Abstract: The limitations of the eddy covariance method contribute substantially to surface energy imbalance. This paper evaluates two existing approaches (large-eddy averaging method and the Durand method) for improving the computational accuracy of the eddy covariance method and enhancing surface energy balance. Through five sets of daytime experiments, their similarities, differences, comprehensive effects, and improvement strategies of the two methods were comparatively studied. Results reveal the following: (1) the level of diurnal surface energy closure is related to the sequence length. If the sequence length is too long, the closure level deteriorates; if it is too short, the closure level becomes inconsistent. Thus, a turning point exists between the two extremes, at which an optimal closure level can be achieved. (2) The original EBR (energy balance ratio) is 0.80, with an energy Res (residual) of 64.9 W m⁻², accounting for 23.8% of the available energy, indicating a severe imbalance. (3) After considering the large-eddy averaging method, the EBR increases by 0.18 and the Res decreases to 14.5 W m⁻² (representing 5.3% of the available energy), thereby achieving balance. (4) After considering the Durand method, the EBR increases by 0.15, and the Res decreases to 25.1 W m⁻², accounting for 9.2% of the available energy. This effectiveness is inferior to that of the large-eddy averaging method. (5) When considering both methods, the EBR increases to 1.17, and the Res decreases to −35.1 W m⁻², resulting in severe over-closure. Subsequent analysis reveals that this over-closure is due to the excessive accumulation of the additional sensible-heat term in the Durand method during the day. (6) By removing the accumulation part and only considering the additional latent-heat term while applying both methods, the EBR can reach 1.0, and the Res can be reduced to 8.7 W m⁻², accounting for only 3.2% of the available energy, thereby achieving an ideal closure state. Further analysis confirms that the large-eddy averaging method is the dominant factor in achieving surface energy closure.