Content uploaded by Christopher G. Lowe
Author content
All content in this area was uploaded by Christopher G. Lowe
Content may be subject to copyright.
The effects of barotrauma on the catch-and-
release survival of southern California nearshore
and shelf rockfish (Scorpaenidae, Sebastes spp.)
Erica T. Jarvis and Christopher G. Lowe
Abstract: Two experiments were used to assess the effects of barotrauma on initial capture survival and short-term post-
recompression survival of line-caught (range 18–225 m) southern California rockfish (Sebastes spp.). Occurrence of exter-
nal and internal signs of barotrauma was characterized across all species. Despite species-specific differences in the extent
of barotrauma observed, initial capture survival of rockfish held in a live well for a 10-min period following capture was
68% overall (19 species, n= 168). Overall 2-day survival of rockfish following recompression in cages was also 68%
(17 species, n= 257). Short-term survival varied across species (range 36% to 82%), as did the occurrence of external
signs of barotrauma. The degree of external signs of barotrauma was not a significant predictor of initial capture survival
or short-term survival. The most significant predictor of short-term survival was surface holding time, with short-term sur-
vival increasing with decreasing surface holding time. These results suggest that rapid recompression of rockfish can sig-
nificantly decrease discard mortality and could potentially enhance rockfish conservation.
Re
´sume
´:Deux expe
´riences nous ont servi e
´valuer les effets du barotraumatisme sur la survie initiale a
`la capture et la
survie a
`court terme apre
`s la recompression de se
´bastes (Sebastes spp.) du sud de la Californie capture
´sa
`la ligne (e
´tendue
des profondeurs de 18–225 m). Nous avons observe
´des signes externes et internes de barotraumatisme chez toutes les es-
pe
`ces. Malgre
´des diffe
´rences spe
´cifiques de l’importance du barotraumatisme, la survie initiale a
`la capture des se
´bastes
garde
´s dans un vivier pendant 10 min suivant la capture est globalement de 68 % (19 espe
`ces, n= 168). La survie globale
des se
´bastes garde
´s dans des cages pendant 2 jours apre
`s la recompression est aussi de 68 % (17 espe
`ces, n= 257). La
survie a
`court terme varie d’une espe
`ce a
`l’autre (e
´tendue de 36 a
`82 %), de me
ˆme que la pre
´sence de signes externes de
barotraumatisme. L’importance des signes externes de barotraumatisme ne permet pas de pre
´dire avec assurance la survie
initiale a
`la capture, ni la survie a
`court terme. La variable la plus significative pour pre
´dire la survie a
`court terme est la
dure
´e de la retenue du poisson en surface, la survie augmentant en fonction inverse de la dure
´e de la retenue en surface.
Ces re
´sultats indiquent qu’une recompression rapide des se
´bastes peut re
´duire de fac¸on significative la mortalite
´lors de
leur rejet a
`la mer et pourrait potentiellement favoriser la conservation des se
´bastes.
[Traduit par la Re
´daction]
Introduction
Rockfish of the genus Sebastes are an extremely diverse
group, inhabiting a variety of depths and habitats worldwide
(Love et al. 2002). The majority of these species occur in
the North Pacific, and historically, many have comprised
major commercial and recreational fisheries. In southern
California, rockfish catches have been an important compo-
nent of the winter recreational fishery since the 1950s
(Dotson and Charter 1998); however, there is strong evidence
indicating that many rockfish populations have suffered
from excessive harvest impacts over the last two decades
(Love et al. 1998; Mason 1998; Butler et al. 2003). Growing
concern regarding bycatch (i.e., incidentals and discards)
mortality in the recreational rockfish fishery has prompted
research into rockfish physiology and the effects of angling-
induced barotrauma on their survival (Parker et al. 2006;
Hannah and Matteson 2007). Upon rapid decompression,
rockfish, which are physoclistic, suffer internal injury and
positive buoyancy as a result of overexpansion of gases
within the swim bladder and other highly vascularized tis-
sues. Consequently, many discarded rockfish are unable to
swim down deep enough to force recompression of gases
within their bodies. Discard mortality in the recreational
fishery is assumed to be high, as floating fish at the sur-
face eventually succumb to thermal shock and (or) bird or
marine mammal predation.
Recent research on Northeast Pacific coast rockfish sug-
gests that assisted release of discards may increase postre-
lease survival, as external signs of barotrauma were
alleviated following recompression of fish to depth (Hannah
and Matteson 2007). However, several studies on other
Received 9 July 2007. Accepted 27 November 2007. Published on the NRC Research Press Web site at cjfas.nrc.ca on 2 June 2008.
J20092
E.T. Jarvis1,2 and C.G. Lowe. California State University, Long Beach, Department of Biological Sciences, 1250 Bellflower Boulevard,
Long Beach, CA 90840-0004, USA.
1Corresponding author (e-mail: EJarvis@dfg.ca.gov).
2Present address: California Department of Fish and Game, 4665 Lampson Avenue, Suite C, Los Alamitos, CA 90720-5188, USA.
1286
Can. J. Fish. Aquat. Sci. 65: 1286–1296 (2008) doi:10.1139/F08-071
#
2008 NRC Canada
physoclistic species have found no correlation between de-
gree of barotrauma and survival (Gitschlag and Renaud
1994; Rummer and Bennett 2005). External signs of baro-
trauma documented for a few rockfish species include stom-
ach eversion, exophthalmia (bulging eyes), corneal gas
bubbles, and subcutaneous gas bubbles (Gotshall 1964; Lea
et al. 1999; Hannah and Matteson 2007). Internal signs of
barotrauma have not been characterized for rockfish, but in
other physoclists, these signs have included arterial embo-
lism, swim bladder rupture, hematoma, hemorrhage, and
organ torsion (Gitschlag and Renaud 1994; Rummer and
Bennett 2005). The effects of internal barotrauma on imme-
diate and delayed mortality in the field have not yet been
reported, and the condition and longer-term survival of line-
caught rockfish recompressed in the field is still largely
unknown.
Releasing rockfish at depth aims to decrease the probabil-
ity of discard mortality, but the practice can only be effec-
tive if evidence indicates that a high enough percentage of
fish returned to depth recover and survive. The primary
objectives of this study were (i) to characterize external and
internal signs of angling-induced barotrauma in rockfish and
(ii) to quantify initial survival (within 10 min of capture)
and short-term (2-day) survival following recompression in
cages. An understanding of postrecompression survival in
rockfish will enable fisheries managers to assess the utility
of catch-and-release regulations, not only in southern Cali-
fornia, but also along the Northeast Pacific coast, and offer
resource managers more accurate population parameters for
stock assessments.
Materials and methods
Characterization of barotrauma and initial capture
survival
Nearshore and shelf rockfish were captured by hook-and-
line from the Palos Verdes peninsula (depth 30–70 m),
Southeast Bank off Huntington Beach (50–90 m), at the off-
shore petroleum platform Gilda off the Ventura coast
(55 m), and off Santa Catalina Island (18–96 m) between
October 2004 and March 2006 (Fig. 1). Fishing was designed
to target demersal species; all fishing lines were fished on
the sea floor and arranged with one to two dropper loops
tied approximately 22 cm above an 8–12 oz torpedo weight.
We recorded bottom depth and the time of day that each
fish was landed. Each fish was measured (standard length)
and the presence of external signs of barotrauma including
stomach eversion (SE), exophthalmia (EX), corneal gas
bubbles (CB), subcutaneous gas bubbles (SB), and pro-
lapsed cloaca (PC) were recorded. Handling time to meas-
ure and examine fish was kept to less than 2 min. Fish
were placed in a live well or cooler with fresh seawater
for 10 min, after which each fish was assessed for signs
of gill ventilation as an indicator of initial capture survival.
After noting survival, fish were euthanized and dissected
within 24–48 h after capture. Internal signs of barotrauma
were recorded as presence of visible tears in the tunica in-
ternus of the swim bladder (ST), organ torsion (OT), hem-
orrhage (HE), and presence and location of arterial gas
embolisms in the pericardial chamber and at the swim
bladder gas gland and oval (AE). Subcutaneous gas bub-
bles and arterial embolism were not initially documented at
the onset of this experiment, thus sample sizes for these signs
vary.
Analysis
Percent occurrence and percent survival (±95% binary
confidence intervals) were reported for each species and for
all species combined (adjusted Wald method; Sauro and
Lewis 2005). We used a general linear model (GLM,
STATISTICA 7.1; StatSoft, Inc., Tulsa, Oklahoma) to
examine the relationship between depth of capture and the
total number of external and internal signs of barotrauma,
including species in the model as a possible effect. Baro-
trauma effects were assessed using a Fisher exact test,
comparing the proportion of survivors with and without a
given sign of barotrauma. Binary logistic regression
(MINITAB release 14.20; Minitab Inc., State College,
Pennsylvania) was used to model the extent of barotrauma
as a predictor of initial survival (barotrauma model I,
external signs; barotrauma model II, internal signs). The
extent of barotrauma or barotrauma category was defined
by specific combinations of different signs of barotrauma
observed in captured fish and was treated as a single,
categorical, independent variable. The models were limited
to those specific combinations of barotrauma that occurred
in at least 10 fish. Additionally, survival of fish showing
each specific combination of barotrauma was reported rela-
tive to fish showing no signs of barotrauma. In these models,
we included only individuals for which all signs were
accounted.
Characterization of barotrauma and 2-day
postrecompression survival
During the summers of 2004 and 2005, nearshore and
shelf rockfish were captured by hook-and-line aboard the
R/V Yellowfin near Southeast Bank (50–89 m; Fig. 1). For
each fish, we recorded bottom depth, time of capture,
standard length, and the presence of external signs of baro-
trauma. Each fish was externally tagged for individual
identification with a 4 cm long dart tag (Hallprint) and
placed in a live well prior to release. Fish were then trans-
ferred to polyvinyl chloride (PVC) coated wire mesh cages
(4 cm 4 cm mesh, 1.2 m 1.2 m 1.5 m). Up to six
cages were deployed during each trip, with as many as
12 fish per cage. Each cage was lowered to the original
fish capture depth and left for 2 days on soft substratum
adjacent to fished reefs. To account for differences in total
surface holding duration among individuals, we recorded
the exact time of each cage deployment. Two cages per
trip were deployed with StowAway TidbiT temperature
data loggers (Onset Computer Corp., Bourne, Mass.) to re-
cord seafloor water temperatures. Differences in water tem-
perature between the sea floor and sea surface (i.e.,
temperature differential) were recorded for each trip to ac-
count for temperature change experienced by rockfish upon
capture.
Two days following fish collection and recompression,
each cage was pulled up to a depth of 20 m where a team
of divers met the cage and assessed each fish for mortality
and external signs of barotrauma. As the greatest change in
pressure and, hence, gas expansion occurs within the first
Jarvis and Lowe 1287
#
2008 NRC Canada
10 m of the surface, this observation depth of 20 m was
chosen to greatly reduce the probability of barotrauma injury
resulting from the second decompression event. All dead
fish and a subset of live fish per cage were saved for exami-
nation of internal barotrauma injury as described above; all
remaining live fish were released.
Longer-term survival
In addition to a unique tag number, all external tags used
in the cage experiments contained a phone number for
reporting recaptured rockfish. To record evidence of longer-
term survival, all tag returns were logged by tag identifica-
tion number, species, date of recapture, location of recapture,
and fisher contact.
Analysis
Chi-square tests of independence were used to test for
species-specific differences in both the percent occurrence
of each sign of barotrauma and percent survival. Fisher
exact tests and logistic regression were again used to test
for barotrauma effects as described in the initial capture
study. An additional logistic regression model (PROC
LOGISTIC, SAS 9.1.3; SAS Institute Inc., Cary, North
Carolina) tested the effects of depth, species, fish length,
cage density, temperature differential, and surface holding
duration on survival (overall model). For this model, we
included only data for species with a sample size greater
than 15. A reduced model was then designed using only
significant variables from the overall model. Because
several variables, including surface holding duration and
temperature differential, could not be controlled in the
field, we did not include interactions. To account for
differences in species survival due to different temperature
differentials or surface holding times, an analysis of
variance (ANOVA) was used to compare temperature
differential experienced by each species upon capture, and
Kruskal–Wallis was used to test for differences in median
surface times experienced by each species prior to recom-
pression.
Results
Barotrauma and initial capture survival
One-hundred and sixty-eight rockfish representing 21 spe-
cies were captured over a 1.5-year period from depths of
18–96 m; vermilion rockfish, Sebastes miniatus; greenspot-
ted rockfish, Sebastes chloristictus; olive rockfish, (Sebastes
serranoides); halfbanded rockfish (Sebastes semicinctus),
rosy rockfish (Sebastes rosaceous), and honeycomb rockfish
(Sebastes umbrosus) comprised the majority of the catch
(Table 1). Initial capture survival of rockfish was 68% over-
all (95% confidence interval, CI: 60% to 75%).
The most common external signs of barotrauma observed
in the overall catch (n= 168) were subcutaneous gas bubbles
(SB, 76% of n= 157), stomach eversion (SE, 63%), and
exophthalmia (EX, 52%); the most common signs of inter-
nal barotrauma were arterial embolism (AE, 68% of n=
144), hemorrhage (HE, 64%), and organ torsion (OT, 33%).
The percent occurrence of each barotrauma type and the
number of frequently occurring signs of barotrauma were
species-specific (Table 2). Stomach eversion occurred fre-
quently among brown rockfish (Sebastes auriculatus), flag
rockfish (Sebastes rubrivinctus), greenspotted, halfbanded,
honeycomb, and rosy rockfish, starry rockfish (Sebastes
constellatus), treefish (Sebastes serriceps), and widow
rockfish (Sebastes ensifer), and swim bladder tears (ST)
occurred frequently (82%) only in olive rockfish. For the
catch as a whole, there was a significant, yet weak, posi-
tive correlation between the number of external and inter-
nal signs of barotrauma observed in each fish (r= 0.28,
Fig. 1. Fishing locations of nearshore and shelf rockfish captured by hook-and-line in southern California from depths of 18–220 m
(October 2004 to August 2006): (1) White’s Point outfall pipe, (2) Rockpile, (3) Ship Rock, (4) platform Gilda, and (5) Southeast Bank.
Shaded lines represent 60 m and 90 m depth contours. Inset depicts southern California in relation to the Pacific coast of the United States.
1288 Can. J. Fish. Aquat. Sci. Vol. 65, 2008
#
2008 NRC Canada
p< 0.01). The degree of barotrauma was greatest for the
two species caught deepest, rosy and greenspotted rockfish.
However, regardless of species, depth had a significant
positive relationship with the total number of traumas ob-
served in rockfish individuals (GLM, n= 168, R2= 0.35,
F= 3.61, p< 0.001).
Initial capture survival varied by species (Fig. 2). Eight of
12 species (with n> 3) had greater than 75% initial capture
survival, whereas olive and rosy rockfish had low survival.
Although fish caught at deeper depths generally showed
higher numbers of trauma, species caught at shallow depths
showed relatively similar survival proportions as species
caught in deeper depths.
Overall percent survival was significantly higher for fish
Table 1. Sample size (n), catch (%), and depth (m) of capture of nearshore and shelf
rockfish captured by hook-and-line in southern California (initial capture survival
experiment, October 2004 to March 2006).
Common name Scientific name nCatch (%) Depth (m)
Vermilion rockfish Sebastes miniatus 35 21 35–96
Greenspotted rockfish Sebastes chlorostictus 19 11 75–189
Olive rockfish Sebastes serranoides 16 10 23–53
Halfbanded rockfish Sebastes semicinctus 15 9 53–64
Rosy rockfish Sebastes rosaceous 12 7 54–152
Honeycomb rockfish Sebastes umbrosus 12 7 46–76
Widow rockfish Sebastes entomelas 95 53
Flag rockfish Sebastes rubrivinctus 9 5 55–60
Copper rockfish Sebastes caurinus 7 4 53–69
Calico rockfish Sebastes dalli 74 53
Treefish Sebastes serriceps 5 3 18–53
Brown rockfish Sebastes auriculatus 4 2 24–53
Starry rockfish Sebastes constellatus 4 2 46–68
Chilipepper Sebastes goodei 3 2 140
Greenstriped rockfish Sebastes elongatus 2 1 177
Squarespot rockfish Sebastes hopkinsi 2 1 61–69
Freckled rockfish Sebastes lentiginosus 2 1 61–76
Bocaccio Sebastes paucispinis 2 1 76–79
Kelp rockfish Sebastes atrovirens 11 30
Yellowtail rockfish Sebastes flavidus 11 53
Canary rockfish Sebastes pinniger 11 53
Total 168 100 .
Table 2. Percent occurrence of barotrauma in nearshore and shelf rockfish (only species with n>3)
following hook-and-line capture from depths of 18–189 m in southern California (initial capture sur-
vival experiment, October 2004 to March 2006).
Occurrence of barotrauma (%)
External signs Internal signs
Common name nSE EX CB PC ST HE OT
No. of traumas
> 50%
Olive rockfish 16 19 6 6 0 88 13 0 1
Treefish 5 60 40 20 0 20 80 02
Halfbanded rockfish 15 67 13 0 13 27 60 53 3
Widow rockfish 9 100 0 0 11 33 33 100 2
Brown rockfish 4 75 25 0 25 0 25 0 1
Calico rockfish 7 43 29 14 43 14 29 43 0
Copper rockfish 7 29 86 71 14 0 43 29 2
Flag rockfish 9 100 33 22 11 0 100 78 3
Starry rockfish 4 100 100 25 0 0 100 03
Honeycomb rockfish 12 92 42 25 8 0 67 17 2
Vermilion rockfish 35 40 71 57 14 6 74 26 3
Rosy rockfish 12 50 83 67 00 75 50 5
Greenspotted rock-
fish
19 84 89 79 11 6 89 32 4
Note: Species are arranged in order of most frequent capture depth. Values in italics indicate signs of baro-
trauma occurring in at least 50% of individuals. SE, stomach eversion; EX, exophthalmia; CB, corneal gas bubbles;
SE, stomach eversion; PC, prolapsed cloaca; ST, swim bladder tear; HE, hemorrhage; OT, organ torsion.
Jarvis and Lowe 1289
#
2008 NRC Canada
without arterial embolism (85%) than for fish with arterial
embolism (58%; two-sided Fisher exact test, p= 0.002).
Percent survival was also significantly higher for fish with-
out swim bladder tears (73%) than for fish with swim blad-
der tears (42%; two-sided Fisher exact test, p= 0.002).
Fifty-four percent of dead fish showing arterial embolism
were rosy and olive rockfish, and 87% of dead fish showing
swim bladder tears were olive rockfish. Externally, rockfish
commonly showed five different categories of barotrauma:
(i) the absence of all signs of barotrauma, (ii) the presence
of only EX and CB, (iii) only SE, (iv) only SE and EX, or
(v) all three signs together (SE, EX, and CB). Internally,
rockfish commonly showed eight different categories of
barotrauma: (i) the absence of all signs of barotrauma,
(ii) the presence of only AE, (iii) only HE, (iv) only AE
and HE, (v) only AE and OT, (vi) only AE and ST,
(vii) only HE and OT, and (viii) HE, OT, and AE together.
The effect of external or internal barotrauma category on
initial survival was not found to be statistically significant
(barotrauma model I, external signs, logistic regression, 2=
8.82, df = 4, p= 0.07; barotrauma model II, internal signs,
logistic regression, 2= 13.71, df = 7, p= 0.06). Within
barotrauma model II, fish showing both arterial embolism
and swim bladder tears had significantly lower survival
than fish showing no signs of barotrauma (p=0.02, OR =
0.05); however, olive rockfish was the only species show-
ing both AE and ST.
Barotrauma and 2-day postrecompression survival
We captured 344 rockfish comprising 17 species over
seven trips. Of these, 328 fish were assessed for external
signs of barotrauma, and 257 fish were recompressed to
original capture depth in 42 cage deployments. A minimum
of three fish and a maximum of 12 were recompressed in a
single cage; the average number of fish per cage (i.e., cage
density) was seven. The average capture depth was 71 m
and ranged from 55 m to 81 m. Five species, vermilion
rockfish, bocaccio (Sebastes paucispinis), and flag, square-
spot (Sebastes hopkinsi), and honeycomb rockfish, com-
prised the majority (82%) of the catch (Table 3). Overall
short-term survival of rockfish was 68% (95% CI: 62% to
73%). Although external signs of barotrauma were common
upon capture, virtually none of the caged fish (<1%) showed
external signs of barotrauma 2 days after recompression.
Of 328 fish, the most common external signs of baro-
trauma observed upon capture were stomach eversion (SE,
88%), exophthalmia (EX, 47%), and corneal gas bubbles
(CB, 36%). The occurrence of prolapsed cloaca was not
common (PC, 7%). Two days after recompression, the ma-
jority of fish were observed with cloudy corneas. Initially,
the cloudy corneas were assumed to be a residual effect of
ocular trauma as a result of the initial capture event, but
75% of fish having no signs of ocular trauma upon capture
also showed cloudy corneas.
The extent of external signs of barotrauma observed upon
capture was species-specific (Table 4). Flag, starry, and ver-
milion rockfish each frequently showed three signs of baro-
trauma. In contrast, only one sign of barotrauma, stomach
eversion, frequently occurred in bocaccio and halfbanded,
speckled (Sebastes ovalis), and squarespot rockfish. The
occurrence of exophthalmia and corneal gas bubbles signifi-
cantly differed among the five most abundant species
(exophthalmia, 2= 44.61, df = 4, p£0.001; corneal gas
bubbles, 2= 24.86, df = 4, p£0.001); however, there was
no significant difference in the occurrence of stomach ever-
sion among these species (2= 3.05, df = 4, p> 0.05).
Overall short-term survival (68%) was similar to that
found in the initial capture survival experiment (also 68%).
Two-day postrecompression survival was species-specific
and ranged from 36% (95% CI: 21% to 56%) for squarespot
rockfish (n= 27) to 82% (95% CI: 51% to 96%) for starry
rockfish (n= 11). There was a significant difference in spe-
cies survival among the five most abundant species (2=
21.6, df = 5, p£0.0001) (Fig. 3); squarespot rockfish
showed the lowest survival, whereas bocaccio showed the
highest survival (95% CI: 79 to 89%). By species, there
Fig. 2. Initial capture survival (±95% binary confidence intervals) of southern California nearshore and shelf rockfish (only species with n>
3) captured by hook-and-line from depths of 18–96 m after a 10-min holding period on deck in a live well (initial capture survival experi-
ment, October 2004 to March 2006). Species are arranged in order of capture depth. Numbers in parentheses indicate sample sizes. Solid
circles represent LaPlace Point estimates for species with survival of either 0% or 100%.
1290 Can. J. Fish. Aquat. Sci. Vol. 65, 2008
#
2008 NRC Canada
were no significant differences in the proportion of survivors
among fish with or without each external sign of barotrauma
(two-sided Fisher exact test, p> 0.05). However, the results
of these analyses for stomach eversion were inconclusive
because sample sizes within species for fish without stom-
ach eversion were too small. When all species were pooled,
there were significantly higher mortalities in fish having
stomach eversion (33%) than in fish without (0%) (two-
sided Fisher exact test, p< 0.001), with bocaccio and vermi-
lion and squarespot rockfish comprising 62% of these mor-
talities. Externally, rockfish commonly showed the same
five categories of external signs of barotrauma as reported
in the initial capture survival experiment. Also similar to
what was reported for initial capture survival, external baro-
trauma category was not a predictor of 2-day postrecompres-
sion survival, although the model results in this case were
much more significant (logistic regression, 2= 1.33744,
df = 4, p= 0.855).
A total of 51 rockfish were dissected to assess internal
signs of barotrauma (survivors, n= 36; dead, n= 15)
2 days after being recompressed in cages. Fewer dead fish
than live fish were dissected because 24% of the dead fish
observed in cages were completely clean carcasses that
could not be assessed. Overall, 50% of fish showing organ
torsion suffered mortality compared with 28% of fish with-
out organ torsion; however, this difference was not signifi-
cant (two-sided Fisher exact test, p= 0.22). There were also
no significant differences in percent mortality for fish with
and without hemorrhage, swim bladder tears, or arterial em-
bolism (two-sided Fisher exact test, p> 0.05). Sample sizes
Table 3. Sample size (n), catch (%), and depth (m) of capture of nearshore and
shelf rockfish captured by hook-and-line in southern California (2-day postrecom-
pression survival experiment, summer 2005 and 2006).
Common name Scientific name nCatch (%) Depth (m)
Vermilion rockfish Sebastes miniatus 73 28.5 55–86
Bocaccio Sebastes paucispinis 64 25.0 57–89
Flag rockfish Sebastes rubrivinctus 29 11.3 55–89
Squarespot rockfish Sebastes hopkinsi 28 10.9 55–83
Honeycomb rockfish Sebastes umbrosus 17 6.6 56–84
Starry rockfish Sebastes constellatus 11 4.3 58–89
Speckled rockfish Sebastes ovalis 11 4.3 80–84
Chilipepper Sebastes goodei 7 2.7 57–84
Halfbanded rockfish Sebastes semicinctus 5 1.6 59–84
Greenspotted rockfish Sebastes chlorostictus 3 1.2 80–83
Olive rockfish Sebastes serranoides 2 0.8 56, 83
Copper rockfish Sebastes caurinus 2 0.8 56, 84
Yellowtail rockfish Sebastes flavidus 1 0.4 84
Rosy rockfish Sebastes rosaceous 1 0.4 82
Greenstriped rockfish Sebastes elongatus 1 0.4 84
Freckled rockfish Sebastes lentiginosus 1 0.4 82
Canary rockfish Sebastes pinniger 1 0.4 89
Total 257 100.0 .
Table 4. Percent occurrence of barotrauma in nearshore and shelf rockfish (only species with n>
10) following hook-and-line capture from depths of 55–89 m in southern California (2-day post-
recompression survival experiment, summer 2005 and 2006).
Occurrence of barotrauma (%)
External signs Internal signsa
Common name nEX CB SE PC nST HE OT AE
Bocaccio 66 33 29 92 0120 100 33 25
Flag rockfish 32 63 59 84 13 5 0 80 040
Halfbanded rockfish 41 12 7 93 12 0 — — — —
Honeycomb rockfish 28 57 46 82 42— ———
Speckled rockfish 13 8 15 85 23 0 — — — —
Squarespot rockfish 40 25 15 88 15 7 43 100 29 29
Starry rockfish 11 82 64 91 0425100 050
Vermilion rockfish 75 81 52 92 0138 85 15 69
Note: Values in italics indicate signs of barotrauma occurring in at least 50% of individuals. EX, exophthal-
mia; CB, corneal gas bubbles; SE, stomach eversion; PC, prolapsed cloaca; ST, swim bladder tear; HE, hemor-
rhage; OT, organ torsion; AE, arterial embolism.
aInternal signs of barotrauma were assessed in a subsample of line-caught rockfish recompressed to original
capture depth in cages.
Jarvis and Lowe 1291
#
2008 NRC Canada
were too small among groups of fish showing specific com-
binations of internal barotrauma to perform logistic
regression analysis.
In the overall model, cage density, depth, and fish length
had no significant effect on survival (Table 5) and, hence,
were removed from the model. Mean values for each variable
and species in the model are reported in Table 6. Species,
surface holding duration, and temperature differential were
significant predictors of survival in the reduced model
(Table 5). Surface holding duration and temperature differen-
tial showed a negative relationship with the predicted proba-
bility of 2-day postrecompression survival (Fig. 4). The odds
of mortality 2 days after recompression increased 1.7 times
with every 10 min increase in surface holding time (e.g., 1/
0.9510; Table 5). The odds of mortality 2 days after recom-
pression increased 1.96 times with every 1 8C increase in
seafloor–surface temperature differential (e.g., 1/0.511).
There was no significant difference in median surface
holding duration among species (Kruskal–Wallis, H= 2.96,
p= 0.56) and no significant difference in average seafloor–
surface temperature differential experienced among species
(ANOVA, F= 0.60, p= 0.66).
Longer-term survival
Of the 125 fish released alive from cages at 20 m 2 days
after capture, 3% were recaptured by local anglers near the
areas where they were released. Two of eight honeycomb
rockfish, two of 41 bocaccio, and one of 40 vermilion rock-
fish that were released alive were recaptured. Days at liberty
for the recaptured honeycomb rockfish were 14 and
208 days. The third recaptured fish, a bocaccio, was reported
after 28 days at liberty. The second recaptured bocaccio
was reported approximately 90 days after the last tagging
event; however, the angler provided no other additional
information (i.e., tag number, date of capture) and so actual
days at liberty are unknown. The vermilion rockfish was at
liberty for 447 days.
Discussion
Barotrauma and initial capture survival
Despite species-specific differences in the types and
degree of angling-induced barotrauma, most rockfish
showed greater than 75% initial capture survival, suggesting
that degree of barotrauma is not a good predictor of mortality
within the first 10 min of capture. Other studies report
high initial capture mortality as a result of gas bubble for-
mation and exhaustion (Feathers and Knable 1983; Kieffer
2000; Stephens et al. 2002). However, in this study, the
initial survival of rockfish overall did not appear dependent
on the presence of gas bubbles. Beyer et al. (1976) reported
that some bubbles present in the blood could be tolerated by
coho salmon (Onchorhynchus kisutch), whereas large gas
bubbles in vital areas such as the heart were lethal. Rosy
and olive rockfish comprised the majority of dead fish with
arterial embolism, suggesting these species may be more
susceptible to lethal arterial embolisms than other species.
Exhaustion may restrict the ability of the opercular
muscles to pump water past the gills and may also account
for species-specific differences in initial capture mortality
(Kieffer 2000). All of the olive rockfish and many rosy
rockfish were moribund directly upon capture, suggesting
that these two species may be susceptible to exhaustion.
This exhaustion, in combination with the high incidence of
arterial embolism observed in dead individuals, may account
for the higher initial capture mortality of these two species
relative to the other rockfish species.
Intraspecific variability in barotrauma responses of fish
captured at similar depths are likely due to differences in
the relative volume of the swim bladders when fish are
caught (Arnold and Walker 1992; Rummer and Bennett
2005; Parker et al. 2006). For example, extent of barotrauma
will vary depending on whether the fish is neutrally buoyant
at the depth of capture (Parker et al. 2006). Further, inter-
specific variability in swim bladder morphology may also
influence interspecific variation in the occurrence of swim
bladder tears. During examination of internal signs of baro-
trauma, olive rockfish swim bladders were found to be rela-
tively thin compared with the more robust swim bladders
observed in vermilion, copper, and brown rockfish. Given
the high occurrence of swim bladder tears in olive rockfish
observed in this study relative to the other species, a thinner
swim bladder may be more prone to severe rupture than a
robust swim bladder, as seen in other species (Feathers and
Knable 1983). Although olive rockfish showed high mortal-
ity and high occurrence of swim bladder tears, all other
rockfish with swim bladder tears lived except for two (n=
17). Nevertheless, longer-term survival may be compro-
mised by structural damage to the swim bladder and (or)
other organs (Parker et al. 2006).
Barotrauma and 2-day postrecompression survival
The species-specific differences in external signs of baro-
trauma reported in this study appear to be related to species
Fig. 3. Two-day survival (±95% binary confidence intervals) of the
five most abundant species captured by hook-and-line in southern
California from depths of 55–89 m and recompressed to original
capture depth in cages (2-day postrecompression survival experi-
ment, summer 2005 and 2006). Numbers in parentheses represent
samples sizes. Percent survival was significantly different among
species (2= 21.6, df = 5, p< 0.0001).
1292 Can. J. Fish. Aquat. Sci. Vol. 65, 2008
#
2008 NRC Canada
differences in body morphology and also to the degree of
vertical movement within the water column. For example,
species that showed few signs of barotrauma (e.g., bocaccio
and squarespot, speckled, and halfbanded rockfish) can be
characterized as having relatively elongate, laterally com-
pressed bodies and occurring in schools up off the seafloor
(Love et al. 2002). In contrast, flag, honeycomb, vermilion,
and starry rockfish, which are all relatively deep-bodied and
more demersal, showed a high degree of barotrauma. Species
that show a high degree of barotrauma might be expected
to have low survival following recompression relative to
those species showing very few external signs of baro-
trauma; however, we did not observe this trend. Further,
barotrauma category (e.g., specific combinations of external
signs of barotrauma) was not a significant predictor of sur-
vival. Although the presence of specific external signs of
barotrauma was not a clear indication of 2-day postrecom-
pression survival, as has been found in other species (Gits-
chlag and Renaud 1994; Rummer and Bennett 2005), the
absence of a specific sign of barotrauma may be more telling.
For instance, 2-day postrecompression survival was signifi-
cantly higher for fish without stomach eversion.
Stomach eversion may occur as a result of swim bladder
leaks or ruptures. Although visible swim bladder tears were
not common for the catch as a whole, we did observe many
fish with partially deflated swim bladders, suggesting that
escapement of gas at high pressure likely occurred during
ascent. Although the mechanism by which swim bladders re-
seal after overinflation is not well understood, this may ex-
plain why not all rockfish with stomach eversion showed
visible tears in the swim bladder wall. Delayed mortality in
fish with stomach eversion may be a result of internal organ
torsion associated with the occurrence of stomach eversion
and (or) internal organ damage resulting from the over-
inflated swim bladder crushing organs (Keniry et al. 1996;
Rummer and Bennett 2005). Fifty percent (4 of 8 fish)
exhibiting organ torsion died versus only 26% (11 of
42 fish) without organ torsion. Low sample size among fish
with organ torsion (n= 8) may explain why this difference
was not significant. Further analyses of internal signs of bar-
otrauma as predictors of postrecompression survival could
not be conducted because of low sample sizes among dead
specimens (n= 15).
Rummer and Bennett (2005) reported that internal organ
damage of red snapper (Lutjanus campechanus) increased
with depth of capture because of the progressive expansion
of the swim bladder from posterior to anterior regions of
the body cavity. Although not specifically tested, it is likely
that swim bladder rupture thresholds and the space allow-
able for swim bladder expansion may also vary across rock-
fish species. Thus, the extent of organ damage in rockfish
may differ among species with different body shapes and
(or) swim bladder morphologies, even at similar capture
depths. For example, squarespot rockfish (low survival) and
bocaccio (high survival) frequently showed stomach ever-
sion. Squarespot rockfish were observed as having thinner
swim bladder morphology similar to that of olive, widow,
and halfbanded rockfish, and three of six fish showed swim
bladder tears. In contrast, bocaccio were observed to have
relatively thicker swim bladders. Bocaccio swim bladders
may rupture only minutely, allowing gas escapement into
the body cavity at high pressure, which would force the
stomach out of the mouth but nevertheless, leave the swim
bladder visibly intact. Only one of 12 bocaccio showed
swim bladder tears. It is important to note that swim bladder
ruptures, although common in some species, may not be
deleterious, at least in the short term. Several studies indi-
cate a remarkable ability of physoclistic species to repair
ruptured swim bladders within a few days or even hours
(Burns and Restrepo 2002; Nichol and Chilton 2006; Parker
et al. 2006).
In the initial capture survival experiment, the degree of
barotrauma in rockfish was partially attributed to depth;
however, depth was not a significant predictor of 2-day
postrecompression survival of rockfish. Depth has been
shown in numerous studies to significantly affect postrelease
survival in other physoclistic species (Wilson and Burns
1996; Morrissey et al. 2005; St. John and Syers 2005).
Nevertheless, because rockfish exhibit a wide range of
foraging behavior (e.g., benthic ambush predators versus
water-column planktivores) and were observed to differ in
swim bladder morphology as discussed above, depth effects
on rockfish survival corresponding to the incidence and
degree of barotrauma are likely to differ by species (Lea et
al. 1999; Parker et al. 2006; Hannah and Matteson 2007).
In addition to depth, fish size was not a significant pre-
dictor of 2-day postrecompression survival. It has been
hypothesized that smaller fish may be more susceptible to
gas embolism because of the relationship between critical
blood vessel size and the size of expanded bubbles in the
blood (Beyer et al. 1976), and thermal shock may occur
more quickly in smaller fish than in larger fish (Davis
2002); however, other studies report no effect of fish length
on short-term survival (Gitschlag and Renaud 1994; Collins
et al. 1999).
Cage density was not found to affect 2-day postrecom-
pression survival; nevertheless, we cannot dismiss that
Table 5. Logistic regression results of the overall and reduced
models of 2-day survival of rockfish captured by hook-and-line
in southern California from depths of 55–89 m and recom-
pressed in cages to original capture depth (2-day postrecompres-
sion survival experiment, summer 2005 and 2006).
Predictor of
survival df X2pOR
Overall model
Species 4 — 11.23 0.024* —
Surface time (min) 1 –0.626 9.69 0.002* 0.94
Depth (m) 1 –0.0277 2.22 0.136 0.97
Fish length
(SL, cm)
1 0.0419 0.99 0.319 1.04
Cage density 1 –0.0462 0.26 0.614 0.96
Temperature
difference (8C)
1 –0.6512 7.57 0.006* 0.52
Reduced model
Species 4 — 19.76 0.001* —
Surface time (min) 1 –0.0566 8.6312 0.003* 0.95
Temperature
differential (8C)
1 –0.6705 8.4990 0.004* 0.51
Note: Asterisks (*) indicate significance at p< 0.05. df, degrees of
freedom; OR, odds ratio; SL, standard length.
Jarvis and Lowe 1293
#
2008 NRC Canada
observed cage abrasion and potentially high stress levels due
to the confinement of the cages may have contributed to
some mortality (Gliniak et al. 2006). The presence of scav-
engers such as crustacean zooplankton or hagfish, as sug-
gested by the clean carcasses of several fish, may have
further exacerbated stress levels in other fish that had not
suffered immediate mortality (Stepien and Brusca 1985).
Seventy-five percent of fish showed cloudy corneas
2 days after recompression, but fish were observed bumping
into the sides of the cages during cage retrievals. We con-
clude that cloudy corneas were likely a symptom of keratitis
as a result of cage abrasion because even fish without initial
ocular trauma exhibited cloudy corneas after recompression
(St. John and Syers 2005). Recompression appeared to
reverse or alleviate external signs of barotrauma observed in
rockfish upon capture, which has also been noted in pre-
vious studies (St. John and Syers 2005; Parker et al. 2006;
Hannah and Matteson 2007). Still, it is unclear whether
ocular trauma may result in long-term visual impairment.
Rapid recompression of line-caught rockfish can reduce
the extent of injury caused by arterial embolism and hemor-
rhaging and thereby increase survival. In this study, surface
holding time was found to have a significant effect on
recompression survival. Fish held at the surface for 10 min
or less had a 78% probability of survival following recom-
pression, and this probability of survival increased to 83%
if fish were released within 2 min of landing. Parker et al.
(2006) reported 97% survival (up to 21 days) of black rock-
fish (Sebastes melanops) recompressed within 30 s of
decompression (from 4 atm) in pressure chambers. Surface
holding duration may explain species-specific differences in
survival in our study. Of the five most abundant species
caught (e.g., bocaccio and vermilion, flag, honeycomb, and
squarespot rockfish), squarespot rockfish showed the lowest
survival. It is possible that surface holding duration may
have affected squarespot rockfish survival because these
fish were held at the surface an average 5 min longer than
the other four species (Table 6). This difference in surface
holding duration, although not statistically significant, may
have been biologically significant.
Fish held at the surface for long periods of time may also
experience thermal stress (Feathers and Knable 1983).
Reduced blood flow as a result of intravascular bubble
formation following decompression may be especially detri-
mental in warmer surface waters where oxygen demand is
higher and oxygen concentration is lower. Feathers and
Knable (1983) reported a synergistic effect of barotrauma
and surface temperature on mortality of released largemouth
bass (Micropterus salmoides). Because of the physiological
effects of temperature on rockfish and the negative affect of
increased temperature differential on short-term survival, it
is possible that postrelease survival probabilities relative to
surface holding duration may increase during winter months
when sea surface temperatures differ less from seafloor tem-
peratures.
Longer-term survival
Although tag recaptures were few, they provided evidence
of longer-term postrelease survival (at least up to 1.5 years)
in line-caught rockfish following recompression. Although
these data provide no indication of actual delayed mortality
following release, other studies document delayed capture
mortality following recompression up to the first several
days of release (Gitschlag and Renaud 1994; Wilson and
Burns 1996; Lowe et al. 2007). For example, 93% of line-
caught rockfish that were acoustically tagged at the surface,
recompressed to depth, and monitored for a 2-year period
were detected within the first 2 days after their release
(Lowe et al. 2007). Subsequently, the decline in detections
leveled out 6 days after release. Cage recompression studies
on line-caught red snapper and red grouper (Epinephelus
morio) found that delayed mortality continued up to
10 days, although the majority of deaths occurred within
Table 6. Species catch-and-release summary data (mean ± standard deviation, SD) for the five most
abundant rockfish captured by hook-and-line in southern California and recompressed to original cap-
ture depth in cages (2-day postrecompression survival experiment, summer 2005 and 2006).
Common name
Standard
length (cm) Depth (m)
Cage
density
Temperature
differential (8C)
Surface holding
duration (min)
Bocaccio 35.2 (7.4) 83.3 (4.3) 7 (2) 9.7 (0.7) 14.4 (6.7)
Flag rockfish 21.5 (2.7) 69.9 (13.1) 8 (2) 9.4 (0.6) 13.6 (9.3)
Honeycomb rockfish 16.3 (1.7) 76.7 (10.8) 8 (3) 8.2 (0.7) 14.9 (9.1)
Squarespot rockfish 19.9 (1.3) 70.2 (12.0) 7 (2) 9.3 (0.7) 21.5 (21.2)
Vermilion rockfish 24.3 (4.0) 63.8 (11.5) 7 (2) 9.0 (1.0) 16.0 (9.7)
Overall mean (±1 SD) 26.0 (8.1) 72.4 (12.9) 7 (2) 9.2 (0.9) 15.8 (11.3)
Fig. 4. Predicted probability (±95% confidence interval) of rockfish
2-day postrecompression survival as a function of surface holding
time (min) (2-day postrecompression survival experiment, summer
2005 and 2006). The probability curve is based on the five most
abundant rockfish captured during the 2-day cage trials; n= 211 fish).
1294 Can. J. Fish. Aquat. Sci. Vol. 65, 2008
#
2008 NRC Canada
the first 2 days (Gitschlag and Renaud 1994; Wilson and
Burns 1996).
Fisheries management implications
The results of this research provide evidence of both
short-term and long-term postrelease survival of line-caught
southern California nearshore and shelf rockfish recom-
pressed to capture depth (from 55 to 89 m). Findings
suggest that the utility of recompression devices is high if
used within minutes of capture. A variety of assisted release
(i.e., recompression) methods, including inverted weighted
milk crates and ‘‘fish descenders’’, currently exist to return
overinflated fish to depth following capture (Theberge and
Parker 2005). Although the use of recompression devices
would not be practical for a strictly catch-and-release fish-
ery, recompression would provide a practical means for de-
creasing discard mortality of incidentally caught rockfish.
Between 1993 and 2002, an estimated two million rock-
fish were discarded in southern California (Pacific States
Marine Fisheries Commission 2007). Fifty-eight percent of
fish were discarded from private boats. Dwarf species (i.e.,
honeycomb, squarespot, and halfbanded rockfish) comprise
a large proportion of rockfish discards (M. Horeczko,
California Department of Fish and Game, 4665 Lampson
Avenue, Los Alamitos, CA 90720, USA, personal communi-
cation), which likely suffer high mortality. These species are
now among the common rockfish caught in southern Cali-
fornia (Love et al. 1998), although they are seldom kept by
fishers because of their small size. The practice of using
assisted release in the recreational rockfish fishery may help
restore rockfish diversity on rocky reefs that have suffered
decades of intense fishing pressure, which in turn would
help to increase ecosystem function and resilience to fishery
collapse (Worm et al. 2006).
Minimum size limits have not been considered as a
management tool for rockfish because of the perceived high
release mortality attributed to barotrauma (J. Ugoretz,
California Department of Fish and Game, 20 Lower
Ragsdale Drive, Monterey, CA 93940, USA, personal com-
munication), but federal and state mandates outlined in the
Magnuson Fishery Conservation and Management Act
(MFCMA) [301 (a)(9)] and the Marine Life Management
Act (MLMA) [7056(d)] require a reduction in groundfish
bycatch (i.e., incidental catch + discards) mortality. Some
studies have evaluated the effectiveness of deflating swim
bladders to increase survival of overinflated fish by relieving
excess gas with a hypodermic needle (‘‘venting’’), which
enables fish to swim to depth on their own following release
(Bruesewitz et al. 1993; Keniry et al. 1996; Collins et al.
1999). However, venting is still controversial because the
results of these studies are not definitive and experience in
using this method is necessary to avoid accidental puncture
of vital organs (Kerr 2001). Implementation of less invasive
assisted release methods would increase the practicality of
size limits as an alternative to current nearshore and shelf
rockfish regulations. Nevertheless, for minimum size limits
to be effective in conserving rockfish stocks, fish that are re-
compressed to depth after capture must live long enough to
reach size at maturity. For long-lived species such as rock-
fish, it may be important to consider cumulative mortality
risk, which increases exponentially with every recapture
(Bartholomew and Bohnsack 2005). Based on rockfish re-
capture rates from this study and those reported from the
Nearshore Groundfish Tagging Project, rockfish recapture
rates in southern California appear low (3%; Hanan and As-
sociates Inc. 2006). Furthermore, the possibility of subse-
quent mortality of a recaptured immature fish is likely
offset by the decrease in overall fishing mortality that would
be provided by implementing recompression. Other factors,
including vision impairment and (or) angling-induced
physiological impacts on growth and reproduction that may
decrease rockfish longevity following capture and recom-
pression, need further exploration.
Acknowledgments
The authors thank E. Ferna
´ndez-Juricic and J. Archie of
California State University, Long Beach (CSULB), and
K. Ritter of Southern California Coastal Water Research
Project for guidance in statistical analyses and J. Cvitano-
vich and the support staff on the R/V Yellowfin of Southern
California Marine Institute for assistance with cage experi-
ments. The authors also gratefully acknowledge all of the
volunteer fishers and divers in the CSULB Shark Lab and
the Long Beach Aquarium of the Pacific. Support for this
research was provided, in part, by the University of South-
ern California Sea Grant College Program of NOAA under
grant No. NA06OAR4170012 and the State of California
Resources Agency and, in part, by the Minerals Manage-
ment Service under cooperative agreement No. 1435-01-04-
CA-34196. Additional funding was provided by CSULB
CNSM – Boeing Scholarship, Southern California Academy
of Sciences Research Grant, SCTC Marine Biology Foun-
dation, and the CSULB Richard B. Loomis Research
Award.
References
Arnold, G.P., and Walker, M.G. 1992. Vertical movements of cod
(Gadus morhua L.) in the open sea and the hydrostatic function
of the swimbladder. ICES J. Mar. Sci. 49: 357–372. doi:10.
1093/icesjms/49.3.357.
Bartholomew, A., and Bohnsack, J.A. 2005. A review of catch-and-
release angling mortality with implications for no-take reserves.
Rev. Fish Biol. Fish. 15: 129–154. doi:10.1007/s11160-005-
2175-1.
Beyer, D.L., D’Aoust, B.G., and Smith, L.S. 1976. Decompression-
induced bubble formation in salmonids: comparison to gas bubble
disease. Undersea Biomed. Res. 3: 121–124. PMID: 951822.
Bruesewitz, R.E., Coble, D.W., and Copes, F. 1993. Effects of
deflating the expanded swim bladder on survival of burbot. N.
Am. J. Fish. Manag. 13: 346–348. doi:10.1577/1548-8675(1993)
013<0346:EODTES>2.3.CO;2.
Burns, K.M., and Restrepo, V. 2002. Survival of reef fish after
rapid depressurization: field and laboratory studies. Am. Fish.
Soc. Symp. 30: 148–151. [Extended abstract.]
Butler, J.L., Jacobson, L.D., Barnes, J.T., and Moser, H.G. 2003.
Biology and population dynamics of cowcod (Sebastes levis)in
the southern California Bight. Fish. Bull. (Washington, D.C.),
101: 260–280.
Collins, M.R., McGovern, J.C., Sedberry, G.R., Meister, H.S., and
Pardieck, R. 1999. Swim bladder deflation in black sea bass and
vermilion snapper: potential for increasing postrelease survival. N.
Am. J. Fish. Manag. 19: 828–832. doi:10.1577/1548-8675(1999)
019<0828:SBDIBS>2.0.CO;2.
Jarvis and Lowe 1295
#
2008 NRC Canada
Davis, M.W. 2002. Key principles for understanding fish bycatch
discard mortality. Can. J. Fish. Aquat. Sci. 59: 1834–1843.
doi:10.1139/f02-139.
Dotson, R.C., and Charter, R.L. 1998. Trends in the Southern Cali-
fornia sport fishery. Cal. Coop. Ocean. Fish. Investig. Rep. 44:
94–106.
Feathers, M.G., and Knable, A.E. 1983. Effects of depressurization
upon largemouth bass. N. Am. J. Fish. Manag. 3: 86–90. doi:10.
1577/1548-8659(1983)3<86:EODULB>2.0.CO;2.
Gitschlag, G.R., and Renaud, M.L. 1994. Field experiments on sur-
vival rates of caged and released red snapper. N. Am. J. Fish.
Manag. 14: 131–136. doi:10.1577/1548-8675(1994)014<0131:
FEOSRO>2.3.CO;2.
Gliniak, H., Lowe, C.G., and Kelley, K.M. 2006. Catching-related
stressors and cortisol response in eastern Pacific rockfishes. In
VIIth International Congress on the Biology of Fish, St. John’s,
Newfoundland, Canada. Available online at http://www.
fishbiologycongress.org/. p. 102. [Abstract.]
Gotshall, D.W. 1964. Increasing tagged rockfish (genus Sebastodes)
survival by deflating the swim bladder. Calif. Fish Game, 50:
253–260.
Hanan and Associates Inc. 2006. Utilizing commercial passenger
fishing vessels to catch and tag nearshore groundfish. Pacific
States Marine Fisheries Commission, West Coast Cooperative
Research Program, Portland, Oregon, PSMFC Contract
No. 05-23.
Hannah, R.W., and Matteson, K.M. 2007. Behavior of nine species
of Pacific rockfish after hook-and-line capture, recompression,
and release. Trans. Am. Fish. Soc. 136: 24–33. doi:10.1577/
T06-022.1.
Keniry, M.J., Brofka, W.A., Horns, W.H., and Marsden, J.E. 1996.
Effects of decompression and puncturing the gas bladder on survi-
val of tagged yellow perch. N. Am. J. Fish. Manag. 16: 201–206.
doi:10.1577/1548-8675(1996)016<0201:EODAPT>2.3.CO;2.
Kerr, S.J. 2001. A review of ‘‘fizzing’’ — a technique for swim
bladder deflation. Fish and Wildlife Branch, Ontario Ministry of
Natural Resources, Peterborough, Ontario.
Kieffer, J.D. 2000. Limits to exhaustive exercise in fish. Comp.
Biochem. Physiol. A, 126: 161–179.
Lea, R.N., McAllister, R.D., and Ventresca, D.A. 1999. Biological
aspects of nearshore rockfishes of the genus Sebastes from
Central California. Calif. Dep. Fish Game Fish Bull. 177:3–
109.
Love, M.S., Caselle, J.E., and Van Buskirk, W. 1998. A severe
decline in the commercial passenger fishing vessel rockfish
(Sebastes spp.) catch in the southern California Bight, 1980–
1996. Cal. Coop. Ocean. Fish. Investig. Rep. 39: 180–195.
Love, M.S., Yoklavich, M., and Thorsteinson, L. 2002. The rock-
fishes of the Northeast Pacific. University of California Press,
Berkeley, California.
Lowe, C.G., Anthony, K.M., Jarvis, E.T., Bellquist, L.F., and Love,
M.S. 2007. Site fidelity of characteristic fish species at offshore
petroleum platforms in the Santa Barbara Channel. Minerals
Management Service, Pacific OCS Region, MMS OCS Study
2004-006, California State University Long Beach, Long Beach,
California. MMS Cooperative Agreement No. 1435-01-04-CA-
34196.
Mason, J. 1998. Declining rockfish lengths in the Monterey Bay,
California, recreational fishery in 1959–94. Mar. Fish. Rev. 60:
15–28.
Morrissey, M.B., Suski, C.D., Esseltine, K.R., and Tufts, B.L.
2005. Incidence and physiological consequences of decompres-
sion in smallmouth bass after live-release angling tournaments.
Trans. Am. Fish. Soc. 134: 1038–1047. doi:10.1577/T05-010.1.
Nichol, D.G., and Chilton, E.A. 2006. Recuperation and behaviour
of Pacific cod after barotrauma. ICES J. Mar. Sci. 63: 83–94.
doi:10.1016/j.icesjms.2005.05.021.
Parker, S.J., McElderry, H.I., Rankin, P.S., and Hannah, R.W.
2006. Buoyancy regulation and barotrauma in two species of
nearshore rockfish. Trans. Am. Fish. Soc. 135: 1213–1223. doi:
10.1577/T06-014.1.
Pacific States Marine Fisheries Commission. 2007. Recreational
Fisheries Information Network: RecFIN estimate summary re-
sults. Available from the RecFIN database online at http://www.
psmfc.org/recfin. [Accessed 26 March 2007.]
Rummer, J.L., and Bennett, W.A. 2005. Physiological effects of
swim bladder overexpansion and catastrophic decompression on
red snapper. Trans. Am. Fish. Soc. 134: 1457–1470. doi:10.1577/
T04-235.1.
Sauro, J., and Lewis, J.R. 2005. Estimating completion rates from
small samples using binomial confidence intervals: comparisons
and recommendations. In Proceedings of the Human Factors and
Ergonomics Society 49th Annual Meeting, Orlando, Florida, 26–
30 September 2005. Human Factors and Ergonomics Society,
Santa Monica, Calif. pp. 2100–2104.
Stephens, F.J., Cleary, J.J., Jenkins, G., Jones, B., Raidal, S.R.,
and Thomas, J.B. 2002. Haemoglobin and oxygen transport of
the West Australian dhufish, Glaucosoma hebraicum Richard-
son, and other species. J. Fish Dis. 25: 409–414. doi:10.1046/j.
1365-2761.2002.00389.x.
Stepien, C.A., and Brusca, R.C. 1985. Nocturnal attacks on near-
shore fishes in southern California by crustacean zooplankton.
Mar. Ecol. Prog. Ser. 25: 91–105. doi:10.3354/meps025091.
St. John, J., and Syers, C.J. 2005. Mortality of the demersal West
Australian dhufish, Glaucosoma hebraicum (Richardson 1845),
following catch and release: the influence of capture depth,
venting and hook type. Fish. Res. 76: 106–116. doi:10.1016/j.
fishres.2005.05.014.
Theberge, S., and Parker, S.J. 2005. Release methods for rockfish.
Oregon Sea Grant, Corvallis, Oregon, No. ORESU-G-05-001.
Wilson, R.R., and Burns, K.M. 1996. Potential survival of released
groupers caught deeper than 40 m based on shipboard and in-
situ observations, and tag-recapture data. Bull. Mar. Sci. 58:
234–247.
Worm, B., Barbier, E.B., Duffy, J.E., Folke, C., Halpern, B.S.,
Jackson, J.B.C., Lotze, H.K., Micheli, F., Palumbi, S.R., Sala,
E., Selkoe, K.A., Stachowicz, J.J., and Watson, R. 2006. Impacts
of biodiversity loss on ecosystem services. Science (Washington,
D.C.), 314: 787–790. doi:10.1126/science.1132294. PMID:
17082450.
1296 Can. J. Fish. Aquat. Sci. Vol. 65, 2008
#
2008 NRC Canada
