Deep learning and geochemical modelling as tools for solute geothermometry

Geothermometry is constituted one of the most important geochemical tools for reservoir exploration and development. Solute geothermometers are used to estimate the temperature in the subsurface. Therefore, the chemical composition of a discharging geothermal fluid is used to infer the temperature of the reservoir. Changes in the chemical composition because of boiling, degassing, and dilution are disturbing the equilibrium state within the fluid leading to uncertainties in the temperature estimation. Especially, the pH value, the aluminium concentration, as well as boiling and dilution are parameters prone to changes. These parameters are elaborated in the geochemical modelling process to optimise these values to fit their in-situ reservoir conditions again. This geochemical modelling method can be used for multicomponent geothermometers leading to more robust and precise temperature estimations. However, this process is time-consuming, and geochemical as well as mineralogical knowledge is beneficial. Consequently, the field of artificial intelligence offers powerful methods to solve complex issues, even considering multiple unknowns. ... mehrTherefore, a new solute geothermometer based on a deep learning algorithm is developed. This neural solute geothermometer is tested and compared to the optimised multicomponent geothermometer and in-situ temperature measurements concluding in a new generation of solute geothermometer as precise as an optimised multicomponent geothermometer but much easier and faster in its applicability.