SOIL TEMPERATURE PROFILE INVESTIGATION UNDER ARID CLIMATE OF KUWAIT USING MECHANISTIC AND MIXED MODELS

Abstract

Soil temperature prediction at several depths is crucial for environmental, engineering, and management applications. In this study, two approaches, namely the mechanistic model and the mixed model, are used to estimate the soil temperature at depths from 0.05 m to 8 m and compared with measured data. The Carslaw-Jaeger equation s parameters, the average, the amplitude, and the phase lag, that suit an arid environment such as Kuwait are determined by the two techniques. Heat flux analyses of the soil are presented. The generated models are tested against the measured data with different statistical tests such as R2, RMSE, and MAPE. Values of R2 for the mechanistic and the mixed models range from 0.9119 to 0.9315 and 0.9068 to 0.9268 respectively, for the depths from 0.05 m to 1 m.