Canadian Journal of Fisheries and Aquatic Sciences
! "#
$"
( #
)
%
%
&
%
'
(
*
+, !
!
) / *! / +!
$. /
$0 *
+, / $. /
*(
)
*2 $ +. /
( 0
3 0# *
+!#
4
. /
"
*
+. /
(
/
3 55
*
%+
) $ 0
) "
$. /
% 1 /
*
#
6
$
"
*
af
Dr
2
*
$* #
% #
$
t
https://mc06.manuscriptcentral.com/cjfas-pubs
*( 7
Page 1 of 36
Canadian Journal of Fisheries and Aquatic Sciences
1
2
3
4
Aaron B. Carlisle1*, Steven Y. Litvin1, Daniel J. Madigan 2, Kady Lyons3, Jennifer S. Bigman4, Melissa
5
Ibarra5 & Joseph J. Bizzarro6
6
7
1
Hopkins Marine Station of Stanford University, 120 Oceanview Blvd, Pacific Grove, CA, USA 93950
8
2
Harvard University Center for the Environment, 24 Oxford Street, Cambridge, MA, USA 02138
9
3
University of Calgary, 2500 University Dr NW, Calgary, Alberta, Canada T2N 4N1
10
4
Simon Fraser University, 8888 University Drive, Burnaby, British Columbia, Canada V5A 1S6
11
5
University of California Davis, One Shields Avenue, Davis, CA, USA 95616
12
6
Moss Landing Marine Laboratories, 8272 Moss Landing Road, Moss Landing, CA 95039
13
*
Corresponding author
Dr
14
Aaron B. Carlisle – aaroncar@stanford.edu
16
Steven Y. Litvin – litvin@stanford.edu
17
Daniel J. Madigan – danieljmadigan@fas.harvard.edu
18
Kady Lyons – kady.lyons@sbcglobal.net
19
Jennifer S. Bigman – jbigman@sfu.ca
20
Melissa Ibarra – mibarra08@yahoo.com
21
Joseph J. Bizzarro – jbizzarro@mlml.calstate.edu
t
af
15
22
23
24
25
26
27
28
29
1
https://mc06.manuscriptcentral.com/cjfas-pubs
Canadian Journal of Fisheries and Aquatic Sciences
Page 2 of 36
30
Stable isotope analysis (SIA) is becoming a commonly used tool to study the ecology of elasmobranchs.
32
However, the retention of urea by elasmobranchs for osmoregulatory purposes may bias the analysis and
33
interpretation of SIA data. We examined the effects of removing urea and lipid on the stable isotope
34
composition of fourteen species of sharks, skates, and rays from the eastern North Pacific Ocean. While
35
effects were variable across taxa, removal of urea generally increased δ15N and C:N. Urea removal had
36
less influence on δ13C, whereas extracting urea and lipid generally increased δ15N and C:N while also
37
increasing δ13C. Because C:N values of nonFextracted tissues are often used to infer lipid content and
38
adjust δ13C, shifts in C:N following urea extraction will change the inferred lipid content and bias any
39
mathematical adjustment of δ13C. These results highlight the importance of urea and lipid extraction and
40
demonstrate the confounding effects of these compounds, making it impossible to use C:N of nonFureaF
41
extracted samples as a diagnostic tool to estimate and correct for lipid content in elasmobranch tissues.
42
43
Stable isotopes, urea, lipid, carbon, nitrogen, C:N, elasmobranch, mathematical lipid
correction, elasmobranch
t
44
af
Dr
31
45
46
47
48
49
50
51
52
53
54
55
2
https://mc06.manuscriptcentral.com/cjfas-pubs
Page 3 of 36
Canadian Journal of Fisheries and Aquatic Sciences
56
57
Stable isotope analysis (SIA) uses the stable isotope composition of organismal tissue to
58
understand a diverse suite of biological and ecological processes. SIA is increasingly being used to
59
investigate the ecology of marine taxa (Peterson and Fry 1987, Michener and Kaufman 2007), including
60
sharks, skates, and rays (elasmobranchs) (Hussey et al. 2012b). Since SIA makes inferences based on the
61
chemical composition of tissues, certain compounds found in specific taxa can interfere with analysis and,
62
therefore, conclusions. Here, we investigate the effects of urea and lipid extraction on tissues from
63
fourteen elasmobranch species and report results that demonstrate the necessity to account for these
64
compounds when using SIA in elasmobranch studies.
65
The physiology and anatomy of elasmobranchs present unique challenges when applying SIA to
study their ecology. In particular, elasmobranchs retain urea ((NH2)2CO) and trimethylamine oxide
67
(TMAO (C3H9NO)) in their tissues for osmoregulatory processes (Ballantyne 1997, Olson 1999, Hazon et
68
al. 2003). This retention of urea can differentially bias stable isotope results depending upon the tissue
69
type examined (Hazon et al. 2008, Kim and Koch 2011, Hussey et al. 2012b, Churchill et al. 2015). As a
70
waste product, urea is expected to have low δ15N values (Minagawa and Wada 1984, Balter et al. 2006)
71
because 14N is preferentially concentrated in urea by deaminases and transaminases (Gannes et al. 1998).
72
We were unable to find any comparable data on TMAO, but as a waste product it also would be expected
73
to be depleted in 15N. As a result, the relative concentrations of urea and TMAO in a tissue may influence
74
the δ15N value of that tissue. As urea and TMAO (hereafter referred to together as urea) both contain
75
carbon, they could potentially affect δ13C. Kim and Koch (2011) reported that the carbon in urea is
76
enriched in δ13C in some terrestrial taxa; however information on the isotopic composition of these waste
77
products, especially in aquatic taxa, remains lacking. Further complicating the effect of urea on SIA is its
78
varying concentration within organisms, which is influenced by a variety of factors including tissue type
79
(Ballantyne 1997), ambient salinity (Hazon et al. 2003, Pillans et al. 2005) and diet (Wood et al. 2010).
80
Information on how to address the effects of urea on SIA results is needed, both in terms of appropriate
t
af
Dr
66
3
https://mc06.manuscriptcentral.com/cjfas-pubs
Canadian Journal of Fisheries and Aquatic Sciences
81
sample treatment methodology and data interpretation (Martinez del Rio et al. 2009, Logan and
82
Lutcavage 2010, Kim and Koch 2011, Hussey et al. 2012b, Li et al. 2016).
83
Page 4 of 36
In addition to the potential effect of urea on the stable isotope composition of elasmobranchs, the
84
presence of lipids is known to influence the δ13C values of tissues (Post et al. 2007, Martinez del Rio et al.
85
2009, Hussey et al. 2012a). Because lipids are depleted in 13C relative to protein, the presence of lipid in
86
tissues can bias δ13C values and increase the tissue carbonFtoFnitrogen ratio (C:N) (Pinnegar and Polunin
87
1999, Post et al. 2007). Tissue samples with high lipid concentrations have lower δ13C values than
88
samples of the same tissue with lipids removed (Post et al. 2007). To account for variation in lipids across
89
tissue types, researchers either chemically extract or mathematically correct for lipids based on the tissue
90
C:N, which has been used as a proxy for relative lipid content in tissues (Post et al. 2007).
91
The influence of lipid content on SIA data of elasmobranch tissues has been relatively well
studied (Kim and Koch 2011, Hussey et al. 2012a) compared to that of urea (Hussey et al. 2012b). Logan
93
and Lutcavage (2010) and Kim and Koch (2011) directly assess the effects of urea extraction on SIA data
94
of elasmobranchs. Logan and Lutcavage (2010) reported no effect of urea extraction on elasmobranch
95
tissues, whereas Kim and Koch (2011) reported a significant increase in δ15N in ureaFextracted tissues.
96
However, treatment methods differed between studies, with Kim and Koch (2011) using a more extensive
97
deionized water (DIW) extraction, which potentially resulted in more complete urea removal. Given that
98
lipid has a high C:N and urea has low C:N (0.5), removal of these compounds will influence tissue C:N.
99
Several studies examining the effect of lipid extraction on elasmobranch tissue noted increases in δ15N
t
af
Dr
92
100
and C:N following lipid extraction in a manner consistent with the removal of urea, suggesting that lipid
101
extraction may effectively remove urea as well as lipid (Hussey et al. 2010, Kim and Koch 2011, Hussey
102
et al. 2012a, Churchill et al. 2015, Li et al. 2016). However Kim and Koch (2011) reported that
103
elasmobranch tissues should have both urea and lipidFextracted to obtain the most reliable results. Li et al.
104
(2016) recently conducted the most thorough study of the interactive effects of urea and lipid toFdate,
105
examining the effects of urea and lipid extraction on six species of pelagic sharks. They reported
106
significant increases in δ15N and C:N following lipid extraction, urea extraction, and lipid and urea
4
https://mc06.manuscriptcentral.com/cjfas-pubs
Page 5 of 36
Canadian Journal of Fisheries and Aquatic Sciences
107
extraction, with δ13C also increasing significantly in treatments with lipid extraction. Li et al. (2016)
108
supported the conclusion of Kim and Koch (2011) that both urea and lipid should be removed when
109
analyzing elasmobranch tissues for SIA.
110
Despite the reported shifts in C:N following urea and lipid extraction and recommendations to
make lipid extraction a standard practice when processing elasmobranch tissues for SIA, estimating lipid
112
content based on the C:N of unextracted bulk samples remains a common practice. Specifically, it is
113
typically assumed that tissues of aquatic organisms with C:N values < 3.3 – 3.5 have low lipid content
114
and do not warrant lipid extraction (Post et al. 2007). This is of potential concern since this does not
115
account for the influence of urea on C:N, which is then used to mathematically adjust δ13C to account for
116
inferred (based on C:N) lipid content. As a result, the δ15N and potentially δ13C values of samples
117
processed without urea and/or lipid extraction may be biased, with any resulting analyses or ecological
118
interpretations being potentially based on inaccurate δ15N and δ13C values. Given the common use of δ15N
119
to estimate trophic level and δ15N and δ13C to understand habitat use and trophic relationships in
120
elasmobranchs (Fisk et al. 2002, Estrada et al. 2003, Dale et al. 2011, Kim and Koch 2011, Vaudo and
121
Heithaus 2011, Carlisle et al. 2012, Hussey et al. 2012b), these biases may have important effects on the
122
ecological interpretation of SIA data.
t
af
123
Dr
111
To examine how common this issue may be, we surveyed 50 recent scientific publications (2013F
124
present) that used SIA to study the ecology of elasmobranchs (Google scholar; search terms: “stable
125
isotope”, “elasmobranch”, “shark”, “ray”; selected the first 50 pertinent results, Table S1). We found 28%
126
used low C:N values (< 3.5) of tissues containing urea to support not extracting lipid from their samples,
127
another 16% used mathematic corrections to adjust δ13C based on lipid content estimates inferred from
128
C:N values of tissues containing urea, and 12% did not account for urea or lipid in any manner. Thus,
129
56% of the surveyed studies potentially had results biased due to not accounting for the combined effects
130
of urea and lipids. While it is not possible to know if the lack of urea or lipid extraction had any
131
meaningful effect on the isotopic results or their interpretation in these studies, it is clear previous
5
https://mc06.manuscriptcentral.com/cjfas-pubs
Canadian Journal of Fisheries and Aquatic Sciences
132
recommendations to make urea and lipid extraction the standard practice when analyzing elasmobranch
133
tissues for SIA (Kim & Koch 2011, Hussey et al. 2012a, Li et al. 2016) have not been fully adopted.
134
Page 6 of 36
In this study, we expand upon previous work to better understand and account for the interactive
effects of urea and lipids on SIA in elasmobranch tissues. The importance of urea and lipid extraction has
136
been demonstrated in leopard sharks (Triakis semifasciata, Kim and Koch (2011) and a suite of pelagic
137
sharks (Li et al. 2016), yet published studies commonly do not appropriately account for the potential
138
effects of urea and lipid on elasmobranch tissues. In addition, the effects of applying multiple chemical
139
treatments to remove urea and lipid, and how to determine an appropriate methodological course for
140
individual species or taxa of interest, remain unaddressed for many elasmobranch taxa across broad
141
ranges of tissue compositions and habitats. In particular, while the interactive effects of urea and lipid
142
have been explored in sharks, they have not been investigated in batoids (skates and rays), a group that
143
comprises over 50% of extant elasmobranchs (Dulvy et al. 2014). Finally, it may not always be feasible or
144
desirable to perform lipid extraction (i.e. avoid the cost, chemical waste generated, and time associated
145
with chemical extraction or to preserve information on the movement of lipids through foodwebs). While
146
Li et al. (2016) provides species specific isotopic correction models to account for urea content in lipid
147
extracted samples, there is a lack of specific guidance in the literature on the appropriate development and
148
application of mathematical correction models for δ13C based on inferred tissue lipid content (C:N) for
149
elasmobranchs that account for urea’s effects on C:N, δ13C and δ15N.
t
af
150
Dr
135
The goals of this study were to 1) assess the relative effects of urea and lipid extraction on the
151
stable isotope composition of muscle tissue from a variety of shark and batoid species, including pelagic,
152
demersal and benthic species, with variable lipid content, 2) address the utility of using C:N as a
153
diagnostic tool to understand and adjust for lipid content in elasmobranch tissue, particularly in the
154
context of the influence of urea on C:N, 3) develop models to mathematically adjust δ13C of ureaF
155
extracted samples to account for lipid content and 4) provide a conceptual framework to understand how
156
urea and lipid interact to influence SIA results in elasmobranchs in order to facilitate proper application of
157
the technique.
6
https://mc06.manuscriptcentral.com/cjfas-pubs
Page 7 of 36
Canadian Journal of Fisheries and Aquatic Sciences
158
159
To examine the effects of urea extraction and lipid extraction on elasmobranch tissues, we
collected white muscle samples from fourteen species, including six species of sharks from the families
161
Lamnidae, Carcharhinidae, Squalidae, and Triakidae, and eight species of batoids (skates, rays, and their
162
allies) from the families Arhynchobatidae, Rajidae, Myliobatidae, Gynmnuridae, Dasyatidae, and
163
Urolophidae. All samples were collected in the eastern North Pacific, primarily off California but ranging
164
as far north as the Gulf of Alaska. Samples were collected from juvenile salmon sharks (Lamna ditropis)
165
stranded on beaches in California and Oregon as described by Carlisle et al. (2015). Juvenile white sharks
166
(Carcharodon carcharias) caught as incidental bycatch in the coastal gillnet fisheries in southern
167
California were sampled as part of the Monterey Bay Aquarium juvenile white shark research program as
168
described in Mull et al. (2012). Samples from shortfin makos (Isurus oxyrinchus), blue sharks (Prionace
169
glauca), and pelagic rays (Pteroplatytrygon violacea) were collected during the annual National Oceanic
170
and Atmospheric Administration (NOAA) Juvenile Shark Longline Survey off southern California.
171
Round stingray samples (Urobatis halleri), leopard shark (Triakis semifasciata), butterfly ray (Gymnura
172
marmorata), and bat rays (Myliobatis californica) were collected in southern California as described by
173
Lyons et al. (2014). Skates (Bathyraja aleutica, Bathyraja interrupta, Beringraja binoculata, Raja rhina)
174
were collected from the western Gulf of Alaska as described in Bizzarro et al. (2014). Spotted spiny
175
dogfish (Squalus suckleyi) were collected off central California during the National Marine Fisheries
176
Service Northwest Fisheries Science Center West Coast Groundfish Bottom Trawl Survey. All muscle
177
samples were collected from the dorsal musculature and stored frozen (F20°C) until processed and
178
analyzed.
t
af
179
Dr
160
Individual tissue samples were homogenized and subdivided into three parts for analysis, with
180
one as the control sample (Control), one for urea extraction (U), and one for both urea and lipid extraction
181
(UL). Methods to process tissues and extract lipids and urea generally followed Kim and Koch (2011). To
182
extract urea, tissue samples were placed in scintillation vials and rinsed three times in 10 mL of DIW
183
(Kim and Koch 2011). A rinse entailed sonication for 15 minutes and then decanting the supernatant.
7
https://mc06.manuscriptcentral.com/cjfas-pubs
Canadian Journal of Fisheries and Aquatic Sciences
184
Lipids were extracted from all tissues except skate tissues using a 2:1 chloroform:methanol solution
185
(Bligh and Dyer 1959, Logan and Lutcavage 2010) by immersing tissues in the solution for 24 hours in
186
glass scintillation vials (Bligh and Dyer 1959, Logan and Lutcavage 2010). Following both treatments,
187
tissue samples were lyophilized and homogenized using a Spex/CertiPrep 5100 mill.
Page 8 of 36
Skate samples were processed slightly differently than the other species as part of another study.
189
For urea extraction of skate tissue, 10 mL of DIW were added to each homogenized sample, and then the
190
samples were mixed using a vortex mixer (Fisher Scientific). After 30 minutes, the sample was
191
centrifuged and the supernatant was decanted. For samples that were lipid and ureaFextracted, lipids were
192
extracted using Petroleum Ether (PE) following Kim and Koch (2011). Briefly, samples were immersed
193
in PE, mixed in a vortex mixer and left uncapped in a fume hood for 8 hours, centrifuged for 10 minutes,
194
and the supernatant decanted. The sample was then rinsed in DIW using the method described for urea
195
extraction of skate tissue. Following urea or urea and lipid extraction samples were dried in an oven at
196
60˚C for 24 hours.
af
197
Dr
188
For each treatment, approximately 500 Qg of tissue was weighed into tin boats and analyzed at
the Stable Isotope Laboratory at the University of California Santa Cruz (UCSC) using an elemental
199
analyzer coupled to an isotope ratio monitoring mass spectrometer (Delta XPFEA,ThermoF Finnagen
200
IRMS). For skate and dogfish samples, 500 Qg of tissue was weighed into tin boats and analyzed at Idaho
201
State University (ISU) using an elemental analyzer coupled to an isotope ratio monitoring mass
202
spectrometer (Elemental Combustion System (ECS) 4010 interfaced with a Delta V Advantage mass
203
spectrometer through the ConFlo IV System). Isotopic composition is expressed using standard δ
204
notation, using Vienna Pee Dee Belemnite limestone as the standard for carbon and AIR for nitrogen. For
205
runs at UCSC, analytical precision, based on an internal lab standard (Pugel), was 0.11‰ for δ15N and
206
0.07‰ for δ13C across multiple runs. For runs at ISU, analytical precision, based on internal lab standards
207
of ISU Peptone, Costech Acetanilide, and DORMF3, was 0.08, 0.04, and 0.04 ‰ for δ15N respectively and
208
0.05, 0.05, and 0.04 for δ13C, respectively. Where parametric assumptions were met (assessed with OneF
209
Sample KolmogorovFSmirnov and Levene’s Tests and visual inspection of residuals) a single factor
t
198
8
https://mc06.manuscriptcentral.com/cjfas-pubs
Page 9 of 36
Canadian Journal of Fisheries and Aquatic Sciences
210
ANOVA, followed by Tukey’s post hoc tests, was used to test for differences in δ13C and δ15N among
211
treatments for species with sample sizes > 3. When assumptions were not met for δ13C, δ15N, or C:N
212
differences were tested using MannFWhitney 2Fsample tests with sequential Bonferroni adjustments (Rice
213
1989). To show the magnitude and direction of the effects of treatments U and UL, differences between
214
treatment and control samples (U – Control, UL – Control, UL F U) were calculated.
215
We considered four previously used lipid correction models (Post et al. 2007, Logan et al. 2008,
Reum 2011) to examine the utility of using C:N as a diagnostic tool to understand and adjust for lipid
217
content in ureaFextracted elasmobranch tissue. Lipid correction models were used to characterize U13C,
218
the difference between lipid and urea (UL) and ureaFextracted (U) δ13C values (U13C = δ13CU F δ13CUL) as
219
a function of the C:N of ureaFextracted tissue (C:NU). The first (model 1) is a threeFparameter model
220
derived by Logan et al. (2008) from McConnaughey and McRoy (1979): U13C = (aC:NU + b)( C:NU + c)F1
221
, where a, the yFasymptote, corresponds to proteinFlipid δ13C discrimination and –baF1, the xFintercept, is
222
the urea and lipid free C:N value (C:NUL), and bcF1, the y'intercept, is the value of U13C at C:NU = 0. The
223
second (model 2) is a two parameter model (Fry 2002): U13C = P F PF(C:NU)F1, where P represents
224
proteinFlipid δ13C discrimination and F is C:NUL. The third and fourth are linear models: (model 3, Logan
225
et al. 2008) U13C = β0 + β1Ln(C:NU) and (model 4, Post et al. 2007) U13C = b + aC:NU, where
t
227
af
226
Dr
216
(
)
and –baF1 are estimates of C:NUL, respectfully.
We modeled the relationship between C:NU and U13C for five groups: all species, batoids, all
228
sharks other than S. suckleyi, S. suckleyi, and C. carcharias. We modeled S. suckleyi independently since
229
its lipid content was higher than all other taxa and its ureaFextracted samples had the widest range of C:N
230
values (Results). For C. carcharias, we wanted to attempt to develop a speciesFspecific relationship and
231
this species had the largest sample size. To compare the performance of potential lipid correction models,
232
the corrected Akaike Information Criterion, AICc (Burnham and Anderson 2002), was calculated for each
233
model. The model with the lowest AICc is considered the best fit, but any model(s) with AICc values
234
within two units of the lowest value have strong support as well (Burnham and Anderson 2002). In
9
https://mc06.manuscriptcentral.com/cjfas-pubs
Canadian Journal of Fisheries and Aquatic Sciences
235
addition for those models that preformed best based on AICc, we calculated the mean and standard
236
deviation of the absolute values of the residuals errors, to further evaluate model fit, and compared
237
estimates of proteinFlipid δ13C discrimination and C:NUL (Logan et al. 2008, Reum 2011). All models
238
were fitted with leastFsquares procedures using R and the libraries nlme and AICcmodavg (www.rF
239
project.org).
Page 10 of 36
240
241
The removal of urea and lipid from elasmobranch tissue influenced δ15N, δ13C, and C:N in most
species, although the direction and magnitude of the effects varied by species and with the lipid content of
243
the sample (Table 1, Figures 1F2, Figure S1). The C:N of control samples was consistently low for all
244
species, with 13 of the 14 taxa having values < 3.5 (mean ± SD of all species 3.0 ± 0.4). The only
245
exception to this was S. suckleyi, which had a high C:N of 4.5 due to higher lipid content of its muscle.
246
Urea extraction (treatment U) generally increased δ15N and C:N, but generally did not
Dr
242
significantly change δ13C, results that are consistent with the removal of isotopically light nitrogen present
248
in urea (Table 1, Figures 2a and 3a). In seven of the ten species (4 of 5 sharks, 3 of 5 batoids) that were
249
statistically tested, δ15N increased significantly following urea extraction (mean 0.8‰ ± 0.2). This result
250
was very similar to the overall trend across all taxa, which showed an average increase of 0.7‰ ± 0.2.
251
Three taxa had increases in δ15N greater than 1‰ (B. aleutica 1.1‰, L. ditropis 1.1‰ and I. oxyrinchus
252
1.0 ‰). δ13C only changed significantly in B. binoculata (F0.9‰) and U. halleri (F0.5‰). Overall, there
253
was a consistent, though generally nonFsignificant, decrease in δ13C across the batoids that was not
254
evident in sharks (mean F0.4‰ for batoids, 0.0‰ for sharks, and F0.2‰ ± 0.3 for all taxa). C:N increased
255
significantly in nine of the ten taxa statistically tested (4 of 5 sharks, 5 of 5 batoids), with the exception
256
being S. suckleyi, which had a high initial C:N that increased from 4.5 to 5.6 (+ 1.1‰ ) following urea
257
extraction. Overall, C:N increased by an average of 0.7 (± 0.2) across all taxa following urea extraction.
258
C:N values increased to values above 3 (mean C:N of 3.6) in all species, and the C:N of C. carcharias, L.
259
ditropis, G. marmorata, and U. halleri increased to values above 3.5.
t
af
247
10
https://mc06.manuscriptcentral.com/cjfas-pubs
Page 11 of 36
260
Canadian Journal of Fisheries and Aquatic Sciences
Extracting urea and lipid (treatment UL) from samples consistently increased δ15N and C:N in a
fashion similar to what was observed with urea extraction only, while also generally increasing δ13C
262
(Table 1, Figures 2b and 3b). Seven out of the ten taxa tested had significantly higher δ15N values
263
following urea and lipid extraction, although all taxa showed some increase (mean 0.8‰ ± 0.2). L.
264
ditropis showed the largest increase in δ15N following lipid extraction (1.1‰). Four of the five
265
statistically tested sharks had significantly higher δ13C following urea and lipid extraction, with dogfish
266
(mean 2.0‰), salmon sharks (1.5‰) and white sharks (1.1‰) having the largest increases. Three batoids
267
showed a decrease in δ13C following urea and lipid extraction, with B. binoculata exhibiting a significant
268
decrease (mean F0.9‰). Except for S. suckleyi, all taxa exhibited an increase in C:N (mean increase 0.3 ±
269
0.3, mean of species C:N 3.3), with nine of ten taxa tested statistically having significant changes. C:N of
270
S. suckleyi decreased, though nonFsignificantly, following urea and lipid extraction (control C:N 4:5, UL
271
C:N 3.9).
272
Dr
261
The differences between the U and UL treatments were more obvious in sharks than in the
af
batoids, in which the differences were relatively small (Table 1). All taxa showed an increase in δ13C in
274
the UL treatment relative to the U treatment, although only three of the five sharks (C. carcharias, L.
275
ditropis and S. suckleyi), and one of five batoids (U. halleri), had significant increases. There were no
276
significant differences in δ15N between U and UL treatments in any species examined (p > 0.05),
277
indicating that lipid extraction did not affect δ15N. C:N was generally lower in the UL treatment relative
278
to the U treatment, especially in the sharks. In the five shark species tested, all had significant decreases
279
in C:N in UL treatments relative to U treatments, whereas only two of the five tested batoids had
280
significant decreases.
281
t
273
Lipid correction models were created to adjust ureaFextracted tissue δ13C to account for lipid
282
content. Model performance varied across elasmobranch groups (Figure 3, Figure S2), with no single
283
model amongst those exhibiting the lowest AICc values across all groups (see supplementary Table S2 for
284
AICc Values, r2 (linear models only) and model parameters). For all species pooled, models 1 and 2 (nonF
285
linear models) had the lowest AICc, with identical mean ± SD of the absolute values of the residual errors
11
https://mc06.manuscriptcentral.com/cjfas-pubs
Canadian Journal of Fisheries and Aquatic Sciences
Page 12 of 36
(MRE, 0.29 ± 0.24) and similar estimates of proteinFlipid δ13C discrimination (5.39 and 6.12) and C:NUL
287
(3.20 and 3.18). All four models for sharks (excluding S. suckleyi) had similarly low AICc values (i.e.
288
within 2 units), MREs (0.20 – 0.25 ± 0.25 – 0.29) and estimates of C:NUL (3.10 – 3.13). However,
289
estimates of proteinFlipid δ13C discrimination for sharks varied widely between models 1 (1.94) and 2
290
(7.86, Table S2). The two linear models (3 and 4) performed equally for batoids (Table S2), with identical
291
MRE (0.21 ± 0.16) and estimates of C:NUL (3.27). Model performance varied between the two single
292
species groups. For S. suckleyi, which had both the widest range and highest C:N values, model 2
293
provided the singular best fit (MRE 0.26 ± 0.24) with estimates of proteinFlipid δ13C discrimination and
294
C:NUL of 6.31 and 3.35, respectively (Table S2) . In contrast, for C. carcharias, which had only two ureaF
295
extracted C:N values > 4, models 2, 3 and 4 had similarly low AICc, MREs (0.12 – 0.13 ± 0.12 – 0.13)
296
and estimates of C:NUL (3.05 – 3.14). The estimate of urea extracted proteinFlipid δ13C discrimination for
297
C. carcharias was the highest in the study (8.30, Table S2).
298
af
Dr
286
Our results provide further evidence of the substantial, direct affect that urea can have on δ15N.
300
We also show the important, and often unconsidered, indirect role urea plays in influencing δ13C values
301
by lowering the C:N, effectively masking lipid content and leading to inaccurate assessments of lipid
302
content and inappropriate mathematical corrections (Figure 4). Our results indicate that urea must be
303
removed to obtain reliable δ15N and C:N values, and that only with ureaFextracted tissues can C:N be used
304
as a diagnostic tool for understanding and mathematically adjusting for lipid content.
305
t
299
Urea extraction resulted in an increase in δ15N across all taxa, and a significant increase in 7 of 10
306
statistically tested taxa, ranging from ~0.5 to 1.1‰ (mean 0.7 or 0.8‰ for U and UL respectively), with
307
urea and lipid extraction producing similar changes. When using δ15N to infer trophic level, this shift is
308
equivalent to an inferred trophic level difference of ~22 – 50% or ~15 – 30% assuming a trophic
309
discrimination factor for nitrogen of 2.3‰ (Hussey et al. 2010) or 3.7‰ (Kim et al. 2012), respectively.
310
This shift is similar to that reported by Hussey et al. (2012a) for elasmobranch tissues following lipid
311
extraction as well as by Li et al. (2016) following both urea extraction and urea and lipid extraction.
12
https://mc06.manuscriptcentral.com/cjfas-pubs
Page 13 of 36
Canadian Journal of Fisheries and Aquatic Sciences
312
Clearly the presence of urea in analyzed tissues will directly bias the use of δ15N as a tracer in ecological
313
studies, whether it is being used to assess trophic level, reconstruct diet, habitat or migration patterns, or
314
even for simple qualitative comparisons.
315
The urea effects we describe here also have important implications for the use of C:N as a metric
of lipid content of elasmobranch tissue. The C:N of nonFextracted elasmobranch tissues are consistently
317
very low (< 3) across studies (Logan and Lutcavage 2010, Matich et al. 2010, Dale et al. 2011, Kim and
318
Koch 2011, Vaudo and Heithaus 2011, Hussey et al. 2012a, Hussey et al. 2012b), and are often much
319
lower than would be expected of pure protein. Frequently, when tissues with C:N values < 3.5 are
320
assumed to have little lipid content (Post et al. 2007), the low C:N values are used to infer that lipid
321
extraction is not warranted. However, results from this study and previous work (Kim and Koch 2011,
322
Hussey et al. 2012a, Li et al. 2016) demonstrate that urea extraction generally increases the C:N as
323
nitrogen is removed (Figures 1 and 2). In this study, extracting urea through DIW rinses caused
324
significant increases in C:N in every species tested, increasing it by as much as 1.2‰ (mean 0.7‰).
af
325
Dr
316
Removal of the nitrogen contributed by urea will increase the C:N value of a sample, thereby
changing the estimated lipid content that are based on C:N (Figure 4). Following removal of urea, C:N
327
can increase from very low values to values above threshold levels that are used to indicate low lipid
328
content (e.g. 3.5). In effect, the presence of urea and its lowering of C:N has the potential to mask lipid
329
content. In four of the species examined in this study (C. carcharias, L. ditropis, G. marmorata and U.
330
halleri) the C:N value shifted from values < 3 to values > 3.5, and other species had C:N values of ~3.3
331
following DIW rinses, which is similar to pure protein values (Post et al. 2007). The interpretation of
332
these ureaFextracted samples would then be that lipid extraction is warranted, and in two of these species
333
(C. carcharias, L. ditropis), there was a significant increase in δ13C following lipid extraction.
334
Importantly, despite having C:N values ~3.3 following urea extraction, most of the species exhibited an
335
increase in δ13C following lipid extraction. These findings suggest that lipid extraction can significantly
336
affect δ13C even when C:N is < 3.5. Hence, lipid extraction may be required even in a tissue that is
337
relatively lean, a result that is concordant with the findings of Li et al. (2016). Our results indicate that
t
326
13
https://mc06.manuscriptcentral.com/cjfas-pubs
Canadian Journal of Fisheries and Aquatic Sciences
Page 14 of 36
338
failure to lipid extract elasmobranch tissues based on low C:N values of untreated tissue, where urea has
339
not been removed, will result in biased δ15N values, due to the inclusion of the isotopically light nitrogen
340
of urea, and potentially biased δ13C values as well, due to inclusion of lipid content that was masked by
341
low C:N values.
342
An additional important, and generally unrecognized implication of the effect of urea on C:N is
that it will bias C:N based arithmetic corrections that are used to adjust δ13C in lieu of lipid extracting
344
tissues. Any adjustment to δ13C values that is based on C:N values from nonFureaFextracted tissue will be
345
biased, although the magnitude of the effect will vary based on the urea and lipid content of the tissue.
346
Thus, failing to extract urea will not only bias δ15N, but by affecting C:N it will lead to incorrect estimates
347
of inferred lipid content upon which mathematical correction models rely. As our results indicate,
348
however, it is possible to develop models to adjust ureaFextracted tissues to account for lipid content
349
when lipid extraction is not feasible. For example, the model we developed for C. carcharias
350
demonstrates that even with a relatively small sample size (n = 19), we were able to generate a robust
351
speciesFspecific simple linear model based on ureaFextracted tissue C:N (model 4, r2 = 0.92, with no
352
systematic prediction biases based on visual inspection of residuals). Deriving taxaFspecific relationships
353
is always desirable, but for sharks (excluding dogfish) all four models examined seem to provide
354
potentially suitable lipid correction models. However, examination of model parameters reveals that for
355
model 1 the estimation of proteinFlipid δ13C discrimination is unrealistically low, ~3 times, or more, lower
356
than other estimates from this study and the generally reported range of 5–8‰ (Fry 2002, Post et al. 2007,
357
Logan et al. 2008, Reum 2011), demonstrating the need to consider other factors beyond AICc and fit (r2
358
and residual distribution) when determining the suitability of a correction model. Sharks (excluding
359
dogfish) have C:N of ureaFextracted tissues below ~4.5 and there appears to be a linear relationship
360
between C:N and U13C. This suggests that linear correction models might be most appropriate for sharks
361
with relatively low C:N values (~4.5), which is concordant with the findings of Post et al. (2007) for
362
aquatic organisms over a similar C:N range. This relationship is likely nonFlinear when tissues span a
363
wide range of C:N values, such as with S. suckleyi (Logan et al. 2008, Reum 2011). It is therefore
t
af
Dr
343
14
https://mc06.manuscriptcentral.com/cjfas-pubs
Page 15 of 36
Canadian Journal of Fisheries and Aquatic Sciences
364
important to consider the range in lipid content in species and tissues of interest when developing and
365
applying lipid correction models. Batoids had a less clear C:N relationship relative to other taxa in the
366
study. While exhibiting a significant linear relationship between C:Nu and U13C, the models explained a
367
relatively low proportion of the variability in U13C (models 3 and 4, r2 = 0.56 and 0.58) and may not
368
provide the same relative correction across all batoids (Figure 3). This emphasizes that the interaction
369
between urea and lipid content may change across disparate elasmobranch taxonomic groups.
370
Although our results indicate that urea directly influences δ15N values and potentially indirectly
influences the δ13C of elasmobranch tissues by affecting the C:N and inferred lipid content, the effects are
372
variable and speciesFdependent (Figures 1, 2). For species that have low lipid content in their muscle,
373
such as batoids (e.g. ~0.2% in U. halleri, (Lyons unpublished data), ~1F2% in B. binoculata and R. rhina
374
(Farrugia et al. 2015)), the effect on δ13C will be minimal, but δ15N may change substantially. In species
375
with higher lipid content, such as L. ditropis, which can have lipid content as high as 6.5 to 14.6% (mean
376
9%) in their muscle (data source: https://dec.alaska.gov/eh/vet/fish.htm), the effect on δ13C and δ15N will
377
be significant. The relatively low lipid content of batoid muscle compared to shark muscle may underlie
378
the observed general decrease in δ13C of batoids following urea extraction while urea and lipid extraction
379
showed less overall change in δ13C relative to the control samples. As described by Kim and Koch (2011),
380
the carbon in urea is enriched in 13C relative to the diet in humans (enriched 3 – 5‰) and cattle (0 –
381
3.5‰), suggesting that its removal may reduce the δ13C (Ivlev et al. 1996, Knobbe et al. 2006). In batoids
382
with low lipid content, removal of the 13C enriched urea would reduce δ13C, while lipid extraction, by
383
removing a small amount of 13C depleted lipids, would offset the removal of urea and result in little net
384
change in δ13C. This effect would vary based on both the concentration of urea as well as lipid content of
385
the tissue and again highlights the importance of understanding how these relationships change across
386
taxa. It is possible that other differences in the composition of batoid tissue may play a role as well, such
387
as differences in urea concentration or the presence of ceratotrichia, but this remains unclear.
t
389
af
388
Dr
371
Although lipid extraction by itself may remove lipids and much of the urea present in tissues
(Hussey et al. 2012a, Churchill et al. 2015), we reiterate the recommendations of Kim and Koch (2011)
15
https://mc06.manuscriptcentral.com/cjfas-pubs
Canadian Journal of Fisheries and Aquatic Sciences
Page 16 of 36
and Li et al. (2016) to make lipid and urea extraction the standard practice when analyzing elasmobranch
391
tissues for SIA. Since extracting urea is simple and inexpensive, there is no practical reason not to remove
392
it. In addition, our results indicate that δ13C may change significantly following lipid extraction even in
393
apparently lean tissues with relatively low C:N, suggesting that lipid extraction may be warranted in all
394
situations as suggested by Li et al. (2016). However, in some taxa that are very lean, such as the batoids in
395
this study, lipid extraction may not be required. However, urea would still need to be extracted to evaluate
396
the need for lipid extraction or correction. In instances where it is not feasible or desirable to lipid extract
397
every sample, we demonstrate that it is possible to develop speciesF or groupFspecific correction curves to
398
adjust for lipid content in ureaFextracted tissues. Though lipid extraction did not affect δ15N in our study,
399
a potential benefit of using mathematical correction models is the ability to account for the effect of lipid
400
content on δ13C while avoiding potential effects of chemical extraction on δ15N, which have been reported
401
previously in other taxa (Post et al. 2007). However, our results show that the confounding effects of urea
402
and lipids make it impossible to use C:N of nonFureaFextracted samples as a diagnostic tool to determine
403
the proper method of tissue treatment, something that occurs regularly in the literature.
af
The changes in the stable isotope composition of elasmobranch tissue resulting from urea and
t
404
Dr
390
405
lipid extraction will be mediated by the relative concentration of those compounds (Figure 4), which vary
406
across taxa, and their differential effects on δ13C and δ15N (Figures 1 & 2). We conclude that at a
407
minimum, urea should be removed to evaluate lipid content and a species or group specific lipid
408
correction relationship created to account for lipid content. The most robust approach to most confidently
409
eliminate bias and to facilitate comparisons across studies will be to apply urea and lipid removal
410
techniques in SIAFbased ecological studies of elasmobranchs.
411
412
We would like to thank Owyn Snodgrass, James Wraith, and Heidi Dewar (NOAA) for assistance
413
in obtaining samples from the NOAA Juvenile Shark Longline Survey, and John O’Sullivan, Christopher
414
Lowe, Salvador Jorgensen and the Monterey Bay Aquarium for help collecting juvenile white shark
415
samples. Funding was provided by PADI Foundation and the Dr. Earl Myers and Ethel Myers
16
https://mc06.manuscriptcentral.com/cjfas-pubs
Page 17 of 36
Canadian Journal of Fisheries and Aquatic Sciences
416
Oceanographic and Marine Biology Trust. For funding for materials and processing dogfish samples at
417
ISU, we would like to thank Bruce Finney. Funding and materials for skate processing were provided by
418
the Pacific Shark Research Center at Moss Landing Marine Laboratories and the North Pacific Research
419
Board. Support for M. Ibarra was provided by Mount Holyoke College.
420
Ballantyne, J.S. 1997. Jaws: the inside story. The metabolism of elasmobranch fishes. Comp Biochem
Physiol B Biochem Mol Biol !(4): 703F742.
Balter, V., Simon, L., Fouillet, H., and Lécuyer, C. 2006. BoxFmodeling of 15N/14N in mammals.
Oecologia "#(2): 212F222.
Bizzarro, J.J., Broms, K.M., Logsdon, M.G., Ebert, D.A., Yoklavich, M.M., Kuhnz, L.A., and Summers,
A.P. 2014. Spatial segregation in eastern north Pacific skate assemblages.
Bligh, E.G., and Dyer, W.J. 1959. A rapid method of total lipid extraction and purification. Canadian
Journal of Biochemistry and Physiology $#: 911F917.
Burnham, K.P., and Anderson, D.R. 2002. Model selection and multimodel inference: a practical
informationFtheoretic approach. Springer Science & Business Media.
Carlisle, A.B., Kim, S.L., Semmens, B.X., Madigan, D.J., Jorgensen, S.J., Perle, C.R., Anderson, S.D.,
Chapple, T.K., Kanive, P.E., and Block, B.A. 2012. Using stable isotope analysis to understand
the migration and trophic ecology of northeastern Pacific white sharks (Carcharodon carcharias).
PLoS ONE #(2): e30492.
Carlisle, A.B., Litvin, S.Y., Hazen, E.L., Madigan, D.J., Goldman, K.J., Lea, R.N., and Block, B.A. 2015.
Reconstructing habitat use by juvenile salmon sharks links upwelling to strandings in the
California Current. Mar Ecol Prog Ser %&%: 217F228.
Churchill, D.A., Heithaus, M.R., and Grubbs, R.D. 2015. Effects of lipid and urea extraction on δ15N
values of deepFsea sharks and hagfish: Can mathematical correction factors be generated? Deep
Sea Research Part II: Topical Studies in Oceanography %: 103F108.
Dale, J.J., Wallsgrove, N.J., Popp, B.N., and Holland, K.N. 2011. Nursery habitat use and foraging
ecology of the brown stingray Dasyatis lata determined from stomach contents, bulk and amino
acid stable isotopes. Mar Ecol Prog Ser "$$: 221F236.
Dulvy, N.K., Fowler, S.L., Musick, J.A., Cavanagh, R.D., Kyne, P.M., Harrison, L.R., Carlson, J.K.,
Davidson, L.N.K., Fordham, S.V., Francis, M.P., Pollock, C.M., Simpfendorfer, C.A., Burgess,
G.H., Carpenter, K.E., Compagno, L.J.V., Ebert, D.A., Gibson, C., Heupel, M.R., Livingstone,
S.R., Sanciangco, J.C., Stevens, J.D., Valenti, S., and White, W.T. 2014. Extinction risk and
conservation of the world’s sharks and rays. eLIFE $: e00590.
Estrada, J.A., Rice, A.N., Lutcavage, M.E., and Skomall, G.B. 2003. Predicting trophic position in sharks
of the northFwest Atlantic Ocean using stable isotope analysis. J Mar Biol Assoc U K !$(6):
1347F1350.
Farrugia, T.J., Oliveira, A.C., Knue, J.F., and Seitz, A.C. 2015. Nutritional content, mercury, and trace
element analyses of two skate (Rajidae) species in the Gulf of Alaska. Journal of Food
Composition and Analysis "&: 152F163.
Fisk, A.T., Tittlemier, S.A., Pranschke, J.L., and Norstrom, R.J. 2002. Using anthropogenic contaminants
and stable isotopes to assess the feeding ecology of Greenland sharks. Ecology !$(8): 2162F2172.
Fry, B. 2002. Stable isotopic indicators of habitat use by Mississippi River fish. J North Am Benthol Soc
& (4): 676F685.
Gannes, L.Z., Martinez Del Rio, C., and Koch, P. 1998. Natural abundance variations in stable isotopes
and their potential uses in animal physiological ecology. Comparative Biochemistry and
Physiology ' : 725F737.
t
af
Dr
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
17
https://mc06.manuscriptcentral.com/cjfas-pubs
Canadian Journal of Fisheries and Aquatic Sciences
Hazon, N., Wells, A., Anderson, G., Good, J., Pillans, R., and Franklin, C. 2008. Osmoregulation in
elasmobranch fishFA review. Comparative Biochemistry and Physiology AFMolecular, pp. S87F
S87.
Hazon, N., Wells, A., Pillans, R.D., Good, J.P., Anderson, W.G., and Franklin, C.E. 2003. Urea based
osmoregulation and endocrine control in elasmobranch fish with special reference to euryhalinity.
Comparative Biochemistry and Physiology BFBiochemistry & Molecular Biology $((4): 685F
700.
Hussey, H.E., Olin, J.A., Kinney, M.J., McMeans, B.C., and Fisk, A.T. 2012a. Lipid extraction effects on
stable isotope values (δ13C and δ15N) of elasmobranch muscle tissue. J Exp Mar Biol Ecol "$")
"$%: 7F15.
Hussey, N.E., Brush, J., McCarthy, I.D., and Fisk, A.T. 2010. δ15N and δ13C dietFtissue discrimination
factors for large sharks under semiFcontrolled conditions. Comp. Biochem. Physiol. AFMol.
Integr. Physiol. %%(4): 445F453.
Hussey, N.E., MacNeil, M.A., Olin, J.A., McMeans, B.C., Kinney, M.J., Chapman, D.D., and Fisk, A.T.
2012b. Stable isotopes and elasmobranchs: tissue types, methods, applications and assumptions. J
Fish Biol: 1449F1484.
Ivlev, A., Knyazev, Y.A., and Logachev, M. 1996. Daily average carbon isotope composition of CO2 of
expired air and urine urea in norm and some endocrine pathologies in man. Biofizika " (2): 508F
516.
Kim, S.L., Casper, D.R., GalvánFMagaña, F., OchoaFDíaz, R., HernándezFAguilar, S.B., and Koch, P.L.
2012. Carbon and nitrogen discrimination factors for elasmobranch soft tissues based on a longF
term controlled feeding study. Environ Biol Fishes '%(1): 37F52.
Kim, S.L., and Koch, P.L. 2011. Methods to collect, preserve, and prepare elasmobranch tissues for stable
isotope analysis. Environ Biol Fishes '%(1): 53F63.
Knobbe, N., Vogl, J., Pritzkow, W., Panne, U., Fry, H., Lochotzke, H., and PreissFWeigert, A. 2006. C
and N stable isotope variation in urine and milk of cattle depending on the diet. Analytical and
bioanalytical chemistry $!((1): 104F108.
Li, Y., Zhang, Y., Hussey, N.E., and Dai, X. 2016. Urea and lipid extraction treatment effects on δ15N
and δ13C values in pelagic sharks. Rapid Communications in Mass Spectrometry $*(1): 1F8.
Logan, J.M., Jardine, T.D., Miller, T.J., Bunn, S.E., Cunjak, R.A., and Lutcavage, M.E. 2008. Lipid
corrections in carbon and nitrogen stable isotope analyses: comparison of chemical extraction and
modelling methods. J Anim Ecol ##(4): 838F846.
Logan, J.M., and Lutcavage, M. 2010. Stable isotope dynamics in elasmobranch fishes. Hydrobiologia
("": 231F244.
Lyons, K., Lavado, R., Schlenk, D., and Lowe, C. 2014. Bioaccumulation of organochlorine contaminants
and EROD activity in southern California round stingrays (Urobatis halleri) exposed to planar
aromatic compounds. Environ Toxicol Chem $$(6): 1380F1390.
Martinez del Rio, C., Wolf, N., Carleton, S.A., and Gannes, L.Z. 2009. Isotopic ecology ten years after a
call for more laboratory experiments. Biol Rev !": 91F111.
Matich, P., Heithaus, M.R., and Layman, C.A. 2010. SizeFbased variation in intertissue comparisons of
stable carbon and nitrogen isotopic signatures of bull sharks (Carcharhinus leucas) and tiger
sharks (Galeocerdo cuvier). Can J Fish Aquat Sci (#: 877F885.
McConnaughey, T., and McRoy, C. 1979. FoodFweb structure and the fractionation of carbon isotopes in
the Bering Sea. Mar Biol %$(3): 257F262.
Michener, R.H., and Kaufman, L. 2007. Stable isotope ratios as tracers in marine food webs: an update. In
Stable isotopes in ecology and environmental science. Edited by R. Michener and K. Lajtha.
Blackwell, Boston. pp. 238 F 282.
Minagawa, M., and Wada, E. 1984. Stepwise enrichment of 15 N along food chains: further evidence and
the relation between δ 15 N and animal age. Geochim Cosmochim Acta "!(5): 1135F1140.
Mull, C.G., Blasius, M.E., O’Sullivan, J.B., and Lowe, C.G. 2012. Heavy metals, trace elements, and
organochlorine contaminants in muscle and liver tissue of juvenile White Sharks, Carcharodon
t
af
Dr
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
Page 18 of 36
18
https://mc06.manuscriptcentral.com/cjfas-pubs
Page 19 of 36
carcharias, from the Southern California Bight. Global perspectives on the biology and life
history of the White Shark. CRC Press, Boca Raton, Florida: 59F75.
Peterson, B.J., and Fry, B. 1987. Stable isotopes in ecosystem studies. Annu Rev Ecol Syst !: 293F320.
Pillans, R.D., Good, J.P., Anderson, W.G., Hazon, N., and Franklin, C.E. 2005. Freshwater to seawater
acclimation of juvenile bull sharks (Carcharhinus leucas): plasma osmolytes and Na+/K+F
ATPase activity in gill, rectal gland, kidney and intestine. Journal of Comparative Physiology B
#%(1): 37F44.
Pinnegar, P.K., and Polunin, N.V.C. 1999. Differential fractionation of δ13C and δ15N among fish tissues:
implications for the study of trophic interactions. Funct Ecol $: 225F231.
Post, D.M., Layman, C.A., Arrington, D.A., Takimoto, G., Quattrochi, J., and Montana, C.G. 2007.
Getting to the fat of the matter: models, methods and assumptions for dealing with lipids in stable
isotope analyses. Oecologia %&(1): 179F189.
Reum, J.C.P. 2011. Lipid correction model of carbon stable isotopes for a cosmopolitan predator, spiny
dogfish Squalus acanthias. J Fish Biol #'(7): 2060F2066.
Rice, W.R. 1989. Analyzing tables of statistical tests. Evolution "$: 223F225.
Vaudo, J.J., and Heithaus, M.R. 2011. Dietary niche overlap in a nearshore elasmobranch mesopredator
community. Mar Ecol Prog Ser "&%: 247F260.
Wood, C.M., Walsh, P.J., Kajimura, M., McClelland, G.B., and Chew, S.F. 2010. The influence of
feeding and fasting on plasma metabolites in the dogfish shark (Squalus acanthias). Comparative
Biochemistry and Physiology Part A: Molecular & Integrative Physiology %%(4): 435F444.
t
af
Dr
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
Canadian Journal of Fisheries and Aquatic Sciences
19
https://mc06.manuscriptcentral.com/cjfas-pubs
Canadian Journal of Fisheries and Aquatic Sciences
560
561
+
,-
,
Species
Carcharodon
carcharias
ID
CC
Isurus
oxyrinchus
IO
Lamna
ditropis
LD
Prionace
glauca
PG
Squalus
suckleyi
SS
Triakis
semifasciata
TS
. +/
Treatment
C
U
UL
C
U
UL
C
U
UL
C
U
UL
C
U
UL
C
U
UL
δ13C
mean
F16.8
F16.6
F15.6
F18.0
F17.8
F17.6
F19.4
F19.0
F17.9
F18.0
F18.1
F17.8
F19.4
F19.7
F17.4
F16.0
F16.1
F15.7
sd
0.9
0.8
0.5
0.2
0.7
0.2
0.4
0.4
0.6
0.2
0.3
0.2
1.1
1.0
0.2
0.3
0.5
0.4
Treatment
C
U
UL
C
U
UL
C
U
UL
C
U
UL
C
U
UL
C
U
UL
C
U
UL
C
U
UL
δ13C
mean
F16.7
F16.9
F16.7
F16.3
F16.6
F16.5
F17.0
F17.2
F16.1
F16.4
F16.5
F16.2
F18.6
F18.8
F18.5
F15.2
F16.2
F16.1
F16.2
F16.8
F16.7
F15.0
F15.4
F14.9
sd
0.5
0.4
0.4
0.4
0.3
0.4
1.1
1.5
0.9
1.2
1.2
1.2
0.5
0.5
0.4
0.5
0.4
0.4
1.0
0.6
0.6
0.3
0.3
0.3
C
test
F
ns
*
F
ns
ns
F
ns
*
F
ns
*
F
ns
*
U
test
F
F
*
F
F
ns
F
F
*
F
F
ns
F
F
*
δ15N
mean
16.8
17.3
17.3
16.4
17.4
17.4
14.4
15.4
15.5
16.1
16.9
17.0
14.0
14.5
14.9
15.3
16.0
16.0
sd
0.6
0.7
0.7
0.2
0.3
0.3
0.3
0.4
0.4
0.6
0.5
0.6
0.7
0.6
0.6
0.2
0.2
0.2
δ15N
mean
15.0
16.1
15.9
15.7
16.2
16.1
16.6
17.0
17.2
16.2
16.9
17.0
13.5
13.9
14.1
15.0
15.5
15.5
15.7
16.5
16.4
15.7
16.5
16.5
sd
0.5
0.5
0.5
0.4
0.5
0.2
0.9
0.9
0.7
0.4
0.1
0.0
0.8
1.2
0.9
0.1
0.3
0.3
0.7
0.8
0.7
0.3
0.3
0.4
C
test
F
*
*
F
*
*
F
*
*
F
*
*
F
ns
*
U
test
F
F
ns
F
F
ns
F
F
ns
F
F
ns
F
F
ns
C:N
mean
2.8
3.6
3.2
2.8
3.3
3.2
2.9
3.7
3.2
2.6
3.2
3.1
4.5
5.6
3.9
2.9
3.3
3.1
sd
0.2
0.3
0.0
0.1
0.1
0.1
0.2
0.2
0.1
0.1
0.0
0.0
1.1
1.2
0.1
0.3
0.0
0.0
C:N
mean
2.7
3.4
3.4
2.9
3.4
3.3
3.1
3.7
3.2
3.0
3.3
3.1
2.6
3.4
3.2
2.9
3.4
3.4
2.9
3.4
3.4
2.7
3.8
3.4
sd
0.0
0.0
0.0
0.0
0.0
0.0
0.3
0.4
0.0
0.4
0.0
0.0
0.1
0.1
0.1
0.1
0.0
0.0
0.0
0.0
0.1
0.1
0.1
0.1
C
test
F
*
*
F
*
*
F
*
*
F
*
*
F
ns
ns
U
test
F
F
*
F
F
*
F
F
*
F
F
*
F
F
*
n
19
TL (cm)
mean (SD)
161 (21)
8
147 (22)
10
106 (9)
10
116 (26)
9
73 (15)
2
U
n
7
TL (cm)
mean (SD)
128 (20)
3
72 (14)
3
U
2
U
7
55 (9)
5
132 (24)
6
104 (11)
11
15 (4)
,
Bathyraja
interrupta
BI
Gymnura
marmorata
GM
Myliobatis
californica
MC
Pteroplatytrygon
violacea
PV
Beringraja
binoculata
BB
Raja
rhina
RR
Urobatis
halleri
UH
C
test
F
ns
ns
U
test
F
F
ns
C
test
F
*
*
U
test
F
F
ns
t
ID
BA
af
Species
Bathyraja
aleutica
Dr
562
563
564
565
566
567
568
569
570
571
572
Page 20 of 36
F
ns
ns
F
*
*
F
ns
ns
F
*
ns
F
F
ns
F
F
ns
F
F
ns
F
F
*
F
ns
ns
F
*
*
F
ns
ns
F
*
*
F
F
ns
F
F
ns
F
F
ns
F
F
ns
C
test
F
*
*
F
*
*
F
*
*
F
*
*
F
*
*
U
test
F
F
ns
F
F
*
F
F
ns
F
F
ns
F
F
*
Table 1: Effects of urea and lipid extraction on stable isotope composition of various sharks (top panel)
and batoids (bottom panel). Treatments are C (control), U (ureaFextracted), UL (urea & lipidFextracted).
C test shows results of statistical comparisons between the control (C) and U and UL treatments, whereas
U test shows comparison between U and UL (ns = not significant, * = significant, p <0.05). Note that for
taxa with low sample sizes (<=3) we did not test for statistical differences.
20
https://mc06.manuscriptcentral.com/cjfas-pubs
Page 21 of 36
0
1
Figure 1: Differences in δ13C, δ15N, and C:N between ureaFextracted and control samples (A) and ureaF
and lipidFextracted and control samples (B) in sharks. Statistically significant differences between control
and U and UL treatments are indicated (* p <= 0.05). Species are Prionace glauca (PG), Squalus suckleyi
(SS), Triakis semifasciata (TS), Isurus oxyrinchus (IO), Lamna ditropis (LD), and Carcharodon
carcharias (CC). We did not test for differences in TS due to low sample size.
Figure 2: Differences in δ13C, δ15N, and C:N between ureaFextracted and control samples (A) and ureaF
and lipidF extracted and control samples (B) in batoids. Statistically significant differences between
control and U and UL treatments are indicated (* p <= 0.05). Species are Bathyraja aleutica (BA),
Bathyraja interrupta (BI), Beringraja binoculata (BB), Raja rhina (RR), Myliobatis californica (IO),
Gymnura marmorata (TS), Pteroplatytrygon violacea (PV), and Urobatis halleri (UH). We did not test
for differences in BI, MC and GM due to low sample sizes.
Figure 3: Relationships between the C:N of ureaFextracted tissue (treatment U, C:Nu) and changes in δ13C
between the U and UL treatments (Uδ13C = δ13CU F δ13CUL). Lines show selected best fit modeled
relationship between C:Nu and Uδ13C using best fit models based on AICc, which for groups with
relatively low lipid content and C:Nu (batoids and all sharks except for dogfish) was a linear model, but
for groups with a higher lipid content and C:Nu was based on the two parameter model from Fry (2002).
Figure 4: Conceptual diagram showing relative effects of urea and lipid extraction on elasmobranch
muscle tissue. Axes show relative change in δ13C and δ15N following different treatments, and color bar
shows C:N. The secondary axes show the relative effect of urea and lipid removal on δ13C and δ15N, with
the size of arrows indicating relative magnitude of effect. Urea extraction will generally increase δ15N and
C:N, and potentially also affect δ13C as 13C enriched urea is removed. Lipid extraction does not influence
δ15N, but increases δ13C and reduces C:N. The degree to which the different treatments affect δ13C, δ15N
(depicted by the magnitude and direction of the “Urea extraction” and “Urea & lipid extraction” arrows)
and C:N (depicted by the shading gradient within each treatment arrow) will vary based on the urea and
lipid (dashed arrow) content of the tissue.
t
af
Dr
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
Canadian Journal of Fisheries and Aquatic Sciences
Figure S1: Changes in δ13C and δ15N of ureaFextracted (grey) and ureaF & lipidFextracted samples (black)
relative to control samples in sharks (A) and batoids (B). Dotted lines show relative shift in values
between ureaFextracted and urea & lipidFextracted samples. Shark species are Prionace glauca (PG),
Squalus suckleyi (SS), Triakis semifasciata (TS), Isurus oxyrinchus (IO), Lamna ditropis (LD), and
Carcharodon carcharias (CC). Batoid species are Bathyraja aleutica (BA), Bathyraja interrupta (BI),
Beringraja binoculata (BB), Raja rhina (RR), Myliobatis californica (IO), Gymnura marmorata (TS),
Pteroplatytrygon violacea (PV), and Urobatis halleri (UH).
Figure S2: Relationship between the C:N of urea extracted tissue (U, C:Nu) and changes in δ13C between
the U and UL treatments (Uδ13C = δ13CU F δ13CUL) for individual species. Inset: magnified view of data
with low C:N (< 4.4).
21
https://mc06.manuscriptcentral.com/cjfas-pubs
Canadian Journal of Fisheries and Aquatic Sciences
t
af
Dr
Figure 1: Differences in δ13C, δ15N, and C:N between urea extracted and control samples (A) and urea and
lipid extracted and control samples (B) in sharks. Statistically significant differences between control and U
and UL treatments are indicated (* p <= 0.05). Species are
(PG),
(SS),
(TS),
(IO),
(LD), and
(CC).We
did not test for differences in TS due to low sample size.
205x256mm (300 x 300 DPI)
https://mc06.manuscriptcentral.com/cjfas-pubs
Page 22 of 36
Page 23 of 36
Canadian Journal of Fisheries and Aquatic Sciences
t
af
Dr
Figure 2: Differences in δ13C, δ15N, and C:N between urea extracted and control samples (A) and urea and
lipid extracted and control samples (B) in batoids. Statistically significant differences between control and
U and UL treatments are indicated (* p <= 0.05). Species are
(BA),
(BI),
(BB),
(RR),
(IO),
(TS),
(PV), and
(UH). We did not test for differences in BI, MC and GM
due to low sample sizes.
207x261mm (300 x 300 DPI)
https://mc06.manuscriptcentral.com/cjfas-pubs
Canadian Journal of Fisheries and Aquatic Sciences
af
Dr
t
Figure 3: Relationships between the C:N of urea extracted tissue (treatment U, C:Nu) and changes in δ13C
between the U and UL treatments (∆δ13C = δ13CU δ13CUL). Lines show selected best fit modeled
relationship between C:Nu and ∆δ13C using best fit models based on AICc, which for groups with relatively
low lipid content and C:Nu (batoids and all sharks except for dogfish) was a linear model, but for groups with
a higher lipid content and C:Nu was based on the two parameter model from Fry (2002).
122x98mm (300 x 300 DPI)
https://mc06.manuscriptcentral.com/cjfas-pubs
Page 24 of 36
Page 25 of 36
Canadian Journal of Fisheries and Aquatic Sciences
Dr
t
af
Figure 4: Conceptual diagram showing relative effects of urea and lipid extraction on elasmobranch muscle
tissue. Axes show relative change in δ13C and δ15N following different treatments, and color bar shows C:N.
The secondary axes show the relative effect of urea and lipid removal on δ13C and δ15N, with the size of
arrows indicating relative magnitude of effect. Urea extraction will generally increase δ15N and C:N, and
potentially also affect δ13C as 13C enriched urea is removed. Lipid extraction does not influence δ15N, but
increases δ13C and reduces C:N. The degree to which the different treatments affect δ13C, δ15N (depicted by
the magnitude and direction of the “Urea extraction” and “Urea & lipid extraction” arrows) and C:N (depicted
by the shading gradient within each treatment arrow) will vary based on the urea and lipid (dashed arrow)
content of the tissue.
246x171mm (300 x 300 DPI)
https://mc06.manuscriptcentral.com/cjfas-pubs
Canadian Journal of Fisheries and Aquatic Sciences
t
af
Dr
Figure S1: Changes in δ13C and δ15N of urea extracted (grey) and urea & lipid extracted samples (black)
relative to control samples in sharks (A) and batoids (B). Dotted lines show relative shift in values between
urea extracted and urea & lipid extracted samples. Shark species are
(PG),
(SS),
(TS),
s (IO),
(LD), and
(CC). Batoid species are
(BA),
(BI),
(BB),
(RR), !
(IO), "
(TS),
#
(PV), and
$
(UH).
209x273mm (300 x 300 DPI)
https://mc06.manuscriptcentral.com/cjfas-pubs
Page 26 of 36
Page 27 of 36
Canadian Journal of Fisheries and Aquatic Sciences
Dr
Figure S2: Relationship between the C:N of urea extracted tissue (U, C:Nu) and changes in δ13C between
the U and UL treatments (∆δ13C = δ13CU ' δ13CUL) for individual species. Inset: magnified view of data with
low C:N (< 4.4).
af
123x78mm (300 x 300 DPI)
t
https://mc06.manuscriptcentral.com/cjfas-pubs
Canadian Journal of Fisheries and Aquatic Sciences
Study #
C:N metric
Adjustment
Nothing
Lipid ext.
0
0
0
0
0
1
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
1
0
0
1
0
0
0
0
0
0
0
0
0
0
0
Urea ext.
0
0
1
0
1
0
0
1
0
1
0
0
1
1
1
0
1
1
1
1
0
1
0
0
0
0
0
0
0
0
0
1
0
0
0
0
1
0
0
0
0
0
1
1
1
1
0
t
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
1
1
1
0
0
1
0
0
0
0
1
0
0
0
0
0
0
0
1
af
0
1
0
1
0
0
1
0
0
0
1
1
0
0
0
0
0
0
0
0
0
0
0
0
1
1
1
1
0
0
0
0
0
0
0
0
0
1
0
0
1
1
0
0
0
0
1
Dr
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
1
0
1
0
0
0
0
0
1
0
0
0
0
0
0
0
1
0
0
1
0
0
0
0
1
0
0
0
0
0
0
0
https://mc06.manuscriptcentral.com/cjfas-pubs
Page 28 of 36
Page 29 of 36
Canadian Journal of Fisheries and Aquatic Sciences
48
49
50
Number
0
1
0
14
0
0
0
8
0
0
1
6
1
0
0
19
0
0
0
7
Table S2: Survey of 50 recent (2013-present) studies using stable isotope analysis to study the ecology of elasmobranchs. We c
“stable isotope”, “elasmobranch”, “shark”, “ray”; and selected the first 50 pertinent results. We assessed each study to determin
indicates the study used the C:N of bulk, non-extracted samples (i.e. contain urea and lipid) to assess the lipid content of their sa
whether lipid extraction was required or not. “Adjustment” indicates the study used the C:N of non-extracted samples to mathem
t
af
Dr
https://mc06.manuscriptcentral.com/cjfas-pubs
Canadian Journal of Fisheries and Aquatic Sciences
Page 30 of 36
t
af
Dr
Study
Couturier et al. 2013. Stable Isotope and Signature Fatty Acid Analyses Suggest Reef Manta Rays Feed on Demersal Zooplankt
Matich & Heithaus 2014. Multi ‐tissue stable isotope analysis and acoustic telemetry reveal seasonal variability in the trophic i
Hussey et al. 2014. Rescaling the trophic structure of marine food webs. Ecology Letters, 17(2), pp.239-250.
Daly et al. 2013. Comparative feeding ecology of bull sharks (Carcharhinus leucas) in the coastal waters of the Southwest India
McMeans et al. 2013. The role of Greenland sharks (Somniosus microcephalus) in an Arctic ecosystem: assessed via stable isot
Meneses et al. 2016. Trophic overlap between blue sharks (Prionace glauca) and shortfin makos (Isurus oxyrinchus): Trophic li
Heithaus et al. 2013. Apparent resource partitioning and trophic structure of large-bodied marine predators in a relatively pristin
Shiffman et al. 2014. Feeding ecology of the sandbar shark in south carolina estuaries revealed through δ13C and δ15N stable i
De Lecea et al. 2013. Processes controlling the benthic food web of a mesotrophic bight (KwaZulu-Natal, South Africa) reveale
Caut et al. 2013. Diet-and tissue-specific incorporation of isotopes in the shark Scyliorhinus stellaris, a North Sea mesopredator
Tilley et al. 2013. Diet reconstruction and resource partitioning of a Caribbean marine mesopredator using stable isotope Bayes
Navarroet al. 2014. Short-and long-term importance of small sharks in the diet of the rare deep-sea shark Dalatias licha. Marine
Polo-Silva et al. 2013. Trophic shift in the diet of the pelagic thresher shark based on stomach contents and stable isotope analy
Olin et al. 2013. Seasonal variability in stable isotopes of estuarine consumers under different freshwater flow regimes. Marine
Munroe et al. 2015. Regional movement patterns of a small ‐bodied shark revealed by stable ‐isotope analysis. Journal of fis
Li et al. 2014. Trophic ecology of sharks in the mid-east Pacific ocean inferred from stable isotopes. Journal of Ocean Universit
Churchill et al. 2015. Trophic interactions of common elasmobranchs in deep-sea communities of the Gulf of Mexico revealed
Kiszka et al. 2015. Plasticity of trophic interactions among sharks from the oceanic south-western Indian Ocean revealed by sta
Espinoza et al. 2015. Feeding ecology of common demersal elasmobranch species in the Pacific coast of Costa Rica inferred fro
McMean et al. 2015. Impacts of food web structure and feeding behavior on mercury exposure in Greenland Sharks (Somniosu
Malpica‐Cruz et al. 2013. Tissue ‐specific stable isotope ratios of shortfin mako (Isurus oxyrinchus) and white (Carcharodon ca
Torres et al. 2014. Trophic ecology and bioindicator potential of the North Atlantic tope shark. Science of The Total Environme
McCauley et al. 2014. Reliance of mobile species on sensitive habitats: a case study of manta rays (Manta alfredi) and lagoons.
Papastamatiou et al. 2015. Movements and foraging of predators associated with mesophotic coral reefs and their potential for l
de Moura et al. 2015. Assessment of trace elements, POPs, 210 Po and stable isotopes (15 N and 13 C) in a rare filter-feeding sh
Madigan et al. 2015. Diet shift and site-fidelity of oceanic whitetip sharks Carcharhinus longimanus along the Great Bahama B
Speed et al. 2012. Trophic ecology of reef sharks determined using stable isotopes and telemetry. Coral Reefs, 31(2), pp.357-36
Drymon et al. 2012. Trophic plasticity in the Atlantic sharpnose shark (Rhizoprionodon terraenovae) from the north central Gul
Fanelli et al. 2013. Trophic webs of deep-sea megafauna on mainland and insular slopes of the NW Mediterranean: a comparis
Courtney & Foy, R., 2012. Pacific sleeper shark Somniosus pacificus trophic ecology in the eastern North Pacific Ocean inferre
McCauley et al. 2012. Assessing the effects of large mobile predators on ecosystem connectivity. Ecological Applications, 22(6
Lyons et al. 2013. Effects of trophic ecology and habitat use on maternal transfer of contaminants in four species of young of th
Matich et al. 2015. Short-term shifts of stable isotope (δ 13 C, δ 15 N) values in juvenile sharks within nursery areas suggest rap
Albo-Puigserver et al. 2015. Feeding ecology and trophic position of three sympatric demersal chondrichthyans in the northwes
Rojas et al. 2014. Feeding grounds of juvenile scalloped hammerhead sharks (Sphyrna lewini) in the south-eastern Gulf of Calif
Reum & Marshall 2013. Evaluating δ15N–body size relationships across taxonomic levels using hierarchical models. Oecologia
Hussey et al. 2015. Expanded trophic complexity among large sharks. Food Webs, 4, pp.1-7.
Valls et al. 2014. Structure and dynamics of food webs in the water column on shelf and slope grounds of the western Mediterra
Teffer et al. 2014. Trophic influences on mercury accumulation in top pelagic predators from offshore New England waters of t
Kiszka et al. 2014. Trophic ecology of common elasmobranchs exploited by artisanal shark fisheries off south-western Madaga
Cresson et al. 2014. Mercury in organisms from the Northwestern Mediterranean slope: Importance of food sources. Science of
Barría et al. 2015. Unravelling the ecological role and trophic relationships of uncommon and threatened elasmobranchs in the
Lopez et al. 2013. Trophic ecology of the dusky catshark Bythaelurus canescens (Günther, 1878)(Chondrychthyes: Scyliorhinid
Freedman et al. 2015. Connectivity and movements of juvenile predatory fishes between discrete restored estuaries in southern
Shaw et al. 2016. Trophic Ecology of a Predatory Community in a Shallow-Water, High-Salinity Estuary Assessed by Stable Is
Jaime-Rivera et al. 2014. Feeding and migration habits of white shark Carcharodon carcharias (Lamniformes: Lamnidae) from
Torres-Rojas et al. 2015. Diet and trophic level of scalloped hammerhead shark (Sphyrna lewini) from the Gulf of California an
https://mc06.manuscriptcentral.com/cjfas-pubs
Page 31 of 36
Canadian Journal of Fisheries and Aquatic Sciences
Hernández-Aguilar et al. 2015. Trophic ecology of the blue shark (Prionace glauca) based on stable isotopes (
δ 13 C and δ
Raoult et al. 2015. Not all sawsharks are equal: species of co-existing sawsharks show plasticity in trophic consumption both w
Frisch et al. 2016. Reassessing the trophic role of reef sharks as apex predators on coral reefs. Coral Reefs, pp.1-14.
e analysis to study the ecology of elasmobranchs. We conducted the search on Google scholar using the search terms:
50 pertinent results. We assessed each study to determine how they treated samples for urea and lipid. “C:N metric”
ntain urea and lipid) to assess the lipid content of their samples and use that inferred lipid content to inform them
13
e study used the C:N of non-extracted samples to mathematically adjust δ
C to account for inferred lipid content.
t
af
Dr
https://mc06.manuscriptcentral.com/cjfas-pubs
Canadian Journal of Fisheries and Aquatic Sciences
Page 32 of 36
on Demersal Zooplankton. PLoS ONE 8(10): e77152
ariability in the trophic interactions of juvenile bull sharks in a coastal estuary. Journal of Animal Ecology, 83(1), pp.199-213.
t
af
Dr
of the Southwest Indian Ocean inferred from stable isotope analysis. PloS one, 8(10), p.e78229.
assessed via stable isotopes and fatty acids. Marine biology, 160(5), pp.1223-1238.
oxyrinchus): Trophic linkages between two shark species In the Eastern Pacific Ocean food web. Food Webs.
ors in a relatively pristine seagrass ecosystem. Mar Ecol Prog Ser, 481, pp.225-237.
δ13C and δ15N stable isotope analysis. Marine and Coastal Fisheries, 6(1), pp.156-169.
l, South Africa) revealed by stable isotope analysis. Marine Ecology Progress Series, 484, pp.97-114.
North Sea mesopredator. Marine Ecology. Progress Series, 492.
ng stable isotope Bayesian modelling. PloS one, 8(11), p.e79560.
k Dalatias licha. Marine biology, 161(7), pp.1697-1707.
and stable isotope analyses. Marine Biology Research, 9(10), pp.958-971.
r flow regimes. Marine Ecology Progress Series, 487, pp.55-69.
analysis. Journal of fish biology, 86(5), pp.1567-1586.
rnal of Ocean University of China, 13(2), pp.278-282.
ulf of Mexico revealed through stable isotope and stomach content analysis. Deep Sea Research Part II: Topical Studies in Oceanography, 1
n Ocean revealed by stable isotope and mercury analyses. Deep Sea Research Part I: Oceanographic Research Papers, 96, pp.49-58.
Costa Rica inferred from stable isotope and stomach content analyses. Journal of Experimental Marine Biology and Ecology, 470, pp.12-25
nland Sharks (Somniosus microcephalus). Science of The Total Environment, 509, pp.216-225.
d white (Carcharodon carcharias) sharks as indicators of size
‐based differences in foraging habitat and trophic level. Fisheries Oceanog
of The Total Environment, 481, pp.574-581.
ta alfredi) and lagoons. Marine biology, 161(9), pp.1987-1998.
and their potential for linking ecological habitats. Marine Ecology Progress Series.
n a rare filter-feeding shark: The megamouth. Marine pollution bulletin, 95(1), pp.402-406.
ng the Great Bahama Bank. Marine Ecology Progress Series, 529, p.185.
Reefs, 31(2), pp.357-367.
m the north central Gulf of Mexico. Environmental biology of fishes, 95(1), pp.21-35.
diterranean: a comparison by stable isotope analysis. Marine ecology. Progress series, 490, pp.199-221.
th Pacific Ocean inferred from nitrogen and carbon stable
‐isotope ratios and diet. Journal of fish biology, 80(5), pp.1508-1545.
gical Applications, 22(6), pp.1711-1717.
r species of young of the year lamniform sharks. Marine environmental research, 90, pp.27-38.
ursery areas suggest rapid shifts in energy pathways. Journal of Experimental Marine Biology and Ecology, 465, pp.83-91.
hthyans in the northwestern Mediterranean. Marine Ecology Progress Series, 524, p.255.
uth-eastern Gulf of California. Hydrobiologia, 726(1), pp.81-94.
hical models. Oecologia, 173(4), pp.1159-1168.
of the western Mediterranean. Journal of Marine Systems, 138, pp.171-181.
New England waters of the northwest Atlantic Ocean. Marine environmental research, 101, pp.124-134.
south-western Madagascar inferred from stable isotopes. Aquatic Biology, 23, pp.29-38.
ood sources. Science of The Total Environment, 497, pp.229-238.
d elasmobranchs in the western Mediterranean Sea. Marine Ecology Progress Series, 539, p.225.
drychthyes: Scyliorhinidae) in the southeast Pacific Ocean. Journal of Applied Ichthyology, 29(4), pp.751-756.
d estuaries in southern California. Marine Ecology Progress Series, 520, p.191.
y Assessed by Stable Isotope Analysis. Marine and Coastal Fisheries, 8(1), pp.46-61.
rmes: Lamnidae) from Isla Guadalupe inferred by analysis of stable isotopes δ15N and δ13C. Revista de Biolog
he Gulf of California and Gulf of Tehuantepec, Mexico. Iranian Journal of Fisheries Sciences, 14(3), pp.767-785.
https://mc06.manuscriptcentral.com/cjfas-pubs
ía Tropical, 62(2), p
Page 33 of 36
Canadian Journal of Fisheries and Aquatic Sciences
δ 13 C and δ 15 N) and stomach content. Journal of the Marine Biological Association of the United Kingdom, pp.1-8.
hic consumption both within and between species. Canadian Journal of Fisheries and Aquatic Sciences, 72(11), pp.1769-1775.
fs, pp.1-14.
t
af
Dr
https://mc06.manuscriptcentral.com/cjfas-pubs
Canadian Journal of Fisheries and Aquatic Sciences
pp.199-213.
Studies in Oceanography, 115, pp.92-102.
apers, 96, pp.49-58.
and Ecology, 470, pp.12-25.
t
af
Dr
hic level. Fisheries Oceanography, 22(6), pp.429-445.
, 80(5), pp.1508-1545.
pp.83-91.
ía Tropical, 62(2), pp.637-647.
https://mc06.manuscriptcentral.com/cjfas-pubs
Page 34 of 36
Page 35 of 36
Canadian Journal of Fisheries and Aquatic Sciences
Kingdom, pp.1-8.
p.1769-1775.
t
af
Dr
https://mc06.manuscriptcentral.com/cjfas-pubs
Canadian Journal of Fisheries and Aquatic Sciences
data
All species
All sharks
(no Dogfish)
All batoids
Dogfish
S. suckleyi
Great White
C. carcharias
Δ13C = (a C:NU + b )( C:NU + c )-1
AICc:
94.02
a=
b=
5.39 ± 0.85
-17.25 ± 2.62
c=
-0.68 ± 0.75
AICc:
a=
b=
38.58
1.94 ± 0.23
-6.27 ± 0.75
c=
-2.98 ± 0.06
a=
b=
nf
-
c=
-
AICc:
c=
-1.90 ± 1.03
a=
b=
-0.46
14.18 ± 16.56
-43.96 ± 50.48
c=
3.22 ± 9.12
Δ13C = β 0 + β 1 Ln(C:NU)
AICc:
102.33
β0 =
β1 =
6.12 ± 0.30
3.18 ± 0.03
AICc:
P=
F=
38.97
7.86 ± 0.70
3.15 ± 0.03
AICc:
β0 =
β1 =
a=
b=
0.88 ± 0.06
-2.57 ± 0.22
0.76
r2 :
0.70
39.54
-7.69 ± 0.82
6.74 ± 0.66
2
r:
AICc:
P=
F=
15.41
5.63 ± 0.77
3.28 ± 0.04
Dr
AICc:
af
β0 =
β1 =
17.30
6.31 ± 0.62
3.35 ± 0.20
0.69
13.87
-6.15 ± 0.91
5.19 ± 0.73
r2 :
t
AICc:
β0 =
β1 =
0.56
20.93
-4.08 ± 1.27
3.79 ± 0.74
r2 :
AICc:
P=
F=
-5.15
8.30 ± 0.52
3.14 ± 0.03
AICc:
β0 =
β1 =
Δ13C = b + a C:NU
AICc:
121.93
-4.89 ± 0.33
4.34 ± 0.25
r2 :
P=
F=
a=
b=
AICc:
P=
F=
AICc:
22.46
4.76 ± 1.03
-16.96 ± 3.44
AICc:
Δ13C = P - PF (C:NU)-1
AICc:
92.53
Page 36 of 36
0.76
-3.88
-7.80 ± 0.61
6.90 ± 0.48
2
r:
0.92
AICc:
a=
b=
40.50
1.81 ± 0.18
-5.61 ± 0.63
2
r:
0.68
a=
b=
12.26
1.45 ± 0.19
-4.74 ± 0.69
r2 :
0.58
AICc:
a=
b=
24.39
0.62 ± 0.16
-1.18 ± 0.91
r2 :
0.65
AICc:
AICc:
a=
b=
-3.27
1.80 ± 0.13
-5.49 ± 0.46
2
r:
2
Table S1. Parameter estimates (± SE), r (linear models only) and corrected Akaike Information Criteria valeus (AICc) for models of
Δ13C, the difference between lipid and urea (UL) and urea (U) extracted δ13C values (Δ13C = δ13CU - δ13CUL), fit to all species,
batoids, all sharks other than S. suckleyi (dogfish), S. suckleyi , and C. carcharias . nf = model failed to converg.
https://mc06.manuscriptcentral.com/cjfas-pubs
0.92