Attributes of seaweeds and seagrasses that may influence seaweed impact on seagrass.

<p>We had <i>a priori</i> expectations about the direction of impact for the first seven attributes (above the dotted line). These directional hypotheses are based on simple rules; we expect a large impact when there is (a) <i>more</i> of a stressor (the seaweed) in either space or time, or (b) <i>less</i> of the impacted organism (the seagrass). Summary of tests-results are shown in the table (significant values in bold, see also <a href="http://www.plosone.org/article/info:doi/10.1371/journal.pone.0028595#pone-0028595-g001" target="_blank">Fig. 1</a>, <a href="http://www.plosone.org/article/info:doi/10.1371/journal.pone.0028595#pone-0028595-g002" target="_blank">2</a> and <a href="http://www.plosone.org/article/info:doi/10.1371/journal.pone.0028595#pone.0028595.s007" target="_blank">Appendix S3</a>; NT = not tested because data were inadequate).</p>1<p>Impact of seaweeds on seagrasses may also be modified by habitat attributes, including the resource levels (e.g., nutrients, light, O₂, space), abiotic conditions (e.g., temperature, salinity, desiccation, sedimentation, substrate conditions, day-length) and resident animals living in and around the seagrass habitat <a href="http://www.plosone.org/article/info:doi/10.1371/journal.pone.0028595#pone.0028595-Thomsen1" target="_blank">[21]</a>.</p>2<p>The categorical test based on experiments that explicitly tested for abundance effect was significant, but the correlation conducted across all experiments was not significant.</p>3<p>We assume that abundant seagrasses have more resources to withstand stress. Alternatively, abundant seagrass may suffer from intra-specific competition resulting in abundant seagrass being more susceptible to stress (i.e. the opposite expectation may be equally valid).</p>4<p>We assume that invaders have superior impact (seaweeds) and resistance (seagrass), e.g., as novel weapons <a href="http://www.plosone.org/article/info:doi/10.1371/journal.pone.0028595#pone.0028595-Catford1" target="_blank">[44]</a>.</p>5<p>Poor ‘condition/health’ of the seaweed results in decomposition and production of anoxia, sulphide and ammonia. Unattached mats often decompose when lower layers are shaded by higher layers.</p>6<p>For seagrasses, integration is a continuous attribute that encompasses below ground storage products and ability to translocate products between ramets.</p>7<p>Dri = Drift/unattached, Epi = epiphytic to seagrass leaves, Roo = rooted in sediment with rhizoids and rhizomes.</p>8<p>A few seagrasses can attach to rocks, but no studies have quantified seaweed impacts on attached seagrass.</p>9<p>Adapted from Littler and Littler (1980); She = sheets, Coa = Coarsely branches, Fil = filaments, Coe = coenocytic.</p>10<p>Ulv = <i>Ulva</i>, Gra = <i>Gracilaria</i>, Lau = <i>Laurencia</i>, Cau = <i>Caulerpa.</i></p>11<p>Amp = <i>Amphibolis</i>, Tha = <i>Thalassia</i>, Cym = <i>Cymodocea</i>, Zos = <i>Zostera</i>, Had = <i>Halodule</i>, Hap = <i>Halophila</i>.