Spatial patterns of seed predation by harvester ants ( Messor Forel) in Mediterranean grassland and scrubland

Abstract

Seed predation by harvester ants is one of the main processes involved in the seed bank dynamics of Mediterranean grassland and scrubland. We investigated the spatial patterns of seed predation by Messor spp. in these systems from a multi-level approach (nest-site selection, foraging patch selection and seed predation risk), using ten experimental plots (20 m ¥ 40 m) set in central Spain. All habitats in grasslands are dominated by M. barbarus, while in scrubland this species shares seed exploitation with M. bouvieri. In this type of system, the nest-hole and foraging territory distribution of the two species show a certain degree of segregation. Messor bouvieri focuses on the exploitation of dry scrub, while M. barbarus also forages on subhumid and humid patches. We interpret this as a consequence of the different foraging behaviour and nest site requirements of the two species. In general, the spatial distribution of seed predation risk reflects the patterns detected for the nest-holes and foraging areas, although scrubland has a more heterogeneous predation pattern than grassland.

A stochastic model of seed dispersal pattern to assess seed predation by ants in annual dry grasslands

“…Therefore, densely aggregated groups of seeds were much more attractive for ants than sparse seeds, and the probability to be predated decreased with increasing reciprocal distance of seeds. Although this tendency was already recorded for Messor (Hensen 2002;Azca´rate and Peco 2003) and other ants (Kaspari 1993), it has been here statistically demonstrated for the first time.…”

“…All other things being equal, they tend to gather seeds of the kinds that are more abundant, passing from one set to the next across the growing season when those usually favoured become less available (Hahn and Maschwitz 1985;Cerdan et al 1986;Neumayer 1994;Detrain and Pasteels 2000). Messor species recruit foraging trails and harvest seeds mainly from clumps with higher seed densities (Hensen 2002;Azca´rate and Peco 2003). However, a large variety of Mediterranean species are exposed to seed predation by these ants (Hensen 2002;Mossa et al 2004).…”

Effects of harvester ants on seed availability and dispersal of Lavandula stoechas subsp. pedunculata in a Mediterranean grassland-scrubland mosaic

“…) colonies have also been recorded in the study area, but the activity of this species is restricted to scrub communities. The density of M. barbarus nest-holes varies between 200 and 300 per hectare, concentrated on the ecotones and adjacent scrub areas (Azca´rate and Peco 2003). At distances higher than fifty meters from the ecotone, M. barbarus is almost absent (Azca´rate 2003).…”

“…) have found high Lavandula seed availability in the scrubland, and almost none in the adjacent grassland as a result of the low dispersal range through autochory . The high density of M. barbarus colonies in the ecotones between scrubland and grasslands (Azca´rate and Peco 2003) could accentuate seed limitation in the potential expansion area of Lavandula, as ants collect seeds from a number of species from both the grassland and the adjacent scrub communities (Azca´rate and Peco 2003). At the same time, M. barbarus activity may influence seed dispersal if some of the predated seeds are accidentally returned to the soil surface at points impossible to reach by autochory.…”

Life history traits of a European Messor harvester ant

“…Being major seed consumers in xeric habitats (Azcárate and Peco, 2003), these keystone species infl uence the vegetation by depleting and dispersing seeds (e.g., Pacini, 1990;MacMahon et al, 2000) to an extent disproportionate to their numbers or biomass (Hölldobler and Wilson, 1990). Despite their ecological impact, little is known about the life history of Central European Messor ants.…”