Mueter and Norcross: Spatial and temporal patterns in the demersal fish community off Alaska 



579 



Ocean between the period of the 1960s and 70s and the 

 1980s and 90s. Polovina et al. (1995) suggested that the 

 increase in primary and secondary productivity in the NE 

 Pacific Ocean resulted from a shoahng of the mixed layer 

 in the GOA by 20-309f associated with variations in the 

 strengtli and position of the Aleutian low pressure system 

 from the late 1970s to the late 1980s. Over the same time 

 period large increases occurred in chlorophyll a levels 

 (Venrick, 1995). zooplankton stocks (Brodeur and Pearcy, 

 1992; Brodeur et al., 1996), and upper trophic level species 

 (Brodeur and Ware, 1995; Francis et al., 1998). Our study 

 suggests that there has been a parallel increase in the bio- 

 mass of demersal fish communities on the GOA shelf and 

 upper slope since 1984, in response to the overall increase 

 in productivity in the NE Pacific Ocean. 



Given the increasing emphasis on ecosystem-based 

 management, managers will have to rely on a variety of 

 indicators in addition to traditional single-species biomass 

 estimates to assess the status and health of an ecosystem 

 (Yaffee et al., 1996). Indices of species composition like 

 those used in this study, as well as survey-based assess- 

 ments of noncommercial species, provide indicators that 

 can help researchers in assessing changes in the commu- 

 nity and can help managers in responding to such changes 

 in a timely fashion. 



Species richness and diversity are particularly simple, 

 yet potentially very useful indices. The concept of marine 

 biodiversity has received considerable attention in recent 

 years (NRC, 1995) and should be included in planning 

 and policy-making (Bengtsson et al., 1997). There is some 

 evidence that bottom fishing can reduce the diversity of 

 benthic communities (Collie et al., 1997) and, in turn, 

 that changes in diversity can impact ecosystem function 

 and productivity (Naeem et al., 1994). There was some 

 indication in our study that the number of species and the 

 diversity of the groundfish community in the GOA as mea- 

 sured by the Shannon-Wiener index decreased between 

 1993 and 1996 (Fig. 4). Although gear effects may have 

 confounded differences among earlier surveys, the same 

 gear and identical survey designs were used in both 1993 

 and 1996; thus the differences in species richness and di- 

 versity cannot be attributed to sampling effects and merit 

 further study. The simple diversity indices used in our 

 study were primarily chosen for practical considerations. 

 Other indices, e.g. indices of beta-diversity, may be more 

 appropriate for describing diversity of marine ecosystems 

 and should be explored. 



Multivariate techniques that summarize the major 

 variation in species composition with a small number 

 of indices (dimensions) provide another useful tool for 

 monitoring changes in groundfish communities. Such in- 

 dices provide sensitive indicators of change in community 

 structure and can identify general trends that may not be 

 apparent when using univariate measures (Austen and 

 Warwick, 1989). The indices reduce the noise from random 

 fluctuations of many individual stocks and the resulting 

 "signal" may be used to test statistically for differences in 

 species composition among areas or time periods, as well 

 as helping to identify those species that show significant 

 trends. 



Another important part of an ecosy.stem approach to 

 management is the monitoring of the abundance of non- 

 commercial species, in addition to commercial species. 

 Changes in the abundance of skates, sculpins, and other 

 noncommercial species are examples of potentially im- 

 portant trends that may go undetected if management 

 agencies focus on the assessment of commercial species 

 only. Although these species comprise a small proportion 

 of total biomass, such trends may be an early indication 

 of changes in the groundfish community due to changes in 

 the environment or fishing. Such monitoring, along with 

 the monitoring of environmental variables, will further 

 help to understand how and why groundfish communities 

 change in response to environmental variation. 



Current fisheries management often fails to account for 

 the ecological complexities inherent in fish communities 

 (Roberts, 1997). This is hardly surprising because of the 

 focus of most stock assessment on individual stocks and 

 on stock-recruitment relationships. Prediction becomes 

 increasingly difficult as other species and environmental re- 

 lationships are added to models. Nevertheless, the complexi- 

 ties should not discourage researchers from trying to under- 

 stand species relationships and the response of species and 

 communities to environmental variation and harvesting. 



Knowledge of such relationships can improve manage- 

 ment in several ways. First, it can help to explain why 

 stocks are fluctuating in particular ways and help manag- 

 ers devise management strategies that take into account 

 such fluctuations. Second, it can improve our understand- 

 ing of multispecies or community-level relationships and 

 can lead to better assessment methods that take such 

 relationships into account. Third, it can lead to predictive 

 hypotheses that can be explored and tested through rigor- 

 ous research programs and adaptive management proce- 

 dures. Ultimately, improved understanding of ecosystem 

 processes will improve fisheries management only if it can 

 reduce uncertainties in prediction. Thus the challenge re- 

 mains to develop improved models that account for species 

 relationships and environmental influences. 



Acknowledgments 



We thank Eric Brown and Michael Martin at the Alaska 

 Fisheries Science Center, NOAA, Seattle, for granting us 

 access to Gulf of Alaska bottom trawl survey data and for 

 help with retrieving the data. The manuscript was greatly 

 improved thanks to thorough reviews of earlier drafts 

 by Ron Barry, Sue Hills, Mark Johnson, Terry Quinn, 

 Bob Foy, and five anonymous reviewers. This work was 

 funded in part by NOAA Saltonstall-Kennedy grant no. 

 50ABNF700086. 



Literature cited 



Anderson, P. J., and J. F. Piatt. 



1999. Community reorganization in the Gulf of Alaska fol- 

 lowing ocean climate regime shift. Mar. Ecol. Prog. Sen 189: 

 117-123. 



