Get reliable science

Models predict that desertification, the loss of perennial grasses in arid and semiarid environments, will be difficult to reverse due to the positive feedback between plant cover and water infiltration; reductions in plant cover concurrently reduce water infiltration to a level that is insufficient for grass growth and recruitment. Thus, a desertified state becomes stable. However, these models cannot account for recently reported perennial grass recovery in desertified habitats. Here, we suggest a novel mechanism to explain these observations: the natural increase in water infiltration rate associated with recovery from livestock-induced soil compaction. We empirically evaluate this mechanism by comparing vegetation, water infiltration rates and soil compaction at three nearby, desertified livestock exclosures that differ in time since livestock removal and perennial grass recovery; the oldest exclosure (54 years) has experienced a significant increase in perennial grasses, while there has been no such increase at the two younger exclosures (25 and 10 years). Across sites, relative differences (inside compared to outside) in water infiltration rate and soil compaction increased across each grazing fence with time since livestock removal. The relative difference in soil compaction and relative difference in water infiltration rate were significantly greatest at the sole exclosure in which perennial grasses have recovered. These data support a key aspect of desertification models: the importance of water infiltration rate for resilience of vegetation in desertified systems. Although water infiltration rate plays a critical role in understanding the stability of desertified systems, theoretical models of desertification have only incorporated plant cover as the key mechanism affecting infiltration rate. Our work suggests soil compaction may also be an important mechanism.