The transpiration coefficient (K_c) method estimates evapotranspiration as a proportion of the evapotranspiration of a reference crop growing in ideal conditions. This approach was designed for irrigated crops and assumes that plants are not subjected to resource limitations. Other assumptions are that plants have high leaf area index and little stomatal resistance to water loss. These conditions are not common for arid-land vegetation. However, mainly due to its simplicity, some studies have proposed the use of transpiration coefficients as a method of determining evapotranspiration in arid environments. In this article, the documented applications of the K_c method in arid environments and their accuracy are reviewed. We also critically discuss the physiological and agronomic concepts that support the K_c method as they relate to water-limited environments. The K_c method typically overestimates water use when plants encounter suboptimal conditions of soil water because it does not consider stomatal regulation and plant adaptations to drought. We conclude that, although the transpiration coefficient method is simple to implement and widely recognized, it is not suitable for determining evapotranspiration of vegetation adapted to arid conditions.

Vegetation cover and production were evaluated after nearly 7 years of livestock grazing exclusion and shrub control in an area with a long history of heavy livestock grazing in the southern Chihuahuan Desert, Mexico. An exclosure was established to prevent livestock grazing. In half of the excluded area, the main shrub, Larrea tridentata, was mechanically controlled. Outside the exclosure, heavy livestock grazing occurred as customary and shrubs were not controlled. Absence of grazing resulted in 50% higher grass cover and 35% higher total biomass. Larrea tridentata cover was twice as high on the grazed area as on the ungrazed area. Vegetation cover was dominated by grasses (42%) in the ungrazed area, whereas in the grazed area, cover was equally divided between grasses (28%) and shrubs (27%). Shrub control did not affect vegetation cover or herbage production. Multivariate analysis confirmed that inside the excluded area, shrub control had little impact on the plant community. The effect of gra...