Sensible and latent heat fluxes are important parameters for understanding energy/mass interactions. At present, there are a number of methods available to estimate these fluxes, with varying degrees of difficulty and success. In the present study, we explore the use of the flux variance method and the surface renewal method to estimate the sensible and latent heat fluxes in two agricultural sites in China. The results of these two methods were compared to direct measurements using eddy covariance. The study sites, a wheat field in a semi-arid area and a rice paddy in a humid area provide an opportunity to examine two extreme conditions. In the arid conditions, both estimation methods provided similar sensible heat flux estimations but higher latent heat flux estimations with respect to the eddy covariance measurements. The sum of the estimated sensible and latent heat fluxes was in good agreement with measurements. In the rice paddy, sensible heat flux estimated by both methods showed similar results, whereas latent heat fluxes estimation were 1.2 times greater than eddy covariance measurements. The surface renewal method underestimated the Bowen ratio over the rice paddy, but overestimated it over the wheat field. In conclusion, both methods appear more appropriate for the estimation of sensible heat flux rather than for latent heat flux, in particular in humid environments.