Driggers and Hannan: Efficacy of 2 common bait types in reducing bycatch of coastal sharks 
191 
The mean number of a given species captured on all sets, 
regardless of bait, was calculated. All sets where the 
total number of individuals captured was greater than 
3 standard deviations from the mean, were considered 
outliers and removed from analyses. For those species 
for which outliers were identified, chi-square tests based 
on revised expected values were conducted. Revised 
expected values were calculated by subtracting sets 
with identified outliers from the total number of sets 
conducted with the corresponding bait and then divid¬ 
ing the remaining number of sets with Atlantic mackerel 
bait by the number of sets with northern shortfin squid 
bait. All statistical tests were considered significant at 
an a level of 0.05. 
Results 
Of the 104 longline sets conducted for this study, 52 sets 
used northern shortfin squid as bait and 52 sets used 
Atlantic mackerel as bait (Fig. 1). There was no signifi¬ 
cant difference in the mean depth (£=-0.58, P=0„56) or 
temperature (£=-0.30, P- 0.77) between sets that used the 
2 types of bait (Table 1). Similarly, there was no signifi¬ 
cant difference in the median bottom salinity (W=1426.0, 
P=0.4G) among sampling locations where the 2 baits were 
used (Table 1). Soak times ranged from 57 to 65 min, and 
there was no significant difference between soak times on 
sets that used mackerel or squid for bait (£=0.28, P- 0.78) 
(Table 1). Among all sets, regardless of bait, 1572 fish from 
25 species were captured; however, 6 of these species made 
up 91.3% of the total catch. Those species were the Gulf 
smoothhound (Mustelus sinusmexicanus ), Atlantic sharp- 
nose shark, blacknose shark, biacktip shark, scalloped 
hammerhead (Sphyrna lewini), and red snapper (Table 2). 
Species that constituted less than 1% of the total catch or 
were caught on fewer than 10 sets were not considered in 
subsequent analyses. 
All captured fish had corresponding body length mea¬ 
surements with the exception of 23 specimens damaged 
by sharks (14 red snapper and 9 Atlantic sharpnose 
sharks). The mean or median fork length (FL) of all spe¬ 
cies examined did not differ by bait, with the exception 
of those for the Atlantic sharpnose shark, specimens of 
which were larger on sets with bait of Atlantic mackerel 
(median: 806 mm FL) than on sets with bait of north¬ 
ern shortfin squid (median: 781 mm FL) (W=49994.4, 
PcO.Ol) (Fig. 2, Table 2). However, there was a greater 
size range (323-970 mm FL) of Atlantic sharpnose 
sharks captured on sets baited with Atlantic mackerel 
compared with the size range on sets baited with north¬ 
ern shortfin squid (487-961 mm FL). 
The total catch of all species on each bait was not dif¬ 
ferent from the expected ratio of 1:1, with the exception 
of the catch of Atlantic sharpnose sharks, which were 
more frequently captured on sets baited with Atlantic 
mackerel (rc=488) than on northern shortfin squid-baited 
sets ( tz=302) (x 2 =43.32, PcO.Ol) (Table 2). Outliers indi¬ 
cating unusually large catches were detected for Atlan¬ 
tic sharpnose sharks, blacknose sharks, biacktip sharks, 
and scalloped hammerheads (Table 3). After removing 
outliers, Atlantic sharpnose sharks (x 2 =12.14, PcO.Ol) 
and blacknose sharks (% 2 =4.92, P=G,03) were caught 
more frequently on sets baited with Atlantic mackerel. 
However, there remained no significant relationship 
between bait type and catch rates of biacktip sharks 
(% 2 =0.02, P=0.87) or scalloped hammerheads (x 2 =0.01, 
P= 0.92). 
Discussion 
Results of our study indicate that for Gulf smoothhounds, 
biacktip sharks, scalloped hammerheads, and red snap¬ 
per, there was no difference in capture rates related to 
bait. However, preference for Atlantic mackerel bait over 
Table 1 
Comparison of abiotic variables and soak times associated with bottom longline sets conducted with baits of Atlantic mackerel 
{Scomber scombrus) or northern shortfin squid (Ilex illecebrosus) in the northern Gulf of Mexico during March-April 2017. Means 
are given with standard errors of the mean (SEs). Results from f-tests and Mann-Whitney (M-W) W tests are provided with their 
associated P-values in parentheses. 
Variable 
Bait type 
Range 
Mean (SE) 
Median 
£-value (P) 
M-W test (P) 
Depth (m) 
Mackerel 
15.3-83.1 
40.51 (2.22) 
39.0 
-0.58 (0.56) 
1425.5 (0.64) 
Squid 
15.9-77.3 
42.41 (2.34) 
37.3 
Temperature (°C) 
Mackerel 
19.8-23.0 
21.48 (0.13) 
21.4 
-0.30 (0.77) 
1369.5 (0.64) 
Squid 
19.5-23.1 
21.53 (0.13) 
21.5 
Salinity 
Mackerel 
31.3-36.4 
35.74 (0.14) 
36.2 
1426.0 (0.40) 
Squid 
32.5-36.5 
35.90 (0.10) 
36.2 
Soak time (min) 
Mackerel 
57.0-65.0 
60.61 (0.25) 
60.5 
0.28 (0.78) 
1308.0 (0.77) 
Squid 
57.0-65.0 
60.52 (0.24) 
61.0 
