673 



Abstract — Spatial variation in demo- 

 graphic parameters of the red throat 

 emperor (Lethrinus miniatus) was 

 examined among 12 coral reefs in 

 three geographic regions (Townsville, 

 Mackay, and Storm Cay) spanning 

 over 3° of latitude of the Great Bar- 

 rier Reef, Australia. Estimates of 

 demographic parameters were based 

 on age estimates from counts of annuli 

 in whole otoliths because there was no 

 significant difference in age estimates 

 between whole and sectioned otoliths. 

 There were significant regional differ- 

 ences in age structures, rates of somatic 

 and otolith growth, and total mortality. 

 The Townsville region was character- 

 ized by the greatest proportion of older 

 fish, the smallest maximum size, and 

 the lowest rates of otolith growth and 

 total mortality. In contrast the Mackay 

 region was characterized by the highest 

 proportion of younger fish, the largest 

 maximum size, and the highest rates 

 of otolith growth and total mortality. 

 Demographic parameters for the Storm 

 Cay region were intermediate between 

 the other two regions. Historic differ- 

 ences in fishing pressure and regional 

 differences in productivity are two 

 alternative hypotheses given to explain 

 the regional patterns in demographic 

 parameters. All demographic param- 

 eters were similar among the four reefs 

 within each region. Thus, subpopula- 

 tions with relatively homogeneous 

 demographic parameters occurred on 

 scales of reef clusters. Previous stud- 

 ies, by contrast, have found substantial 

 between-reef variation in demographic 

 parameters within regions. Thus spa- 

 tial variation in demographic param- 

 eters for L. miniatus may differ from 

 what is assumed typical for a coral-reef 

 fish metapopulation. 



Scales of spatial variation in demography of a large 

 coral-reef fish— an exception to the typical model? 



Ashley J. Williams 



School of Marine Biology and Aquaculture 

 and 



CRC Reef Research Centre 



James Cook University 



Townsville, Queensland. 4811, Australia 



E-mail address, astiley williams(a'|cu edu au 



Campbell R. Davies 



Bruce D. Mapstone 



CRC Reef Research Centre 



James Cook University 



Townsville, Queensland 4811, Australia 



Garry R. Russ 



School of Marine Biology and Aquaculture 



James Cook University 



Townsville, Queensland, 481 1, Australia 



Manuscript approved for publication 

 17 December 2002 by Scientific Editor. 



Manuscript received 4 April 2003 at 

 NMFS Scientific Publications Office. 



Fish Bull. 101:673-683 (2003). 



Estimates of demographic parameters, 

 such as growth and mortahty rates, are 

 fundamental to the understanding of a 

 species population dynamics and for 

 predicting responses of populations to 

 exploitation. Processes affecting popu- 

 lation dynamics operate at a number 

 of spatial and temporal scales (Levin, 

 1992) and can result in subpopulations 

 with distinct demographics. Differences 

 in demography between populations 

 may suggest geographic or reproduc- 

 tive isolation (or both) and as such 

 have been used in stock identification 

 for fisheries assessment and manage- 

 ment purposes (e.g. Begg et al., 1999). 



Identifying the "unit stock" has been 

 the primary focus of studies of spatial 

 structure of harvested populations in 

 most fisheries. Knowledge of spatial 

 structure within a unit stock is impor- 

 tant for both fisheries management, 

 because potential yields may vary 

 spatially within a population (Caddy, 

 1975), and for conservation, in order to 

 maintain intraspecific diversity (Niel- 

 sen, 1998). Hence, it is important to 

 estimate demographic parameters over 

 a range of temporal and spatial scales 

 to determine the scale(s) at which the 

 parameters vary significantly (Caley et 



al., 1996) and, therefore, to infer which 

 scales are of greatest importance for 

 assessment and management purposes 

 (Sale, 1998). 



Most coral-reef fish exist as metapop- 

 ulations of sedentary adult populations 

 linked by pelagic larval dispersal (Sale, 

 1998). Consequently, adult populations 

 of reef fish are commonly spatially seg- 

 regated and may be exposed to different 

 environmental, biological, and ecologi- 

 cal processes, resulting in spatial dif- 

 ferences in demographic parameters 

 at a range of spatial scales. Relatively 

 few studies, however, have focussed on 

 spatial variation in demographic pa- 

 rameters of harvested species of coral- 

 reef fish. Those that have, have gener- 

 ally focussed on spatial scales within 

 individual reefs or among reefs within 

 a single region (e.g. Ferreira and Russ, 

 1995; Hart and Russ, 1996; Newman et 

 al., 1996). Comprehensive multiscale 

 approaches are rare (but see Adams et 

 al., 2000; Meekan et al., 2001). 



The spatial structure of coral-reef 

 populations has generated considerable 

 interest in terms of the use of spatial 

 closures, or marine protected areas 

 (MPAs), as an effective tool for their 

 management (Roberts and Polunin, 



