Informations et ressources scientifiques
sur le développement des zones arides et semi-arides

Accueil du site → Doctorat → États-Unis → 2012 → A new model for reversal of desertification : Long-term livestock removal affects soil and vegetation dynamics

Saint Louis University (2012)

A new model for reversal of desertification : Long-term livestock removal affects soil and vegetation dynamics

Allington, Ginger R.H.

Titre : A new model for reversal of desertification : Long-term livestock removal affects soil and vegetation dynamics

Auteur : Allington, Ginger R.H.

Université de soutenance : Saint Louis University

Grade : Doctor of Philosophy (PhD) 2012

Résumé
Desertification, the conversion of arid grasslands to shrublands due to overgrazing and drought, is occurring on over a quarter of the land on earth. Conceptual models of desertification predict desertified shrublands are in a stable state, and grass recovery is unlikely. These models hypothesize that as grass is lost, feedback loops involving low soil nutrients or insufficient water infiltration rates prevent the return of perennial grasses to desertified sites. However, these models cannot account for recent reports of reversals of desertification following long-term livestock removal. This dissertation investigates changes in soil physical and chemical properties associated with perennial grass recovery at a desertified site. I use data from this site to develop a new model for desertification. This new model postulates that in the absence of livestock there is a slow release from compaction that, over decades, results in an increase in water infiltration and a concomitant decrease in erosion, which promotes nutrient accumulation in the soil. I tested this model at two additional long-term exclosures with different vegetation histories. Data from these sites support the new model. Additionally, at all three sites I failed to detect islands of fertility under shrub canopies. Fertile islands are considered an inherent property of desert shrub systems. In order to investigate further I conducted an extensive literature review of the empirical examples of islands of fertility. I found that while the pattern was very common, at all sites where livestock had been removed for more than thirty years did not exhibit fertile islands. This suggests a relationship between fertile islands and the activities of livestock, and also further corroborates the assertion of the new desertification model that with sufficient rest from grazing soil nutrients between shrubs can increase significantly. This suggests that given sufficient time and removal of livestock, soil properties at some desertified sites can improve sufficiently to support the re-establishment of perennial grasses. This has significant implications for our understanding of the dynamics of vegetation and soil in arid systems and the potential for restoration from desertification.

Search Oxford Libraries Online (SOLO)

ProQuest Dissertations Publishing

Page publiée le 17 septembre 2012, mise à jour le 5 décembre 2018