method was utilized to estimate subsurface temperature by analyzing three-dimensional (3-D) resistivity models derived from audio-magnetotelluric (AMT) data and temperature logs from the Mount Meager Volcanic Complex (MMVC), southwestern British Columbia, Canada. A multi-layer perceptron algorithm was utilized to capture the complexity of the data and estimate the subsurface temperature to 3 km. The model was trained with 70% of 1160 data points, validated using the remaining 30%, and fine-tuned based on the data and error analysis. Subsequently, it was tested on three temperature logs that were not part of the training process, to ensure the robustness and reliability of the model predictions. The final model achieved a root mean square (RMS) of 13.1 °C (5% error) and an R2 of 0.97 when estimating subsurface temperature using the training dataset, which is much more promising than using conventional analytical models that show an RMS of 61%. The 3-D temperature model of the MMVC is correlated with available geological data. This methodology offers a cost-effective and non-invasive alternative for the thermal characterization of potential geothermal reserves, providing a powerful tool for resource development.#xD;