Evaluating the importance of spatiotemporal variation in plant establishment

Background/Question/Methods

Experiments under carefully controlled conditions have been criticized as having limited applicability because they do not adequately reflect the heterogeneity of natural environments. This heterogeneity, arising from multiple sources, can affect the magnitude and direction of ecological interactions. We evaluated the importance and magnitude of spatiotemporal variability relating to plant establishment in a field restoration experiment in western Washington. Our experiment was conducted at four sites (two each in North and South Puget Sound) and in three seeding years (2009-2011), for a total of ten site-year combinations. Sites were abandoned agricultural fields initially dominated by non-native grasses and agricultural weeds. We killed the extant vegetation using two site preparation treatments (glyphosate-burn-glyphosate (burned) or glyphosate-mow-till-solarize (solarized)). All plots were broadcast seeded with a forb-rich mixture in autumn, keeping the seeding mixture as consistent as possible among site-year combinations (mean of 23 species seeded (range: 20-26), with a mean density of 711 seeds m^-2 (range: 603-970)). Establishment (species richness and density of seeded species) was measured the following spring. Data were summarized as proportions of the sown richness and sown density, and analyzed using generalized linear models.

Results/Conclusions

Establishment differed more strongly among site-year combinations than site preparation treatments, but numerous interactions were evident. Effects of site preparation on seedling density differed more strongly among years than sites; for example, seedling densities were higher in solarized than burned plots for 5 of 6 site-year combinations in 2009-2010 but were higher in burned plots at all sites in 2011. The proportion of seeded species that established ranged almost three-fold among site-year combinations (23-61%; mean 42%); more sown species established in the North than South Sound (55 vs. 32%). Seedling density averaged 7% of the sown density. Establishment was greater in North than South Sound (10 vs. 5%), but spatiotemporal variability was also more pronounced in the North Sound. For example, seedling density ranged from 2-15% of sown densities in North Sound site-year combinations compared to 3-7% in South Sound site-year combinations. The higher variability in the North Sound is due in part to its lower mean annual precipitation (540 vs. 1300 mm) and thus higher sensitivity to intra-annual precipitation patterns.  These results highlight the importance of replicating experiments both spatially and temporally, as the outcome of ecological interactions can depend on when and where they occur.