U.S. Dept. of Commerce / NOAA / OAR / PMEL / Publications
The patchiness of ichthyoplankton is influenced by many biological factors including spawning mode (pelagic vs. demersal), frequency of the spawning (single vs. batch), fecundity, daily mortality rates and ontogenetic changes in behavior (Frank et al., 1993; McGurk, 1986; Matsuura and Hewitt, 1995). The low patchiness values for all stages of eggs compared to early larvae was surprising, considering the nonuniformity of spawning aggregations. Apparently, the eggs diffuse quickly in the horizontal but remain layered vertically due to differences in egg buoyancy (Kendall and Kim, 1989; Kim, 1987). Our calculations for egg patchiness (3.9-6.1) were similar to those (4.3-6.8) determined from finer scale net camera observations (Brodeur et al., 1996b, see pp. 92-111 in this supplement). They are also similar to values (2.5-7.1) for different egg stages of Browns Bank haddock (Melanogrammus aeglefinus), and a demersal-spawning gadid with pelagic larvae (Koslow et al., 1985). They are, however, substantially lower than estimates for eggs (3-61) calculated for pelagic-spawning Pacific sardine (Sardinops sagax) off California (Smith, 1973). The patchiness values calculated for larvae are consistent with previous results for walleye pollock (Kim, 1987). Those calculated for larvae and early juveniles compare well to values obtained for other species (McGurk, 1986).
A U-shaped pattern in larval patchiness as a function of age is characteristic of the larvae of many marine fish (Frank et al., 1993; Matsuura and Hewitt, 1995). The high patchiness we observed in the early larval stage during some years resulted from larvae interacting with the physical environment (eddies and currents). The decrease in patchiness with age resulted from diffusion and the dissipation of the eddies. Only in the late larval and juvenile stages could swimming and schooling influence large scale (~10 km) distributions.
Patches can have important implications for predator-prey interactions of larval pollock. Given that other similar-sized zooplankton are more abundant (Napp et al., 1996), it is unlikely that patches of pollock larvae are of sufficient density to attract swarming predators. Conversely, if pollock larvae coexist in patches with more numerous zooplankton, those predators feeding within the patch may saturate quickly, reducing predation pressure on larvae.
Mechanisms that aggregate larvae may likewise aggregate their prey. To date, the evidence that it is advantageous for larvae to feed in a patch is equivocal. In late April 1989, larvae in a patch (and eddy) had higher prey concentrations, had better nutritional condition and fuller guts than larvae not found in a patch. Later that year, prey concentrations and the associated larval feeding were equivalent in or out of patches (Canino et al., 1991). A similar situation existed in 1990 (Bailey et al., 1996).
Results from modeling studies suggest that wind mixing of the upper water column can be either beneficial or detrimental to larval survival (Davis et al., 1991). A relationship between wind-mixing, stratification within an eddy and larval behavior, and their subsequent survival has been suggested for Shelikof Strait (Bailey and Macklin, 1994). Observations within an eddy revealed enhanced prey and feeding conditions together with a low-salinity core relative to surrounding waters (Canino et al., 1991). The vertical stratification of this feature suppressed wind-induced turbulence relative to that in adjacent waters. In laboratory experiments, pollock larvae avoid turbulence (Olla and Davis, 1990) by going deeper. Light intensity decreases with depth, which has detrimental effects on the ability of larvae to search for and capture prey (Heath, 1989). This coupling of biology and physics provides a potential mechanism for enhanced larval survival in a patch maintained by an eddy, or any region with enhanced stratification. Conservation of metabolic energy has also been suggested for young salmon in an eddy (Freeland, 1988). With the present observations, however, it is not possible to ascertain quantitatively the importance of patches associated with eddies to larval survival.
Larval aggregations maintained by dynamics that increase retention within the patch may convey definite advantages to larvae. The selection of pollock life history traits and spawning location would appear to be coupled to an area where eddies are common and island retention features also exist. The former coupling also occurs in the eastern Bering Sea (Schumacher and Stabeno, 1994). For those larvae associated with recurrent patches around Sutwik Island, removal from the shelf is highly unlikely and transport to nursery grounds on the shelf (Hinckley et al., 1991) assured. Furthermore, waters shoreward of the ACC may be more stratified than offshore waters and stratification is related to enhanced plankton production and vertical layering of prey (Napp et al., 1996). Both of these conditions allow the Sutwik Island patch to contribute to larval survival and later recruitment.
We wish to thank all members of FOCI who helped to collect and process the data used in this paper. Special thanks to A. Hermann for free exchange of ideas and thought-provoking conversations, W. Rugen for providing the data in a useful format, and Dr. L. Ejsymont and the staff of the Polish Plankton Sorting and Identification Center, Sczcecin, Poland who processed the ichthyoplankton data. This research is contribution FOCI-0244 to NOAA's Fisheries Oceanography Coordinated Investigations, and contribution 1646 from NOAA's Pacific Marine Environmental Laboratory.
Return to previous section or go to next section