Soil ice content impacts winter vadose zone hydrology. It may be possible to estimate changes in soil ice content with a sensible heat balance (SHB) method, using measurements from heat pulse (HP) sensors. Feasibility of the SHB method is unknown because of difficulties in measuring soil thermal properties in partially frozen soils. The objectives of this study were (i) to examine the SHB method for determining in situ ice content, and (ii) to evaluate the required accuracy of HP sensors for use in the SHB method. Heat pulse sensors were installed in a bare field to measure soil temperatures and thermal properties during freezing and thawing events. In situ soil ice contents were determined at 60-min intervals with SHB theory. Sensitivity of the SHB method to temperature, heat capacity, thermal conductivity, and time step size was analyzed based on numerically produced soil freezing and thawing events. The in situ ice contents determined with the SHB method were sometimes unrealistically large or even negative. Thermal conductivity accuracy and time step size were the key factors contributing to SHB errors, while temperature and heat capacity accuracy had less influence. Ice content estimated with a 15-min SHB time step was more accurate than that estimated with a 60-min time step. Sensitivity analysis indicated that measurement errors in soil temperature and thermal conductivity should be less than ±0.05°C and ±20%, respectively, but the error in the soil heat capacity could vary by ±50%. Thus, improving the accuracy of thermal conductivity measurements and using short time steps are required to accurately estimate soil ice contents with the SHB method.