Steves et al : Settlement and nursery habitats for demersal fishes 



185 



shelves, and the middle front, dividing the middle 

 and outer shelves. A third front, the shelf-break 

 front, separates the outer shelf from the continental 

 slope (McRoy et al., 1986). These three hydrographic 

 features can be utilized as an effective means for 

 describing the general nursery areas for gi-ound- 

 fishes within the NYB. 



Differences in environmental conditions occur not 

 only along the shelf, but also temporally. Patterns 

 in surficial sediments and bathymetry change on 

 the order of hundreds to thousands of years. These 

 characteristics are relatively stable in their distri- 

 butions on the shelf Other factors such as benthic 

 faunal distributions may change from year to year 

 but should remain unvarying during the first year of 

 a fish's life. Hydrographic variables and the presence 

 of ephemeral predator and prey species, however, 

 may change on a seasonal or even daily basis. The 

 scale and extent to which these variables change are 

 important in understanding the nature of the settle- 

 ment and nursery areas. 



The turnover of the cold pool in the fall represents 

 an abrupt change in bottom temperature that sub- 

 sequently may have biological significance. In our 

 study, higher mean catch of age-0 P. ferrugineus cor- 

 responded to the distribution of the cold pool, and 

 catch dropped dramatically in the fall as bottom 

 temperature increased during cold pool turnover. 

 For a species with a preference for low tempera- 

 tures, increased temperature may be a density-inde- 

 pendent source of mortality. In at least one case, 

 an increase in mortality of larval P. ferrugineus on 

 Georges Bank has been associated with an increase 

 in temperature from a warm-core ring ( Colton, 1959 ). 

 Although such abrupt temperature fluctuations at 

 greater depths are not common, they may occa- 

 sionally occur near the edge of the shelf in some 

 areas, perhaps associated with Gulf Stream mean- 

 ders (Able et al., 1993). Able et al. (1993) found a 

 temperature increase of 6°C over a two-day period, 

 and they suggested that such temperature fluctua- 

 tions may cause a cessation of feeding for tilefish, 

 Lopholatilus chamaeleonticeps. In contrast, some 

 species may benefit from increased bottom temper- 

 atures associated with turnover. For example, in 

 our study, settlement of M. bilinearis did not peak 

 until after turnover, when bottom temperature was 

 greater than 9'"C. Likewise, spawning distribution 

 for Atlantic croaker (Micropogonias undulatus) in 

 the southern portions of the MAB has a positive 

 association with the presence and location of warmer 

 waters coincident with the cold pool turnover (Nor- 

 cross and Austin, 1988). 



This study and another from Georges Bank (Frank 

 et al., 1992) indicate that the distribution of P. fer- 



rugineus corresponds well with bottom temperature. 

 The presence of P. ferrugineus is negatively corre- 

 lated with bottom temperature in the NYB, with 

 highest abundance in the coldest waters (<8°C). 

 However, on the Grand Banks the correlation with 

 temperature was positive, with most fish collected 

 above the warmest bottom waters (>3°C). Juveniles 

 from the Grand Banks, however, were collected off 

 the bottom, and highest abundances were found just 

 below the thermocline, in waters closer to 6°C. Tem- 

 peratures between 4° and 8°C have been noted as 

 the preferred temperatures for older P. ferrugineus 

 (Scott, 1982; Ross and Nelson, 1992; Walsh, 1992), 

 and the evidence from these studies suggests that 

 this may be true for age-0 P. ferrugineus as well. 



Besides P. ferrugineus, several other boreal spe- 

 cies, American plaice iHippoglossoides platessoides), 

 haddock (Melanogrammus aeglefinus), and Atlantic 

 cod {Gadus morhua), were collected in our samples 

 from the cold pool. Although not as numerically 

 dominant as P. ferrugineus, the presence of these 

 species this far down the coast suggests that the 

 cold pool may act as a temporary refuge from the 

 warmer waters normally associated with lower lati- 

 tude. Metapopulations of such species, wherein local 

 populations occupy small areas of suitable habitat 

 outside the main population, are a potential confound- 

 ing factor to the management of these species (Bailey, 

 1997). In the extreme, these juveniles may not be 

 viable additions to the population if environmental 

 conditions are not suitable for them to reach matu- 

 rity. Just as warm-core eddies and eddy streamers 

 may bring expatriates into this area from the south 

 in the summer (Hare and Cowen, 1991), the cold pool 

 may allow boreal species to temporarily extend their 

 range southward during the summer. It is possible 

 that conditions may occasionally enhance survival in 

 such potentially marginal habitats to the extent that 

 success of the population year class is facilitated. 



In our study, we used trawl data to describe the 

 general distribution and large-scale habitat associa- 

 tions of age-0 fishes. The resolution obtainable from 

 trawl data is limited to the area swept (about 700 

 m^). Such resolution is sufficient to determine differ- 

 ences between stations that are 10 km apart but not 

 to deterine the heterogeneity in habitat within a sta- 

 tion. In situ methods, including the use of manned 

 submersibles, are required for these types of small- 

 scale studies (see Auster et al., 1991, 1997; Adams 

 et al., 1995). One other drawback of using a trawl to 

 quantify fish abundance is that trawls are inherently 

 poor at collecting all fishes present in a given area; 

 gear selection is size- and species-specific. However, 

 if gear selection is constant between sites, trawl sam- 

 ples should be comparable (Kuipers et al., 1992). The 



