Over the years, electromagnetic induction sensors, such as EM38, have been used to monitor soil salinity or local electrical conductivity (ECa) and their output has been instrumented in establishing models for depth profiling of ECa. In the previous work both the forward propagation and inverse matrix approaches offered potential to produce depth profiles of soil ECa. However, it remains a question whether EM38 is able to measure v in different depths. This present study concerns itself with underlying volumetric soil moisture (v) and in our previous study it was demonstrated that v is a key driver of ECa in the deep Vertisol soils. Therefore, the objective of this study was to investigate the ability of the EM38 sensor to estimate vertical variations in the soil moisture profile using both the forward propagation and the inverse matrix procedures. The EM38 was calibrated for determining v using the pit calibration method. In the pit sampling process, undisturbed soil cores were taken at different depths by digging a pit at 1.8 m depth and subsequently v was determined from these samples according to the gravimetric method. Before digging the pit, the EM38 readings (ECa) at specified heights were taken in both vertical and horizontal dipole orientations from the same places where core samples were taken. All ECa values were then converted to the local moisture-driven ECaθ. The forward-propagation models of Rhoades & Corwin (1981) [Soil Sci. Soc. Am. J. 45: 255-260] and Slavich (1990) [Aust. J. Soil Res. 28: 443 - 352] were refined and tested to see how well they could directly predict the vertical profile of soil moisture content along with the inverse-matrix Thikonov regularization procedure described by Borchers et. al. (1997) [Soil Sci Soc Am J. 61: 1004 -1009]. The relationship between v and both ECa & ECaθ for all depth groups was statistically significant for both the Slavich model and Rhoades & Corwin model. However, the Slavich model, incorporating both vertical and horizontal dipole configurations, produced the best predictions with an error of approximately 10%. This is might be due to the integration of response of the instrument where it occurs ~80% and ~62% for the horizontal and vertical dipole configurations respectively from the surface to 1.2 m depth. Again the integrated response of vertical mode reaches maximum at 0.4 m depth and it diminishes to zero at the surface while it reaches at maximum at the surface for horizontal mode of operation. A complex inversion process involving Thikonov regularization was also tested and was found to consistently under-predict v by approximately 50 – 75%. Therefore, the forward propagation model of the EM38 could be used to directly measure the v¬ at different depths.