Measuring the Length of a Pelagic Longline Set: Applications for Management

—Recent U.S. federal fisheries management actions aimed at reducing bycatch have established or proposed limits on the length of a pelagic longline set. However, such management measures are unenforceable if the criteria for measuring the length of a pelagic longline set have not been defined. Differences in the physical gear and deployment strategies suggest high variability among the several methods that can be used to measure set length. This study uses a combination of scientifically monitored sets, fisheries observer data, and federal pelagic logbook data to compare four possible metrics of longline set length: the geographic track of the vessel while deploying the gear, a ‘‘folding ruler’’ approximation based on straight-line distance between ends of the sections of gear, the straight-line distance between the beginning and end of the gear, and a deadreckoning distance based on vessel speed and the time duration of the gear deployment. Results indicate that while set lengths determined from self-reported data via logbook submissions are not significantly different from the other proposed methods, the inclusion of an ending location for each set within the logbook form could provide a basis for evaluation of the regulatory measure. Pelagic longline fishing gear is used worldwide commercially to target such species as tunas Thunnus spp. and swordfish Xiphias gladius. In its simplest construction, it consists of a mainline, hook-tipped leaders (gangions), and floats to suspend the line below the water surface (see Bjordal and Løkkeborg 1996). During gear deployment, the mainline is deployed over the stern as the vessel moves forward, as the crew attaches individual baited leaders onto the mainline at predetermined intervals. Radar-reflecting buoys (‘‘high-flyers’’) and buoys that transmit a signal at a known radio frequency (‘‘beeper buoys’’) are spaced throughout the gear to maintain contact with the freefloating gear during its active fishing period, usually overnight for gear targeting swordfish. Each individual deployment of the gear is known as a ‘‘set.’’ Although perhaps intended to mean the distance of the sea surface that a longline spans or some equivalent, the term ‘‘set length’’ has not yet been expressly defined. This lack of definition has not precluded the use of the term in fisheries management measures, however. Regulatory measures based on the length of a type of fishing gear are common in state and federal commercial fisheries (e.g., gill-net fisheries), but rare in domestic pelagic fisheries managed by the U.S. National Marine Fisheries Service (NMFS). One exception is a 1999 regulatory measure that was intended to reduce incidental catches of bluefin tuna T. thynnus thynnus by limiting the lengths of pelagic longline sets to 24 nautical miles (nmi, 1 International nmi 1⁄4 1.85 km) in the Mid-Atlantic Bight from 1 July 1999 through 30 June 2000 (NMFS 1999). (Although Richards [1999] correctly noted the technical problems associated with the use of such poorly-defined geographic terms as ‘‘Mid-Atlantic Bight,’’ the use of such terms herein only refers to the specific NMFS pelagic fishery statistical areas.) More recently, the Atlantic Pelagic Longline Take Reduction Plan included a proposed measure that would limit sets to no longer than 20 nmi, also within the Mid-Atlantic Bight statistical area. Although based on several tenuous assumptions and limited data across time–area strata, simulations suggested that this restriction would reduce the total pilot whale Globicephala spp. interaction rate in the Mid-Atlantic Bight by approximately 26% (APLTRT 2006). The 2006 Take Reduction Plan measure for reducing pilot whale interactions, like the similar 1999 bluefin tuna measure, also did not define the term ‘‘set length,’’ even in an implicit fashion. There are at least two general types of measurement for the length of a longline set: (1) the length of the mainline itself, and (2) the length of the course-over-ground distance traveled by the vessel during gear deployment. Both have peculiarities concerning the variability of the measurement and the possible techniques used to measure the length. To complicate matters, the behavior of pelagic longline gear during the fishing period is highly variable and not necessarily dependent on the amount of mainline deployed. Several authors (e.g., Yano and Abe 1998; Bigelow et al. 2006) have examined the * E-mail: [email protected] 1 Present address: Nova Southeastern University Oceanographic Center, 8000 North Ocean Drive, Dania Beach, Florida 33004, USA. Received October 10, 2006; accepted September 24, 2007 Published online March 6, 2008 378 North American Journal of Fisheries Management 28:378–385, 2008 Copyright by the American Fisheries Society 2008 DOI: 10.1577/M06-238.1 [Management Brief] D ow nl oa de d by [ N ov a So ut he as te rn U ni ve rs ity ], [ D av id K er st et te r] a t 0 6: 55 1 2 Ju ly 2 01 1 relationship between vessel speed, mainline sag, and the effective fishing depths of longline gear. This gear type relies on a certain amount of ‘‘sag,’’ or slack gear, between floats to create a concave ‘‘caternary’’ curve that allows the hooks to fish at varying depths. Characterizing this sag accurately has been challenging for workers attempting to standardize the gear type among longline fisheries. In the Gulf of Mexico, Wathne (1959) demonstrated that sag or ‘‘reduction rates’’, another measure of the caternary curve formation between floats, might be as important as leader length in determining the final depth of individual hooks. Using small bathythermographs, Mizuno et al. (1997) found that the reduction rate was an important determinant in the final depth distribution of hooks even in the presence of vertical shear from a nearby current. In the absence of other metrics, several authors (e.g., Yano and Abe 1998; Bigelow et al. 2006) have used the great-circle distance (the shortest distance between points on a sphere measured along a path on the surface) between the start and end latitude and longitude positions as one of two factors in calculating the reduction rate. The other reduction rate factor used by these authors was the captain–operator-reported estimate of the amount of mainline deployed for each set. Presumably, these experienced fishers would be best qualified to estimate the amount of deployed gear, but even here the estimates vary, and the ratio of these two numbers is very rarely equal to the 1.00:1 that would signify agreement between the measured and estimated values. Several possible measuring devices could be used if the designation of the ‘‘set length’’ is the type of measurement based on the physical length of mainline deployed. For example, because the mainline spool has a mechanism that winds the mainline onto the spool evenly, one possibility could be to measure the difference in diameters of the mainline wound on the spool before and after the gear deployment. However, as the mainline is wound onto the spool at varying speeds and tension, this can affect the diameter of the mainline strand itself up to 50% (J. Lindgren, Lindgren-Pitman, Inc., personal communication) and, therefore, the spool diameter would probably result in a highly variable measurement. Another method could use the so-called ‘‘line shooter’’ device that is positioned near the stern of the vessel and pulls mainline off the spool faster than the vessel is moving forwards, thereby allowing the captain to more accurately generate a deep caternary curve ‘‘sag’’ between the buoy floats attached to the mainline. As shown by Bigelow et al. (2006), knowledge of both the predetermined speed at which this device operates and the duration of the setting procedure allows a direct calculation of the amount of mainline deployed. (Equations for such a calculation are shown in the Lindgren-Pitman, Inc., Pompano Beach, Florida, on-line product instructions for the line setter device; see www.lindgren-pitman.com/pdfs/ LS-5LinesetterManual.pdf.) However, and unlike the fishing practices within the Hawaii fleet, line shooters are infrequently used in the U.S. Atlantic pelagic longline fleet, accounting for only 0.07% of all logbook-reported longline sets during the 2001 to 2004 period (NMFS, unpublished data). Without physical mainline measuring devices, an alternative type of measurement that employs vessel speed or geographic position must be used to estimate the length of a pelagic longline set. This paper describes and compares four different methods of this measurement type that were used to estimate the set length of commercial pelagic longline sets in the western Atlantic Ocean between 2001 and 2006.