Parrish and Kazama: Ghost fishing in the Hawaiian lobster fishery 



721 



Hawaiian Islands (NWHI); an es- 

 timated 2000 of these traps were 

 unrecovered (Landgraf et al. 

 1989). The annual accumulation 

 of lost plastic traps on the banks 

 where commercial trapping oc- 

 curs must be considered a poten- 

 tial hazard to the lobster stocks. 

 No field studies have been done 

 on the interactions between lost 

 traps and the adults of the two 

 target lobster species. The objec- 

 tives of this study were to (1) 

 evaluate the persistence of lost 

 commercial traps under field con- 

 ditions, (2) estimate retention of 

 target species in plastic traps 

 with bait depleted, and (3) assess 

 mortality of lobsters unable to 

 exit traps. 



Methods 



:^^^i£s*S 





Figure 1 



Plastic trap used in the Hawaiian commercial lobster fishery. 



Study sites 



The prohibitive cost of prolonged, ship-supported 

 diving operations in the NWHI dictated that all field 

 experiments be conducted at Oahu. Sites close to the 

 windward shore of Oahu provided appropriate depths 

 (30-40 m) and habitat consistent with the NWHI com- 

 mercial banks. The area is known to harbor exploitable 

 numbers of the lobster fishery's target species. Its 

 heavy seas, strong bottom surge, and swift currents 

 (Bathen 1978) might mimic NWHI conditions and thus 

 could test the stability of lost traps. Traps placed in 

 such rough conditions, without surface markers, were 

 unlikely to be disturbed by fishermen and other recrea- 

 tional users. 



Trap stability and faunal interactions 



A string of eight empty traps was deployed in a linear 

 orientation from February to July 1990. The selected 

 area afforded two types of adjacent habitat— high-relief 

 rugose bottom, and hard relatively-flat bottom— allow- 

 ing comparisons of trap stability and use by lobsters 

 in the different environments. Individual traps were 

 set 10m apart, four on high-relief bottom and four on 

 adjacent even bottom. The molded-plastic traps (Fig. 

 1) used in the study were a standard commercial model 

 employed throughout the commercial fishery in Hawaii 

 (Fathoms Plus Marine Implementation, P.O. Box 6307, 

 San Diego, CA 92106). Each trap consisted of a single 

 chamber with two side entrances and was weighted 

 with about 10 kg of lead, as is conventional in the 



fishery. Traps remained in place over a 6-month period. 

 They were observed monthly by scuba divers during 

 three dives conducted at 48-hour intervals. Physical 

 condition, movement, and contents of the traps were 

 noted on each dive, along with general observations of 

 the surrounding area. The monthly censuses recorded 

 the initial presence of lobsters from the surrounding 

 study site and any exits or entries over the following 

 4 days. The area in and under the traps was examined 

 for exoskeleton remnants that might indicate molting 

 or mortality. One additional trap with its hinge pins 

 removed was deployed on flat bottom near the trap 

 string to mimic a trap with corrodible hinge pins that 

 had deteriorated. 



Trap stocl<ing experiment 



In the summers of 1990 and 1991, traps of the same 

 type (Fig. 1) were deployed in the field and laboratory, 

 and stocked with spiny and slipper lobsters from the 

 NWHI to evaluate their ability to exit and the extent 

 of mortality. Prior to the traps being stocked, lobsters 

 had spent 3-8 days in transit in continuously-flushing 

 shaded bait wells where they were fed daily. Mean 

 carapace lengths were 87.6 mm (A^ 96, range 67.4- 

 121.7mm) for spiny lobster, and 83.3mm (A'' 96, range 

 50. 1-99.7 mm) for slipper lobster. Antennae were tagg- 

 ed with color-coded, plastic, self-locking, electrical ties 

 to permit visual identification of individuals without 

 their being handled during the experiment. Molt stage 



