Temperature predictions at 5 kilometers depth based mostly on the Stanford Temperature Mannequin (supply: Web site screenshot, Stanford Temperature Mannequin)
Researchers from Stanford College have created a temperature-by-depth mannequin for continental US that predicts subsurface temperatures to 7 kilometers depth.
A research by Mohammad Aljubran and Professor Roland Horne from Stanford College has resulted within the creation of a temperature-at-depth mannequin masking 0 to 7 kilometers depth (at 1-km intervals) for the continental United States. The map might be accessed by way of a web-based API or ArcGIS. The info may also be accessed by way of the Geothermal Information Repository.
A preprint outlining the strategies and the findings might be accessed right here: https://doi.org/10.48550/arXiv.2403.09961
The Stanford Thermal Earth Mannequin was generated utilizing inputs from greater than 400,000 bottomhole temperature measurements. Different inputs included information on depth, elevation, geographic coordinates, sediment thickness, magnetic anomaly, gravity anomaly, gamma-ray flux of radioactive parts, seismicity, and electrical conductivity.
The info was then processed utilizing physics-informed graph neural networks. The mannequin was skilled to concurrently predict subsurface temperatures, floor warmth movement, and rock thermal conductivity, therefore satisfying the three-dimensional warmth conduction regulation.
The mannequin confirmed superior temperature, floor warmth movement and thermal conductivity imply absolute errors of 4.8C, 5.817 mW/m2 and 0.022 W/(C-m), respectively.
The mannequin by Aljubran & Horne joins a rising checklist of works that intention to make information on subsurface temperatures publicly accessible. Such efforts are instrumental within the identification of potential websites for geothermal tasks, in addition to within the communication of dangers and alternatives to stakeholders. Noteworthy examples embrace the next:
