Hannah: A new method for indexing spawning stock and recruitment for Pandalus /ordani 



491 



strates that attempting to maintain a 

 spawner threshold by increasing es- 

 capement of age- 1 shrimp is unhkely 

 to be an effective management strat- 

 egy. The reason for this is that pre- 

 cisely when a boost in spawner abun- 

 dance is needed, the age-1 year class 

 is too depressed to offer significant 

 help, even with a complete cessation 

 of fishing. This result suggests that 

 when a year-class failure is detected 

 in this stock, reducing fishery impacts 

 on all age classes of shrimp may be a 

 better strategy to maintain spawner 

 abundance than the standard manage- 

 ment measures (e.g. mimimum codend 

 mesh or maximum count per pound) 

 which protect the escapement of age-1 

 shrimp. 



Discussion 



e 



10.0 



7.5- 



5.0- 



2.5 



Age-1 spawners 

 Age-2 and age-3 spawners 

 — Age-1 recruitment 



0,0 



I m I I H 



82 83 84 85 86 87 88 89 90 91 92 93 94 

 Year of spawning 



Figure 8 



Ocean shrimp spawner index, brolten down by age. and age-1 recruitment 

 inde.x for the same year, 1982-94. All data are calculated assuming elemen- 

 tal trawl efficiency of 0.50. Horizontal line indicates a hypothetical manage- 

 ment threshold for spawner abundance. 



This analysis supports the hypothesis 

 that the spring transition in coastal currents is im- 

 portant for successful ocean shrimp recruitment 

 (Hannah, 1993). Although data from this effort and 

 the prior study overlap for 10 of the 16 years pre- 

 sented here, new methods were used in this study 

 that did not depend on constant catchability and were 

 insensitive to failure in the assumption of constant 

 natural mortality and insensitive to assumptions 

 about trawl efficiency. The fact that this study and 

 the earlier effort both found a strong correlation be- 

 tween recruitment and SLH in April of the year of 

 larval release suggests that the interannual varia- 

 tion in shrimp recruitment is substantial, creating a 

 strong "signal" that can be observed through consid- 

 erable "noise" from various simplifying assumptions. 



The precise mechanisms by which a large drop in 

 spring sea levels coincides with strong recruitment 

 of ocean shrimp remains a matter for speculation. 

 This study did not attempt to investigate this issue. 

 A more in-depth discussion of the alternative hypoth- 

 eses can be found in the earlier paper ( Hannah, 1993 ). 

 The continued importance of April SLH as a strong 

 correlate of shrimp recruitment suggests that fur- 

 ther investigation of the potential mechanisms is 

 warranted. 



The principal novel finding of this study is the evi- 

 dence for a stock-recruitment relationship for ocean 

 shrimp. This evidence is statistical in nature and the 

 data series is fairly short, suggesting that this find- 

 ing should be considered preliminary at this time. 

 The fact that much of the environmental variation 



in recruitment has been accounted for in the mod- 

 els, in combination with the lack of a serial 

 autocorrelation in the recruitment index, supports 

 the contention that an actual stock-recruitment re- 

 lationship has been detected (Garcia, 1983). The form 

 of the detected relationship corresponds to the as- 

 cending leg of the curve, indicating that some reduc- 

 tion in average recruitment has resulted from low 

 spawner abundance. Unfortunately, the relationship 

 demonstrated gives little indication of the shape of 

 the right hand portion of the cui-ve, leaving the proper 

 functional form an open question. 



With environmental variation strongly influenc- 

 ing ocean shrimp recruitment, the detection of a 

 stock-recuitment relationship has different implica- 

 tions from those in a classical application of stock- 

 recruitment theory. In stocks with reasonably stable 

 recruitment, stock-recruitment curve parameters 

 can, in theory, be used to determine a maximum sus- 

 tainable yield. For shrimp, the multivariate recruit- 

 ment models (Table 2) define a surface over which 

 recruitment varies in response to the environment 

 and changes in the level of spawning stock. The en- 

 vironmental influence on recruitment also implies a 

 very strong recruitment-stock relationship; the points 

 residing in the lower left corner of Figure 4 are not 

 simply the result of fishing but are due to low re- 

 cruitment in the parent years, in combination with 

 fishing (Fig. 8). The main value of knowing that a 

 stock-recruitment relationship exists for ocean 

 shrimp is that it suggests strongly that there is a 



