Asch and Collie: Changes in a benthic megafaunal community due to disturbance from bottom fishing 
451 
Shallow sites 
Figure 7 
Species richness (number of species) and Simpson’s index of diversity fl-A') for noncolo- 
nial epifauna at shallow and deep sites. Black squares indicate mean species richness 
or diversity for a given level of disturbance; standard error bars are shown. Different 
units of measurement are plotted on the x axis for shallow and deep sites, reflecting 
the fact that fishing effort was estimated by different methods in U.S. and Canadian 
waters. Results of linear regression for species richness versus disturbance at shallow 
sites (A): F=17.69; df=l,3; P= 0.02. All other linear and polynomial regressions were 
not significant. 
ing at shallow depths. Most sponges at our study sites 
had branched or mounding-forming structures, which 
contributed to their sensitivity to mobile fishing gear. 
Other characteristics that make sponges vulnerable to 
mobile fishing gear are aperiodic recruitment and a 
perennial life cycle that allows colonies to persist over 
several years (Hughes, 1989). In the southern portion of 
CA-II that is dominated by sandy substrate, sponges are 
one of the microhabitats whose abundance differs sig- 
nificantly between fished and unfished areas (Lindholm 
et al., 2004) as was also recorded in our study. Several 
studies from other regions have documented that spong- 
es are among the most sensitive phyla to bottom fishing 
(Van Dolah et al., 1987; Freese et al., 1999). 
Arborescent epifauna react inconsistently to bottom 
fishing, reflecting the fact that these organisms possess 
life history traits that both facilitate and hinder their 
recovery from physical disturbances. Past research in- 
dicates that the effect of bottom fishing on arborescent 
epifauna varies between regions and species (Auster 
et al., 1996; Bradshaw et al., 2002). Like sponges, hy- 
droids and bushy bryozoans have an emergent growth 
form that makes them vulnerable to being knocked 
over by trawls and dredges. Yet these types of arbores- 
cent epifauna do not generally grow as tall as sponges, 
thus providing them with a degree of resistance to 
disturbance. When mobile fishing gear do succeed in 
removing arborescent epifauna from their substrate, 
recolonization may occur quickly because these organ- 
isms are known for rapid growth and typically have 
short life spans, ranging in length from ten days to 
one year (Boero, 1983; Hughes, 1989). Other species of 
hydroids and erect bryozoans are perennial but exhibit 
seasonal regression. Despite the swift turnover of hy- 
droid and bushy bryozoan colonies, the spatial extent of 
recovery may be restricted because many species have 
limited larval dispersal (i.e., <50 m) or tend to settle 
close to parent colonies (Hughes, 1989; Bradshaw et al., 
2002). The dispersal potential of other species, such as 
hydroids in the family Sertularidae, is even more con- 
strained, because they do not have a pelagic medusa 
stage (Boero, 1983). Our study indicates that depth may 
be an additional factor affecting how hydroids react to 
disturbance caused by bottom fishing, because deeper 
