424 
Fishery Bulletin 115(3) 
Table I 
Summary information from video surveys conducted with a remotely operated vehicle during 2012-2015 at 
the Pulley Ridge mesophotic coral ecosystem off southwestern Florida in the Gulf of Mexico. N/A indicates 
that information was not available for that year. 
Year 
No. of 1-km 2 
blocks surveyed 
Average distance 
surveyed (km) 
No. of grouper 
pits surveyed 
Dates of 
survey 
Depth 
range (m) 
Bottom temperature 
range (°C) 
2012 
10 
35.2 
80 
14-25 Aug 
62.8-75.5 
N/A 
2013 
15 
29.4 
41 
12-27 Aug 
60.3-93.9 
N/A 
2014 
17 
23.8 
35 
14-28 Aug 
63.1-86.1 
18.1-29.2 
2015 
28 
33.4 
52 
23 Aug-2 Sept 
59.3-105.5 
19.0-22.52 
were not abundant enough to examine individually. For 
all analyses, the replicate was each grouper pit, and 
average abundances (with standard errors) for each 
factor were calculated and compared. Significant dif- 
ferences existed when P<0.05. 
The small fish group consisted of taxa of small, ben- 
thic fish that typically reside inside the pit to use the 
structural complexity it offers. Taxa in the small fish 
category were cardinalfish ( Apogon spp.); damselfish 
( Chromis spp.); small sea basses (Serranidae), includ- 
ing anthiids (Anthiinae); wrasses ( Halichoeres spp.); 
and parrotfish (Labridae). Not only do these species use 
the grouper pits in the same manner, but they are all 
possible prey of lionfish and red grouper. Abundances 
of small fish were log e transformed to correct for nor- 
mality and then tested with one-way analysis of vari- 
ance (AN OVA) to examine the effects of year, predator 
presence or absence, region, and HAPC. Categories for 
each factor were the same as the multivariate analy- 
ses: year was 2012-2015; predator presence or absence 
was either no predator, lionfish only, red grouper only, 
or both; region was main ridge, off main ridge, central 
basin, or west ridge; and HAPC was either inside or 
outside the protected area. 
Taxa in the schooling fish category hover just above 
a grouper pit and usually travel in schools of greater 
than 50 individuals. This group included the striped 
grunt ( Haemulon striatum), the school bass ( Schul - 
tzea beta), and bonnetmouths (Haemulidae). Because 
of the nature of the abundance data for the schooling 
group (large variances among pits) and because trans- 
formations did not correct for normality, nonparamet- 
ric Kruskal-Wallis tests were used to test for effects 
of year, predator presence, region, and HAPC on the 
abundances of taxa in the schooling group. 
The final category, large fish, consisted of the larger 
taxa, most of which are managed by the Gulf of Mex- 
ico Fishery Management Council. This group included 
grunts ( Haemulon spp.), except for the striped grunt; 
snapper ( Lutjanus spp.); grouper ( Mycteroperca spp., 
Epinephelus spp., and the graysby, Cephalopholis cru- 
entata), except for the red grouper; triggerfish ( Batistes 
spp.); and the hogfish ( Lachnolaimus maximus). Again, 
transformations did not correct for non-normality this 
time because of low abundances; therefore, nonpara- 
metric Kruskal-Wallis tests were used to test the ef- 
fects of the 4 factors (year, predator presence, region, 
and HAPC) on the abundance of taxa in the large fish 
group. 
Results 
In August of each year between 2012 and 2015, 70 ran- 
dom blocks (each 1 km 2 ) were surveyed over the en- 
tire Pulley Ridge MCE region (Table 1). Within those 
blocks, 208 grouper pits were encountered and ana- 
lyzed. The number of grouper pits observed by year was 
80 in 2012, 41 in 2013, 35 in 2014, and 52 in 2015. The 
average distance surveyed per year was 30.5 km (SE 
2.5). The temperature range was 18.1-29.2°C in 2014 
and 19.0-22.5°C in 2015; however, these data were not 
available for 2012 and 2013 because a conductivity, 
temperature, and depth profiler was not on the ROV 
during those years. Depths sampled were slightly shal- 
lower in 2012 because dives that year were conducted 
primarily on the main ridge, which is shallower than 
the other regions. The central basin and west ridge ar- 
eas of the Pulley Ridge MCE are slightly deeper (10-30 
m) and were sampled during the remaining years of 
the surveys (Fig. 2). 
Grouper pits were distributed throughout the sam- 
pling region. Their locations are shown in Figure 2 by 
year, region, predator presence, and HAPC. In general, 
all the regions surveyed have very similar habitat, 
except the off main ridge region, which is east of the 
main ridge and is a predominately soft-bottom sub- 
strate mixed with rock rubble and cobble. The other 
regions (main ridge, central basin, and west ridge) 
all have low relief (0-1 m) and a substrate consisting 
primarily of rock pavement (probably old, dead coral 
plates), and rock rubble and cobble (5-20 cm rock), or 
a combination of the latter 2 substrate types. 
Bathymetric maps show grouper pits 8-15 m in di- 
ameter, 1-2 m deep, and evenly spaced about 100 m 
apart over much of the area. Up to 340 grouper pits 
were visible in a single 1-km 2 block on high-resolution 
bathymetric maps of main Pulley Ridge (NOAA 7 ). 
