Water 

 Type 



ASW 

 BSW 

 CRW 



Number of 

 Transects 



42 



47 



9 



TABLE 1 



Allocation of census effort by water type. 



Total time 

 (min) 



X 



Speed 



420 



470 



70 



15.0 

 15.0 

 15.0 



Range 



Area 



(km-) 



58.38 



65.33 



9.73 



X Area/ 

 transect 



1.39 

 1.39 

 1.39 



consumed, as reported in the literature (Bedard, 1969; Sowls 

 et al.. 1978; Hunt er ah. 1981; Springer & Roseneau. 1985). 

 Additionally, birds were classified by the principal foraging 

 method each species employs — that is, surface feeders, shallow 

 divers (<5 m) and deep divers (>5 m) (Ashmole, 1971 ; Sowls 

 etal., 1978; Brown, 1989). Only observations of sitting birds 

 were used for comparison with habitat parameters to provide 

 the most conservative estimate of birds actually using a water 

 type habitat. Both sets of reclassified bird data were examined 

 along with the other biological and physical data to determine 

 the degree of pattern concurrence and differences in average 

 abundance among water types. 



Hydrographic data were collected with a Sea-Bird(R) 

 CTD at all stations (Coachman & Shigaev, Subchapter 2. 1 , this 

 volume). Seawater density is arguably the most useful physical 

 property for differentiating between water types and masses. 

 Density is primarily a function of salinity in this and other high- 

 latitude baroclinic tlow systems (Royer, 1981). Further, the 

 communities that inhabit these water types differ, reflecting the 

 different sources of these waters, and this information can also 

 beusedinadefmitivemanner(Springer£'r(;/., 1989). Therefore, 

 salinity values and, secondarily, zooplankter distributions were 

 used collectively to delineate differences among water types. 

 Anadyr Stream water was defined as waters having salinities 

 >32.5, BSW salinities were <32.5 and >3 1 .0, and CCW salinity 

 values were <31.0. These definitions approximate those 

 described for water masses by Coachman and Shigaev 

 (Subchapter 2. 1 , this volume) forthiscruise of the RA'AAat/ew/A: 

 Korplev and also take into account water type positions as 

 defined by a more biological parameter, zooplankton 

 communities (see Wyllie Echeverria & McRoy, 

 Subchapter 5.3.1. this volume). 



The diving depth capabilities of planktivorous seabirds in 

 this region, primarily auklets (Aethia spp. ), are largely unknown. 

 However, water depths throughout the northern Bering Sea 

 Shelf region do not exceed 50 m, and the most conservative 

 prediction of diving capabilities based upon currently available 

 information (Piatt & Nettleship, 1985) puts most of the water 

 column within reach of even the smallest of these birds. 

 Auklets depend heavily on large deep-water copepods advected 

 onto the Bering Sea Shelf via ASW (Springer & Roseneau, 

 1985). and so they may be able to exploit prey stocks carried in 

 the ASW at depth. For this reason the position of each bird 

 transect was determined with respect to water type distribution 

 at the surface, and distribution of auklets in particular was also 



compared with water type distributions at each of 5, 10. 15, and 

 20 m depths. 



Zooplankton were collected by vertical tow of a 

 1-m- diameter ring net at all oceanographic stations (Fig. 1). 

 These tows were made from the ocean bottom through the full 

 water column, so the data reported here give an integrated view 

 of zooplankton distribution and abundance, but it is not possible 

 to derive depth of occurrence for any given zooplankter. 

 Evidence from previous oceanographic cruises suggests that 

 the majority of zooplankters concentrate near the pycnocline, 

 which is generally located in the upper 15-20 m in the north 

 Bering Sea in summer (Coachman, 1986; Hunt et al., 1990; 

 ISHTAR group, unpubl. data). Zooplankton samples were 

 preserved at sea in 5% formalin in seawater. In the lab, these 

 samples were sorted by the lowest taxonomic level possible, 

 usually to species (Springer et al., 1989). To compare the 

 distribution and abundance of zooplankton with that of seabirds, 

 only zooplankton species reported to be most important in the 

 diets of planktivorous seabirds are considered here. These are 

 calanoid copepods, including Calamts marshallae, Neocalanus 

 plumchni.'i, N. cristatus. and the euphausiid genus Thyssanoessa 

 (Bedard, 1969; Piatt et al., 1988; Hunt & Harrison, 1990), 

 Because it is unknown which particular zooplankton species 

 birds may have been feeding on during this cruise, if any. these 

 four species will be considered collectively and their quantities 

 summed as an estimate, or index, of available prey. 



Statistical comparisons were made between categorically 

 classified seabird data and water type position at the surface. 

 Except for the case of auklets as discussed above, more 

 species-specitlc comparisons generally were not often possible 

 because of the high variability associated with at-sea 

 observations. Significant differences in seabird abundances 

 among water types were detected using a series of Kruskal- 

 Wallis nonparametric tests (g.< 0.05) (Zar, 1984). 



Results 



Densities of all seabird species combined were not 

 significantly different among water types (p > 0.05. Table 2). 

 However, average densities of surface-feeding birds were 

 significantly more abundant in ASW than BSW or CCW 

 (Fig. 3). Densities of this group of birds were not highest in 

 BSW but significantly different between BSW and CCW. 

 Abundances of shallow-feeding birds, predominantly short- 

 tailed shearwaters and black-legged kittiwakes, were not 



391 



