SMITH and HOWELL: BOTTOM TRAWLING EFFECTS ON AMERICAN LOBSTERS 



American lobsters <81 mm CL, taken from pot 

 and trawl vessels, were held to estimate trawl- 

 induced delayed mortality (Table 3). Of 374 

 trawl-caught lobsters held in the laboratory, 47 

 (12.6%) had sustained major damage. Eighteen 

 were newshell and were treated as one category 

 regardless of damage sustained. Two of 309 

 (0.6%) undamaged, hardshell trawl-caught lob- 

 sters died whereas 55.3% of damaged ones died 

 within the 14-d period, most within the first 7 

 days. Six of 18 newshell lobsters died (33.3%). Of 

 153 pot-caught lobsters, 8 (5.2%) had major dam- 

 age; none were newshell. No pot-caught lobsters 

 experienced delayed mortality. 



Laboratory delayed mortality rates for dam- 

 aged hardshell American lobsters were pooled in 

 May, July, and August because of the small sam- 

 ple sizes (n = 14). Delayed mortality rates ranged 

 from 50 to 100% in the three months. The mean 

 (85.7%) was applied to the proportion of damaged 

 hardshell lobsters in the commercial trawl fish- 

 ery in each of those months. In November, a 

 42.4% delayed mortality rate (n = 33) was ap- 

 plied to the damaged lobster category in Novem- 

 ber commercial samples. Although only two un- 

 damaged hardshell lobsters died, both occurred in 

 July (2 of 19). This proportion (10.5%) was applied 

 to the proportion of undamaged hardshell lobsters 

 in the July commercial catches. Since no delayed 

 mortality occurred to undamaged hardshell lob- 

 sters in May, August, or November, no delayed 

 mortality rate was applied to that category for 

 those months. The delayed mortality rate for 

 newshell lobsters (33.3%, n = 18) was applied to 

 the proportion of newshell lobsters in commercial 

 catches (1-8%). Immediate (on-deck) and delayed 

 mortality rates for lobsters <81 mm CL and for 

 each shell condition (undamaged hard, damaged 

 hard, and newshell) were summed, resulting in 

 an estimate of total mortality for each season 

 (Table 3). 



Estimated trawl-induced mortality rates for 

 the four sample periods were tested to determine 

 the relative importance of high seasonal water 

 temperatures (July and August) and molting 

 (November and July). November and July had 

 significantly higher rates of mortality than May 

 and August (Tables 3, 4), indicating that molt 

 condition (July and November) is more important 

 in determining the extent of mortality than water 

 temperature. There was no significant difference 

 between May and August delayed and immediate 

 mortality estimates (Table 4) despite a 10°C dif- 

 ference in water temperature. However, warm 



temperatures appeared to increase the incidence 

 of mortality after molting since the July rate was 

 significantly higher than the November rate 

 (21.3% vs. 6.3%, Table 4). 



Table 4. — Log-likelihood ratio test of expected trawl- 

 induced mortality to American lobsters <81 mm CL, by 

 season. Underlining denotes no significant difference 

 at P = 0.05. x2 = 178.4* overall, 44.8* (July- 



November), 

 August). 



11.8* (November-May), and 1.8 (May- 



*= P< 0.005. 



Mortality Due to Freezing 



Exposing undamaged American lobsters to am- 

 bient air temperatures of -9.5°C produced no 

 mortality at 30-min exposure, and 70% and 100% 

 mortality at 60 minutes and 120 minutes, respec- 

 tively. Damaged lobsters were not tested since 

 they occur so infrequently during cold-water peri- 

 ods of the year (Table 1). Freezing temperatures 

 may also induce reflex amputation (autotomy) of 

 claws. One lobster of 20 in the 30-min sample and 

 four American lobsters of 20 in the 60-min sample 

 autotomized one claw. 



DISCUSSION 



Jamieson and Campbell (1985) found that sea 

 scallop dragging in eastern Canada could damage 

 American lobsters, but since the sea scallop and 

 lobster fisheries generally were not simultaneous 

 (lobsters tended to emigrate from the scalloping 

 areas each season prior to the advent of the drag 

 fishery), the use of scallop gear over beds that 

 hold lobsters at other times posed no significant 

 impact to the resource. Scarratt (1972) observed a 

 similar situation in the eastern Canadian Irish 

 moss rake fishery. Although American lobsters 

 did suffer damage from the gear, most lobsters 

 emigrated before the moss harvest season, so the 

 damage associated with the gear was minimal. In 

 western Long Island Sound, the lobster pot and 

 mixed species trawl fisheries both operate 

 throughout the year. Since the seabed in this area 

 is relatively uniform and generally free of 



741 



