Communities are influenced by multiple stressors that can interact in surprising ways, yet trampling studies typically ignore other sources of disturbance. While there is some evidence that disturbed communities may resist trampling because of species shifts, we lack an understanding of how previous disturbance interacts with trampling in monospecific stands. Furthermore, recent studies emphasized experimental trampling additions with assumed ecological realism. We monitored barnacle cover in unmanipulated and experimentally disturbed plots at 2 sites separated by only 300 m and with disparate levels of human visitation. During the summer, barnacle cover increased by 6 and 15% at the less-visited site but decreased by 5 and 4% at the heavily visited site (unmanipulated and disturbed plots, respectively). Interestingly, the influence of site on barnacle cover was greater in disturbed than in unmanipulated plots. New recruits represented a greater proportion of the barnacles at the heavily visited site. Neither plot slope nor predator abundance differed between these sites. To isolate the influence of trampling on barnacles, we monitored barnacle cover in caged, human exclusion plots and in uncaged controls at both sites. Summer barnacle cover increased in all treatments except in uncaged plots at the heavily visited site. During 9 paired surveys coinciding with the exclusion experiment, as many as 70 humans (km −1 shoreline h −1 ) entered the intertidal zone of the heavily visited site but no one did so at the less-visited site. Given that (1) visitation was the dominant factor explaining differences in barnacle cover between the 2 sites, and (2) caging enhanced barnacle cover only at the heavily visited site, we conclude that previous disturbance mediated the effects of human trampling on barnacle populations, with more disturbed plots displaying greater trampling vulnerability.
KEY WORDS: Multiple stressor effects · Human disturbance · Rocky shore · Intertidal
Resale or republication not permitted without written consent of the publisherMar Ecol Prog Ser 437: 165-173, 2011 166 for an exception). Previous disturbance may influence trampling resistance via at least 3 mechanisms. First, recently disturbed communities may be less susceptible to trampling because they are dominated by trampling-resistant species that survived the initial disturbance (Liddle 1991, Povey & Keough 1991, Brosnan 1994, Brown & Taylor 1999, Hill & Pickering 2009). For example, disturbed communities dominated by lawn grasses are 2 to 10× more resistant to trampling than undisturbed communities dominated by ferns or tussock grasses (Yorks et al. 1997, Hill & Pickering 2009). Second, disturbed populations may contain trampling-resistant genotypes (Warwick & Briggs 1978). For example, regularly mown bowling greens contained prostrate genotypes of Poa annua while less-disturbed populations contained a mixture of prostrate and erect genotypes (Warwick & Briggs 1978). Third, disturbed pop ulations may contain phenotypes that are di...