It is important to obtain the soil water content threshold (theta(THR)) for agricultural water management. However, the measurement of theta(THR) is time consuming and needs specialized and expensive equipment. The accuracy of the empirical estimates is often low. Therefore, the development of a simple, rapid, and accurate prediction method for theta(THR) is the focus of the present study. The value of theta(THR) is regarded as the soil water content at the capillary break capacity (theta(CB)). A formula based on field capacity (theta(FC)) and soil bulk density (D-b) is proposed to calculate theta(CB), expressed as theta(CB)=theta(FC)-0.21*1-D-b/2.65. Six soils from six published studies on the response of tree physiological processes to water deficit were used to calculate theta(CB) using this formula. The calculated theta(CB) values were compared with the measured theta(THR). The results showed that the calculated theta(CB) values were nearly equal to the measured theta(THR). A highly significant (adj R-2 = 0.9826, p < 0.001) linear relationship with a slope of 0.9506 and a y intercept of 0.0072 was found between the calculated theta(CB) and measured theta(THR). The formula proposed in this study provides a novel approach for estimating the theta(THR) of trees in the semi-arid regions on the Loess Plateau.