The effects of foundation species on community assembly: a global study on alpine cushion plant communities.

Foundation species can change plant community structure by modulating important ecological processes such as community assembly, yet this topic is poorly understood. In alpine systems, cushion plants commonly act as foundation species by ameliorating local conditions. Here, we analyze diversity patterns of species' assembly within cushions and in adjacent surrounding open substrates (83 sites across five continents) calculating floristic dissimilarity between replicate plots, and using linear models to analyze relationships between microhabitats and species diversity. Floristic dissimilarity did not change across biogeographic regions, but was consistently lower in the cushions than in the open microhabitat. Cushion plants appear to enable recruitment of many relatively stress-intolerant species that otherwise would not establish in these communities, yet the niche space constructed by cushion plants supports a more homogeneous composition of species than the niche space beyond the cushion's influence. As a result, cushion plants support higher α-diversity and a larger species pool, but harbor assemblies with lower ß-diversity than open microhabitats. We conclude that habitats with and without dominant foundation species can strongly differ in the processes that drive species recruitment, and thus the relationship between local and regional species diversity.

Facilitative plant interactions and climate simultaneously drive alpine plant diversity.

Interactions among species determine local-scale diversity, but local interactions are thought to have minor effects at larger scales. However, quantitative comparisons of the importance of biotic interactions relative to other drivers are rarely made at larger scales. Using a data set spanning 78 sites and five continents, we assessed the relative importance of biotic interactions and climate in determining plant diversity in alpine ecosystems dominated by nurse-plant cushion species. Climate variables related with water balance showed the highest correlation with richness at the global scale. Strikingly, although the effect of cushion species on diversity was lower than that of climate, its contribution was still substantial. In particular, cushion species enhanced species richness more in systems with inherently impoverished local diversity. Nurse species appear to act as a 'safety net' sustaining diversity under harsh conditions, demonstrating that climate and species interactions should be integrated when predicting future biodiversity effects of climate change.

A realized facilitation cascade mediated by biological soil crusts in a sagebrush steppe community.

Biological soil crusts can have strong effects on vascular plant communities which have been inferred from short-term germination and early establishment responses. However, biocrusts are often assumed to function as an "organizing principle" in communities because their effects can "cascade" to interactions among crust-associated plant species. We conducted surveys and experiments to explore these cascades and found that biocrusts were positively associated with large patches (> 10 m diameter) of a dominant shrub Artemisia tridentata. At the smaller scale of individual shrubs and the open matrices between shrubs, biocrusts were negatively associated with Artemisia. Juveniles of Artemisia were found only in biocrusts in intershrub spaces and never under shrubs or in soil without biocrusts. In two-year field experiments, biocrusts increased the growth of Festuca and the photosynthetic rates of Artemisia. Festuca planted under Artemisia were also at least twice as large as those planted in open sites without crusts or where Artemisia were removed. Thus, biocrusts can facilitate vascular plants over long time periods and can contribute to a "realized" cascade with nested negative and positive interactions for a range of species, but unusual among documented cascades in that it includes only autotrophs.

Plant invasions, generalist herbivores, and novel defense weapons.

One commonly accepted mechanism for biological invasions is that species, after introduction to a new region, leave behind their natural enemies and therefore increase in distribution and abundance. However, which enemies are escaped remains unclear. Escape from specialist invertebrate herbivores has been examined in detail, but despite the profound effects of generalist herbivores in natural communities their potential to control invasive species is poorly understood. We carried out parallel laboratory feeding bioassays with generalist invertebrate herbivores from the native (Europe) and from the introduced (North America) range using native and nonnative tetraploid populations of the invasive spotted knapweed, Centaurea stoebe. We found that the growth of North American generalist herbivores was far lower when feeding on C. stoebe than the growth of European generalists. In contrast, North American and European generalists grew equally well on European and North American tetraploid C. stoebe plants, lending no support for an evolutionary change in resistance of North American tetraploid C. stoebe populations against generalist herbivores. These results suggest that biogeographical differences in the response of generalist herbivores to novel plant species have the potential to affect plant invasions.

Novel weapons and invasion: biogeographic differences in the competitive effects of Centaurea maculosa and its root exudate (+/-)-catechin.

Recent studies suggest that the invasive success of Centaurea maculosa may be related to its stronger allelopathic effects on native North American species than on related European species, one component of the "novel weapons" hypothesis. Other research indicates that C. maculosa plants from the invasive range in North America have evolved to be larger and better competitors than conspecifics from the native range in Europe, a component of the "evolution of increased competitive ability" hypothesis. These hypotheses are not mutually exclusive, but this evidence sets the stage for comparing the relative importance of evolved competitive ability to inherent competitive traits. In a competition experiment with a large number of C. maculosa populations, we found no difference in the competitive effects of C. maculosa plants from North America and Europe on other species. However, both North American and European C. maculosa were much better competitors against plants native to North America than congeners native to Romania, collected in areas where C. maculosa is also native. These results are consistent with the novel weapons hypothesis. But, in a second experiment using just one population from North America and Europe, and where North American and European species were collected from a broader range of sites, competitive interactions were weaker overall, and the competitive effects of C. maculosa were slightly stronger against European species than against North American species. Also consistent with the novel weapons hypothesis, (+/-)-catechin had stronger effects on native North American species than on native European species in two experiments. Our results suggest that the regional composition of the plant communities being invaded by C. maculosa may be more important for invasive success than the evolution of increased size and competitive ability.

The relative importance of allelopathy in interference: the effects of an invasive weed on a native bunchgrass.

The relative importance of allelopathy and resource competition in plant-plant interactions has been vigorously debated but seldom tested. We used activated carbon to manipulate the effects of root exudates of Centaurea maculosa, a noxious weed in much of western North America, on root elongation rates and growth of the native bunchgrass Festuca idahoensis in order to investigate the relative importance of allelopathy in the total interference of Centaurea. In root observation chambers, Festuca root elongation rates decreased to ≈50% of the control, beginning 4 days before contacting Centaurea roots in silica sand. However, when activated carbon, which has a high affinity for adsorbing to organic compounds, was added to the sand the effects of Centaurea roots on Festuca root elongation were reduced. In other experiments, Festuca plants were 50% smaller when grown with Centaurea than with conspecifics in pure silica sand. However, Festuca grown with Centaurea in mixtures of sand and activated carbon were 85% larger than Festuca grown with Centaurea in silica sand without carbon. These results suggest that allelopathy accounts for a substantial proportion of the total interference of Centaurea on Festuca, shifting the balance of competition in favor of Centaurea. However, Centaurea outperformed Festuca even in the presence of activated carbon, demonstrating the importance of the combined roles of resource competition and allelopathy.

Concentrations of the Allelochemical (+/-)-catechin IN Centaurea maculosa soils.

The phytotoxin (+/-)-catechin has been proposed to mediate invasion and autoinhibition by the Eurasian plant Centaurea maculosa (spotted knapweed). The importance of (+/-)-catechin to C. maculosa ecology depends in part on whether sufficient catechin concentrations occur at appropriate times and locations within C. maculosa soil to influence neighboring plants. Previous research on catechin in C. maculosa soils has yielded conflicting results, with some studies finding high soil catechin concentrations and other, more recent studies finding little or no catechin in field soils. Here, we report the most extensive study of soil catechin concentrations to date. We examined soil catechin concentrations in 402 samples from 11 C. maculosa sites in North America sampled in consecutive months over 1 yr, excluding winter months. One site was sampled on seven dates, another was sampled twice, and the remaining nine sites were each sampled once on a range of sampling dates. Methods used were similar to those with which we previously measured high soil catechin concentrations. We detected catechin only in the site that was sampled on seven dates and only on one sampling date in that site (May 16 2006), but in all samples collected on that date. The mean soil catechin concentration on that date was 0.65 +/- 0.45 (SD) mg g(-1), comparable to previously reported high concentrations. There are a number of possible explanations for the infrequency with which we detected soil catechin in this work compared to previous studies. Differences in results could reflect spatial and temporal variation in catechin exudation or degradation, as we examined different sites in a different year from most previous studies. Also, large quantities of catechin were detected in blanks for two sampling periods in the present study, leading us to discard those data. This contamination suggests that previous reports of high catechin concentrations that did not include blanks should be viewed with caution. Our results suggest that pure catechin is only rarely present in C. maculosa bulk soils. Thus, although catechin may play a role in C. maculosa invasion, the infrequency of soil catechin that we determined in this study suggests that we cannot be as certain of its role as previous reports of high soil catechin concentrations suggested.

Oxalate contributes to the resistance of Gaillardia grandiflora and Lupinus sericeus to a phytotoxin produced by Centaurea maculosa.

Centaurea maculosa Lam. is a noxious weed in western North America that produces a phytotoxin, (+/-)-catechin, which is thought to contribute to its invasiveness. Areas invaded by C. maculosa often result in monocultures of the weed, however; in some areas, North American natives stand their ground against C. maculosa and show varying degrees of resistance to its phytotoxin. Two of these resistant native species, Lupinus sericeus Pursh and Gaillardia grandiflora Van Houtte, were found to secrete increased amounts of oxalate in response to catechin exposure. Mechanistically, we found that oxalate works exogenously by blocking generation of reactive oxygen species in susceptible plants and reducing oxidative damage generated in response to catechin. Furthermore, field experiments show that L. sericeus indirectly facilitates native grasses in grasslands invaded by C. maculosa, and this facilitation can be correlated with the presence of oxalate in soil. Addition of exogenous oxalate to native grasses and Arabidopsis thaliana (L.) Heynh grown in vitro alleviated the phytotoxic effects of catechin, supporting the field experiments and suggesting that root-secreted oxalate may also act as a chemical facilitator for plant species that do not secrete the compound.