Subsurface heat transport simulation with periodic surface temperature signals and groundwater flow

Vertical heat conduction in the subsurface is typically delineated by the conventional heat conduction equation, which incorporates records of surface temperature and thermal properties of the soil and subsoil. However, a new field experiment using a fiber-optic distributed temperature sensing (FO-DTS) system in a 100-m borehole returned a different temperature profile than predicted by the typical temperature diffusion equation, suggesting the need for a new heat transport model. The test site in east-central Illinois lies above the Mahomet Bedrock Valley ( approximately 100 m deep), which is filled with glacial and nonglacial sediments, including deposits of glacial sand and gravel forming an important buried valley aquifer, the Mahomet aquifer, in the region. Another shallower aquifer, the Upper Glasford aquifer, lies approximately 35 m below the ground surface. The 100-m borehole is grout sealed and extends through both aquifers to bedrock. Consecutive temperature measurements at 1-m and 0.1 degrees C resolution were made over several time scales, ranging from 30-minute to 1-month intervals from June to December 2015. Temperature-depth profiles showed monthly variations in the first 15 m and then reached a certain temperature, which can be explained by a simple one-dimensional transient heat conduction equation with annual surface temperature forcing. Influenced by geothermal processes, the temperature below 15 m increased almost linearly with depth, but deviation occurred with a concave-upward profile at greater depth, which might be the result of vertical groundwater flow and geologic structure changes. This interplay could be accounted for by an integrated approach that simulates the vertical temperature profile. To fully understand how these sources affect the profiles, future studies will incorporate the impacts of (1) heat content in infiltrating surface water; (2) depth-dependent thermal properties of geologic materials; and (3) low-frequency surface temperature variations. A hybrid model, validated by high-resolution FO-DTS measurements, could better simulate the heat transport and be applied to varying subsurface environments.