Comparison of two simplified approaches for ground temperature estimations in Australia

Abstract

Developing an accurate and practical method for ground temperature estimations are critical for the ground source heat pump system design and energy calculation procedures. In Australia, Baggs' method is a common procedure for ground temperatures predictions as a function of depth and time of year. Xing and Spitler developed a new procedure for ground temperature estimations for engineering applications at 4112 sites worldwide. This new procedure considers the variations of surface cover conditions (bare soil, vegetated, asphalt or concrete), effects of snow cover and soil freezing or melting. These important factors, which significantly affect the ground temperature results accuracy either are neglected or are simplified in Baggs' method. In this paper, we selected 6 sites in Australia which belongs to two climates: warm climates and arid or dry summer climates. Xing and Spitler's method and Baggs' method are used respectively to calculate the ground temperatures at depths of 10cm, 20cm, 50cm and 100cm. Calculation results of two methods are both compared to the 3-14 years of measurement results at the 6 sites and validation results are discussed and investigated. Results demonstrate the Xing and Spitler's method averaged root mean square error (RMSE) is 2.2°C of the 6 sites; Baggs' method averaged RMSE is 3.4°C of the 6 sites. This paper presents a new and improved procedure for ground temperature estimations in Australia. It enables a more accurate design of the ground heat exchangers so as to reduce the capital cost of the installed ground source heat pump systems.