CONTENTS Introduction Data Sources and Methods Results and Discussion References Footnotes
Northeast Fisheries Science Center Reference Document 03-18
Kathryn D. Bisack
Estimates of marine mammal bycatch in the Northeast (New England) multispecies sink gillnet fishery in 1996
National Marine Fisheries Serv., Woods Hole Lab., 166 Water St., Woods Hole, MA 02543
Web version posted October 6, 2003Citation: Bisack, K.D. 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; 21 p.
Information Quality Act Compliance: In accordance with section 515 of Public Law 106-554, the Northeast Fisheries Science Center completed both technical and policy reviews for this report. These predissemination reviews are on file at the NEFSC Editorial Office.
This report provides bycatch estimates of six marine mammal species incidentally taken in the 1996 Northeast (New England) multispecies sink gillnet fishery. The estimated total take of all marine mammals in the 1996 fishery was 2,364 animals (CV = 17%). This included 1,185 harbor porpoise (CV = 25%), 855 harbor seals (CV = 27%), 116 white-sided dolphins (CV = 114%), 69 common dolphins (CV = 139%), 49 gray seals (CV = 56%), and 90 harp seals (CV = 55%).
During 1996, incidental takes of six marine mammal species were documented by fishery observers in the Northeast (New England) sink gillnet fishery. The purpose of this paper is to present estimates of the total bycatch of these six species in the 1996 Northeast sink gillnet fishery as required by the US Marine Mammal Protection Act (MMPA) of 1972 and its amendments of 1994.
The US sink gillnet fishery extends from Maine to North Carolina. This paper deals with the component of the fishery in New England waters north of 40°N. Sink gillnet gear fished in these waters consists of nets with 6-10 inch monofilament stretch mesh suspended between a buoyed head rope and a weighted ground line. A sink gillnet vessel generally deploys four to seven strings per trip, on average. One string typically consists of five to twelve nets strung together, in which the standard net length averages three hundred feet and height averages eleven feet. The gear normally soaks in the water for 24 to 72 hours, is hauled, the catch is removed and then reset. Target species include pollock and cod, flatfish, monkfish and dogfish. Marine mammals may become entangled in the gear and suffocate and die.
Harbor porpoise (Phocoena phocoena), white-sided dolphins (Lagenorhynchus acutus), common dolphins (Delphinus delphis), harbor seals (Phoca vitulina), gray seals (Halichoerus grypus), and harp seals (Phoca groenlandica) were incidentally taken in the Northeast sink gillnet fishery in 1996. Of these species, two are listed as strategic stocks (harbor porpoise and common dolpin), four are listed as nonstrategic (white-sided dolphins, harbor seals, gray seals, and harp seals) (Table 1) (Waring et al., 2002). A species is listed as 'strategic' if the total fisheries bycatch is greater than the Potential Biological Removal Rate (PBR).
Estimates of annual marine mammal bycatch in the Northeast sink gillnet fishery are available since 1990 (Bisack, 1993; Smith et. al. 1993; Bravington and Bisack, 1996; Bisack, 1997). Several methods have been used to derive bycatch estimates. The method presented in this paper is a modification of previous approaches and was developed because of changes implemented in the data collection system.
In June 1994, the National Marine Fisheries Service/Northeast Region's commercial fisheries data collection system, in which catch, effort and fishing location data were collected from dealer reports and voluntary dockside interviews of vessel captain or crew, was replaced by one in which fishing trip information was recorded in mandatory logbooks (Wigley et al., 1998). The logbook data were not available at the time the 1994 and 1995 marine mammal bycatch estimates were made (Bisack, 1997). The marine mammal bycatch estimates presented here for 1996 utilize the logbook data and account for area/season closures in the analyses.
Data Sources and Methods
Three databases were used in estimating the marine mammal bycatches in 1996: (1) the NEFSC Sea Sampling (SS) program [Fishery Observer Program] data bases was used to estimate the bycatch of marine mammals per observed ton of fish caught; (2) the Northeast Dealer Report Weighout (WO) data base was used to determine the total landings in 1996 of all finfish caught in the Northeast sink gillnet fishery; and (3) the Northeast Vessel Trip (VTR) data base was used to allocate (prorate) the sink gillnet landings in the WO data base into spatially/temporally defined strata (season/port group or area closure) (Rossman and Merrick, 1999)(1)
To ensure similar fishing practices are grouped together to estimate the bycatch rates, spatial and temporal strata were delimited. The spatial stratification includes seven port groups (Bisack, 1997), an offshore area (SA 515, 464 and 522), and closed area strata. The seven port groups are northern Maine, southern Maine, New Hampshire, north of Boston, south of Boston, east of Cape Cod, and south of Cape Cod. The temporal stratification used was: winter (January - May), summer (June - August), and fall (September - December).
Five time/area closures (Figure 1) were added to the stratification scheme in 1996: 1) Downeast Maine (August 13 to September 15; 2) Mid-Coast (March 26 to April 25 and September 15 to December 31); 3) Mass Bay (March 1 - March 31); and 4) South Cape Cod (March 1 - March 31). Data collected during these closures were removed from the port/season strata (defined above) and placed in individual time/area closure strata (see Table 2). For example, sink gillnet trips in the March 26th to April 25th Mid-Coast closure were removed from three port strata (Southern Maine, New Hampshire, North of Boston) and placed in the Mid-Coast winter closure strata. A stratum was not created for the Downeast Closure as no sink gillnet landings were reported from this closure area in 1996.
The bycatch rate in each stratum was calculated as the number of observed takes of marine mammals on sea sampling trips in the Northeast sink gillnet fishery divided by the total weight of fish caught during these trips. The bycatch rate is expressed as the number of marine mammals caught per ton of fish landed.(2) Only dedicated marine mammal trips were used. (3) These trips were divided into two groups: trips with active pingers attached to the gear, and trips that did not use pingers. Bycatch rates were estimated for both types of trips. Vessels allowed to fish in any of the closure areas were required to use active pingers.
Observer coverage was defined as the ratio (in weight) of fish caught on marine mammal sampling trips in a time/area stratum to the total fish landings in that time/area stratum. The ratio therefore represents the fraction of total landings caught during observed trips.
Marine mammals incidentally taken during sampling trips were identified to species by the onboard fishery observers. However, nine (9) seals were captured whose species identification was recorded as unknown. For the bycatch estimates, these unknown seals were assigned to one of the identified seal species (harbor, gray, and harp) based on the species composition of the seals taken in the strata in which the unknown animals were captured. For example in the winter Mid-Coast closure stratum, 14 seals were taken: 5 unknown seals, 8 harbor seals, and 1 gray seal. Of the 9 known seals taken, 89% (8/9=0.89) were harbor seals and 11% (1/9) were gray seals. The 5 unknown seals were therefore assigned to species according to these percentages (i.e., 4.4 harbor seals and 0.6 gray seals).
Bycatch Estimate Adjustment
The winter (January to May) 1996 offshore bycatch estimate was adjusted because observer coverage was extremely low in this time/area stratum. In February, only 2 trips (20 hauls) were observed with 8 harbor porpoise takes. During March through May combined, another 20 sink gillnet hauls were observed with no harbor porpoise takes (there were no trips in April). According to the WO data base, total offshore sink gillnet fishery landings increased from 145 tons in February to 586 tons in May. Because of the low observer coverage, the bycatch estimate would be extremely biased if the February to May bycatch rate were multiplied by the total landings during these months. Therefore, the winter offshore bycatch estimate was based only on the observer data (and fishery landings) in February.
CV and Confidence Intervals
Standard bootstrapping techniques were used to derive confidence intervals and coefficients of variation (CV) for all of the strata bycatch estimates. The resampling unit was an entire trip (rather than an individual haul) as this ensured that any within trip dependence in the original data was carried over into the bycatch estimates.
The bootstrap results were adjusted in two strata where observer coverage was greater than 10% and where marine mammal takes occurred (Mid-Coast and Mass Bay strata) (Table 2 and Table 3). These adjustments applied the finite correction factors so that the variance of the bycatch estimates would be accurately calculated.
Results and Discussion
Northeast Sink Gillnet Fishery Landings
In 1996, 24,277 tons of fish were landed in the Northeast sink gillnet fishery (Table 2). These were prorated to season/area strata based on the time/area distribution of sink gillnet landings in the 1996 VTR logbook database (Appendix A, Tables A1 and A2). Summer landings were greater than those in the winter and fall seasons in almost all strata.
Observer coverage was 4.1% in the winter, 1.8% in summer, and 3.3% in the fall (Table 2). A total of 96 takes of marine mammals were observed. These included 44 harbor porpoise, 31 harbor seals, 2 white-sided dolphins, 1 common dolphin, 3 gray seals, 6 harp seals and 9 unidentified seals (Table 3).
Observed harbor porpoise takes in 1996 occurred in some new areas and seasons compared to previous years. The new time and areas and port group were: (1) Winter/North Boston port stratum; (2) Fall/New Hampshire port stratum; and (3) Observed Fall/South of Cape Cod port stratum.4
Marine Mammal Incidental Take Estimates
The estimated total incidental take of all marine mammals in the Northeast sink gillnet fishery was 2,364 animals (CV = 17%). This included 1,185 harbor porpoise (CV = 25%) , 855 harbor seals (CV = 27%), 116 white-sided dolphins (CV = 61%), 69 common dolphins (CV = 139%), 49 gray seals (CV = 56%), and 90 harp seals (CV = 55%). Bycatch rates and total estimated bycatch by species and strata are presented in Table 4 and Table 5, respectively. Species-specific bycatch estimates for 1996 are summarized in Table 6.
Fifty five percent of the estimated takes in 1996 occurred in winter (1,298, CV=21%) , twelve percent in summer (293, CV=48%), and thirty three percent during fall (773, CV=32%) (Table 5). The south of Cape Cod and North Boston port strata accounted for the largest percentage of the estimated takes in 1996, 32% (=764/2364) and 16% (=372/2364), respectively.
Closures - Pingers
During the winter season, bycatch rates of harbor porpoise and harbor seals were highest in marine mammal closure areas where pingers were required (e.g., Mid-Coast closure area) than in the areas where pingers were not required (Table 4). This is based on the assumption that all the gear fishing during the closure in the Mid-Coast area had active pingers, and outside of the closure there were no pingers being used or they were not active.
Bootstrap Replicate Analysis
The distribution of the bootstrap replicates differed by species (Table 7; Figure 2: Frame 2a to 2f). The null hypothesis that the bootstrap replicates followed a normal distribution was rejected for all six species. The bootstrap replicates for harbor porpoise, harbor seals, and harp seals followed a lognormal distribution (p>0.15).(4)
Summary of marine mammal bycatch estimates, 1993 to 1996
Annual estimates of marine mammal bycatch in the Northeast sink gillnet fishery during 1993-1996 range from a low of 2,321 animals in 1993 to a high of 4,550 animals in 1994 (Table 8). Changes in the annual estimates are due to inter-annual variability in bycatch rates and changes in annual landings (Table 9).
Harbor porpoise, harbor seals, white-sided dolphins, and gray seals have been consistently taken in the sink gillnet fishery over the years. Harp seal takes first occurred in 1995, while hooded seal and common dolphin takes first occurred in 1995 and 1996, respectively.1995 by takes of. The increase in the number of species of seals taken in the sink gillnet fishery is probably due to improved species identification methods (use of photographs, tooth and skull examination protocols, etc) implemented in 1994.
Bisack, K., and G. Dinardo. 1992. Estimating total effort in the Gulf of Maine demersal gillnet fishery. Appendix, Report of the Fall Stock Assessment Workshop (13th SAW), Research document SAW 13/3, Northeast Fisheries Science Center Reference Doc. 92-02.
Bisack, K. 1993. Estimates of total U.S. harbor porpoise bycatch in the Gulf of Maine sink gillnet fishery. Northeast Fisheries Science Center Reference Doc. 93-11. 23p.
Bisack, K.D. 1997 Harbor porpoise bycatch estimates in the New England multispecies sink gillnet fishery: 1994 and 1995. Rep. int. Whal. Commn 47: 705-714.
Bravington, M.V., and K.D. Bisack. 1996. Estimates of harbor porpoise bycatch in the Gulf of Maine sink gillnet fishery, 1990-1993. Rep. int. Whal. Commn 46: 567-574.
Johnson D.R., C. Yeung, and C.A. Brown. 1999. Estimates of marine mammal and marine turtle bycatch by the U.S. Atlantic pelagic longline fleet in 1992-1997. NOAA Tech. Memo. NMFS-SEFSC-418, 70 pp.
Rossman, M.C. and R.L. Merrick. 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 Fisheries Science Center Reference Doc. 99-17, 36p.
Smith, T., D. Palka, and K. Bisack. 1993. Biological significance of bycatch of harbor porpoise in the Gulf of Maine demersal gillnet fishery. NOAA, NMFS, Northeast Fisheries Science Center Reference Doc. 93-23, 15p.
Wade, P. 1998. Calculating limits to the allowable human-caused mortality of cetaceans and pinnipeds. Marine Mammal Science, 14(1):1-37 (Jaunuary 1998)
Waring, G.T., D.L. Palka, P.J. Clapham, S. Swartz, S., R.C. Rossman, M.C., T.V. Cole, K.D. Bisack, and L.J. Hansen. 1999. US. Atlantic Marine Mammal Stock Assessments -1998. NOAA Tech. Memo. NMFS-NE-116, 196p.
Waring, G.T., J.M. Quintal, and C. Fairfield (Eds.). 2002. US Atlantic and Gulf of Mexico marine mammal stock assessments -2002. NOAA Tech. Memo. NMFS-NE-169, 180p.
Wigley, S., M. Terceiro, A. Delong, and K. Sosebee.1998. Proration of 1994-96 USA Commercial Landings of Atlantic Cod, Haddock, and Yellowtail Flounder to Unit Stock Areas. Northeast Fisheries Science Center Reference Doc. 98-02, 32p.
1. The Rossman and Merrick (1999) paper is an alternative source for the proration algorithm. However, their proration method was based on the method presented in this paper. Details are in the appendix.
2. A ton is equivalent to 2000 pounds.
3. On dedicated marine mammal trips the observer watches the gear being hauled back. On non-dedicated marine mammal trips, the observer collects other biological data and relies on the captain and crew to notify him/her if a marine mammal has been caught in the gear.
4. Some studies have assumed the distribution of the total bycatch to be normal or lognormal (Wade 1998; Johnson et. al. 1999).