The Dropout Problem

During the driftnet retrieval process, catch and by-catch only loosely ensnared may become dislodged and drop back into the ocean before the net is brought on board. Many of these animals are already dead. It is probable that the catch starts dropping out of the net as soon as the lead line is pulled tight immediately prior to the onset of the long retrieval process. Still more dropouts occur during the several hours that it takes to retrieve the driftnet. If there are still unwanted species entangled when the net clears the water, driftnet fishermen try to shake them free before hauling the nets on board (Gooder 1989; Tsunoda 1989). Albacore fishermen observing large-mesh driftnet fisheries in the South Pacific estimate the total dropout rate to be 4060% of the catch (Vanderpool pers. comm. 1991).

Dropout figures recorded by observers in the North Pacific joint observer program were the result of observing a subset of net sections selected randomly each day. Only those species that fell out of the net after it cleared the water were recorded. In 1990, observers on board squid vessels in the North Pacific reported rough dropout estimates of 20% for albacore, 15% for swordfish and 5% for skipjack tuna after the net cleared the water (Sidney Report 1991). Research vessels estimate large-mesh dropout rates at 530% for albacore. The results of experiments in wave tanks indicate total dropout rates between 50-70% under the rough sea states frequently observed in the driftnet fishing area (Sidney Report 1991; Bartoo and Watanabe 1989). Scientists, evaluating the dropout data available through June 1991, estimate that overall dropout rates for some species "may approach 50%" (U.S. Summary Report 1991).

Since a large portion of the catch and by-catch may go unrecorded due to dropouts occurring under water, the total catch and by-catch may be much greater than reported. The impact on the target species, the by-catch, and the ecosystem is, therefore, likely to be much greater than estimated and the actual waste involved in fishing with driftnets much larger than people have been led to believe. Albacore trollers reported that large-mesh driftnets typically did not have many albacore in the lower half of the nets. They theorize that the acts of pulling the lead lines tight and dragging the nets through the water for the many hours that it takes to retrieve them results in the loss of many of the fish caught in the nets. Experiments on the effects of lowering the gear in the water column tend to support this supposition. Although a driftnet lowered two meters in the water resulted in a reduced catch of skipjack tuna, there was no apparent effect on the catch of albacore.


In the squid fishery, as much as 30% of the driftnets on each fishing voyage wear out or are damaged and are thrown overboard (World Fishing 1991). The U. S. National Marine Fisheries Service estimates that .06% of driftnets are lost each time they are set, resulting in 12 miles of net lost each night of the season and 639 miles of net lost in the North Pacific Ocean alone each year (Davis L. 1991). It has been asserted that lost and discarded sections of driftnet ball up fairly quickly and cease to ghost fish in a short period of time (Mio, Domon, Yoshida, and Matsumura 1990). Although monofilament driftnet fragments may eventually form a loose ball, they frequently trail long streamers of torn net and continue to ensnare animals attracted to the floating mass. A two kilometer section of driftnet can form a mass more than seven meters in diameter (Hayworth pers. comm. 1991). Much evidence exists to indicate these tangled masses of net continue to ghost fish for long periods, both on the surface (Gooder 1989; Ignell, Bailey, and Joyce 1986; von Brandt 1984; Degange and Newby 1980) and on the bottom (Carr and Cooper 1987; Way 1977).

Multifilament driftnets do not ball up, but stay stretched out as long as their floats and weights stay attached. Data from the 1990 North Pacific joint observer program revealed that large-mesh nets catch 20 times the number of dolphins and 50 times the number of turtles that small mesh squid nets catch. Driftnet fragments often end up in the EEZs of coastal States, threatening other fisheries and rare and endangered species (Stewart and Yochem 1990; Henderson 1984). Although the "act" of losing this deadly gear may have occurred on the high seas, the effect constitutes a "taking" of the resources of coastal States that is in direct contravention of both international and many domestic laws (Paul 1990). Lost and discarded driftnets may continue to ghost fish for years, silently and wastefully depleting the oceans of their living resources.

In 1978, a 3500 meter section of lost driftnet was found floating in the North Pacific (49 deg 15 min N, 168 deg 14 min E). Entangled within the 1500 meters brought on board were 75 newly snared salmon, at least twice that many rotten ones, plus assorted other fish and some 99 seabirds (Degange and Newby 1980). Degange and Newby (1980) identified birds from six species listed in the Migratory Bird Convention Annex (4 Laysan albatrosses, 15 northern fulmars, 15 tufted puffins, 14 sooty shearwaters, 40 slender-billed (short tailed) shearwaters and several fork-tailed storm petrels). Endangered Hawaiian monk seals have also been found entangled in masses of monofilament driftnet (Henderson 1984) and young California sea lions were frequently found entangled in monofilament driftnet fragments (Stewart and Yochem 1990). In 1985, United States observers monitoring Japanese high seas squid driftnet vessels recovered four sections of renegade net measuring 115 mm in width and 30-86 meters in length. All the sections had both live and dead animals ensnared in them, including yellowtail, pomfret, two pelagic hammerhead sharks, three blue sharks, an ocean sunfish, and a fur seal (Ignell, Bailey, and Joyce 1986).

Although there have been few observations of torn netting being discarded at sea, this may have been influenced by the fact that observers were watching. Annex V of MARPOL prohibits discarding old driftnets at sea; however, not all driftnet States are party to this convention and enforcement is difficult at sea. In 1990, Japan tested a net disposal furnace capable of incinerating a five kilometer driftnet in under 48 hours. The furnace was to be installed on board key vessels throughout the driftnet fleet (World Fishing 1991). In October 1991, one to two kilometers of very new monofilament 134 mm mesh driftnet washed up on Oahu, Hawaii's northeastern shore. The knots of the net were lined up and there was no algal growth on the nylon filaments, indicating that the net probably had never been used, but for some reason had been discarded very recently and well within the U.S. EEZ. Fishing vessels are not permitted to bring driftnets into state waters under Hawaii law (Hawaii Revised Statutes 188/30.5). On November 17, 1992, the U.S. Coast Guard Cutter Jarvis encountered an 800 pound mass of gill nets, trawl nets and longline five miles west of Kauai, Hawaii. Because drifting nets are a hazard to navigation, the crew hauled it on board and found a dead dolphin entangled in the mass.

The By-Catch

Driftnets fish indiscriminately. While mesh size may be targeted for a particular species of squid or fish, the nets are capable of catching just about anything that swims or dives and is equal to or larger than the mesh size being used. Estimates of incidental catch rates of non targeted species vary greatly depending on the year, the vessel, its location, the time of year, who is doing the reporting, the proximity of the observer, and sections of net selected for monitoring. The total amount of net that has been actually monitored and recorded is extremely small compared to the total amount of driftnet that has been set. In the North Pacific over 100 species have been reported taken as by-catch by squid driftnets (U.S. Summary Report 1991). In the South Pacific large-mesh driftnets have caught at least 46 species (SPF 1991). Most of the incidental catch is edible fish and almost all of it is discarded. Out of the 46 species caught by large-mesh driftnets in the South Pacific, up to 35 species may have been discarded as unwanted by-catch (SPF 1991). Discards also included target fish that were putrefied, badly damaged, or partly eaten by sharks.

Obtaining statistically significant North Pacific driftnet data on populations of some long-lived, slow-reproducing by-catch species may no longer be possible. Between 1979 and 1981 the amount of fishing effort by the driftnet fleet rose dramatically and exponentially; since then it has increased much more gradually. Thus the maximum incidental catch of species such as marine mammals, seabirds, and turtles would have occurred in the early eighties at a time when the amount of by-catch data being collected was not statistically significant for those species. Incidental catches today probably reflect the impact of driftnet fishing on much reduced populations. By-catch data from the 1989 and 1990 observer reports can be found in Tables 7-14.


Diving seabirds are attracted to squid and fish hanging in the nets several meters below the surface. Seabird entanglement rates are affected by species behavior, feeding habits and soak times. "The number of seabirds attracted to a driftnet appears to be directly (but non linearly) related to the length and soak time of the gear. Entanglement rates appear to depend on bird density in the fishing area and are extremely variable, because of the patchy at-sea distribution of seabirds" (Wetherall 1989). At least 30 of the 70 species found in the North Pacific driftnet fishing grounds have been caught in driftnets. Seabirds such as shearwaters, boobies and the northern fulmar regularly dive down three to seven meters to catch fish; some, such as penguins, dive much deeper. Albatrosses regularly feed on squid and fish caught in nets up to 1.5 meters below the sea surface.

Data already collected indicates that large numbers of various species of seabirds were killed annually by driftnets. Entanglement data on birds were collected mostly from the salmon and squid driftnet fisheries. Data from the large vessel component of the Japanese landbased salmon driftnet fishery indicates that more than eight million seabirds have been killed by driftnet fishing since 1952 (DeGange and Day 1991). In the early to mid 1980s the Japanese mothership fleet killed up to 250,000 seabirds each year (Jones and DeGange 1988). Rough estimates of the total number of seabirds killed by the three squid driftnet fisheries are between 875,000 and 1,660,000 seabirds annually (Anon. 1989; DeGange et al in press). There are no overall estimates as yet of the seabird by-catch by the large-mesh fisheries. There are also no estimates of the commercial dropout rates, although Ainley et al. (1981) estimated a 513% dropout rate based on data taken by research vessels.
Although the seabird entanglement rate may be higher when the fleet fishes closer to nesting colonies, driftnet fleets hundreds of miles from land also catch large numbers of seabirds. Many seabirds migrate the entire length of the Pacific Ocean and may spend 310 years at sea before returning to land to breed (Eilerts pers. comm. 1991). During periods of migration more birds are exposed to fishing gear set in the migratory path and the risk of entanglement probably increases (i.e. during the September-October southward migration period for sooty and short-tailed shearwaters). Of the six species most frequently entangled in North Pacific squid driftnets, three breed in the North Pacific (Laysan albatross, tufted puffin, and horned puffin) and three breed in the South Pacific (flesh-footed shearwater, sooty shearwater, and short-tailed shearwater) (Wetherall 1989). Data from the 1990 observer reports indicates that the Japanese high seas squid fleet took 270,000 seabirds as by-catch. Eighty four percent were sooty shearwaters; 5% were short-tailed shearwaters.

The actual impact of driftnet mortality on the various populations of seabirds is not known, but potentially it is devastating. Seabirds have a low natural mortality. They are also long-lived and slow to reproduce. Some birds such as the albatrosses live 40-60 years. Most seabirds take several years to reach breeding maturity, some as long as 6-10 years (Eilerts pers. comm. 1991). The impact of driftnets on individual species is determined by a variety of factors, but most particularly on the catch as a percentage of a breeding population. The take of only a few birds of a relatively rare species can have a severe impact on the ability of the species to survive. Scientists are particularly concerned about the black-footed albatross which has a low population size and a relatively high driftnet catch rate (U.S. Summary Report 1991). The North Pacific driftnet fisheries together may have taken up to three percent of the black-footed albatross population as a whole, and if driftnets were disproportionately impacting a particular breeding colony that had only a few breeding pairs, they could have decimated entire colonies. Seabirds are part of the overall ecological balance of the ocean. They are top predators, recycle nutrients, and serve as indicators of ocean health. If their populations are decimated, the entire oceanic ecosystem changes.

Marine Mammals

By-catch of marine mammals have been recorded in nearly all driftnet fisheries and almost all of them drowned. The FAO (1990) estimates that the total cetacean by-catch taken by all pelagic driftnets during the 1988-89 season was between 300,000 and 1.06 million. The U.N. Secretary-General's Report (1990) cites rates of interception by the large mesh driftnet fishery at 0.6 cetaceans per 10 km of driftnet per day for the South Pacific and 0.4 per 10 km per day for the North Pacific squid driftnet fishery. In the North Pacific salmon mothership fishery, data collected between 1962 and 1971 by the Japanese Fisheries Agency (JFA) indicates that 0.370.65 cetaceans were caught per 1000 tans, with a maximum of 1.42 per 1000. Data collected by United States researchers indicate an average of 6.66 cetaceans caught per 1000 tans by the salmon drift net fishery data during the 1960s, (approximately the amount of driftnet set by one squid vessel each night). The FAO (1990) cites rates of interception at 0.9 cetaceans per 10 km of driftnet per day in the Tasman Sea and 0.21 cetaceans per 10 km east of New Zealand. An observer recording by-catch by a commercial driftnetter fishing in the EEZ of the Federated States of Micronesia recorded one dolphin caught for every nine tuna landed. The area had never been fished with driftnets before.

In 1989, scientists noted that data collected prior to the Canada-Japan-United States 1989 and 1990 Observer Programs indicated that five important marine mammal species were threatened by the North Pacific driftnet fisheries: northern fur seals, Dall's porpoises, northern right whale dolphins, Pacific white-sided dolphin, and common dolphins. In 1987, Japan reported a total of 676 Dall's porpoise taken as by-catch by the salmon mothership fishery and 458 Dall's porpoise taken by the land-land-based fishery. Incidental catches of other species were not reported (INPFC 1988). U.S. estimates were higher (Jones et at 1990). (Table 5) In 1989, observers monitoring only 27 Japanese squid driftnet vessels and 1,402 nightly operations recorded 208 northern fur seals, 141 Dall's porpoise, 455 northern right whale dolphin, 254 Pacific white-sided dolphin, 12 common dolphin, plus 52 unidentified dolphins. (Table 7) In 1990, observers monitoring 10% of just the Japanese squid driftnet fleet recorded 4,312 marine mammals taken. These included 545 northern fur seals, 1,736 dolphins, including 840 northern right whale dolphin, 459 Pacific white-sided dolphin, 318 Dall's porpoise, 69 common dolphin, and 22 other cetaceans, including beaked whales, pilot whales and sperm whales. (Table 8) A disproportionate number of juveniles were recorded in the by-catch of northern right whale dolphin and the Pacific white-sided dolphin. Overall an estimated 26,000 marine mammals were taken as by-catch by the Japanese squid fishery alone in 1990 (US Summary Report 1991). Analysis of data from the large-mesh fishery indicates its marine mammal by-catch may be 20 times higher than the squid by-catch. The by-catch of marine mammals caught by driftnets in other parts of the world is equally appalling. Scientists speculate that several species of whales, including the near-extinct Northern Pacific right whale and the endangered humpback whale may also have been caught in driftnets, but broke loose while still entangled and later died unobserved (U.S. Summary Report 1991). Researchers reported that despite a moratorium on the harvesting of humpback whales, many juveniles did not return to their wintering grounds in Hawaii and the North Pacific population as a whole was not increasing. In the Mediterranean, a sperm whale was videotaped hopelessly entangled in a large section of driftnet (Di Natale and Notarbartolo-di-Sciara 1990). In the Federated States of Micronesia (FSM), 11 whales were caught by one large mesh driftnet vessel fishing during one month's worth of fishing, including nine caught during one night's fishing (Goldblatt 1989). In the Tasman Sea, a rare Southern bottlenose whale was discovered entangled, but alive, and set free by divers (Coffey and Grace 1990). Another was cut loose with the net still wrapped around it (Watanabe 1990). In the North Pacific, observers recorded pygmy sperm whale, dwarf sperm whale, black whale, Cuvier's beaked whale, short-finned pilot whale and false killer whale in the by-catch (Joint Report - Japan Squid 1990; Joint Report - Taiwan 1990). In April 1993, a small fin whale and a small sei whale were reported caught in a driftnet off the Noto peninsula in the Ishikawa prefecture in Japan.

The marine mammal by-catch sampling size and method may also be inadequate. If the populations of marine mammals being taken in the by-catch are evenly distributed across the ocean, sampling 10% of the entire fishery may be sufficient. The sampling done by the joint observer program in the North Pacific did not meet this requirement. In 1989, 4% of the Japanese squid fleet was sampled; in 1990, 10% was sampled; in 1991, only 7% of the Japanese fleet was sampled. In 1991, a smaller percentage of the ROK and Taiwanese fleets were sampled. In addition, by prior agreement, not all sections of the nets were monitored. If a cetacean was caught in one of the sections not scheduled to be monitored, it was not recorded. In addition, existing data on cetacean by-catch indicates that it is highly sporadic. The animals travel in pods that are unevenly distributed across the ocean. None will be caught for several operations, then several will be caught at once (Joint Report - Taiwan 1990; Goldblatt 1989). (Table 15)

Marine mammals are also long-lived and slow to reproduce. The driftnet catch rose exponentially in the early 1980s and increase only gradually after that, despite increasing fishing effort. It is highly likely that large numbers of marine mammals were taken by the nets in the early 1980s and what was caught later, when by-catch data was recorded, represented only a sampling of the remnants of once much larger populations. As noted previously, an observer monitoring large-mesh driftnets set in Micronesian waters that had never been fished with a driftnet before recorded one dolphin for every nine tuna caught (Goldblatt 1989).

Scientists from Japan, South Korea and Taiwan have acknowledged that their fleets are significantly depleting at least one species, the northern right whale dolphin. At the INPFC meeting held in Sydney, B.C. in June, 1991, the delegates agreed that the North Pacific driftnets could soon exterminate the species (U.S. Summary Report 1991). In 1990, observers on board just 10% the North Pacific Japanese squid driftnet fleet recorded over 840 northern right whale dolphins killed. (Table 8) In addition, the combined effects of driftnets with other sources of man-made mortality may prevent populations of other species such a striped dolphins and northern fur seals from recovering.


The early life-stages of all seven sea turtles involve some pelagic existence and driftnets clearly pose a substantial threat to those species that reside in or migrate through the fishing area. By-catch data from the North Pacific driftnet fisheries indicates that three species are particularly vulnerable: the leatherback, the loggerhead, and the green sea turtle. The leatherback turtle spends most of its life in the pelagic environment and is particularly at risk. It was known for more than a decade prior to the passage of the U.N. driftnet resolutions that leatherback turtles were vulnerable to driftnets. In 1980, an albacore trolling vessel observed at least five dead leatherback turtles entangled in loose sections of driftnet, floating at the surface (Balazs 1982). They were apparently too large and difficult to land and were cut adrift still entangled in the net. The joint report for the 1989 observer program listed as squid by-catch 22 marine turtles, including nine leatherbacks. The Japanese squid joint report for the 1990 observer program listed as squid by-catch 35 marine turtles taken, including 29 leatherbacks. Fifty five percent of the turtles entangled in the squid nets were dead on landing. The Taiwanese joint report for the 1990 observer program listed 76 taken as by-catch by the 11 large mesh vessels. Preliminary analysis of the CPUE from the large-mesh driftnet fishery indicated a turtle catch rate 50 times higher than that of the squid fishery (Balazs and Wetherall 1991). Reports from other driftnet fisheries also noted marine turtles taken as part of the by-catch. Some were kept in violation of the CITES Convention (Goldblatt 1989). Both the leatherback and the green sea turtle are listed as endangered by the International Union for Conservation and Natural Resources (IUCN). The loggerhead is listed as "vulnerable". All three are listed in Appendix I of the CITES convention.


It is conservatively estimated that by-catch of blue sharks taken in 1990 by the Japanese squid fleet in the North Pacific alone totaled 700,000 (U.S. Summary Report). Catch rates were highest in waters between 14 and 18 deg C (Wetherall and Seki 1991). In 1988, the Japanese estimated that 1.61 million blue shark were taken by squid driftnet gear in the North Pacific and 3.27 million by longline gear. In 1988, the Japanese large mesh fleet took 150,000 blue shark. Estimates for the total amount of blue shark taken in the North Pacific by the Taiwanese squid and large mesh fleet and the South Korean squid fleet are not available. Blue sharks were and are also taken by driftnet fleets and longliners in the South Pacific, Atlantic and Indian Oceans. The world wide catch of blue shark probably reaches several million. Except for a few shark fins collected for soup, almost all of this protein was discarded (U.S. Summary Report 1991).

Sharks are long-lived, slow-reproducing top level predators. They are sustainable only at relatively low levels of exploitation. Although the blue shark has a relatively high growth rate and matures earlier than most sharks, it still has a low reproductive rate (U.S. Summary Report 1991; Wetherall and Seki 1991). In the North Pacific the blue shark pupping grounds overlap with the squid driftnet grounds (Wetherall and Seki 1991).

Return to Table of Contents

Return to Earthtrust's DriftNetwork Page