The small-scale spatiotemporal pattern of the seedbank and vegetation of a highly invasive weed, Centaurea solstitialis: strength in numbers

LORTIE, C.J.; M. MUNSHAW; J. DITOMASO; J.L. HIERRO

OIKOS

WILEY-BLACKWELL PUBLISHING, INC

Año: 2010 p. 428 - 436

0030-1299

The dynamics of invasive plant populations are intriguing and informative of the importance of population and community-level processes. A dominant approach to understanding and describing invasion has been the development of unique hypotheses to explain invasion. However, here we directly explore the relevance of the small-scale, spatiotemporal pattern in seedbanks and plants of the highly invasive weed, Centaurea solstitialis, to determine whether pattern can be used to contrast predictions associated with the simple ecological hypotheses of seed versus microsite limitations. At three invaded grasslands in California, highly invaded (> 20 adult plants present), invaded (< 10 adults), and uninvaded (no C. solstitialis plants) sites were selected. The spatial pattern of the seedbank was assessed using fine-scale, 2 cm diameter contiguous cores and geostatistical statistics, and the number of C. solstitialis seeds in the seedbank was recorded in addition to the total community seedbank density. Three of the four critical predictions associated with the seed limitation hypothesis were clearly supported as an explanation for the patterns of C. solstitialis invasion observed in the field. The density of C. solstitialis seeds decreased from high to low extents of invasion, there was no relationship between the community seedbank and C. solstitialis seeds, and the distances between C. solstitialis plants was inversely related to the density of C. solstitialis seeds. However, both the persistent and transient seedbanks of C. solstitialis were spatially aggregated with autocorrelation up to 12 cm2 which suggests that aggregation is a consistent attribute of this species in the seedbank regardless of extent of invasion. This basic pattern-based approach clearly detected an ecological signal of invasive seedbank dynamics and is thus a useful tool for subsequent studies of invasions in grasslands.