Intraspecific interactions affect the spatial pattern of a dominant shrub in a semiarid shrubland: A prospective approach.

Dispersal, physical conditions and biotic interactions contribute to determine the spatial distribution of individuals in plant populations. Much of what we know has been learned from studies that retrospectively posit mechanisms presumed to have generated the observed spatial patterns. Here we present a prospective approach. We start by measuring spatial demographic effects and evaluate if they can generate observed spatial patterns. We evaluated the influence of interactions among conspecifics on vital rates, demography and spatial distribution of Croton aff. wagneri, a dominant shrub in dry Andean ecosystems. Recruitment, survival and growth varied in relation with distance to conspecifics neighbours and with their summed cover. We built a spatial individual‐based model and simulated its population dynamics in 30 × 30 m plots for a 30‐year period. We compared the predicted spatial pattern from these demographic models with that observed among plants in 16 independent plots with the same area. Simulated populations mimicked observed spatial patterns, although in plots at high elevations the simulated populations did not reproduce the observed inhibition at small scales. Observed and simulated patterns indicated differences between elevations in maximum aggregation and location of the distances with higher aggregation. We discuss how consideration of critical seed and juvenile stages and interspecific interactions could further improve our understanding of spatial pattern and recommend that these factors be considered in future models. We developed a prospective approach to evaluate the influence of interactions among conspecifics on vital rates, demography and spatial distribution of Croton aff. wagneri, a dominant shrub in dry Andean ecosystems. We built a spatial individual‐based model and simulated its population dynamics to compare predicted spatial pattern from demographic models with that observed among plants in 16 independent plots. Simulated populations mimicked observed spatial patterns, although in plots at high elevations the simulated populations did not reproduce the observed inhibition at small scales. [ABSTRACT FROM AUTHOR]