224 
Fishery Bulletin 119(4) 
_ CD THI[,}|;=FL},U'TI PE. [D> DID >I DI DE DE D-IEEEE————————————————————————————— — — t  oaaa aST 
Standardized count 
0 25 
Vehicle range (m) 
Figure 6 
Change in standardized fish counts observed at selected 
reef sites in the Florida Middle Grounds of the benthic pur- 
suers guild as a function of vehicle range (meters) and by 
vehicle type. Survey vehicles used in this experiment were 
an autonomous underwater vehicle (AUV), a remotely 
operated vehicle (ROV), and a towed vehicle (TV). Predic- 
tions were made by using the GAM predict function in the 
R package mgcv, and parameters were set in the models as 
follows: first transect transited, the relative vehicle alti- 
tude from the seafloor was set at the middle level (2 m), 
and habitat complexity was set at the average qualitative 
rating (2.5). The gray shaded areas represent the 95% con- 
fidence intervals. Surveys were conducted during August 
2014 and July and August 2015. 
effects on relative abundance for RVA, transect number, 
and habitat complexity, but vehicle range was not signifi- 
cant. The TV model explained 26.8% of the deviance with 
an r” of 0.73 (Table 2). Results from the ROV model indi- 
cate significant effects on relative abundance for RVA, 
transect number, habitat complexity, and vehicle range. 
The ROV model explained 38.3% of the deviance with an 
r” of 0.14. There were not enough interactions with the 
AUV to analyze effects for the lateral movers guild. Lat- 
eral movers guild fish abundance decreased significantly 
with decreasing vehicle range to the MOUSS platform 
sampling volume (Fig. 7). Following passage of the TV, fish 
tended to return to the MOUSS platform sampling vol- 
ume; whereas, they did not return following ROV passage. 
Relative abundance of the lateral movers guild decreased 
with increasing number of transects surveyed with the 
ROV. Relative vehicle altitude above the seafloor had 
opposite effects by vehicle, with low transect numbers for 
the TV resulting in increased relative abundance and with 
increased transect number for the ROV resulting in 
increased relative abundance. This result might be due to 
the differences in habitat preferences between vermilion 
snapper (semi-pelagic) and porgies (benthic) having been 
mixed into the same model. 
From a qualitative perspective, the down movers guild 
exhibited negative reactions to the vehicles, and fish in this 
Standardized count 
0 
Vehicle range (m) 
Figure 7 
Change in standardized fish counts observed at selected 
reef sites in the Florida Middle Grounds of the lateral 
movers guild as a function of vehicle range (meters) and 
by vehicle type. Survey vehicles used in this experiment 
were a remotely operated vehicle (ROV) and a towed 
vehicle (TV). Predictions were made by using the GAM 
predict function in the R package mgcv, and parameters 
were set in the models as follows: first transect tran- 
sited, the relative vehicle altitude from the seafloor was 
set at the middle level (2 m), and habitat complexity 
was set at the average qualitative rating (2.5). The gray 
shaded areas represent the 95% confidence intervals. 
Surveys were conducted during August 2014 and July 
and August 2015. 
group fled down into available interstitial spaces in the 
local habitat but did not leave the MOUSS platform sam- 
pling volume (Fig. 8). Relative abundance of the down mov- 
ers guild decreased with decreasing vehicle range, and the 
steepness of the decline and subsequent increase appears to 
be dependent on vehicle type. For the down movers guild, 
results from the AUV model indicate significant effects on 
relative abundance for RVA, transect number, and habitat 
complexity, but range was not significant. The AUV model 
explained 98.0% of the deviance with an r” of 0.98 (Table 2). 
Results from the TV model indicate significant effects on 
down movers guild relative abundance for RVA, transect 
number, habitat complexity, and vehicle range. The TV 
model explained 60.8% of the deviance with an r” of 0.38. 
Results from the ROV model indicate significant effects on 
relative abundance for RVA, transect number, habitat com- 
plexity, and vehicle range. The ROV model explained 18.6% 
of the deviance with an r” of 0.14. Strongest responses were 
observed for the TV and ROV (fastest vehicles), and the 
weakest responses were observed for the AUV (slowest 
vehicle). For both the AUV and ROV, down movers guild 
relative abundance decreased with increasing transect 
number. In contrast, results from the TV model indicate 
that relative abundance increased through the first 5 tran- 
sects followed by decreasing relative abundance thereafter. 
Increasing RVA indicates increasing down movers guild 
