
|
|
|
Spatially explicit population modelling of the gypsum endemic Helianthemum squamatum. Quintana-Ascencio, Pedro*,1, Caballero, Idoia2, Olano, Jose 3, Escudero, Adrian4, Albert, Maria4, 1 University of Central Florida, Orlando, FL2 Laboratorio de Botánica., Bilbao, Spain3 Universidad de Valladolid, Soria, Spain4 Universidad Rey Juan Carlos., Madrid, Spain ABSTRACT- We use spatio-temporal Markov chains to model the population dynamics of Helianthemum squamatum in gypsum habitats of central Spain. We evaluate changes in plant cover, population persistence and population spatial structure in different scenarios by varying spatially-explicit habitat structure, dispersal functions, and levels of demographic and environmental stochasticity. Simulated populations under basal conditions declined showing a trend of decreasing spatial structure across population stages with seeds having the highest autocorrelation between neighboring squares and small plants having the lowest spatial autocorrelation. This spatial trend was stronger in simulations without demographic stochasticity, weaker in simulations without dispersal and disappeared under uniform dispersal. Quasiextinction (threshold of 20 individuals) was very similar among scenarios under basal conditions and different levels of dispersal. Uniform microhabitat had a detrimental effect on population growth and cover. Since we found no effect of dispersal function on population growth, this result can not be interpreted as a consequence of the arrangement of adult plants. A plausible hypothesis is a change in microhabitat availability (safe sites) for recruitment. The model indicates a change of habitat from shrub cover to crust and litter microhabitat. The increase of crust may be responsible for the higher growth rate in the baseline simulation. Key words: spatial structure, matrix models, ,Helianthemum |
All materials copyright The Ecological Society of America (ESA), and may not be used without written permission.