Asch and Collie: Changes in a benthic megafaunal community due to disturbance from bottom fishing 
447 
T ransect depth 
A Shallow 
▼ Deep 
Shallow transects 
Site 16 
A Site 17 
♦ Site 17W 
• Site 18 
Deep transects 
A Site 10 disturbed 
A Site 10 undisturbed 
■ Site 13 disturbed 
□ Site 13 undisturbed 
O Site 20 undisturbed 
Figure 5 
Nonmetric multidimensional scaling (MDS) plot describing varia- 
tions in noncolonial species composition among photographic 
transects. MDS is an ordination technique that produces two- 
dimensional “maps” of samples reflecting differences in their bio- 
logical communities. The distance between points is proportional 
to the dissimilarity between different photographic transects in 
terms of their noncolonial epifaunal community structure. The 
axes on MDS diagrams are arbitrary and can be flipped and 
rotated. Stress refers to the ordination’s goodness-of-fit when 
multidimensioal data are projected onto a two-dimensional plane. 
(A) All transects, (B) Shallow transects, (C) Deep transects. 
The majority of noncolonial species were most abun- 
dant at undisturbed sites, but hard-shelled mollusks 
and scavengers tended to dominate in heavily fished 
areas. Significant variations in community structure 
were detected at both depths between different dis- 
turbance levels and years by means of ANOSIM tests 
(Tables 2A and 3A). A total of six taxa contributed to 
50% of the cumulative dissimilarity between disturbed 
and undisturbed sites at the shallow depth stratum 
(Table 2B). Of these six taxa, P. magellanicus, the green 
sea urchin ( Strongylocentrotus droebachiensis), and the 
hermit crabs Pagurus spp. exhibited higher levels of 
abundance at shallow sites inside CA-II. Scavenging 
seastars ( Asterias spp.) and bivalves (astartes Astarte 
spp. and the northern cyclocardia Cyclocardia borealis) 
were more abundant along shallow, disturbed transects 
recorded outside the closed area. The elevated numbers 
of the latter two species at site 18, as well as siphons 
from an unidentified, infaunal bivalve, explained why 
transects from site 18 were separated from other shal- 
low sites in the MDS (Fig. 5B). Of the four species 
that defined 88% of the similarity among undisturbed 
