Previous Page Table of Contents Next Page


Regional Break-Outs

Table 5 documents discard weight by major world region. As can be seen, just over one-third of total discards originate in fisheries conducted within the Northwest Pacific. Key contributors to this regional total of 9.1 of million mt are the large crab, true and jack mackerels (Scomber spp. and Trachurus spp.), Alaska pollock, and cod fisheries of the region, as well as discards associated with the roughly 560,000 mt of wild shrimp harvested in the area.

Table 5. Discard weight* by major world region. Source: NRC bycatch database.
AreaDiscard Weight (mt)
Northwest Pacific9,131,752
Northeast Atlantic3,671,346
West Central Pacific2,776,726
Southeast Pacific2,601,640
West Central Atlantic1,600,897
West Indian Ocean1,471,274
Northeast Pacific924,783
Southwest Atlantic802,884
East Indian Ocean802,189
East Central Pacific767,444
Northwest Atlantic685,949
East Central Atlantic594,232
Mediterranean and Black Sea564,613
Southwest Pacific293,394
Southeast Atlantic277,730
Atlantic Antarctic35,119
Indian Ocean Antarctic10,018
Pacific Antarctic109

* Includes bycatch landed but unreported by species in industrial fisheries.

Discards associated with haddock, whiting, cod, pout (Trisopterus esmarkii), plaice, and other flatfish fisheries make a significant contribution to the 3.7 million mt regional bycatch total for the Northeast Atlantic, a total ranking the region as the second largest source of total removals. Capelin (Mallotus villosus), with measured discard rates of from 44% to 133%, were also a surprising contributor to aggregate bycatch in the Northeast Atlantic. Discards in this fishery were attributed primarily to size, quality, and market-condition parameters.

Discards in shrimp fisheries accounted for more than 50% of the total bycatch in West Central Pacific fisheries (Table 4). Shrimp activities off Thailand, Indonesia, and the Philippines produced the bulk of this shrimp fishery discard. Discards associated with harvests of scad, crabs, and tuna also added significantly to the regional total.

The Southeast Pacific ranked fourth among world fishing areas in terms of discards, with 2.6 million mt. Even at very low bycatch rates of 1% to 3%, the large anchoveta and pilchard fisheries in the area produced several hundred thousand tons of discards. Shrimp and mackerel fisheries also added sizable amounts of discard.

Wild shrimp fisheries were responsible for roughly 50% and 80% of regional discard totals in the West Indian Ocean and West Central Atlantic, respectively. In the West Indian Ocean, Indian and Pakistani shrimp fisheries represented the largest sources of discards, whereas in the West Central Atlantic, shrimp fisheries off the southeastern U.S. and in the Gulf of Mexico were responsible for most of the reported bycatch. Crab fisheries also added substantial quantities of discards to the West Central Atlantic total, while various species of carangids and mugilids, considered together, made a sizable contribution to aggregate removals from the West Indian Ocean.

Throughout the temperate and subarctic North Pacific and Atlantic region, discards were generally dominated by sublegal and legal sizes of commercially important cods (Gadidae), right-eyed flounders (Pleuronectidae), hakes (Merluccidae), and rockfishes (Scorpaenidae). Exceptions included discards of various species of sharks, eel pouts (Zoarcidae), and sculpins (Cottidae) and a wide variety of invertebrates. By contrast, in the subtropical and tropical regions of the world, a significant portion of the discards occurring in the shrimp fisheries involved small jacks, pompanos (Carangidae), goatfishes (Mullidae), lizardfishes (Synodontidae), mojarras (Gerreidae), threadfins (Nemipteridae), tooth ponies (Leiognathidae), left-eyed flounders (Bothidae), and rays (Dasayatidae) (Figure 3). Discards from most tropical shrimp fisheries are mainly comprised of species maturing at sizes under 20 cm and predominantly weighing under 100 grams, or about 1/5 of a pound. However, Conolly (1992) states that off Brazil 50% of the bycatch species are made up of juvenile commercial species.

Figure 3

Figure 3. Major discard families by region. Source: NRC bycatch database.

Species/Group Break-Outs

Table 6 documents global discards on the basis of the FAO International Standard Statistical Classification of Aquatic Animals and Plants (ISSCAAP) species groups.

Table 6. Global marine discards* on the basis of the FAO International Standard Statistical Classification of Aquatic Animals and Plants (ISSCAAP) species groups.
ISSCAAPMean Discard Weight (mt)Landed Catch Weight (mt)Ratio of Discarded Weight to Landed WeightRatio of Discarded Weight to Total Weight
Shrimps, prawns9,511,9731,827,5685.200.84
Redfishes, basses, congers3,631,0575,739,7430.630.39
Herrings, sardines, anchovies2,789,20123,792,6080.120.10
Jacks, mullets, sauries2,607,7489,349,0550.280.22
Cods, hakes, haddocks2,539,06812,808,6580.200.17
Miscellaneous marine fishes992,3569,923,5600.100.09
Flounders, halibuts, soles946,4361,257,8580.750.43
Tunas, bonitos, billfishes739,5804,177,6530.180.15
Squids, cuttlefishes, octopuses191,8012,073,5230.090.08
Lobsters, spiny-rock lobsters113,216205,8510.550.35
Mackerels, snooks, cutlassfishes102,3773,722,8180.030.03
Salmons, trouts, smelt38,323766,4620.050.05

* Includes bycatch landed but unreported by species in industrial fisheries.

Immediately apparent from the table, and of little surprise, is the contribution of discards in shrimp fisheries to global removals. Using estimates of mean species discards as the basis for comparison, shrimp fisheries account for about 9.5 million mt of the 27.0 million mt of discards, or 35% of global commercial fisheries discards. Furthermore, out of the 15 ISSCAAP species categories, discards in shrimp fisheries represented more than one-third of the group discard total. In eight of the 18 FAO statistical areas evaluated, shrimp fisheries accounted for more than one-third of the total discards, and in four areas, for more than two-thirds of the discard. Shrimp fisheries, as a gear type, were also characterized by the highest average discard ratios.

The second highest average discard ratio was also associated with a shellfish group--crab. However, the low total harvest of crab drops their global rank in terms of aggregate discard removals to fourth. In the North Pacific crab fisheries, sublegal and immature crab comprise the majority of discards.

The highest discard ratio associated with a finfish species group is tied to flatfishes, although this species group ranks sixth on the list of total removals due to its lower landed total. More than two-thirds of removals of this species groups originate from within the Northeast Atlantic and Northeast Pacific.

Jacks, mullets, and sauries account for 2.6 million mt of bycatch from a landed total of 9.3 million mt. The total is largely a function of documented rates for capelin discards ranging from 44% to 133%. As mentioned already, discards in this fishery were attributed primarily to quality and market-condition parameters. Other species in this group exhibit bycatch rates of less than 5%.

Discards associated with ISSCAAP 32 (cods, hakes, and haddocks) contribute 2.5 million mt of bycatch to the global total. Whiting, cod (Atlantic and Pacific), pout, and Alaska pollock fisheries account for the vast majority of these removals. The large whitefish fisheries of the Northeast Atlantic are alone responsible for 52% of the global total for this species group.

Redfishes and herrings, sardines, and anchovies round out the set of species groups with aggregate bycatch weights in excess of 1 million mt. Despite the very low bycatch ratios associated with the schooling species in ISSCAAP 35, the high annual harvest volumes of this group (23.8 million mt average over the 1988–1990 period) result in substantial discard volumes.

Other species groups of note are the tunas and billfishes with an aggregate bycatch weight of 739,580 mt. This figure may rise, however, if purse seine sets on logs or other fish-aggregating devices (FAD) account for an increasing proportion of total sets in the future. While dolphin removals on log/FAD sets are much lower than when setting on dolphin pods, aggregate removals of other species are much higher. This characteristic will be discussed more fully in an upcoming section covering ecological impacts.

Squid driftnet fisheries have received a great deal of attention in recent years owing to their capture of species having low reproductive rates and relatively long life spans. As documented in Table 6, mean discard weight in all squid fisheries (trawl, jig, and formerly net fisheries) totals approximately 191,801 mt, or 9% of the landed weight, substantially below the global all species rate of 35.1%.

Fishery-Specific Discard Ratios

Tables 7a and 7b identify the twenty fisheries with the highest recorded discard ratios by weight or by numbers. Both formats are presented because in some cases it was impossible accurately to convert rate estimates in one format into values for the other. Consequently, a fisheries presence on one list and absence from the other does not necessarily indicate its rank according to the latter format is greater than 20 or higher. To avoid such confusion in the future, bycatch researchers should be encouraged to document data in a manner allowing for ready conversion from one measure to the other.

Table 7a. Top twenty fisheries with the highest recorded discard ratios by weight (discard weight per landed target catch weight). Source: NRC bycatch database.
Fishery DescriptionKg Discard per Kg Landed
Trinidadian Shrimp Trawl14.71
Indonesian Shrimp Trawl12.01
Australian Northern Prawn Trawl11.10
Sri Lankan Shrimp Trawl10.96
U.S. Gulf of Mexico Shrimp Trawl10.30
Sea of Cortes Shrimp Trawl9.70
Brazilian Shrimp Trawl9.30
West Indian Shrimp Trawl8.52
U.S. Southeast Shrimp Trawl8.00
Northwest Atlantic Fish Trawl5.28
Persian Gulf Shrimp Trawl4.17
Southwest Atlantic Shrimp Trawl4.10
East Indian Shrimp Trawl3.79
Bering Sea Sablefish Pot3.51
Malaysian Shrimp Trawl3.03
Senegalese Shrimp Trawl2.72
Bering Sea Rock Sole Trawl2.61
British Columbia Cod Trawl2.21
Gulf of Alaska Flatfish Trawl2.08
Northeast Atlantic Dab Trawl2.01
Table 7b. Top twenty fisheries with the highest recorded discard ratios by number (discard number per landed target catch number). Source: NRC bycatch database.
Fishery DescriptionNumber Discard per Number Landed
West Central Atlantic Shrimp Trawl12.13
Bering Sea King Crab Pot9.71
California Halibut Net4.83
Northeast Atlantic Whiting Trawl2.83
Bering Sea Tanner Crab Pot2.34
Northeast Atlantic Haddock Trawl1.94
Arabian Gulf FinfLsh Trawl1.75
Northeast Atlantic Nephrops Trawl1.70
East Central Pacific Spiny Lobster Pot1.68
East Central Pacific Swordfish Longline1.58
Northeast Atlantic Hake Trawl1.18
East Indian Ocean Tuna Longline1.13
Northeast Atlantic Cod Danish Seine0.79
Northeast Atlantic Haddock Danish Seine0.70
East Central Pacific Slipper Lobster Pot0.67
Northeast Atlantic Whiting Danish Seine0.64
Northeast Atlantic Cod Trawl0.51
Northeast Atlantic Plaice Trawl0.42
Carribbean Tuna Longline0.40
Japanes High Seas Squid Net0.39

Shellfish fisheries (shrimp, crab, Nephrops) as a whole comprise 14 of the top 20 weight-based discard ratios and 20 of the top 40 ratios when looking across both classifications. Shrimp fisheries, of course, predominate, with shrimp trawl ratios accounting for the top nine weight-based ratios and the highest numbers-based estimate. Only six finfish fisheries occur on the list of highest weight-based discard ratios.

In terms of gear type, trawl fisheries dominate the weight-based breakout (19 of the top 20). The top discard ratios are more evenly distributed across gear types in the numbers-based list. Line and net fisheries, absent from the top 20 weight-based list, contribute five ratios to the top 20 numbers-based set. Four pot fisheries and three Danish seines are also a part of the numbers-based breakout.

Trawl fisheries, however, are also well represented on lists of the ten lowest weight-based and numbers-based discard ratios (Tables 8a and 8b). Eight of the ten lowest weight-based records are associated with trawl fisheries. One pot fishery for Pacific cod and an Atlantic menhaden (Brevootia tyrannsus) seine fishery round out the bottom ten list. In terms of numbers, driftnet fisheries for tuna and squid constitute five of the ten lowest ratios and are joined by two midwater Alaska pollock fisheries, one Pacific cod pot fishery, and one ETP tuna purse seine fishery.

Table 8a. The ten lowest observed weight-based discard ratios in fisheries other than shrimp (discard weight per landed target catch weight). Source: NRC bycatch database.
Fishery DescriptionKg Discard per Kg Landed
Northwest Atlantic Hake Trawl0.011
West Central Atlantic Menhaden Seine0.029
Bering Sea Cod Pot0.041
Northeast Pacific Whiting Trawl0.043
Northwest Atlantic Cod Trawl0.058
Bering Sea Pelagic Pollock Trawl*0.062
Northwest Atlantic Redfish Trawl0.063
Northeast Atlantic Groundfish Trawl0.083
Gulf of Alaska Midwater Pollock Trawl0.086
Northwest Atlantic Plaice Trawl0.118

* Includes some on-bottom fishing.

Table 8b. The ten lowest observed numbers-based discard ratios in fisheries other than shrimp (discard number per landed target catch weight). Source: NRC bycatch database.
Fishery DescriptionNumber Discarded per Number Landed
Bering Sea Midwater Trawl Pollock0.005
Northeast Atlantic Tuna Driftnet0.009
Gulf of Alaska Midwater Trawl Pollock0.018
(Korea) Northwest Pacific Squid Driftnet0.037
(Taiwan) Northwest Pacific Squid Driftnet0.068
Sub-Tropical Convergence Zone Tuna Driftnet0.08
Tasman Sea Tuna Driftnet0.123
Bering Sea King Crab Pot0.132
Eastern Tropical Pacific Tuna Purse Seine0.18
Bering Sea Cod Pot0.18

Gear-Specific Discard Rates

Tables 9a and 9b take a closer look at weight-based and numbers-based bycatch ratios, respectively, from a gear-specific perspective. Although differences in discard ratios can be observed between gear types, four major ratios groups can be identified. Shrimp trawls stand alone at the top of the list of high discard ratios, while relatively low ratios are reflected for pelagic trawls, purse seines targeting on menhaden, sardines, and anchoveta, and some of the high seas driftnet fisheries. Between these high and low values lie two groups. The first of these is comprised of bottom trawls, unspecified (probably near-bottom) trawls, longline gear, and the majority of the pot fisheries. The remaining group falls in between the very low ratios of the midwater trawls group and the moderate ratios of the aforementioned bottom trawl/line/pot group. This last set of gear types includes the Japanese high seas driftnet fisheries, Danish seines, and purse seines for capelin.

The highest discard ratios by region, across gear type, are presented in Table 10. In the Northwest Atlantic, rates are led by the 1.58 kg discard per kg of target landings in the New England yellowtail flounder trawl fishery, while a dab trawl rate of 2.01 kg discard/kg target is at the top of the Northeast Atlantic list. Three gear types are represented in the top five list for the Mediterranean/Black Sea region, while shrimp trawls alone comprise the set of five highest records for the West Central Atlantic and West Central Pacific. In the Northeast Pacific, the region containing perhaps the most comprehensive discard database, the top five rates range from 3.51 kg discard/kg target in the Bering Sea sablefish pot fishery to 2.08 kg discard/kg target in the Gulf of Alaska flatfish trawl fishery. Regions for which data gaps preclude a full set of records are the West Central, East Central, and Southwest Atlantic, the Eastern Indian Ocean, the Eastern Central Pacific, and the Southwest Pacific.

Table 9a. The top weight-based discard to landed target catch ratios by gear type. Source: NRC bycatch database.
Fishery DescriptionKg Bycatch per Kg Landed
Non-Pelagic Fish Trawl 
 Northwest Atlantic Fish Trawl5.28
 Bering Sea Rock Sole2.61
 British Columbia Pacific Cod2.21
 Gulf of Alaska Flatfish2.08
 Northeast Atlantic Dab2.01
Pelagic Fish Trawl 
 Bering Sea Pollock-19880.01
 Northeast Atlantic Cod0.00
 Bering Sea Pollock-19890.00
 Gulf of Alaska Pollock-19890.00
 Bering Sea Pollock-19870.00
Shrimp Trawl 
 Sri Lanka10.96
 U.S. Gulf of Mexico10.30
 Eastern Central Pacific Swordfish-19901.13
 Bering Sea Greenland Turbot1.03
 Eastern Central Pacific Swordfish-19911.00
 Bering Sea Sablefish0.50
 Gulf of Alaska Cod0.26
Purse Seine 
 Northwest Atlantic Capelin-19830.81
 Northwest Atlantic Capelin-19810.37
 East Central Atlantic Sardine0.03
 U.S. Gulf of Mexico Menhaden0.03
 West Central Pacific Tuna0.00
Danish Seine 
 Northeast Atlantic Haddock0.50
 Northeast Atlantic Whiting0.45
 Northeast Atlantic Cod0.36
 Bering Sea Sablefish3.51
 Bering Sea King Crab3.39
 Bering Sea Tanner Crab1.78
 Northwest Atlantic Capelin0.80
 East Central Pacific Spiny Lobster0.36
Table 9b. The top numbers-based discard to landed target catch ratios by gear type. Source NRC bycatch database.
Fishery DescriptionBycatch Number per Target Number
 Caribbean Shrimp12.13
 Northeast Atlantic Whiting2.83
 Northeast Atlantic Haddock1.94
 Arabian Sea Finfish Trawl1.75
 Northeast Atlantic Nephrops1.70
High Sea Driftnet 
 Japanese North Pacific Squid0.37
 Japanese Tasman Sea Tuna0.12
 Japanese Sub-Trop Convergence Zone Tuna0.08
 Taiwan North Pacific Squid0.07
 Korean North Pacific Squid0.06
 Eastern Central Pacific Swordfish1.58
 Eastern Indian Ocean1.13
 Danish Seine 
 Northeast Atlantic Cod0.79
 Northeast Atlantic Haddock0.70
 Northeast Atlantic Whiting0.64
 Bering Sea King Crab9.71
 Bering Sea Tanner Crab2.35
 Eastern Central Pacific Spiny Lobster1.68
 Eastern Central Pacific Slipper Lobster0.67
 American Lobster0.22
Table 10. The highest discard to target catch ratios by region (discard weight per landed target catch weight). Source: NRC bycatch database
Fishery DescriptionKg Bycatch per Kg Landed
Northwest Atlantic 
 New England Yellowtail Flounder Trawl1.58
 Newfoundland Shrimp Trawl1.38
 Atlantic Canada Capelin Seine1.33
 Atlantic Canada Groundfish Trawl0.80
 Atlantic Canada Capelin Trap0.80
Northeast Atlantic 
 Baltic Sea Dab Trawl2.01
 Irish Sea Nephrops Trawl1.95
 Baltic Sea Flounder Trawl1.60
 North Sea Shrimp Trawl1.44
 North Sea Whiting Trawl1.34
West Central Atlantic 
 Trinidad Shrimp Trawl14.71
 US Gulf of Mexico Shrimp Trawl10.30
 US Southeast Shrimp Trawl8.00
East Central Atlantic 
 Shrimp Trawl off Senegal2.72
 Finfish Trawl off Senegal1.48
Mediterranean and Black Seas 
 Finfish Trawl0.85
 Tuna Longline0.10
 Hake Trawl0.04
 Sardine Purse Seine0.03
 Flatfish Trawl0.03
Southwest Atlantic 
 Brazilian Shrimp Trawl4.10
West Indian Ocean 
 Sri Lankan Shrimp Trawl10.96
 West Coast Indian Shrimp Trawl8.52
 Persian Gulf Shrimp Trawl4.17
 Arabian Sea Finfish Trawl0.64
East Indian Ocean 
 East Coast Indian Shrimp Trawl3.79
Northeast Pacific 
 Bering Sea Sablefish Pot3.51
 Bering Sea King Crab Trawl3.39
 Bering Sea Rock Sole Trawl2.61
 British Columbia Cod Trawl2.21
 Gulf of Alaska Flatfish Trawl2.08
West Central Pacific 
 Indonesian Shrimp Trawl12.01
 Australian Northern Prawn Trawl11.10
 Malaysian Shrimp Trawl3.03
 Malaysian Finfish Trawl0.44
 Thailand Finfish Trawl0.15
East Central Pacific 
 Mexican Shrimp Trawl9.70
 Hawaiian Swordfish Longline1.12
 Hawaiian Spiny Lobster Pot0.36
 Hawaiian Slipper Lobster Pot0.17
Southwest Pacific 
 Hoki Trawl0.30

Bycatch of Marine Mammals

Our quantitative assessment of weight- and numbers-based bycatch and discard levels in world fisheries considers only levels and ratios for finfish and some invertebrate populations. Nevertheless, many of the oceansí living resource management conflicts surfacing in recent years have been concerned with marine mammals, turtles, or birds. In the past decade it has been well documented that the capture of marine mammals and other marine life in fishing gear can be a serious problem in terms of impacts imposed on marine mammal populations and public ethical concerns. (Northridge 1991a, 1991b, 1991c; Nelson and Lien 1992; IWC Workshop Report 1991; F. Alverson, in press; DeGange and Day 1991; Perrin 1988; Jones 1988; Dawson 1991a, 1991b; Snow et al. 1987; Brownell et al. 1989; Oshumi 1972; Manning 1989; Polacheck 1989; Barlow et al. 1990; Mohan 1985; Reynolds 1985; Perkins et al. 1990–1991, 1992; Read and Gaskin 1988; Lien et al. 1987; Mate 1980; and many others).

Although it is known by marine scientists, conservation groups, and the general public that a wide variety of interactions occur between marine mammals and fisheries (Northridge, 1991b), the quantities of discards involved are generally less well known and more speculative than those for many commercially harvested fish or shellfish.

Northridge (1991b) notes “most marine mammals, with the exception of the rarer ocean beaked whales, for example, have been recorded at some time or other caught in some type of fishing gear”. In a detailed examination of mortalities of cetaceans in passive fishing gears during the IWC Workshop held at La Jolla, California, in 1990, scientists were able to identify 54 species/populations/regions (SPRs) for which abundance estimates and data on incidental mortality in passive gears were available. Rough estimates were made for many other species/populations/areas (see IWC Workshop 1990, Table 1). This listing of species involves only cetaceans interacting with passive gear and does not include species of bycatch taken by active fishing gears or species from other taxonomic groups, such as seals, sea lions, manatees, and dugongs.

The IWC reports make it apparent that considerable qualitative data are available concerning observed bycatch of marine mammals in a variety of fisheries. However, few references provide specific bycatch ratios, and for most fisheries only a small portion of marine mammal catch is recorded (Northridge 1991b). Thus, it is much more difficult to assess global harvest and impacts of the marine mammal discards than for fisheries. Nevertheless, data provided by Northridge (1991a, 1991b) and the IWC Workshop (1991) provide an indication of the potential magnitude of such takes.

The IWC Workshop data can account for potential marine mammal discard mortalities ranging from 65,000 to 86,000 animals annually. These figures are based on summing the range of estimates of mortalities regardless of years, and it is questionable whether higher estimates of mortalities occurred simultaneously in any one year. On the other hand, it must be noted that the IWC scientists, because of inadequate data, could not provide any information on seven SPRs, could do no better than report “some” as the mortality level for 86, and for many other SPRs noted reported mortalities as being less than or greater than some specific number. Thus, the estimate for mortalities associated with passive gears needs to be considered as conservative (Northridge 1991b). The actual catch must be considerably greater.

Cetacean species and areas for which relatively large numbers of bycatch have been observed include Dall's porpoise (Phocoenoides dalli) in the Western and Central North Pacific, Risso's (Grampus griseus) and spinner dolphins (Stenlla longirostris) off Sri Lanka, harbor porpoise (Phocoena phocoena) off Eastern Canada, common dolphin (Delphinus delphis) in the Southwest Indian Basin and Tasman Sea, Pacific white-sided dolphin (Lagenorhynchus obliquidens) off coastal Japan, dusty dolphin (Lagenorhynchus obscurus) off Peru, northern right dolphin (Lissodelphis borealis) in the Central Pacific, pan-tropical spotted dolphin (Stenella attenuata) in the North Indian Ocean, striped dolphin (Stenella coeruleoalba) in the Mediterranean, spinner and bottlenose dolphin (Tursiops truncatus) off Australia, and spotted, spinner, and common dolphin in the ETP (Northridge 1991b).

In addition to the cetacean discard estimates provided by the IWC for passive gears, a considerable incidental catch is reported for various active fishing gears. Northridge (1991b) and the National Research Council (1992) indicate bycatch dolphin mortality in the ETP tuna purse seine fishery may have approached 100,000 animals in 1989, a decline from an estimate of about 140,000 animals in 1986, and perhaps much larger numbers in the 1960s. Discard mortalities in the ETP tuna purse seine fishery have continued to decline since 1989 (National Research Council 1992) (Figure 4). Nevertheless, from the ecological viewpoint, the gains made with the decline in dolphin mortality have been offset by a substantial increase in the discards of young tuna and other pelagic fish species (Table 11). Documented dolphin mortalities in the ETP tuna purse seine fishery involve a single fishery, and F. Alverson (in press) notes some dolphins are taken in association with almost all world seine fisheries. Further, it is well known that seals, sea lions, and other marine mammals are incidentally taken in many of the world's fisheries. Examples of discard mortality as a function of fishery production (i.e., target catch) for purse seining and gillnets by areas are provided in Table 12.

Figure 4

Figure 4. Discard mortalities (numbers of porpoise) in the Eastern Tropical Pacific tuna purse seine fisheries, 1959–1993. Source: Hall/IATTC per. comm. 1993.

Table 11. Discards of young tuna and other pelagic fish species in the ETP tuna fishery (number of fish). Source: Hall/IATTC per. comm. 1993.
Species School SetsLog Sets*Dolphin Sets
(n = 1,060)(n = 324)(n = 764)
Small tunas5,230,000128,700,000833,000
Mahi mahi1,811402,577628
Rainbow runners25542,068 
Other small fish1,70810,123523
Sea turtles471617143
Other large fish1,1518,611 
Trigger fish28334,938 
Sample size1,060324764

* Estimates in this column are likely to be overestimates because more fishing on existing objects oraddition of objects will not result in linear increases of the catches.

Table 12. Dolphin mortality and fishery production. Source: Hall/IATTC pers. comm. 1993.
Gear TypeDolphin Mortality/Fishery Production
Purse Seining 
Eastern Pacific1 dolphin/40–50 tons of tuna
Philippines1 dolphin/2.2 tons of tuna
Ivory Coast1 dolphin/2.2 tons of tuna
Sri Lanka1 dolphin/4.4 tons of “fish”
Bay of Biscay1 dolphin/2.7 tons of albacore
Gulf of Maine1 dolphin/10 tons of demersal fish

Information about the quantities of seals, sea lions, and other marine mammals taken incidentally in fishing gear is even less quantifiable than for cetaceans. Northridge (1991b), in his review of interactions between marine mammals and fisheries, lists numerous species for which documented bycatches have been observed. Significant interactions are noted between fisheries and the northern sea lion (Eumetopias jubatus) northern fur seal (Callorhinus ursinus), harp seal (Phoca groenlandica), gray seal (Halichoerus grypus), Caribbean manatee (Trichechus manatus), Mediterranean monk seal (Monachus monachus), franciscana (Pontopocia blainvillei), dugong (Dugong dugon), harbor seal (Phoca vitulina), and New Zealand fur seal (Arctocephalus forsteri). Perez and Loughlin (1991) also provide an excellent review of the numbers of incidental catches of marine mammals taken in North Pacific trawl fisheries, an area where thousands of animals were taken yearly during the 1970s and 1980s.

Because of the lack of discard rate data for marine mammals, we, like others, are hesitant to estimate a total number and weight of discards for marine mammals. It seems conservative to conclude that several hundreds of thousands of animals are involved annually. However, technological developments in the ETP region, the disappearance of several major high seas driftnet fisheries, national regulations controlling marine mammal bycatch, and fishermen's education programs supervised by scientists may have helped to reduce the total levels of marine mammal mortalities in recent years.

It would appear marine mammal bycatch occurs in almost all major gear types, including driftnets/gillnets, bottom trawls, pots, longlines, tangle/trammel nets, fish traps, purse seine and other round nets, pelagic trawls, setnets, etc. The largest numbers of marine mammal bycatch appear to occur incidentally or as the result of sets on porpoises in various driftnets/gillnets and tuna seine fisheries (Northridge, 1991b; National Research Council 1992).

Sea turtle takes have also been reported in ocean longline operations. Annual incidental takes have been reported at 40,000 worldwide and 21,200 in the Western Pacific and South China Sea alone. Mortality rates have been estimated at 42% for turtles incidentally taken by longline gear (Nishemura and Nakahigashi 1990). Some scientists feel this estimate is too low (Paul Dayton, personal comm.).

Because of the lack of broad-scale quantitative data on birds, we have not attempted to establish regional or global losses of these species groups. However, comments on impacts on birds are provided in the following chapter.

Discard Mortality Rates

Biological and ecological impacts, as well as other impacts resulting from discards, will vary between species depending on (1) the quantities taken, (2) the survival rates of discarded species, and (3) the population characteristics and life histories of impacted species. During the review of the bycatch and discard literature, it became evident that over the past decade considerable energy has been expended to compile and improve discard information. On the other hand, relatively little effort seems to have been made to research and chronicle the mortality rates associated with discards. Of course, the ultimate consequence of discarding depends on the mortilities imposed on the relevant populations. Nevertheless, recorded mortality rates documented in studies reviewed are presented. The most complete report identified by the authors on discard mortalities for the Northeast Pacific is entitled “The Nature and Scope of Fishery Dependent Mortalities in the Commercial Fisheries of the Northeast Pacific” (NRC 1990). This report contains information on available mortality data for halibut, crab, and salmon. Information from this paper is summarized below.


Halibut discard mortality estimates are highly variable (Table 13). Values range from a low of 2% reported in the North Pacific foreign longline fisheries to a high of 100% in several trawl and pot fisheries. Thirty-one percent to 100% of all halibut incidentally taken in crab pots are expected to die. Halibut taken by small and large foreign trawlers in the Bering Sea/Aleutian Islands in the 1970s and 1980s were reported to have experienced slightly higher mortality rates (25% to 54%) than those incidentally taken by similar operations conducted in the Gulf of Alaska (9% to 31%). Other reports of halibut mortality in foreign trawl fisheries in the North Pacific reach as high as 100%. Ninety percent of the halibut taken by joint venture trawlers in the Bering Sea/Aleutian Islands were judged as dead, while only 60% were so described in observed Gulf of Alaska joint venture trawl operations.8 Bycatch mortality of U.S. and Canadian trawl-caught halibut ranges from 10% to 72%. In a study using tag returns (Williams et al. 1989; Hoag 1971, 1975) to estimate mortality, smallar, sublegal halibut (<80 cm) suffered a substantially higher mortality rate (72%) in the domestic trawl fishery off British Columbia than did larger (>80 cm) halibut (45%). As much as 50% to 100% of all halibut caught in shrimp trawls die.

Discard mortalities within foreign longline fisheries have been estimated to range from 2% to 50%. Halibut taken by domestic longliners experienced a minimum bycatch mortality of 32% and a maximum of 50%. Forty percent of the halibut taken in the Pacific cod longline fishery were judged as dead or in poor condition. Sublegal halibut caught incidentally within the U.S. longline fishery for halibut experienced a mortality range of 25% to 40%. Over the past several years, scientists working with the North Pacific Fishery Management Council have evaluated observer data and attempted to provide their best estimates of halibut mortality rates imposed by the various North Pacific demersal fisheries. The mortality rates used in 1992 and recommended for 1994 for the Gulf of Alaska and the Bering Sea are somewhat lower than for earlier observations (Table 14).

8Catches by U.S. vessels were subsequently offloaded onto foreign vessels for processing.

Table 13. Halibut discard mortality rates estimated by fishery. Source: NRC 1990.
Bycatch SpeciesAreaFisheryMortality Percentage
HalibutBS/AITrawl, JV90
HalibutBS/AITrawl, large25–54
HalibutBS/AITrawl, mothership/freezer90
HalibutBS/AITrawl, small25–54
HalibutGOACrab Pot, Tanner45
HalibutGOACrab Pot, Tanner31
HalibutGOAGroundfish trawl36
HalibutGOATrawl, JV60
HalibutGOATrawl, large9–31
HalibutGOATrawl, small9–31
HalibutNorth PacificCrab pot, domestic45–100
HalibutNorth PacificJapanese groundfish trawl90–100
HalibutNorth PacificLongline25
HalibutNorth PacificLongline, domestic32–50
HalibutNorth PacificLongline, domestic32–50
HalibutNorth PacificLongline, foreign10–50
HalibutNorth PacificLongline, foreign2–14
HalibutNorth PacificLongline, foreign10–50
HalibutNorth PacificShrimp100
HalibutNorth PacificShrimp trawl, domestic50
HalibutNorth PacificTrawl, domestic10–42
HalibutNorth PacificTrawl, foreign50–100
Halibut<80 cmBritish ColumbiaTrawl, domestic72
Halibut<80 cmBritish ColumbiaTrawl, domestic45
Sublegal halibutGOALongline, domestic halibut33–40
Sublegal halibutBS/AI and GOALongline, domestic halibut25
Table 14. Trend in halibut discard mortality rates during 1990 through 1993 and recommendations for 1994 preseason rates. Change are noted with an asterisk (*). Source: NPFMC SAFE Document, 1994.
Region/Target1990199119921990–92 AveragePrelim 1993Used In 19931994 Recommendation
BS/AI TRAWL       
MWT Pollock81818783868080
Atka mackerel69736268697070
Rock sole/Oflats**58687868747070
Yellowfin sole**73747775817075*
Pacific cod68606765656065*
BT pollock65597667876065*
Arrowtooth5741 49 4050*
Grnld. turbot5838 48 4045*
Other sp.3629 33724040
GOA TRAWL       
MWT Pollock63746969617570*
BT Pollock65567064435565*
Shallwtr. flatfish62616262636060
Other sp.63596462556060
Pacific cod61555958585555
Deepwtr. flatfish57525956585555
BS/AI H&L17211819171815*
Pacific cod131819171812.5/1515*
Sablefish1829 24 12.5/1515*
Rockfish  1717812.5/1515*
Grnld. turbot       
GOA H&L       
Pacific cod1317302081615*
Rockfish1520 18 11.5/1415*
BS/AI POT       
Pacific cod73127455
GOA POT       
Pacific cod105161022510*

** During 1990 and 1991, “Other flatfish” was grouped with yellowfin sole.Since 1992, the target has been grouped with rock sole.

Nielson et al. (1989), in a field experiment on the viability of Atlantic halibut taken by trawls and longline, report 35% of the trawl-caught halibut and 75% of the longline-caught halibut below a proposed size limit survive for 48 hours after capture.


Like those for halibut, reports of king crab discard mortalities are extremely variable (Table 15). Mortality estimates for trawl-caught king crab (Paralithodes spp.) range from 2% to 81%. Tanner crab (Chionoecetes spp.) mortality estimates are no less unpredictable, and values of as low as 12% and as high as 82% have been reported. In a study investigating king and Tanner crab taken by trawlers off Kodiak (Blackburn and Schmidt 1988), only 2% of the king and 36% of the Tanner crab were judged injured or dead. Dead king crab were much more common during the February through June period (19%) than in the remainder of the year (0.9%). In a separate study (Stevens 1989), king and Tanner crab were held in live-tanks for up to 17 hours after capture by Bering Sea/Aleutian Islands joint venture sole trawlers. Mortalities of 77% to 81% for king crab and 74% to 82% for Tanner crab were observed. Suffocation, not noticeable external injuries, was found to contribute most substantially to the recorded mortalities. Crab mortality studies which did not distinguish between the types of crab taken in the trawl reported mortality rates of 50% to 100%. Finally, 0% to 50% of all crab taken with longline fisheries are expected to die.

Dungeness crab (Cancer magister) mortality estimates are surprisingly low. In trawls, hardshell and softshell Dungeness mortality has been estimated at 2% and 9%, respectively. Female Dungeness crab suffer a slightly higher trawl bycatch mortality (6% to 11%) than do males (0% to 4%). Softshell mortality within the Dungeness crab pot fishery has been reported at 22% to 25%. Mortality of sublegal hardshell Dungeness crab within the Dungeness fishery is much lower (2% to 4%).

Table 15. Reported crab bycatch mortality rates in North Pacific fisheries. Source: NRC 1990.
Bycatch SpeciesAreaFisheryMortality Percentage
CrabNorth PacificLongline, domestic halibut50
CrabNorth PacificTrawl50–100
Dungeness crab (hardshell)   
Dungeness crab (hardshell)   
 West CoastCrab pot, dungeness2–4
Dungeness crab (softshell)   
Dungeness crab (softshell)   
 West CoastCrab pot, dungeness22–25
Dungeness crab, females   
Dungeness crab, malesGOATrawl0–4
King crabBS/AITrawl, JV sole77–81
King crabGOATrawl2
King crabNorth PacificTrawl19
King crabNorth PacificTrawl79
Softshell King crabGOATrawl21
Tanner crabBS/AITrawl, JV sole74–82
Tanner crabGOATrawl36
Tanner crabNorth PacificTrawl12
Tanner crabNorth PacificTrawl78


Salmon mortality rates range from a low of 8% for legal-sized chinook (Oncorhynchus tshawytscha) taken in troll fisheries for other salmon in Southeast Alaska to a high of 100% for salmon taken in longline operations in the North Pacific (Table 16). However, the latter fishery takes only a small amount of the salmon taken by the regional bottom fisheries. Most are taken in trawls and are presumed to be dead. Legal-sized chinook losses in troll fisheries for other salmon are only slightly lower (8% to 25%) than mortalities associated with sublegal-sized chinook (19% to 28%). Illegal troll-caught coho (Oncorhynchus kisutch) are reported to suffer a much higher rate of mortality (34% to 52%) than illegal chinook. As would be expected, barbed hooks raise this mortality substantially. Seventy-seven percent of the juvenile coho caught on barbed hooks died, compared to only a 33% rate of mortality when caught on barbless hooks. Recent work has demonstrated that mortality rates range from 6% to 11% of coho and chinook, respectively, hooked and released by recreational anglers (NRC 1993).

Table 16. Reported salmon mortality rates in North Pacific fisheries. Source: NRC 1990.
Bycatch SpeciesAreaFisheryMortality Percentage
ChinookNorth PacificPurse seine50–90
ChinookNorth PacificTroll20–30
Chinook, largeSE AlaskaPurse seine24
Chinook, mediumSE AlaskaPurse seine68
Chinook, smallSE AlaskaPurse seine60
Chinook, illegalSE AlaskaPurse seine50–90
Chinook, legalGOATroll25
Chinook, legalGOATroll, coho20
Chinook, legalSE AlaskaTroll8–13
Chinook, sublegalGOATroll28
Chinook, sublegalGOATroll, coho24
Chinook, sublegalSE AlaskaTroll19–28
Coho, illegalWest CoastTroll34–52
Coho, illegalWest CoastTroll42
Coho, juvenileWest Coast(barbless hooks)77
Coho, juvenileWest Coast(barbed hooks)33
SalmonNorth PacificLongline100

Reported purse seine chinook mortalities have generally been significantly higher than troll mortalities. Up to 50% to 90% of all legal- and illegal-sized chinook caught by purse seiners die. In a study in which purse seine mortality was separated by size (Rowse and Marshall 1988), 68% of all medium-sized (21" to 28") and 60% of all small-sized (<21") chinook caught by purse seiners died. Only 24% of the large (>28") chinook taken as bycatch died.

Chinook dropout losses in Puget Sound gillnet fisheries have been estimated at 2% to 3%. Predation losses accounted for an additional 2% to 9% of observed mortalities, but reached as high as 28% in areas of high seal populations.

A report of the NOAA/NMFS (1990/1991) on recreational fisheries note about 50.5% of the total marine recreational catch was released alive, but it is not apparent what percentage of the total catch was discarded dead. Estimates of “released alive” come from the fishermen and the ultimate amount of discards is unknown. Recent studies, however, by NRC (1993) indicate a relatively high rate of survival for chinook and silver salmon (Oncorhynchus kisutch) taken in recreational fisheries, approximating 90%.

Other Observations Regarding Discard Mortality

Many researchers believe survival rates for fishes, particularly those species left floating on the sea's surface, are low (Wassenberg and Hill 1990). For cods, whiting, and rockfishes (Sebastes) whose air bladders expand and trap them at the surface, survival rates are expected to be very low. Cushing (1984) assumed that of the estimated 40 million whiting discarded each year, all die. Similarly, for many tropical species of the snapper and grouper families, survival rates must also be low (Seidel 1975; Caddy 1982; Howell and Langan 1987). Hill and Wassenberg (1990) examined the fate of discards from prawn trawlers in Torres Strait, Australia, and noted about half of the discarded fish and 25 % of the cephalopods floated, but none of the non-commercial crustaceans floated. The percentage of each group examined that survived ranged from 2% for finfish and cephalopods up to 51% for crustaceans. Results of their work for fish subjected to 10–30 minute exposures on deck and subsequently discarded (into holding tanks) are shown in Tables 17a and 17b. Factors contributing to mortality of discards include (a) the character of the gear deployed, (b) sorting methods and duration of sorting time, (c) fishing time, (d) size of catch for trawls and seines, (e) associated species in the catch, and (f) environmental conditions (Wassenberg and Hill 1989; Nielson et al. 1989; Van Beek et al. 1990).

Table 17a. Percentage (by weight) of three taxonomic groups in a subset of discards (fish, non-commercial crustaceans, and cephalopods) from prawn trawls, the percentage of each group that floated, and the percentage of each group that was alive 12 hours after a 30-minute exposure to air on deck. Source: Hill and Wassenberg 1990.
 Composition (%)Floating (%)Surviving (%)
Fish78 (62–90)45 (29–59)2 (1–3)
Crustacea18 (9–33)0 (0)51 (23–80)
Cephalopods3 (1–6)26 (0–59)2 (0–12)
Total 37 (23–49)11 (7–17)
Table 17b. Percentage of animals surviving for 12 hours after exposure for 10 minutes on deck. Note: n is at least 30 in all cases; animals collected from trawls of 30-minute duration. Source: Hill and Wassenberg 1990.
TaxonPercentage survival
Stomatopods (Squilla spp.)50
Crabs (mainly portunids)100
Cuttlefish (Sepia spp.)0*
Squid (Loligo spp.)0*
Starfish (Astropecten spp.)88
Carangidae: Carangoides caerulopinnatus13
Caranx bucculentus29
Selaroides leptolepis66
Leiognathidae: Leiognathus splendens0*
Lethrinidae: Lethrinus fraenatus43
Lutjanidae: Lutjanus japonicus33
Mullidae: Upeneus sulphurus0*
Upeneus sundaicus0*
Nemipteridae: Nemipterus furcosus0
Priacanthidae: Priacanthus spp.3
Siganidae: Siganus canaliculatus56
Teraponidae: Pelates quadrilineatus50
Terapon theraps97
Tetraodontidae: Lagocephalus scleratus0*

*All dead within 1 hour.

Berghahn et al. (1992) provide discard survival data for a number of bycatch species taken by shrimp vessels in the North Sea. Survival of flatfish is noted to depend strongly on the species and the size of specimens, as well as the catch processing conditions. A series of experiments on dab survival resulted in discard mortalities ranging from 0% to 67%, with an average mortality of 32.6% for fish collected after “sieving”9 and 11.9% for dabs collected from the catch before “sieving” (Table 18). Additional data on the mortality of fish taken in shrimp fisheries are also provided in the Berghahn work (1990) in the Wadden Sea (Table 19).

9Sieving refers to an onboard sorting system.

Table 18. Dab mortality from shrimp fishery bycatch. A = after sieving out; B = results of controls with samples collected from the catch before sieving; TL = total length; Catch = total catch in one codend; Nb = number in the beginning; Ne = number at the end of the experiment. Source: Berghahn et al. 1992.
Date TL Range (cm)Catch (kg)Water Temperature (celcius)Duration of Experiments (d)NbNeMortality (%)
 B12–22   262312
5/21/89B10.5–23.5   20200
5/28/89B12–26   454011
7/49/89B11.5–25.5   11109
5/16/90B11–16   220
Average   A 29119632.6
    B 15914011.9
Table 19. Mortality of selected fish species in the bycatch of the shrimp fishery after five days of maintenance. The trials without sieving are given in brackets. Source: Berghahn 1990.
SpeciesNumber of TrialsTotal Catch One Codend kgWater Temp. CNumber of SpecimensTotal Length cmMortality %
 (3)  (78) (0.0–11)
 (3)  (17) (0.00)
 (5)  (37) (0.0–5)
 (4)  (28) (0.0–13)

Chapter Summary

The term bycatch has different meanings among scientists. It may refer to (1) that portion of a catch taken by fishermen during the capture of a target species, (2) an assemblage of target species either retained or discarded, (3) discarded species only, or (4) that portion of non-target catch which is retained. In this study, we have chosen to use the operational definitions proposed by the Newport National Industry Bycatch Workshop. However, we recognize the need for an international effort to standardize bycatch and discard terminology and reporting modes.

Data for the world bycatch study were retrieved by the authors from literature searches, publications provided by fishery organizations throughout the world, and many independent scientists who have been concerned with the world bycatch issue. Books, publications, and gray literature acquired by the investigators have been placed in a special collection in the NRC library and can be accessed through a FileMaker® computerized information retrieval system. Additionally, a database has been created to access qualitative and quantitative information regarding discard rates in various areas of the world.

A total of 28.7 million mt of bycatch and 27.0 million mt of discard is estimated for world marine fisheries, although this total is considered to be conservative in that data are not included for many invertebrate fisheries, recreational fisheries, and subsistence fisheries. A great share (37.2%) of world discard occurs in various shrimp fisheries. Much of the discard in the shrimp fisheries is comprised of small tropical fishes, frequently weighing less than 100 grams. In temperate and subarctic waters, a great portion of the discard is comprised of juvenile and adult species of commercial value. Significant amounts of discard occur in many cod, hake, flounder, and crab fisheries.

Shrimp trawling almost universally results in high discard rates. High discard rates also occur in many of the world's trawl fisheries, although mid-water trawl rates fall among the lowest observed discard rates. Those for high seas gillnets were frequently among the lowest observed rates. Marine mammal discards probably constitute several hundred thousand a year. However, some reduction in marine mammal discards must have resulted from technical and operational changes in the ETP tuna seine fisheries and the discontinuation of high seas driftnet fisheries in the Pacific.

Survival of discards depends on a number of biological, environmental, and fishing operational patterns, but is believed to be low for most species.

Previous Page Top of Page Next Page