Vol. 382: 211–219, 2009 MARINE ECOLOGY PROGRESS SERIES Published April 30 doi: 10.3354/meps07997 Mar Ecol Prog Ser Seabird seasonal trophodynamics: isotopic patterns in a community of Pacific alcids W. E. Davies1, 3, J. M. Hipfner1, K. A. Hobson2, R. C. Ydenberg1,* 1Centre for Wildlife Ecology, Simon Fraser University, 8888 University Blvd., Burnaby, British Columbia V5A 1S6, Canada 2Environment Canada, 11 Innovation Blvd., Saskatoon, Saskatchewan S7N 3H5, Canada 3Present address: Department of Ecology & Evolutionary Biology, University of Toronto, 25 Willcocks Street, Toronto, Ontario M5S 3B2, Canada ABSTRACT: We measured δ15N and δ13C values in the blood of breeding adults and nestlings of 5 species of alcids at Triangle Island, British Columbia, to estimate the extent to which these seabirds alter their foraging ecology across successive breeding stages. Considerable intraspecific (stage-to- stage) and interspecific variation was found. Two species—common murre Uria aalge and pigeon guillemot Cepphus columba—fed consistently at high trophic levels (i.e. diets of fish) in inshore or benthically linked habitats. The foraging ecology of 3 others—Cassin’s auklet Ptychoramphus aleu- ticus, rhinoceros auklet Cerorhinca monocerata and tufted puffin Fratercula cirrhata—was more variable. Tufted puffins exhibited especially dramatic trophic and habitat shifts between early and late-season diets. With the exception of tufted puffin, the diet of provisioning adults differed from that fed to their nestlings. Trophic level of the community as a whole increased as the season progressed due to the combination of trophic shifting by rhinoceros auklets and tufted puffins, and earlier breed- ing by zooplanktivorous Cassin’s auklets than by piscivorous murres and guillemots. Our results con- tribute to a growing body of evidence that marine bird species exhibit considerable flexibility in their foraging behaviour and also shed new light on seasonal patterns in the trophic relations within marine bird communities. KEY WORDS: Alcids · Breeding stages · 13C · Community ecology · Foraging ecology · 15N · Trophic relationships · Stable isotope analysis Resale or republication not permitted without written consent of the publisher INTRODUCTION reproductive traits remains, to a large extent, unre- solved. Seabirds play critical roles in the transfer of energy In recent years, new techniques (stable isotope and nutrients within marine ecosystems and also analysis, fatty acid analysis) have provided novel between marine and terrestrial ecosystems. Despite insights into previously less-accessible facets of verte- decades of intensive study, however, key facets of the brate foraging ecology (Kelly 2000). Stable isotopic reproductive ecology of marine birds remain poorly analysis (SIA) is based on the fact that organisms incor- understood. In particular, there is a notable discrep- porate into their body tissues the isotopic signals of the ancy between the depth of our understanding of on- environments in which they live, as well as the foods colony versus off-colony (including the non-breeding they eat. This analytical approach has a particularly season) facets of their biology, stemming from the fact rich history in marine ecological research (Hobson et that seabirds are relatively easy to study on land, but al. 1994) and has revealed that some marine birds con- much more difficult to study at sea. As a result, the sidered to be narrow trophic specialists actually utilise question of how the foraging ecology of marine birds is a surprisingly wide range of trophic levels and marine adaptively related to the form and function of their habitats while breeding (Forero et al. 2004), with vari- *Email: [email protected] © Inter-Research 2009 · www.int-res.com 212 Mar Ecol Prog Ser 382: 211–219, 2009 ation found among colonies (Forero et al. 2002), sub- provisioning adults with those of offspring, we aimed colonies (Hipfner et al. 2007), foraging trip duration to provide a uniquely comprehensive view. (Cherel et al. 2005), age (Hodum & Hobson 2000) and gender (Forero et al. 2005). To date, however, few studies based on SIA have MATERIALS AND METHODS examined the extent to which marine birds alter their foraging behaviour across discrete breeding stages Triangle Island (120 ha; 50° 52’ N, 129° 05’ W) lies (but see Quillfeldt et al. 2005). Flexible foraging 46 km off the northwest tip of Vancouver Island, British behaviour could be an important mechanism by which Columbia, Canada, in shallow (<200 m) waters some seabirds meet the unique nutritional demands of each 25 km from the edge of the continental shelf. The stage and, within the larger ecological context set out site lies within the transitional zone between the by the seasonal progression of events, in the marine upwelling-driven California Current ecosystem to the environment (Levinton 2001). While several SIA stud- south and the downwelling-driven Alaska Gyre system ies have documented the trophic relationships among to the north. Approximately 1.2 million alcids breed on the component members of seabird communities (Hob- Triangle Island, including sizeable populations of son et al. 1994, Forero et al. 2004), few have considered 5 species: common murre Uria aalge, pigeon guillemot temporal variation in the relationships across the com- Cepphus columba, Cassin’s auklet Ptychoramphus munity as a whole (Sydeman et al. 1997, Cherel et al. aleuticus, and 2 species of puffins, rhinoceros auklet 2008). Combined with more conventional ‘snapshot’ Cerorhinca monocerata and tufted puffin Fratercula studies, the examination of seasonal trophodynamics cirrhata. Two other alcids, thick-billed murre Uria could provide a more comprehensive understanding of lomvia and horned puffin Fratercula corniculata, typi- factors that shape and constrain the life histories of this cally associated with arctic and subarctic marine envi- group of marine organisms. Such approaches could ronments, breed in very small numbers on Triangle also improve our ability to understand and predict Island (Rodway 1991). species-level and community-level responses to nat- Field sampling protocol. Sampling took place ural and anthropogenic climate change, which is between late March and late August in 2002, a year in having dramatic effects on marine trophic interactions which oceanographic conditions in the vicinity of Tri- (Edwards & Richardson 2004). angle Island, including both abiotic and biotic factors, We used SIA on blood samples taken from breeding were close to long-term averages (Mackas et al. 2007). adults of 5 species of seabirds belonging to the family Prey items: We collected prey items delivered to Alcidae, to track trophic level (based on δ15N measure- Cassin’s auklet and rhinoceros auklet nestlings at the ments) and habitat use (i.e. inshore vs. offshore food- colony, using standard methods (Hipfner et al. 2004). webs based on δ13C measurements) from the prelaying Provisioning Cassin’s auklets were intercepted in through the incubation and offspring-provisioning pheasant nets set up at the base of the breeding periods. In addition, we collected prey species deliv- colony, removed by hand and induced to regurgitate ered by adults to nestlings, including both zooplankton food loads into plastic bottles. From these samples, and fish, and used this conventional dietary informa- small subsamples for SIA were removed and frozen at tion, in combination with the isotopic values in nestling –10°C. Later examination (Hipfner 2008) revealed blood, to calculate discrimination factors appropriate to that individual samples were dominated (≥64% by our study site and study species. As we will show, this wet mass) by of 1 of 3 types of zooplankton: copepods proved to be necessary, since published discrimination (almost exclusively Neocalanus cristatus), euphausi- values did not result in realistic solutions to simple iso- ids (mainly Thysanoessa spinifera) and ichthyoplank- topic mixing models. ton (unidentified age 0+ larval and juvenile fish; here- Our main objective was to quantify the extent of after ‘juvenile fish’). We selected 5 regurgitant trophic and habitat shifting that these species exhibit samples dominated by each of these 3 prey types for from one breeding stage to the next. Secondarily, we SIA. also used intraspecific information to investigate sea- Provisioning rhinoceros auklets were caught on the sonal variation in trophic level and habitat use across ground with long-handled fishing nets, and the whole the community as a whole. Previous studies suggest prey items they carried (age 0+ to adult Pacific sand- that alcid communities tend to converge on fish-feed- lance Ammodytes hexapterus, post-smolt Pacific ing late in the season, based on nestling diets, and this salmon Oncorhynchus spp., age 0+ rockfish Sebastes has been used to make inference about the ecological spp., adult Pacific saury Cololabis saida, squid of factors that structure seabird communities (Ainley & unknown species) were collected. All prey items were Boekelheide 1990, Hatch & Hatch 1990). By bringing weighed, measured and identified in the field. Muscle together isotopic measurements of self-feeding and samples (~1 g) were cut from the mid-body, placed into Davies et al.: Alcid trophodynamics 213 vials and frozen at –10°C. Squid were processed in the of powdered material at the stable isotope facility same way, with muscle removed from the mantle. of the Department of Soil Science, University of Seabird blood samples: Blood (0.5 to 1.0 ml) was Saskatchewan.