Technical Abstract: Energy balance studies generally require knowledge of the heat flux at the soil surface (G0). Typically, the rate of change of heat storage in the soil (S) is used together with an estimate of the reference soil heat flux (Gr) at some depth to determine G0. The near-surface soil volumetric heat capacity (C) must be determined in order to calculate S. The objective of this paper is to quantify the effects of the method of C determination on S. The methods evaluated were estimation of C by soil sampling every 2-3 days, estimation of C based on Theta Probe water content measurement every 2-3 days, and hourly measurement of C using heat pulse sensors. Measurements were performed under a bare soil surface, a soybean [Glycine max (L.) Merr.] canopy, and a corn (Zea mays L.) canopy. When C was determined using all three methods simultaneously, the three C estimates agreed to within 6% on average. However, temporal variability of C was best recorded with the heat pulse sensors. Sampling three times per week was not frequent enough to consistently record the rate of change of C (dC/dt) for the near-surface soil. With sampling intervals of 2-3 days, dC/dt is unknown and neglected when calculating S. Neglecting dC/dt introduced errors occasionally exceeding 200 W m-2 during infiltration and persistent but much smaller errors during periods of soil drying. Hourly measurements with the heat pulse sensors avoided these errors and provided the most accurate S data.