Lindholm et al.: Ecological effects of bottom trawling on fish habitat along the central California outer continental shelf 
91 
1.0 - 
i 
i 
0.8 - 
// 
i 
i 
i 
Power 
o 
(T> 
. L.. 
i 
i 
i 
i 
i 
0.4 - 
/ / 
i 
i 
0.2 - 
t 
i 
i 
i 
i 
1 1 1 1 I 
1 5 10 50 100 
Sample size 
1 1 1 
500 1000 5000 
Figure 5 
Power curves for Z-test conducted on percent cover of microtopographic 
features (solid curve) and /-tests conducted on densities of sessile and 
mobile invertebrates (dotted curve) from surveys conducted off central 
California between 2009 and 2012. The x-axis was log transformed to 
resolve differences between the 2 curves. The sample size for all 3 met- 
rics is also provided for overall tests (vertical dashed and dotted lines) 
and interannual comparisons (vertical dotted line). 
2009), global reviews (e.g., Auster and Langton, 1999; 
Hall, 1994; NRC, 2002; Barnes and Thomas, 2005), and 
the fact that, at a depth of 170 m, the entire study area 
was well below the effective depth of storm penetra- 
tion. For instance, elsewhere along the central coast of 
California, approximately 375 km to the north of Morro 
Bay, de Marignac et al. (2009) conducted a study at 
similar depths and with similar substrate composition 
and found that numbers of biogenic mounds and de- 
pressions were significantly higher in a recovering area 
than in an area that continued to be actively trawled. 
Yet the expectation of clearly discernible impacts on 
the seafloor was largely not borne out by the results of 
our study. After low-intensity trawling, the small but 
persistent difference in the microtopographic complex- 
ity of the seafloor between control and trawled plots 
was indicative of an impact from bottom trawling. We 
attributed this difference to the smoothing of habitat 
features by the trawl footrope as it passed over the bot- 
tom of the seafloor (Auster and Langton, 1999), as well 
as to the removal of the mobile organisms responsible 
for bioturbation of the sediment (Lohrer et al., 2004; 
Meysman et al., 2006). However, the difference was not 
statistically significant in our analyses, despite a high 
statistical power (Fig. 5). We expected that any impact 
from trawling would be most pronounced in this study 
after high-intensity trawling was conducted a year lat- 
er at the same study plots; however, the trajectories 
of the differences in microtopographic complexity at 
the control and trawled plots were even less clear than 
those for plots where low-intensity trawling was con- 
ducted, converging at 2 weeks after trawling, increas- 
ing while diverging significantly at 6 months, and then 
converging again at 1 year after trawling. 
Insofar as our trawling activities represented a type-I 
disturbance, where a relatively small disturbed area is 
surrounded on one or more sides by undisturbed habitats 
or organisms (Connell and Keough, 1985), it is possible 
that the lack of a significant impact to microtopographic 
structure on the seafloor resulted from the rapid colo- 
nization of the patches by bioturbating organisms from 
surrounding areas. This colonization may also explain 
why the temporal variation in the data was more pro- 
nounced than the spatial variation. Despite the minimal 
reduction in microtopographic complexity observed in 
the trawled plots, we did find, on a larger scale, altera- 
tion of the seafloor in the form of scour marks from trawl 
doors that were visible immediately after both low- and 
high-intensity trawling and that persisted for up to a 
year after low-intensity trawling. 
