Northeast Fisheries Science Center Reference Document 06-13
Estimates
of Cetacean and Seal Bycatch
in the 2004 Northeast Sink Gillnet
and
Mid-Atlantic Coastal Gillnet
Fisheries
by Dana
L. Belden, Christopher
D. Orphanides, Marjorie C. Rossman,
and Debra
L. Palka
National Marine Fisheries Serv., 166 Water St., Woods Hole MA 02543
Print
publication date July 2006;
web version posted July 17, 2006
Citation: Belden
DL, Orphanides CD, Rossman MC, Palka DL. 2006. Estimates of Cetacean
and Seal Bycatch in the 2004 Northeast Sink Gillnet and Mid-Atlantic
Coastal Gillnet
Fisheries. U.S. Dep. Commer., Northeast Fish. Sci. Cent. Ref. Doc.
06-13; 24 p.
Download complete PDF/print version
Abstract: This
report provides incidental take estimates for six marine mammal
species taken in the 2004 Northeast Sink Gillnet (NESG) and Mid-Atlantic
Coastal Gillnet (MACG) fisheries, and documents the methodology used
to produce the estimates. The total estimated marine mammal takes
in the 2004 NESG (2,292 animals, coefficient of variation [CV] =
18%) and MACG (231 animals, CV = 60%) fisheries was 2,523 animals (CV
= 17%). For
the NESG fishery, the species included in this estimate are: 655
harbor porpoises (CV = 36%); 7 Atlantic white-sided dolphins (CV =
98%); 498 gray seals (CV = 35%); 786 harbor seals (CV = 34%); 303 harp
seals (CV = 30%); and 43 hooded seals (CV = 95%). For the
MACG fishery, the species included in this estimate are: 135 harbor
porpoises (CV = 91%); 68 gray seals (CV = 92%); and 28 harbor seals
(CV = 77%).
INTRODUCTION
Pursuant to the 1994 amendments of the Marine Mammal Protection
Act (MMPA), section 117 states that estimates of annual human-caused
mortality and serious injury to marine mammal stocks must be reported
in annual stock assessment reports (SAR) for each stock of marine mammal
that occurs in waters under US jurisdiction.
The Northeast Fisheries Science Center (NEFSC) sea sampling observer
program (SSOP), presently known as the Northeast Fisheries Observer Program
(NEFOP), was initiated in 1989 to document the bycatch of marine mammals
taken incidentally to commercial fishing operations (Waring et al. 2003). Since
the initiation of the observer program, the estimation of total takes
for harbor porpoise (Phocoena phocoena) has been the focus of
much attention due to frequent observations of incidental takes occurring
in the Northeast Sink Gillnet (NESG) fishery (NMFS 1998). This
attention led to the development of a stratification method designed
to estimate the total annual takes of harbor porpoise (Bisack 1993, Smith
et al. 1993, Bravington and Bisack 1996, Bisack 1997, Rossman and Merrick
1999, Bisack 2003). The regional scope of the SSOP was expanded
into the mid-Atlantic (MA) region in 1995 as an effort to learn more
about marine mammal interactions occurring in MA gillnet fisheries.
Rossman and Merrick (1999) documents the methods used to estimate harbor
porpoise bycatch in the NESG and Mid-Atlantic Coastal Gillnet (MACG)
fisheries. The methods used to estimate harbor porpoise bycatch
have also been used by the NEFSC to estimate the bycatch of other marine
mammals observed bycaught in the NESG and MACG fisheries (Blaylock et
al. 1995, Waring et al. 1997, Waring et al. 2003). However, to
date there has been no documentation of the methods used to estimate
takes for all marine mammals incidentally bycaught in both the NESG and
MACG fisheries to support the estimates that have been reported in the
marine mammal SARs.
The NESG fishery extends from Maine to Rhode Island and is dominated
by bottom-tending sink gillnets. Less than 1% of the fishery utilizes
a drift gillnet (not tending the ocean bottom). Monofilament twine
is typically used with stretched mesh sizes ranging from 6 to 12 inches. String
lengths range from 600 to 10,500 feet. Mesh size and string length
vary by the primary fish species targeted for catch. The MACG fishery
ranges from Connecticut to North Carolina and utilizes both drift and
sink gillnets. These nets are most frequently attached to the bottom,
although unanchored drift or sink nets are also utilized to target specific
species. Monofilament twine is again the dominant material and
is used with stretched mesh sizes ranging from 2.5 to 12 inches. String
lengths range from 150 to 8400 feet. The mesh size and string length
vary by the primary fish species targeted for catch (Waring et al. 2003).
The purpose of this manuscript is to first document incidental take
estimates for all marine mammals that were observed in the NESG and MACG
fisheries during 2004. Secondly, this manuscript’s purpose
is to discuss the shortcomings of a 10-year-old method that was spatially
and temporally designed to estimate harbor porpoise takes, but has been
used to estimate takes on a variety of other marine mammal species. Suggestions
are provided for improvements to reflect current fishing patterns in
gillnet fisheries and improved data collection.
METHODS
Data Sources
Three databases were used to estimate the total marine mammal takes
in 2004: the NEFOP database, Northeast Dealer Reports, and Northeast
Vessel Trip Reports.
First, the NEFOP database provided data on the observed bycatch of marine
mammals. The NEFOP has two types of sampling protocols when observing
fishing trips: (1) complete fish sampled trips where the observer samples
the catch for fish discard information (the observer is not able to watch
the net as it is being hauled) and (2) limited fish sampled trips where
the observer watches the net for incidental takes as it is being hauled. In
the NESG fishery only, hauls observed from both trip sampling protocols
were used to estimate the bycatch rates from observed incidental takes. Only
limited fish sampling trips were used in the MACG fishery to estimate
the bycatch rates. For purposes of this manuscript a “take” is
defined as any observed incidental take where the animal’s condition
was recorded as either alive with injuries or dead (fresh or under various
stages of decomposition). All incidental takes are identified to
species whenever possible by the fishery observer. There were several
incidental takes that were not identified to species: one unknown dolphin
species; four unknown porpoise/dolphin species; and nine unknown seal
species. These animals were not included in the bycatch estimates
for the strata they were caught in.
Second, the Northeast (NE) Dealer Report landings database was used
to determine the total landings in 2004 of all finfish caught in the
Northeast gillnet fishery.
Lastly, the NE Vessel Trip Report (VTR) database was used to assign
(prorate) the NE Dealer Report landings from the NESG fishery to spatial
and temporal strata historically used to estimate takes of harbor porpoise
in the NESG fishery (Rossman and Merrick 1999, Bisack 2003).
The level of sampling (observer coverage) for each stratum was calculated
by dividing the observed tons of fish caught by the prorated tons of
fish recorded in the dealer database. This value represented the
fraction of total landings that were sampled.
Analysis
The strata defined in Rossman and Merrick (1999) was used to estimate
takes in 2004. The NESG data was stratified temporally by season,
spatially by port group-area and time/area closures, and by bycatch avoidance
techniques via the use of pingers (Table 1; Figure 1). Seasons
are defined as: winter (January to May), summer (June to August), and
fall (September to December). Fishing effort observed in time/area
closures was analyzed in separate time/area closure strata. The
MACG data were not stratified, but were examined by month and state for
each bycaught species. See Figure 2 for the time/area closures
for the MA region.
The number of marine mammal takes (B) is the product of the observed
bycatch rate multiplied by the total effort in each stratum (S). The
bycatch rate for each stratum is defined as the number of observed takes
divided by the observed tons (effort) of fish landed.
There is a possibility that strings could be either equipped or not
equipped with pingers in the NESG fishery. Therefore, a weighted
bycatch rate was calculated for strata where there were hauls with and
without pingers. The weighted bycatch rate was calculated as the
sum of two weighted bycatch rates -- one from hauls with pingers and
one from hauls without pingers -- within a stratum. Each bycatch
rate was weighted by the proportion of hauls sampled with or without
pingers within its respective stratum.
Standard bootstrapping techniques were used to derive the confidence
intervals and coefficients of variation (CV) for the bycatch estimates
for each stratum. The resampling unit used was an entire trip rather
than individual hauls, to ensure that any within-trip dependence was
carried over into the bycatch estimates (Bisack, 2003).
RESULTS
For the NESG fishery, observer coverage was 3%, 8%, and 10% during the
winter, summer, and fall seasons, respectively (Table 2). The total
annual coverage for the NESG fishery was 6% (Table
2). There were
110 observed incidental takes of marine mammals: 27 harbor porpoise,
one Atlantic white-sided dolphin, 21 gray seals, 45 harbor seals, 15
harp seals, and one hooded seal (Figure
3 and Figure
4). For the MACG
fishery, observer coverage was 2%, 2%, and 3% during the winter, summer,
and fall, respectively (Table
3). The total annual coverage was
2% in the MACG fishery. There were four observed incidental takes
of marine mammals: two harbor porpoise, one gray seal, and one harbor
seal (Figure
5 and Figure
6).
For the NESG fishery there were 19,083.59 tons of fish landed with the
seasonal breakdown as follows: winter (53%); summer (24%); and fall (23%;
Table
2). For the MACG fishery there were 14,627.14 tons of fish
landed with the seasonal breakdown, which is the same as that used in
the NESG fishery, as follows: winter (50%); fall (31%); and summer (19%;
Table
3).
In the NESG fishery only one of the 11 observed harbor porpoise takes
that were in time/area closures requiring pingers was actually caught
in a net equipped with pingers (Table 4). Outside of the NESG time/area
closures, three of the 16 observed harbor porpoise takes were caught
in nets equipped with pingers (Table 4). The only observed Atlantic
white-sided dolphin take was in a NESG time/area closure and was caught
in a net equipped with pingers (Table 5). One of the four observed
gray seal takes in NESG time/area closures requiring pingers was taken
in a net equipped with pingers (Table 6). Eight of the 17 observed
harbor seal takes in NESG time/area closures requiring pingers were caught
in nets equipped with pingers (Table 7). All three of the observed
harp seal takes in NESG time/area closures requiring pingers were caught
in nets not equipped with pingers (Table 8). The one observed hooded
seal take in NESG time/area closures requiring pingers was caught in
a net not equipped with pingers (Table 9).
The 2004 estimated total takes of cetaceans in the NESG and the MACG
fisheries was 797 animals (CV = 33%). The NESG fishery had 655
(CV = 36%) harbor porpoise (Table 4) and 7 (CV = 98%) Atlantic white-sided
dolphin estimated takes (Table 5). The MACG fishery had 135 (CV
= 91%) harbor porpoise estimated takes (Table 10). One incidental
take of a bottlenose dolphin in the NESG Fishery was also observed, but
takes of Atlantic bottlenose dolphins are analyzed with different methods
that are not reported here (Palka and Rossman 2001).
The majority (81%) of the NESG fishery harbor porpoise takes occurred
in the winter, with 19% in the fall, and no takes observed in the summer
(though some summer incidental takes have been observed in previous years). During
the winter, the South of Cape Cod port group-area and the South Cape
closure had 46% and 20% of the entire NESG fishery estimated harbor porpoise
takes, respectively (66% of the regions annual total for harbor porpoise)
During the fall season 67% of harbor porpoise takes from the NESG fishery
occurred in the Mid-Coast closure area (13% of the region’s annual
total for harbor porpoise). The only observed take of an Atlantic
white-sided dolphin occurred in the fall in the Mid-Coast closure area.
In the MACG fishery the total estimated takes for harbor porpoise in
April off the coast of Virginia was 135 animals (Table 10).
The 2004 estimated total takes of seals in the NESG and MACG fisheries
was 1,726 (CV = 20%). For the NESG fishery, this included 498 gray
seals (CV = 35%; Table 6); 786 harbor seals (CV = 34%; Table 7); 303
harp seals (CV = 30%; Table 8); and 43 hooded seals (CV = 95%; Table
9). For the MACG fishery, the species included in this estimate
are: 68 gray seals (CV = 92%) and 28 harbor seals (CV = 77%; Table 10).
The majority of the NESG gray seal takes (89%) occurred in the winter
season, with the remaining takes occurring in the summer (6%) and fall
(5%). The South of Cape Cod port group-area had 45% of the estimated
takes in winter, the South Cape time/area closure had 31% of the takes
in winter, and the remaining 24% of the estimated takes took place in
the remaining port group-areas in other seasons (Table 6).
For harbor seals in the NESG fishery, 43% of the estimated takes occurred
in the summer, 37% occurred in the winter, and the remaining 20% occurred
in the fall. The South of Cape Cod port group-area in the winter
had 28% of the annual estimated takes, the New Hampshire port group-area
in the summer had 28% of the annual estimated takes, the Mid-Coast time/area
closure in the fall had 18% of the annual estimated takes, and the remaining
26% of estimated takes occurred in the remaining time/area closures and
port group-areas in all seasons (Table 7).
All of the estimated harp seal takes in the NESG fishery took place
in the winter season. North of Boston (39%) and South of Cape Cod
(38%) port group-areas had the majority of the estimated harp seal takes. The
remaining 20% of the takes occurred in the Massachusetts Bay time/area
closure and 3% in the East Cape port group-area (Table 8).
All of the estimated hooded seal takes in the NESG fishery occurred
in the winter season in the South Cape time/area closure (Table 9).
There were two pinniped species caught in the MACG fishery, where an
estimated 68 gray seals were caught in April off Virginia and an estimated
28 harbor seals were caught in December off New Jersey (Table 10).
DISCUSSION
The current stratification method (ratio estimator) being used
to estimate marine mammal bycatch in U.S. gillnet fisheries
has been in use for some time. The current methodology was
established as an acceptable method to estimate harbor porpoise
takes in the NESG fishery (Bravington and Bisack, 1996). However,
since this method was developed, spatial patterns of the NESG fishery
and sampling of the fishery by the NEFOP has changed, and self
reported fishery dependent data collection (VTR) has improved. In
addition, there are potential problems with this approach when
applied to other species. Each issue is discussed below.
For all marine mammal take estimates, one of the potential problems
with the current methodology involves a bias associated with using
observed hauls from complete fish sampling trips as well as marine
mammal sampling trips to estimate takes. Observers on a complete
fish sampling trip are engaged in sampling the catch for fish discards
during the haul back and are not watching the net as it is being
hauled in. Consequently, they may not see marine mammals
that fall out of the net before it reaches the deck (NEFOP Observer
Manual, http://www.nefsc.noaa.gov/femad/fishsamp/fsb/). In
contrast, observed trips dedicated to sampling for marine mammal
interactions have observers watching the gear at all times. See
Table 11 for the complete breakdown of trips, hauls, and incidental
takes by trip types.
In 2002 the NEFOP was charged with increasing the sampling of
fish discards for the NESG fishery. As a result the number
of complete fish sampling trips in the NESG fishery increased from
~10% in 2002 to ~60% and ~91% in 2003 and 2004, respectively. Based
on this change, it was decided that both complete fish sampling
and marine mammal sampling trips would need to be used to compute
the marine mammal bycatch rates for the NESG fishery, starting
with the take estimates calculated from the 2003 data, which were
reported in the 2005 SAR. These take estimates have not been
corrected to account for the potential bias associated with trips
dedicated to fish sampling (Table 11). We anticipate updating
the work of Bravington and Bisack (1996) utilizing new data from
recent years to develop revised correction factors. Prior
to 2003 only limited fish sampling trips were used to estimate
marine mammal bycatch in the NESG fishery. The increase in
trips sampled for fish discards did not impact the MACG fishery
as bycatch rates from the MACG fishery were only derived from limited
sampling trips.
For both cetacean and pinniped bycatch estimates, the following
methodology issues require further attention: (1) the stratification
and proration scheme used for estimating takes in NE fisheries
may need to be revised to account for improved VTR data reporting
and the spatio-temporal patterns of animals other than harbor porpoise;
(2) the observer data are likely over-stratified, which means that
the data may not reflect the changes that have occurred in fishing
patterns which also influences the accuracy and precision of estimated
bycatch rates; and (3) the port-based stratification scheme may
no longer reflect current fishing patterns. The current stratification
method was specifically designed to estimate takes of harbor porpoise
in the NESG fishery. Therefore, this method may not be appropriate
for other cetacean or pinniped species due to their different spatial/temporal
patterns within the habitat of the Northeast region. In addition,
the effectiveness of pingers was tested by experimental design
and proven to successfully deter harbor porpoise away from gillnets
(Kraus et al. 1997). The weighting of bycatch rates for other
species incidentally bycaught in gillnets implies that pingers
are also effective in deterring these other species. However,
reports on the effectiveness of pingers to deter or attract other
small cetaceans and pinnipeds are varied and some were inconclusive
(Dawson 1998, IWC 1999, Jefferson and Curry 1996). Therefore,
the use of weighted bycatch rates for species other than harbor
porpoise bycaught in nets with and/or without pingers should be
considered on a case-by-case basis, given the information that
is available in the literature for each individual species.
In conclusion, we are in the process of evaluating our take estimation
methodology to determine whether the current method can be improved
given the changes in the data. For cetaceans, we are looking
to combine the NESG and MACG data into a predictive model that
will improve stratification as well as the accuracy and precision
of our take estimates. For pinnipeds, we hope to design a
new stratification method that utilizes pinniped spatio/temporal
patterns in relation to the fishery.
REFERENCES
Bisack KD. 1993. Estimates of total US harbor porpoise bycatch
in the Gulf of Maine sink gillnet fishery. Northeast Fish. Sci.
Cent. Ref. Doc. 93-11; 23 p.
Bisack KD. 1997. Harbor porpoise bycatch estimates in the
New England multispecies sink gillnet fishery: 1994 and 1995. Rep. int. Whal. Commn.
47: 705-714.
Bisack KD. 2003. Estimates of marine mammal bycatch in the
Northeast (New England) multispecies sink gillnet fishery in 1996. Northeast
Fish. Sci. Cent. Ref. Doc. 03-18; 18 p.
Blaylock RA, Hain JW, Hansen LJ, Palka DL, Waring GT. 1995. US
Atlantic and Gulf of Mexico Marine Mammal Stock Assessments. NOAA
Tech. Mem. NMFS-SEFSC-363; 211 p.
Bravington MV, Bisack KD. 1996. Estimates of harbor porpoise
bycatch in the Gulf of Maine sink gillnet fishery, 1990-1993. Rep. int. Whal. Commn. 46:
567-574.
Dawson SM, Read A, Slooten E. 1998. Pingers, porpoises and
power: uncertainties with using pingers to reduce bycatch of small cetaceans. Biol.
Conservation 84:141-146.
International Whaling Commission (IWC). 1999. Report of
the Scientific Committee. IWC/51/4, p. 62-65.
Jefferson TA, Curry BE. 1996. Acoustic methods of reducing
or eliminating marine mammal-fishery interactions: do they work? Ocean & Coastal
Management 31(1): 41-70.
Kraus SD, Read A, Anderson E, Baldwin K, Solow A, Spradlin T, Williamson
J. 1997. Acoustic alarms reduce porpoise mortality. Nature 388:525.
National Marine Fisheries Service [NMFS]. 1998. Taking of
marine mammals incidental to commercial fishing operations; harbor porpoise
Take Reduction Plan regulations. Federal Register 63(231):66464-66490,
December 2, 1998.
Palka DL, Rossman MC. 2001. Bycatch estimates of coastal
bottlenose dolphin (Tursiops truncatus) in US mid-Atlantic gillnet
fisheries for 1996 to 2000. Northeast Fish. Sci. Cent. Ref.
Doc. 01-15; 77 p.
Rossman MC, Merrick RL. 1999. Harbor porpoise bycatch in
the Northeast multispecies sink gillnet fishery and the mid-Atlantic
coastal gillnet fishery in 1998 and during January-May 1999. Northeast
Fish. Sci. Cent. Ref. Doc. 99-17; 36 p.
Smith T, Palka D, Bisack K. 1993. Biological significance
of bycatch of harbor porpoise in the Gulf of Maine demersal gillnet fishery. Northeast
Fish. Sci. Cent. Ref. Doc. 93-23; 15 p.
Waring GT, Palka, DL, Mullin K, Hain J, Hansen L, Bisack K. 1997. US
Atlantic and Gulf of Mexico marine mammal stock assessments – 1996. NOAA
Tech. Mem. NMFS-NE-114; 250 p.
Waring GT, Pace R, Quintal J, Fairfield C, Maze-Foley K. 2003. US
Atlantic and Gulf of Mexico marine mammal stock assessments – 2003. NOAA
Tech. Memo. NMFS-NE-182; 287 p.
