100 
Fishery Bulletin 110(1) 
32°46'0”N' 
32°45’0"N- 
32°44'0”N- 
32°43'0"N- 
32°42'0”N- 
" 0 “ 
-\V 
& 
<3* 
A* 
Charleston Harbor 
0 1 2 Kilometers 
D3 
79°52'0"W 79°51'0"W 79°50'0"W 79°49'0"W 79°48'0"W 79°47'0"W 79°46'0 ,, W 
Figure 1 
Trawling in the Charleston, South Carolina, shipping channel during 2004-2007 
was completed at seven index stations (A1-A3; Bl, B3; Dl, D3) within three arbi- 
trary blocks previously established by Van Dolah and Maier (1993). 
sured from the nuchal notch to the posterior tip of the 
carapace (SCLnt, in cm). SCLnt for five loggerheads 
captured with healed posterior carapace amputations 
was estimated using the following relationship between 
carapace length and maximum straight-line carapace 
width (SCW) determined for 1497 loggerheads <80 cm 
SCLnt captured in our various studies between 2000 
and 2010: SCLnt = 0.496 + 1.23(SCW); coefficient of 
determination (r 2 ) = 0.76. 
Blood samples were collected from the dorsal cervical 
sinus (Owens and Ruiz, 1980) with a 21-gauge, 3.5- 
cm needle to measure three standard health metrics: 
blood glucose (mg/mL), hematocrit (%), and serum pro- 
tein (g/dL) at sea. Blood samples were also analyzed 
in the laboratory to assess sex and genetic origin. Sex 
was assigned by using serum testosterone concentra- 
tions measured by radioimmunoassay, as described in 
Braun-McNeill et al. (2007) and considered reliable at 
water temperatures >23°C. Loggerheads with serum 
testosterone concentrations <450 pg/mL were identi- 
fied as female, and those between 450 and 550 pg/ 
mL, as undetermined, and those >550 pg/mL as male; 
however, two probable adult loggerheads >90 cm SCLnt 
with testosterone levels >1200 pg/mL were reclassified 
as female given tail length measurements consistent 
with adult females. Whole blood samples were prepped 
with lysis buffer solution before a 378 base-pair frag- 
ment of the mitochondrial DNA (mtDNA) control region 
was sequenced (see Roberts et al., 2005) to determine 
haplotypes for comparison with haplotypes reported 
for regional rookeries (Encalada et al., 1998; Bowen et 
al., 2004). 
Station data consisted of towing speed (in knots, kn) 
at the start of each trawling event; surface water tem- 
perature (°C); wave height (m); wind speed (kn); wind 
direction (numeric); cloud cover (%); and barometric 
pressure (millibars, mb). Surface water temperature 
was recorded with a transducer located on each ship’s 
hull approximately 1.5 m below the water surface. 
Wind direction was converted to a numeric value as 
follows: N (0°); NNE (22.5°); NE (45°); etc. Tide-stage 
data (15-min intervals) were obtained from the United 
States Geological Survey (USGS) for Fort Sumter, SC 
(station 02172100), which was located approximately 
2 km directly inshore of the shoreward boundary of 
the shipping channel survey area. Three metrics cor- 
responding to the start of each trawling event were 
subsequently computed: tide stage (ebb, flood); water 
level difference (m) between high and low tide; and 
the percentage of tide stage expired at the start of the 
trawling event. 
Invertebrate and fish bycatch captured during each 
trawling event were identified to the lowest possible 
taxon and the actual or estimated counts for each taxon 
were also recorded. Total counts of potential inverte- 
brate prey of turtles (Plotkin et al., 1993; Seney and 
Musick, 2007) per trawling event were included in mul- 
tivariate analyses as follows: blue crab (Callinectes 
