A simple procedure to obtain the in situ determination of heat loss from an underground heat distribution system is described. The proposed procedure for estimating the heat loss from underground heat distribution systems consists of making a series of temperature and thermal conductivity measurements on the ground grid in the vicinity of the pipe system and deducing the heat loss by statistically determining the parameters of a mathematical model using the gathered data. Verification of the procedure was tested in a simulated scale model of an underground system using an insulated sandbox. The scale model is a two-pipe system buried 0.3045 m (12 in.) below the top surface of a 0.66-m (26-in.)-wide, 0.71-m (28-in.)-high, and 1.22-m (48-in.)-long sandbox. The test pipes are two electrically heated steel pipes having an outside diameter of 3.37 cm (1.325 in.). The walls of the sandbox are insulated by 5.08-cm (2-in.)-thick expanded polystyrene board. The heat loss measurements from this simulated underground heat distribution system were obtained for pipe surface temperatures of 66°C (150.9°F), 93.8°C (200.8°F), and 149.2°C (300.7°F) for three different sand moisture contents (dry, 10%, and 13%). Relatively good agreement was obtained between the measured electrical power input to the test pipes and the system heat loss estimated by the sand temperature profile around the pipes and sand thermal conductivity measured by the thermal needle technique. The paper also describes an application of the technique to a field measurement conducted at the James Madison University, Harrisonburg, Virginia.