2010 SOUTHEASTERN NATURALIST 9(1):47–62 Challenges of Documenting Tursiops truncatus Montagu (Bottlenose Dolphin) Bycatch in the Stop Net Fishery along Bogue Banks, North Carolina Barbie L. Byrd1,* and Aleta A. Hohn1 Abstract - Each fishery presents its own challenges for observers to document bycatch. The North Carolina (NC) stop net fishery is especially challenging because it uses anchored gear (the stop net) that soaks up to 15 days to herd fish, which are then hauled to shore via another gear (a beach seine). Three Tursiops truncatus (Bottlenose Dolphin) entanglements in stop nets and six Bottlenose Dolphin strandings, each suspected of having been entangled in stop net gear based on injuries noted (lesions) and spatio-temporal overlap with the fishery, were documented by the NC Marine Mammal Network between 1992 and 2007. In 2001–2002, new observational techniques and surveys were used to observe this fishery to estimate bycatch and to document dolphin behavior around the gear. Techniques included observations from the beach during net retrievals and in situ surveys using a vessel with a sonar-video camera system. No entangled dolphins were observed, and, in fact, observations indicated dolphins were not attracted to stop nets and generally changed direction to avoid the gear. Nonetheless, characteristics of the fishery impose severe limitations on the efficacy of bycatch observer methods, rendering those results unreliable. Given low levels of known or suspected entanglements and the challenges of observing this fishery, stranding network data may be the most practical and effective method to monitor dolphin bycatch. Introduction Observer programs document bycatch of a variety of species in commercial fisheries to estimate bycatch mortality incidental to those fisheries (Edwards and Perrin 1993, Epperly et al. 1995, Julian and Beeson 1998, NMFS 2009). Bycatch is defined as fishery discards, retained incidental catch, and unobserved mortalities resulting from a direct encounter with fishing gear (NMFS 1998). Traditionally, observers are placed on board fishing vessels, but not all fisheries are conducive to having observers on board because of the size of the boat and the way the gear is fished. Observing fisheries via an alternative platform, where observers use a separate vessel to observe fishing vessels on the water (Kolkmeyer et al. 2007) or observe beach-based fishing operations from the beach, has proven successful when observers cannot be accommodated on board. In the case of traditional and alternative platforms, observers document the catch and bycatch during the retrieval of gear. Observations made only during retrievals are insufficient for fisheries with anchored gear that soaks, i.e., remains in the water, for long 1National Marine Fisheries Service, Southeast Fisheries Science Center, NOAA Beaufort Laboratory, 101 Pivers Island Road, Beaufort, NC 28516. *Corresponding author - Barbie.Byrd@noaa.gov. 48 Southeastern Naturalist Vol. 9, No. 1 periods of time (e.g., pound nets [Epperly et al. 2007]), because bycaught animals may not remain entangled until the gear is retrieved. For these gear types, observations also must be made in situ, i.e., while the net is soaking. The North Carolina (NC) stop net fishery is one such fishery. The NC stop net fishery is regulated by the NC Division of Marine Fisheries (NCDMF) and is restricted to the ocean side of one barrier island, Bogue Banks, in Carteret County, NC (Steve et al. 2001) (Fig. 1). The stop net fishery occurs annually during October and November to harvest Mugil cephalus L. (Striped Mullet) for their roe (Steve et al. 2001). Two crews participate in this fishery, each allowed to set two nets of specified mesh size and dimensions within the six designated sites. Unlike most fisheries, the stop net fishery incorporates two gear types (a stop net and beach seine) that are used concurrently with differing material and fishing techniques. The stop net is an anchored, multifilament gear set roughly in a “L” shape from the beach to herd schools of Striped Mullet moving westward along the beach (Fig. 2). The 365.8-m length net is comprised of an inshore section (suds and backstaff; 182.9 m, 20.3-cm stretched mesh) set roughly Figure 1. Boundaries of the six sites along Bogue Banks, NC, designated for the stop net fishery. The sites from west to east are Bogue Inlet (BI), Emerald Isle (EI), Salter Path (SP), Pine Knoll Shores (PKS), Atlantic Beach (AB), and Fort Macon (FM). Bottlenose Dolphins entangled in stop nets or stranded on or near Bogue Banks during the stop net seasons of 1992–2007 are shown as plus symbols: gray for entangled dolphins (n = 3), white for stranded dolphins with entanglement lesions consistent with multifilament gear similar to stop nets (n = 6), and black for stranded dolphins with entanglement lesions consistent with monofilament gear (not used in stop nets; n = 2). 2010 B.L. Byrd and A.A. Hohn 49 perpendicular to the beach and an offshore section (lead; 182.9 m, 15.2-cm stretched mesh) set roughly parallel to the beach. The beach seine is then used to encircle the herded fish and pull them to shore, a method commonly known as a beach strike (Fig. 2). The seine used for the beach strikes is constructed of multifilament and monofilament panels with a stretched mesh size of 8.3 to 9.5 cm. Depending on weather (e.g., north-northeast wind and low wave height), the stop net may soak unattended from one to fifteen days. In contrast, beach strikes occur only when fish are corralled in the stop net, and the seines are actively set, remaining in the water only briefl y (generally less than 45 minutes) while hauled in. The stop net fishery is one of six NC fisheries in which coastal Tursiops truncatus Montagu (Bottlenose Dolphin; the coastal morphotype [Hoelzel et al. 1998, Mead and Potter 1995]) bycatch is known to occur (Waring et al. 2002). The other fisheries are the mid-Atlantic coastal (i.e., ocean) gillnet, NC inshore gillnet, mid-Atlantic haul/beach seine, NC long haul seine, and mid-Atlantic blue crab trap/pot fisheries (Federal Register 2006). Because of high overall bycatch level in commercial fisheries and the “depleted” designation status of the stock, the National Marine Fisheries Service (NMFS), in accordance with Section 118(f) of the MMPA, convened the Bottlenose Dolphin Take Reduction Team (BDTRT) in November 2001 to prepare a plan to reduce bycatch (Federal Register 2006). Bycatch mortality estimates, however, are only available for the coastal gillnet fishery (Waring et al. Figure 2. The study area in 2002 at the Fort Macon stop net was bounded by a pier to the west of the net and a jetty to the east, and extended approximately 950 m from shore. The observer was positioned at the highest point of the beach, without entering the dunes, and in line with where the stop net was attached to shore. The 365.8-m anchored, stationary stop net was comprised of the suds, backstaff, and lead sections to impede the movement of fish schools migrating along shore. A beach seine was used to capture fish corralled by the stop net and then hauled ashore using tractors. 50 Southeastern Naturalist Vol. 9, No. 1 2009). Anecdotal information indicates that bycatch occurs in the other five fisheries, and estimates of the level of that bycatch are needed. Since 1994, researchers have tried various approaches to observe the stop net fishery to document the occurrence and frequency of dolphin entanglements. Researchers’ efforts primarily monitored beach strikes and stop net retrievals (i.e., haulbacks) to determine if dolphins were entangled in the gear (Asher 2001, Padgett 1995). Although not necessarily indicative of an entanglement or lack thereof, researchers also scanned the fl oat line of soaking stop nets to look for any obvious signs, such as submerged fl oats, that a large animal was entangled. Asher (2001) investigated in situ methods of observing the stop net fishery using sonar, snorkelers, and scuba divers. Sonar proved inadequate because it did not always detect fish in the nets even when they were visually detected, and the lack of available, qualified personnel and poor weather conditions prevented the use of snorkelers and divers from being a viable method to observe this fishery (Asher 2001). Dolphins are susceptible to entanglement in the stop net fishery because of its nearshore fishing location and gear use practices (i.e., large-mesh webbing with long soak times). The fishery is difficult to observe because it incorporates two gears and because one of those gears, the stop net, soaks for extended and unpredictable periods. The purpose of this paper is to report stranding data for known or suspected entanglements of dolphins in the stop net fishery, describe dolphin behavior around stop net gear, and explore alternative observer methods to document mortality in the fishery. Methods Field-site description Beach and vessel surveys in the present study occurred along Bogue Banks, NC at six sites designated by the NCDMF as locations where stop nets can be placed: Bogue Inlet (BI), Emerald Isle (EI), Salter Path (SP), Pine Knoll Shores (PKS), Atlantic Beach (AB), and Fort Macon (FM) (Fig. 1). Stranding data The NC Marine Mammal Stranding Database maintained at the NMFS, National Oceanic and Atmospheric Administration (NOAA) Beaufort Laboratory was queried to determine the number of Bottlenose Dolphin strandings in Carteret County, NC during October and November, 1992 through 2007. From that list, a subset of records was reviewed consisting of all strandings on or near Bogue Banks categorized as positive for fishery interaction (fi), i.e., showed lesions consistent with fisheries entanglement or had gear attached (Byrd et al. 2008, Kuiken et al. 1994, Read and Murray 2000). For each dolphin in the subset of strandings, lesions were categorized as associated with monofilament or multifilament twine (Geraci and Lounsbury 2005) based on descriptions and photographs. Observations of stop nets This study had two objectives for the stop net observations: 1) to document dolphin behavior around stop nets, and 2) to document dolphin 2010 B.L. Byrd and A.A. Hohn 51 entanglements in stop net gear. The stop nets, not the beach seines, were the focus of investigations because of their long, unattended soak times. Although landings have been used as the measure of fishing effort for Bottlenose Dolphin bycatch estimates in gill nets (Palka and Rossman 2001, Waring et al. 2009), landings in the stop net fishery occur in the beach seine. Total soak time for each stop net, therefore, was deemed a more appropriate measure of fishing effort. To obtain estimates of soak time during both years, we requested that the leaders of the two fishing crews notify us soon after a stop net had been deployed and before it would be removed. Knowing beforehand that a stop net was going to be removed also would allow us to be present to observe the haulbacks for entangled dolphins and accurately record the end of the soaking period for that net. Daily phone calls were made to crew leaders to facilitate this communication. 2001 season. Beach observations of deployed stop nets occurred two to three times per week by one of two observers throughout the fishery season, depending on the presence of stop nets and excluding the last week when scheduling confl icts and short soak times prevented observations. Observers started in the early morning at the crews’ local gathering place and followed them to a specific stop net if a haulback was anticipated. If the crews were not at the gathering location, observers started at the westernmost stop net location and moved eastward to each net location, observing each deployed net. At each stop net, observers scanned the area around the net with the naked eye, looking for evidence of dolphin entanglement (i.e., submerged fl oats) and recording presence and behavior of dolphins around the nets. Fujinon 7 x 50 reticle binoculars (Fujinon Inc., Trenton, GA) were used if needed. Multiple dolphins observed in the survey area at the same time were defined as a single sighting; however, a sighting sometimes consisted of two or more subgroups of dolphins moving independently of each other. Dolphins were considered part of the same subgroup if they were within 10 m of any other dolphin within the group (a 10-m “chain” rule) (Smolker et al. 1992). If one sighting was comprised of separate subgroups, each subgroup was tracked separately. For each subgroup within a sighting, observers recorded time, estimated distance from the net, direction of travel, and whether any individuals entered the perimeter of the stop net. If dolphins were not present, observers monitored the net for 30 minutes before moving to the next net. 2002 season. Building on the successes and challenges of the previous year, observation methods were altered in 2002. Observation effort increased, and new methods were used to more accurately document dolphin behavior. Observations occurred only at the Fort Macon net because: 1) its location provided the best remote viewing of an active stop net, 2) it had the closest proximity to a beach access point, and 3) it was more consistently deployed at the beginning of the season. The goal was to conduct observations three to four times per week, weather permitting. During each observation period, a single observer was at the site continuously from two hours before 52 Southeastern Naturalist Vol. 9, No. 1 to two hours after high or low tide, depending on which tidal cycle occurred during daylight hours, limiting any effects varying tidal levels may have had on Bottlenose Dolphin behavior. To investigate whether or not the net affected dolphin behavior in the area, observations occurred when the stop net was soaking (present) and when it had been removed (absent). Conveniently, the Fort Macon site was fl anked by two barriers, a jetty to the east and a pier to the west, creating a sighting area approximately 1349 m wide (Fig. 2). The sighting area extended from shore approximately 950 m, and was delineated by the length of the pier. The observer was located at the highest point of the beach, without entering the dunes, and in line with the suds-backstaff section attached to shore, approximately 649 m from the jetty and 700 m from the pier. The observer scanned for dolphins continuously within the survey area, using the naked eye. Fujinon 7 x 50 reticle binoculars were used if needed. Sightings and subgroups were defined the same as in 2001. For each subgroup within a sighting, the observer recorded time of day, number of individuals, surface activity, direction of travel, distance from the net, and distance from shore at one-minute intervals. Two-sample t-tests were used to determine differences between the number of subgroups observed per hour and number of dolphins per hour for each observation period when the stop net was present versus absent and when the tide was high versus low. Two-sample t-tests were also used to determine differences in subgroup size and the time (in hours) spent in the sighting area for each subgroup observed when the stop net was present versus absent and when the tide was high versus low. The observer estimated distances of dolphins from shore or from boundaries of the survey area. To calculate error in estimated distances, estimates were compared to actual distances measured with a Bushnell Yardage Pro 800 rangefinder (Bushnell Corporation, Overland Park, KS) at the beginning and the middle of the season. Estimated distances were on average within 50 m of the actual distance, with the most accurate measurements being within 100 m from shore. Due to estimate error, tests were not performed to determine differences between when the net was present versus absent for minimum distance to the net and minimum distance from shore. Estimated distances relative to the net and shore were used only to track dolphin movements within the study area to examine patterns of movements. In addition to beach observations, a 5.5-m NOAA vessel served as a platform for a sonar-video camera system used to scan for entangled dolphins in all deployed stop nets (i.e., in situ). The system consisted of a sonar transducer and video camera mounted sideways onto an extendable fiberglass pole (2.2–3.8 m) that could be lowered over the side of a boat. The forwardfacing sonar (Interphase Probe, Interphase Technologies, Soquel, CA) transmitted at a frequency of 200 kHz at 400 watts. The sonar was connected to a 12.7-cm monochrome screen. The video camera (Seaviewer Sea-Drop 2010 B.L. Byrd and A.A. Hohn 53 6500, Seaviewer, Tampa, FL) was designed for low visibility and was hardwired to a monitor on the boat. Field tests with the sonar on various objects (e.g., divers to mimic approximate size of a dolphin, buoys, nets) occurred at the NOAA Laboratories in Pascagoula, MS and Beaufort, NC to determine its efficacy and optimal settings for detecting fish and other objects. A target of at least 4 x 6 pixels on the sonar display was considered large enough to warrant visual inspection. Boat surveys occurred when at least one stop net was in the water, and the marine weather consisted of winds less than 18.5 km/hr (10 knots) and wave height less than 0.9 m. The boat idled parallel to and outside of the lead sections of the nets from a distance of 9 to 12 m, and the transducer was extended to a depth of 70–75 cm to eliminate surface or boat-wake interference. This configuration allowed the sonar to scan from 70–75 cm below the water’s surface to the bottom of the net while the top 1 m of the net was visually observed. If strong targets were detected by the sonar, the boat immediately repeated the pass to determine if the target was a false return or a return of fish swimming by the transducer. If the target was detected during the second pass, the video camera was moved to 0.25–1.25 m from the net, depending on water clarity, to identify the target. The leads were surveyed three times each observation day to ensure that no part of the net was missed. The suds-backstaff section could not be surveyed consistently with the sonar-video system when high surf made it difficult for the vessel to remain a safe distance outside the net. Results Stranding data From 1992 through 2007, the stranding network in NC recorded three dolphins entangled in stop net gear and eight dolphins stranded on or near Bogue Banks with lesions consistent with fisheries entanglement (Fig. 1). The first known entanglement occurred in November 1993 when two dolphins became entangled simultaneously in the inshore section. One was released alive by stranding network personnel, and the other died in the net. In November 1999, another dolphin became entangled and drowned in the inshore section of a stop net. Of the eight stranded dolphins with entanglement lesions, two had lesions indicative of monofilament gear. The other six had lesions indicative of multifilament gear similar to that used in stop nets; three of them were stranded in 1994, and one stranded in each of the years 1997, 1998, and 2007. Observations of stop nets 2001 season. During 2001, stop net crews were active from 15 October to 15 November, as weather permitted. The nets were set at four different locations: Fort Macon, Atlantic Beach, Salter Path, and Bogue Inlet. Information from the crew leaders regarding stop net deployment and removal was inconsistent despite regular calls and messages to them. As a result, it was not possible to calculate soak times for the stop nets. 54 Southeastern Naturalist Vol. 9, No. 1 Twenty-five beach observations of stop nets occurred during eleven days of effort. Observers documented three stop net haulbacks and three beach strikes, during which no dolphins were seen in the vicinity of or entangled in the stop nets or beach seines. Dolphins occurred within 500 m of a stop net during eight different observations. For five of the eight observations, dolphin subgroups approached within 50 m of the net; two of these were within three body lengths (approximately 9 m) of the lead. One of these closest approaches consisted of the subgroup entering the inside perimeter of the stop net, changing direction, and exiting. The other close approach occurred outside of the lead. 2002 season. During 2002, stop net crews were active from 14 October to 29 November, as weather permitted. Discussions with crew leaders at the beginning of the season and almost daily phone calls resulted in better communication, allowing for soak times to be documented for all nets. Despite the improved communication, crew leaders called only once before an anticipated haulback and often responded to messages days afterwards to relay deployment and removal times. Soak times, therefore, were not exact and often were rounded up to the nearest half hour. Stop nets were deployed at all six locations authorized by the NCDMF. Average soak time for each set was 60.8 hrs (SD = 45.1; range = 23.5–138; n = 17). From the beginning to the end of the season, fishing activity shifted from the east end of Bogue Banks to the west end of the island. The Fort Macon net was last fished with a beach seine on 13 November, and the Bogue Inlet net was last fished on 29 November. Table 1. Beach observations of Bottlenose Dolphin subgroups (n = 26) at the Fort Macon site of the stop net fishery on 16 days (total effort = 61.6 hrs) during the 2002 season. Observations were categorized as to whether the stop net was present (n = 7) or absent (n = 11); on two days, data were divided into 2 observation periods each because the net was either deployed or retrieved during the observation (n = 18 total observation periods). Period = observation period, Hours = hours observed. Stop net Dolphin subgroups Period Date Tide Hours Present Absent Number Average size 1 24 Oct High 4.0 x 3 5.7 2 30 Oct High 3.4 x 5 7.0 3 30 Oct High 0.6 x 0 - 4 31 Oct High 4.0 x 3 2.7 5 1 Nov Low 4.0 x 1 10.0 6 2 Nov Low 4.0 x 3 3.0 7 4 Nov Low 3.9 x 1 2.0 8 11 Nov High 4.0 x 0 - 9 12 Nov High 1.0 x 0 - 10 14 Nov Low 2.6 x 1 1.0 11 14 Nov Low 2.1 x 1 10.0 12 15 Nov Low 4.0 x 1 10.0 13 18 Nov Low 4.0 x 1 2.0 14 26 Nov High 4.0 x 2 5.0 15 27 Nov High 4.1 x 3 8.3 16 28 Nov High 4.0 x 0 - 17 29 Nov High 4.0 x 0 - 18 2 Dec Low 4.0 x 1 18.0 2010 B.L. Byrd and A.A. Hohn 55 Beach observations for dolphins occurred only at the Fort Macon site and for 16 days, totaling 61.6 hrs of observed effort (Table 1). No dolphin entanglement was observed. Sightings of dolphins occurred on all but four of the sixteen observation days, and one of those four survey days lasted only 1 hour due to rain. On 30 October, the stop net was deployed during approximately the last 40 minutes of the 4-hr observation period, and no dolphins were seen in the area during deployment. On 14 November, the stop net was removed for the last time, about halfway through the observation period. The observer remained to scan for dolphins in the area, but also watched for possible dolphin entanglements in the net during its removal. For these two days, data were divided into net present and net absent, resulting in 2 observation periods for each of those days and a total of 18 observation periods overall. There was no significant difference between observation periods when the net was present (n = 7) versus absent (n = 11) for the number of dolphins per hour or the number of subgroups per hour (Table 2). There also was no significant difference between high tide (n = 10) and low tide (n = 18) for the number of dolphins per hour or the number of subgroups per hour (Table 2). Twenty-six subgroups of dolphins were recorded, totaling 156 individuals. When the net was present (n = 7), dolphins passed just offshore beyond Table 2. Results of t-tests comparing various parameters for Bottlenose Dolphin sightings during stop net presence versus absence and during high tide versus low tide in 2002. Sample size (n = 18) for the categories of number of dolphins per hr and subgroups per hour represents the number of observation periods. Sample size (n = 26) for the other categories represents the number of subgroups observed during the 18 observation periods. * denotes significance. Observation period Dolphin subgroup Minutes Dolphins Subgroups Subgroup observed Test parameters per hour per hour size per subgroup Net present versus absent Mean (SD) Net present 1.7 (1.5) 0.4 (0.3) 3.8 (3.2) 21.8 (16.8) Net absent 2.9 (3.4) 0.4 (0.4) 7.9 (5.2) 19.7 (14.5) n Net present 7 7 12 12 Net absent 11 11 14 14 t critical value 2.13 2.12 2.07 2.06 P value 0.32 0.65 *0.02 0.73 High tide versus low tide Mean (SD) High tide 2.5 (3.5) 0.4 (0.5) 5.9 (4.5) 21.4 (15.1) Low tide 2.2 (1.8) 0.4 (0.2) 6.1 (6.0) 19.6 (18.0) n High tide 10 10 16 16 Low tide 8 8 10 10 t critical value 2.14 2.18 2.06 2.06 P value 0.81 0.73 0.93 0.78 56 Southeastern Naturalist Vol. 9, No. 1 the net seven times. Dolphins were not seen inside the perimeter of the net; rather, they altered direction to move around the net on three occasions (Fig. 3). During the remaining observations when the net was present, dolphins were seen far offshore from the net or swimming fairly close to shore, but west or east of the net. When the net was absent (n = 11), dolphins traversed seven times close to shore through the study area where the net would be soaking if deployed (Fig. 3). There was no significant difference between net presence versus absence in the amount of time subgroups remained in the study area (Table 2). Subgroup size, however, was smaller when the net was present compared to its absence (n = 26, t = 2.07, P = 0.02) (Table 2). There was no significant difference between high tide (n = 16) and low tide (n = 10) for the amount of time subgroups remained in the study area or the subgroup size (Table 2). Boat surveys of stop nets using the sonar-video camera system occurred on six days between 31 October and 26 November (Table 3). During the first two days, only two of the four soaking nets were observed due to deteriorating weather conditions. All deployed stop nets were observed during the latter surveys. In total, eight sonar surveys occurred on six days. However, the backstaff section of a stop net was surveyed only twice, and a suds section was never surveyed due to high surf. No submerged buoys in the suds-backstaff sections were seen. On two days, stop nets had been removed within a few hours before arrival, including the haulback observed during the beach survey on 14 November. That same day, the survey boat Figure 3. Examples of visually estimated locations of Bottlenose Dolphins in the study area during 2002 showing that dolphins moving westward avoided the stop net (i.e., 4 November) but moved through the same area when the net was absent (i.e., 15 November). 2010 B.L. Byrd and A.A. Hohn 57 opportunistically arrived in time to observe the haulback at the Bogue Inlet location, and no entanglements were observed. Strong sonar targets were detected on six of the eight sonar surveys. On two occasions, the targets were from fish schools inside the perimeter of the net. Fish were seen jumping at the water’s surface, and the sonar screen showed the distance of the target from the boat as being further than the net. The identities of the other targets were confirmed with the video camera, which showed fish caught in the stop net as the source of the return signal. No dolphins were observed entangled using the sonar-video camera system. Discussion Dolphins were observed frequently in the vicinity of stop nets in 2001 and 2002; however, no entanglements or dolphins stranded with lesions suggesting entanglements occurred during the study. Nonetheless, historical strandings data and documented entanglements from 1993 to 2007 show that lethal entanglements in stop net gear have occurred despite the fact that they have not been observed during dedicated observations of this study and others (Asher 2001) (Fig. 1). Various explanations for dolphins entangling in nets have been proposed. Dolphins may not always detect the presence of the nets visually or acoustically or they may accidentally blunder into nets during behaviors such as depredation (i.e., actively feeding out of the net) (Au 1994, Dawson 1994, Read et al. 2003). In 2001 and 2002, dolphins appeared to be aware of stop nets, changing direction to avoid them and swimming just offshore of the net. These observations support Asher’s (2001) earlier observations that dolphins generally avoided stop nets. Dolphins in NC also have been seen avoiding gill nets (Read et al. 2003), yet dolphin mortality in gill nets has been high in some years (Waring et al. 2009). Similarly, entanglement has occurred in the stop net despite observed avoidance of the gear. Table 3. Boat surveys of stop nets using the sonar-video camera system occurred on six days during the 2002 season. On two occasions, low swell height and favorable wind direction allowed the backstaff section to be surveyed with the echosounder, but not the suds section. Portion of stop Date Net location Sonar survey net surveyed Comments 31 Oct Fort Macon Yes Lead 31 Oct Atlantic Beach Yes Lead 1 Nov Emerald Isle Yes Lead; backstaff 1 Nov Bogue Inlet Yes Lead 14 Nov Fort Macon No N/A Arrived after stop net removal 14 Nov Salter Path Yes Lead 14 Nov Bogue Inlet No N/A Arrived during stop net removal 19 Nov Salter Path Yes Lead 19 Nov Bogue Inlet Yes Lead 20 Nov Salter Path No N/A Arrived after stop net removal 20 Nov Bogue Inlet No N/A Arrived after stop net removal 26 Nov Bogue Inlet Yes Lead; backstaff 58 Southeastern Naturalist Vol. 9, No. 1 Dolphin behavior, such as depredation, may be responsible for entanglement in fishing gear such as crab pots (Noke and Odell 2002) and gill nets (Hagedorn 2002, Read et al. 2003). However, depredation has never been documented for the stop net fishery. Depredation of the stop net itself is unlikely because it corrals, not catches, the target species, Striped Mullet, and because Striped Mullet is an uncommon prey item for dolphins in NC (Gannon and Waples 2004). Furthermore, the primary bycaught animals in stop nets are sharks and rays (Asher 2001), which are rarely consumed by dolphins in NC (Gannon and Waples 2004). Although Striped Mullet can account for more than 99% of fish landed in the fishery (Asher 2001), nontarget fishes may be corralled in the stop net and attract dolphins to venture inside the perimeter of the net, albeit that behavior was observed only once. In 2001, stop net fishermen said they could often detect which species of fish were corralled in the net, and chose not to deploy their beach seine for fish other than Striped Mullet. As a result, landed fishes may not be representative of all species corralled in the net. In addition, past observations of beach strikes (Asher 2001) documented bycaught fish species that are common dolphin prey, such as Menticirrhus sp. (kingfish) and Leiostomus xanthurus Lacepède (Spot). The preponderance of non-primary prey species entangled in or corralled by stop nets may lessen the chance that dolphins will interact with and, subsequently, entangle in the gear. Observations at the Fort Macon stop net site indicated the presence of the net did not affect the length of time dolphins spent in the area. This may indicate that dolphins are neither attracted to nor repelled by the presence of the gear. In fact, the only significant difference found between observations during net presence versus absence was the decrease in subgroup size during net presence. The difference may not refl ect any biological significance and may be an artifact of sample size or the effect of parameters not considered. For example, the net was absent more often towards the end of the season, and temporal changes in dolphin subgroup size may have occurred. The stop net fishery presents inherent difficulties for observer coverage. To properly observe the fishery, two types of information are required: 1) total soak time for each stop net as a measure of fishing effort, and 2) reliable data on the presence of entangled dolphins in the stop net during observations. Obtaining data on fishing effort was challenging in 2001 and 2002 because of poor communication with the fishermen; however, communications improved to some extent in 2002, allowing estimates of soak time to be recorded to the nearest 0.5 hour. Soak time data are available only from the fishermen because they are not collected by state fishery organizations, including in NC. Communication challenges also impeded observation of stop net haulbacks. For example, crew leaders rarely contacted observers in time to observe haulbacks. This communication failure was due to several possible factors: 1) the decision to remove a net was sometimes quickly made; 2) crew leaders often were focused on the task at hand and forgot to 2010 B.L. Byrd and A.A. Hohn 59 call observers until after nets were hauled; 3) one of the crew leaders did not have a cell phone, so contact was inconvenient; and 4) there was perhaps a lack of appreciation of the need for haulback observations. It cannot be ruled out that fishermen had an interest in preventing the observation of haulbacks because that is when entanglements may be discovered or may occur; however, their cooperation of providing contact information and meeting with observers at the beginning of the seasons makes purposeful non-cooperation unlikely. Because of the lack of consistent communication and the sporadic nature of the stop net haulbacks, daily observations would be necessary throughout the season to guarantee that observers would be present when haulbacks occur. In addition, observing haulbacks provides only partial information for monitoring the gear for entangled dolphins. Entangled dolphins may be inadvertently released from the net during long soak times due to strong currents, tidal changes, wave action, etc. Daily, dedicated observations would provide increased opportunity to detect submerged floats or other signs of an entangled dolphin. In situ observation of soaking stop nets is another method to investigate the presence of entangled dolphins, but with inherent difficulties. In Asher’s (2001) study, snorkelers and divers were able to scan the entire net from the beach to the terminal end of the lead to look for entangled dolphins; however, weather, surf conditions, and personnel availability severely limited the number (n = 3) of these surveys. It would be necessary, but costly, to use qualified dive personnel dedicated to observing the stop net fishery. The sonar-video camera unit worked well in 2002 to investigate the lead sections of stop nets for entangled dolphins. Observations of the suds-backstaff section, where the three known entanglements occurred, were rarely possible because wind direction and high surf made approach to the outside of the net unsafe, and fishermen objected to approaches from inside the net because of concerns for negative impacts on their catch. The inability to observe the suds-backstaff section of the stop net is a serious limitation for monitoring bycatch. Stranding data and opportunistically observed entanglements currently provide the only information about stop net entanglements. Observation limitations include actual observation time versus total net soak time. Entanglement frequency data are also lacking. Not all animals that die at sea, including those entangled in fishing gear, reach the beach (McLellan et al. 2002, Renaud et al. 1990, Tregenza 1996). Furthermore, some strandings are too decomposed to determine the presence or absence of entanglement lesions (Kuiken et al. 1994, Read and Murray 2000). The operation of the stop net fishery close to shore when beachgoers are common should increase the likelihood that entanglements are observed and reported; however, only one stranding with lesions indicative of multifilament gear has been recovered concurrent with the stop net fishery since 1999. This finding may indicate that the bycatch rate is too low to effectively observe and provide an estimate of annual bycatch. In the absence of an acceptable means or level of observer 60 Southeastern Naturalist Vol. 9, No. 1 coverage and despite limitations, marine mammal stranding data may represent the best way to monitor dolphin bycatch in the stop net fishery (Byrd et al. 2008) Acknowledgments We are grateful to Robin Baird, Brandi Biehl, Annie Gorgone, and Jennifer Lawrence (NMFS, Southeast Fisheries Science Center [SEFSC], Beaufort, NC); Allen Brooks (NC Maritime Museum, Beaufort, NC); and Debi Palka (NMFS, Northeast Fisheries Science Center, Woods Hole, MA) for assistance in planning, observing, and operating the boat for this project. We also would like to thank John Mitchell and David Ross (NMFS, SEFSC, Pascagoula, MS) for providing the sonar and training to operate the system in 2002, and Mark Fonseca (National Ocean Service, Beaufort, NC) for providing the video camera and monitor for the system. Information on stranded marine mammals would not be possible without the dedicated participants in the North Carolina Marine Mammal Stranding Network. Red Munden (NCDMF, Morehead City, NC), Stacey Horstman (NMFS, Southeast Regional Office [SERO], St. Petersburg, FL), Larry Hansen (NMFS, SEFSC, Beaufort, NC), Jim Franks (University of Southern Mississippi, Ocean Springs, MS), and two anonymous reviewers provided thoughtful comments on the manuscript. This work was funded by the NMFS, SERO in St. Petersburg, FL. No official endorsement of any product or company name herein is made or implied. Literature Cited Asher, J.M. 2001. The North Carolina stop net fishery: A case study of interaction between fisheries and Bottlenose Dolphins. Master’s Project. Duke University, Durham, NC. 75 pp. Au, W.W.L. 1994. Sonar detection of gillnets by dolphins: Theoretical predictions. Report of the International Whaling Commission Special Issue 15:565–571. Byrd, B.L., A.A. Hohn, F.H. Munden, G.N. Lovewell, and R.E. Lo Piccolo. 2008. Effects of commercial fishing regulations on stranding rates of Bottlenose Dolphin (Tursiops truncatus). Fishery Bulletin 106:72–81. Dawson, S.M. 1994. The potential for reducing entanglement of dolphins and porpoises with acoustic modifications to gillnets. Report of the International Whaling Commission Special Issue 15:573–578. Edwards, E.F., and C. Perrin. 1993. Effects of dolphin group type, percent coverage, and fl eet size on estimates of annual dolphin mortality derived from 1987 US tuna-vessel observer data. Fishery Bulletin 91:628–640. Epperly, S.P., J. Braun, A.J. Chester, F.A. Cross, J.V. Merriner, and P.A. Tester. 1995. Winter distribution of sea turtles in the vicinity of Cape Hatteras and their interactions with the summer fl ounder trawl fishery. Bulletin of Marine Science 56:547–567. Epperly, S.P., J. Braun-McNeill, and P.M. Richards. 2007. Trends in catch rates of sea turtles in North Carolina, USA. Endangered Species Research 3:283–293. Federal Register. 2006. Taking of marine mammals incidental to commercial fishing operations; Bottlenose DolphinTake Reduction Plan regulations; Sea turtle conservation; Restriction to fishing activities. Federal Register 71(80):24776–24797. Gannon, D.P., and D.M. Waples. 2004. Diets of coastal Bottlenose Dolphins from the US Mid-Atlantic coast differ by habitat. Marine Mammal Science 20:527–545. 2010 B.L. Byrd and A.A. Hohn 61 Geraci, J.R., and V.J. Lounsbury. 2005. Marine Mammals Ashore: A Field Guide for Strandings, 2nd Edition. National Aquarium in Baltimore, Baltimore, MD. 371 pp. Hagedorn, S.C. 2002. Depredation by Bottlenose Dolphins on gill nets in Dare County, North Carolina. Master’s Project. Duke University, Durham, NC. 30 pp. Hoelzel, A.R., C.W. Potter, and P.B. Best. 1998. Genetic differentiation between parapatric “nearshore” and “offshore” populations of the Bottlenose Dolphin. Proceedings of the Royal Society (London) B 265:1177–1183. Julian, F., and M. Beeson. 1998. Estimates of marine mammal, turtle, and seabird mortality for two California gillnet fisheries: 1990–1995. Fishery Bulletin 96:271–284. Kolkmeyer, T., B. Guthrie, B.L. Byrd, and A.A. Hohn. 2007. Report on the alternative platform observer program in North Carolina: March 2006 to March 2007. US Department of Commerce, NOAA Technical Memorandum NMFS-SEFSC-558. Beaufort, NC. 20 pp. Kuiken, T., V.R. Simpson, C.R. Allchin, P.M. Bennett, G.A. Codd, E.A. Harris, G.H. Howes, S. Kennedy, J.K. Kirkwood, R.J. Law, N.R. Merrett, and S. Phillips. 1994. Mass mortality of Common Dolphins (Delphinus delphis) in southwest England due to incidental capture in fishing gear. The Veterinary Record 134:81–89. Mead, J.G., and C.W. Potter. 1995. Recognizing two populations of the Bottlenose Dolphin (Tursiops truncatus) off the Atlantic coast of North America: Morphological and ecological considerations. International Marine Biolog Research Institute: IBI Reports 5:31–44. McLellan, W.A., A.S. Friedlaender, J.G. Mead, C.W. Potter, and D.A. Pabst. 2002. Analysing 25 years of Bottlenose Dolphin (Tursiops truncatus) strandings along the Atlantic coast of the USA: Do historic records support the coastal migratory stock hypothesis? Journal of Cetacean Research and Management 4:297–304. National Marine Fisheries Service (NMFS). 1998. Managing the Nation’s bycatch: Priorities, programs, and actions for the National Marine Fisheries Service. 192 pp. NOAA, US Department of Commerce, Washington, DC. NMFS. 2009. Description of National Observer Program. Available online at http:// www.st.nmfs.noaa.gov/st4/nop/index.html. Accessed 13 April 2009. Noke, W.D., and D.K. Odell. 2002. Interactions between the Indian River Lagoon Blue Crab fishery and the Bottlenose Dolphin, Tursiops truncatus. Marine Mammal Science 18:819–832. Padgett, E.H. 1995. Bogue Bank stop net fishery: A case study in coastal resource-use confl ict. Master’s Project. Duke University, Durham, NC. 31 pp. Palka, D., and M. Rossman. 2001. Bycatch estimates of coastal Bottlenose Dolphin (Tursiops truncatus) in US Mid-Atlantic gillnet fisheries for 1996 to 2000. Northeast Fisheries Science Center, Woods Hole, MA. Reference Document 01-15. 77 pp. Available online at http://www.nefsc.noaa.gov/publications/crd/ crd0115/0115.htm. Accessed 15 January 2010. Read, A.J., and K.T. Murray. 2000. Gross evidence of human-induced mortality in small cetaceans. US Department of Commerce, NOAA Technical Memorandum NMFS-OPR-15. Silver Spring, MD. 21 pp. Read, A.J., D.M. Waples, K.W. Urian, and D. Swanner. 2003. Fine-scale behaviour of Bottlenose Dolphins around gillnets. Proceedings of the Royal Society, Biology Letters 270:90–92. 62 Southeastern Naturalist Vol. 9, No. 1 Renaud, M., G. Gitschlag, E. Klima, A. Shah, J. Nance, C. Caillouet, Z. Zein-Eldin, D. Koi, and F. Patella. 1990. Evaluation of the impacts of turtle excluder devices (TEDs) on shrimp catch rates in the Gulf of Mexico and South Atlantic, March 1988 through July 1989. US Department of Commerce, NOAA Technical Memorandum NMFS-SEFC-254. Galveston, TX. 165 pp. Smolker, R.A., A.F. Richards, R.C. Connor, and J.W. Pepper. 1992. Sex differences in patterns of association among Indian Ocean Bottlenose Dolphins. Behaviour 123:38–69. Steve C., J. Gearhart, D. Borggaard, L. Sabo, and A.A. Hohn. 2001. Characterization of North Carolina commercial fisheries with occasional interactions with marine mammals. US Department of Commerce, NOAA Technical Memorandum NMFS-SEFSC-458. Beaufort, NC. 57 pp. Tregenza, N. 1996. By-catch pathology as seen from the fishing boat. Pp. 10–12, In T. Kuiken (Ed.). Diagnosis of By-catch in Cetaceans: Proceedings of the Second European Cetacean Society Workshop on Cetacean Pathology, Montpellier, France, 2 March 1994. European Cetacean Society, Saskatoon, Saskatchewan, Canada. 44 pp. Waring, G.T., J.M. Quital, and C,P, Fairfield (Eds.). 2002. Bottlenose Dolphin (Tursiops truncatus): Western North Atlantic Coastal stock. Pp. 169–180, In US Atlantic and Gulf of Mexico Marine Mammal Stock Assessments: 2002. US Department of Commerce, NOAA Technical Memorandum NMFS-NE-169. Woods Hole, MA. 318 pp. Waring, G.T., E. Josephson, C.P. Fairfield-Walsh, and K. Maze-Foley (Eds.). 2009. Bottlenose Dolphin (Tursiops truncatus): Western North Atlantic coastal morphotype stocks. Pp. 132–146, In US Atlantic and Gulf of Mexico Marine Mammal Stock Assessments: 2008. US Department of Commerce, NOAA Technical Memorandum NMFS-NE-201. Woods Hole, MA. 440 pp.