Estimation of global thermal conditions via cosimulation of machine learning outputs and observed data

A non-transitory, computer-readable medium having instructions stored thereon that are executable by a computing device to perform operations comprising:

• for each supervised learning model in a plurality of supervised learning models, train the supervised learning model to estimate a plurality of heat flow values based, at least in part, on a plurality of preprocessed geological parameter values that at least include a plurality of observed heat flow values;
• determine which of the plurality of trained supervised learning models most satisfies a corresponding model selection criterion;
• and
• based on a first of the plurality of trained supervised learning models most satisfying the corresponding model selection criterion, cosimulate the estimated plurality of heat flow values of the first trained supervised learning model and the plurality of observed heat flow values to generate a cosimulated heat flow map.

An apparatus comprising:

• a processor;
• and
• a computer-readable medium having instructions stored thereon that are executable by the processor to cause the apparatus to,
• for each supervised learning model in a plurality of supervised learning models, train the supervised learning model to estimate a plurality of geothermal gradient values based, at least in part, on a plurality of preprocessed geological parameter values that at least include a plurality of observed geothermal gradient values;
• determine which of the plurality of trained supervised learning models most satisfies a corresponding model selection criterion;
• and
• based on a first of the plurality of trained supervised learning models most satisfying the corresponding model selection criterion, cosimulate the estimated plurality of geothermal gradient values of the first trained supervised learning model and the plurality of observed geothermal gradient values to generate a cosimulated geothermal gradient map.