surface-air temperature differences are commonly used in a bulk resistance equation for estimating sensible heat flux, h. this estimate of h is inserted in the one-dimensional energy balance equation to solve for the latent heat flux, le, as a residual serious discrepancies between estimated and measured le have been observed for partial-canopy-cover conditions, which are mainly attributed to inappropriate estimates of h. in an attempt to improve the estimates of h over sparse canopies and obtain more accurate values of le, one- and two-layer resistance models that account for some of the factors causing poor agreement are developed. the utility of the two models for estimating le at the field scale is tested with remotely sensed and micrometeorological data collected in an arid environment from a furrowed cotton field with 20 cover and a dry soil surface. it is found that the one-layer model performs better than the two-layer model when a theoretical bluff-body correction for heat transfer instead of an empirical adjustment, thought to be applicable to a fairly wide range of conditions, is used. otherwise, the two-layer model is better. this result is attributed to the significant size of the furrows relative to the height of the vegetation. 21 refs