Forest Investment Account (FIA) - Forest Science Program
FIA Project Y092238

    Reconstructing historic diets and population dynamics of the Marbled Murrelet
Project lead: Arcese, Peter (University of British Columbia)
Author: Arcese, Peter
Subject: Forest Investment Account (FIA), British Columbia
Series: Forest Investment Account (FIA) - Forest Science Program
The Marbled Murrelet (MAMU) nests in mature coastal forest but resides most of its life at sea. MAMUs are listed as Threatened in Canada (COSEWIC) and Red-listed in British Columbia. Causes of population decline are thought to include the loss and fragmentation of forest habitat and increased mortality related to over-fishing, climate change and other factors related to declines in the quality of habitat at sea. Of these causes, the influence of diet quality on MAMU populations is the least understood aspect of this species and represents a critical information gap.

The overarching goal of our research is to test if historic declines in the diet and marine habitat quality of MAMUs now limit population growth rate in this species in the Georgia Basin, and in less heavily impacted marine habitats west of Vancouver Island and the Queen Charlotte Sound. Our preliminary results for the Georgia Basin support this hypothesis. Confirmation of these preliminary results, and conducting parallel analyses in other regions of BC supporting the bulk of the Canadian MAMU population, represent critical next steps towards the identification of an optimal conservation strategy for this species in Canada.

Specifically, our research will facilitate decisions about the optimal allocation of conservation investment in the protection or restoration of 1) mature forest habitats in which this species nests and 2) marine environments in which this species spends most of its life. Because conservation efforts in support of MAMU now focus exclusively on the protection of high value mature forest habitat, our research has the potential to influence greatly sustainable forest management plans in coastal forest regions of British Columbia, and to provide critical advice to managers charged with the development of a reliable and defensible recovery plan for MAMUs in Canada. In the remainder of this introduction we review relevant results to date and our approach to test the hypothesis that diet and marine habitat quality currently limit population growth in this species.

Recent evidence from California suggests that MAMU breeding success and habitat use is influenced by marine habitat quality (Peery et al. 2004, Becker and Beissinger 2006, Becker et al. 2007). In California, up to 70% of adult MAMU failed to breed when diet quality in spring was low, suggesting that reproductive rate is limited by prey availability (Peery et al. 2004). Variation in prey type and abundance is also known to be related to climate, and to limit reproduction in seabirds related to MAMU in BC (Bertram et al. 2001). Our results based on the analysis of stable isotopes of nitrogen and carbon in MAMU feathers taken from museum specimens collected in the Georgia Basin of BC from 1890 to 1997 also suggest strongly that reductions in the biomass of forage fishes over the last 100 years have caused parallel declines in diet quality of MAMUs just prior to breeding and, furthermore, that these declines in diet quality limit the population growth rate of MAMUs in this region (Norris et al. 2007). If diet quality does limit reproductive rate in MAMUs in BC, recovery teams will need to consider re-focusing their recovery action plans on rebuilding fish stocks, in addition to securing sufficient nesting habitat in mature forest stands onshore.

The potential influence of ocean habitat on MAMU recovery is also emphasized in the most recent Conservation Assessment (Burger 2002), which identifies 'changes in prey' and 'over-fishing' as factors potentially limiting MAMU populations and in need of research. As in many other seabirds, high quality prey required for successful breeding in MAMU are those at higher trophic levels (i.e. Pacific sandlance, herring and anchovy). Although MAMU also consume prey at lower trophic levels, such as euphasiids, the fraction of prey at lower trophic levels in diet is related to poor breeding performance in California.

In the remaining year (2008/09) of our ongoing workplan, we will complete our second year of data collection to rigorously test how diet quality influences MAMU population dynamics using stable isotopes. To do so, we are now analyzing stable-nitrogen and carbon isotopes in the feathers and blood of 160 live MAMUs captured in ‘dip-nets’from April to June 2007 to identify the trophic feeding level of MAMU breeding in the Georgia Basin. Collaborators at Simon Fraser University are currently conducting assays of blood hormones collected from these same MAMU to diagnose reproductive state in females, sexed using genetic markers. Combining our isotope and hormone results will allow us to test directly for statistical links between trophic feeding level and breeding state.

Second, we have succeded in collecting an additional c. 200 historical feather samples from MAMU collected in coastal BC for the analysis of stable isotopes in feathers. These samples include specimens from Vancouver Island, Queen Charlotte Sound and the Georgia Basin and, as originally proposed, also include specimens collected by Captain Vancouver in 1778. These samples are currently being prepared for isotope analysis by Dr. Kurt Kaiser, Queens University. As originally proposed, we will contrast estimates of diet quality and prey composition from these locations to describe with greater precision diet declines through time, to compare these diets with those estimated from contemporary samples from each region.

Overall, we remain confident that our research will allow us to test definitively the hypothesis that MAMU populations are currently limited by diet quality in the Georgia Basin, and to also test if historic declines in diet quality already described in a preliminary fashion within the Georgia Basin have also occurred in the less impacted marine regions of west Vancouver Island and Queen Charlotte Sound. If at-sea feeding conditions limit MAMU population growth, we predict that females feeding at higher trophic levels in late winter to early spring will be in better physical condition at the beginning of the breeding season and more likely to breed. Second, we predict that long-term reconstruction of historic diet based on the analysis of feathers collected from museum specimens will be related positively to observed variation in population size as estimated from long-term surveys conducted since 1957 (Christmas Bird Count) and more recently via radar survey and incidental counts by ourselves and others since 1967 in the southern Georgia Basin. Overall, our work tests a critical management assumption that the conservation of nesting habitat in mature forest stands will be sufficient to affect species recovery, as well as the alternate hypothesis that MAMU populations are unlikely to grow where marine resources are severely depleted. These tests are critical to the development of a reliable and defensible recovery plan, and to the rational allocation of conservation efforts in support of species recovery planning
Related projects:  FSP_Y081238


Technical report (0.8Mb)

To view PDF documents you need Adobe Acrobat Reader, available free from the Adobe Web Site.

Updated August 16, 2010 

Search for other  FIA reports or other Ministry of Forests and Range publications.

Please direct questions or comments regarding publications to