Space conditioning has one of the highest end-use service demands in the building sector. To avoid negative effects on the energy system and the environment, efficient energy sources and technologies must be implemented to provide future heating and cooling requirements. Geothermal heat pump systems (GHPS) are one of the technologies with highest thermodynamic and cost performance; nevertheless, its performance highly depends on local geological characteristics. In this study, a thermodynamic assessment of different types of sands, that could potentially be used as energy sources for GHPS in dry regions, has been conducted. The experiment focuses on understanding the thermal behaviour of five dry sand samples with different standard sieve sizes according to ASTM designations (Nos. 50, 45, 30, 16, and 14) and moisture content capacities. Based on the obtained data, a mathematical model to predict sand temperatures has been derived considering grain size, shape and moisture content. Compared to previous models, our results show that the developed model computed more accurate approximations compared to actual temperatures, providing a robust thermal behaviour model of dry regions' sands that could be used in building simulation tools more effectively. (C) 2019 Elsevier B.V. All rights reserved.