On the mechanisms of coexistence among annual-plant species, using neighbourhood techniques and simulation models

Abstract

Many studies have investigated the density-dependent regulation ofannual-plant populations on coastal sand dunes, but few have explored theconsequences of competition for the coexistence of plants in these simplecommunities. We used neighbourhood techniques to parameterize competition anddispersal functions from field data collected for two species of dune annual(Aira praecox and Erodium cicutarium)over three successive years, and then combined these functions into spatiallyexplicit simulation models. The population size of Airavaried enormously between years, while Erodium remainedsteady. Competition with neighbours reduced the spike length ofAira plants only in one of the three years (when itspopulation density was highest), while competition with neighbouringErodium plants appeared to result in the local death ofAira plants. However, these density-dependent effects werefar too weak to generate the observed changes in the population size ofAira among years, or to maintain populations below theupper limits observed. The large-seeded Erodium wasaffected by intraspecific competition but was unaffected by small-seededAira plants. Therefore, the larger-seeded species wascompetitively superior to the smaller-seeded species, an affect that couldpromote coexistence (albeit weakly) by a competition-colonisation trade-off.Modal dispersion distances of Aira andErodium were 45 and 60 mm respectively,greater than the radius within which competitive interactions occurred (40mm). Theoretical studies suggest that under these conditions thespatial arrangement of plants should be nearly random. In factAira was spatially aggregated, especially when rare,suggesting that patchy mortality across the dunes was important in generatingspatial structure. The study suggests that density dependence only weaklyregulates dune annual communities, while year-to-year environmental variationexert major influences on population sizes and spatial structures.