Schiel and Breen: Population, ageing, and fishing mortality of Haliotts ins 



683 



and H. d. aquatilis (Kim and Cheung 1985). Inoue and 

 Oba (1980) suggest that shell circuli are annual for 

 Japanese H. gigantea. Prince et al. (1988) describe a 

 method for ageing Tasmanian H. rubra, based on the 

 method of Munoz-Lopez (1976). 



In two quota management areas that support sub- 

 stantial landings, we compare estimated levels of fish- 

 ing mortality with estimates of appropriate levels of 

 fishing mortality obtained from per-recruit modeling. 

 We also compare growth and mortality estimates from 

 shell growth rings with those obtained from other 

 methods. Finally, we explore the consequences of alter- 

 ing the present MLS. 



Materials and methods 



Field sampling 



The size structure of emergent H . iris populations was 

 examined around Stewart I., which comprises a major 

 part of the southern fishery, and the Marlborough 

 Sounds, which are a major part of the fishery in the 

 northern part of the South I. (Fig. 1). Stewart I. is ex- 

 posed to a range of weather conditions. The east coast 

 is in the lee of prevailing westerly winds, but is occa- 

 sionally subjected to severe easterly storms. The west 

 coast of the island is often inaccessible due to the 

 prevailing winds and a large onshore swell. During the 

 time of our sampling, the swell was reduced to about 

 1 m. The northern side of the island is protected from 

 the westerly winds but has a refracted westerly swell. 

 Paua occur along the entire coastline of Stewart I. 



The Marlborough Sounds are generally less exposed 

 than Stewart I. Paua occur only at the outer margins 

 of the Sounds, as the inner shores are protected and 

 provide generally unsuitable habitat for paua. Norther- 

 ly winds funnelling through Cook Strait are occasional- 

 ly severe and can bring exposed conditions to most 

 places where paua are fished. The eastern side of D'Ur- 

 ville I. is protected from westerly winds. Sites in the 

 central portion of the Sounds are generally subjected 

 to similar conditions of exposure to the north, while 

 southeastern sites are protected from the west and 

 north. One other site was used in this study. 



Karori Rock is on the south coast of the North I. It 

 is within an area closed to commercial fishing since 

 1974 after severe depletion of paua populations caused 

 by both commercial and recreational fishing. This site 

 is generally protected from severe swells but occa- 

 sionally experiences severe northerly and southerly 

 storms in Cook Strait. 



Except at Karori Rock, sampling sites were chosen 

 to be representative of commercially harvested sites. 

 This was done in consultation with commercial divers 



and, at Stewart I., a commercial diver assisted us in 

 the field. During 15-25 September 1989, 17 sites were 

 assessed at Stewart I., and 19 sites during 17-23 Oc- 

 tober 1989, at the Marlborough Sounds. Karori Rock 

 was sampled on 30 November 1989. Each sampling site 

 was roughly 1 ha in area. Paua were collected, taken 

 to a boat, measured to the nearest mm (shell length), 

 and a subsample from each site was frozen and taken 

 back to Wellington for weighing and ring counting. Re- 

 maining paua were returned to their habitat. The 

 numbers collected at each site varied (range 125-548), 

 depending on sea conditions and the abundance of 

 paua. The goal was to collect several hundred paua at 

 each site. To remove paua, we used a blunt trowel of 

 a type commonly used by commercial divers, and took 

 care not to cause damage. Paua less than 70mm were 

 not representatively sampled because they occupy cryp- 

 tic inshore habitats and require different sampling 

 techniques. We attempted to sample paua >70mm 

 representatively by having each diver remove all paua 

 as they were encountered. A total of 5790 (Stewart I.) 

 and 8731 (Marlborough Sounds) paua were measured. 

 For analyses, we grouped sites by regions of Stew- 

 art I. and the Marlborough Sounds (Fig. 1) because 

 we did not wish to assess fine-scale variability in 

 parameters. 



Ageing and growth 



Shells in the ring-count subsample were measured in 

 length and then sectioned sagittally through the spire 

 with a lapidary saw. Because the rings are obvious (Fig. 

 2) no further enhancement was necessary. Ring counts 

 for all shells were done by the same experienced 

 technician. 



From the ring-count sample data we estimated the 

 von Bertalanffy growth parameters with the non-linear 

 least-squares procedure FISHPARM (Saila et al. 1988). 

 We also estimated growth parameters from mark- 

 recapture data obtained at several sites (Fig. 1) in the 

 northern Marlborough Sounds and around D'Urville I. 

 (Talbot Murray, Min. Agric. Fish., P.O. Box 297, Well- 

 ington, pers. commun.). Abalone had been collected, 

 tagged with a numbered plastic disc attached to the 

 shell with epoxy putty, replaced, and recovered later. 

 Most of the times-at-liberty were one year; increments 

 for shorter and longer times (range 92-519 days) were 

 adjusted pro rata. We estimated asymptotic length 

 L^, and the Brody growth coefficient K from a Ford- 

 Walford regression (Ricker 1975). For comparison with 

 these results, we also calculated growth parameters 

 from age estimates (ring counts) obtained from abalone 

 in the same experiments (Talbot Murray, pers. com- 

 mun.). The ring counts were done by the same techni- 

 cian, as explained above. 



