Peltonen (1969) reported on two holding experiments conducted in 1958 and 1960 in which tagged setline-caught halibut ranging in length from 80 to 119 cm were held in captivity in sea pens for the purpose of evaluating the magnitude of possible mortality resulting from capture and tagging. Fish were caught on J-style hooks and held in shallow sea-pens near Prince Rupert, Canada. The design called for holding fish for 14 days, although some fish were held up to 2 ½ months, during which hooking injuries were noted to heal and little feeding was observed. Peltonen concluded that there was little if any mortality associated with the capture or tagging operation, and gave a maximum estimate of this mortality to be .038. Survival was positively associated with fish length. This is the source of the 96.5% survival for excellent condition fish.

Myhre (1974) reported on a 1966 experiment where 2,042 longline-caught halibut were tagged and released. Fish were subjectively partitioned into four condition codes; dead (471 fish), poor (233 fish), and good or excellent (1,388 fish). Using relative tag recovery rates, Myhre estimated that the return from the poor condition fish was half that from the fish in good or excellent condition, independent of fish length.

Hoag (1975) used tag recovery rates from 2,057 trawl-caught fish to estimate survival of fish released from domestic trawl fisheries. Fish were classified into one of five condition codes, based on degree of physical activity and injuries. Survival was positively associated with fish length. Estimated overall trawl survival at 50%. Although Hoag used five condition codes in his study, he concluded that three codes would be sufficient to define discard condition.

Note: this paragraph describes our collection of condition information on surveys The IPHC has conducted many tagging operations throughout our history, during both setline and trawl research operations. We have condition code data in our database going back to 1961, and continuing through 1985. While data may have been recorded prior to this, that data is not in our computer database. Halibut were coded into one of three condition codes based on a subjective observations at release. Halibut were coded as excellent, average, or poor. Condition codes were next used by the IPHC during 1993 and 1994 on a setline study (see Kaimmer, 1994, below). In this and subsequent applications, the same codes were applied but were determined using criteria developed by IPHC for use by NMFS observers. They were also used in 1995 on a trawl tagging study (Forum Star) and in a study of survival of bycatch in trawls (Pikitch ??).

Prior to 1990, bycatch mortality was determined using gear specific mortalities which were applied to all fisheries using similar capture gear (see Table 1).

Although data on bycatch halibut condition in domestic fisheries has been collected by U.S. observers since the early 1980’s, in 1990 these data were used to determine mortality in these fisheries. The IPHC developed a set of criteria for these observers to use in examining bycatch halibut. The codes ‘average’ and ‘poor’ were changed to ‘poor’ and ‘dead’, and each of the three condition codes, excellent, poor, and dead, were defined more critically by a combination of capture injuries and physical condition. The criteria included capture injuries, muscle tone, presence of bleeding wounds, gill color, and evidence of predation. Separate criteria were developed for halibut caught by setline and trawl gear.

Williams and Wilderbuer (1991), using methods proposed by Williams (1990), used the observer observations to determine condition based estimates of halibut mortality based on frequencies of observed condition codes. Trawl mortalities were estimated from the study of Hoag. Conservatively applying the mortalities determined for small fish (< 81cm) to all sizes of halibut, trawl mortality rates of 50%, 73%, and 97% were used for excellent, poor, and dead fish, respectively. Mortalities for longlines were based on the results of Peltonen (96.5% survival for excellent fish), and Myhre (survival of poor fish is approximately half that of fish in excellent condition. They additionally assumed that all fish judged as dead would actually die. From these relationships, longline mortality rates of 3.5%, 52%, and 100% are used for excellent, poor, and dead fish, respectively (Williams and Wilderbuer, 1994).

Kaimmer (1994) reported on a 1987 experiment to compare and determine mortalities in sublegal size halibut caught by the directed halibut fishery and released by either careful shaking or using an automated hook stripper. A total of 2,098 halibut averaging 81 cm in length (± 17 cm) were tagged and released. Return rates were examined both by hook removal injury, and by hook removal method. Kaimmer concluded that hook strippers can result in mortalities up to 9 times more severe than those resulting from carefully shaking fish from hooks. Applying current condition code criteria to these recoveries, and assuming the 96.5 percent survival for excellent condition fish, mortality rates for poor and dead condition fish released from longlines are 45 percent and 77 percent, respectively.

Kaimmer and Trumble (1997, in press) reported on an experiment conducted in 1993 and 1994 to determine mortalities associated with careful release from two popular styles of sablefish and cod longline gear. A total of 12,851 halibut averaging 78 cm (±11 cm, circle hooks) and 76 cm (±9 cm, autoline hooks) were caught and removed from the hooks using one of three careful release methods prescribed by IPHC regulations and U.S. government regulations. As well, fish were removed using an automated hook stripper. Hook removal injuries and condition code were recorded for each fish. Return rates were examined by gear style, hook removal method and injury, and condition code. Still assuming a 96.5 percent survival for excellent condition fish, estimated survival rates for poor and dead condition fish released from longlines are 74 percent and 44 percent, respectively.

Table 1. Gear and area specific mortality rates used through 1989. Since 1990, domestic groundfish fisheries have used fishery-specific rates.

Fishery Years Mortality rate Foreign Trawl all 100% Joint venture 1979-1990 100% Catcher Processor (BSAI only) 1986-1989 100% U.S. Domestic trawl thru 1989 50% Bering Sea trawl 1990+ 75% Gulf of Alaska trawl 1990+ 65% B.C. Domestic trawl 1962+ 50% Foreign longline thru 1987 25% Domestic longline thru 1989 25% DAP longline 1990+ 16% Crab pot/shrimp trawl (all) all 100% Groundfish pot all 10%