Community dominance by a canopy species controls the relationship between macroalgal production and species richness

By field manipulations of nutrients, propagules, and canopy cover in a shallow rocky subtidal, we show that dominance by a leathery macrophyte (Fucus vesiculosus) controls net production and the propagation of nutrient effects on understory algal species richness. On substrates with propagules, canopy cover prevented a significant loss of diversity from nutrient enrichment demonstrated outside canopy cover. On sterile substrates, negative canopy effects on diversity were counteracted by nutrient enrichment. Experimental manipulations produced a significant hump-shaped relationship between species richness and log net production, with low richness at low or high biomass production and highest richness at intermediate biomass production. By strong control of net primary production, canopy cover changed the algal understory community to an alternative production state and thereby generated a switch in the relationship between net primary production and species richness from negative outside canopy cover to positive inside. This demonstrates that species contributing to biological structure and habitat diversity can determine the responses of a coastal ecosystem to resource loading. In a world where large-scale species trends are dominated by species erosion, increasing evidence from diverse ecosystems shows that species richness has significant effects on central ecological processes (Chapin et al. 2000). Important ecosystem properties such as primary production and nitrogen storage can increase with both species richness and functional-group diversity of plants and algae (Loreau et al. 2001; Kinzig et al. 2002; Bruno et al. 2005; van Ruijven and Berendse 2005). Consumermediated trophic interactions that control many ecosystems also seem highly dependent on richness, both of prey and consumer levels (Duffy et al. 2003; Hillebrand and Cardinale 2004). Such effects of species diversity on ecosystem function can be a simple linear relation between species richness and trait diversity (Loreau et al. 2001). However, higher diversity can also cause nonadditive changes in community processes, by increasing the probability of interspecific interactions such as facilitation (Cardinale et al. 2002) and intraguild predation (Finke and Denno 2004), or by including highly productive and dominant species that control the assembly (Chapin et al. 1997). Species that produce canopies dominate many ecosystems because of their often massive, persistent, and extensive structures (Jones et al. 1997). Canopy-forming species can control ecosystem processes both by being competitively superior and dominating primary production (Paine 2002), and by increasing habitat complexity and modifying the environment for associated organisms (Dayton 1975; Bruno et al. 2003). Thus, in many ecosystems a few dominant species may fundamentally alter ecosystem properties, which should generate highly skewed effects of biodiversity loss on ecosystem functioning (Solan et al. 2004). Habitats generated by canopy-forming species are often threatened by pollution, development, or overexploitation (Bruno et al. 2003), making the need to understand interactions between canopy-mediated processes, biodiversity loss, and habitat change a major challenge for conservation. Here we explore the influence of a canopy-forming seaweed and resource availability on a macroalgal understory by field manipulations of canopy presence, nutrients, and recruitment history on a subtidal rocky shore in the southern Baltic Sea. Canopy-mediated processes are an important part of subtidal community dynamics, since subtidal communities are commonly patchy with large variations in the abundance of canopyforming species. Nutrients often determine marine community production (Menge 2000) and nutrient loading is a common cause of habitat change in coastal ecosystems (Cloern 2001). On rocky shores, nutrient effects on macroalgal production depend on the presence of microscopic resting stages (overwintering propagules, Worm et al. 2001). We therefore included recruitment history in the design by exposing plots consisting of both a sterile artificial substrate and a bare natural stone with propagules (no macroscopic vegetation) to a nutrient enrichment gradient. Half the number of plots from each level of nutrient treatment were placed under seaweed canopy cover from adult subtidal Fucus vesiculosus L., the dominant leathery macrophyte in the Baltic Sea. A parallel experiment at the field site including artificial shadings and one level of nutrient enrichment revealed that canopy cover prevented increases in understory net biomass accumula1 Corresponding author ([email protected]). 2 Present address: Department of Systems Ecology, Stockholm University, SE-106 91 Stockholm, Sweden. Acknowledgments We thank the Leibniz Institute for Marine Sciences for access to the field station in Maasholm, and S. Flöder, J. Chase, H. Rydin, M. H. Graham, and an anonymous reviewer for valuable comments on the manuscript. This study was supported by a postdoctoral grant from the Swedish Research Council for Environment, Agricultural Sciences, and Spatial Planning, Formas, through contract 20030275 (BKE). Limnol. Oceanogr., 51(4), 2006, 1813–1818 E 2006, by the American Society of Limnology and Oceanography, Inc.