Development and Implementation of Fisheries Bycatch Monitoring Programs In The Gulf Of Mexico Acknowledgment Many individuals contributed to the development of this report. We would like to thank all Gulf of Mexico Program partners who gave of their time, energy and expertise in creating this document. Special thanks are also due to the numerous state and federal agency personnel who provided the authors with many hard-to-find references. This document has been funded by the United States Environmental Protection Agency under Cooperative Agreement Number MX-994717-95-0 awarded to the Mississippi Agricultural and Forestry Experiment Station of Mississippi State University. The contents of this document do not necessarily represent the views and policies of the Environmental Protection Agency, nor does the mention of trade names or commercial products constitute an endorsement or recommendation.   David D. Burrage Mississippi State University Coastal Research & Extension Center Sea Grant Advisory Service 2710 Beach Boulevard, Suite 1-E Biloxi, MS 39531 Steven G. Branstetter Gulf & South Atlantic Fisheries Development Foundation, Inc. Lincoln Center, Suite 997 5401 West Kennedy Boulevard Tampa, FL 33609 Gary Graham Texas Agricultural Extension Service Sea Grant College Program P.O. Box 1675 Galveston, TX 77553-1675 Richard K. Wallace Auburn University Marine Extension & Research Center Sea Grant Extension 4170 Commanders Drive Mobile, AL 36615 TABLE OF CONTENTS THE ROLE OF BYCATCH IN FISHERIES CONSERVATION AND MANAGEMENT OVERVIEW OF GULF OF MEXICO FISHERIES OBJECTIVES AND METHODOLOGY THE COMMERCIAL SHRIMP FISHERY Shrimp Trawl Bycatch Characterization Studies Species-Specific Characterization Studies Shrimp Trawl Bycatch Reduction Implications Non-Gear Shrimp Trawl Bycatch Reduction Efforts Turtle Excluder Devices (TEDs) -Finfish Bycatch Reduction Shrimp Trawl Bycatch Reduction Devices (BRDs) TRAP FISHERIES Fish Traps Crab Trap Fisheries Related Material From Outside The Region Lobster Trap Fisheries Related Material From Outside The Region PELAGIC LONGLINE FISHERIES BOTTOM LONGLINE AND COMMERCIAL HOOK-AND-LINE FISHERIES Florida Grouper Longline Fishery Red Snapper Longline Fishery Golden Tilefish Longline Fishery Western Gulf Grouper Longline Fishery Kali Pole Longline Fishery Traditional Hook and Line Fishery RECREATIONAL SHRIMP FISHERIES PURSE SEINE FISHERIES GILL AND TRAMMEL NET FISHERIES FINFISH TRAWLS RECREATIONAL HOOK AND LINE FISHERIES SUMMARY AND RECOMMENDATIONS Additional References     THE ROLE OF BYCATCH IN FISHERIES CONSERVATION AND MANAGEMENT Fishery resources are harvested from Gulf of Mexico waters using a variety of gears and methods. With few exceptions, most of the fisheries have an element of bycatch associated with them. For purposes of this report, "bycatch" includes discarded fish, shellfish, or other organisms which are taken as non-target incidental catch in fisheries. Bycatch includes those fish and shellfish that have no market value, are damaged during harvest, or cannot be legally retained, landed, or sold. Other organisms such as marine mammals, birds, and turtles are accidentally caught and discarded in some fisheries. The fishing event may cause either immediate mortality or the potential for future mortality as a result of gear interactions or handling. Fisheries in the Gulf of Mexico are prosecuted under a wide range of management regimes. In waters beyond state jurisdiction, many fisheries are managed under federal regulations promulgated by the National Marine Fisheries Service (NMFS) and the Gulf of Mexico Fishery Management Council (Council). Fishery resources taken primarily from state waters are managed by the respective Gulf states, and some species are managed through interstate compacts initiated and developed by the Gulf States Marine Fisheries Commission (Commission). Fishery management, in the Gulf of Mexico and worldwide, has intensified and undergone many changes. Limited entry, ITQ (individual transferable quotas which privatize harvesting rights), trip limits (frequency and landing limits), gear restrictions, area closures, and seasons further restrict the fishing industry. Often more than one management method is employed simultaneously. All of these management options strongly influence bycatch and discard rates either by changing fishermen's behavior or altering the type of fishing technology used. The discard of bycatch or lower-valued fish (high-grading) is among the most difficult fishery management challenges, making attainment of conservation and economic goals of fishery managers, the fishing industry, and the public problematic (Dewees and Ueber, 1990). Probably more common is bycatch from non-selective harvesting gear. These discards contribute to the fishing-related mortality of many species and are a factor in management decisions such as setting catch quotas and fishing seasons. Fishery bycatch also plays a larger role in the overall balance of the Gulf ecosystem when viewed in light of the fate of discards, predator-prey relationships, and environmental quality. Bycatch in the Gulf of Mexico is an issue of great concern when viewed in light of the many fisheries which are conducted in the region and the level of effort being expended to harvest fishery resources. Most bycatch research has focused on the commercial shrimp fishery. According to a recent FAO report (Alverson et al., 1994), this fishery ranks fifth in the world in bycatch generated (discard weight per landed target catch weight). Many other fisheries and fishing gears within the Gulf also impact non-target species. These include pelagic and bottom longlines, commercial hook and line, purse seines, trap fisheries, gill and trammel nets, recreational hook and line, finfish trawls, and recreational shrimp trawls. Bycatch in these fisheries is generated by incidental catch of non-target species and release of regulated species which are under- or over-sized or out of season. Given the large area covered, the multitude of fisheries, and an increasing population that heavily uses marine resources, addressing the bycatch issue is a timely endeavor. OVERVIEW OF GULF OF MEXICO FISHERIES The fishery resources in the Gulf of Mexico support an extensive commercial and recreational fishery. Due to increased demands for fishery products in the marketplace and an increase in individual leisure time and discretionary income, more pressure is being brought to bear on the fishery resources of the Gulf. It is important to understand the status of the fishery resources and the demands being placed on them. The total United States commercial harvest in 1995 was 9.9 billion pounds or 4.5 million metric tons with an ex-vessel value (price paid to the fishermen) of $3.8 billion. The five Gulf states produced 15 percent (1.4 billion pounds) of this volume and accounted for 19 percent ($725 million) of the value of these landings (NMFS, 1996). Included in these landings are shrimp, the most valuable fishery in the nation; and menhaden, the second largest fishery by volume in the nation. In 1995, marine recreational fishermen in Florida, Alabama, Mississippi, and Louisiana participated in 17 million fishing trips and landed approximately 136 million fish. Thirty percent of all marine recreational angling trips in the nation occurred in the Gulf and these anglers landed 44 percent of all fish reported in 1995 (NMFS, 1996). According to the latest estimates, there are 33,696 commercial fishing vessels registered or documented in the five Gulf states (NMFS, 1996). Approximately 25,000 of these are classified as "boats" (under five net tons) which are typically used in inshore fisheries such as oyster tonging, gill netting, and crabbing. Over 8,000 units of the commercial fishing fleet are classified as "vessels" (over five net tons). These are the offshore shrimp trawlers, longline and bandit-rigged reef fish vessels, pelagic longliners, oyster dredgers, purse seiners, and finfish draggers. Many of the larger vessels are typically rigged to participate in more than one fishery. The amount of effort expended in Gulf fisheries can be approximated by examining license sales in each of the Gulf states. The National Marine Fisheries Service also issues permits for several fisheries which are conducted primarily in federal waters. The number of licenses sold in each fishery by state is not a true indicator of the total number of fishermen harvesting Gulf resources because some fishermen purchase several licenses in their own or other states for the same fishery. However, if one distinguishes between resident and nonresident sales, a better approximation can be obtained. Additionally, states vary in the way licenses are issued. For example, some states license vessels, others license individuals, and others do both. Some states have exemptions for certain classes of people regarding license requirements in some fisheries. The following tables are compiled from information obtained from the Gulf States Marine Fisheries Commission (1996) and the National Marine Fisheries Service (1996): FLORIDA Type of License Number Sold FY95 Resident Annual Salt Water Sport Fishing 542,378 Resident 10-Day Salt Water Sport Fishing 56 Nonresident Annual Salt Water Sport Fishing 75,395 Nonresident 7-Day Salt Water Sport Fishing 51,578 Nonresident 3-Day Salt Water Sport Fishing 207,571 Blue Crab Permit 6,082 Stone Crab Permit 7,258 Crawfish Permit 2,463 Spiny Lobster Trap Certificate -- Each Trap 63,470 Shellfish Relaying Permit 20 Shellfish Leases 386 Resident Apalachicola Bay Oyster Harvesting 748 Nonresident Apalachicola Bay Oyster Harvesting 2 Noncommercial Lobster Permits 520 Noncommercial Shrimp Permits 403 Resident Indian River Clam Permit 900 Nonresident Indian River Clam Permit 74 Dead Shrimp Production License - -St. Johns River Only 74 Bait Shrimp -- Statewide 111 Live Shrimp Production License 73 Noncommercial Shrimp Trawling License -- St. Johns River Only 17 Purse Seine* 208 Manatee County Gill Net* 219 Pinellas County Gill Net* 55 Nassau County Gill Net* 0 Hillsborough County Gill Net* 45 Sarasota County Gill Net* 82 St. Johns County Beach Seine 7 License to Take Sardine-like Fish from Pinellas County Waters 5 Pleasure Vessel Registrations 713,413 Commercial Vessel Registrations 34,188 Nonresident or Alien Commercial Vessel Fees 674 Resident Saltwater Products 18,933 Nonresident Saltwater Products 728 Alien Saltwater Products 136 Restricted Species Endorsement 9,503 *These types of license will probably decline drastically due to Florida's recent net ban which was instituted on July 1, 1995.   MISSISSIPPI Type of License Number Sold FY95 Resident Salt Water Sport Fishing 48,444 Nonresident Salt Water Sport Fishing 6,645 Resident Gill and Trammel Net 220 Nonresident Gill and Trammel Net 4 Resident Recreational Shrimp 503 Nonresident Recreational Shrimp 1 Resident Shrimp Under 30' 347 Nonresident Shrimp Under 30' 3 Resident Shrimp 30' to 45' 389 Nonresident Shrimp 30' to 45' 39 Resident Shrimp Over 45' 449 Nonresident Shrimp Over 45' 65 Resident Commercial Crab 148 Nonresident Commercial Crab 18 Recreational Crab (Resident Only) 3 Charter/Party Boat (Resident Only) 84 Commercial Hook and Line 86 Recreational Oyster (Resident Only) 105 Resident Oyster Tonging 46 Nonresident Oyster Tonging 25 Resident Oyster Dredging 119 Nonresident Oyster Dredging 15 Live Bait Boat 37   ALABAMA Type of License Number Sold FY95 Resident Annual Salt Water Rod and Reel 39,245 Nonresident Annual Salt Water Rod and Reel 3,769 Resident Commercial Oyster Catcher 707 Nonresident Commercial Oyster Catcher 5 Shrimp Under 30' 757 Shrimp 30' to 45' 222 Shrimp Over 45' 199 Nonresident Commercial Shrimp 242 Resident Recreational Shrimp Boat 1,727 Nonresident Recreational Shrimp Boat 90 Resident Commercial Crab Fisherman 150 Nonresident Commercial Crab Fisherman 3 Resident Commercial Net License (1,200' or Less) 362 Nonresident Commercial Net License (1,200' or Less) 30 Resident Commercial Net License (1,201' to 2,400') 204 Nonresident Commercial Net License (1,201' to 2,400') 42 Resident Purse Seine 2 Nonresident Purse Seine 8 Resident Recreational Net (Not to Exceed 300') 384 Nonresident Recreational Net (Not to Exceed 300') 17 Resident Commercial Hook and Line 60 Nonresident Commercial Hook and Line 0 Resident Charter Boat -- 6 Passenger 70 Nonresident Charter Boat -- 6 Passenger 8 Resident Charter Boat -- 25 Passenger 14 Nonresident Charter Boat -- 25 Passenger 0 Resident Charter Boat -- Over 25 Passenger 1 Nonresident Charter Boat -- Over 25 Passenger 2   LOUISIANA Type of License Number Sold FY95 Resident Oyster Tong -- Per Tong 181 Resident Oyster Dredge -- Per Dredge 1,084 Nonresident Oyster Dredge -- Per Dredge 45 Resident Commercial Fisherman 15,062 Resident Hoop Net -- Any Legal Number 1,753 Resident Fish Seine -- Any Legal Number 162 Resident Trammel Net -- Any Legal Number 467 Resident Freshwater Gill Net -- Any Legal Number* 1,000 Nonresident Hoop Net -- Any Legal Number 34 Nonresident Trammel Net -- Any Legal Number* 17 Resident Vessel License 14,323 Nonresident Vessel License 1,581 Resident Purse/Menhaden Seine -- Per Seine 57 Resident Shrimp Trawl -- Per Trawl 10,095 Nonresident Shrimp Trawl -- Per Trawl 3,553 Resident Oyster Harvester 940 Nonresident Oyster Harvester 28 Nonresident Commercial Fisherman 1,625 Nonresident Fish Seine -- Any Legal Number 1 Resident Butterfly Net -- Per Net 3,050 Nonresident Butterfly Net -- Per Net 37 Resident Slat Trap -- Any Legal Number 218 Nonresident Slat Trap -- Any Legal Number 0 Nonresident Purse/Menhaden Seine -- Per Seine 1 Resident Crab Trap -- Any Legal Number 3,423 Nonresident Crab Trap -- Any Legal Number 65 Resident Crab Trap Attached to Trotline 321 Resident Eel Pot License 8 Resident Minnow Trap License 136 Resident Mullet Permit 582 Nonresident Mullet Permit 79 Resident Spear Gun -- Per Gun 29 Resident Set Line License 1,185 Nonresident Set Line License 133 Resident Dip/Cast Net License 383 Nonresident Dip/Cast Net -- Per Net 1 Resident Flounder Gig License 25 Nonresident Flounder Gig -- Per Gig 0 Resident Can, Bucket, Pipe, Drum, Tire 66 Nonresident Can, Bucket, Pipe, Drum, Tire 0 Resident Skimmer Net 5,447 Nonresident Skimmer Net 68 Resident Saltwater Gill Net* 781 Nonresident Saltwater Gill Net* 73 Resident Mullet Strike Net* 755 Nonresident Mullet Strike Net* 80 Resident Pompano Strike Net* 34 Resident Saltwater Rod and Reel 3 Resident Spotted Seatrout Permit 73 Resident Saltwater Gill Net for EEZ* 2 Nonresident Saltwater Gill Net for EEZ* 1 Out of State Oyster Landing Permit 13 Resident Soft Shell Crab Shedder 36 Resident Pompano Permit 8 Resident Restricted Species Permit 46 Resident Recreational Saltwater Fishing Season 280,360 Nonresident Recreational Saltwater Fishing Season 6,510 Nonresident Recreational Saltwater Trip -- 7 Days 1,269 Nonresident Recreational Fresh/Saltwater Trip -- 2 Days 27,618 Resident Recreational Hoop Net -- No More Than 5 Nets 4,288 Nonresident Recreational Hoop Net -- No More Than 5 Nets 87 Resident Recreational Slat Traps -- No More Than 5 Traps 746 Nonresident Recreational Slat Traps -- No More Than 5 Traps 25 Resident Recreational Crab Traps -- No More Than 10 Traps 3,116 Nonresident Recreational Crab Traps -- No More Than 10 Traps 15 Resident Recreational Shrimp Trawl -- Per 16' Trawl 4,389 Nonresident Recreational Shrimp Trawl -- Per 16' Trawl 44 Resident Recreational Oyster Tong -- Per Tong 63 Nonresident Recreational Oyster Tong -- Per Tong 1 Resident Recreational Crab Trap -- Per Trap on Trotline 1,549 Nonresident Recreational Crab Trap -- Per Trap on Trotline 10 *These types of license will probably decline drastically due to Louisiana's recent net ban which will be phased in by 1997.   TEXAS Type of License Number Sold FY95 Resident Recreational Fishing 1,043,764 Lifetime Resident Recreational Fishing 14 Temporary Resident Recreational Fishing -- 14 Day 77,784 Saltwater Sportfishing Stamp 624,218 Special Resident Fishing 7,121 Nonresident Recreational Fishing 47,109 Temporary Nonresident Recreational Fishing -- 5 Day 63,236 Commercial Crab Trap Tag 79,723 Saltwater Trotline Tag 10,395 Resident Commercial Oyster Fisherman 5 Nonresident Commercial Oyster Fisherman 0 Resident Commercial Oyster Boat Captain 413 Nonresident Commercial Oyster Boat Captain 18 Resident Commercial Oyster Boat 343 Nonresident Commercial Oyster Boat 12 Resident Sport Oyster Boat 78 Nonresident Sport Oyster Boat 1 Resident Commercial Fishing Boat (Fresh and Saltwater) 1,334 Nonresident Commercial Fishing Boat (Fresh and Saltwater) 29 Resident Commercial Mussel and Clam Fisherman 108 Nonresident Commercial Mussel and Clam Fisherman 7 Commercial Fishing Boat (Menhaden Only) 15 Resident Commercial Gulf Shrimp Boat 896 Nonresident Commercial Gulf Shrimp Boat 441 Resident Commercial Bay Shrimp Boat 1,529 Nonresident Commercial Bay Shrimp Boat 0 Resident Commercial Bait Shrimp Boat 1,614 Nonresident Commercial Bait Shrimp Boat 0 Shrimp House Operator's Individual Bait Shrimp Trawl Tag 974 Resident General Commercial Fisherman's 4,175 Nonresident General Commercial Fisherman's 45 Resident Commercial Finfish Fisherman's 1,380 Nonresident Commercial Finfish Fisherman's 11   FEDERAL Type of Permit/Endorsement Number Issued (Most Recent Fishing Year) Gulf and South Atlantic Shark 1,841 Swordfish 984 Coastal Pelagics (Charter Boats Only) 1,455 Commercial Mackerel (Includes Gill Net Endorsement) 2,839 Commercial Mackerel Gill Net Endorsement 115 Reef Fish (Charter Boats Only) 515 Commercial Reef Fish (Includes Bandit, Longline and Trap) 1,451 Commercial Reef Fish Trap 95 Commercial Reef Fish Bandit 939 Commercial Reef Fish Bottom Longline 436 Red Snapper 2,000 Pound Endorsement 131 Commercial Spiny Lobster 294 Spiny Lobster Tailing Permit 377       OBJECTIVES AND METHODOLOGY The general goals of the Gulf of Mexico Program regarding fisheries bycatch are to 1) conserve and restore species diversity and health of aquatic resources while allowing sustainable development, 2) assess and monitor the effects of fishing mortality on the health and abundance of living aquatic resources in the Gulf of Mexico, 3) enhance the sustainability of Gulf commercial and recreational fisheries, and 4) identify and assess existing commercial and recreational bycatch data to determine research needs. The specific objectives of this project were as follows: Conduct a Gulfwide survey of agencies and organizations to determine existing and on-going data and develop a data catalog; Compile and analyze these data, identify data gaps, and develop a summary document for peer review and publication; and Assist the Gulf of Mexico Program with information and technology transfer activities addressing fishery bycatch in the Gulf of Mexico by disseminating information developed under this project to a wide variety of user groups, fishery managers, and the general public. The methodology used in preparation of this report encompasses more than the original scope of work specified by the Gulf of Mexico Program. The authors made a conscious decision to expand the search for relevant information beyond "a Gulfwide survey of agencies and organizations" in order to include work done outside of the region involving species which are also indigenous to the Gulf of Mexico. Similarly, work done in other areas with the same fishing gear types used in the Gulf but targeting different species was sometimes included in the analyses so that a variety of research techniques and protocols could be compared and evaluated. It is hoped that experience gleaned from this wider examination of bycatch research will be useful in shaping future efforts in the Gulf of Mexico region. Cooperators in the project conducted a survey of agencies and organizations to identify existing data. Contact points were state and federal fishery management agencies, private sector fishery organizations, academic researchers, environmental entities, the Sea Grant advisory network, the Gulf States Marine Fisheries Commission, and public/private organizations such as the National Fisheries Institute and fisheries development foundations. Data were compiled using computer and literature searches as well as individual personal contacts throughout the Gulf region. Much of the documentation associated with this effort is in the form of grant final reports, or reports to agencies, and some of it is preliminary. More detailed and conclusive peer-reviewed literature is likely forthcoming. The project collaborators divided their efforts based on their respective knowledge of the fisheries, fishing gears, and fishing techniques used to harvest living marine resources from the Gulf of Mexico region. The findings in this report are presented according to a classification based on individual fisheries and major gear types used. Wherever practical, citations include mailing addresses to facilitate document acquisition by interested readers.   THE COMMERCIAL SHRIMP FISHERY Shrimp trawling has long been identified as a non-selective fishing activity, with numerous species being vulnerable to the nets. Bycatch in this fishery usually exceeds the catch of shrimp. Since 1990, a comprehensive multi-organizational effort, funded through federal, state and private sources, has addressed shrimp trawl bycatch. Much of the documentation of this effort is in the form of grant final reports, or reports to agencies, and some of it is preliminary. To date, the program has generated information on nearly 6,000 commercial shrimp trawl tows in the southeastern U.S., with a focus on the Gulf of Mexico. In addition, several fishery-independent surveys have been conducted. All totaled, there is a substantial database that can be reviewed for characterization of the catch of the commercial shrimp fishery, and more importantly, much research has now been completed on options to reduce bycatch. Reduction of finfish bycatch, especially for certain heavily fished species such as red snapper, is expected to help rebuild the stock. In the past few years, the shrimp industry has modified its gear configurations and operational techniques; these changes, though not specifically designed to reduce finfish bycatch, achieved that goal as well. For example, the addition of turtle excluder devices (TEDs) in the fishery also reduced the catch of finfish. Several bycatch reductions devices (BRDs) have been tested in the Gulf of Mexico, and some designs have shown good results at substantially reducing finfish while losing only minimal amounts of shrimp. Shrimp Trawl Bycatch Characterization Studies:   Adkins, G. 1993. A comprehensive assessment of bycatch in the Louisiana shrimp fishery. Technical Bulletin No. 42, Louisiana Depart. Wildl. & Fisheries, Marine Fisheries Div., Bourg, Louisiana 70343. Also available as a MARFIN Final Report (Award NA89WC-H-MF006), for the period 1 January 1989 through 31 December 1989. The study compared the catch rates of offshore and inshore trawlers and wingnet efforts during 108 commercial tows. Brown and white shrimp comprised 60 percent of the catch by number. By weight the average fish/shrimp ratio was 3.2:1 with inshore trawl bycatch higher (3.0 vs. 2.2:1) than offshore trawling; wingnet bycatch was 4.7:1. Although the wingnet catches were higher, the shorter tow times and the handling procedures meant more was released alive than from otter trawling. The study reviews much of the bycatch literature available at the time, and notes the possibilities of reducing bycatch through various methods such as area/time closures, and use of excluder devices.   Baltz, D. M. 1993. Patterns in the distribution and abundance of fishes and macroinvertebrates in a Louisiana marsh: shrimp bycatch in the inshore, fishery-independent trawl samples. MARFIN Final Report (Award NA17FF0263-01) by Louisiana State University, Coastal Fisheries Institute, Baton Rouge, Louisiana 70803 Using a 20-year fishery-independent trawl survey of three stations in coastal Louisiana, 141 taxa were identified from over 2,000 tows. Of these taxa, 90 were considered rare, with less than 100 individuals being documented during the entire period; thus only about 50 species occurred with some regularity. Of the 141 taxa, fish comprised 110 species. Two species, the bay anchovy and the Atlantic croaker, comprised 72 percent of the catch by number. The study indicated a stable community structure even with the highly fluctuating environmental conditions that occurred on an annual or longer basis. One disappointing point in this study is a lack of CPUE data by year to indicate trends in relative abundance over such a long time span. This analysis would have suggested the long-term effects of trawling on bycatch species found in inshore Louisiana waters (see Perret et al. 1995; this section for that information).   Boylan, J.M., R.P. Webster, H.R. Beatty, and E.L. Wenner. 1990. Results of trawling efforts in the coastal habitat of the South Atlantic Bight. SEAMAP--SA Final Report, FY-1990. Marine Resources Research Div., South Carolina Wildl. & Mar. Resources, Dept., P.O. Box 12559, Charleston, SC 29412. This study, looking at the South Atlantic Bight, stratified analyses in an inner and outer area by depth. Diversity and abundance was higher in the inner areas. The dominant species overall was spot, occurring in 71 percent of the samples. Atlantic croaker ranked second in frequency of occurrence. The report focused analyses on the mackerels. Spanish mackerel densities were estimated at 1.5 individuals per hectare and included fish from 3-51 cm (mean = 20 cm); length frequencies indicated young-of-the-year and early age-I fish in the catch. King mackerel were estimated at one individual per hectare, and ranged from 4-44 cm (mean = 18 cm), and length frequencies indicated the presence of young-of-the-year and a strong representation of age-I individuals.   Chittenden, M.E. Jr., and J.D. McEachran. 1976. Composition, ecology, and dynamics of demersal fish communities on the northwestern Gulf of Mexico continental shelf, with a similar synopsis for the entire Gulf. A Final Report to Texas Sea Grant (TAMU-SG-76-208), Texas A&M University, College Station, TX 77843. Collections aboard commercial shrimp trawlers during normal working conditions on 4 seasonal trips monitored 21 tows on white shrimp grounds and 39 tows on brown shrimp grounds. 103 species were taken in the 18 kg/tow samples; an additional 58 species were found in the culled catch of the net. The discard to shrimp volume ratio was 11.35:1. The document notes that the authors assumed invertebrates made up 10-20 percent of the discard, thus the fish to shrimp ratio was estimated at 10:1. The document does note that the discard catch in the white shrimp grounds may have been biased due to some large discard catches in June.   Coleman, F.C., C.C. Koenig, and W.F. Herinkind. 1992. Annual report: survey of Florida inshore shrimp trawling by-catch and preliminary test of by-catch reduction devices.(Copy received from Gulf of Mexico Fishery Management Council who received it from the Florida Marine Fisheries Commission. This study continued for at least 2 more years; more recent results are probably available from the authors at: Dept. Biological Sciences, Florida St. Univ., Tallahassee, FL 32306). This document presents the results of the first year's analysis of an ongoing project. It provides a table of the species composition of the catch in 10 different Florida sites, and preliminary results of bycatch reduction devices testing. The tabular material on species composition is not ranked by abundance, although the text does list the most abundant species. Given the preliminary nature of this report, and its limited sample sizes, few conclusions can be drawn; species composition and abundance changed dramatically between seasonal sampling trips, and differed substantially among sites.   Fuls, B. 1995. Assessment of composition and magnitude of bycatch associated with the commercial shrimp trawling industry in central lower Texas coastal bays during spring and fall Texas commercial bay-shrimp open seasons. Saltonstall-Kennedy Program Final Report (Award NA37FD0083) by Texas Parks and Wildl. Dept., Austin, TX. Monitoring three bays in lower Texas - Aransas Bay, Corpus Christi Bay, and lower Laguna Madre - this study reported that bycatch was higher in the spring than in the fall. Bycatch to shrimp ratios were 4:1 to 6.8:1 depending upon season and area. Finfish to shrimp ratios were 1:1 to 5.1:1, again varying by season and area. Bycatch ratios were highest in Corpus Christi Bay and lowest in Laguna Madre. The report also noted that the quantity and composition of the bycatch in this fishery-independent survey was very different from concurrent fishery-dependent surveys. Bait shrimp bycatch surveys (9.8 m trawl) in Lower Laguna Madre during the spring of 1993 showed that four species (lesser blue crab, Callinectes similis; Atlantic croaker, Micropogonias undulatus; spot, Leiostomus xanthurus; and sand seatrout, Cynoscion arenarius) comprised 62 percent and 43 percent of the mean CPUE in number and weight, respectively. The overall mean CPUE for bycatch was 2,966 individuals/h/net in number and 54.643 kg/h/net in weight. In the fall, five species (sand seatrout, Cynoscion arenarius; lesser blue crab, Callinectes similis; spotfin mojara, Eucinostomus argenteus; hardhead catfish, Arius felis; and Atlantic cutlassfish, Trichiurus lepturus) comprised 65 percent and 53 percent of the mean CPUE in number and weight, respectively. The overall mean CPUE for bycatch organisms was 1,597 individuals/h/net in number and 27.775 kg/h/net in weight.   Griffin, W.L., and A.K. Shah. 1995. Estimation of standardized effort in the heterogeneous Gulf of Mexico shrimp fleet. MARFIN final report (Award #NA37FF0053). Because of concerns about shrimp effort estimates performed by NMFS, this study was conducted to examine alternative estimation techniques. Compared to the current NMFS practice of expanding effort using simple average CPUEs and extrapolating these data for empty cells, the models used in this analysis correct for potential biases associated with blank cells and non-proportional reporting between interviews and landings. Models produced a similarity in estimates to those of NMFS through 1980, but a divergence since that time. The study noted that since 1980 there has been substantial underestimation of "boat" effort and overestimation of "vessel" effort because of non-proportional interviews between these two sectors. The authors estimate that during the period 1965-1993 inshore effort (as nominal days fished) tripled, but boat interviews declined. In contrast offshore effort doubled, but vessel interviews were proportionally too high in the estimates. (See Nance 1992 and 1995 below; this section).   Gulf and South Atlantic Fisheries Development Foundation. 1994. Organization and management of a Gulf of Mexico and South Atlantic Ocean fishery bycatch management program (Year II). Saltonstall-Kennedy Grant Program, Final report to the National Marine Fisheries Service (Award NA37FD0032) by the Foundation (Ste. 997, 5401 W. Kennedy, Tampa, FL 33609). As part of this grant, observers logged 744 days on 63 commercial fishing trips gathering bycatch data for characterization of the catch and evaluation of various BRDs under actual operating conditions. A total of 362 nets were sampled for characterization, and 653 tows compared the catch of a "control" (without a BRD) net to the catch of a BRD-equipped net. The report notes that finfish comprised 67 percent of the catch by weight, while shrimp represented 19 percent; no South Atlantic characterization data were available for analysis. Red snapper were noted to make up less than 1 percent of the catch by weight. BRD testing under this project is reported under the "Bycatch Reduction" section of this report. See also NMFS 1995 under this section for more detailed characterization analyses completed using the entire data set, including foundation and other research efforts.   Huner, B., and G. Faulkner. 1995. Energy conservation in the Louisiana shrimp trawling industry. Final report to the Louisiana Dept. Nat. Resources, Energy Div., P.O. Box 44156, Baton Rouge, LA 70804 Focusing on various webbing materials (spectra, knotted, and unknotted polypropylene), this report notes that shrimp catch was similar among webbing types as was finfish catch. Fish to shrimp ratios were 2.2 lb fish to 1 lb of shrimp.   Keiser, R.K. Jr. 1976. Species composition, magnitude, and utilization of the incidental catch of the South Carolina shrimp fishery. Technical Report 16, South Carolina Marine Resources Center, Charleston, SC. A total of 294 tows from 120 trips aboard commercial shrimp boats ranging in length from 35 to 75 feet were sampled in 1974 and 1975. Monthly bycatch to shrimp ratios were 1:1 to 3:1. Fish CPUE ranged from 15 kg/hr to 244 kg/hr while shrimp ranged from 17 to 160 kg/hr. A total of 105 fish species were identified, and only a few species comprised the majority of the catch. Mean total lengths of 25 species ranged from 6.9 to 18.6 cm. Sciaenids made up from 50-80 percent of the catch; spot was the most abundant making up over 30 percent of the yearly catch, followed by star drum at 12 percent, and Atlantic croaker was the fourth most abundant. This report includes numerous tables and figures representing the catch analyses, and provides detailed discussions of the catch and implications of its potential for better utilization.   Keiser, R.K. Jr. 1977. The incidental catch from commercial shrimp trawlers of the South Atlantic states. South Carolina Marine Resources Center, Technical Report 26, South Carolina Wildl. & Marine Resources Dept., Charleston, SC 29412. This report documented, through available literature, the catch in shrimp trawls for the South Atlantic states, North Carolina to Florida. Fish to heads-on shrimp ratios ranged from 1.2:1 to 4:1. For North Carolina, results ranged as high as 100:1, but the average was 4:1. Night time ratios were lower than daytime ratios; not because less fish were caught at night, but because more shrimp (17 percent vs. 13 percent of catch) were taken at night. The two most common species, spot and Atlantic croaker, comprised 63 percent of the finfish catch. For South Carolina, ratios ranged from 2.6:1 in summer to 1.2:1 in fall. Spot and Atlantic croaker were the dominant species, comprising 50 percent of the catch. Bycatch ratios in Georgia were estimated at approximately 2.5:1, and spot, Atlantic croaker, star drum, and bay anchovy were the dominant species. In the Atlantic Florida region, the ratio was approximately 3.8:1. No current estimates of species composition were cited.   Martinez, E.X., and J.M. Nance. 1993. Trawling Bycatch in the Galveston Bay System. The Galveston Bay National Estuary Program Publication GBNEP-34. NMFS/SEFSC, Galveston, Texas. The characterization study, performed by the National Marine Fishery Service (NMFS) Galveston Laboratory, was conducted in three phases: 1) a review of historical bycatch studies, 2) initiation of new data collection efforts on commercial vessels and 3) a comparison of new data collected with fishery-independent surveys of the Texas Parks and Wildlife Department. Three historical studies regarding bycatch in Galveston Bay were identified and reviewed. Matlock (1982) analyzed the catch of gulf and southern flounder Paralichthys albigutta and P. lethostigma, respectively) in 34 tows from a commercial shrimp vessel during April-November 1978. He concluded bycatch of flounder was lower in Galveston Bay than in other Texas bay systems. Lamkin (1984) reviewed bycatch in tows sampled from one bait shrimp vessel in lower and West Galveston Bay during July 1981-June 1982. He identified 56 bycatch species (52 finfish species) from 62 samples (34 trips); bycatch averaged 27.2 percent of total catch weight (range = 17-42 percent). Lamkin observed that five species accounted for about 71 percent of the bycatch by number and 65 percent of bycatch biomass. These species included Atlantic croaker, sand seatrout, blue crab, spot and gulf menhaden. Bessette (1985) accompanied six different bait shrimpers throughout five areas of Galveston Bay during May-November 1984. In 107 tows sampled, Bessette identified 66 species of finfish and eight invertebrates. Bycatch comprised 3-99 percent of total catch by weight with an average of 65 percent. Bessette observed 4.1 kg of fish captured for each kg of shrimp landed. New data collection was initiated in 1992 by NMFS. A total of 296 samples were collected during March-November 1992 and 85 finfish species and 49 invertebrates were identified. Overall, bycatch species comprised 38 percent of the catch by number and averaged 71 percent of total catch by weight. Nine species (of 134 total) accounted for 80 percent of the bycatch by number and 79 percent by weight. These included gulf menhaden, Atlantic croaker, spot, cutlassfish, sand seatrout, bay anchovy, Atlantic brief squid, hardhead catfish and blue crab. Gulf menhaden, Atlantic croaker and sand seatrout were the only species of commercial or recreational value which were captured in great numbers.   Nance, J.M. 1992. Estimation of effort in the Gulf of Mexico shrimp fishery. NOAA Tech. Mem., NMFS-SEFSC-300. Given the size of the shrimp fleet in the Gulf of Mexico, including inshore, nearshore, and offshore vessels, actual documentation of effort is impossible. Boats and vessels are interviewed by NMFS port agents to gather information concerning specific trips in the various area-season-depth matrix. These data are then used to calculate effort for the fleet by dividing the average CPUE (catch per fishing day) of these interview trips into the total landings for the particular region, resulting in an estimate of effort for the fishery. When specific information is lacking for a particular cell in the matrix, the average historical value for that cell is used in a model to estimate the catch for that particular cell for the given time period. From 1969-1989, effort of the Gulf fleet has increased from approximately 125,000 days fished to about 300,000 days fished. The greatest increases are in offshore effort, especially in the north central area (areas 10-12) and off Texas (areas 18-21). Effort in Louisiana shifted in the mid-70's from inshore to offshore while inshore effort in the north central Gulf and off Texas increased. Conversely, inshore effort in Louisiana dropped, but offshore effort increased markedly. (See Griffin and Shah, 1995 for comparison; this section).   Nance, J.M. 1993. Effort trends for the Gulf of Mexico shrimp fishery. NOAA Technical Memorandum, NMFS SEFSC 337. Two types of data are used to estimate shrimp effort: dealer data (landings through a recognized dealer) and interview data (actual interviews with captains following a fishing trip). The fishing trip is considered a unit of effort; in 1992 an estimated 291,954 trips occurred. Interviews have declined substantially from about 20,000 to 6,000 since the early 1980's due to several logistic problems. For offshore efforts, average days fished per trip for all areas have increased since 1980 from 3 - 6 days (a day is towing trawls for 24 hours; this may include several actual days fishing effort {i.e. four, 6-hr tows made over a 2-day period would equal 1 netday}), and CPUE has declined slightly since the early 1980's for all areas. For nearshore areas, days fished have increased over time, driven by increases in Louisiana which logs nearly two to three times as much nearshore effort as the rest of the Gulf combined. Offshore boat and vessel trips have been stable over time, although offshore vessel trips in the eastern Gulf have declined. The data in this report are presented in several subunits by statistical area groups and specific groups of ports.   Nance, J.M. 1993. Shrimp trawl bycatch characterization study.(93NMFS20). NMFS/SEFC Galveston Laboratory, Galveston, Texas. This document presents the results obtained by an onboard observer program. Sixty-seven trips were completed from May 1992 through September 1993. Fourteen of the trips were along the eastern coast of the United States, while the other 53 trips were in the Gulf of Mexico. Trip length varied from 1 - 27 days. A total of 770 sea days were used to collect the data from 1,027 tows. One hundred and forty-five of the sea days were along the eastern coast of the United States, and the other 625 sea days were in the Gulf. Of the 625 sea days in the Gulf, 59 were off Florida, 67 were off Alabama/Mississippi, 340 were off Louisiana, and 159 were off Texas. Thirty-nine different vessels were used in the study. NMFS-approved observers were used to collect the trawl haul subsamples and record the data. Appendix I summarizes the findings by season and statistical area; Appendix II summarizes the data by season, statistical area, and depth; Appendix III gives data for red snapper by season, statistical area, and depth. In the Gulf of Mexico, 10 species accounted for 71 percent of the bycatch by weight and 68 percent by number in the trawls sampled. Dominant among these were Atlantic croaker and longspine porgy. This work was later incorporated into the industry/government cooperative bycatch research program database (see National Marine Fisheries Service, 1995 below).   National Marine Fisheries Service. 1995. Cooperative research program addressing finfish bycatch in the Gulf of Mexico and South Atlantic shrimp fisheries: a report to Congress, April 1995. USDOC, NOAA, NMFS. National Marine Fisheries Service, 9721 Executive Center Drive, St. Petersburg, FL 33702 The document outlines the goals, objectives, and results to date for a federally mandated bycatch reduction research program. Eight program objectives are discussed in detail -- characterization, improved stock assessments, evaluation of bycatch reduction devices (BRDs), non-gear options, management options, information and education programs, identification of other mortality, and development of a centralized database. This booklet provides a good overview of the program and the status of the research. Substantial advances have been made in characterizing the catch through a large-scale fishery-dependent survey. Over 450 taxa have been identified in Gulf of Mexico trawls, with an average catch of 27 kg/net-hour. Shrimp represented 16 percent of the catch by weight; fish, 68 percent. The 150 taxa in South Atlantic shrimp trawls constituted a catch of 29 kg/net-hour of which shrimp represented 20 percent and fish represented 47 percent. These fish to shrimp ratios (4.25:1 for the Gulf and 2.4:1 for the South Atlantic) are much lower than previous (older) estimates of 10:1. Results of two types of bycatch reduction devices (BRDs) are summarized: fisheyes and expanded mesh-extended funnel. Both have minimal shrimp loss with substantial finfish reduction including reductions for key species of concern such as red snapper and weakfish. Additional information is found in this report under "Bycatch Reduction Devices".   Nichols, S., A. Shaw, G.J. Pellegrin, Jr., and K. Mullen. 1987. Estimates of shrimp fleet bycatch for thirteen finfish species in the offshore waters of the U.S. Gulf of Mexico. National Marine Fisheries Service, P.O. Drawer 1207, Pascagoula, MS 39568. This report provided estimates of the offshore catch from 1971-1985 for some commonly occurring fish species, including Atlantic croaker (1-2 billion individuals), spot (150-200 million), longspined porgy (250 million), red snapper (10-15 million), king mackerel (200-250 thousand), Spanish mackerel (ca. 1.5 million), and red drum (120,000), as part of the shrimp catch (100-120 million individuals).   Nichols, S., A. Shaw, G.J. Pellegrin Jr., and K. Mullen. 1990. Updated estimates of shrimp fleet bycatch in the offshore waters of the U.S. Gulf of Mexico 1972-1989. National Marine Fisheries Service, P.O. Drawer 1207, Pascagoula, MS 39568. This report updated an earlier similar report (Nichols et al. 1987), noting an error in the earlier calculations, and provided substantially different values for the species. Total offshore catch of finfish was ca. 400 million pounds annually. The annual catch in numbers for some commonly occurring species included Atlantic croaker (5-6 billion individuals), spot (600 million), seatrouts (two species, 1.5 billion), longspined porgy (1 billion), red snapper (25 million), king mackerel (1 million), Spanish mackerel (3 million), and red drum (20,000), as part of the shrimp catch (100-120 million individuals).   Pellegrin, G. Jr. 1982. Fish discards from the southeastern United States shrimp fishery. pp. 51-54 In: Fish by-catch...bonus from the sea; report of a technical consultation on shrimp by-catch utilization held in Georgetown, Guyana, 27-30 October 1981. FAO and International Development Research Centre, Ottawa (IDRC 198-e). This report divided the Gulf of Mexico into four zones and the South Atlantic into four zones and noted the bycatch taken in each area. Bycatch ratios in the South Atlantic were highest in North Carolina (4:1) and lowest in South Carolina (1.6:1). For the Gulf, bycatch ratios west of Mobile were substantially higher (ca. 15:1) than in the eastern Gulf (5-6:1). The estimate of total discard on an annual basis was 33,000 tons for the South Atlantic and was estimated to be 15 times higher in the Gulf of Mexico due to its larger amount of estuarine-dependent fauna.   Pellegrin, G.J. Jr., S.B. Drummond, and R.S. Ford Jr. (no date). The incidental catch of fish by the northern Gulf of Mexico shrimp fleet. Draft manuscript by the National Marine Fisheries Service, P.O. Drawer 1207, Pascagoula, MS 39568. At-sea observers collected data aboard commercial shrimp vessels during 1972 - 1980. The highest bycatch to shrimp (heads-on) ratio by weight occurred during cool weather (21.1:1), and the lowest occurred in the same area in offshore waters during both cool and warm seasons (2.0:1) {This is confusing in the draft; in the introductory material it notes the area as "eastern Gulf" but later in the document notes the same information related to "area 4" which is all waters west of 92 degrees longitude -- Texas and western Louisiana. This latter region is likelier to be correct}. Annual mean ratio for the area was 10.3:1. Sciaenids dominated the catch at 52.5 percent, with Atlantic croaker at 33.6 percent. The authors estimate that the northern Gulf fleet catches 576,000 tons of fish annually with nearly 80 percent of this caught during the warmer months. Of interest is the note that red snapper comprised less than 0.4 percent of the total finfish weight. This is a similar value to more recent bycatch observer studies.   Perret, W.S., P.E. Bowman, and L.B. Savoie. 1996. Bycatch in the shrimp fishery of Louisiana. pp. 137-143 In: Baxter, B., and S. Keller (eds.). Bycatch: considerations for today and tomorrow. Alaska Sea Grant College Program Report No. 96-03, University of Alaska, Fairbanks. Fishery-independent surveys of Louisiana territorial waters have continued since 1967 using a 4.9 m otter trawl (flat net) towed for 10 minutes at set stations. This data set provides long-term information to assess any changes in abundance. 268 species including 183 fishes, 62 crustaceans, 14 mollusks, and 9 miscellaneous groups comprise the database. Several dominant or fishery important species were examined -- blue crab: high annual fluctuation with long-term trend of abundance nearly doubling over the study period; bay anchovy: high annual fluctuation in abundance with an increasing trend over time; spotted seatrout: catch rates were so low that no trends could be determined, but CPUE was stable over time; sand seatrout: general increase which has nearly tripled the relative abundance of this species in the samples over the study period; Gulf menhaden: long-term trend of slight but steady increase with high levels of fluctuation during the early 1970's through mid-1980's. Two new gears were also examined -- skimmer and butterfly nets. Both have a lower bycatch to shrimp ratio than the standard shrimp otter trawl, and because of the way the catch is handled, much of the bycatch is released alive.   Pueser, R. (ed.). 1996. Estimates of finfish bycatch in the South Atlantic shrimp fishery. Final Report, SEAMAP South Atlantic Committee Shrimp Bycatch Work Group (NOAA Award NA47FS0035), submitted to the Atlantic States Marine Fisheries Commission, 1444 Eye St., NW, Sixth Floor, Washington D.C. 20005. This report summarizes the catch in the South Atlantic shrimp fishery including background information on the biology of the shrimps, the fishery, and the current status of management efforts. Landings data by state along with effort information were used to estimate the bycatch in the fishery based on the NMFS Bycatch Research Program results. Atlantic croaker and spot were the dominant species in most area-season-year analyses, but weakfish were abundant in North Carolina during summer and fall, and in the offshore Florida area in winter. Limitations in this report were noted by the editor as: 1) differences between landings records and observer data as to the definitionof a "trip"; and 2) small sample sizes for many strata. Although the document contains detailed analyses and extrapolations for each area-season-year for the bycatch taken by the fishery, the editor notes that nothing in this report should be used to represent the actual estimate of bycatch in the southeast Atlantic.   South Carolina Department of Natural Resources. 1996. Results of trawling efforts in the coastal habitat of the South Atlantic Bight, FY-1995. SEAMAP-South Atlantic Annual Report. South Carolina Dept. Nat. Resources, P.O. Box 12559, Charleston, SC 29422 Information was collected on the composition, abundance, and biomass of shrimp trawl catches, as well as seasonal and regional trends in environmental parameters during 1995. Sampling collected 202 species. Inner strata sampling of 234 trawl tows produced 186 species of which 142 were fishes. Twenty-seven trawl samples in outer strata produced 135 species, of which 98 were fishes. Spot and Atlantic croaker made up 38 percent of number of individuals and 20 percent of the biomass. White shrimp was third in abundance. Weakfish densities were high (46 individuals/hectare) in Raleigh Bay, but were substantially less for more southerly regions. In South Carolina, Georgia, and Florida, the densities were less than five fish per hectare except for a fall spike of larger fish off Florida (7.78 fish/hectare).   Stender, B.W., and C.A. Barans. 1991. A comparison of the catch from two types of shrimp nets off South Carolina, USA. Saltonstall-Kennedy Program Final Report (GASAFDI # 40-11-44769/22494) (Award #NA90AA-H-SK006) by the South Carolina Wildl. and Mar. Resource Dept., P.O. Box 12559, Charleston, SC 29422. In tests conducted off Charleston, SC, catch rates between a two-seam net and a tongue trawl were compared. The study documented differences in catch rates for eight of 182 taxa collected. Bycatch in the tongue trawl, which has a higher vertical opening, was substantially higher. Catches in both nets were dominated by sciaenids. Bycatch to shrimp ratios in this fishery-independent study were much higher than that documented for the fishery. Substantial catches of Spanish and king mackerels were taken during this study.   Wallace, R.K., and W. Hosking. 1991. Documentation of bycatch from small inshore shrimp vessels and evaluation of appropriate bycatch reduction devices. Final Report (NOAA Grant Award NA90AAH-SK120) by the Auburn Marine Extension and Research Center, 4170 Commanders Drive, Mobile, AL 36615. This report combined field sampling (fishery-independent) with a mail survey to assess the bycatch and effort of the small boat recreational fishery. From the field sampling bycatch to shrimp ratios were nearly 15:1 (range 1.2:1 to 93:1). Four hundred seventy-four surveys were returned (19.5 percent); usable surveys indicated that, in 1990, recreational shrimpers averaged 5.6 trips, 4.3 tows per trip, and 38.2 minutes per tow equaling approximately 40,000 net-hours. Based on their estimates of 16.2 kg of bycatch per net-hour, Alabama recreational shrimping contributed to an estimated 648,000 kg of bycatch and 49,000 kg of shrimp for a 13:1 bycatch to shrimp ratio. Tests of bycatch reduction devices included "fish shooter" (a slit in the bag), and two sizes of "fisheyes" placed on the bottom of the bag. The fisheyes in this configuration reduced fish, but lost 14 and 19 percent of the shrimp. Bycatch from recreational shrimping was estimated from fishery-independent trawling and through a survey of licensed recreational shrimpers in Alabama. The mean fish bycatch was 5.4 kg per 20-minute tow and contained 426 fish primarily from three families (Sciaenidae, Engraulidae, Clupeidae). Based on the survey of recreational effort, the total fish bycatch was estimated at 603,000 kg or 47.6 million fish. Tests of two bycatch reduction devices resulted in significant reduction in bycatch for the Florida Fish Eye, but no significant reduction for the Fish Shooter. Species-Specific Characterization Studies:   Gregory, D.R. Jr. (Draft) An annotated bibliography of literature pertaining to reef fish bycatch in shrimp trawls. Gulf of Mexico Fishery Management Council. 1988. Thirteen documents, both published and unpublished, are listed with brief synopses of their contents. Most of the listed documents are unpublished, but addresses or contacts are listed. Some of the documents are also listed in this report, but copies of the unpublished material, some of which appears to be expanded correspondence to the Gulf Council, were not requested for review and inclusion in this bibliography.   Gutherz, E.J., and G.J. Pellegrin. 1986. Report on snapper-grouper mortality by shrimp trawlers in the U.S. Gulf of Mexico. (Unpublished report to Gulf of Mexico Fishery Management Council.) This document is more formally presented in Gutherz and Pellegrin 1988, listed below. Although the addition of groupers to this report would appear to provide additional information, that information is contained in two sentences on page 7: "Small gag (Mycteroperca microlepis) have been taken infrequently by bait shrimpers in Tampa Bay and other Florida bay systems, but most groupers reside in areas not suitable for trawling. Commercial shrimping activities, therefore, probably exert little influence on populations of Mycteroperca and Epinephelus groupers." Otherwise, Gutherz, and Pellegrin 1988 is a more obtainable reference.   Gutherz, E.J., and G.J. Pellegrin. 1988. Estimate of the catch of red snapper, Lutjanus campechanus. by shrimp trawlers in the U.S. Gulf of Mexico. Fisheries. 50(1):17-25. In an analysis of 1972-1983 resource survey (fishery-independent) and 1972-1981 commercial fishery (fishery-dependent) data sets, this study indicated that resource survey data shows a much higher catch and catch rate of juvenile red snapper than that of actual catch by the shrimp fishery. The two methods both indicated that the majority of snapper are taken from September through November, and catches were primarily west of the Mississippi River delta. By region, highest catch rates occurred off Texas. By depth, highest catch rates were in the 11-20 fathom region; few juvenile red snapper were taken shallower than 10 fm, or deeper than 30 fm. Based on the commercial fishery data, annual catch appeared to be about 5 million juvenile red snapper.   McCarty, G. 1995. Biological benefits of the 200-mile closure for red Snapper and brown shrimp. (GMFMC Briefing Book Addition, Tab l, No. 7). Texas Parks and Wildlife Department, Austin, Texas. The summer closure of the shrimping grounds off Texas has been monitored through a Texas Parks and Wildlife Department (TPWD) sampling program since 1978. From 1977-1980 the closure distance was 9 miles; from 1986-1988, 15 miles; and from 1981-1985 and 1989-1993, 200 miles. Data for this study come from the TPWD standardized fishery-independent monitoring program. The mean number per hour of juvenile red snapper caught in trawls was significantly greater when the 200-mile closure was in effect, showing a greater than 400 percent increase over the years with a 15-mile closure. No significant differences were found when comparing the years following a 9-mile and a 15-mile closure. For the years when the 200-mile closure was in effect there were significant increases in the number of juvenile red snapper in trawl samples and in the number of juvenile brown shrimp in the estuaries.   Nichols, S. 1990. The spatial and temporal distribution of the bycatch of red snapper by the shrimp fishery in the offshore waters of the U.S. Gulf of Mexico. (Unpublished report of the National Marine Fisheries Service, Pascagoula Lab, P.O. Drawer 1207, Pascagoula, MS 39568.) The report provides estimates of red snapper abundance according to an area-season matrix based on fishery-independent and fishery-dependent sources collected from 1972 - 1982. The purpose of the report was to investigate the possibility of area or season closures to reduce the bycatch mortality on red snapper. The general conclusion was that either area or seasonal closures would only transfer the mortality to another cell in the matrix. The report notes that predicting fleet behavior was unsuccessful. The fleet did not respond to regional differences in shrimp catch rates in a predictable manner; cost-benefits ratios and personal preferences may influence the extent of fleet migrations. Thus, without any predictability, such closures would likely have limited benefit.   Shrimp Trawl Bycatch Reduction Implications:   Alverson, D.L., M.H. Freeber, S.A. Murawski, and J.G. Pope. 1994. A global assessment of fisheries bycatch and discards. FAO (Food and Agricultural Organization of the United Nations) Technical Paper 339. Shrimp trawls are only one of many fishery efforts categorized in this report, however they are identified as the dominant source of bycatch and discard in world fisheries. Globally, shrimp fishing is categorized as having a 5.2:1 bycatch to shrimp ratio, with the highest catch ratio from Trinidad at about 15:1. The Gulf of Mexico ranked fifth at 10.3:1. This extensive document is divided into several sections addressing various issues including: estimates of bycatch and discard; biological, economic, socio-cultural, and ecological impacts; a summary of international policies; and a detailed discussion on various options that can help achieve bycatch reduction.   Colura. R.L., and B.W. Bumguardner. 1996. The Texas shrimp industry salt-box catch separation procedure effect on bycatch survival. (MARFIN NA57FF0047). Texas Parks and Wildlife Department, Perry R. Bass Marine Fisheries Research Station, Palacios, Texas. The use of salt-boxes by the Texas shrimping industry to separate bycatch from shrimp was described and bycatch survival evaluated. Commercial and bait shrimpers were interviewed about their use of salt-boxes. Bioassays were conducted for lethal exposure time of important sport and commercial species. Bycatch samples were taken from trawling operations to determine bycatch survival for the salt-box and no salt-box separation methods. Salt-box salinities averaged 67 ppt to which bycatch was exposed an average of 1.7 minutes. Red drum was the most easily affected species requiring 17 minutes of exposure to 70 ppt to kill 59 percent within 48 hours. Survival was mainly affected by "cull" time on the boat deck and length of trawling time rather than the use of salt-boxes.   Fowle, S., and R. Bierce (eds.). 1992. Proceedings of the shrimp trawl bycatch workshop; November 22-23, 1991. Center for Marine Conservation. These proceedings provide relatively complete transcripts of oral presentations made by a variety of speakers on three topics: 1) Effects of shrimp trawl bycatch on finfish populations and ecosystems (Nichols, Browder, Muller, Teehan presenters); 2) Socioeconomic effects of shrimp trawl bycatch (Ward, Griffin, Ditton, Dyer, Margavio presenters), and 3) shrimp bycatch and fishery management (Swingle, Seidel, Apricio, Easley, Gauvin presenters). Other reports covering this material, by many of these presenters, are included in this bibliography. This workshop was held early during the development of the NMFS Shrimp Trawl Bycatch Reduction Research Program, and highlighted the various concerns about bycatch and its implications in fishery management. Many of the presentations discussed how to address the issue, not the results of what had been accomplished. At this stage of the program, little information was available as to the most feasible ways to actually accomplish bycatch reduction.   Griffin, W.L., D. Tolman, and C. Oliver. 1993. Economic impacts of TEDs on the shrimp production sector. Society and Natural Resources, Vol. 6:291-308. A simulation modeling technique is used which estimates the changes in landings, revenues, costs, and the economic rents. A base scenario in which no TEDs are used is compared with five different scenarios where the TED is used by vessels in the Gulf of Mexico. The analysis was based on a single year impact. The implementation of the TED comes with costs to the vessel owners and crew. All other things remaining equal, some vessel owners and crew will leave the industry depending on how successful they are at learning to use the TED effectively.   Gulf of Mexico Fishery Management Council. 1990. Report of the workshop to evaluate potential management alternatives for reducing directed effort and shrimp trawl bycatch of red snapper. A workshop of interested and involved parties was held in Pascagoula, MS, in May of 1990 to address shrimp trawling and red snapper bycatch. Participants concluded that reduction in shrimp bycatch of about 60 percent of the red snapper catch would be required to increase ABC (allowable biological catch) for red snapper fisheries. Options to achieve this reduction included area-season closures and TED modifications. These options are discussed in some detail, especially the various area-season closures that could be implemented.   Hendrickson, H.M., and W.L. Griffin. 1993. An analysis of management policies for reducing shrimp by-catch in the Gulf of Mexico. North American Journal of Fisheries Management 13:686-697. The general bioeconomic fisheries simulation was used to estimate the changes in economic rent and bycatch of red snapper, king mackerel, and Atlantic croaker that would result under two fishery management policies: use of bycatch reduction devices (BRDs) and season-area closures. The BRDs were found to be more effective than closures at reducing bycatch and less costly to shrimpers. Under the BRD scenarios, red snapper discards were reduced 20.2-42.5 percent, king mackerel discards fell approximately 89 percent, and Atlantic croaker discards fell about 45 percent. Under closure policies, the change in discards was a 2.1-15 percent decline for red snapper, a 1.9 percent increase to a 39.3 percent decrease for king mackerel, and a 0.1-12.9 percent decline for Atlantic croaker.   Kennelly, S.J. (draft manuscript). The issue of by-catch in Australia's trawl fisheries. State of the Marine Environment Report for Australia: Technical Annex. Bycatch is defined as two types: interfishery bycatch where an important fishery species is taken as unwanted catch in a fishery targeting another species, and intrafishery bycatch where undersized (or oversized) individuals of the target species are taken. The report notes the need to characterize the catch through fishery-dependent surveys, and develop good stock assessments through fishery-independent faunal surveys. The latter is essential in interpreting the impact of bycatch on the biomass of the "impacted" stocks. Changes occur in faunal composition due to trawling. Therefore, the impacts on benthic communities through habitat alteration must be identified. The report notes that much of the finfish bycatch suffers mortality, but the crustacean bycatch likely survives. Removal of the fish predators on shrimp should positively impact shrimp stocks. On the other hand, discards probably do little to benefit shrimp stocks as they do not prey or scavenge on discards, but other species, such as crabs, sharks, pelagic fishes do; thus, discard may actually benefit the stock of these populations. Better utilization of bycatch is not often possible, but management of bycatch is an increasing concern. The report highlights standard mechanisms such as closures or gear modifications, but notes that regional issues must drive the system to ensure that measures are effective and practical to specific local situations.   Lunz, G.R., J.L. McHugh, E.W. Roelofs, R.E. Tiller, and C.E. Atkinson. 1951. The destruction of small fish by the shrimp trawlers in Pamlico Sound, North Carolina. Report to the Chesapeake Bay and South Atlantic sections, Atlantic States Marine Fisheries Commission. Noting a decline in the catch of several commercial species, this report attributes the declines to bycatch of juvenile finfishes by shrimp trawlers, although it notes that such mortality is only accountable if it contributes to additional mortality from natural causes instead of just replacing natural mortality. Interestingly, bycatch mortality and its effect on commercial finfishes was the concern of this period, and it was written up in several newspapers and other general media outlets. This report debunks some of the hyped-up stories generated by such media coverage. The report concludes that such fluctuations in abundance may or may not be from incidental mortalities, and do happen occasionally. Several appendix documents discuss specific studies: one of note measured fish to shrimp ratios which in July were 2 pounds of shrimp per pound of fish, but by the end of August were 1 pound of shrimp to 3 pounds of fish. A second study listed fish/shrimp ratios at 3:1 to 15:1 in an October sampling period.   Goodyear, C.P. 1992. Red snappers in U.S. waters of the Gulf of Mexico. Contribution MIA91/92-70 of the NMFS Southeast Fisheries Center, Miami Lab, 75 Virginia Beach Drive, Miami, FL 33149. This stock assessment was provided for Gulf of Mexico Fishery Management Council consideration as to options for both the red snapper fishery and the shrimp fishery. The assessment estimated that survival of age-0 and age-I fish to the directed fishery at later ages is reduced by 83 percent because of trawling. As much as one-third to one-half of the age-0 class suffers mortality from trawling. The report notes that without 50 percent reductions in bycatch mortality on these age groups, the snapper fishery cannot continue with current total allowable catches (TACs).   Goodyear, C.P. 1994. Red snappers in U.S. waters of the Gulf of Mexico. Contribution MIA93/94-63 of the NMFS Southeast Fisheries Center, Miami Lab, 75 Virginia Beach Drive, Miami, FL 33149. This stock assessment was provided for Gulf of Mexico Fishery Management Council consideration as to options for both the red snapper fishery and the shrimp fishery. This assessment estimated that survival of age-0 and age-I fish to the directed fishery at later ages is reduced by 82 percent because of trawling. As much as one-third to one-half of the age-0 class suffers mortality from trawling. The report notes that without 50 percent reductions in bycatch mortality on these age groups, the snapper fishery cannot continue with current TACs, especially due to over-harvesting the quotas by both recreational and commercial sectors. These projections are more pessimistic than the previous estimates.   Goodyear, C.P. 1995. Red snappers in U.S. waters of the Gulf of Mexico. Contribution MIA995/96-05 of the NMFS Southeast Fisheries Center, Miami Lab, 75 Virginia Beach Drive, Miami, FL 33149. This stock assessment was provided for Gulf of Mexico Fishery Management Council consideration as to options for both the red snapper fishery and the shrimp fishery. This assessment estimated that survival of age-0 and age-I fish to the directed fishery at later ages is reduced by 80-88 percent because of trawling. As much as one-third to one-half of the age-0 class suffers mortality from trawling. The report notes that without 50 percent reductions in bycatch mortality on these age groups, the snapper fishery cannot continue with current TACs, especially due to excessive over-harvesting by the recreational sector. If the reductions are met, then a quota of approximately 10 million pounds could be taken and still meet recovery target date requirements.   Hoar, P., J. Hoey, J. Nance, and C. Nelson (eds.) 1992. A research plan addressing finfish bycatch in the Gulf of Mexico and South Atlantic shrimp fisheries. Gulf and South Atlantic Fisheries Development Foundation, 5401 W. Kennedy, Tampa, FL 33609. To address the concerns of a multitude of user and interest groups associated with the bycatch issue in the southeastern shrimp fishery, this consensus document expanded upon a "research requirements" document developed by NMFS (1991), and outlined a strategic research plan that would address the bycatch issue in the shrimp fishery. This included eight program objectives: 1) update bycatch estimates; 2) improve stock assessments; 3) identify and develop gear options for bycatch reduction; 4) identify and evaluate non-gear options; 5) evaluate biological, social, and economic impacts of management options; 6) allow for multi-organizational oversight and information transfer opportunities; 7) identify other sources of fishing mortality; and 8) develop a centralized database. These eight objectives included 17 specific tasks that would require 44 different projects to be completed. The funding costs for these objectives, tasks, and projects were estimated, and the priorities for each project were identified.   Jones, R.P. (ed.). 1993. International conference on shrimp bycatch (proceedings) {May 24-27, 1992}. Southeastern Fisheries Association (under NOAA/NMFS Award NA90AAHMF7345), 312 E. Georgia St., Tallahassee, FL 32301. This was one of several conferences that initially addressed the issue of southeast U.S. shrimp trawl bycatch: its quantity, composition, distribution, and impacts on marine resources. The conference brought together scientists, management agencies, industry, and other stakeholders to discuss the bycatch situation worldwide; the focus of most of the presentations was on the southeastern U.S. Presentations from all the represented groups focused on the qualitative and quantitative aspects of bycatch. Concerns focused on defining the goal of bycatch reduction, the quality and quantity of data that existed, how to improve those data through cooperative partnerships, working toward simple solutions, and accepting current reductions through existing gear modifications. Also discussed are some techniques that have been used to reduce bycatch through fishing effort changes, gear modifications, etc.   Martinez, E.X., J.M. Nance, and R.J. Zimmerman. 1996. A model for assessment of ecological interactions among living marine resources in the Gulf of Mexico: implications for bycatch management and shrimp production. Executive summary of a report to the Gulf of Mexico Fishery Management Council. As an update of an earlier modeling attempt by Sheridan et al. (1984), an ecosystem-based model to assess the impacts of bycatch reduction and shrimp production indicated that the release of additional fish which are predators on shrimp may impact the shrimp stock. Only 14 of 161 fish species examined have been identified as predators on shrimp; however these include some of the more abundant species taken in shrimp trawls, including the Atlantic croaker and seatrouts. Sand seatrouts represent the dominant shrimp predator. Using a nitrogen-cycle based model and looking at the northwest Gulf of Mexico (west of Mobile Bay) where the majority of the sciaenid fish predators exist, various scenarios were developed depending on the amount of bycatch reduced. With a 10 percent reduction of all fish species equally by number, shrimp stock would decline only 1 percent (it is important to note here that these shrimp values represent stock of shrimp, not fishery yield), but with a 50 percent reduction in catch of predatory fish, shrimp stock would decline as much as 10 percent. Using the actual reduction values for various finfish species, related to the bycatch reduction gear (BRD) type, a 6-7 percent reduction in shrimp stock would occur for "fisheye" BRDs, and an 8 percent reduction in shrimp stock would occur for the expanded mesh BRDs. Using various predation rates, which change ontogenetically for various fish species, shrimp stock would decline between 8 percent and 17 percent. Lastly, as the fish matured, and their dietary changes moved away from shrimp, the decreased predation would increase shrimp stock by 5 percent. The report concludes that, on average over the last 5 years, shrimp stock has fluctuated naturally by as much as 12 percent, thus the above estimates would fall within normal ranges of production.   Murray, J.D., J.J. Bahen, and R.A. Rulifson. 1992. Management considerations for by-catch in the North Carolina and southeast shrimp fishery. Fisheries 17(1):21-26. The document sets a background using the available characterization studies and the impetus behind the bycatch issue in the southeastern shrimp fishery. It notes that with the possible exception of red snapper and weakfish, there is no conclusive evidence that shrimp bycatch is a biological problem. It does however note that a recent North Carolina study (Miller et al. 1990 - FAO Fish. Biol. Tech. Paper 314) suggests that estuarine species do not demonstrate density-dependent responses to juvenile mortality (increased growth or survival). The estuaries are not saturated with larvae or young; the limiting factor to the area is colonization, thus juvenile bycatch may result in reduced adult populations. The report concludes that managers have several options such as seasonal and area closures or gear modifications and restrictions with which to address the issue. The article is careful to point out that many problems encountered during TED implementation concerning poor user and interest group interaction and communication should be addressed when introducing bycatch reduction to the fishing industry.   National Marine Fisheries Service. 1991. Shrimp trawl bycatch research requirements. USDOC, NOAA, NMFS. National Marine Fisheries Service, 9721 Executive Center Drive, St. Petersburg, FL 33702 In response to the mandate outlined in the 1990 revision of the Magnuson Fishery Conservation and Management Act, NMFS developed a strategic plan outlining a research program to address the bycatch issue in the southeastern shrimp fishery. This initial document summarized what was known about the quantity and composition of bycatch, how and why it was (perceived) an issue, current research on bycatch and its reduction, and the impacts that bycatch reduction would have on shrimp stocks. The document noted the need for a multi-organizational interactive and cooperative effort to address this issue on a region-wide basis.   National Marine Fisheries Service. 1995. Cooperative research program addressing finfish bycatch in the Gulf of Mexico and South Atlantic shrimp fisheries: a report to Congress, April 1995. USDOC, NOAA, NMFS. National Marine Fisheries Service, 9721 Executive Center Drive, St. Petersburg, FL 33702 The document outlines the goals, objectives, and results to date for a federally mandated bycatch reduction research program. Eight program objectives are discussed in detail - characterization, improved stock assessments, evaluation of bycatch reduction devices (BRDs), non-gear options, management options, information and education programs, identification of other mortality, and development of a centralized database. This booklet provides a good overview of the program and the status of the research. Substantial advances have been made in characterizing the catch through a large-scale fishery-dependent survey, and the effectiveness and efficiency of numerous BRDs (over 80 types or configurations) have been evaluated. Two types of BRDs have been identified as meeting program goals: expanded mesh-extended funnel (large meshes which allow escapement surrounding a funnel), and fisheyes (metal-framed cones which provide a permanent hole for escapement). The report also highlights some socio-economic work that has been completed characterizing the fishery and the fishers.   Nichols, S., J. Nance, C.P. Goodyear, A. Shah, and J. Watson. 1995. Some considerations in determining bycatch reduction requirements. (Unpublished report of the National Marine Fisheries Service, P.O. Drawer 1207, Pascagoula, MS.) Based on the requirements outlined in the red snapper stock assessment, a baseline for reduction is established, and current reduction capabilities are examined. Although the majority of red snapper taken are age-0, and are not excluded with any efficiency, the majority of bycatch mortality (vs. natural mortality) occurs on age-1 fish, thus the potential for mortality reduction lies in this less numerous group. The fraction of age-I fish removed (excluded from the gear) is greater, thus the potential to increase stock size. Fifty percent reductions in mortality from the 1982-1986 level of 1.82 are necessary, and according to the research on bycatch reduction gears, this reduction in overall F for age-0 and age-I combined is achieved. The report notes that non-gear options (closures) would only work if there was an actual reduction in fishing effort; any closure which simply moves fishing effort to other red snapper grounds only transfers or delays mortality.   Powers, J.E., C.P. Goodyear, and G.P. Scott. 1987. The potential effect of shrimp fleet bycatch on fisheries production of selected fish stocks in the Gulf of Mexico. NMFS unpublished report, contribution No. CRD-87/88-06 of the Coastal Resources Division, Miami Lab, 75 Virginia Beach Dr., Miami, FL 33149. According to this companion report to Nichols et al. (1987), bycatch reduction has potential for increasing stocks of fishery-important finfish species such as red snapper, the mackerels, and red drum. Reduction of red snapper bycatch has the potential to increase yield by 30-90 percent depending on the level of natural mortality. For Spanish mackerel, the potential is 40-60 percent, and for king mackerel, the potential is 20-30 percent. No specific computations were made for red drum other than to note bycatch reduction of adults vs. juveniles was unknown, and other than noting that bycatch reduction would benefit juvenile recruitment. (NOTE -- the implementation of TEDs in the late 1980's all but eliminated the catch of large red drum). This paper hedges every statement with phrasing such as "given the variability of the estimate" and "due to uncertainty"; apparently although statistically valid in exercise, the results should be considered speculative projections.   Sheridan, P.F., J.A. Browder, and J.E. Powers. 1984. Ecological interactions between penaeid shrimp and bottomfish assemblages. pp. 235-254 In: J.A. Gulland and B.J. Rothschild, (eds.). Penaeid shrimps -- their biology and management. Fishing News Books, Farnham, England. To assess the effects of better utilization of shrimp trawl bycatch, two models were used to evaluate the possible impact on shrimp stocks through reductions in quantity of discards. Elimination of bottomfish discards back to the ecosystem would reduce shrimp stock by as much as 25 percent through reduced nutrients available for the ecosystem and food web. Contrastingly, if bycatch {dead} discard were reduced through gear modifications that reduced the catch (bycatch reduction devices), the resulting shrimp stock reduction would be approximately 8 percent. The report does note that shrimp production is more likely influenced by environmental changes resulting in annual fluctuations in production. Any changes in discard-shrimp interactions would be masked by natural variation.   Texas A&M Sea Grant. 1991. Bycatch - a matter of opinion. Texas Shores 23(3). This Sea Grant quarterly publication is entirely dedicated to the bycatch issue. As noted in an introductory article, it is not a Sea Grant position, but everyone who is interested will find something in the issue to agree with, much to disagree about, and a lot to think about. Articles include information on TEDs to BEDs (bycatch excluder devices), recreational snapper fishing, snapper bycatch in trawls, commercial snapper fishing, the conflicts among various interest and user groups concerning snapper, and the policy issues and who is responsible for addressing the issues.   Thomas, J.S., G.D. Johnson, and C. Formichella. 1996. Bycatch: the social dimensions. University of South Alabama, Mobile, Alabama. (Unpublished at the time of this report). Social scientists conducted a Gulf-wide survey of shrimpers to describe their current social conditions, and discuss perceptions fishermen have about bycatch. Among the more interesting findings, 39.9 percent of the shrimp fishermen interviewed felt they make enough money to support their family under current conditions. This number was reduced to 5.4 percent if bycatch regulations were enacted. When asked about various bycatch regulatory preferences, 15.4 percent of the fishermen preferred closed areas, 11.3 percent preferred closed seasons, 6.1 percent preferred bycatch reduction devices, and 22.7 percent preferred some form of license limitation. One serious shortcoming of this study is that no Vietnamese fishermen were interviewed.   Ward, J.M. 1994. The bioeconomic implications of a bycatch reduction device as a stock conservation management measure. Marine Resource Economics 9:227-240. (A manuscript by this author in press with the Southern Journal of Business and Economics entitled "Static and dynamic implications of a gear modification designed to reduce bycatch in a stylized fishery" was also reviewed. Both documents present similar material.) Based on bioeconomic modeling, bycatch reduction in the shrimp fishery, especially for species of recreational and commercial importance, will not necessarily lead to enhanced fish stocks. All savings accrued from bycatch reduction will be negated after the fish recruit to the directed fisheries. With increased recruitment and availability, catch in the directed fishery will increase, leading to subsequent increased effort by those fisheries. This reallocation of stock harvest, if unregulated, will eventually reduce stock to the previously existing level. This report notes that only if catch and effort in the directed fisheries are regulated can bycatch reduction actually have a beneficial effect on fish stocks. The extensive list of literature cited in this report (includes several not listed in this bibliography) is additionally very good, providing numerous references to all aspects of bycatch, its reduction, and the implications of those management efforts.   Ward, J.M. 1994. Stock conservation implications of proposed bycatch reduction management regulations: social and economic research panel trawl bycatch session. Gulf of Mexico Fishery Management Council. NMFS, St. Petersburg, Florida. This amplification of the above-referenced work makes the following observations in light of bioeconomic modeling principles: 1) Gear modifications should reduce finfish bycatch levels in the shrimp fishery; 2) The increase in vessel operating costs caused by BRD adoption should reduce shrimp fishing effort levels and lead to reduced finfish bycatch levels; 3) The shrimp loss associated with a particular BRD design induces increased shrimp fishing effort levels with concomitant increased bycatch levels; 4) While finfish bycatch levels are reduced, bycatch reduction devices in and of themselves do not result in long run increases in finfish stock sizes; 5) Short run increases in finfish stock size induce increases in recreational and commercial finfish fishing effort levels; and 6) Long run equilibrium stock size returns to its initial equilibrium level with increased commercial and recreational fishing effort levels, increased fishing costs and slightly increased harvest levels.   Non-Gear Shrimp Trawl Bycatch Reduction Efforts:   Dawson, C.E. 1957. Preliminary report on the effects of closing Calibogue Sound, South Carolina, to shrimp trawling. (Typewritten manuscript - apparently to South Carolina Wildlife Resources Commission.) Closure of a traditional shrimp and crab trawl area in 1957 allowed for examination of the effects of no trawling on that area compared to areas which remained open. Catch rates during the one-year closure were compared to the mean catch rates of 1953-1956. The report noted that perhaps the largest variable was the annual fluctuations of local populations of shrimp and crabs (this type of fluctuation probably precludes any good comparison of one year to a four-year average). The author noted that there were no substantial or significant increases in productivity for closed vs. open areas.   Nichols, S., J. Nance, C.P. Goodyear, A. Shah, and J. Watson. 1995. Some considerations in determining bycatch reduction requirements. (Unpublished report of the National Marine Fisheries Service, P.O. Drawer 1207, Pascagoula, MS.) Based on the requirements outlined in the red snapper stock assessment, a baseline for reduction is established, and current reduction capabilities are examined. Although the majority of red snapper taken are age-0, and are not excluded with any efficiency, the majority of bycatch mortality (vs. natural mortality) occurs on age-I fish, thus the potential for mortality reduction lies in this less numerous group. The fraction of age-I fish removed (excluded from the gear) is greater, thus the potential to increase stock size. Fifty percent reductions in mortality from the 1982-1986 level of 1.82 are necessary, and according to the research on bycatch reduction gears, this reduction in overall F for age-0 and age-I fish combined is achieved. The report notes that non-gear options (closures) would work only if there was an actual reduction in fishing effort; any closure which simply moves fishing effort to other red snapper grounds only transfers or delays mortality.   Whitaker, J.D., L.B. DeLancey, and J.E. Jenkins. 1989. A study of the experimental closure of South Carolina's sounds and bays to commercial trawling. Technical Report 72, Comm. Crustacean Mgmt. Sect., Off. Fish. Mgmt., Div. Marine Res., S.C. Wildl. and Mar. Res. Dept. During a two-year study of closed inshore areas, there appeared to be no effect of long-term (55 years) trawling in the areas. Catch rates of most finfish (especially those of recreational and commercial importance) indicated that the stocks appeared to be in relatively good condition. The same was true for white shrimp. Catches and relative abundance of these species compared to areas which had never been open were not different. The conclusion reached in this study reflected that of Dawson 1957 - that stocks of estuarine dependent finfish fluctuate primarily in response to local and seasonal environmental conditions.   Turtle Excluder Devices (TEDs) -Finfish Bycatch Reduction:   Andrew, N.L., S.J. Kennelly, and M.K. Broadhurst. 1993. An application of the Morrison soft TED to the offshore prawn fishery in New South Wales, Australia. Fisheries Research 16:101-111. Comparisons of the catch in a net equipped with a Morrison TED to a net without a TED indicated no significant alteration of the catch of shrimp, but a reduction in the unwanted finfish catch. Total biomass was reduced by approximately 32 percent, or 9 kg/90-minute tow. It was noted that the catch of commercially valuable finfishes was substantially reduced, and that the income earned by fishermen was reduced approximately 4 percent.   Christian P.A., and D.L. Harrington. 1987. Loggerhead turtle, finfish, and shrimp retention studies on four turtle excluder devices (TEDs). pp. 114-127 In: Proceedings of the non-game and endangered wildlife symposium, 8-10 Sept., Georgia Dept. Natural Resources, Social Circle, Georgia. Four TEDs (NMFS collapsible, Georgia, Louisiana, Texas) were tested for their efficiency. There was 100 percent turtle exclusion for all TEDs, and total biomass was reduced from 23-45 percent. The various TEDs had different shrimp retention rates with only one (Texas TED) having a statistically significant 23 percent shrimp loss. Fish exclusions varied by species by TED with the Texas TED reducing the most and the Georgia TED having the least reduction.   Holland, B.F. Jr. 1989. Evaluation of certified trawl efficiency devices (TEDs) in North Carolina's nearshore ocean. Final Report project 2-439-R (funded in part by NOAA, NMFS Award NA87WCD06100), North Carolina Division Mar. Fish., P.O. Box 769, Morehead City, NC 28557 Four different TED designs (2 configurations of a Georgia TED, a Parrish TED, and Morrison TED) were tested for their efficiency at turtle exclusion, finfish exclusion, and shrimp retention. For a 4" Georgia TED, total finfish was reduced about 15 percent while shrimp loss in pounds was about 3-5 percent. With a 2 5/16" grid, this TED reduced finfish by 20 percent and lost 5 percent of the shrimp by weight. The Parrish TED reduced finfish by 75 percent and lost over 50 percent of the shrimp (the report notes no reasons were discerned as to why this TED worked as it did). The Morrison TED reduced finfish and shrimp by about 25 percent.   Kendall, D. 1990. Shrimp retention characteristics of the Morrison soft TED: a selective webbing exclusion panel inserted in a shrimp trawl net. Fisheries Research 9:13-21. Fishery-independent surveys, mimicking commercial operations, tested the Morrison TED for its turtle exclusion and bycatch reduction capabilities. This report analyzed the latter of these concepts. Forty-eight tows were made, and using a minimum shrimp catch of 4.5 kg/hr (as per commercial fishermen's notes that this was an economic minimum), 27 of the tows were used to compare shrimp catch rates and bycatch reduction against a net without a TED. There was no difference in shrimp catch rates when catches exceeded the minimum threshold; total biomass was reduced by 24 percent. The report noted that the Morrison was not a preferred TED at the time, but that with proper (emphasized) installation, it provided optimal results. Fishers were concerned that it excluded many marketable fishes.   Murray, J.D. 1990. Laboratory and field experimentation of three TED designs to eliminate shrimp loss. Final report Saltonstall-Kennedy Grant Program award (S-K NA89WC -H- SK036). UNC Sea Grant College Program, Raleigh, North Carolina. Flume tank testing of scale model nets and TEDs as well field observations of full-scale gear led to the development of TED modifications to help reduce shrimp loss. Although the project was targeted at minimizing shrimp loss, a spin-off publication "Blueprints" from UNC Sea Grant contains a table depicting the differences in total biomass between a TED net and a control (net with no TED) during the tests.   National Marine Fisheries Service. 1993. Cruise results: shrimp trawl bycatch reduction, NOAA Ship Oregon II Cruise 92-05 (201) 09/04-29/92. NMFS Pascagoula Lab, P.O. Drawer 1207, Pascagoula, MS 39568. Three excluder devices were tested: a large mesh surrounding a funnel, a TED with side openings, and a fisheye. The side-opening TED had a 46 percent finfish reduction and an 8 percent shrimp loss. A fisheye had an 7 percent finfish reduction and a 3 percent shrimp gain. The large mesh design was only tested for water flow and performance; it was not compared to other catches. (this is only one of several cruise reports that are available over time on this topic; contact NMFS).   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. NOAA Techn. Memorandum NMFS-SEFC-254. During 3,808 tows, observers onboard commercial shrimp vessels tested two TED types for efficiency. Overall, a 10 percent shrimp loss was found for quad-rigged vessels, and a 2 percent loss for twin-rigged vessels; finfish reduction was about 10-15 percent. For a Georgia TED with a funnel, finfish catch was 3.9 lb per hour, whereas without a funnel it was 12 lb per hour. Additional detailed information is available in this document comparing efforts in the Gulf and the South Atlantic. (results from this document led to Renaud et al. 1992; see below)   Renaud, M., G. Gitschlag, E. Klima, A. Shah, D. Koi, and J. Nance. 1991. Evaluation of the impacts of turtle excluder devices (TEDs) on shrimp catch rates in the Gulf of Mexico and South Atlantic, September 1989 through August 1990. NOAA Techn. Memorandum NMFS-SEFC-288. A mean shrimp loss of 0.7 lb/hr was documented for Georgia and Super Shooter TEDs combined. The Georgia TED gained nominally (0.05 lb/hr) while the Super Shooter lost 0.16 lb/hr; these are minimal differences. Differences in finfish catch was about 10 lb/hr (210 vs. 200 lb/hr), although the results were not significantly different. There was a more substantial loss of shrimp from the pink shrimp fishery; the brown and white shrimp fishery efforts which comprise the bulk of the efforts did not show decreased yield. (results from this document led to Renaud et al. 1992; see below)   Renaud, M., G. Gitschlag, E. Klima, A. Shah, D. Koi, and J. Nance. 1992. Loss of shrimp by turtle excluder devices (TEDs) in coastal waters of the United States, North Carolina to Texas: March 1988 - August 1990. Fish. Bull. U.S. 91:129-137. Three TEDs were tested aboard commercial vessels during normal working conditions. These included a Georgia TED with and without a funnel, and a Super Shooter with a funnel. Both configurations of the Georgia TED lost statistically significant amounts of shrimp; with a funnel the loss was 4 percent, without a funnel the loss was 14 percent. The Super Shooter did not lose statistically significant amounts of shrimp (-1.4 percent). All TEDs lost more shrimp in the Florida area, and the Georgia TED with a funnel also lost shrimp off Louisiana and in all seasons except winter. Without a funnel, the Georgia TED consistently lost shrimp in all areas and seasons.   Vendetti, R.A., R.G. Overman, L.G. Parker, and D.L. Harrington. 1996. Improved methods and procedures for the transfer of technology and the education of constituency groups for devices that will reduce the bycatch in shrimp trawls. MARFIN final report (Award NA57FF0051) by the University of Georgia Marine Extension Service, 715 Bay Street, Brunswick GA 31523 to the National Marine Fisheries Service. Several bycatch reduction devices (BRDs) and turtle-excluder-devices (TEDS) were tested for their abilities to reduce unwanted bycatch from shrimp trawls. BRDs tested included expanded mesh, Kiffe BRD, and fisheyes. Some TEDs were also examined. Tests were conducted in various South Atlantic Bight areas from South Carolina to northeast Florida, and finfish and biomass reductions with the various BRDs were substantial (20-40 percent). Shrimp losses were minimal and not usually significantly different. Only limited numbers of a key species, weakfish, were collected, thus reduction rates (which were not great) may have been more influenced by the scarcity of the species than the gear's ability to exclude them. Each set of tests is reported independently, thus it is hard to present general quantitative results of this study.   Watson, J.W., and C.W. Taylor. 1990. Research on selective shrimp trawl design for penaeid shrimp in the United States; a review of selective shrimp trawl research in the United States since 1973. Proceedings of the Fisheries Conservation Engineering Workshop, Narragansett, RI, April 4-5 1990. (also available through NMFS Lab, P.O. Drawer 1207, Pascagoula, MS 39568). This is a good summary document of the bycatch research that has occurred over time. In the early 1970's, separator panels, used elsewhere, were tested in the shrimp fishery with little success because many of the small fishes were gilled in the apparatus. Shrimp losses were high as well and tests on this design were discontinued. Electrical stimulators were shown to be effective, but the high cost was prohibitive. Developments of the NMFS TED, and subsequent research, led to separation rates of as high as 78 percent in the daytime and 50 percent at night. Other TEDs developed later did not have the same capabilities, but research was continuing on ways to improve their efficiency.   Watson, J.W. 1980, 1981. Sea turtle excluder trawl project; milestone reports. NMFS, Pascagoula, P.O. Drawer 1207, Pascagoula, MS. 39568. These two reports outline the research during 1978 through 1980. Starting with front end deflectors which worked poorly, NMFS developed TEDs in the bags. Focused on the turtle excluding capabilities, these reports also note shrimp loss with the TEDs; in most cases shrimp loss was negligible. Finfish reductions are not reported.   Watson, J.W. 1981, 1983, 1983, 1984. Sea turtle excluder trawl development, annual reports (FY81, FY82, FY83, FY84). NMFS, Pascagoula, P.O. Drawer 1207, Pascagoula, MS. 39568. These reports outline research during each of the fiscal years, noting the progress towards developing efficient TEDs (NMFS design). For the major shrimp grounds, the TED nets caught nominally more shrimp than the non-TED nets. The 1981 report notes little bycatch reduction; either for total biomass or finfish. The FY82 results indicated an approximate 10 percent increase in shrimp catch for the TED nets, and even with modifications for a finfish deflector, little finfish were excluded. For FY83, emphasis was on making the TED smaller and lighter with different construction and material. With additional modifications, finfish reduction was over 50 percent for daytime towing and 10 percent for nighttime; several other modifications such as hummerwires were also examined and show potential. For FY84, further modifications were made to make the TED lighter and less bulky, and to increase finfish reduction after dark, cyalume light sticks were attached to the deflectors; this produced about 50 percent reductions in finfish with nighttime towing.   Watson, J.W., J.F. Mitchell, and A.K. Shah. 1986. Trawling efficiency device: a new concept for selective shrimp trawling gear. Fisheries 48(1):1-9. Trawl-efficiency-devices (TEDs) {which later became turtle-excluder-devices} were tested through both fishery-independent and fishery-dependent sampling on the commercial shrimp grounds. Three TED designs were tested: two collapsible hard TEDs, and a rigid frame TED. The TED itself serves as a mechanical separator for large organisms; the primary target being turtles, but including large fishes such as red drum, sharks, etc. Several variations of designs with additional flaps or leading panels were tested to further evaluate finfish exclusion with these gears. The collapsible steel TED lost a non-significant 2 percent of the shrimp and 51 percent of the finfish, with common species such as Atlantic croaker, spot, butterfish and bumper being excluded at 50-70 percent. A solid fiberglass TED lost a non-significant 5 percent of the shrimp and 53 percent of the finfish, again with common fishes being excluded at rates as high as 70-80 percent. The collapsible version of the fiberglass TED had a nominal shrimp gain, and lost 52 percent of the finfish, with common species being excluded at rates better than 60 percent. Comparative commercial efforts had 1 percent shrimp losses and 30-55 percent finfish losses.   Wenner, C.A. 1987. Results of tests conducted on two different trawl efficiency devices (TED) in South Carolina coastal waters. A final report (to whom unknown). Marine Research Institute, S. Carolina Wildl. and Marine Resources Dept. P.O. Box 12559, Charleston, SC 29412. Several TED tests were conducted during this study. Using 48-foot nets in St. Helena Sound, during four tows, fishes were reduced by about 66 percent, blue crab by 75 percent, without any loss of shrimp using a NMFS TED. In a follow-up study to these preliminary estimates, during less than 60 tows in brown shrimp, the NMFS TED and the Georgia TED were compared against a non-TED net, and against each other. Against a non-TED net, the NMFS TED lost approximately 5 percent of the brown shrimp, and the Georgia TED lost about 16 percent. Although this was quite different, when the two TEDs were tested against each other, the Georgia TED only had about 3 percent less shrimp than the NMFS TED net. Against a non-TED net, the NMFS TED reduced finfish by 55 percent by weight, and the Georgia TED 37 percent. Against each other, the NMFS TED lost 30 percent more fish than the Georgia TED. During 10 tows in the white shrimp season, the NMFS TED caught about 3 percent less shrimp than the non-TED net, and the Georgia TED caught 30 percent less by weight. Against each other, the Georgia TED had 15 percent less white shrimp by weight than the NMFS TED. Finfish were reduced during the white shrimp sampling by 53 percent using the NMFS TED, and by 57 percent with the Georgia TED. This last number was biased in that the net was not rigged with chafing gear, and the dominant species (star drum) could exit through the webbing of the bag. The authors note this was the primary difference in the finfish catch during the Georgia TED tests.   Shrimp Trawl Bycatch Reduction Devices (BRDs):   Bahen, J.J., J.D. Murray, and R.A. Rulifson. 1993. Development and evaluation of finfish separator device and TED combination to reduce bycatch in the shrimp fishery. Final Report, NMFS Award NA17FD0101, by Univ. North Carolina Sea Grant Program, Box 8605, North Carolina State University, Raleigh, NC 27695. A large mesh panel located over a funnel and diamond cut-outs (snake-eyes) over a funnel were tested. These were first examined for their effect on net integrity; after that, field testing monitored the gear efficiency. The diamond mesh BRD reduced fish by 51 percent without a shrimp loss in one trial and by 38 percent in another trial but shrimp loss was 7 percent. In a third test, fish were reduced by 37 percent; no shrimp values are given. The square mesh BRD reduced finfish by 70 percent, but shrimp catches were too low to be representative.   Christian, P.A., D.L. Harrington, D.R. Amos, R.G. Overman, L.G. Parker, and J.B. Rivers. 1993. The reduction of finfish capture in South Atlantic shrimp trawls. Final report of a NOAA/NMFS Saltonstall-Kennedy Award (NA27FD0070) to University of Georgia Marine Extension Service, 715 Bay Street, Brunswick, Georgia 31523. During the study a low profile trawl, three configurations of fisheyes, and three colors of expanded mesh BRDs were tested. The low profile trawl did not show any substantial finfish reductions, although there was good reduction of some species at night, especially those that are more pelagic such as mackerels. For the various BRDs tested, the authors note that an improper TED angle in the "control" net probably allowed for greater exclusion by the TED in that net, thus the values generated here for the "experimental" nets are suspect.   Coale, J.S., R.A. Rulifson, J.D. Murray, and R. Hines. 1994. Comparisons of shrimp catch and bycatch between a skimmer trawl and an otter trawl in the North Carolina inshore shrimp fishery. N. Amer. J. Fish. Management 14:751-768. Brown and pink shrimp catches were better in an otter trawl; in part because of gear problems with the initial skimmer design and because it could not fish in deep water. When the two gears were fished in similar depths the catch rates were more comparable. For white shrimp, the skimmer caught six times more by weight than the otter trawl. The skimmer caught 0.47 kg/min bycatch vs. 0.66 kg/min for the otter trawl. During the brown shrimp efforts skimmer fish to shrimp ratios were 7:1 vs. 8.4:1 for the otter trawl (under differing sampling as noted). During the white shrimp season, the fish to shrimp ratio was 1.4:1 vs. 12.5:1 for the otter trawl. Twelve of 16 finfish species observed for survivability showed increased survival with the skimmer trawl because of shorter fishing times and handling practices on deck. The skimmer does take a larger percentage of pelagic fishes such as menhaden, bluefish, and mackerels because it fishes the entire water column. (More detailed information available in Coale, J.S. 1992. Changes in bycatch using a skimmer trawl in the North Carolina shrimp fishery. MS Thesis, Dept. of Biology, East Carolina University, Greenville, NC 27858.)   Gulf and South Atlantic Fisheries Development Foundation. 1994. Organization and management of a Gulf of Mexico and South Atlantic Ocean fishery bycatch management program (Year II). Saltonstall-Kennedy Grant Program, final report to the National Marine Fisheries Service (Award NA37FD0032) by the Foundation (Ste. 997, 5401 W. Kennedy, Tampa, FL 33609). As part of this grant, observers logged 744 days on 63 commercial fishing trips gathering bycatch data for characterization of the catch and evaluating various BRDs under actual operating conditions. A total of 362 nets were sampled for characterization, and 653 tows compared the catch of a "control" (without a BRD) net to the catch of a BRD-equipped net. The report notes that finfish comprised 67 percent of the catch by weight, while shrimp represented 19 percent. Two BRDs were extensively tested in the Gulf of Mexico: a fisheye placed in the top-center of the bag, 45 meshes back from the start of the bag, reduced total biomass by about one-third, excluded 20 percent of the red snapper, and had a minimal shrimp loss; the extended funnel-expanded mesh BRD excluded more than 25 percent of the red snapper with no shrimp loss, but overall total biomass reduction was limited. In the South Atlantic, the fisheye was tested in two top center positions (30 meshes and 45 meshes from the start of the bag). Results were similar to those obtained in the Gulf of Mexico. Weakfish reductions were good (70 percent) in summer, but declined to 20 percent in the fall. Shrimp loss during brown shrimp season was about 6-7 percent, but declined to nearly zero loss during white shrimp season. Minimal testing in south Florida during pink shrimp season indicated a 7 percent shrimp loss with about 40 percent finfish reduction for a fisheye.   Gulf and South Atlantic Fisheries Development Foundation. 1995a. Continued implementation of high priority objectives outlined in a Gulf of Mexico and South Atlantic fishery bycatch research program. MARFIN Grant Program final report to the National Marine Fisheries Service (Award NA47FF0007) by the Foundation (Ste. 997, 5401 W. Kennedy, Tampa, FL 33609). This report updated the foundation's efforts (see Gulf and South Atlantic Fisheries Development Foundation 1994) to characterize the shrimp trawl fishery catch and to evaluate various BRDs. During 304 tows comparing the catch of a net with a BRD to the catch of a net without a BRD, several various configurations of fisheye shapes and placements were tested; other gears tested included snake-eyes (diamond holes in the net outside a funnel), and versions of expanded mesh. Small fisheyes (ca. 4" high by 7" wide) showed little fish reduction; a medium sized fisheye (5" x 12") at 30 meshes from the front of the bag showed a 23 percent reduction in total biomass, a 4 percent shrimp loss and fish were excluded at various rates; red snapper were reduced by 47 percent by weight. Initial tests of a modified expanded mesh-extended funnel BRD (two additional bars of expanded mesh) indicated a 14 percent total biomass reduction, a 1 percent shrimp loss, and 20-80 percent reductions in finfish. A side-shooting TED had a 15 percent reduction in total biomass, a 3 percent shrimp loss, a 6 percent reduction in red snapper, and 20-75 percent finfish reductions depending on species. The report also summarizes the evaluations of fisheyes and expanded mesh (all configurations combined) over the entire study period. For fisheyes (a total of 341 tows), shrimp loss was 1 percent, red snapper reduction was 27 percent, and total finfish reduction was 33 percent; for expanded mesh (a total of 162 tows), there was no shrimp loss, a 26 percent reduction in red snapper and a 23 percent reduction in total finfish.   Gulf and South Atlantic Fisheries Development Foundation. 1995b. Continued observer coverage of the Gulf of Mexico and South Atlantic shrimp fisheries to characterize the catch and evaluate the efficiency of bycatch reduction devices. Final report to the National Marine Fisheries of a special unallied authorization award (NA47FM0131) by the Foundation (Ste. 997, 5401 W. Kennedy, Tampa, FL 33609). This report furthers the work completed by the Foundation (see 1994 and 1995a) under the bycatch program. During the study, 1,010 tows compared the efficiency of various BRDs which contributed to a total Foundation database of 1,441 tows in the Gulf of Mexico and 542 tows in the South Atlantic. Using the entire database, fisheyes (seven configurations) were determined to exclude finfish by 15-30 percent with less than a 4 percent shrimp loss; red snapper were excluded at 25-40 percent and weakfish by 10-30 percent. Expanded mesh-extended funnel (two configurations) had a 20-25 percent finfish reduction without a shrimp loss; red snapper were reduced by 25 percent and weakfish by 20 percent. Preliminary results on several tests comparing a "naked" net (without a TED) to a TED-net indicated that some TEDs exclude fish well, especially soft TEDs.   Hines, B., S. Coale, R. Rulifson, and J. Murray. 1993. The skimmer trawl in North Carolina estuaries. Univ. North Carolina Sea Grant College Program, publication UNC-SG-93-01 (funded by NOAA/NMFS Saltonstall Kennedy grant program NA90AADSG062). This booklet was developed to convey information concerning the skimmer trawl to the general public and the fishing community. Catches in skimmers were compared to otter trawls, and for white shrimp the skimmer trawl was more efficient. For brown shrimp, the skimmer could not be deployed in deep enough water; catches were 6 percent of the total catch vs. 17 percent for otter trawls. Bycatch was compared between gears as well; fish to shrimp ratio was 1:1 vs. 8:1 for the otter trawl. Skimmer trawl bycatch was more likely to survive because of the way the gear is fished and the catch is handled. The booklet also tells how to build a skimmer and outlines fishing techniques. (The scientific results are better described in Coale et al. 1994.)   McKenna, S.A., and J.P. Monoghan Jr. 1991. Gear development to improve management of commercial fisheries in North Carolina. Saltonstall-Kennedy grant program (award NA90AAHSK052) annual contract report to the Gulf and South Atlantic Fisheries Development Foundation (contract 43-01), Suite 997, Lincoln Center, 5401 W. Kennedy, Tampa, FL 33609. Three bycatch reduction device (BRD) designs were examined in the North Carolina trawl and flynet fishery. These included fisheyes, accelerator funnels, and large mesh in the tailbag. The fisheyes reduced finfish catch by 50-60 percent with minimal shrimp losses. Larger mesh (1" bar) vs. smaller mesh (3/4" bar) did not reduce fish in the trawls, but 1-3/8" bar allowed many smaller fishes out vs. 1" bar in the flynet fishery. Four TEDs were also tested and showed that total finfish catch varied from +35 to -35 percent and shrimp catches varied from +17 percent to -4 percent compared to a net with no TED.   Murray, J.D., S.L. Diamond, and J.J. Bahan. 1994. A program to distribute and evaluate bycatch reduction devices in inshore waters of North Carolina. MARFIN final report (Award NA37FF038) by the University of North Carolina Sea Grant Program to the National Marine Fisheries Service. North Carolina State University, P.O. Box 8605. Raleigh, North Carolina 27695. Following a series of industry-oriented workshops to introduce various bycatch reduction devices (BRDs) to the shrimp industry, four BRD types were distributed to 25 fishermen for their use and evaluation during the 1993 fishing season. These were accompanied by questionnaires which were to be filled out before and after the season. Most fishermen had long-term experience, fished on vessels larger than 35 feet, trawled an average of 130 days per year, and preferred the fisheye type BRDs. The fishermen noted that the device lost shrimp and reduced bycatch by less than 10 percent; however they would use the BRDs during the next shrimping season. To compare the results, four shrimpers were contracted to conduct BRD evaluations with onboard observers with data collection following the Bycatch Program protocols. The results of this study confirmed the initial evaluations by fishers. The report notes that 91 percent of the fishers reported they would use BRDs without regulations in order to address the conservation concept of bycatch reduction.   Murray, J.D., J.L. Gearhardt, R.A. Rulifson, and C.W. Wescott. 1995. Introduction of larger mesh webbing in the belly and wings of traditional shrimp trawls to reduce bycatch in inshore waters. Final report, Saltonstall-Kennedy grant NA37FD0088 by Univ. North Carolina Sea Grant Program, Box 8605, North Carolina State University, Raleigh, NC 27695. Given previous research on large mesh work (see several other citations in this section), large sections of large mesh were installed in a trawl to allow fish to escape. Shrimp loss using this design was substantial, and deemed unacceptable for application to the fishery. However, blue crab escapement was substantial as well as catches of summer flounder. Thus, this design may be applicable for these fisheries, allowing escapement of undersized animals.   National Marine Fisheries Service. 1995. Cooperative research program addressing finfish bycatch in the Gulf of Mexico and South Atlantic shrimp fisheries: a report to Congress, April 1995. USDOC, NOAA, NMFS. National Marine Fisheries Service, 9721 Executive Center Drive, St. Petersburg, FL 33702. The document outlines the goals, objectives, and results to date for a federally mandated bycatch reduction research program. Eight program objectives are discussed in detail - characterization, improved stock assessments, evaluation of bycatch reduction devices (BRDs), non-gear options, management options, information and education programs, identification of other mortality, and development of a centralized database. This booklet provides a good overview of the program and the status of the research. A substantial database including nearly 4,000 commercial shrimp trawl tows has been accumulated. Substantial advances have been made in characterizing the catch through a large-scale fishery-dependent survey. Results of two types of bycatch reduction devices (BRDs) are summarized: fisheyes and expanded mesh-extended funnel. Fisheyes were reported to exclude 33 percent of the fish, 27 percent of the red snapper, and lose 1 percent of the shrimp. Expanded mesh reduced 23 percent of the fish, 26 percent of the red snapper, and had no shrimp loss. No South Atlantic testing evaluations are reported in this document.   Pearce, K.B., D.W. Moye, and S.K. Strasser. 1989. Evaluation of trawl excluder devices in the Pamlico Sound shrimp fishery. Report 88-07 North Carolina Dept. Natural Resources and Comm. Development, Division of Marine Fisheries, Morehead City, NC 28557. Four BRDs were tested in Pamlico Sound: 1) Scottish separator trawl {SST}; 2) bottom- positioned fisheye; 3) Georgia TED; and 4) Parrish TED. The SST separated fish but lost shrimp, the Georgia TED and the fisheye reduced fish without any substantial shrimp loss, and the Parrish TED lost both fish and shrimp. None but the SST reduced the catch of weakfish.   Rogers, D., B.D. Rogers, J.A. de Silva, and V.L. Wright. 1994. Evaluation of shrimp trawls designed to reduce bycatch in inshore waters of Louisiana. Final report MARFIN award NA17FF0375 to NMFS by School of Forestry, Wildl. & Fish., LSU, Baton Rouge, LA 70803. Industry-developed bycatch reduction devices were tested in inshore and nearshore waters of Louisiana. These tests were made with experimental nets compared to "naked" (without TEDs) nets. Over the study period, two configurations of two different BRDs were tested; later in the study expanded mesh configurations, as developed by NMFS, were also tested. The four industry BRDs and the NMFS modifications to the expanded mesh all had good finfish reduction but lost unacceptable amounts of shrimp (15-25 percent).   Rulifson, R.A., J.D. Murray, and J.J. Bahen. 1992. Finfish catch reduction in South Atlantic shrimp trawls using three designs of by-catch reduction devices. Fisheries 17(1):9-19. Three BRDs, all working with large mesh escape openings in the bag (modified Parrish TED, expanded mesh on top of the bag over a funnel, and square-mesh "snake-eyes" around a funnel) were tested. None of the BRDs demonstrated a change in large fish weight compared to the control net. A major drawback to this paper is that it contains numerous statements such as "significant difference...compared to its control...(df =4, F =3.02, P=0.0367)," but no values are ever given in the text for the reader to understand what the catch was, only that it was significantly different. Additional problems included modifications to the BRDs during the survey, which meant that five BRDs were tested, not just three, and comparisons between unmodified and modified gears are reported. The text mentions problems with sampling design where port and starboard nets may not have been calibrated.   Vendetti, R.A., R.G. Overman, L.G. Parker, and D.L. Harrington. 1996. Improved methods and procedures for the transfer of technology and the education of constituency groups for devices that will reduce the bycatch in shrimp trawls. MARFIN final report (Award NA57FF0051) by the University of Georgia Marine Extension Service, 715 Bay Street, Brunswick GA 31523 to the National Marine Fisheries Service. Several bycatch reduction devices (BRDs) and turtle-excluder-devices (TEDs) were tested for their abilities to reduce unwanted bycatch from shrimp trawls. BRDs tested included expanded mesh, Kiffe BRD, and fisheyes. Tests were conducted in various South Atlantic Bight areas from South Carolina to northeastern Florida, and finfish and total biomass reductions with the various BRDs were substantial (20-40 percent). Shrimp losses were minimal and not usually significantly different. Only limited numbers of a key species, weakfish, were collected, thus reduction rates (which were not great) may have been more influenced by the scarcity of the species than the gear's ability to exclude them. Each set of tests is reported independently, thus it is hard to present general quantitative results of this study.   Watson, J.W., and C.W. Taylor. 1986. Research on selective shrimp trawl designs for penaeid shrimp in the United States: A review of selective shrimp trawl research in the United States since 1973. Food and Agriculture Organization of the United Nations, Rome, Italy.   This historical review chronicles some of the early work done in efforts to separate finfish bycatch from shrimp trawls using webbing panels, electric trawls, webbing skylights, and early TEDs. The authors point out that many of the early attempts at finfish separation were abandoned due to unacceptable shrimp loss or prohibitive gear costs. At the time this document was written, the NMFS TED appeared to be the best option available with the authors noting "reduced finfish catches by as much as 85 percent during daytime fishing and 54 percent during nighttime fishing with no significant difference in shrimp catch rates" when comparing a TED-equipped net against a control net.   Watson, J.W. 1989. Fish behavior and trawl design: potential for selective trawl development. pp. 25-29 In: Campbell, C.M. (ed.). Proceedings of the World Symposium on Fishing Gear and Fishing Vessels. Marine Institute, St. Johns, Newfoundland, Canada. The National Marine Fisheries Service began researching the feasibility of separator trawls for the penaeid shrimp fishery of the southeast U.S. in the 1960's. Separator panels, used elsewhere, provided limited success; the small size of the fishes led to gilling in the web panels. Examining fish behavior during TED tests resulted in several concepts that were proposed for further testing. Fishes, in general, orient to flow in a trawl, and swim parallel to moving backgrounds (optomotor response); this may be cued from both visual and lateral line reception. Modified TEDs with webbing panels that led fish to escape openings worked well; fish reductions were 85 percent in daytime and 54 percent at night.   Watson, J., I. Workman, D. Foster, C. Taylor, A. Shah, J. Barbour, and D. Hataway. 1993. Status report on the potential of gear modifications to reduce finfish bycatch in shrimp trawls in the southeastern United States. 1990-1992. NOAA Techn. Mem. NMFS SEFC 327. During 1990-1992, NMFS gear specialists tested 51 BRD conceptual designs for efficiency and functionability. Designs included gears developed by industry, NMFS, and other researchers. The report summarizes the designs for 39 BRDs, and the reduction capabilities of 30 prototypes tested on commercial fishing grounds. Of these, 12 had finfish reductions of 40-60 percent, and seven had shrimp retention rates of 90+ percent. BRD designs of expanded mesh, expanded mesh-extended funnel, HSB (a modified TED with fish exclusion holes in the side), and fisheyes all showed promise for more detailed testing. As much as 30+ percent of the dominant fish species were excluded by all these designs. Good detailed drawings and descriptions of the various BRDs tested are included in the back of this document.   Workman, I.K., J.W. Watson, and C.W. Taylor. 1992. Trawl gear modifications to reduce bycatch in the southeastern United States shrimp fishery (draft manuscript). NMFS, P.O. Drawer 1207, Pascagoula, MS 39568. Beginning in 1990, NMFS began looking at bycatch excluder devices. The most promising designs tested included fisheyes, an expanded mesh around a funnel, and side openings incorporated behind a TED. A fisheye located on the top reduced fish by 68 percent but lost 17 percent of the shrimp. Double fisheyes on the side reduced fish by 56 percent with no shrimp loss. The extended funnel excluded 46 percent of the fish with no shrimp loss, and the side openings behind a TED reduced fish by 43 percent.   Workman, I., J. Watson, D. Foster, C. Taylor, A. Shah, C. Taylor, and J. Barbour. 1994. Status report on the potential of gear modifications to reduce finfish bycatch in shrimp trawls in the southeastern United States. 1993 annual report by the NMFS Pascagoula Lab, P.O. Drawer 1207, Pascagoula, MS. 39568. This report updates Watson et al. 1993 and reports on evaluations of 25 BRD designs. Of these, four (three of which were modified TEDs) showed good fish reduction (>40 percent) without significant shrimp loss. Good detailed drawings and descriptions of the various BRDs tested are included in the back of this document; this is an essential part of the document to avoid future researchers from "reinventing the wheel." TRAP FISHERIES   Fish Traps: Fish traps have been in existence for many years, and their use has increased recently, especially in the eastern Gulf of Mexico. Few comprehensive studies on this method of fishing exist. Studies available indicate that much of the bycatch in this fishery is released alive, thus the impact on the faunal community is not substantial. However, other logistic problems with this fishery, including lost traps and enforcement problems, led the Gulf of Mexico Fishery Management Council to recently approve discontinuing this fishery over the next 10 years (approved at the July 1996 Gulf Council meeting).   Harper, D.E., J.A. Bohnsack, and D.B. McClellan. 1994. Investigations of bycatch from the wire fish-trap fishery in federal waters of southern Florida. Proc. 43rd annual Gulf Caribb. Fish. Inst.:3-25. Observations from 417 commercially fished wire fish traps in 1990 documented the catch and bycatch from these efforts. In 199 completely sampled traps, 84 percent of the total weight and 66 percent of the individuals were landed. A total of 1,884 live fishes were released, and at least 79 percent of these swam down immediately; the fate of the rest was not documented, thus survival could have been higher. In addition, 1,080 crustaceans and 15 mollusks were released; 68 percent of this catch was spiny lobster released during the closed season for this fishery (most were noted as legal and no egg-bearing females were seen). Discard mortality was recorded for 101 fish. Sublegal fish releases included 144 snapper and grouper (14 percent by number of all captured fishes for which size limits exist).   National Marine Fisheries Service. 1995. Characterization of the reef fish fishery of the eastern Gulf of Mexico. A report to the Gulf of Mexico Fishery Management Council, Reef Fish Management Committee. Prepared by the Galveston and Miami NMFS Laboratories. Available through Gulf of Mexico Fishery Management Council, Ste. 331, 5401 W. Kennedy, Tampa, FL 33609. Observers aboard commercial fish trap and bottom longline vessels off Florida documented the catch and bycatch of the fisheries. For the fish trap fishery, observers aboard 12 trips on six vessels monitored 517 sets totaling 10,654 traps; 36 percent of the traps (3,867) were processed by the observers. A total of 15,148 fish of 63 taxa were documented. Approximately 55 percent of the individuals were released alive, 35 percent were kept, 7 percent were retained for bait, and 2 percent were released dead or with an unknown fate. Seven species comprised 88 percent of the 8,400 fish released, and the majority were red grouper. Although red grouper is the primary target fish, more lane snapper were caught, and they accounted for 34 percent of the fishes kept, followed by red grouper at 22 percent. Of the 242 dead discards, 45 percent were red grouper. 75 percent of the red grouper were undersized (< 20"). These data were compared to previous fish trap studies in Florida, where from 50-80 percent of the bycatch was released, and 2-20 percent was dead. The observer data were compared to logbook data representing a total of 1,168 fish trap trips. Five species comprised 78 percent of the total landings, led by red grouper at 43 percent.   Sutherland, D.L., and D.E. Harper. 1983. The wire fish trap fishery of Dade and Broward counties, Florida. FL Mar. Res. Inst. Publ. 40. During 45 commercial trips, 538 traps were hauled and sampled. A total of 5,984 fish were caught; 62 percent by number and 77 percent by weight were target species. Snapper and grouper species dominated; the 10 most abundant target species comprised 51 percent of the total catch and the 10 most abundant non-target species comprised 13 percent. A total of 478 fish died in the traps. Approximately 20 percent of the traps are lost annually, but other work suggests that about 50 percent of the fish can escape by 14 days; at that point ingress equals egress.   Taylor, R.G., and R.H. McMichael, Jr. 1983. The wire fish-trap fisheries in Monroe and Collier counties, Florida. FL Mar. Res. Inst. Publ. 39. Over a one-year period, 1,694 trap hauls caught 10,226 fishes in Monroe County (MC) and 270 trap hauls in Collier County (CC) produced 3,111 fishes. Target species constituted 51 percent of the catch in MC and 70 percent of the weight; in CC the targets comprised 30 percent of the catch by number and 70 percent by weight. Five species comprised 44 percent of the catch by number; serranids comprised 50 percent of the weight and 15 percent of the number of fishes taken. Four percent of the fish were dead or injured. Of 619 released fish, 53 percent swam down within one minute; 20 percent died, and 27 percent had done neither at the end of one minute.   Crab Trap Fisheries: Crab and lobster trap fishing in the Gulf of Mexico is common, with strong recreational and commercial components. As with fish traps, regular monitoring of the traps produces a live release for the majority of the bycatch, including sublegal target species. The major concern is lost traps that continue to "ghost" fish. Recent studies have been completed, and some controls have been implemented, concerning escape vents and biodegradable materials. These measures allow the escape of sublegal target species and non-targeted species, and disable the gear from continuing to trap organisms after an extended period in the water. Comprehensive studies are available concerning these issues in both fisheries.   Arcement, G., and V. Guillory. 1993. Ghost fishing in vented and unvented blue crab traps. Proc. La. Acad. Sci. 56:1-17. Mortality in unvented traps was approximately three times higher than in vented traps due to the higher number of sublegal crabs.   Bishop, J.M. 1983. Incidental capture of diamondback terrapin by crab pots. Estuaries 6(4):426-430. During tests on pre-molt blue crabs in South Carolina, diamondback terrapin captures in crab traps were documented. Males outnumbered females by greater than 2:1, perhaps because large females could not enter the traps. Most (over 80 percent) were captured in April and May in a wide range of salinities and temperatures. Based on the catch rates, and the estimated number of pots fished in South Carolina waters, along with the estimated 624,000 pots being fished in the U.S. at the time, the author estimates that blue crab traps may account for the majority of adult terrapin mortality.   Eldridge, P.J., V.G. Burrell Jr., and G. Steele. 1979. Development of a self-culling blue crab pot. Mar. Fish. Rev. 41:21-27. In tests in South Carolina, crab pots with two escape ports of 2.5" diameters in the top chamber and one in the bottom chamber reduced the catch of sublegal crabs by 82 percent. Use of these vents was recommended as a management measure in blue crab fisheries.   Guillory, V. 1989. An evaluation of different escape vents in blue crab traps. Proc. La. Acad. Sci. 52:29-34. Square and circular escape vents were tested in Louisiana estuarine crab traps. Both shapes worked equally well (80 percent reduction of sublegals), but square vents were noted as easier to construct and install.   Guillory, V. 1990. Effects of escape vents on catch rates of pre-molt blue crabs (Callinectes sapidus). Proc. La. Acad. Sci. 53:20-28. Premolt crabs (peelers) were reduced in vented traps by approximately 80 percent; similar to sublegal reductions.   Guillory, V. 1993. Ghost fishing by blue crab traps. N. Amer. J. Fish. Mgmt. 13:459-466. 25 traps were baited and after three days, all 307 crabs in the traps were examined and tagged. An average of 13 crabs were caught per trap, and an additional 35 crabs recruited to the traps over a year period (with no bait). Approximately 55 percent of the crabs died. Mortality was highest during the first month of ghost fishing. Smaller crabs escaped more effectively from the entrances. Fish bycatch in the traps included 190 individuals of 11 species; sheepshead were 74 percent of the total followed distantly by spot, southern flounder and Atlantic spadefish each at 5 percent.   Guillory, V., and J. Merrell. 1993. An evaluation of escape rings in blue crab traps. Louisiana Dept. Wildl. and Fish. Techn. Bull. 44. P.O. Box 189, Bourg, Louisiana 70343. Escape ring size and location were evaluated in Louisiana estuaries. Sublegal crab reductions increased with number of rings. Three rings of a minimum of 6.03 cm in diameter were recommended; this produced reductions of 75-80 percent.   Guillory, V., and S. Hein. 1995. A review and evaluation of escape vents in blue crab traps. A manuscript report to Louisiana Department of Wildlife and Fisheries, P.O. Box 189, Bourg, Louisiana 70343. To reduce undersized catches, requirements for escape rings have been incorporated into numerous management strategies. Based on results of the study, three 2-3/8" diameter rings located on vertical outside walls of the upper chamber were recommended. Benefits of the rings include: 1) a reduction of 75-80 percent of undersized catch; because of differences in the shape of the carapace, small legal female crabs will have higher escapement rates too; 2) delayed increase in catch of legal crabs due to decreased handling mortality on sub-legal crabs; 3) decreased injury and stress to sublegal crabs; 4) reduction of ghost fishing; 5) reduced culling time; 6) enforcement reduction in effort; and 7) reduction in sale of small crabs. Disadvantages are the loss of small legal females, and loss of peeler crabs. This paper also has a good review of other related studies; many of which are summarized below or in this document.   Casey, J.F., S. Doctor, and A.E. Wesche. 1992. Manuscript on file at Maryland Dept. Nat. Resources, Tawes St. Office Bldg, Annapolis, MD 21401. Different sizes and placements of vents provided for sublegal reductions of as much as 50-90 percent   Virginia Marine Resources Commission. 1994a, 1994b. Manuscripts on file at Va. Marine Resources Commission, P.O. Box 756, Newport News, VA 23607. In unvented traps, 34 percent of the small crabs could escape, while in vented traps 50 percent of sublegal crabs escaped. Based on measurements and sexes of crabs collected from commercial gear, approximately half the immature females, over one-third of the sub-legal males, and 18 percent of the legal mature females could escape through a 5.87 cm vent. Reduction of legal female escapement was accomplished with a vent of only 5.56 cm.   Whitaker, D.K. 1978. Data report for escape ring study. Manuscript on file at S. Carolina Wildl. and Mar. Resources Dept., P.O. Box 12559, Charleston, SC 29412. Two sized rings (6.03 and 6.35 cm) were tested in South Carolina; sublegal catch was reduced by 75 and 83 percent respectively.   Whitaker, D.K. 1980. Escape ring project - summer 1979. Manuscript on file at S. Carolina Wildl. and Mar. Resources Dept., P.O. Box 12559, Charleston, SC 29412. Escape rings were tested in both high and low salinity waters of South Carolina; sublegal catch was reduced by 90 percent in both areas, and legal catch increased by 12 percent in low salinity, but decreased substantially (42 percent) in high salinities due to the loss of legal females.   Steele, P., and H.M. Perry (eds.). 1990. The blue crab fishery of the Gulf of Mexico, United States: a regional management plan. Publ. No. 21 of the Gulf States Marine Fisheries Commission, P.O. Box 726, Ocean Springs, Mississippi 39546. Although not specifically a bycatch document, this document provides detailed information on a state-by-state basis for the blue crab fishery of the Gulf of Mexico, including management considerations for reducing sublegal take, and bycatch reduction in general. Related Material From Outside The Region:   Breen, P.A. 1987. Mortality of Dungeness crabs caused by lost traps in the Fraser River estuary, British Columbia. N. Amer. J. Fish. Mgmt. 7:429-435. Simulating lost traps, traps were baited then left in place for a year. Overall, 169 Dungeness crabs were caught and monitored; nearly all were male and the majority died, including legal and sublegal crabs. Based on an estimated 11 percent loss rate annually in the fishery, losses of resources due to ghost fishing might equal 7 percent of the Fraser River District catch. Use of escape mechanisms was recommended.   Gagnon, M., and M. Boudreau. 1991. Sea trials of a galvanic corrosion delayed release mechanism for snow crab traps. Rep. 1803, Can. Techn. Rep. Fish. Aquat. Sci. Galvanic releases were tested in conditions of the St. Lawrence snow crab fishery. Results showed that predictable releases could be obtained, and accidental release was unlikely. Authors' note: In another unpublished report by Boudreau, these were tested on 333 traps in the fishery; none released before the predicted time, and 89 percent of the crabs escaped within two weeks after the release opened.   High, W.L. 1976. Escape of Dungeness crabs from pots. Mar. Fish. Rev. 38(4):19-23. Good escapement of females and males, legal and sublegal, occurred from disabled (open triggers and escape rings), but was limited for operational traps. The latter was especially true for larger crabs. At the end of 74 days, 21 percent of the large and 67 percent of the small crabs had escaped. When the triggers were released, most left alive escaped within three days. The results strongly suggested that the use of triggers reduces escapement.   Kimker, A. 1990. Biodegradable twine report to the Alaska Board of Fisheries. Regional information report 2H90-05. Alaska Dept. Fish Game, 333 Raspberry Rd., Anchorage, Alaska 99518. This report, noted as containing interim and thus preliminary data which might change, summarized tests on various sizes of twine used in the crab fisheries. A recommendation of reducing twine size to 30 thread was made, and the report noted although there was no statistical difference in breaking time between this and much larger thread twine, on average, the release might be 6-18 days sooner.   Kruse, G.H., and A. Kimker. 1993. Degradable escape mechanisms for pot gear: a summary report to the Alaska Board of Fisheries. Regional information report 5J93-01, Alaska Dept. Fish. Game, P.O. Box 25526, Juneau, Alaska 99802. This report, noted as containing interim and thus preliminary data which might change, summarized tests on four release mechanisms: 1) cotton twine; 2) galvanic time releases {GTR}; 3) cotton twine or GTR; and 4) GTR after an initial twine/GTR phase-in period. Cotton twine was less expensive but less predictable, and GTR while being more predictable were probably harder to enforce due to differences in construction. The recommendation was some kind of a release mechanism, with an optimum 30-day breakage, although a longer time could be used as a trade-off for the unpredictability.   Muir, W.D., J.T. Durkin, T.C. Coley, and G.T. McCabe, Jr. 1984. Escape of captured Dungeness crabs from commercial crab pots in the Columbia River estuary. N. Amer. J. Fish. Mgmt. 4:552-555. Because of ghost fishing concerns, escapement rates were examined during a 28-day experiment. Sixty percent of the crabs escaped; size, leg loss, and leg regeneration were not factors in escapement.   Paul, J.M., A.J. Paul, and A. Kimker. 1993. Starvation resistance in Alaskan crabs. Interim report. Alaska Dept. Fish Game, 333 Raspberry Rd., Anchorage, Alaska 99518. This report, noted as containing interim and thus preliminary data which might change, noted that starvation for as much as 30 days impacted subsequent survival of tanner crabs even though feeding was reintroduced. This has implications for developing release mechanisms on traps.   Paul, J.M., A.J. Paul, and A. Kimker. 1993. Tests of galvanic release for escape device in crab pots. Interim report. Alaska Dept. Fish Game, 333 Raspberry Rd., Anchorage, Alaska 99518. This report, noted as containing interim and thus preliminary data which might change, indicated that the use of galvanic release mechanisms had a linear decay rate over time, and thus could be fine-tuned for use in Alaska pot fisheries so that they separate at any set time interval needed to reduce crab mortality from lost gear.     Lobster Trap Fisheries:   Hunt, J.H., R.D. Bertelsen, C. Cox, T.R. Matthews, and W.C. Sharp. 1995. Commercial and recreational harvest of the spiny lobster, Panulirus argus (Latreille), in Florida during the 1993-94 season. A report to the Florida Marine Fisheries Commission by the Florida Department of Environmental Protection, South Florida Lab, 2796 Overseas Highway, Suite 119, Marathon, FL 33050. This report is a detailed characterization of the fishery; bycatch monitoring was only a part of the study. Monitoring among nine areas, about 400 traps per area from at least three different fishers were sampled on a monthly basis. Contents of 100-150 traps per trip were sampled. All totaled, 192 trips with 116 fishers monitored the catch of 21,309 lobster traps. Bycatch included 43 species or groups which occurred in more than 0.1 percent of the traps; 45 other species were rarer than this occurrence. Stone crab was the most abundant and valuable species. Grunts, grouper, snapper, and spider crabs were also taken and occasionally harvested. Also taken and sold were aquarium trade fishes and invertebrates. Only 28 dead fish and invertebrates were observed.   Matthews, T.R., C. Cox, and D. Eaken. 1994 (draft manuscript). By-catch in Florida's spiny lobster trap fishery. (submitted to Gulf and Caribbean Fisheries Institute for publication in the proceedings of that organization). (formal publication of the Hunt et al. 1995 report). An analysis of 21,309 lobster traps found 6,972 traps with 172 species of organisms besides lobsters. Lobster catch was consistently lower in traps that contained bycatch. Wooden traps (90 percent of the fishery) did not have as many fish as wire-reinforced traps; the latter caught more commercial fishes, aquarium trade fishes, and incidental species, although this may be because they were fished in deeper waters than wooden traps. All totaled, 20 dead crabs, 8 dead fish, and 5 dead cormorants were observed in traps. Bycatch was retained for sale or released alive. Related Material From Outside The Region:   Blott, A.J. 1978. A preliminary study of timed release mechanisms for lobster traps. Marine Fish. Rev. 40:44-49. Release mechanisms for New England lobster traps were examined to reduce ghost fishing. In simulated fishing efforts, several options for a degradable release mechanism were tested; jute twine and steel wire were recommended for further study in actual fishing operations.   Pecci, K.J., R.A. Cooper, C.D. Newell, R.A Clifford, and R.J. Smolowitz. 1978. Ghost fishing of vented and unvented lobster, Homarus americanus, traps. Mar. Fish. Rev. 40:9-24. Unvented, sublegal escape vent, and escape panels were tested for American lobster traps that were tended or left as "ghost" traps. Vented traps caught fewer and larger lobsters. Ghost pots selectively caught large lobsters. Approximately 25 percent of the ghost trap lobsters suffered mortality. Bycatch of numerous species groups was recorded, but was dominated by cancer crabs (Cancer spp.) in hauled pots; ghost traps had little bycatch.   Smolowitz, R.J. 1978. Trap design and ghost fishing: discussion. Mar. Fish. Rev. 40:59-67. An estimated 2 million pots were fished for New England lobsters in 1973, and estimates of lost pots were 20-30 percent annually; all lost pots are considered actively fishing units due to their construction. To be disabled, the wooden pots must be damaged by wood-boring organisms. Escapement of lobsters from ghost gear was estimated at 30 percent. Ghost pot mortality in 1976 was estimated at over 1.4 million pounds. No degradable sections were required at the time, but were recommended; however these would produce long-term (one-year) disablement. It was estimated that sublegal escape vents would reduce mortality by over one-half million pounds annually. (See also Smolowitz paper in same issue -pp. 2-8)     PELAGIC LONGLINE FISHERIES Pelagic longlines, especially those targeting tuna and swordfish, have recently been scrutinized for their unavoidable and unwanted bycatch of prized recreational marlins and other billfishes. Pelagic longlining is not a species-specific effort, thus the definition of bycatch in this fishery becomes a little more arbitrary. Numerous marketable species may be taken, even if only one species is being targeted. Differences in fishing practices also change the target species. For example, tuna fishing efforts are usually daytime sets, but swordfishing is done at night. However, some tuna may be taken on swordfish sets, and some swordfish may be taken on tuna sets. Thus, although perhaps not targeted, these species would classify as bycatch; on the other hand, given their desirability, these types of catches should not be considered bycatch. Perhaps a more appropriate measure of "bycatch" in this fishery is the quantity that is discarded dead due to either no market value or regulatory restrictions. This latter category, although providing for an overall beneficial concept to stock management, does on the surface appear to be wasteful. Substantial fishery-dependent and fishery-independent data are available to characterize the catch of the various fishing efforts, but actual measures of bycatch and the impact of fishing on bycatch have only recently begun to be addressed. Available information suggests that the discard, whether voluntary or regulatory, is less than 20 percent of the catch, and about half is released alive. This, of course, varies by species and region.   Anderson, E.D. 1985. Analysis of various sources of pelagic shark catches in the northwest and western central Atlantic Ocean and Gulf of Mexico with comments on catches of other large pelagics. NOAA Techn. Rep. NMFS 31:1-14. Primarily a paper dealing with quantifying the volume and number of shark catches and landings from various sources, this document does note that in one survey, sharks were 234 percent of the swordfish catch north of Cape Hatteras, 296 percent of the catch from the southeast U.S. Atlantic, and 213 percent of the catch from the Gulf of Mexico. These data were then used to extrapolate the total shark catch based on the swordfish landings. The document is a good historical summary of shark catches and landings from a variety of fishery sources.   Berkeley, S.A., and W.L. Campos. 1984. Shark by-catch in the Florida swordfish longline fishery. A final report to Florida Sea Grant. This grant report was more formally presented in Berkeley and Campos 1988. Some interesting information is included here that is not in the 1988 paper, but no additional information on bycatch, release/discard rate, etc. was mentioned.   Berkeley, S.A., and W.L. Campos. 1988. Relative abundance and fishery potential of pelagic sharks along Florida's east coast. Mar. Fish. Rev. 50(1):9-16. Catch rates of sharks were examined during 111 commercial swordfish longline sets in the southeast Atlantic and Florida Straits. Six hundered, thirteen sharks of 13 species were taken along with 523 swordfish; 86 percent of the sharks were three species; 66 percent of the sharks brought alongside were dead after overnight soaks.   Branstetter, S. 1981. Biological notes on the sharks of the north central Gulf of Mexico. Contrib. Mar. Sci. 24:13-34. During 69 fishery-independent shark longline sets, this survey caught 381 sharks of 15 species and 22 teleosts of 8 species on 6,476 hooks. Non-targeted teleosts comprised 5 percent of the catch by number.   Branstetter, S., and G. Burgess. 1995. Bycatch on directed shark longlines January 1994 - June 1995. A memo provided to "An industry workshop addressing bycatch issues in southeastern U.S. fisheries." Copy on file at the Gulf and South Atlantic Fisheries Development Foundation, 5401 W. Kennedy, Tampa, FL 33609. For 144,716 hooks of commercial shark longline gear, 7,688 sharks were caught and 317 other fishes were taken along with 15 sea turtles. All turtles were released alive. Totals of 104 marketable fish of 6 species were harvested; all other bycatch was released or discarded. Specific information on percentages released/discarded not available. (For shark catches and discards, see Gulf and South Atlantic Fisheries Development Foundation, 1996; this section.)   Guitart-Manday, D. 1975. Los pesquerios pelagico - Oceanicas de corto rodio de accion en la region nortoccidental de Cuba. Cuban Sci. Acad. Oceanogr., Inst. Oceanogr. Serv. 31 (translated U.S. Dept. Com document NMFSTT7755012). In the short-range Cuban longline fishery, sharks comprised 41 percent of the catch, billfish 33 percent, and swordfish 26 percent.   Hoey, J.J. 1992. Bycatch in U.S. Atlantic longline fisheries for swordfish and tuna. pp. 61-70 In: Schoning, R.W., R.W. Jacobson, D.L. Alverson, T.H. Gentle, and J. Auyong (eds.). Proceedings of the national industry bycatch workshop, Feb. 4-6, 1992, Newport, OR. Natural Resource Consultants, 4055 21st W., Seattle, WA 98199. In a summary of longline efforts, this report notes that bycatches were reduced in the 1970's when rope/steel (Yankee) gear was replaced by the more efficient monofilament gear. That bycatch taken was more likely to be alive using the latter gear as well, and thus could be released. From an observer program and logbooks, bycatch or catch and release information were -- northeast U.S.: 54 percent retained, 36 percent released, and 10 percent discarded dead (this last category included shark damage or size/quota restrictions); six species comprised 98 percent of the retained catch -- Grand Banks: marketable fish constituted over 80 percent of the catch, and the bycatch was primarily blue shark which are usually released alive -- southeast U.S.: 81 percent was marketable in one data set, and two others indicated that approximately 33 percent to 66 percent of the catch was shark of which in one study, 66 percent was dead -- Gulf of Mexico: in one data set, about 25 percent of the total catch was non-targeted swordfish/tuna species; no disposition data on released or retained was available; in another study of 2,185 fish caught, 69 percent was retained, 23 percent discarded dead, and 8 percent was released alive; in older data sets concerning Yankee gear, swordfish were about 24 percent of the catch, sharks 73 percent of the catch; most likely the sharks were discarded both dead and alive.   Hoey, J.J. 1996. Bycatch in western Atlantic pelagic longline fisheries. pp. 193-203 In: Baxter, B., and S. Keller (eds.). Solving bycatch: considerations for today and tomorrow. Alaska Sea Grant College Program Report No. 96-03, University of Alaska, Fairbanks. This report summarizes the catches of over 5,000 Japanese longline sets in the northwest Atlantic and 1,500 observed U.S. longline sets. Target species vary, but focus on tuna and swordfish; bycatch consists primarily of marlins and sharks with a few miscellaneous bony fishes taken. In the Caribbean, 60 percent of the catch was retained, 17 percent was released alive, and 23 percent was discarded dead. For the Gulf of Mexico, for U.S. fishing efforts, 51 percent was retained, 31 percent was discarded dead, and 18 percent was released alive. For Japanese efforts in the Gulf in the 1970's, approximately 50 percent of the catch was retained (tunas) and other species were required to be released or discarded; 19 percent of the swordfish were alive, 44 percent of the marlin were alive, and over 75 percent of the sharks were alive when released. In another Gulf survey where live bait was used, 69 percent was retained, 23 percent was discarded dead, and 9 percent was released alive. In the southeast U.S., 48 percent was kept, 18 percent released alive, and 34 percent was discarded dead. In the northeast U.S., blue sharks accounted for 19 percent of the total catch, and most were released; overall, 49 percent was retained, 35 percent released, and 15 percent discarded. For the Grand Banks area, blue sharks were dominant at 44 percent of the catch and most were alive and released; 36 percent of the catch was retained, 34 percent released, and 30 percent discarded. For the current U.S. fishery, a total of 1,523 sets accounting for the capture of 50,104 individuals had over a 50 percent survival rate at haulback. Overall, dead discard is about 20-30 percent by number, and includes shark- and whale-damaged fish which may account for as much as 5 percent of the total tuna/swordfish catch. Releasing dead fish (undersized swordfish and all marlin) is wasteful.   Hoey, J.J., and J.G. Casey. 1983. Distribution and relative abundance of sharks in the western North Atlantic as indicated by longline catch data. Manuscript noted as MARMAP Contribution MED/NEFC 83-14. Using 2,700 pelagic longline sets, the relative abundance of nine shark species is discussed. Total catch was 105,123 teleosts and sharks, including 27,140 swordfish, 4,366 tuna, 544 billfish, and 71,827 sharks. Of the sharks, nearly 40,500 were blue sharks, and in other Hoey papers listed here, he notes over 90 percent of these are released alive.   Hoey, J.J., and J.G. Casey. 1984. Shark longline fisheries: gear and production characteristics. Proc. 37th Gulf and Caribbean Fisheries Institute:169-186. Using a large-scale database from a variety of fishery-dependent and independent sources, this paper summarizes the species composition of catches. In the Japanese tuna fishery, of 22,316 fish caught, approximately 13,000 tuna, 1,600 swordfish (regulatory discard), 1,500 billfish, and 3,000 sharks were the majority of the catch. In the U.S. swordfish catch, out of 19,200 fish caught, species composition was dominated by 5,000 swordfish and 13,000 sharks; tuna and billfish together were about 250 individuals. No specific disposition of the catches was noted.   Gulf and South Atlantic Fisheries Development Foundation. 1996. Commercial shark fishery observer program. A final report to the National Marine Fisheries Service for MARFIN award NA47FF0008 submitted by the Foundation, Suite 997, 5401 W. Kennedy, Tampa, FL 33609. During 1994 and 1995, observers monitored 96 fishing trips where 276 commercial shark longlines were fished. Nearly 11,000 sharks were documented. Of 7,836 large coastal sharks caught, 5,979 (76 percent) were retained for sale, 563 (7 percent) were discarded, and 1,294 (17 percent) were released alive. For small coastal species, 3,037 individuals were taken, 1,773 (58 percent) were marketed, 1,232 (40 percent) were discarded (primarily used for bait), and 32 (1 percent) were released alive. (For finfish bycatch records, see Branstetter and Burgess 1995; this section.)   Kawaguchi, K. 1974. Exploratory tuna longline fishing in the Caribbean and adjacent waters. Mar. Fish. Rev. 36(9):61-66. Exploratory fishery-independent surveys throughout the Caribbean produced poor results. In a year's effort around the region 126 tunas of several species were taken, along with 56 billfishes. The records in this paper may not be complete, as they are focusing on total foodfish production, not species composition.   McEachran, J.D. 1983. Fishery assessment of underutilized shark resources in the northwestern Gulf of Mexico. A final report to the Gulf and South Atlantic Fisheries Development Foundation (GSAFDF contract 21-13-21375), available through GSAFDF. As part of a larger survey, the shark catch on commercial swordfish longlines was monitored. During three trips, 6,000 pounds of shark carcass were produced from 113 sharks compared to 3,475 pounds of swordfish from 86 fish; one marlin was also caught. Two cruises, fishing on the continental shelf, produced 40 sharks, two swordfish, one grouper, one marlin, two wahoo, and two cobia. Other commercial shark operations had no bycatch recorded.   McEachran., J.D. 1984. Fishery assessment of underutilized shark resources in the northwestern Gulf of Mexico. A final report to the Gulf and South Atlantic Fisheries Development Foundation (GSAFDF contract 22-13-22881), available through GSAFDF. As a continuation of McEachran 1983, this report noted that for commercial swordfish boats, an estimated shark to swordfish ratio had been calculated at 1.25:1 (2:1 in pounds). Monitoring two commercial trips for swordfish/tuna produced 52 sharks, 60 swordfish, 47 tuna, 16 escolar/oilfish, and four billfish.   Power, J.H. 1993. Analysis of the longline fishery effort, catch, and bycatch in the southwest Atlantic and Gulf of Mexico. A final report to the National Marine Fisheries Service, MARFIN award NA37FF0040 available from the Coastal Fisheries Institute, Louisiana State University, Baton Rouge, LA 70803. This report analyzed the fishery effort and how it changed seasonally and annually, but some bycatch information can be gleaned from the tables; 18 species were considered for multiple species association analyses. Swordfish, bigeye tuna, yellowfin tuna, and dolphin occurred on more than 30 percent of the sets; blue shark, mako sharks, and hammerhead sharks (several species), occurred on more than 10 percent of the sets. All other species occurred on less than 10 percent of the sets. In the northwest Atlantic, except for swordfish, occurrence of a species on a set was less frequent than its non-occurrence by a large margin. For the Caribbean, swordfish, bigeye tuna, yellowfin tuna occurred on more than 50 percent of the sets; other species occurred on 25 percent (blue shark) to <0.5 percent (bluefin tuna) of the sets. For the Gulf of Mexico, swordfish and yellowfin tuna were the only species that occurred on more than 50 percent of the sets; most other species occurred on less than 10 percent of the sets. No disposition (release, retain, discard) rates were presented.   Russell, S.J. 1993. Shark bycatch in the northern Gulf of Mexico tuna longline fishery, 1988-91, with observations on the nearshore directed shark fishery. pp. 19-29 In: Branstetter, S. (ed.) Conservation Biology of Elasmobranchs. NOAA Techn. Rep. NMFS 115. During 87 tuna trips in the Gulf of Mexico, 516 sharks were recorded for 302 longline sets. Of the sharks captured, 120 were retained for sale. Mortality rate of discarded sharks was 46 percent; higher than previous records of 15-35 percent (see Witzell, 1985) probably because of finning (legal at the time). No percent of total catch is given (no target species numbers). In shark-directed sets, 1,181 sharks of 1,449 caught were retained; 18 percent was discarded, and 92 percent of that discard was dead.   Wathne, F. 1959. Summary report of exploratory long-line fishing for tuna in Gulf of Mexico and Caribbean Sea, 1954-1957. Comm. Fish. Rev. 21(4):26. Exploratory longline fishing throughout the Gulf of Mexico and Caribbean during 1955 through 1957 was used to determine the potential to develop fisheries in this region. During several cruises, 203 longlines were set in the region. Species composition, at least that recorded, was: 2,963 yellowfin tuna (72 percent), 235 other tuna (5 percent), 370 billfish including 17 swordfish (9 percent), and 562 sharks (14 percent).   Wilson, C.A. and J.H. Render. 1993. The application of pelagic longline data in reducing billfish by-catch and resource monitoring. (MARFIN NA89WC -H- MF013). Coastal Fisheries Institute, Louisiana State University, Baton Rouge, Louisiana. Statistical analyses were performed on foreign and domestic data sets generated by fishery observers. Foreign data from Japanese longline vessels were collected from March 1978 through December 1981 when the Japanese abandoned the fishery. Monitoring of the domestic longline fleet was implemented in 1985 prior to final acceptance of the Swordfish Fishery Management Plan in 1986. This program was conducted by NMFS from May 1985 through March 1986 and provided observer coverage of 114 sets. Since fishermen in this fishery targeted swordfish, most of the sets were at night. The focus of the domestic program shifted in 1987 to address concerns over increased incidental catches of billfish and sharks associated with increased efforts to catch yellowfin tuna. Most of the observer coverage was in 1987. Data for the Gulf of Mexico region were generated from 728 foreign longline sets. Domestic data collected in the Gulf region were from 80 sets where yellowfin tunas were targeted. There is evidence to show the association and co-occurrence of certain species of tunas and billfish and that these relationships may be predictable based on water temperature. This relationship has often been noted for pelagic resources, particularly tunas. There may be clear implications from these relationships that are useful for management purposes. Billfish bycatch may be reduced through seasonal closures or changes in fishing practices during the summer months. Seasonal closures would impact yellowfin catch, therefore, closure length should maximize billfish bycatch reduction while minimizing impacts to the domestic yellowfin fleet.   Witzell, W.B. 1985. The incidental capture of sharks in the Atlantic United States fishery conservation zone reported by the Japanese tuna longline fleet. NOAA Techn. Rep. NMFS 31. Using observer data and Japanese logbooks, the catches of sharks from the Japanese fleet operating in U.S. waters during the 1970's was estimated. The percent contribution of various catch species is not listed, but shark mortality was listed at 15 percent for Atlantic and 7 percent for the Gulf.     BOTTOM LONGLINE AND COMMERCIAL HOOK-AND-LINE FISHERIES A diversity of bottom longline fisheries exists throughout the Gulf of Mexico. These different longline fisheries vary in their take of bycatch. Gulf demersal longlining operations vary by target species which differ dramatically in depth and bottom type. As a result of these differences, it is impractical to characterize bycatch associated with longlining from a generic perspective. One exception seems to be that sharks are a common bycatch component. From the literature reviewed, sharks account for a high percentage of numbers and weight of non-target species produced from various bottom longlining practices. Because sharks are often alive and can be effectively released unharmed, the impacts of longlining on this resource may not have strong detrimental implications. Because of the various differences in demersal longlining, the following discussion of the fisheries that incorporate this technique should be helpful. Note that some bottom longlining practices no longer exist due to introduced fishery legislation or as in the case of Kali longline gear, non-adoption by industry.   Florida Grouper Longline Fishery: The grouper industry on the west coast of Florida is the largest fishery in the Gulf of Mexico employing bottom longline gear. As a result of information from mandatory logbook reports, 1,223 bottom longlining trips were recorded between March 1994 and February 1995 (NMFS, 1995). Reported landings for these trips totaled approximately 4.53 million pounds. Groupers dominated landings and accounted for 70.3 percent of the total production. Five species accounted for 83 percent of the total bottom longline landings. These were: red grouper (Epinephelus morio), 50.3 percent; unclassified sharks (Chondrichthyes), 14.7 percent; yellowedge grouper (Epinephelus flavolimbatus), 10.2 percent; golden tilefish (Lopholatilus chamaeleonticeps), 5.1 percent; and gag (Mycteroperca microlepis), 2.8 percent. As indicated in Characterization of the Reef Fish Fishery of the Eastern U.S. Gulf of Mexico, the harvest of bycatch appears to be minimal in this fishery -- 4.5 percent of the catch were released dead. Another major consideration regarding take of unwanted species regards harvest of undersized target species (discards). Red grouper have a 20-inch minimum size limit which requires that smaller fish be returned. Mortality rates for released grouper do not seem to be high in this fishery.   Red Snapper Longline Fishery: A substantial bottom longline fishery for red snapper (Lutjanus campechanus) was established in the Western Gulf of Mexico in 1980 and peaked in 1983. The landings from bottom longlining have subsequently declined. This fishery remained viable for several years until the resource was over-exploited by a number of vessels which entered into the fishery. The bottom longline fishery targeted very mature fish which were discovered on relatively smooth bottom contours away from reef structures. A tremendous decline in red snapper produced from longlines occurred after 1991 due in large part to the 1990 prohibition against fishing this gear inside of 50 fathoms (Goodyear, 1995). The majority of red snapper produced by bottom longlines were within the 50 fathom contour. When federal regulations were passed to ban longlines from inside this depth, the vast majority of longliners either changed their fishing operation (converted back to power reels) or redirected bottom longlining to other fisheries (grouper or tilefish). Red snapper are still harvested with longlines outside of the 50-fathom contour in the Western Gulf of Mexico, but these fish are taken incidentally to grouper operations. Because there is a restricted commercial season for red snapper, it is perceived that the majority of incidentally taken red snapper outside of 50 fathoms are released. There is no known study or documentation regarding survivability of these fish when captured from such depths. Furthermore, documentation of quantities of red snapper taken in the western grouper longline fishery does not exist. A need for such data merits consideration. Golden Tilefish Longline Fishery: The bottom longline fishery for golden tilefish (Lopholatilus chamaeleonticeps) has fluctuated since the late 1970's. Compared to the grouper fisheries, it is dwarfed substantially by its limited landings. As indicated in NMFS (1995), tilefish are often harvested as an incidental species in the Eastern Gulf. Only a few vessels target golden tilefish in the western Gulf, and these efforts are contingent upon receiving adequate prices for a seafood species which has been traditionally subjected to extensive market fluctuations. Another problem with this fishery relates to the very slow biological growth and development of this species. The tilefish longline fisheries throughout the Gulf and Atlantic have never been able to maintain sustainability when pursued by a number of vessels. Bycatch production from the tilefish fishery appears to consist mostly of shark. The catch composition of these sharks are compromised mostly of dogfish sharks, Squalus and Centrophorus. Western Gulf Grouper Longline Fishery: The bottom longline fishery for grouper is the largest of the demersal longline fisheries in the western Gulf. The magnitude of this fishery is not known, however it appears to have diminished over the last decade. The loss of red snapper longlining inside of 50 fathoms inhibited the viability of this fishery because it reduced its versatility. Data for this fishery are limited, however the bycatch species appear to be compromised primarily of sharks and eels (especially banded shrimp eels). It appears that relatively little other bycatch exists for this fishery except that of red snapper. Because the Gulf red snapper fishery has a quota which is usually met in less than two months, grouper vessels harvesting red snapper must return these fish if they are harvested outside of the season. There is little knowledge relative to the extent of red snapper that are incidentally harvested outside of 50 fathoms, but it is thought to be considerable in some areas of the Western Gulf. A scarcity of information also exists regarding the survival of red snapper that are harvested and then released from grouper longlines in waters outside of the 50-fathom depth. Kali Pole Longline Fishery: Efforts to introduce Kali pole longlines into the Gulf of Mexico were never successful. Difficulties in handling the gear and expensive gear losses greatly inhibited the adaptation of this longlining system. Limited data exists regarding the take of bycatch with this gear (Bankston and Horst 1984) (McEachran 1985) ( Russell, Gutherz, and Barans 1988). Traditional Hook and Line Fishery:   The hook and line fishery for various seafood species represents one of the oldest fishing techniques in the Gulf of Mexico. The fishery for red snapper in the Gulf of Mexico began over 150 years ago off Pensacola, FL (Goodyear, 1995). Currently this fishing practice is limited almost entirely to harvest of reef fish and is employed throughout the Gulf of Mexico. Although commercial hook and line fishing is a long established fishing technique, there is a clear paucity of information concerning its production of bycatch. The lack of information regarding incidental take of non-target species is somewhat surprising because the reef fisheries have undergone considerable studies and management measures under the direction of the Gulf of Mexico Fisheries Management Council (GMFMC). Perhaps the apparent absence of concern regarding bycatch from commercial hook and line fisheries is an indication that no problems are thought to exist. Closely associated with concerns for bycatch in the commercial hook and line fishery are discards within the various fisheries -- especially the red snapper fishery. The introduction of size limits for various species brought about the requirement to release undersized individuals from the targeted fishery. A minimum size limit was imposed upon red snapper in 1984 by the GMFMC, and size limits have increased from 12 inches to 15 inches since that time. Goodyear (1995) reported that an observer program to collect data about the reef fish fisheries in the Gulf of Mexico was initiated in 1995, and included some red snapper trips aboard handline vessels. Among other information, the observers recorded the lengths and fates of the catch, including the discards. Data from these trips during the 1995 season indicated that 40.7 percent of the red snapper caught were discarded. By weight these fish constituted about 18.6 percent of the red snapper catch. The observations were of fish caught at an average depth of 40 m (range 22 to 62 m) generally offshore Louisiana and east Texas. Only about 1.6 percent of the catch was discarded dead, but Goodyear reported that most of the discarded fish had either stomachs or eyes protruding and many of these probably suffered delayed mortality. Goodyear further noted that reef fish logbooks also contain fields for reporting the number of discarded fish since 1993. Most captains have left that field blank. However for those captains who reported discards, the fraction of catch discarded by number was about 31 percent in 1993, 28 percent in 1994, and 30 percent in 1995. These rates are less than the observer data indicated. Small sample sizes or different geographical coverage between the two sources may account for these differences. Resource managers, fisheries scientists, commercial fishermen, and recreational fishermen continue to debate the magnitude of mortality regarding discards -- especially red snapper. Although field experiments have taken place to determine release mortality, this is a very recondite task. Cages have been used to hold discarded red snapper for study. These are lowered back into the water at various depths and held for periods of time to determine survival. Gitschlag and Renaud (1987), Parker (1985) and Render and Wilson (1994) have performed such experiments. The commercial industry and others have questioned the feasibility of lowering fish back into the water column to determine mortality rates of released fish. This practice introduced forced recompression of fish, particularly those with protruding stomachs. Attempts were made by Gitschlag and Renaud, as well as Parker, to obtain mortalities of fish released at the surface. These efforts have been limited in success because of visibility problems with divers and variations in abundance of predators which would impact released discards. The survival of discards presents a dilemma to the red snapper hook and line fishery. Gitschlag and Renaud (1987) report, "Neither cage experiments nor surface release experiments independently provided an accurate estimate of survival for red snapper released at different depths due to inherent differences in methodologies. The resulting survival range did not account for mortality attributable to predation. Efforts to quantify predation on released snapper were unsuccessful, and values for predation rates could only be assigned arbitrarily for the purpose of discussion. It is doubtful that accurate estimates of predation rates will ever be determined due to potentially insurmountable difficulties with in situ methodologies." Because the magnitude of red snapper discards is so extensive, mortality rates of released fish continues to be an important consideration. Research efforts have provided some clarity regarding the impacts of discard mortality, but definitive answers to this important question do not and may not ever exist.   Bankston, D., and J. Horst. 1984. Exploratory bottom longline fishing off Louisiana's coast. Technical Series No. 1. Coastal Ecology and Fisheries Institute, Louisiana State University, Baton Rouge, Louisiana. Exploratory longline sets were made off the Louisiana Coast utilizing sample quantities of gear. The report focuses upon 122 longline sets deployed in various depths of water extending from 41 to 218 fathoms. Approximately 65 of these sets were made utilizing Kali longline gear arrays. The remainder of the sets (approximately 57) incorporated more traditional bottom longlining gear. Target species were red snapper, yellowedge grouper, and tilefish. A description of bycatch species was included for each set. The primary bycatch species consisted of dogfish and other shark species, eels, and hake. From Gary Graham's experience, the majority of shark species (certainly dogfish) and eels are brought to the surface alive and can be released in good condition. According to this report, the hake were not marketed, but a potential for a small market exists. From Gary Graham's experience, hake are excellent bait and should be utilized for such during longlining activities. This report did not address bycatch production except to list the numbers of individual species taken during each set.   Christian, P.A., M.V. Rawson, D.L. Harrington, and L.G. Parker. 1985. Bottom longlining off the southeastern U.S. coast, Report to Gulf and South Atlantic Fisheries Development Foundation, and National Oceanic and Atmospheric Administration, U.S. Department of Commerce through the National Sea Grant Program (Grant # NA80 AA-D-0091 and NA84 AA-D-0072). This report gives a perspective of demersal longlining operations and gear. Technical aspects regarding deployment and retrieval of gear are presented. Fishing locations where commercial quantities of fish were located are listed. Bycatch data does not exist other than a discussion regarding the use of bycatch for bait.   Cody, T.J., B.E. Fuls, G.C. Matlock, and C.E. Bryan. 1981. Assessment of bottom longline fishing off the central Texas coast, a completion report, Texas Parks and Wildlife Department, Coastal Fisheries Branch Management, Management Data Series Number 22. Utilizing sample quantities of bottom longlining gear, a study of potential finfish production was performed from February 1978 to February 1980 along the central Texas coast. Of the 104 samples obtained, 91 came from the inshore zone within the 50-fathom depth contour. Thirteen additional samples were taken in random sets deployed in depths ranging from 55 to 228 fm. Approximately 21 sets were made in depths of 10 fathoms or less. Sample sets normally consisted of 100 hooks, but ranged in numbers from 50 to 250 hooks. Records of all species of finfish harvested from this study are presented. Atlantic sharpnose shark (Rhizoprionodon terraenovae) was the most abundant species caught during the study; it represented 78.4 percent of the number and 69.5 percent of the weight caught. Although this study did incorporate longline sets in waters which were later exploited for red snapper and grouper, only 17 red snapper (Lutjanus campechanus) and five yellowedge grouper (Epinephelus flavolimbatus) were taken from the sampling. Several deepwater sets on the tilefish grounds were made and 29 golden tilefish (Lopholatilus chamaeleonticeps) were taken in these samples.Although some sets were made outside of fishing grounds normally utilized for past or traditional commercial longlining activities, recorded catches from certain areas do contain data which can compliment bycatch data regarding bottom longlining.   Dixon, R.L. (Abstract only) Survival rates of released undersized reef fishes. NMFS Beaufort Laboratory. Beaufort, NC 28516. To prevent growth over-fishing the South Atlantic Fishery Management Council established size limits for 22 species of reef fishes. However, fishermen reported that many released undersized fishes floated and were presumed to die. Observers were placed aboard headboats and commercial handline vessels to directly observe the fraction of released undersized reef fishes that floated so estimates of maximum acute mortality could be made. Approximately 4,900 released fish representing 28 species were observed. Survival was estimated for 15 species: red porgy (Pagrus pagrus), whitebone porgy (Calamus leucosteus), vermilion snapper (Rhomboplites aurorubens), red snapper (Lutjanus campechanus), lane snapper (Lutjanus synagris), gray snapper (Lutjanus griseus), mutton snapper (Lutjanus analis), yellowtail snapper (Ocyurus chrysurus), gag (Mycteroperca microlepis), scamp (Mycteroperca phenax), red grouper (Epinephelus morio), speckled hind (Epinephelus drummondhayi), black sea bass (Centropristis striatus), white grunt (Haemulon plumieri), and greater amberjack (Seriola dumerili). Total survival rates ranged from 70 percent (red snapper) to 100 percent (whitebone porgy). Tests were made for effects of region, surface water temperature, depth of capture, and depth of capture-surface water temperature interaction on rate of survival. There was a correlation between decreasing rate of survival and increasing depth for gag, speckled hind, white grunt, and black sea bass ( = 0.05).   Gitschlag, G.R., and M.L. Renaud. 1987. Field experiments on survival rates of released red snapper with a discussion of their impact on yield models and minimum size limits, NMFS Southeast Fisheries Center, Galveston, Texas. Survival of red snapper (Lutjanus campechanus) discards was estimated using two types of experiments--a cage survival study and a surface release study. In addition, size by depth relationships for red snapper off the Texas coast were investigated. Results from cage studies indicated 68 percent survival of snapper captured at 50 m (165 ft.). The author acknowledged that predation upon normally released fish did not occur with the use of cages in their study. Surface release data showed survival rates of 99 percent, 89 percent, and 64 percent at 21-24 m (70-80 ft.), 30 m (100 ft.), and 37-40 m (120-130 ft.), respectively, for red snapper measuring less than 30 cm (12 inches) fork length.   Goodyear, P.C. 1995. Red snapper in U.S. waters of the Gulf of Mexico, Southeast Fisheries Science Center, Miami Laboratory, NMFS, Contribution: MIA-95/96-05. This comprehensive report of red snapper covers biological characteristics, catch trends, discards, commercial and recreational catches, mortalities, stock assessments, etc. Little information indicating bycatch of other species is included, however mortality rate of snapper discards is discussed.   Horst, J., and D. Bankston. 1987. Bottom longlining fishing off Louisiana's Coast, Prepared for Coastal Fisheries Institute and Louisiana Sea Grant College Program. This document describes production of marketable fish harvested from bottom longlining efforts and limited snapper reel production off the Louisiana Coast. Results from approximately 76 bottom longline sets are documented. A variety of locations and bottom types were fished during this study. Depths ranging from 21 to 218 fathoms were fished. A total of 1,273 salable fish weighing 8,313 lb. (round weight) were caught. Three species, yellowedge grouper (Epinephelus flavolimbatus), red snapper (Lutjanus campechanus), and golden tilefish (Lopholatilus chamaeleonticeps) were the targeted species for the longlining operation. Additional fishing effort was directed toward harvest of reef fish with snapper reels during certain nighttime periods. Vermilion snapper (Rhomboplites aurorubens) was the primary commercial species taken from 14 locations in depths ranging from 26 to 47 fathoms. Bycatch utilizing snapper reels from these fishing efforts is listed in the production tables. Bycatch consisted primarily of giant snake eels (237), sharks (211), hakes (77), and moray eels (31). A cruise log indicating the location and catch of each set is included in the report.   Hueter, R.E. 1994. Bycatch and catch-release mortality of small sharks in the Gulf coast nursery grounds of Tampa Bay and Charlotte Harbor. (MARFIN NA17FF0378). Mote Marine Laboratory, Sarasota, Florida. (This document also contains both recreational hook-and-line and gillnet info). This two-year project assessed the relative importance of two estuaries of the southwest Florida Gulf Coast, Tampa Bay and Charlotte Harbor/Pine Island Sound, as shark nursery areas, and examined potential fishing mortality of these young sharks in the nurseries. Fishery-independent and fishery-dependent surveys for juvenile and small adult sharks were conducted in the study areas from November 1991 to October 1993. This project is discussed in more detail under the headings of "gillnet/trammel net" and "recreational hook-and-line" elsewhere in this report, but of note here is that one of the fishery-independent methods used in this survey was bottom longlining. Longline sets accounted for 108 gear sets fishing for 488.5 hours with a total catch of 87 sharks of 8 species, 121 skates and rays of 5 species, and 518 bony fishes of 16 species. Interestingly, one of the species listed as being taken on the longline gear was yellowfin menhaden (Table 22). In general, the longline was not nearly as effective as gill nets in catching small sharks. Although a classification system was used to assess the condition of the catch when tagged and released, mortality by gear types is not addressed except in the following manner: "As of 7 December, 1993, a total of 52 shark recaptures, representing 4.2 percent of all tagged sharks at liberty, were reported. Longest time at liberty was 358 days and longest distance traveled was a minimum of 105 nautical miles. Based on tag return data, we estimate an average of 34.8 percent of sharks released alive after being caught do not survive the catch-and-release event. The delayed mortality, combined with an immediate at-the-boat observed mortality of 30.6 percent, yields an estimated total mortality from a single fishing event of 54.8 percent of all juvenile and small adult sharks caught. These mortality estimates apply primarily to sharks caught in gill nets within the study areas."   Huntsman, G.R., W.R. Nicholson, and W.W. Fox Jr. 1982. The biological bases for reef fishery management-proceedings of a workshop held October 7-10, 1980 at St. Thomas, Virgin Islands of the United States. Production of snapper and grouper species harvested aboard research cruises utilizing electric and mechanical reels is documented. Areas sampled are Surinam, Guyana, Honduras, Antigua, Barbuda, and Pedro Bank. Production relative to bottom type is characterized. Documentation of bycatch of non-commercial species is not indicated.   Kawaguchi, K. 1974. Handline and longline fishing explorations for snapper and related species in the Caribbean and adjacent waters, Mar. Fish. Rev. 36 (9):8-31. Explorations for snapper and related reef fish were conducted on under-exploited areas in the Caribbean. This report examines catch results obtained relative to fishing grounds, water depth, bottom type, fishing season, and species composition. Handline production was documented and experimental bottom longline sets were deployed. Three project vessels spent a total of 382 days conducting exploratory fishing. The report includes results from 382 days of fishing. Production from this period equaled 291,000 pounds of fish consisting of 51 percent snapper (Lutjanidae), 34 percent jacks (Carangidae), 6 percent groupers (Serranidae) and 9 percent other mixed species. Because this work was conducted in the Caribbean, application to bycatch production in Gulf of Mexico waters may have considerable limitations. It should be noted that red snapper (Lutjanus campechanus) traditionally harvested in the Gulf and South Atlantic do not occur in Caribbean waters. This report includes results from 87 bottom longline sets throughout the Caribbean. It is indicated that production from longline operations yielded 1,888 pounds of fish. The catch weight was about 15 percent (17 percent in number) snapper, 21 percent (19 percent in number) grouper, 43 percent (25 percent in number) jacks, and 21 percent (39 percent in number) sharks and other fish.   McEachran, J.D. 1985. Assessment of demersal reef fish resources and gear technology for the northwest Gulf of Mexico, prepared for the Gulf and South Atlantic Fisheries Development Foundation, Inc. Contract 25-13-24213/9224. Production from Kali and demersal longline sets are documented in this report. Forty-one sets consisting of 20,470 hooks were conducted utilizing Kali gear. Twenty-five sets utilizing 4,490 hooks were made with traditional bottom longlines. Total catch of marketable species is listed from each of the two gear types. Bycatch is indicated as miscellaneous fish and represents a relatively low number in regard to commercial species landed.   Nelson, W.R., and J.S. Carpenter. 1968. Bottom longline explorations in the Gulf of Mexico, A report on Oregon II's first cruise, Commer. Fish. Rev. 39 (10):57-62. One hundred-nineteen sample longline sets consisting of 36,100 hooks were set off Texas/Louisiana, Campeche Bank, West Florida and the Northern Gulf. Bycatch quantities and species are not documented in this report, however incidental shark production is included (32 percent of the total catch). The gear was set in a broad range of depth, extending from 20 to 300 fathoms. This report indicates that food fish consisted of 77 percent of the total catch off Texas. The second most productive area for food fish (74 percent of the total catch) was Campeche Bank.   NMFS. 1995. Characterization of the reef fish fishery of the eastern U.S. Gulf of Mexico Report to the Gulf of Mexico Fishery Management Council Reef Fish Management Committee, Key West, Florida, July 17-18, 1995. Prepared by the Galveston and Miami Laboratories. The data encompass eleven trips aboard seven bottom longlining vessels from April 1994 through February 1995. Three-hundred-eleven sets (227,607 hooks) were sampled in the Eastern Gulf of Mexico. Two-hundred thirty-six sets targeted red grouper with the remaining seeking yellowedge grouper and blueline tilefish. The study indicates that 5,016 fish of 85 taxa were caught. Approximately 55.9 percent of the individuals were kept, 28.3 percent were released alive, 9.4 percent were retained for bait, 4.5 percent were released dead and 1.8 percent were released with an unknown fate. By number, red grouper comprised the majority (51 percent) of the kept category. Yellowedge grouper accounted for 22 percent, followed by gag and blueline tilefish at six percent, scamp at four percent, speckled hind at two percent and red porgy at one percent. All other species combined accounted for nine percent. Three thousand forty-seven red grouper were caught using longlining gear during this study. Forty-seven percent of the fish caught were less than 20 inches in length (undersized). Of these, 83 percent were released alive, 12 percent were released dead, three percent were released with an unknown fate and one percent were used for bait.   Parker, R.O. Jr. 1985. Survival of released red snapper. Progress report to Gulf of Mexico Fisheries Management Council. (Unpublished data). Initially, the relationship of the depth of capture to survival of released fish was tested in the laboratory. Fish captured from various depths (18 to 122m) were transported to the laboratory and observed over a period of three weeks and mortality was recorded. Later, to remove the reduced pressure and transportation effects, which could cause substantial biases, cage studies were initiated on the fishing grounds. Mortality of red snapper discards was evaluated through cage studies in the Atlantic Ocean and the Gulf of Mexico. Fish measuring up to 38 cm (15 in) FL were used and 79 percent (11 of 14) survival rates were observed at 24 m (78 ft) depths in the Atlantic Ocean. An 89 percent (39 of 44) survival rate was noted at 30 m (100 ft) depths off Texas in the Gulf of Mexico.   Prytherch, H.F. 1983. A descriptive survey of the bottom longline fishery in the Gulf of Mexico, NOAA Technical Memorandum NMFS-SEFC-122. A survey was conducted in three main areas of concentration of the longline fisheries: the eastern Gulf off St. Petersburg, FL, the northern Gulf off Panama City, FL and in the western Gulf off the South Texas area. Information was acquired through interviews with longline vessel crews and through use of NMFS observers. Observers were deployed aboard at least one vessel fishing in each of the principal longlining areas and catch data were obtained. In the eastern Gulf, eels and shark comprised the primary bycatch species as represented from one 10-day trip. It seemed apparent from the data that grouper were being targeted. Six different vessels were observed in the northern Gulf and sharks and eels again comprised the bulk of the bycatch (species were not indicated). Grouper and tilefish comprised the bulk of the catch. In the western Gulf of Mexico, one red snapper longline trip was documented. Sharks comprised the majority of the bycatch (species not listed) with an occasional scorpionfish being harvested (four, three, and three harvested respectively in three days, followed by no scorpionfish harvested in the remaining six days of the trip). Very little bycatch was indicated in any of the data from the three areas.   Render, J.H., and C.A. Wilson. 1994. Hook-and-line mortality of caught and released red snapper around oil and gas platform structural habitat. Bulletin of Marine Science. Vol. 55 (2):1106-1111. A study to determine the mortality of released red snapper was conducted off Louisiana. Net cages were used to hold 282 red snapper caught in depths of 21m (69 feet). A 19.7 percent mortality was observed through the study. No observed significant variation in mortality rate was noted for treatment (deflation of gas bladder) or length of time in net.   Russell, M.G., E.J. Gutherz, and C.A. Barans. 1988. Evaluations of demersal longline gear off South Carolina and Puerto Rico with emphasis on deep-water reef fish stocks. Mar. Fish. Rev. 50 (1):26-31. Comparative gear trials were conducted during three cruises in two geographical areas. The RV Oregon sampled in a specific (0.8 km2) area off Charleston, SC in a depth of 100-110 fm. Production of traditional bottom longlining gear was compared to that of Kali pole arrays. Similar studies were conducted in the Caribbean depths ranging from 38-324 fm. Catch analysis is presented and bycatch composition can be abstracted from the data. The data does not indicate that commercial quantities of fish were encountered in these trials which might deter actual determination of bycatch impacts from commercial operations.   Russell, S.J. 1994. Mackerel and reef fish bioprofile and catch/effort data collection from the northern Gulf of Mexico (MARFIN NA37FF0042) Russell Research Associates, Grosse Tete, Louisiana 70740. This report summarizes the results obtained during the project period of April 1, 1993, through December 31, 1993, of a continuing study to record catch/effort and bioprofile data from the mackerel and reef fish fisheries in the northern Gulf of Mexico. A total of 80 interviews (34 from bandit reel boats, 9 from bottom longline boats, 1 from a charter/party boat selling its catch, 23 from mackerel trolling boats, 2 from boats fishing tunas with bandit reels, 1 from a fish trawl boat, 1 from a strike net boat, and 9 from vessels carrying multiple gear such as bandits and longlines, or bandits, and fish traps) were obtained from commercial fishermen targeting reef fish and king mackerel during the 7 ½ months of actual field sampling. A total of 3,833 fish were measured during the project period, and 3,679 of these fish were also weighed. Table 2 gives a comparison between different gear of the proportion of species comprising the sampled catches (percent by weight in gutted pounds except for gray triggerfish which is always landed whole). Since the surveys were conducted dockside, there is no indication of the amount of at-sea release which occurred. However, primary target species landed made up the majority of catch for mackerel trolling (97.8 percent king mackerel), and bottom longlining (82.1 percent grouper and 13 percent tilefish). The catches landed by bandit-reel fishermen were more diverse (39.6 percent snappers, 32 percent jacks, 13 percent groupers, 4.1 percent triggerfish, and 3.2 percent tunas).   Wilson, R.R. Jr., and K.M. Burns. 1996. Potential survival of released groupers caught deeper than 40 m based on shipboard and in-situ observations, and tag-recapture data. Bulletin of Marine Science. 58(1):234-247. (This work also applies to the recreational hook-and-line fishery). In this study, shipboard and in-situ observations were used to determine the potential post-release survival rate of groupers, chiefly red grouper (Epinephelus morio), caught from between 44 and 75 m on the central west Florida shelf. Potential survival rates were then further evaluated in combination with data from a tag and recapture study (3,818 releases) in the same area and time period. Potential survival rates for released red grouper and scamp (Mycteroperca phenax) caught shallower than 44 m were very high (86 percent to 100 percent) for up to eight days following release. Undersized grouper (<50.8 cm) caught from both shallower and deeper than 44 m, then tagged and released, were found to survive long enough to reach legal size. For grouper caught deeper than 44 m, however, tag/recapture data and in-situ observations indicate that potential survival rates are too low (<33 percent) for the 50.8 cm (20 inches) size rule to be effective in increasing yield.   Wolf, R.S. 1974. Minor miscellaneous exploratory/experimental fishing activities in the Caribbean and adjacent waters. Mar. Fish. Rev. 36(9):78-87.   In a follow-up effort to that described by Nelson and Carpenter (1968), one of several miscellaneous fishing tactics used was deepwater bottom longlining. This was attempted off Guyana to French Guiana in 100 to 225+ fathoms. During 43 sets, 7,118 hooks were deployed and caught 224 tilefish, 33 groupers, and 13 other fishes (noted as mostly shark).     RECREATIONAL SHRIMP FISHERIES Authors' note: This section also includes information on bycatch from bait shrimping and other small nets or nearshore shrimping activities. Recreational shrimping is defined by most Gulf Coast states as the taking of shrimp for personal use or consumption (for example: eating or bait) and not for sale, trade, or commercial use. Recreational shrimpers use small shrimp trawls with a maximum headrope length of 16 feet, (20 feet in Texas) or cast nets with a maximum size of 12 to 24 feet, (varies by state) and are required to have a recreational shrimping license and/or a recreational fishing license. The total costs of these for each state are (for residents): $15 in Mississippi, $16 dollars in Alabama, $13.50 in Florida, $36 in Louisiana, and $43 in Texas. In Florida, recreational shrimpers are prohibited from using otter trawls as of January 1, 1996. Thus, Florida recreational shrimpers can only use the following gear: landing or dip nets with a 96-inch maximum perimeter, cast nets with maximum radius of 12.5 feet, push nets, one frame net with a maximum opening of 16 feet around the perimeter, and shrimp traps. State limits for shrimp catches are: 50 pounds per person per day in Mississippi, five gallons per person per day in Alabama, five gallons per vessel per day in Florida, 50 pounds per person per day or 100 pounds per boat per day in Louisiana, and 15 pounds per person per day in inside waters or 100 pounds per boat per day in outside waters in Texas. Recreational shrimp size limits are restricted to 68 shrimp per pound in Mississippi, no size limit in Alabama, 47 shrimp heads-on (or 70 headless) per pound for Florida, 100 shrimp (white only) per pound in Louisiana, and 50 shrimp per pound in inshore waters or 100 shrimp per pound for outside waters in Texas. A 1991 survey of Alabama's licensed recreational shrimpers combined with independent trawling data estimated the recreational shrimp bycatch to be about 1,222,000 lb. of fish compared to 99,000 lb. of shrimp. In the 1970's recreational shrimp catches along the Gulf coast varied from 23,000,000 lb. in Louisiana to 204,000 lbs. in Alabama. During this same time period, recreational shrimping finfish bycatch for Alabama, Mississippi, and Louisiana was estimated at 110,200,000 lb. annually. Trinidad's artisan trawlers were reported to have catch ratios of 9 lb. finfish to 1 lb. shrimp and 14.7 lb. bycatch to 1 lb. shrimp. Reported fish species in the bycatch from recreational and bait shrimping activities include: pinfish (Lagodon rhomboides), pigfish (Orthopristis chrysoptera), silver jenny (Eucinostomus gula), Atlantic croaker (Micropogonias undulatus), weakfish (Cynoscion regalis), spot (Leiostomus xanthurus), yellowfin and Atlantic menhaden (Brevoortia smithi and Brevoortia tyrannus), hard-head and gafftopsail catfish (Arius felis and Bagre marinus), with seasonal catches of gray snapper (Lutjanus griseus), lane snapper (Lutjanus synagris), gag grouper (Mycteroperca microlepis), and hogfish (Lachnolaimus maximus). Although bycatch to shrimp ratios and bycatch mortality rates tend to be highest during the summer and fall, differences in mean CPUE in number and weight exist among seasons and location. Seasonal and location shifts in species composition are attributed to migration and various marine and environmental influences. Net-modification trawl experiments showed that the FFE (Florida Fish Eye), AFAT (Accelerator Funnel mounted in front of a TED), and AFBT (Accelerator Funnel mounted behind TED) designs were relatively effective and consistent at reducing small fish bycatch in various estuarine habitats, while the SWFD (Seaweed Fish Excluder), HWED (Hummer-wire Deflector), FFED (Florida Finfish Excluder Device), and the Fish Shooter were somewhat inconsistent in their efficiencies. The reason for this was suggested to be due to different environmental factors associated with the various habitats. An electrified trawl experiment caught an average of 36.7 percent more shrimp than a roller-frame, while roller frame trawls had lower bycatch mortality rates. Roller frame and otter trawls would be more effective at bycatch reduction if seasonal and temporal closures coincided with times of fish abundance, although otter trawls are thought to need gear modifications as well.   Campos, W.L., and S.A. Berkeley. 1986. Impacts of the commercial fishery on the population of bait shrimp (Penaeus spp.) in Biscayne Bay. Final report to Dade County Department of Environmental Resources Management. This study addressed the concern that the live bait shrimp fishery harvest represented a significant removal of food from Biscayne Bay. Relative contribution of the commercial fishery to total mortality required accurate estimates of shrimp population size. This was accomplished by using an electrified trawl which captured a relatively constant percentage of available shrimp (36.7 percent more than a roller frame trawl). Total catch by the bait shrimp fleet, estimated monthly by monitoring the commercial fishery, was used to calculate the percent harvested by the bait fishery. Total mortality, natural mortality and fishing mortality were also estimated. The commercial fishery and ecosystem absorbed (monthly) 23 percent and 26 percent of the male and female non-migrating population, respectively. Emigrating shrimp stocks had losses of 56 percent (male) and 57 percent (female) due to fishing, natural means and emigration. The biggest source of loss was through natural means, while fishing accounted for only 21 percent. Approximately four percent were lost each month to emigration. Thus, the direct impact of the bait shrimp fishery on the shrimp population was relatively small.   Coleman, F.C., and C.C. Koenig. (n.d.). Florida inshore shrimping: experimental analysis of bycatch reduction. Final report to Florida Department of Natural Resources. 3900 Commonwealth Blvd., Tallahassee, FL 32399. The efficiencies of five bycatch reduction devices (TED - turtle exclusion device, FFED - Florida finfish excluder device, AFAT - accelerator funnel mounted in front of a TED, AFBT - accelerator funnel mounted behind TED and AFAT, HWED - hummer-wire deflector, and SWFE - seaweed fish excluder) were evaluated in three representative estuarine habitats. The AFAT and the AFBT designs were the most effective and consistent at reducing small fish bycatch under various estuarine habitats, while the SWFE, HWED, and the FFED were inconsistent in their efficiencies, possibly due to different environmental factors in the various testing sites.   Coleman, F.C., C. Koenig, and W.F. Herrnkind. 1993. Survey of Florida inshore shrimp trawler bycatch. Annual report to Florida Department of Natural Resources. 3900 Commonwealth Blvd., Tallahassee, FL 32399. The catch of inshore bait shrimpers and food shrimpers was sampled in the Big Bend, South Florida, Northeast Florida, and the Panhandle. The most abundant fish in the bycatch of bait shrimpers (using roller-frame trawls operating in seagrass beds) were pinfish (Lagodon rhomboides), pigfish (Orthopristis chrysoptera), silver jenny (Eucinostomus gula), with gray snapper (Lutjanus griseus), lane snapper (Lutjanus synagris), gag grouper (Mycteroperca microlepis), and hogfish (Lachnolaimus maximus) being more common only during the spring and summer. Bycatch reduction from roller frame trawlers was suggested to be more effective if seasonal and temporal closure of seagrass beds coincided with time of abundance, while otter trawlers were suggested to need gear modification as well as seasonal and temporal closures.   Coleman, F.C., C.C. Koenig, and W.F. Herrnkind. 1995. Survey of Florida inshore shrimp trawling by-catch and preliminary tests of by-catch reduction devices. Annual report to Florida Department of Natural Resources. 3900 Commonwealth Blvd., Tallahassee, FL 32399. Bycatch reduction device experiments (FFED - Florida finfish excluder device, SWFED - seaweed fish excluder device, and AF - accelerator funnel) were conducted in several Florida inshore shrimping areas using otter trawls in coordination with roller frame studies to establish the geographic distributions and seasonality of finfish, portunid crabs, and penaeid shrimp. Although their was no indication of potential influence of bycatch on population sizes, some of the more common components of the bycatch included pinfish (Lagodon rhomboides), pigfish (Orthopristis chrysoptera), spotfin mojara (Eucinostomus argenteus), Atlantic croaker (Micropogonias undulatas), spot (Leiostomus xanthurus), bigheaded searobin (Prionotus tribulus), and gulf toadfish (Opsanus beta).   Continental Shelf Associates, Inc. 1992. Commercial food shrimp fishery impacts on by-catch in the lower St. Johns River, Florida. Draft final report to Florida Department of Natural Resources. 3900 Commonwealth Blvd., Tallahassee, FL 32399. Species composition and mortality of otter trawl bycatch showed that the fishes most commonly caught were sciaenids (Atlantic croaker, Micropogonias undulatas; weakfish, Cynoscion regalis; spot, Leiostomus xanthurus; silver perch, Bairdiella chrysoura), ariids (hard-head and gafftopsail catfish, Arius felis and Bagre marinus) and clupeids (yellowfin and Atlantic menhaden, Brevoortia smithi and Brevoortia tyrannus). Seasonal shifts in species composition were attributed to migration and various other marine and environmental influences. Shrimp catches and bycatch mortality were highest during the summer and fall and lowest in the winter probably because of the higher water and air temperature during the summer.   Continental Shelf Associates, Inc. 1992. Roller-frame trawl bait shrimp fishery impacts on by-catch, seagrass meadows, and habitat quality in Pine Island Sound, Florida. Draft final report to Florida Department of Natural Resources. 3900 Commonwealth Blvd., Tallahassee, FL 32399. Of the 57 species of fish caught as bycatch, the 4 most abundant were pinfish (Lagodon rhomboides), silver jenny (Eucinostomus gula), pigfish (Orthopristis chrysoptera) and silver perch (Bairdiella chrysoura). Floating pen mortality experiments of by-caught fishes showed low mortality among the fishes in both summer and winter because of a combination of short tows, short culling times, and night trawling. No statistically significant effect of repetitive roller-frame trawling (multiple trawl passes) on shoot density, blades/shoot, leaf blade length, or leaf blade area was detected on designated turtle grass sites, thus supporting the hypothesis that roller-frame trawling did not physically or mechanically remove or destroy seagrass.   Fuls, B. 1995. Assessment of composition and magnitude of bycatch associated with the commercial shrimp trawling industry in central lower Texas coastal bays during the spring and fall Texas commercial bay-shrimp open seasons. Final report to the Texas Parks and Wildlife Department. 4200 Smith School Road, Austin, TX 78744. Monitoring three bays in lower Texas -- Aransas Bay, Corpus Christi Bay, and lower Laguna Madre - this study reported that bycatch was higher in the spring than in the fall. Bycatch to shrimp ratios were 4:1 to 6.8:1 depending upon season and area. Finfish to shrimp ratios were 1:1 to 5.1:1; again varying by season and area. Bycatch ratios were highest in Corpus Christi Bay and lowest in Laguna Madre. The report also noted that the quantity and composition of the bycatch in this fishery-independent survey was very different from concurrent fishery-dependent surveys. Bait shrimp bycatch surveys (9.8 m trawl) in Lower Laguna Madre during the spring of 1993 showed that four species (lesser blue crab, Callinectes similis; Atlantic croaker, Micropogonias undulatus; spot, Leiostomus xanthurus; and sand seatrout, Cynoscion arenarius) comprised 62 percent and 43 percent of the mean CPUE in number and weight, respectively. The overall mean CPUE for bycatch was 2,966 organisms/h/net in number and 54.643 kg/h/net in weight. In the fall, five species (sand seatrout; lesser blue crab; spotfin mojara, Eucinostomus argenteus; hardhead catfish, Arius felis; and Atlantic Cutlassfish, Trichiurus lepturus) comprised 65 percent and 53 percent of the mean CPUE in number and weight, respectively. The overall mean CPUE for bycatch organisms was 1,597 organisms/h/net in number and 27.775 kg/h/net in weight.   Gulf of Mexico Fishery Management Council. 1980. Fishery management plan and proposed regulations for the shrimp fishery of the Gulf of Mexico, United States waters. Federal Register 45:74178-74308. During the early 1970's, the largest amount of licensed recreational shrimpers (using 16 to 40 ft. nets) along the Gulf coast resided in Louisiana (30,000) and caught about 23,000,000 lb. of shrimp annually. Louisiana's number of recreational shrimpers was followed by Texas (>10,000), Alabama (5,100), Mississippi (>1,500), and west Florida (>500) who caught approximately 900,000 lb., 204,000 lb., 290,000 lb., and 166,000 lb., respectively.   Gulf of Mexico Fishery Management Council. 1981. Draft fishery management plan/ environmental impact statement and regulatory analysis for groundfish in the Gulf of Mexico. The estimated recreational shrimping groundfish bycatch during the 1970's was calculated to be 50,000 metric tons (110,200,000 lbs.) annually for Alabama, Mississippi, and Louisiana. The bycatch estimate was calculated by multiplying the total annual recreational shrimp catch by the average fish/shrimp ratio of commercial shrimp trawls in inshore waters for the 1973-1977 period.   Maharaj, V., and C. Recksiek. 1991. The by-catch from the artisanal shrimp trawl fishery, Gulf of Paria, Trinidad. Mar. Fish. Rev. 53:10-15. Trinidad's artisanal trawler catches (10 m trawl) of Penaeus schmitti, P. subtillis, P. notialis, and Xiphopenaues kroyeri, in the nearshore (<5 fm isobath) of the heavily developed Gulf of Paria were examined and showed annual weight ratio estimates of 9 kg finfish : 1 kg shrimp and 14.7 kg bycatch : 1 kg shrimp with monthly ratio estimates being the lowest (<10) from late January to May (the dry season) and highest during September to May (the wet season and peak time for P. schmitti). Reported shrimp landings equaled 108,000 kg, while the total annual bycatch estimate was 1,594,000 kg (90 percent estimated to be discarded) containing 974,000 kg of finfish (most caught August -December) and 620,000 kg of crab.   Wallace, R.K., and C.L. Robinson. 1994. Bycatch and bycatch reduction in recreational shrimping. Northeast Gulf Science 13:139-144. Bycatch from recreational shrimping was estimated from fishery-independent trawling and through a survey of licensed recreational shrimpers in Alabama. The mean fish bycatch was 5.4 kg per 20-minute tow and contained 426 fish primarily from three families (Sciaenidae, Engraulidae, Clupeidae). Based on the survey of recreational effort, the total fish bycatch was estimated at 603,000 kg or 47.6 million fish. Tests of two bycatch reduction devices resulted in significant reduction for the Florida Fish Eye, but no significant reduction for the Fish Shooter. From the field sampling, bycatch to shrimp ratios were nearly 15:1 (range 1.2:1 to 93:1). Four hundred, seventy-four surveys were returned (19.5 percent); usable surveys indicated that, in 1990, recreational shrimpers averaged 5.6 trips, 4.3 tows per trip, and 38.2 minutes per tow equaling approximately 40,000 net hours. Based on their estimates of 16.2 kg of bycatch per net-hour, Alabama recreational shrimping contributed to an estimated 648,000 kg of bycatch and 49,000 kg of shrimp for a 13:1 bycatch to shrimp ratio. Tests of bycatch reduction devices included "fish shooter" (a slit in the bag), and two sizes of "fisheyes" placed on the bottom of the bag. The fisheyes in this configuration reduced fish, but lost 14 and 19 percent of the shrimp.     PURSE SEINE FISHERIES At the present time, purse seine gear is used primarily to take non-food finfish species in the Gulf of Mexico. Although a small seasonal fishery exists in the eastern Gulf for species harvested for direct consumption, the vast majority of fish produced using this gear are industrial species such as menhaden and baitfish such as round scad ("cigar minnows"). In the past, purse seine gear was used to take other species like red drum and mullet, but state and federal regulations have effectively shut down these fisheries. For example, four out of the five Gulf states currently have a prohibition against the use of purse seine gear in the striped mullet fishery. All of the Gulf states have regulations regarding the amount of non-target species which can be landed by a purse seine vessel (typically expressed as a percentage of the catch by weight). The use of purse seines for the taking of finfish other than menhaden or herring-like species is prohibited outright in some Gulf states. There is a relatively small-scale fishery for finfish for direct consumption in federal waters in the eastern Gulf. This fishery takes species such as little tunny, blue runner, and crevalle jack. The menhaden fishery is also subject to area and seasonal closures in all Gulf states. With the exception of Florida, which has no closed season, the menhaden season runs from the third Monday in April to November 1 in the Gulf. There is also a provision for a special bait season in Louisiana which can extend the season until December 1 and allow for an opening beginning April 1 subject to special permit regulations. The menhaden fishery is one of the United States' oldest and most valuable fisheries with landing dating to the late 1800's. Data for the fishery are sketchy prior to World War II; thereafter, however, landings generally increased through the mid 1980's as the industry grew. Although there may be considerable annual fluctuations, Gulf landings increased to a record of 2.2 billion pounds in 1984. This figure amounted to 76 percent of U.S. menhaden and 29 percent of total U.S. landings of fish and shellfish (GSMFC, 1995). Due to the way the gear is operated, the menhaden purse seine fishery is considered to be a relatively "clean" fishery with little incidental harvest of non-target species. Indeed, most early research into the effects of the fishery on the ecosystem concentrated on predator-prey relationships. Oviatt (1977) reported that sport fishermen in the Narragansett Bay region or Rhode Island felt that such a large portion of the biomass of menhaden was taken that there was insufficient food for predator species. However, calculations suggested that even when menhaden abundances are so low that it is not commercially feasible to catch them, they are still sufficiently abundant to be a primary food source for predator fish. "Modern" studies of menhaden fishery bycatch date back to the 1950's. Perhaps due to the widely-publicized controversy regarding dolphin interactions in the Pacific tuna purse seine fishery as well as emerging user-group conflicts, the menhaden purse seine fishery has been subject to more recent scrutiny regarding bycatch. Recent Florida regulations regarding the use of commercial gear have had a substantial impact on the eastern Gulf baitfish fishery. Because this is primarily a shallow-water nearshore fishery, and those waters are now closed to purse seine gear, landings are down considerably. One of the largest suppliers of round scad (Decapterus punctatus) reports a drop in landings from one to 2 million pounds per year to 15,000 pounds this year (Lancaster, 1996).   Austin, H., J. Kirkley, and J. Lucy. 1994. By-catch and the fishery for Atlantic menhaden, Brevoortia tyrannus in the mid-Atlantic Bight: An assessment of the nature and extent of by-catch. Virginia Sea Grant Marine Resource Advisory No. 53. College of William and Mary School of Marine Science, Virginia Institute of Marine Science Gloucester Point, Virginia 23062. This study presents a framework for assessing the extent of bycatch and an analysis of bycatch in the menhaden fishery. A regulatory-enforcement type sampling scheme in which all inspections were unannounced was designed and sampling was conducted at the docks during off-loading and on-board vessels during harvesting. Sampling was conducted between June and November 1992. A total of 45 off-loadings and 43 at-sea sets were sampled. Following industry practices , all counts of menhaden caught or landed were measured in terms of standard fish (1,000 standard fish weigh approximately 670 pounds). The total number of menhaden off-loaded during dockside sampling was 13.6 million standard fish; 2.5 million standard menhaden were harvested during at-sea sampling. Total bycatch observed in samples was 1,413 fish or shellfish. Relative to the total harvest, bycatch was estimated to account for 0.04097 percent and bycatch rates had a seasonal component with higher rates (0.14 percent) in August and minimum rates (0.002 percent) in September. There were substantial differences in percentage bycatch observed between dockside and at-sea sampling with higher percentages observed in at-sea samples. Thus, it was concluded that dockside sampling is inadequate to precisely determine the nature and extent of bycatch because at-sea discards and associated mortality cannot be determined. Bycatch of recreational species was extremely low relative to the harvest of menhaden.   Condrey, R. 1994. Bycatch in the U.S. Gulf of Mexico menhaden fishery--results of onboard sampling conducted in the 1992 fishing season. Coastal Fisheries Institute, Louisiana State University, Baton Rouge, Louisiana 70803-7503. Bycatch in the U.S. Gulf of Mexico menhaden was sampled during 10, week-long onboard sampling trips made along the Louisiana through Alabama coast during the 1992 fishing season. Bycatch was estimated in two ways. By one method, a long-handled net was used to sample the stream of fish going through the chute into the hold. From one to five such samples were taken per set, depending on the length of the fish pumping operation. For each sample, the weight and number of menhaden was recorded; as was the identity, length frequency and total weight of each bycatch species. These samples are termed "retained bycatch." By the second method, the identity, estimated length and fate of bycatch species which did not go directly into the hold were recorded. These are termed "released bycatch." The released bycatch were observed in two areas. The first area was in association with the net. Here the identity, fate, and estimated length of fish which were enclosed by the net but not pumped into the hose lowered into the hardened net were recorded. The second area was the deck of the menhaden vessel. Bycatch landed on the deck during normal pumping operations when the chute of the large fish deflector was directed onto the deck. One hundred and eighty-two (182) samples of the retained bycatch were obtained from 49 sets. The intensity of sampling per set ranged from one to five samples with a mode of four samples per set. The weight of menhaden per sample ranged from 10 to 42 pounds with a mode of 16 pounds and an arithmetic mean and standard deviation of 19.6 ± 5.5 pounds per set. The number of menhaden per sample varied from 26 to 237 with an arithmetic mean of 93.21 ± 43.30. The wide range in number of menhaden per sample reflects the variation in individual fish size in the three to four year classes which comprise the fishery. The weight of total bycatch per sample ranged from zero to four pounds per sample with a mode of 0. The number of different species of bycatch encountered per set ranged from zero to four with a mode of zero. While 47 percent of the samples contained no bycatch species, 77 percent contained zero or one bycatch, and 93 percent contained zero to two bycatch species. Atlantic croaker was by far the most frequently observed bycatch, occurring in 30 percent of the samples. Croaker were followed in frequency of occurrence in the samples by Atlantic bumper (10 percent), silver seatrout (9 percent), and gafftopsail catfish (7 percent). Together these four species accounted for 56 percent of the cumulative frequency of species occurrences. The percent of the menhaden catch (by number) was computed as the ratio of the weighted daily mean catch of bycatch per sample to the average menhaden catch per sample, raised to a percentage. Similar computations were performed to determine the bycatch percentage by weight of the menhaden catch. As such, retained bycatch were estimated to account for 1 percent by number and 1.2 percent by weight of the total catch. The most frequently encountered species in the released bycatch were sharks (63 percent), followed by gafftopsail catfish (61 percent) and crevalle jack (48 percent). Dead release was high among the most abundant released bycatch species. For example, 825 gafftopsail catfish were observed. Of these, 57 percent were released dead. Of the 246 crevalle jack observed, 80 percent were released dead. And of the 201 sharks observed, 50 percent were released dead. Adding the released bycatch to the retained bycatch for purposes of obtaining a bycatch to menhaden ratio would not change the above figures of 1 percent by number and 1.2 percent by weight even if all of the released bycatch were released dead. This paper also has a good review of other related studies which are summarized below. The author is critical of these previous studies citing flaws in sampling techniques which led to under-reporting of the released portion of the bycatch. This critique also points to a lack of statistical scrutiny of the data on the retained bycatch, which has likely resulted in an over-estimation of its magnitude. Despite the published reports of highly skewed distributions, previous authors used simple arithmetic means to extrapolate the bycatch samples to the entire catch of the industry.   Christmas, J.Y., G. Gunter, and E.C. Whatley. 1960. Fishes taken in the menhaden fishery of Alabama, Mississippi, and eastern Louisiana. U.S. Fish and Wildlife Service Special Scientific Report--Fisheries No. 339. 10 pp. Documents a survey of menhaden vessels operating out of Mississippi and fishing the waters of the Mississippi, Breton and Chandeleur Sounds, as well as west of the Mississippi River to Grand Isle in 1958. Though not detailed, estimated total bycatch of the Gulf menhaden industry was 15 million pounds per year. This figure may be heavily influenced by the accidental take of a single school of striped mullet.   Dunham, F. 1972. Menhaden fishery investigations. In: A study of commercially important estuarine-dependent industrial fishes. Louisiana Wildlife and Fisheries Commission Technical Bulletin 4:52-60. Dockside sampling of menhaden bycatch in Louisiana was conducted in 1971 and 1972. From 1971 results, a bycatch to menhaden ratio of 0.05 percent by number was calculated. In 1972, a total of 409 kg of menhaden was sampled. The catch by weight consisted of 96 percent menhaden, 2 percent Atlantic thread herring and 2 percent other species. Atlantic thread herring (which had not occurred in the 1971 samples) should not be considered as bycatch, since this is a highly desired fish in the fishery because of its high oil yield.   Guillory, V., and G. Hutton. 1982. A survey of bycatch in the Louisiana gulf menhaden fishery. Proceedings of the Annual Conference of the Southeastern Association of Fish and Wildlife Agencies 36:213-223. At-plant sampling across the Louisiana coast for menhaden bycatch was conducted. Eastern Louisiana had a higher bycatch than did the central and western portion of the state. The two-year average percentage by number was 4.6 percent for eastern Louisiana, as compared to 1.1 percent and 1.3 percent for central and western Louisiana respectively. By weight these percentages were 2.6 percent for eastern Louisiana, as compared to 0.8 percent and 1.4 percent for central and western Louisiana. Using an overall mean percentage by weight ratio, the average annual bycatch for 1970-1975 was estimated to be 14.6 million kg (32.2 million pounds).   Knapp, R.T. 1950. Menhaden utilization in relationship to the conservation of food and game fishes of the Texas Gulf Coast. Transactions of the American Fisheries Society 79:137-144. Onboard sampling of the entire catch of 17 menhaden sets was made off western Louisiana. A total of 1,574 fish other than menhaden were taken in these 17 sets giving a bycatch to menhaden ratio by number of 0.07 percent.     GILL AND TRAMMEL NET FISHERIES The use of entanglement gear such as gillnets and trammel nets has been and continues to be the subject of much controversy. User group conflicts waged on the political front have resulted in restrictions or prohibitions of the use of this type of gear in all five Gulf states. Ironically, while opponents of gill and trammel net use have cited bycatch as one of the major problems in the fishery, very few scientific studies exist to verify or refute this position. Harrison (1996) provides the following summary of current regulations regarding the use of gillnets on a state-by-state basis: Texas was the first Gulf state to enact a gillnet ban. According to the Texas Parks and Wildlife Department (TPWD), during the late 1970's the state's red drum and seatrout populations declined sharply due to harsh winters and overfishing. To counter this trend, TPWD issued a proclamation prohibiting gillnet use and sale of these fish in 1981 over the objection of the commercial fishing industry. TPWD banned all other types of commercial netting for saltwater fish (excluding menhaden) in 1988. Since the Texas gillnet ban, all four remaining Gulf states have enacted various measures to restrict gillnetting. Florida enacted both a constitutional amendment and a statute governing gillnet use which became effective on July 1, 1995. The amendment provides "no gillnets or other entangling nets shall be used in Florida waters." The statute provides in part "no person may take food fish within or without the waters of this state with a purse seine, purse gill net, or other net." Alabama has the distinction of being the only Gulf state where the commercial and recreational fishermen sat down together and worked out a compromise. Although over-fishing of striped mullet was an issue in Alabama, the main concern was the influx of Florida fishermen, ousted by their state's new gillnet ban. The Alabama statute prohibits gillnets over 2,400 feet in length. In addition, the law imposes license restrictions. An applicant must have purchased a license for two years between 1989 and 1993. The applicant must also have reported at least 50 percent of his gross income on an Alabama tax return came from the capture and sale of seafood two of the five years between 1989 and 1993. The law requires higher fees for commercial gillnet permits from non-residents. In Louisiana, the Coastal Conservation Association argued for gillnet restrictions to protect against over-fishing of marine stocks such as spotted seatrout and mullet. The Louisiana legislature enacted gillnet restrictions as a result. The main part of the law is the Louisiana Marine Resources Conservation Act of 1995. The law establishes a qualification system to determine which commercial fishermen may receive a license to use gillnets. To qualify for a license, an applicant must "provide positive proof that they held a valid commercial gear license for gillnets during any two years of the years 1995, 1994, and 1993." In addition, an applicant must also show that they "derived more than 50 percent of their earned income from the capture and sale of seafood species in at least two of the three years, 1995, 1994, or 1993." The law also restricts the length, mesh size, and seasons in which nets can be used, as well as their manner of use (e.g. setnets are prohibited and only strike nets are allowed). Mississippi was the last Gulf state to address commercial gillnet restrictions. On August 15, 1995, the Mississippi Commission on Marine Resources passed an "emergency" regulation restricting the issuance of gillnet licenses. The new restriction provides for issuance of gillnet licenses only to "individuals, firms or corporations that purchased Mississippi gill and trammel net licenses during any license year between May 1, 1990 and April 30, 1995." Contained in Ordinance No. 5.013 adopted on November 16, 1996, is a provision which stipulates in pertinent part that "On and after January 1, 1997, all gill and trammel nets must be constructed of an approved degradable material." Mississippi also regulates where and when gill and trammel nets may be used. There is also a fishery in federal waters for king and Spanish mackerel using run-around gillnets. This fishery is regulated using permit endorsements, trip limits and catch quotas. The Gulf of Mexico is divided into seasonal sub-zones for management purposes. As in the five Gulf states, there are restrictions on net mesh size and size limits on fish taken in the fishery. Mackerel are also taken as a "bycatch" in the shark gillnet fishery in federal waters. Consistent allocation overruns by both commercial and recreational harvesters of king mackerel in the Gulf will force a reduction in fishing effort in the near term. The 1996-97 commercial season for Gulf group king mackerel in the western zone closed in less than two months (July 1-August 26, 1996 ) after the quota for all methods of fishing of 770,000 pounds was taken. Run-around gillnets are also used to take dolphin but little production occurs in Gulf waters.   Acosta, A.R., and R.S. Appeldoorn. 1995. Catching efficiency and selectivity of gillnets and trammel nets in coral reefs from southwestern Puerto Rico. Fisheries Research 22:175-196. A comparative fishing experiment was conducted to investigate the capture efficiency and selectivity of bottom-set gillnets and trammel nets. Twelve gillnets and trammel nets of different mesh size and hanging ratios were fished from May 1990 to September 1991 in coral reef and mangrove areas off La Parguera, Puerto Rico. Significant differences in capture due to mesh size and hanging ratio were observed. Largest catches in gillnets were obtained by combination of high hanging ratio (1:3) with large meshes (12.7 cm) or low hanging ratio (1:1) with small meshes (7.6 cm). For trammel nets the largest catches were obtained with low hanging ratio and large meshes or high hanging ratio with small meshes. This opposite interaction may arise from the different mode of capture of these gears.   Collins, M.R., and T.I.J. Smith. 1994. Bycatch of Atlantic and shortnose sturgeon in the South Carolina shad fishery. (S-K annual report) Marine Resources Research Institute, South Carolina Department of Natural Resources, Charleston, South Carolina. The sturgeon bycatch of the commercial American shad set gillnet fishery in Winyah Bay, SC was documented during the 1994 season. Examination of the bycatch from 2,561 net-hours (1 net-hour = 91.4 m of gillnet fished for 1 hour) of effort resulted in examination of 23 Atlantic and four shortnose sturgeons, for a CPUE of 0.011 sturgeon/net-hour. Three Atlantic sturgeon were dead and four were injured. Total fishing effort in the study area during the 1994 season was estimated as 10,536 net-hours, permitting an expansion of the bycatch to an estimated 115 sturgeons (85 percent Atlantic), of which it was estimated that 12 died in the nets and 17 were released with varying injuries.   Hueter, R.E. 1994. Bycatch and catch-release mortality of small sharks in the Gulf coast nursery grounds of Tampa Bay and Charlotte Harbor. (MARFIN NA17FF0378). Mote Marine Laboratory, Sarasota, Florida. With the cooperation of commercial fishermen working in the Tampa Bay and Charlotte Harbor areas, fishery-dependent surveys were conducted from November 1991 through August 1993 to document total catch in various fisheries, and monitor bycatch of sharks, rays, bony fishes, and other vertebrates in commercial operations. Total catches were documented for commercial fishing trips in the following fisheries: 1) purse seine fishery for baitfish in Tampa Bay area; 2) roller frame trawl fishery for bait shrimp in Pine Island Sound; 3) otter trawl fishery for food shrimp in Charlotte Harbor; 4) gillnet fishery for Spanish mackerel in Tampa Bay area; 5) gillnet fishery for mullet in Tampa Bay and Charlotte Harbor; 6) gillnet fishery for pompano in Tampa Bay and Charlotte Harbor; 7) trammel net fishery for mullet in Tampa Bay; 8) trammel net fishery for crevalle jack in Tampa Bay; and 9) trammel net fishery for seatrout in Tampa Bay. Over the course of the entire study, a total of 48 commercial gear sets fishing for over 58.3 hours was documented. Observed shark bycatch in commercial fisheries was low, with 19 sharks of six species (juvenile blacktip, adult Florida smoothhound, juvenile and adult bonnethead, juvenile Atlantic sharpnose, juvenile lemon, and juvenile nurse) documented. All but three of these sharks were caught in gillnets targeting mackerel (an observed bycatch rate of four sharks per set for this fishery). Estimated total mortality was 54.8 percent of all juvenile and small adult sharks caught. These mortality estimates apply primarily to sharks caught in gill nets within the study areas.   Hueter, R.E. 1994. Early life history and relative abundance of blacktip and other coastal sharks in eastern Gulf of Mexico nursery areas, including bycatch mortality of sharks and associated fishes. (MARFIN NA57FF0034-01) Mote Marine Laboratory, Sarasota, Florida. Although it has a different title, this work is essentially an extension of the above project into other geographic areas for sampling to determine relative abundance and develop a bioprofile of the shark species encountered. This is ongoing work in which the investigator proposes to get an estimate of bycatch mortality by using commercial gillnet and longline to capture and tag sharks. As in the above work, bycatch mortality estimates will be inferred from tag returns.   Kennedy, S. 1994. Impact of nearshore gillnet fishery on marine turtles on Florida's east coast--Report to the Florida Marine Fisheries Commission. Florida Department of Environmental Protection, Florida Marine Research Institute, St. Petersburg, Florida. An observer program was conducted jointly between the National Marine Fisheries Service (NMFS) and Florida's Department of Environmental Protection/Florida Marine Research Institute (FMRI). A fishery-independent survey was also conducted. Observers were posted on boats from February 4, 1994, through May 5, 1994. Anecdotal information suggested that the presence of observers can affect the behavior of commercial or recreational fishermen. In an attempt to identify the direction and magnitude of those biases, a program to observe commercial gillnet fishing from the beach was implemented. Observers patrolled the beaches of Jupiter Island and Hutchinson Island from April to May looking for nearshore netting activity. No sea turtles were caught in any of the observer trips. Twenty sea turtles were seen by NMFS observers in the vicinity of the boats and two were seen by FMRI observers. Median value of soak time observed from the beach (15 minutes) was significantly larger than those recorded from on-board observations (5.75 and 4.67 minutes). Between 27 percent and 40 percent of fishermen who were unaware that they were being watched from the beach did not abide by the zero soak time rule. Three sea turtles were caught during the independent study representing an overall 3.6 percent capture rate. Results of the independent sampling show that when gillnets are set as a stab net, soak time has a definite impact on the catch of fish. Both the catch of fish and the time necessary to retrieve the net were increased.     FINFISH TRAWLS In the past, trawl gear was used to take industrial bottomfish in the Gulf of Mexico for the pet food industry, this fishery has been all but phased out at the time of this writing. The "croaker boats" used modified heavy-twine small-mesh shrimp trawl designs developed in the northern Gulf to take species suitable for meal or pet food production. As a result, "bycatch" in this fishery is a manner of interpretation since nearly everything was retained and used. Annual production rates averaged about 45,000 metric tons, peaked in 1974 and have been declining steadily since then. Atlantic croaker, which is the target species because of its desirability for canned pet food, is the predominant species in landings and accounts for 69 percent by weight. Spot, seatrouts (sand and silver) and cutlassfish are the other primary species in the fishery. In 1963, the groundfish fleet consisted of 50 vessels; today there are only two or three boats in the fishery on a full-time basis. Interestingly, although the fishery is prosecuted on essentially the same grounds as the shrimp trawl fishery using the same gear and incidental capture of sea turtles was mentioned as a problem in the fishery (GMFMC, 1981), TEDs were never mandated. There is a small-scale deep-water fishery using trawl gear for butterfish in the Gulf. The fish are typically captured using both bottom and midwater trawl in depths of over 50 fathoms. The small amount of incidental harvest associated with this fishery is typically retained and sold as bait due to the fact that no domestic or foreign markets have been developed for species other than butterfish in the coastal herring complex. Butterfish have a poorly developed swim bladder and therefore have a unique signature on modern color depth sounders. This helps to make this essentially a single-species fishery. Since there has never been any perception of bycatch in the finfish trawl fisheries operating in the Gulf of Mexico, related studies are rare or non-existent. It may prove beneficial to review some work done in other regions in the event that marketing obstacles can be overcome for the coastal herring complex and the vast resources (over five million metric tons annual potential yield) begin to be exploited.   DeAlteris, J.T., and D.L. Morse. 1994. The effects of sweep design on the species selectivity of trawls in the silver hake fishery of New England (S-K NA26FD0032). Trawl nets in New England's small mesh fishery for silver hake (Merluccius bilinearis) are highly non-selective in the sizes and species of organisms they retain, capturing not only saleable individuals of target species, but juvenile individuals of target species, and individuals of non-utilized species. This unusable catch is returned to the water, but most discarded individuals die as a result of the capture process or post-capture predation. In this study, sea trials to evaluate the effects of trawl sweep rigging on the retention of silver hake, other gadoids, and flatfishes were performed, in the small mesh silver hake fishery of New England. Experiments conducted during November and December of 1992, in Cape Cod Bay, MA, examined the selective effects of a raised fishing line. Trials conducted in June and July of 1993 took place near Block Island, RI, and evaluated designs combining an elevated fishing line with a discontinuous chain sweep. Fishing line heights ranged from 0.45 m to 0.9 m in the bosom of the trawl, with sections removed from the sweep ranging from 2.1 to 4.6 m in length. Designs were evaluated through alternate tow catch comparisons, trawl geometry mensuration, and videographic monitoring. Wilcoxon Signed-Ranks tests detected no significant reductions in bycatch in the elevated fishing line experiments, at the 95 percent confidence level, while dramatic reductions in bycatch occurred in the discontinuous sweep comparisons, though loss of target species generally accompanies bycatch loss. It is concluded that bottom disturbance by the sweep is the primary stimulus controlling flatfish escapement behavior in the trawl mouth.   DeAlteris, J.T., and K.M. Castro. 1992. Recent progress in the development of selective trawls for demersal fisheries. Department of Fisheries, Animal and Veterinary Science, University of Rhode Island, Kingston, Rhode Island 02881.   This paper provides a good overview of work done to develop species and size selective trawls for the northwest Atlantic demersal fisheries. Sections address the survival of escapees, fish size selectivity as a function of mesh size and shape, and species selection through mechanical size sorting and species-specific behavioral patterns. Codend escapee survival appears to be reasonably high, and not related to mesh shape. But for some species of actively swimming fish, a square mesh shape provides a larger opening that leads to a shorter escape time for the small fish, thus presumably reducing the stress placed on the fish during the capture/escape process. The size-selection process occurs in the cod end; mesh size in the body of the trawl has no selective effect. This paper also contains the results of work done with the Nordmore Grate in the northwest Atlantic shrimp fishery. The bycatch of flatfish, cod and skates was nearly eliminated with no shrimp catch reduction and an improvement in the quality of the catch.   Monaghan, J.P. Jr., and J.L. Ross. 1995. Reduction of bycatch of recreationally important fishes in North Carolina. North Carolina Department of Environment, Health, and Natural Resources, Division of Marine Fisheries, Morehead City, NC. A trouser trawl was used to examine the effectiveness of large square and diamond mesh extensions and codends in releasing small finfish from otter trawls used in the nearshore ocean flynet fishery and the offshore black sea bass and scup fisheries. Stretched mesh sizes of codends and extensions tested in the nearshore work were 3-inch square, 3.5-inch square, 4-inch square, 3-inch diamond, and 3.5-inch diamond. Mesh sizes of extensions and codends tested in the offshore work were 3-inch square, 3.5-inch square, 4-inch square, 3-inch diamond and 4-inch diamond. Length frequencies of weakfish (Cynoscion regalis), Atlantic croaker (Micropogonias undulatus), southern kingfish (Menticirrhus americanus), and spot (Leiostomus xanthurus) were compared between the experimental and control samples for the inshore work. Scup (Stenotomus chrysops) and black sea bass (Centropristis striata) lengths were compared in the offshore work. Smaller weakfish were released by the large square and diamond mesh gear for six of the 10 treatments tested. More large southern kingfish and less small southern kingfish were retained by the experimental side. Size distributions of Atlantic croaker were similar between experimental and control samples. The mean weight of the catch in the experimental codends was 35 percent of the mean weight in the control codend, in the nearshore segment of this study. The mean catch weight of the experimental side was 47 percent of the control side for the offshore work. As a result of the work done in this study, the North Carolina Division of Marine Fisheries issued a proclamation that required a 3-inch square or 3.5-inch diamond mesh codend and a 3-inch square mesh extension in flynets for the 1994-95 season.   Robinson, W.E., and H.A. Carr. 1993. Assessment of juvenile bycatch survivability in the northeast fishing industry. (S-K NA16FL0068). University of Massachusetts-Boston, Environmental Sciences Program. Boston, MA 02125-3399. In order to obtain quantitative data on the survival of the deck discard of Atlantic cod (Gadus morhua) and American plaice (Hippoglossoides platessoides), two cruises were conducted aboard a commercial trawler. Survival rates were determined by putting the "discarded" fish in large cages and returning them to the tow depth for a period of about 24 hours. The first cruise, in June 1991, resulted in overall survival rates of 13 percent for cod and 44 percent for plaice. The second cruise, in late April 1992, yielded survival rates of 51 percent for cod and 66 percent for plaice. Measured blood parameters for cod juvenile bycatch were elevated above control values, even for fish measured within three minutes of landing on deck. These data demonstrate that fish had been considerably stressed prior to landing (either during tow or haulback). Length of tow had no effect on any of the measured blood parameters.   Robinson, W.E., H.A. Carr, and J. Harris. 1993. Assessment of juvenile bycatch and cod-end survivability in the northeast fishing industry--second year's study. (S-K NA26FD0039). University of Massachusetts-Boston, Environmental Sciences Program. Boston, MA 02125-3399. In this extension of the above-referenced work, four cruises were completed to assess the survival of the deck discard of Atlantic cod (Gadus morhua), American plaice (Hippoglossoides platessoides) and yellowtail flounder (Pleuronectes ferrugineus). A spring cruise in April 1992 resulted in overall survival rates of 51 percent for cod, 66 percent for plaice, and 77 percent for yellowtail flounder. Two summer cruises conducted in June 1991 and May-June 1992 produced combined total survival rates of 9 percent for cod, 40 percent for plaice and 66 percent for yellowtail flounder. One winter sampling cruise was conducted in January-February 1993; cod survival was 36 percent; plaice, 0 percent; and yellowtail flounder, 50 percent. Primary factors that were determined to influence survival of cod were air temperature, deck time, fish length, tow duration and tow weight. Air temperature, deck time, fish length and tow duration were most critical to plaice survival. Tow duration and deck time affect the survival of yellowtail flounder.   Supan J., and D. Bankston. 1988. Fish trawling in Louisiana inshore waters. Louisiana Sea Grant College Program, Center for Wetland Resources, Louisiana State University. Baton Rouge, LA 70803-7507. The purpose of this trawling project was to experiment with different mesh sizes and trawl designs and compare efficiency with the trawl developed by Noel Usannaz of New Orleans for use in the Lake Borgne area of southeast Louisiana. In 1983, Usannaz began using a 4-seam fish trawl made primarily of 4-inch stretch mesh to catch sheepshead and black drum. Experimental trawls included trawls of the same design made of 5-inch and 6-inch webbing, a bibbed trawl, a two-seam design, and the same design with greater headrope flotation. The highest catches during the project were 20-40 pounds of sheepshead and black drum, produced by 30-minute test trawls. The test trawl catch commonly contained shrimp, croakers, mullet, stingrays, menhaden, and other herring-like species. The resulting fish trawl bycatch consisted mainly of stingrays, mullet, and small sheepshead. Very few, if any, shrimp, spotted seatrout, red drum or other commercial species were caught by the fish trawls, although these species were often found in the test trawl catch. This is remarkably different from shrimp trawling, making fish trawls extremely efficient gear. Over the 12 trawling days, 28,040 pounds of salable fish were caught, including test trawl catch. The total sheepshead catch was 25,618 pounds and the black drum catch was 2,422 pounds. At prices of $0.20/pound for sheepshead and $0.60/pound for black drum, over $6,500 was earned, not including the sale of a small amount of crab and mullet bycatch. Catching fish during colder weather, when the inshore shrimp season is nearly over, helped 12 vessel operators and their crews in Louisiana's Lake Borgne area salvage a below-average shrimp trawling season, allowing many of them economic survival through the following winter.     RECREATIONAL HOOK AND LINE FISHERIES The term "bycatch" is not usually associated with recreational fisheries in the popular media but has been the growing focus of researchers because of the magnitude of effort in the fisheries. The operative definition of bycatch used here and elsewhere is "those fish and shellfish which have no market value, are damaged during harvest, or cannot be legally sold or retained." The fishing event may cause either immediate mortality or the potential for future mortality as a result of gear interactions or handling. The release of undesired or non-legal catch by recreational fishers certainly fits this description of bycatch. Studies by Johnson and Griffith (1985), as well as many others have shown the preferences recreational fishers have for certain species over others in their catch are often based on flawed or incomplete knowledge. Many species considered "trash fish" are, in fact, quite edible and offer excellent sport for recreational fishing. Further complicating the issue of desirability are cultural and geographic differences in the rating of recreationally caught species. For example, smaller bluefish are considered relatively good table fare on the East Coast of the U.S., but are typically discarded as inedible in the northern Gulf of Mexico. Striped mullet (the target of a recreational cast-net fishery) are prized table fare in Florida, Alabama, and Mississippi, but are seldom eaten in Louisiana and Texas. Size and creel limits for many species contribute to a portion of the bycatch in recreational fisheries in the Gulf of Mexico known as "regulatory discards." According to National Marine Fisheries Service statistics, over 17 million saltwater fishing trips were taken by recreational anglers in 1995 by residents in four of the five Gulf states (data do not include Texas). Texas anglers spent 6.6 million days fishing in saltwater in 1991 (Maharaj and Athey, 1996). In 1995, recreational fishers in Alabama, Florida, Mississippi, and Louisiana landed about 136 million fish weighing over 73 million pounds. Recreational harvest of edible finfish species roughly equals the commercial harvest in the Gulf region. Thirty percent of all recreational angling trips in the nation occurred in the Gulf and these anglers landed 44 percent of all fish reported in 1995 (NMFS, 1996).   Campbell, R.P., and P.C. Choucair. 1995. Characterization of finfish bycatch of private-boat recreational anglers in Texas marine waters (S-K Final Report NA37FD0084). Texas Parks and Wildlife Department, 702 Navigation Circle, Rockport, TX 78382. Data were collected during the Texas Parks and Wildlife Department's routine harvest monitoring program. On 50 percent of the surveys, bycatch tally cards were given to anglers before the start of their fishing trip. Interviews were conducted at the completion of each fishing trip. On days when cards were given out (card days), cards were collected at the completion of a fishing trip and the information verified. On days when cards were not given out (non-card days), anglers were interviewed at the end of their trip and data were recorded on a bycatch data sheet. Two independent bycatch estimates were made using the same methodology employed to generate recreational harvest estimates. Total numbers of fish reported caught and released, and total number of species reported were determined and compared between card days and non-card days. A total of 719 survey days produced bycatch estimates of 3,105,100 fish for card days and 3,400,500 fish for non-card days. Estimated landings of fish during the same period were 1,380,200 fish. In addition, numbers of fish reported released again showed no apparent differences between card and non-card days for bays and passes or Gulf areas. Landings to bycatch ratios were 1:2.25 for card days and 1:2.46 for non-card days. Number of fish reported released per interview in bays and passes was 12.14 for card days and 13.11 for non-card days. Ten species of fish make up 96 percent of the total numbers reported caught and released on both card and non-card days. Spotted seatrout were the most reported caught and released species on both card days (51 percent) and non-card days (49 percent). This was followed by red drum (39 percent and 38 percent, respectively) and hardhead catfish (45 percent and 33 percent, respectively). Number of fish reported released per interview from Gulf waters was 8.47 on card days and 8.44 on non-card days. Ten species of fish make up 85 percent of the total numbers reported caught and released on card days and 82 percent on non-card days. Red snapper were reported caught and released the greatest number of times, 34 percent for card days and 28 percent on non-card days.   Clark, R.D. Jr. 1983. Potential effects of voluntary catch and release of fish on recreational fisheries. North American Journal of Fisheries Management 3:306-314. This study examines how the release of legal fish might affect a fishery. The classic yield-per-recruit model was modified so that total mortality rate (Z) was partitioned into three components: natural (M), fishing (F), and hooking (H) mortality rates. Another parameter (p) was used to represent the probability a legal fish was released when captured, to modify the levels of fishing and hooking mortality. Results indicated that the release of legal-size fish reduced the total mortality rates of the populations. As the release rates increased: 1) total catch and catch of trophy fish increased (i.e. fish harvested plus fish caught and released; 2) total harvest decreased; and 3) harvest of trophy fish remained relatively constant. The author states his belief that the effects of voluntary release can be assumed negligible if less than 10 percent of the legal fish caught are released, but release rates higher than 10 percent change the interpretation of conventional creel census estimates of catch and fishing mortality. The actual catch will be higher than indicated by a survey of fish in the creel, and the fishing mortality rates computed from these data will underestimate the true catch rate. Managers of sport fisheries need to estimate the voluntary release rate, along with harvest and fishing effort, if they want to assess a fishery accurately.   Edwards, R.E., A.P. McAllister, and B.D. Fortune. 1989. Billfish mortality and survivability. (Contract No. SA-88-07-MML). Mote Marine Laboratory, 1600 Ken Thompson Parkway, Sarasota, FL 34236. A total of eight blue marlin were tagged with ultrasonic transmitters, six by cooperating fishermen and two by the tracking boat. Six of the eight were successfully tracked. The fish ranged in size from 23 to 190 kg (50 to 420 pounds) and were caught on 50-130 pound test lines. None were hooked deep and none were observed to be bleeding at release. All six marlin were swimming strongly and were judged to be surviving and in good condition when tracking was terminated after periods ranging from 2.2 to 6.4 hours.   Edwards, R.E. 1992. Tarpon release mortality assessment using acoustic tracking. Final project report submitted to Florida Department of Natural Resources. Mote Marine Laboratory, 1600 Ken Thompson Parkway, Sarasota, FL 34236. Tarpon were tracked acoustically (ultrasonic transmitter tags) after they were caught and released by anglers in Boca Grande Pass, FL. A total of 26 of 27 tarpon were surviving at the end tracking periods that averaged about four hours and ranged up to 12.1 hours. Many tarpon remained in the general area of the pass during the tracking. Overall, the study documented the fact that tarpon are relatively hardy and have a high probability of surviving after being caught and released. It was suggested that anglers can decrease tarpon release mortality by minimizing handling.   Edwards, R.E. 1994. Development and evaluation of methods and protocols for determining acute mortality of released red drum, snook, Spanish mackerel and king mackerel. Mote Marine Laboratory Technical Report No. 380. Mote Marine Laboratory, 1600 Ken Thompson Parkway, Sarasota, FL 34236. A total of 24 snook ranging in size from 44 to 99 cm fork length were caught and tagged with transmitters in or near Gulf of Mexico waters. Of these, 23 were tracked for periods ranging from two to 15 hours. All fish were judged to be surviving at the end of the tracking period. A total of 23 red drum were similarly caught and tagged with transmitters from an area of shallow seagrass flats along the western side of Sarasota Bay. All were found to be moving and therefore surviving after periods ranging from two to 19 hours. A total of 29 Spanish mackerel were caught in the nearshore Gulf of Mexico, tagged and tracked for periods ranging up to 5.2 hours. Three were judged not to survive, based on their stationary position. During their tracking periods, the surviving fish moved distances from 0.1 to 2 nautical miles from their release points. A total of 20 king mackerel were caught using a variety of angling techniques and were tagged and tracked for periods up to 30 hours. Four fish died at or almost immediately after release. One fish was lost due to transmitter failure. Fourteen of the surviving fish were tracked for 1.7 to 2.2 hours, at the end of which they were 0.8 to 3.6 nautical miles from their release point. Total distances covered after release were typically much greater for both king and Spanish mackerels.   Edwards, R.E. 1996. King mackerel hooking mortality assessment (S-K NA37FD0087-01). Mote Marine Laboratory, 1600 Ken Thompson Parkway, Sarasota, FL 34236. A study was conducted to provide estimates of release mortality rates of king mackerel, and to augment previous estimates from a companion study of 18 percent mortality within the first two hours after release (see Edwards, 1994 above). In the present study, release mortality during the period from two to twenty-four hours after release was assessed. King mackerel were caught using normal angling techniques and gears, quickly tagged non-invasively with small acoustic transmitters, released, and followed for periods of up to 24 hours to determine whether they survived (as indicated by their continued movement) after being released. A total of 18 fish (17 king mackerel, one large Spanish mackerel) were caught and tracked during the summers of 1994 and 1995. Three fish died within two hours, and two were lost shortly after release due to problems with electronic equipment. Thirteen fish were tracked for more than two hours. Two tracks were abbreviated (3.4 and 10.4 hours) by storms. Of the remaining 11 fish, 9 were judged to have been definitely surviving at the end of the tracking, and 2 others were believed to have survived but were thought to have shed their transmitter tags after 17.9 and 19.9 hours. Combining the results of this study with the previous companion study provides an estimate of 18.2 percent for release mortality within the first two hours and a 95 percent confidence interval of 7 to 35.5 percent. Further combining these results with similar estimates for Spanish mackerel from the companion study provides an estimate of 16.7 percent with a 95 percent confidence interval of 8.6 to 27.9 percent. New information about king mackerel movement patterns was obtained from the paths of the tracked fish, which traveled at an average speed of 1.1 knots and covered distances of up to 32 nautical miles. (Dr. Edwards has also compiled an extensive bibliography of release mortality references for both fresh and salt water studies, some of which are reviewed here).   Fable, W.A. (Ongoing). Reef fish release mortality (1992-1994 progress reports). National Marine Fisheries Service Panama City Laboratory, 3500 Delwood Beach Road, Panama City, FL 32407. During 1992, six fishing trips with rod and reel provided 82 vermilion snapper caught from between 90 and 100 foot depths. Fifty-three of these fish were then punctured to relieve expanded gases from the swimbladder. Twenty of these 53 were punctured with a knife, as a fisherman might do, while the rest were punctured with a hypodermic needle. Fifteen of the 82 fish died before they could be released into holding pens at the Panama City Laboratory docksite. All surviving vermilion snapper were kept in pens for a minimum of 15 days to a maximum of 29 days. Of the 67 fish released into pens, six died the first day, and six more died over the duration of their captivity. Fifty-five fish survived the entire period and were released. This is an overall survival rate of 67 percent. In 1993, 10 groups of ten fish were caught, with half the fish having their swimbladders punctured. Fish were alternately punctured or not punctured after a randomly chosen start, whether they appeared to have an expanded swimbladder or righting difficulties or not. The fish were taken from 90-100 foot depths, returned to the laboratory and kept in floating pens for 14 days. The differences in overall survival between punctured and unpunctured fish was significant at variation = .05 with punctured fish exhibiting a higher survival rate. Overall survival rate was 52 percent.   Hegen, H.E., G.E. Saul, and G.C. Matlock. 1984. Survival of handled and tagged spotted seatrout held in wood and wire cages. Management Data Series No. 61, Texas Parks and Wildlife Department, 4200 Smith School Road, Austin, TX 78744. Spotted seatrout (Cynoscion nebulosus) were captured by hook-and-line in each of six Texas bay systems and placed in wood cages during July-September 1982. During December 1982-April 1983, spotted seatrout were captured by hook-and-line in each of seven bay systems and placed in wire cages. Within each cage type, there were no significant differences (P>0.05) in estimates of survival of handled and tagged fish held for seven days. Mean coastwide survival in wood cages ranged from 37.5 ± 16 (SE) percent to 42.5 ± 12 percent. Coastwide survival in wire cages ranged from 77.1 ± 13 percent to 85.7 ± 9 percent. Mean coastwide survival rates adjusted for controls was 74 percent and 95 percent for wood and wire cages, respectively. Computed survival in wood cages was probably less than in wire cages because some fish escaped the wood cages and were counted as mortalities and because wood cages were less stable than wire.   Jordan, S.R., and A.G. Woodward. 1992. Survival of hook-caught red drum. Proceedings of the Annual Conference of the Southeastern Association of Fisheries and Wildlife Agencies 46:337-344. Sub-legal red drum (<355 mm) collected with angling gear during 1988 and 1989 were held in confinement to evaluate post-hooking survival. During 1988, 217 fish were collected by hook-and-line and 65 control fish were collected with trammel and gill nets. In 1989, a total of 296 hook-caught and 103 net-caught red drum were used in the study. Short-term post-hooking survival was 84 percent in both years. Gear and techniques favored by Georgia anglers were used to collect experimental fish. Single hooks of various materials in sizes 4 to 3/0 were used on both float and bottom rigs deployed from small boats. Live and dead penaeid shrimp were used most often as bait. Experimental fish were examined, measured and tagged. Hook type, size and location were recorded, and the fish was temporarily placed in a nylon mesh holding net until sampling was completed. Location of the hook was defined as follows: maxilla, gill and esophagus. Most fish were hooked in the maxilla area and 92 percent of these individuals survived. Gill-hooked fish comprised 13 percent of the red drum collected and exhibited 68 percent survival. Only 10 percent were hooked in the esophagus; however, these fish had the lowest survival (47 percent). Red drum used as controls were collected with monofilament gill and trammel nets in the same area as the experimental fish. Fish were transported to the holding site in an aerated 114-liter cooler. Both control and hook-caught fish used in the study were held a minimum of 14 days before release. Angler recapture rates of marked hook-caught red drum were similar to those of marked net-captured fish during 1988.   Malchoff, M. 1995. Effects of catch and release angling on important northeast marine fishes: Mortality factors and applications to recreational fisheries (S-K NA36FD0102). New York Sea Grant Extension Program, 3059 Sound Avenue, Riverhead, NY 11901. An investigation into the level of post-angling, catch-and-release mortality of bluefish (Pomatomus saltatrix), striped bass (Morone saxatilis), black seabass (Centropristis striata), scup (Stenotomus chrysops), and weakfish (Cynoscion regalis) in the New York marine recreational fishery was undertaken. Forty angling trials aboard private, charter and party boats, as well as shore-based were conducted in which fish were caged following normal catch-and-release procedures. After 72 hours, mortality counts were made. Mean mortality estimates were 10.3 percent for bluefish, 5 percent for striped bass, and 9.9 percent for scup. Mortality for black seabass subject to a depressurization change of three atmospheres was 21.8 percent. Because of the limited sample size, meaningful estimates of weakfish mortality were not obtained.   Matlock, G.C., L.W. McEachron, J.A. Dailey, P.A. Unger, and P. Chai. 1993. Short-term hooking mortalities of red drums and spotted seatrout caught on single-barb and treble hooks. North American Journal of Fisheries Management 13:186-189. Short-term (three days) hooking mortalities of red drums (Sciaenops ocellatus) and spotted seatrout (Cynoscion nebulosus) caught with single-barb hooks were compared with hooking mortalities for those caught with treble hooks in Texas bays in the summers of 1989 and 1990. Natural baits (primarily penaeid shrimp) and artificial baits (spoons, worms, and fish-shaped lures) were used with both hook types. Overall mortality of 121 red drums was 4.1 percent at the end of three days after capture, and there were no significant (P=0.67) red drum mortality differences between hook types or bait types. Mortality of 124 spotted seatrout was 7.3 percent by three days after capture, and no significant (P=0.27) differences between hook types or bait types were found. Restricting anglers to one hook type (either single-barb or treble hooks) would not reduce unintended fishing mortality for either species.   Muoneke, M.I., and W.M. Childress. 1994. Hooking mortality: A review for recreational fisheries. Reviews in Fisheries Science 2(2):123-156. This is an excellent review of a number studies on catch-and-release (hooking) mortality gathered from the existing fisheries literature and from a survey of fisheries management in all 50 states, the U.S. government, all Canadian provinces, and selected academic and research institutions. Hooking mortality estimates were identified for 32 taxa. Most studies dealt with salmonids, centrarchids and percids. Within and among species, differences in percent mortality were reported in association with bait type (artificial vs. natural), hook type (number of hooks, hook size, and barbs), season/temperature, water depth (depressurization), anatomical location of hook wound, and individual size. Although most hooking mortalities occur within 24 hours, the use of initial plus delayed mortality provides a more complete estimate of mortality. Single hooks (especially when used in conjunction with natural baits) resulted in higher mortalities than treble hooks. Environmental conditions (notably high water temperature and low dissolved oxygen) are important to overall mortality related to hooking, playing and handling. Mortalities were highly variable; occasionally exceeding 30 percent among red drum (Sciaenops ocellatus), smallmouth bass (Micropterus dolomieu), largemouth bass (Micropterus salmoides), cutthroat trout (Oncorhynchus clarki), and catfishes (Ictaluridae). Mortalities were occasionally found to exceed 68 percent among spotted seatrout (Cynoscion nebulosus), bluegills (Lepomis macrochirus), crappies, (Pomoxis spp.), and striped bass (Morone saxatilis). (This report is perhaps most valuable for its summary of other studies and its extensive bibliography. A pertinent excerpt for the Gulf of Mexico region is provided below).   Mortality up to 55.6 percent was indicated for spotted seatrout caught using natural baits on treble hooks as well as artificial lures (Matlock and Dailey, 1981). In another study, Matlock et al. (1993) reported mortalities of 4.13 percent and 7.26 percent for red drum and spotted seatrout, respectively, caught with single and treble hooks on natural baits and artificial lures. Hegen et al. (1983) reported 37 percent mortality for spotted seatrout caught using a variety of gears. Martin et al. (1987) caught spotted seatrout, red drum and black drum using baited circle hooks on trotlines and attributed higher spotted seatrout mortality to hooking in vital organs. Summer mortality was 44.7 percent for red drum caught with natural baits or artificial lures in a freshwater reservoir (Childress, 1989a).   Render J.H., and C. A. Wilson. 1994. Hook-and-line mortality of caught and released red snapper around oil and gas platform structural habitat. Bulletin of Marine Science, 55(2-3): 1106-1111. The study was conducted on an oil and gas platform approximately 90 km south of Cameron, LA, in 21 m water depth. Red snapper were caught by hook-and-line, treated (control, gas bladder deflation, tagging, tagging with deflation) and released into a vertical holding net (9 m deep) for varying lengths of time (24, 30, 36, 48 h). Results indicated an average mortality rate of 20 percent at 21 m depth, with no significant difference between treatments or time-in-net. There was a significant difference in seasonal mortality, with higher mortalities observed in summer versus fall. The long-term effect of air bladder deflation on survival was investigated using 107 red snapper transferred to a large holding tank and held for 30 to 40 days. All fish were tagged for identification. Thirty-five of the red snapper had deflated air bladders, while 72 were tagged only. Gas bladder deflation did not significantly enhance survival of released red snapper at 21 m.   Render J.H., and C. A. Wilson. 1993. Mortality rate and movement of hook-and-line caught and released red snapper (MARFIN NA90AAHMF762 Final Report) Coastal Fisheries Institute, Louisiana State University, Baton Rouge, LA 70803. This report more fully documents the work referenced above, including two years of field work. Interestingly, this report indicates higher seasonal mortalities in the fall than in the summer (opposite of what was reported above) and postulates that colder water temperatures during that time of year may act as a stressor on red snapper. A total of 345 red snapper were tested for the effects of depth on mortality rates at five intervals ranging from 25 to 55+ m. The researchers also used ultrasonic tags as an alternative method to track released red snapper. Results from the variable depth tests did not show evident differences in apparent survival due to deflation until depths of approximately 30 to 40 m were reached. Based on these results it did not appear that air bladder deflation is an effective tool for enhancing survival of released red snapper. The researchers also pointed to some of the shortcomings of using ultrasonic telemetry to locate and position individual red snapper around an oil platform, including variable holding depth of the fish and signal bounce caused by the platform structure. The researchers postulated that the overall mortality rates (approximately 20 percent) were conservative due to the method used to obtain the estimates (i.e. releasing fish into an open net system and tallying the number of red snapper that were successful in submerging).   Saul, G.E. 1992. Recreational fishery by-catch in the Galveston Bay system. Publication GBNEP - 25. FTN Associates, Ltd. 7101 Highway 71 West, Suite 214, Austin, TX 78735. This report does not reflect new research, but provides a thorough review of existing research and data sets for the Galveston Bay, Texas region. A preliminary estimate of the recreational bycatch of sport fishermen was made using a combination of data obtained from the National Marine Fisheries Service's (NMFS) Marine Recreational Fishery Statistics Survey (MRFSS) and routine creel survey data provided by the Texas Parks and Wildlife Department (TPWD). Data from the MRFSS included landings of fish, determined to species by NMFS contractors, and bycatch data based on fishermen recall during intercept surveys. Data from TPWD included estimated annual landings of fish by sport fishermen as determined by TPWD fisheries professionals in intercept surveys. Recreational fishermen caught and released approximately two fish for every fish landed. Because of the limited nature of the data, estimates by species and year were not made. During 1979-1985, the years of concurrent data collection by NMFS and TPWD, it was estimated that sport fishermen caught and released between 1.2 and 3.5 million fish in the Galveston Bay system. Approximately 5 percent of the fish reported released were released dead. Available literature on hooking and handling mortality suggests that less than 15 percent of red drum released alive and up to 30 percent of spotted seatrout released alive die from injuries or stresses related to capture within seven days of being hooked, handled and released. TPWD biologists used sport-fishing techniques to capture spotted seatrout for tagging purposes. These "sport fishermen" had a lower total bycatch ratio than NMFS-surveyed fishermen, catching about one fish for every fish tagged. The author states: "If it is assumed that fishermen fishing specifically for spotted seatrout would retain other desired species, the estimated bycatch by these specialty fishermen would be even less." The author states his belief that typical sampling methods, such as intercept surveys conducted at the end of the fishing trip, do not provide verifiable data. Studies suggested to further explore recreational bycatch included: 1) limiting bycatch recall studies to those species under management regulations; 2) using professionals (e.g. TPWD, NMFS, university or other biologists) to emulate sport-fishermen; 3) using volunteer fishermen to record catch information in logbooks; and 4) conducting hooking and handling mortality studies of selected species. This report also includes a listing of all the bycatch literature holdings maintained by Natural Resources Consultants, Inc. in their Seattle, WA library (Appendix B-"Current North American and International Literature Survey on Fisheries Bycatch"). One interesting fact is apparent from reviewing this bibliography; although fisheries bycatch is a world-wide issue encompassing many fisheries and species, comparatively little work has been done in the Gulf of Mexico region (with the possible exception of the shrimp fishery) in addressing non-target species impacts.   Thomas, R.G., J. Lightner, V. Hebert, and E. Lear. (in review). Release mortality of red drum (Sciaenops ocellatus) and spotted seatrout (Cynoscion nebulosus) from four angling methods. Louisiana Department of Wildlife and Fisheries, Lyle S. St. Amant Marine Laboratory, P.O. Box 37, Grand Isle, LA 70358. A hook-and-release mortality study of red drum and spotted seatrout was conducted between May, 1993 and January 1995 at the Louisiana Wildlife and Fisheries, Lyle S. St. Amant Marine Laboratory. Mortality rates for four different common fishing methods were evaluated. Methods used were: single hook with live bait (SHB), treble hook with live bait (THB), single hook artificial lure (SHA), and treble hook artificial lure (THA). All fishing was done by boat within three miles of the laboratory. No special care was taken in the unhooking and handling process. Hooks were removed by hand in mouth-hooked fish, with pliers in more deeply hooked fish, and were left in place in the deepest hook sites. Fish were transported in 210-liter tanks with 32 liters per minute flow-through. On return to the laboratory, fish were netted into 11 1,900-liter tanks located on a covered wharf, each with 38 liters per minute flow-through. Holding periods of three days in summer and five days during colder months were interspersed with seven-day holding periods to assess delayed mortality. Water temperature, salinity and dissolved oxygen concentration in the holding tanks differed little from those parameters in the adjacent bay. Total catch of spotted seatrout was 1,512; most were taken in May-August, with live bait accounting for the most fish. Total catch of red drum was 743, taken primarily during July-February. Overall survival was 82.5 percent in seatrout and 97.3 percent in red drum. Nearly all mortality in both species occurred within the first 48 hours. Seatrout survival varied widely between methods: 97 percent (THA), 91 percent (SHA), 83 percent (THB), and 74 percent (SHB). Less variation in survival was seen between red drum method groups: 99 percent (SHA), 97 percent (THA), 96 percent (SHB), and 94 percent (THB). Each method was numerically well-represented in seatrout, but limited numbers of red drum were taken with treble hooks. These results indicate that far fewer red drum and spotted seatrout succumb following hook-and-release than commonly believed, and that the use of treble hooks doesn't induce excessive mortality in released fish. The survival rate of released red drum below the minimum size limit (406 mm) was not significantly different from that of legal-sized fish (98 percent vs. 96 percent). Seatrout smaller than the 305 mm limit were more likely to live (87 percent) than were legal fish (81 percent).   Virginia Sea Grant. 1995. Release mortality in marine recreational fisheries. Waterfront News Vol. 2, Nos. 1 & 2, Spring/Summer 1995. Virginia Institute of Marine Science, Gloucester Point, Virginia 23062. This quarterly newsletter provides a good overall summary of a recent conference entitled "Release Mortality in Marine Recreational Fisheries: Current Research and Fishery Management Implications." The conference was held in Virginia Beach, Virginia on May 8-10, 1995. Following is a list of summaries pertinent to species which can be found in the Gulf of Mexico region:   Field and Laboratory Investigations into Striped Bass Mortalities Following Angler Release was addressed by Eric May of the Maryland Department of Natural Resources. May undertook laboratory research to determine the impact of temperature, salinity and fish size on mortality. He found the most significant factor affecting release mortality was temperature. Larger fish (>21"), lower salinities (1.0 ppt) and high temperatures (>21 degrees C) equal higher mortality rates.   Survival Estimates for Demersal Reef Fishes Released by Anglers was addressed by Mark Collins of the South Carolina Marine Resources Institute. He sampled various fish species caught in three depth zones, 21 m, 36 m and 46-54 m, off central South Carolina. The fish were immediately observed for survival in holding tanks, survivors placed in weighted mesh cages and again observed for survival 24 hours later. Immediate survival of the fish was determined by either their ability to swim to the bottom of the holding tank (possible survivors), or their floating on top of the water (probable mortalities). Collins sampled more than 15 species and found the "swim or float" method of predicting survival to be valid for all species and depths except black sea bass in 36 m (119 ft.). Pooled estimates of survival for all species were approximately 88 percent at 21 m, 81 percent at 36 m, and 62 percent at 46-54 m. Survival differed by species, but in general, release mortality increases with increased depth of catch.   Factors Affecting Short and Long Term Hooking Mortality in Virginia's Recreational Fishery for Summer Flounder was the title of research results presented by Jon Lucy and Tracy Holton of the VIMS Sea Grant Marine Advisory Program. Fish caught with an otter trawl are transported to a holding facility and then caught with live bait on hook-and-line. Variables being addressed are hook type, leaving hooks in place in deeply hooked fish vs. removing them, crimping or not crimping the hook barb, location of the hook wound, the size of the fish, and the water temperature. Results so far show an overall release mortality of 18 percent. The most important variables affecting mortality are location of hook wound and water temperature. Of the fish that have been deeply hooked or swallowed the hook, about 50 percent died.   The use of ultrasonic tagging in real-world situations to study release mortality was discussed by two researchers, Randy Edwards (see above reference in this category) and Phillip Bettoli. Edwards reported survival rates for blue marlin, tarpon, red drum and snook were between 96 percent and 100 percent. Spanish and king mackerel had survival rates of 88 percent and 83 percent respectively. Bettoli reported that mortality rates for striped bass in a Tennessee reservoir were strongly influenced by air temperature at time of capture. Mortality rates were approximately 15 percent in November and December, but went as high as 67 percent in July and August. One surprising result of the striped bass research was that the condition of the fish when initially released (i.e. "good" or "poor") was not an accurate indicator of whether the fish lived or died.   Wilson, R.R. Jr., and K.M. Burns. 1996. Potential survival of released groupers caught deeper than 40 m based on shipboard and in-situ observations, and tag-recapture data. Bulletin of Marine Science. 58(1):234-247. (This study is reported under this heading because of the way the subject fish were captured. However, this work also applies to the commercial grouper longline fishery in the western Gulf). In this study, shipboard and in-situ observations were used to determine the potential post-release survival rate of groupers, chiefly red grouper (Epinephelus morio), caught from between 44 and 75 m on the central west Florida shelf. Potential survival rates were then further evaluated in combination with data from a tag and recapture study (3,818 releases) in the same area and time period. Potential survival rates for released red grouper and scamp (Mycteroperca phenax) caught shallower than 44 m were very high (86 percent to 100 percent) for up to eight days following release. Undersized grouper (<50.8 cm) caught from both shallower and deeper than 44 m, then tagged and released, were found to survive long enough to reach legal size. For grouper caught deeper than 44 m, however, tag/recapture data and in-situ observations indicate that potential survival rates are too low (<33 percent) for the 50.8 cm (20 inches) size rule to be effective in increasing yield.     SUMMARY AND RECOMMENDATIONS One of the most serious challenges in addressing Gulf of Mexico fisheries bycatch is the lack of effective communication among researchers, fishery managers, fishery user groups, the environmental community and other stakeholders. This can lead to the deliberate or inadvertent release of misleading information. The new era of electronic communication offers a great potential for the future, but currently is not widely used by researchers and is still inaccessable to many user groups and the general public. Electronic publication with extensive keyword indices would also facilitate information and data-sharing among those with access to computer search capabilities. Research associated with this document pointed to a serious lack of computer-available reports. Relatedly, the authors also found it somewhat vexing that the state of Texas was not included in data generated by the Marine Recreational Fisheries Statistics Survey for the Gulf of Mexico region. Removing these types of bottlenecks which adversely affect information sharing and distribution capabilities should be of the highest priority. Faced with the fiscal reality of dwindling resources for bycatch-related research, it is important for researchers to concentrate their efforts on fisheries which generate the most impacts on fish stocks. This can be determined by amount of participation in the fishery, type of gear used, landings data, relative impacts on non-target stocks, and existing or incipient management regimes under which the fisheries are prosecuted. Based on the studies reviewed herein, three fisheries emerge as appropriate areas for more intense study in the Gulf of Mexico region. They are, in no particular order of significance: 1) the commercial shrimp trawl fishery, due to the inherent non-selectivity of small-mesh trawl gear; 2) the commercial hook-and-line reef fish fishery, due to unanswered questions concerning regulatory discards; and 3) the recreational hook-and-line fishery, due to the vast number of participants and the cumulative pressures on non-target species and non-legal target species. In the Gulf of Mexico region, the majority of bycatch research efforts to date have been directed at the commercial shrimp trawl fishery. While species characterization and magnitude surveys are current, and preliminary research into gear development for bycatch reduction looks promising, much remains to be done in the realm of information and technology transfer to fishermen and other stakeholders such as environmental groups. Further, a baseline must be established from which to measure any bycatch reductions attributable to management measures. Some existing research indicates TEDs are also effective at reducing bycatch; will gear-related reductions be measured against pre-TED or post-TED levels? How will changes in effort in the fishery in both the near term and the future be factored in to the equation? What will change on an ecosystem-wide basis as those species currently being removed by the fishery become available elements of predator/prey relationships and enter into the food chain at one level while simultaneously being removed at other levels? Gear development, installation and operation procedures will continue to evolve as the seafood industry takes a more active role in BRD development in the same manner as it did with TEDs. This information will need to be shared on an industry-wide basis to mitigate any potential adverse economic impact on shrimp producers. With respect to the literature reviewed, the large variation in mortalities within and between experiments suggests the need for a standard experimental protocol for setting up, conducting and monitoring survival experiments. This should include some index of bycatch condition prior to capture so that comparisons can be made between wild and experimental organisms. Although injury type and location are recorded by most authors, a classification system should be set up to record the causes and severity of damages incurred. The same approach should be taken to identify the range and severity of stressors organisms may be subjected to during capture and escape or release. Longer term studies should be aimed at measuring the individual and cumulative effects of stressors. Mortalities which occur after escape or release have been attributed to stress associated with the capture process. Regulatory measures to increase the escape of bycatch such as the use of bycatch reduction devices (BRDs) might also result in increased levels of mortality. Using fishing gear selectivity as a fishery management tool without adequate research into the condition of the organisms escaping from fishing gear might not be the most effective way to protect bycatch species (Chopin and Arimoto, 1995). With the exception of the shrimp fishery, the other fisheries examined had a small bycatch as a percentage of total catch, and mortality associated with this bycatch (or regulatory discard) was low compared to other sources of mortality. There are some concerns centering on the accuracy of stock assessment parameters currently being used, especially regarding sources and magnitudes of mortality associated with activities other than fishing (e.g. habitat loss, pollution) for both targeted species and bycatch species. The significance of bycatch mortality relative to natural mortality is not well understood for most bycatch species. It is apparent that, in many cases, the bycatch issue is being driven by perception. Reducing bycatch is a desirable goal, if for no other reason than maintaining the sustainability of future harvests. However, it appears that except in the case of a few species, there are no documented negative impacts on finfish stocks because of bycatch, and this point must be made clear. Those species which are of concern in the Gulf of Mexico, such as red snapper, king mackerel, and Spanish mackerel, are harvested in several directed fisheries by a variety of user groups, in addition to being taken as bycatch in the shrimp trawl fishery. Undersized fish or fish released in excess of bag and trip limits are also "bycatch" in the directed fisheries. Future management measures and research should address the broad spectrum of factors affecting the status of fishery stocks, of which bycatch is but one component. This type of ecosystem-based approach to research and management is needed in the Gulf of Mexico because of the enormous species diversity and broad spectrum of fisheries in the region.   Additional References: Alabama Department of Conservation and Natural Resources. 1995. Commercial and recreational shrimping regulations. Arfman, C. 1995. Billfish tournament deaths. Taranaki Polytechnic, New Zealand. (Internet Posting, c.arfman@taranaki.ac.nz). Baxter, B., and S. Keller (eds.). 1996. Solving bycatch: considerations for today and tomorrow. Alaska Sea Grant College Program Report No. 96-03. University of Alaska Fairbanks. Branstetter, S. 1995. An industry workshop addressing bycatch issues in southeastern U.S. fisheries: A report to the National Marine Fisheries Service. Gulf & South Atlantic Fisheries Development Foundation, Inc., Tampa, Florida. Bull, A.S. and J.J. Kendall, Jr. 1992. Estimation of fisheries impacts due to underwater explosions used to sever and salvage petroleum platforms. Minerals Management Service, New Orleans, Louisiana. Caillouet, C.W. Jr., M.J. Duronslet, A.M. Landry, Jr., D.B. Revera, D.J. Shaver, K.M. Stanley, R.W. Heinly, and E.K. Stabenau. 1991. Sea turtle stranding and shrimp fishing effort in the northwestern Gulf of Mexico, 1986-89. Fish. Bull. U.S. 89:712-718. Chopin, F.S. and T. Arimoto. 1995. The condition of fish escaping from fishing gears--a review. Fisheries Research 21:315-327. Christmas, J.Y., Etzold, D.J., McIlwain, T.D. and L.B. Simpson. (eds.) 1985. Marine fisheries initiative: Gulf of Mexico phase. Gulf States Marine Fisheries Commission, Ocean Springs, Mississippi. Coleman, E. 1995. A stitch in time...reducing bycatch now. Coast & Sea (Winter 1995). Louisiana Sea Grant College Program, Louisiana State University, Baton Rouge, Louisiana. Crouse, D.T., Donnelly, M., Bean, M.J., Clark, A., Irvin, W.R. and C.E. Williams. 1992. The TED experience: claims and reality. (Internet abstract). Environmental Defense Fund, New York, New York. Dewees, C.M. and E. Ueber. 1990. Effects of different fishery management schemes on bycatch, joint catch, and discards: summary of a national workshop. California Sea Grant College, University of California, La Jolla, California. Edwards, R.E. 1995. Reduction of billfish bycatch in longline fisheries: feasibility of gear modifications based on differences in feeding patterns. (MARFIN NA47FF0019 {proposal}). Edwards, R.E. 1993. Reduction of bluefin tuna and undersize swordfish bycatch in Atlantic longline fisheries. (S-K NA57FF0031 {proposal}). Fable, W.A. Jr. 1980. Tagging studies of red snapper (Lutjanus campechanus) and vermilion snapper (Rhomboplites aurorubens) off the south Texas coast. Contributions in Marine Science, Vol. 23:115-121. FAO Fisheries Technical Paper No. 339. 1994. A global assessment of fisheries bycatch and discards. Unipub, Lanham, Maryland. Florida Administrative Code. Chapter 46-31: Shrimp. Vol. 15:375, 382-3. (Revised 7/96). Fowle, S. and R. Bierce. 1992. Proceedings of the shrimp trawl bycatch workshop, November 22-23, 1991 St. Petersburg, Florida. Center for Marine Conservation, Washington, D.C. Galloway, B.J., J.D. Bryan, L.R. Martin, and J.G. Cole. Sea turtle and shrimp fishery interactions-is a new management strategy needed? A report to the Texas Shrimp Association, P.O. Box 1020, Aransas Pass, Texas 78336. Georgia Department of Natural Resources. 1993. Management plan for reducing bycatch in the Georgia shrimp trawl fishery. Gough, B. 1996. Flounder rescue proposed. South Texas Outdoors with Buddy Gough. (Internet posting, Vaughn@caller.com). Green, D. 1995. Blue Water Offshore Fishing News. Vol. 14, No. 29, December 26, 1995. Gulf States Marine Fisheries Commission. 1996. Licenses/fees required by Alabama, Florida, Louisiana, Mississippi and Texas in their marine waters. Ocean Springs, Mississippi. Gulf States Marine Fisheries Commission. 1995. The striped mullet fishery of the Gulf of Mexico, United States: A regional management plan. (No. 33) GSMFC, Ocean Springs, Mississippi. Gulf States Marine Fisheries Commission. 1995. The Menhaden fishery of the Gulf of Mexico, United States: A regional management plan. (No. 32) GSMFC, Ocean Springs, Mississippi. Harrison, W.C. 1996. Gillnet restrictions in the Gulf of Mexico. Water Log. Vol. 15, No. 4.:1-6 Mississippi-Alabama Sea Grant Legal Program. University of Mississippi Law Center. University, Mississippi. Hein, S. And P. Meier. 1995. Skimmers: their development and use in coastal Louisiana. Marine Fisheries Review 57(1):17-24. Henwood, T.A., and W.E. Stuntz. 1987. Analysis of sea turtle captures and mortalities during commercial shrimp trawling. Fish. Bull. U.S. 85(4):813-817. Henwood, T.A., W.E. Stuntz, and N. Thompson. 1992. Evaluation of U.S. turtle protective measures under existing TED regulations, including estimates of shrimp trawler related mortality in the wider Caribbean. NOAA Tech. Memo. NMFS SEFSC 303. Hoagland, P. and Y. Kaoru. 1995. Economic valuation of marine recreational fishing for large pelagic species (S-K NA37FD0104). Woods Hole Oceanographic Institution, Woods Hole, Massachusetts. Johnson, J.C. and D.C. Griffith. 1985. Perceptions and preferences for marine fish--A study of recreational fishermen in the southeast. UNC Sea Grant College Publication UNC-SG-85-01. Institute for Coastal and Marine Resources, East Carolina University, Greenville, North Carolina. Lancaster, F. 1996. (telephone interview). Raffield Fisheries, Port St. Joe, Florida. Louisiana Department of Wildlife and Fisheries. 1996. Louisiana fishing regulations:1996. Maharaj, V. and M. Athey. 1996. Compendium of sportfishing statistics. American Sportfishing Association. Alexandria, Virginia. Mississippi Department of Wildlife, Fisheries, and Parks. 1996. Mississippi saltwater fishing 1996 rules/regulations. National Marine Fisheries Service. 1996. Marine recreational fishery statistics survey. U.S. Department of Commerce, Silver Spring, Maryland. National Marine Fisheries Service. 1996. Current fishery statistics No. 9500. Fisheries of the United States, 1995. U.S. Department of Commerce, Silver Spring, Maryland. Natural Resources Consultants, Inc. 1991. Hooking mortality study final project report: A report on an expanded study to determine rates and causes of mortality in coho and chinook salmon caught and released with sport fishing tackle. NRC, Seattle, Washington. Natural Resources Consultants, Inc. 1994. 1992-1993 Hooking mortality study final project report: A report on a study to determine rates and causes of mortality in coho and chinook salmon caught and released with sport fishing tackle. NRC, Seattle, Washington. Oviatt, C.A. 1977. Menhaden, sport fish, and fishermen. Marine Technical Report 60. Graduate School of Oceanography, University of Rhode Island, Kingston, Rhode Island. Pikitch, E.K. 1995. Bycatch research. Fisheries Forum. School of Fisheries, College of Ocean & Fishery Sciences, University of Washington. Prado, J. 1993. Selective shrimp-catching devices: A review. INFOFISH International 1/93. Quebec Federal Fisheries Development Program. 1993. The Nordmore grate: adaptation and efficiency in the commercial fishery. Department of Fisheries and Oceans, Quebec, Canada. Reese, G.B. 1990. 1990 Annual report: small pelagics (coastal herrings, butterfish and associated species). National Marine Fisheries Service, Southeast Fisheries Science Center, Mississippi Laboratories, Pascagoula, Mississippi. Risenhoover, A.D. 1990. Federal policy and the Endangered Species Act: the politics, perceptions, and technologies of protecting sea turtles. MS thesis, Dept. Wildl. & Fish. Sci., Texas A&M University, College Station, Texas 77843.