742 



Fishery Bulletin 88(4). 1990 



For king crabs, the relationship between time in cap- 

 tivity and immediate mortality was strong enough that 

 knowledge of average processing times for a particular 

 trawl fishery could be used to estimate overall survival 

 of crabs. It can also allow estimation of the effects of 

 reduced captivity time on survival. In this study, im- 

 mediate survival (0.527) and delayed survival (0.472) 

 of king crabs were approximately equal (Table 2). For 

 illustrative purposes we can assume that this equiva- 

 lency is a general phenomenon, and these two aspects 

 of survival vary proportionally, since increased stress 

 should result in increases in both immediate and 

 delayed mortality. Based on this assumption, the esti- 

 mated overall survival at the median lethal time of 9.3 

 hours would be approximately 0.5 x 0.5 or 0.25, which 

 is very close to the value of 0.21 estimated in Table 2, 

 but differs slightly because the latter was calculated 

 in a stratified manner over all conditions rather than 

 just representing the median lethal time. Overall sur- 

 vival could theoretically be doubled to 50% by reduc- 

 ing CAPTIME such that immediate (and consequent- 

 ly delayed) survival was equal to 0.5"-, or 0.71, i.e., by 

 reducing mortality to 1-0.71, or 0.29. Solving the 

 logistic equation (Eq. 3) for X results in: 



X = X50 - {ln|(l -M)/M]}/r 



(3.1) 



Substituting 0.29 = M (as determined above), 9.34 = 

 Xr,„, and 0.367 = r (the latter two values are fitted 

 parameters from the logistic model for king crab sur- 

 vival) results in X = 6.90 hours. Thus, for this exam- 

 ple, reducing the median total time in captivity from 

 9.34 to 6.9 hours (a 26% reduction) would increase the 

 immediate and delayed survival rates from 0.50 to 0.71 

 (a 42% increase), and consequently increase the overall 

 survival to a value equal to the square of immediate 

 survival, 0.71 x 0.71 = 0.50, i.e., a 100% increase in 

 overall survival, assuming immediate and delayed sur- 

 vival to be approximately equal, as stated above. 



Reduction of CAPTIME, the time crabs spend in the 

 trawlnet or the processing bins, could be accomplished 

 by reducing either towing time, processing time, or 

 both. If towing time cannot effectively be reduced (and 

 may in fact need to be increased to efficiently use the 

 experimental trawls because of reduced fish catch 

 rates) then delay time (between net haulback and 

 delivery to the processing vessel) and processing time 

 (time spent on deck or in the bunkers prior to recovery) 

 should be reduced as much as possible. This could be 

 achieved through better communication and coordina- 

 tion between catcher and processor vessels in order to 

 more evenly distribute deliveries to the processor over 

 time, and reduce or prevent a backup in processing. 

 Towing time may actually contribute less to mortality 

 than processing time, since fish and crabs are less 



densely packed in the net during towing, and have ac- 

 cess to free flowing water. However, neither of these 

 variables (towing or processing time) provided a signifi- 

 cant predictive relationship by themselves. Once the 

 codend is recovered on deck, the packing density and 

 pressure within the codend increase greatly and prob- 

 ably contribute substantially to mortality of fish and 

 crabs therein. 



Observers placed aboard trawlers by NMFS or the 

 State of Alaska have a variety of tasks to perform, and 

 assessing crab vitality has usually been one of low 

 priority. If one objective were to determine the pro- 

 portion of crabs that clearly have the best chance of 

 survival, then a two-step process using both vitality and 

 the presence of injuries, such as that currently used 

 by the NMFS observer program, is a better method, 

 as crabs which fell into the "best" category of this 

 scheme exhibited 93.2% survival. Even better predic- 

 tions could probably be made if degrees of injury (e.g., 

 none, mild, severe) were coded and used to predict sur- 

 vival. However, if the object were to predict survival 

 using the quickest and most efficient method of assess- 

 ment, then use of vitality alone might be the best 

 choice, as it indicated similar survival (92.5%) for the 

 top category and could be assessed adequately in sec- 

 onds, whereas a comprehensive assessment for injuries 

 required a deliberate, systematic examination, consum- 

 ing more time than may be available, given the prior- 

 ity of the task. Even then, some injuries, such as minor 

 cracks in the carapace, were easy to miss. 



Despite the quantity of data collected on injury types 

 and severity, there was no clear, unambiguous rela- 

 tionship between injuries and mortality. The overall 

 physical condition of crabs, which could be assessed by 

 activity level, was a better indicator of future survival 

 than injuries. Leg injuries were only slightly more fatal 

 than bodily injuries to king crabs. Both immediate and 

 delayed mortality increased with the number of legs 

 injured, and proximal injuries to legs were more fatal 

 than distal injuries, which was not unexpected since 

 bleeding from the coxa or basi-ischium (leg segments 

 which are proximal to the autotomy plane) would like- 

 ly be more difficult for the crab to stop than bleeding 

 from a more distal segment, which could be autotom- 

 ized at a later time. In our experience, though, it was 

 uncommon for crabs that had injured legs at the time 

 they were examined to autotomize them in the live 

 tanks. If autotomy had not occurred soon after the in- 

 jury, and prior to examination, it was less likely to do 

 so after a day or two. However, this is a subjective 

 observation and was not documented, because injuries 

 wei'e not reassessed prior to release. The poor relation- 

 shi}) between injuries and death suggests that other fac- 

 tors, such as exposure or suffocation, may have played 

 a significant role in determining survival of crabs. 



