A new system for simulating future belowground temperature increases was conceived, simulated, constructed and tested in a temperate deciduous forest in Oak Ridge, TN, USA. The new system uses low-wattage, 3 m deep heaters installed around the circumference of a defined soil volume. The heaters add the necessary energy to achieve a set soil temperature differential within the treatment area and add exterior energy inputs equal to those, which might be lost from lateral heat conduction. The method, which was designed to work in conjunction with aboveground heated chambers, requires only two control sensor positions one for aboveground air temperatures at 1 m and another for belowground temperatures at 0.8 m. The method is capable of achieving temperature differentials of at least +4.0±0.5 °C for soils to a measured depth of −2 m. These +4 °C differential soil temperatures were sustained in situ throughout 2009, and both diurnal and seasonal cycles at all soil depths were retained using this simple heating approach. Measured mean energy inputs required to sustain the target heating level of +4 °C over the 7.1 m2 target area were substantial for aboveground heating (21.1 kW h day−1 m−2), but 16 times lower for belowground heaters (1.3 kW h day−1 m−2). Observations of soil CO2 efflux from the surface of the target soil volumes showed CO2 losses throughout 2009 that were elevated above the temperature response curve that have been reported in previous near-surface soil warming studies. Stimulation of biological activity within previously undisturbed deep-soil carbon stocks is the hypothesized source. Long-term research programs may be able to apply this new heating method that captures expected future warming and temperature dynamics throughout the soil profile to address uncertainties in process-level responses of microbial, plant and animal communities in whole, intact ecosystems.