Thermal regime at shallow depths of the crust is controlled by the temperature condition at the surface and the heat flowing from deeper part of the Earth. In an idealized homogeneous crust, if the surface temperature is steady, the distribution of ground temperature is a linear function of depth. However, if the surface temperature changes with time, the ground temperature will depart from the linear distribution which is governed by heat flow (q) and thermal conductivity (k). A progressive cooling at the surface will cool down the rocks near to the surface, increase the thermal gradient at shallow depths, and lead to a temperature profile with curvature like the one shown in green in the illustration above. A progressive warming, on the other hand should be responsible for a temperature profile with smaller even negative thermal gradients at shallower depths like the one shown in red. If the surface temperature oscillates with time, oscillations in the ground temperature will follow. The magnitude of the departure of ground temperature from its undisturbed steady state is related to the amplitude of the surface temperature variation, and the depth to which disturbances to the steady state temperature can be measured is related to the timing of the original temperature change at the surface. A ground surface temperature history is therefore recorded in the subsurface. By careful analysis of the variation of temperature with depth, one can reconstruct the past fluctuation at the Earth's surface.
It is a kind of direct temperature - temperature study. Therefore, it is free of any uncertainties due to conversion from proxy data to temperatures.
The time interval over which the geothermal technique has resolving power covers both industrial and pre-industrial periods.
There exist many ground temperature profiles measured from boreholes located in remote areas. Climate information preserved in those temperature profiles contain fewer anthropogenic disturbances such as urban thermal island effect and agriculture.
There exist tens of thousands of borehole in the world which have been systematically temperature logged for heat flow determination. The huge archive of existing borehole temperatures comprises a great potential for a comprehensive world wide paleoclimate study.