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

Titre : Seed limitation vs. water limitation in annual plant communities of semi-arid shrubland

Auteur : Adler, Dov

Université de soutenance : Ben-Gurion University of the Negev

Grade : Master of Science (M. Sc.) 2006

Résumé
Plant communities, in terms of density, primary production and richness, depend on the spatial and temporal distribution of resources and organisms (plants as seeds). In arid and semi-arid ecosystems, the essential resources water, organic matter and nutrients, are patchily distributed in the landscape. Semi-arid shrubland in the northern Negev is composed of two patch types : scattered shrub patches taking up 5-20% of the area and an intershrub matrix of soil covered with a cyanobacterial crust. It is commonly assumed that annual plant density in arid ecosystems is determined mainly by resource availability. I tested the alternative hypotheses, that establishment and thus density of the annual plants is seed-limited by low seed availability. Seed limitation can occur due to low seed production, dispersal, arrival and/or retention, and can be affected by granivory during any stage. I applied two treatments, irrigation (irrigated and not irrigated) and seed manipulation (seeds removed or added). The predictions of the experiment were, that 1) Seed addition will increase plant density and diversity. 2) Irrigation will increase plant density and diversity only at higher seed availability. In the first year of the experiment I did not find any evidence that the annual plant communities of the crusted intershrub matrix are seed or resource limited. However, the decrease of both Simpson’s and Shannon’s equitability measures with irrigation indicates that some water limitation occurred, as addition of water increased the abundance of dominant species relative to the minor species. In 2005, the increase of Shannon’s J with seed addition indicates that some minor species may be seedlimited, resulting in a slight increase relative to the dominants. Species richness increased from 2004 to 2005. Since 2005 was wetter than 2004 (330 vs 190 mm, respectively), this supports the idea of resource limitation, at least for some minor species. Ordination of 2004 of the experiment showed that irrigation caused a consistent shift in species composition, with Stipa capensis prevalent in the unirrigated samples and Plantago coronopus in the irrigated ones. In 2005 Anagallis arvensis and P. coronopus together reflect the irrigated end of the main compositional gradient. Community responses were very variable and most treatment effects were not statistically significant. However, dominance structure of the community changed, due to the responses of individual species. The main dominant grass, S. capensis, and the numerous minor species did not respond to irrigation, which suggests that their densities are seed limited, although seed addition did not in fact increase seed availability for all species. The other species, the dominant dicots (forbs) Anagallis arvensis and Plantago coronopus, on the other hand, showed resource limitation

Mots Clés : Desert ecology. — Seeds — Ecology. — Shrubland ecology — Israel — Negev. — Shrubs — Effect of drought on. — Shrubs — Irrigation.

Présentation (BGU)

Version intégrale