Archived 2014 discussion: Kittlitz’s Murrelet (Brachyramphus brevirostris): is it eligible for downlisting?

This discussion was first published as part of the 2012 update, and remains open for comment in the hope that we will receive enough information to reassess the species in 2014.

BirdLife species factsheet for Kittlitz’s Murrlet

Kittlitz’s Murrelet Brachyramphus brevirostris is distributed in the Bering Sea, Gulf of Alaska and Chukchi Sea, and breeds in parts of Alaska, USA, and far eastern Russia (USFWS 2010a). The species is listed as Critically Endangered under criterion A4b,c,d,e on the basis that it was estimated and projected to be undergoing a decline of at least 80% over a period of 36 years (estimate of three generations) stretching from the past into the future (1986 – 2022).

Apparently large declines in the species’s abundance since the late 1980s and early 1990s have been reported in Alaska. Surveys in much of the Alaskan range have indicated that populations may have declined by 80-90% in c.15 years since the early 1990s (Piatt et al. undated). In Prince William Sound, the population declined by 84% between 1989 (6,436 birds) and 2000 (1,033 birds) following a possible longer-term decline since 1972 when the population was estimated at 63,000 individuals (K. Kuletz in litt. 2002, Kuletz  et al. 2003). The rate of decline in Prince William Sound between 1989 and 2004 has been estimated at 63% (5% per year) (Kuletz et al. 2011b). In the Malaspina Forelands, numbers declined by 38-75% between 1992 and 2002. Total abundance in Icy Bay, Alaska, was estimated to be 1,725-2,372 birds in 2002, suggesting a decline of 59% over a three-year period (Kissling et al. 2006).

Some more recently published studies also report local declines in this species. An analysis of survey data from the Kenai Fjords, Alaska, suggests that a 90% decline occurred between 1989 and 2002; however, this analysis is hampered by changing survey methods over time, few years of survey effort and low population numbers (Arimitsu et al. 2011). In the Lower Cook Inlet, survey data suggest that the breeding population declined by 84% (26% per year) between 1993 and 1999 (Kuletz et al. 2011a). Anecdotally, the apparent recent absence of the species from LeConte Bay may indicate a change in its range in south-eastern Alaska, where the species’s abundance and distribution appears to be influenced by changes in icefields, raising concerns over its future in the region (Kissling et al. 2011).

In contrast, Day et al. (2011) assert that there is a lack of strong evidence for a significant population change in the breeding and non-breeding populations in northern Alaska between 1970-1999 and 2000-2009. Likewise, anecdotal evidence implies that numbers have been stable around Attu Island over recent decades (Madison et al. 2011). Increases have been recorded in other areas, such as Unalaska and Atka islands (Romano et al. 2005a,b, in Day 2011).

Somewhat conflicting reports have emerged from Glacier Bay, where density estimates declined by 89.1% between 1991 and 2000, with c.2,200 birds estimated there in 1999-2000 (K. Kuletz in litt. 2002). Further surveys there suggest a decline of 85-90% between 1991 and 2008, with an annual rate of decline of between -10.7% and -14.4% (Piatt et al. 2011). However, Kirchhoff et al. (2010) recorded an increase in the on-water density of the species in Glacier Bay between 1993 and 2009, from 2.07 birds/km2 to 3.55 birds/km2, although this change is not significant. This apparent stability in the population suggests that any decline had occurred before 1993, most likely in the winter of 1991/1992 (Kirchhoff et al. 2010). Similarly, Hoekman et al. (2011b) recorded an apparent increase in abundance from numbers recorded in 1999-2007 compared with numbers in 2009-2010, although earlier surveys could have suffered from negative biases. Overall, the various studies conducted in Glacier Bay are unlikely to be comparable owing to differences in survey methods and coverage (Kirchhoff et al. 2010, Hoekman et al. 2011b).

According to the US Fish and Wildlife Service, data from Agattu and Kodiak islands indicate that nest success is very low (less than 10%), with few juvenile birds having been documented there, and blood chemistry analysis at Icy Bay indicates that few females there (c.10%) are breeding despite the majority (c.90%) being physiologically prepared to breed (USFWS 2011). It is also suspected that adult mortality is slightly elevated by losses to hydrocarbon contamination, entanglement in gillnets and high predation pressure (USFWS 2011).

There are no trend data from other parts of the species’s range, but anecdotal information suggested population declines were occurring in at least some of these other areas. A population of 500 breeding pairs has been estimated to be present in the Sea of Okhotsk, and it is thought that the Asian coast supports a significant proportion of the global population (Artukhin et al. 2011).

Recently, the reliability of earlier studies that have documented apparent declines has been called into question, and inconsistencies in survey and analytic protocols have cast uncertainty over the species’s population status and trend (Day 2011, Hoekman et al. 2011a). The accuracy of counts conducted in the late 1980s and early 1990s has been questioned owing to problems with study designs and analyses (D. Vincent-Lang in litt. 2010, Day 2011). Such results have been deemed unsuitable for use as baseline data from which to calculate trends (Day 2011). For example, surveys conducted by Piatt et al. (undated) in the early 1990s predominantly sampled near-shore waters and some offshore areas with high densities of Kittlitz’s Murrelet, resulting in inflated baseline data (Day 2011). For many studies that report significant declines, the comparison of data sets from surveys employing disparate methodologies has been strongly criticised (Day 2011).

In recent years, more rigorous, species-specific survey methods have been introduced (D. Vincent-Lang in litt. 2010, Day 2011), and studies relevant to this species have been helped by advances in technology and software, as well as statistical awareness (Day 2011). Population counts in some areas during the past decade have indicated stable or increasing trends in almost every year surveyed, and declining trends that were significant in the past are no longer significant (D. Vincent-Lang in litt. 2010). In addition to indications that the species may not be declining as rapidly as previously thought, larger than expected populations have been located in areas such as the Aleutian Islands, Kodiak and Russia, resulting in a recent upward revision of the global population estimate (D. Vincent-Lang in litt. 2010). The total population has recently been estimated at 30,900-56,800 individuals (USFWS 2010b).

Although Day (2011) criticises many of the analyses that have indicated very rapid declines in this species, he indicates that smaller declines in parts of or the whole of its range may still have taken place, and that shifts in the species’s geographic preferences could also account for local trends. Localised surveys for this species can probably never be said to be capturing data on closed populations, and variation in numbers resulting from movements of birds can occur over a range of temporal scales (e.g. Kirchhoff et al. 2010). Local abundance estimates may also be affected by anomalous oceanographic conditions (Arimitsu et al. 2010). Many potential threats have been linked with previously detected declines in this species; however, there is apparently little empirical evidence to support such links (D. Vincent-Lang in litt. 2010). It has also been pointed out that other marine species that are not undergoing rapid declines are exposed to the same threats of climate change, variation in food availability, oil-spills, fisheries bycatch, shipping and tourism (D. Vincent-Lang in litt. 2011).

With this information in mind, it is proposed that the species’s status be re-evaluated. However, there is a lack of consensus regarding the species’s current population trend, with views that the species may be declining more slowly than previously thought, or that it could be stable or increasing. Further information is requested on the likely trend over a period of three generations (c.36 years) in the past, future or both. A decline of 50-79% over this period would qualify the species for downlisting to Endangered, while a decline of 30-49% would make it eligible for downlisting to Vulnerable. A rate of decline approaching 30% (typically 20-29%) would likely warrant its downlisting to Near Threatened, and typically a decline of less than 20% over three generations would suggest the species should be downlisted to Least Concern. Such a change in category would take place in the 2012 update if it were shown to be based on an improvement in knowledge, rather than a real improvement in the species’s status. If a real improvement has taken place, it needs to be shown that this has been the case for at least five years.

References:

Arimitsu, M. L., Piatt, J. F., Romano, M. D., Madison, E. N. and Conaway, J. S. (2010) Kittlitz’s and Marbled Murrelets in Kenai Fjords National Park, South-Central Alaska: At-Sea Distribution, Abundance, and Foraging Habitat, 2006-08. USGS Numbered Series: 2010-1181. US Geological Survey.

Arimitsu, M., Piatt, J., Romano, M. and van Pelt, T. I. (2011) Status and distribution of the Kittlitz’s Murrelet Brachyramphus brevirostris in Kenai Fjords, Alaska. Marine Ornithology 39: 13-22.

Artukhin, Y. B., Vyatkin, P. S., Andreev, A. V., Konyukhov, N. B. and Van Pelt, T. I. (2011) Status of the Kittlitz’s Murrelet Brachyramphus brevirostris in Russia. Marine Ornithology 39: 23-33.

Day, R. H. (2011) Evaluating Population Trends of Kittlitz’s Murrelets in Alaska. Report for Alaska Department of Fish and Game, Division of Wildlife Conservation, prepared by ABR, Inc. – Environmental Research & Services, Fairbanks, AK.

Day, R. H., Gall, A. E., Prichard, A. K., Divoky, G. J. and Rojek, N. A. (2011) The status and distribution of Kittlitz’s Murrelet Brachyramphus brevirostris in northern Alaska. Marine Ornithology 39: 53-63.

Hoekman, S. T., Moynahan, B. J. and Lindberg, M. S. (2011a) Monitoring Kittlitz’s and marbled murrelets in Glacier Bay National Park: 2009 annual report. Natural Resource Technical Report NPS/SEAN/NRTR—2011/440. Fort Collins, CO: National Park Service.

Hoekman, S. T., Moynahan, B. J. and Lindberg, M. S. (2011b) Monitoring Kittlitz’s and marbled murrelets in Glacier Bay National Park: 2010 annual report. Natural Resource Technical Report NPS/SEAN/NRTR—2011/441. Fort Collins, CO: National Park Service.

Kirchhoff, M., Smith, M. and Wright, S. (2010) Abundance, population trend, and distribution of Marbled Murrelets and Kittlitz’s Murrelets in Glacier Bay National Park. Final Report to the National Park Service and Alaska Department of Fish and Game. Anchorage, AK: Audubon Alaska.

Kissling, M., Lukacs, P. M., Lewis, S. B., Gende, S. M., Kuletz, K. J., Hatch, N. R., Schoen, S. K. and Oehlers, S. (2011) Distribution and abundance of the Kittlitz’s Murrelet Brachyramphus brevirostris in selected areas of south-eastern Alaska. Marine Ornithology 39: 3-11.

Kissling, M. L., Reid, M., Lukacs, P., Gende. S. M. and Lewis, S. B. (2006) Temporal and spatial variability of Kittlitz’s Murrelets in Icy Bay, Alaska. Pp. 175 in Wings without borders: IV North American Ornithological Conference, October 3-7, 2006, Veracruz, Mexico. Waco, TX, USA: American Ornithologists’ Union.

Kuletz, K. J., Nations, C. S., Manly, B., Allyn, A., Irons, D. B. and McKnight, A. (2011) Distribution, abundance, and population trends of the Kittlitz’s Murrelet Brachyramphus brevirostris in Prince William Sound, Alaska. Marine Ornithology 39: 97-109.

Kuletz, K. J., Speckman, S. G., Piatt, J. F. and Labunski, E. A. (2011) Distribution, population status and trends of Kittlitz’s Murrelet Brachyramphus brevirostris in Lower Cook Inlet and Kachemak Bay, Alaska. Marine Ornithology 39: 85-95.

Kuletz, K. J., Stephensen, S. W., Irons, D. B., Labunski, E. A. and Brenneman, K. M. (2003) Changes in distribution and abundance of Kittlitz’s Murrelets Brachyramphus brevirostris relative to glacial recession in Prince William Sound, Alaska. Mar. Ornithol. 31: 133-140.

Madison, E. N., Piatt, J. F., Arimitsu, M. L., Romano, M. D., Van Pelt, T. I., Nelson, S. K., Williams, J. C. and Degange, A. R. (2011) Status and distribution of the Kittlitz’s Murrelet Brachyramphus brevirostris along the Alaska Peninsula and Kodiak and Aleutian islands, Alaska. Marine Ornithology 39: 111-122.

Piatt, J. F., Arimitsu, M., Drew, G., Madison, E. N.M., Bodkin, J. and Romano, M. D. (2011) Status and trend of the Kittlitz’s Murrelet Brachyramphus brevirostris in Glacier Bay, Alaska. Marine Ornithology 39: 65-75.

Piatt, J., Climo, L., Springer, A. and Duncan, T. (undated) Marine bird distribution in Glacier Bay National Park during the summer of 1991. Unpublished report prepared by U.S. Fish and Wildlife Service, Anchorage, AK.

USFWS (2010a) Species Assessment and Listing Priority Assignment Form, for Brachyramphus brevirostris (Kittlitz’s Murrelet). U.S. Fish and Wildlife Service.

USFWS (2010b) Endangered and Threatened Wildlife and Plants; Review of Native Species That Are Candidates for Listing as Endangered or Threatened; Annual Notice of Findings on Resubmitted Petitions; Annual Description of Progress on Listing Actions. Federal Register Vol. 75, No. 217.

USFWS (2011) Endangered and Threatened Wildlife and Plants; Review of Native Species That Are Candidates for Listing as Endangered or Threatened; Annual Notice of Findings on Resubmitted Petitions; Annual Description of Progress on Listing Actions. Federal Register Vol. 76, No. 207.

Vincent-Lang, D. in litt. 2010

Vincent-Lang, D. in litt. 2011

The following document was sent by Jack Hodges on 23 January 2012 in reaction to this discussion topic: Brachyramphus brevirostris Hodges Jan2012

The following letter was sent by Kathy Kuletz on behalf of herself, John Piatt and David Irons, on 31 January 2012, in response to the letter by Jack Hodges: Brachyramphus brevirostris Kuletz Jan12

The following letter from Jack Hodges was requested to be posted on 4 December 2012, in response to the letter by Kuletz et al. posted on 31 January 2012: Hodges Response to Kuletz Posting

Population trend spreadsheet from Matt Kirchhoff, received on 9 February 2014: Population Trends – Post 1998 Kirchhoff Feb14

This entry was posted in Americas, Archive, Asia, North America, Seabirds and tagged . Bookmark the permalink.

19 Responses to Archived 2014 discussion: Kittlitz’s Murrelet (Brachyramphus brevirostris): is it eligible for downlisting?

  1. The proposed new listing for KIMU will be under the A4 criteria, which looks both back and ahead. Joe has suggested 2 generations (24 years) back, and 1 generation (12 years) ahead. For this assessment, I suggest we model into the future based on data gathered in the last 15 years in each area. This would capitalize on our better surveys (more birds identified, intensive effort in KIMU fjords, offshore stratum sampled) and reflect our best estimate of current population trends.

    • Kathy Kuletz says:

      Matt’s suggestion of only using data from the last 15 years would be practical, and might be best for looking towards the future. However, since the major decline appears to have occurred prior to the early 2000s, it effectively leaves us with a shifting baseline. For some purposes, I suggest going back to at least 1989. At least two of the large scale surveys (Prince William Sound and Cook Inlet) included offshore areas in their study design, and despite variable identification rates provide, at minimum, numbers of identified birds and total Brachyramphus trends.

      • I agree. We should go back to at least 1986 for historical perspective. Forecasting trend into the future probably needs a more recent time frame. After looking at the suggested 1995-2010 time frame for a number of survey areas, perhaps the only place we have sufficient data for forward-forecasting is in Glacier Bay and Prince William Sound.

        • A clarification: good forward-forecasting should involve more than extrapolating trend lines. There are other survey areas where we not only have excellent surveys, but we know something about movements, recapture rates, nest success, adult survival, causes of mortality, forage fish and plankton data. All of this helps paint a more complete picture of not only what is happening, but why. Those are the most interesting questions, and the answers researchers come up with will tell us a lot about where this species is heading. All study areas contribute to the big picture.

    • John Piatt says:

      Many of the processes believed to influence marine ecosystems and food webs in Alaska operate at multi-decadal scales, with duration of positive or negative phases sometimes lasting as long as 20-30 years (like the most recent regime). In evaluating the population trends of any seabird species in Alaska it would therefore be prudent to consider population changes integrated over the same time periods. Thus 24 years would be a reasonable minimum time frame.

  2. Estimating percent decline

    How should we characterize the percent decline in KIMU? One option is to report the difference between the highest estimate in the time series and the lowest estimate within a long time series, as some papers do. This method assumes 100% of the difference between two survey points is attributable to population change, when in fact, a significant percentage of the difference is often due to sample variance or process variance (both typically high in our KIMU surveys). I suggest a better method is to fit a regression model through the survey data. This isolates the population trend from the variance, or noise, and the associated regression statistics will convey the confidence we have in that decline.

  3. Weighting percent change

    We have survey data in numerous population centers or areas around the state. If populations are increasing in 2, stable in 5, and decreasing in 4, how do we integrate that information into an assessment of the global population trend for the species? I suggest we weight the percentage change in each survey area by the population in that area, so that a 10% increase in a small population center is properly balanced with a 10% decrease in a very large population center. For weighting purposes, the average population over the survey record may be most appropriate. In estimating population size, not all authors include unidentified birds prorated to species. Given the large influence this can have on abundance estimates when identification rate is low, I suggest unidentified birds be prorated to species for these abundance estimates.

    • John Piatt says:

      Weighting estimates of total population change by population size seems reasonable as long as we have good estimates of both population size and trend in each area. Seems like you would also want to weight trends by variance of the estimates. Not sure how a 10% increase in a small population would balance a 10% decrease in a large population (probably not really what you meant?). Prorating unidentified birds is reasonable only where evidence suggest a credible effort has been made to identify species. How do you evaluate credibility of data where observers claim to identify all birds to species?

      • In my view, the best surveys come from crews that are specifically-trained and highly motivated. Surveys with a particular focus on murrellets, or that allow vessels to slow or stop for accurate identification, should generate reasonably accurate results. Survey experience is also important. I’d trust your results identifying 100% of the murrelets over a novice observer’s identifying 70%.

  4. John Piatt says:

    Mr. Joe Taylor
    Global Species Program
    Birdlife International
    Wellbrook Court
    Girton Road
    Cambridge, CB3 0NA
    United Kingdom,

    January 31, 2012
    Dear Mr. Taylor,

    I have read your online forum material for Kittlitz’s Murrelet, and was disappointed to see so much unpublished information and unsubstantiated opinions being used to shape conclusions about the status of Kittlitz’s Murrelet populations in Alaska. While some folks would steer you towards old reports and old controversies, under the banner of a “new analysis” of data by Dr. Bob Day, I would encourage you to rely instead on published material that has undergone extensive peer review by statistical and murrelet experts, wherein many of the issues raised by Day and other critics have been either accommodated in the analyses or rejected as inapplicable.

    Excuse me if I am overly sensitive, but the wording of your forum piece seems contrived to persuade opinion. Declines that have been established by conducting 10,000’s of km of surveys (using methods similar to those adopted by Tasker et al. many years ago to census seabirds in the North Sea) are variously described as “apparent”, “possible”, “suffering from negative biases”, “unlikely to be comparable”, and otherwise shrouded in uncertainty by “conflicting” reports. Analyses have been “hampered” by changing survey methods, the reliability of earlier studies “has been called into question”, and the accuracy of earlier counts “has been questioned”.

    In contrast, from those who focus on the uncertainties in data, you provide more confident pronouncements: “Such results have been deemed unsuitable for use as baseline data” by Dr. Bob Day. Recent studies are much better: “more rigorous…. helped by advances in technology and software, as well as statistical awareness” (Day 2011). “Declining trends that were significant in the past are no longer significant” (D. Vincent-Lang 2010). Really? I must be lacking the right software because I sure don’t share their confidence or conclusions.

    “With this information in mind, it is proposed that the species status be re-evaluated. However, there is a lack of consensus regarding the species current population trend.”

    There is no other way to say this: Poppycock!

    How is it possible that the field studies and data summarized by 41 knowledgeable scientists in 11 articles recently published in Marine Ornithology, which were peer-reviewed by 22 leading experts in Brachyramphus murrelet biology and/or statistics, and vetted by 4 different editors, can be construed to be questionable science and lacking in consensus? How can it be trumped by the unsubstantiated, unpublished and un-reviewed conclusions of a few individuals? Is it because they “assert” while we just “suggest”?

    I don’t have enough time, and this is not the right forum, to dispute Bob Day’s report line-by-line. But since this commissioned Alaska Department of Fish and Game (ADF&G) report was never published, and yet has been widely distributed as if it were authoritative, I am left with little choice but to comment on it here. Contrary to its title, this report does not evaluate population trends of Kittlitz’s Murrelets. There is not a single analysis of data here, or test of data bias, or statistical comparison of data, or statistical evaluation of trend. Its rather strong conclusions are based entirely on Day’s personal interpretation of other peoples’ work, and he didn’t ask those folks to review his conclusions before he released his report (and so you can see where the lack of consensus truly lies). Several principal investigators have written letters of rebuttal (if interested, you might ask Day or ADF&G to share that correspondence with you). I am one of those investigators, and I will simply summarize here by saying that many of Dr. Day’s conclusions about baseline data are based on flawed information. I found many errors of fact with regard to surveys with which I have been personally involved, including those from Glacier Bay, Kenai Fjords, Kachemak Bay, Alaska Peninsula, and several Aleutian Islands. Indeed, conspicuous errors regarding survey protocols and misleading characterizations of data or its interpretation are found throughout the report. Other investigators raised similar concerns with Day; to my knowledge all remain unresolved.
    With regard to the Glacier Bay “controversy” generated by Day, and perpetuated by you, I would simply say: 1) Day’s assertion that my baseline estimate of murrelets in Glacier Bay was inflated is absurd. For one thing, the “undated” report you keep referring to was a one page summary of what we did and saw on the 1991 survey; there was no population estimate. Surely a “statistically aware” investigator would be quoting our recent paper (Piatt et al. 2011) where we compiled survey data collected in seven different years and analyzed coastal and offshore strata separately? 2) Surveys conducted by me from 1991 to 2008, using the same survey protocols, in the same habitats, show an 85% decline in populations. We could spend a lifetime discussing why surveys conducted in recent years by other investigators, who surveyed at different times of summer, and used different protocols for data collection and analysis, yielded different population estimates. This should not cloud the fact that the surveys I conducted over 20 years revealed a major decline of murrelets in both coastal and offshore habitats (and, interestingly, no change in abundance of most other avian taxa).
    In summary, published information shows that Kittlitz’s Murrelet populations have declined dramatically in core areas of Glacier Bay (85%), Prince William Sound (>63%) and lower Cook Inlet (84%), and declines have leveled off in recent years. Preliminary results of recent land-based studies of demography in three different areas of the Gulf of Alaska and Aleutians indicate that 1) breeding success is very low (at best, 0.17 to .37 chicks/pair), and, 2) adult survival is very low (0.84 during breeding season) [see three abstracts by Kissling, Kaler and Lawonn]. These independent studies all paint a pessimistic picture for Kittlitz’s Murrelet. But I don’t know how these results would affect your listing status. Maybe they should still be down-listed, maybe not. Whatever you do, I would recommend it be based on verifiable information.

    John F. Piatt, Ph.D.
    Research Biologist
    USGS Alaska Science Center
    Anchorage, Alaska

  5. Jack Hodges says:

    This is in response to the Kuletz January 31, 2012 response to the Hodges January 23 posting.

    In Prince William Sound, observers started to consistently identify more than 70% of the murrelets to species in 1996. From 1996 to 2007, the MAMU population estimates declined by roughly 50%. The proportion of murrelets which were KIMU did not remain constant; it increased. For 1996-1998, one in 77 murrelets were KIMU. For 2000-2004, one in 50 murrelets were KIMU. For 2005-2007, one in 13 murrelets were KIMU.

    Kuletz mischaracterized my post when she stated, “…just because we have some anomalous results in 1993, doesn’t mean that all the other years of data are invalidated; it does mean that our analyses need to consider these aspects of the data.” We agree that 1993 was anomalous. My post did not imply that all of the other years are invalid. My post indicated that there is evidence of misidentification problems in one other data point, 1989. Accepting all remaining data points from 1989 to present, there is no basis for reporting a declining population.

  6. Joe Taylor says:

    A letter from Jack Hodges has now been posted below the reference list for this forum topic.

  7. Andrew Allyn has published a very good thesis on KIMU population size, habitat use, and diet in Prince William Sound.

    It is available here:
    http://scholarworks.umass.edu/cgi/viewcontent.cgi?article=1877&context=theses.

    The 2009 Kittlitz’s Murrelet population estimate was 2079 birds (95% CI = 1409 – 2990), reflecting a 48% increase over the 2001 Kittlitz’s Murrelet population estimate of 1400 birds (95% CI = 977 – 1889). The increase was statistically significant.

    The historic population assessments for PWS are notable for how much they vary, with reported increases between some years impossible based on vital rates alone (Kirchhoff 2011- Marine Ornithology). The low survey effort placed in important KIMU fjords likely contributes to these extreme differences.

    Kathy Kuletz rectified this in 2001 with a new survey design expressly for KIMU. Allyn’s 2009 survey replicated this new survey design. It is a strong comparison, albeit based on just 2 years, and should be added to our discussion of the correct conservation status of KIMU.

    Matt

  8. Matt Kirchhoff says:

    I believe previously published estimates of global population size for KIMU can be expressed more accurately given new surveys and new survey techniques. Early published estimates were derived by summing survey results in selected study areas within the global range. First estimates by this method placed the global population in the 9 – 25 thousand range. Newer Fish and Wildlife estimates, that cover more areas, more recent surveys, and a few line-transect surveys, put the global population in the 30-50 thousand range.

    There are three reasons the true global population is likely higher than previously estimated: (1) surveyed areas include most, but not all of the waters occupied by KIMU in summer (e.g., non-breeders offshore, portions of coastal areas with low densities) (2) surveys which extrapolate from strip-transect densities do not account for undetected birds, and (3) summer surveys do not account for birds that are on nests (except for Russia, and they may overestimate this number).

    As a starting point, I suggest (1) increasing strip-transect-based estimates by 33% to account for undetected birds (Ronconi and Burger 2009 suggest 33-50% for murrelets; Piatt et al. 2011 indicate about 33%) (2) increase population estimates by 25% to account for un-surveyed but occupied waters (assuming this is the source of the anomalous large influxes– nonbreeders?–flooding into some areas in some years), and (3) increase by another 5 percent to account for birds not on the water (assuming 10% of adults are actively breeding/nesting).

    By my math, if we start with 30-50 thousand, and make these corrections, the global lower estimate becomes 30 +(30*0.33) + (30*0.25) + (30*0.05), or 48,900. The global upper estimate becomes 50 + (50*0.33) + (50*0.25) + (50*0.05), or 81,500. Rounding, the global population size falls in the ballpark of 50-80 thousand. That makes the 18,000 counted in all of Glacier Bay in July 2012 about 22 – 36 of the global population. A more average year in Glacier Bay might reflect 7 – 12%. Those numbers seem plausible to me.

    Although I share the discomfort associated with estimating “impossible” things, like global population size, I’d say we are in a better position to do this with KIMU than most bird species. Let’s collectively make the most accurate estimate we can, and allow the range reflect our uncertainty. Getting our “best” estimate out on the BLI and IUCN websites will go a long way to cleaning up the older, often conflicting, information carried on other websites.
    Matt

  9. Matt Kirchhoff says:

    Thoughts on KIMU trend time-frame

    The current conservation status of Kittlitz’s Murrelet hinges on its population trend. The IUCN Redbook states the relevant time frame is 10 years or 3 generations prior. If we assume a 12 year generation time, that backs us up 36 years, to the early 1970’s. There is some literature that is critical of a 3-generation backward look for long-lived species because (1) historical surveys tend to be less rigorous, and (2) historic trends have less bearing on current extinction risk if the species is presently abundant, widespread, and stable.

    Joe Taylor suggested we look back two generations, and project forward 1 generation (e.g., 1990-2026). That seems more appropriate because it puts greater emphasis on what is likely to happen given current population trajectory.

    A current or forward-looking emphasis has the further advantage of incorporating surveys in additional areas (e.g., Yakutat, Icy Bay) and surveys designed with murrelets more in mind. Those include modern surveys in Icy Bay (2002-2013), Prince William Sound (2001 and 2009 intensive), Glacier Bay (1993, 2009-2013), Yakutat (2000, 2009), Malaspina Forelands (2002-2009), Kachemak Bay (2005-2007), Kenai Fjords (2003, 2006 – 2008), and Attu Island (2003, 2009).

    Although the number of surveys post 2000 in some areas are few, they are relatively good surveys (better designs, better training, better methods) that have met with wide agreement. It’s not perfect, and we don’t want to ignore the possibility of large declines in some areas in the more distant past. But working with post-2000 data will provide less arguable and still valuable trend information.

    As suggested in previous posts, an analysis that weights results by the percent of population in each area may give the best sense of how the population is trending at the global scale.

    As with global population size, estimating global trend is a challenging exercise. However, we have learned much since 2004 when the species was petitioned for listing. I think we can offer a new synthesis that allows BLI and IUCN to determine the current conservation status of this species with best scientific information.
    Matt

  10. Matt Kirchhoff says:

    Subject: Recent KIMU trends

    After drafting the last comment, I decided to take a look at what the recent survey data show. I gathered all the published population estimates from all survey areas that report 2 or more population estimates in the last 15 years (post 1998), and put them in a spreadsheet (which I’ve asked Joe Taylor to make available). I have also graphed the results by area, and included regression lines and equations. If anyone has more recent unpublished data not in the spreadsheet, I’d encourage you to add it and update the graphs (I realize some may wish to reserve for publication).

    I think it’s safe to conclude from this that there is no compelling evidence of decline in the KIMU population over the last 15 years. One can argue that we don’t have enough data to make a positive statement about what IS going on in areas with just 2 or 3 survey points, and that’s true. But we also can’t conclude they are declining. The power in putting ALL of the data together like this is that if the population was still declining steeply, and broadly, there should be a signal. I don’t see it.

    This doesn’t speak to the valid issue of what might have happened with KIMU in earlier decades, and whether the surveys and papers that comment on those trends are good or bad (we all have our views). I think that ground has been plowed in the literature, and there’s not a lot to add by blogging about it. But I do hope we can agree these recent surveys are better than the old, with high identification rates, good crew training, and better designs (e.g., crossing transects, increased effort in KIMU fjords, line transects). It’s worth looking at these apart.

    I think all are familiar with the papers cited, but I am happy to provide full citations and PDFs if not. Numbers come from the papers directly, but if I have misstated something, please correct. Note, too, that with the exception of Glacier Bay, which has far more birds than any other survey area, the y axis on the graphs are identical. This allows one to visually see the relative “weight” each area might get in a global meta-analysis.

    I have made a rough stab at estimating general abundance of KIMU in each survey area in the last column of the spread sheet, with a lean towards the results observed in more recent surveys (I do not make the corrections recommended in earlier post, but some abundance estimates already adjust for detection).

    Those “current abundance” estimates are approximations. If you work in one of those areas please revise to what you think is more accurate or representative. Of course there are other areas where we have no recent surveys, or only a single survey after 1998 (Aleutians, northern Alaska, places along the lost coast), or where investigators have multiple surveys but did not report abundance figures (Russia, Kodiak). Since this is a comment on trends, those areas were not included.

    That’s plenty to chew on. I hope this information adds to the discussion.

    Matt

  11. Joe Taylor says:

    The spreadsheet sent by Matt Kirchhoff and mentioned in his last forum post can be downloaded via a link below the list of references in the forum topic.

  12. Andy Symes says:

    Preliminary proposals

    Based on available information and comments posted above, our preliminary proposal for the 2014 Red List would be to treat Kittlitz’s Murrelet Brachyramphus brevirostris as Near Threatened, approaching the threshold for Vulnerable under criterion A4.

    There is now a period for further comments until the final deadline of 31 March, after which recommended categorisations will be put forward to IUCN.

    The final Red List categories will be published on the BirdLife and IUCN websites in mid-2014, following further checking of information relevant to the assessments by both BirdLife and IUCN.

  13. Andy Symes says:

    Recommended categorisation to be put forward to IUCN

    Following further review, there has been no change to our preliminary proposal for the 2014 Red List status of this species.

    The final categorisation will be published later in 2014, following further checking of information relevant to the assessment by BirdLife and IUCN.

Comments are closed.