Previous studies have suggested a causal link between 

 seabird reproductive success and changes in fisheries ( Fumess 

 &Cooper, l982;Springer&Byrd, 1988). Because seabirds are 

 closely tied to the Bering Seaecosystem and are its most visible 

 component, results of studies monitoring their foraging and 

 breeding ecology can be used as an index to the "health" of the 

 ecosystem. For this reason alone, seabirds are an important 

 resource to both the USA and the USSR. In addition, native 

 cultures in both countries use birds and their eggs for food. The 

 extensive mobility of seabirds gives further motivation for 

 cooperative study and management of this resource. 



Relatively little research has considered the effect of 

 large-scale habitat patterns in this region on the distribution of 

 seabirds, particularly across existing political boundaries. Hunt 

 et al. ( 1981 ) described distributions of birds in the southeast 

 Bering Sea, and acknowledged the potential importance of 

 oceanographic conditions and the related availability of prey 

 for determining these observed distributions. Wahl et al. 

 ( 1 989) identified differences in seabird community assemblages 

 among broadly defined habitats in the southern Bering Sea. 

 Day et al. (in prep.) considered seabird distributions with 

 respect to water masses in the Benng Sea east of the USA- 

 USSR Convention Line in September 1985. Andrew and 

 Haney (Subchapter 10.1, this volume) describe these patterns 

 across the Chukchi Sea. Much remains to be learned, 

 quantitatively, about how macroscale variations in habitat 

 affect foraging seabirds across the Bering Sea Shelf. 



To this end. this report describes patterns of bird distribution 

 with respect to water types observed at depths of 20 m or less 

 during July and August 1988 from the Soviet research v^^ssel 

 (RA') Akademik Korolev. The term "'water type" is used here 

 to denote a parcel of water characterized by a unique combination 

 of physical and biological properties (i.e.. ranges of salinity 

 values and zooplankton communities). 



Each water type is given the same name as the formally- 

 recognized water mass (Coachman & Shigaev, Subchapter 2. 1 . 

 this volume) with the most similar properties (e.g.. ASW type 

 has similar properties to the water mass known as Anadyr 

 Stream water). The following hypotheses were tested using 

 these data (not stated as null hypotheses): 



1. Birds are unevenly distributed among water types, and 

 these distributions are related to physical features of each water 

 type that may affect availability of prey (zooplankton versus 

 fish) and effectiveness of foraging methods employed. 



2. Auklet distribution is related to the position of potentially 

 prey-rich waters (i.e.. parallels patterns of secondary 

 production), even if these waters occur at depth. 



This study is based on a relatively small number of 

 observations (96 "transect" sampling units). Given the temporal 

 and spatial limitations of these data, this report may best serve 

 as a preliminary "sketch" with which to develop further studies. 

 I am presently working with seabird data collected during 

 1987-1990 ill conjunction with National Science Foundation 

 project ISHTAR (Inner Shelf Transfer and Recycling) to 

 address these questions in greater detail. 



Water Types 



Three principal water types (differing in salinity by as 

 much as I ppt; see Methods section below) define the macroscale 

 ecology of the northern Bering Sea. These are: Anadyr Stream 

 water (ASW) the most western of the three; Bering Shelf water 

 (BSW),centrally located; and Alaska Coastal water (ACW) in 

 the east. Periodically, a fourth hydrographic entity, here called 

 Chukchi Coastal water (CCW), is observed as a shallow 

 surface layer near the western shore. Tlie seabird data considered 

 here were primarily collected from ASW and BSW types and 

 also from areas overlain by CCW. For the purposes of this 

 paper, CCW will be referred to as a "water type," even though 

 it does not have the spatial or temporal magnitude of the other 

 two. Where CCW occurs, it affects the habitat within which 

 seabirds are foraging (e.g., through stratification, stabilization, 

 etc. ) and these areas may be considered a different habitat from 

 areas where ASW or BSW is found at the surface. 



These water types exhibit distinct differences that affect 

 biological patterns across the shelf. Anadyr Stream water, 

 transported from the North Pacific Ocean at depth and upwelled 

 onto the Bering Sea shelf, is the most nutrient-rich of the three. 

 Intermediate nutrient concentrations are typically measured in 

 BSW, which originates in the Bering Sea (Coachman & Shigaev, 

 Subchapter 2. 1 , this volume ). The lowest nutrient concentrations 

 among waters of the Bering Sea Shelf are found in ACW. 

 Chukotski Coastal water is also nutrient-poor in comparison 

 with ASW or BSW. 



The properties of each of these water types are reflected in 

 distinct changes in the east-west distributions of flora and 

 fauna. Types of primary producers and their biomass vary 

 among the water types (Robie et al.. Subchapter 5.1.2, this 

 volume). Zooplankf' n communities are distinctively different 

 among water types ( Springer et al. , 1 989 ), and communities of 

 higher-order consumers such as fishes and mammals also 

 exhibit longitudinal variation that may be related to water type 

 effects (Nasu, 1974; Wyllie Echeverria & McRoy, Subchapter 

 5.3.1, this volume). 



Coachman and Shigaev (Subchapter 2.1, this volume) 

 provide further detail of the origins and physical properties of 

 the northern Bering Sea water masses. In addition, they 

 describe a separate water mass in this region. Gulf of Anadyr 

 water, that is apparently the same "cold core" previously 

 detected by other oceanographers (e.g., Zenkevitch, 1963). 

 This water mass apparently remains near the bottom in the 

 central Gulf of Anadyr and so was not included in this bird- 

 oriented study. 



Methods 



Spatial distributions of seabird densities and zooplankton 

 biomass were compared with each other and with the relative 

 locations of the three water types surveyed. Positions of 

 oceanographic stations and seabird transect counts are shown 

 in Figs. 1 and 2. 



389. 



