Archived 2015 topics: Yelkouan Shearwater (Puffinus yelkouan) – downlist from Vulnerable to Least Concern?

Yelkouan Shearwater Puffinus yelkouan is endemic to the Mediterranean and Black Seas, breeding from S France and E Algeria east to Greece and Bulgaria; breeding is also suspected in Turkey, but still not proven (Carboneras et al. 2014). It is currently listed as Vulnerable, because when last assessed it was considered to be undergoing a rapid population decline.

Globally, it has an extremely large range (c. 3 million km2), and hence does not approach the thresholds for Vulnerable under the range size criteria (B and D2). Its population size is also moderately large (with 39,000–62,000 mature individuals in Europe alone; BirdLife International 2015), and hence does not approach the thresholds for Vulnerable under the population size criteria (C and D1). Therefore, the only potentially relevant criterion is A, which relates to reductions in population size. Until recently, the population was thought to be declining rapidly at a rate which, if continued over three generations (54 years, based on a generation length estimated by BirdLife to be 18 years), may have resulted in an overall population decline of >30% (the threshold for listing as Vulnerable under criterion A).

New data collated from across Europe for the European Red List of Birds (BirdLife International 2015) suggest that the species is no longer declining. A combination of official data reported by 27 EU Member States to the European Commission under Article 12 of the EU Birds Directive and comparable data from other European countries, provided by BirdLife Partners and other leading national ornithologists, suggest that the European breeding population is now increasing overall, with stable or increasing trends in those countries with the largest populations (Italy, Greece and Malta, which together hold c. 95% of the European breeding population). Consequently, the species is now classified as Least Concern at European level (BirdLife International 2015).

Europe holds >95% of the global breeding population and range, with the remainder in NW Africa, so the species’ status in Europe effectively determines its global status. Despite the threats perceived to be facing the species when last assessed (see current global factsheet for details), its population has not continued to decline overall. Furthermore, the increasing numbers reported from the breeding colonies correspond well with those counted moving through the Bosphorus in February (the non-breeding season), which have risen from 73,000 in 2012 to 90,000 in 2014 (D. Sahin in litt. 2014). These increases may in part reflect recent investments in conservation projects to improve the species’ prospects, e.g. in Italy and Malta.

As the species is no longer declining, and seems unlikely to decline sufficiently rapidly in the near future to be listed as Near Threatened, it should be reclassified as globally Least Concern. Comments on this proposal are welcome.


BirdLife International (2015) European Red List of Birds. Luxembourg: Office for Official Publications of the European Communities.

Carboneras, C., Jutglar, F. & Kirwan, G.M. (2014). Yelkouan Shearwater (Puffinus yelkouan). In: del Hoyo, J., Elliott, A., Sargatal, J., Christie, D.A. & de Juana, E. (eds.) (2014). Handbook of the Birds of the World Alive. Lynx Edicions, Barcelona.

This entry was posted in Africa, Archive, Europe & Central Asia, Middle East, Seabirds and tagged . Bookmark the permalink.

21 Responses to Archived 2015 topics: Yelkouan Shearwater (Puffinus yelkouan) – downlist from Vulnerable to Least Concern?

  1. Andy Symes says:

    Comments already received, in relation to the Italian population reported under Article 12:

    Nicola Baccetti, Istituto Superiore per la Protezione e la Ricerca Ambientale (ISPRA) (8/6/15):

    The paper which re-assessed the single largest colony of Tavolara Archipelago (cf Medmaravis Alghero Proceedings, see here) has probably been misinterpreted: the trend there (and nationally, by consequence) remains negative, for unchanged reasons respect to Bourgeois &Vidal’s previous paper)

    Lorenzo Serra Istituto Superiore per la Protezione e la Ricerca Ambientale (ISPRA) (12/6/15):
    ‘’The population estimates published in the Italian report originated from the collection of available data and from expert opinion, in a general framework of poor information and very weak data, despite the efforts invested on the species over the last 20 years. Both methods and quality of the data were therefore scored at the lowest rank.

    However, there are no mistakes or misinterpretation of relevant references in the report. Available information shows that the species is increasing on the short term as all estimates of the largest colony (Tavolara or the Tavolara Archipelago) are increasing since 2002. The population of this colony gives the sign of the national trend. We can speculate on the reliability of these estimates but we have to consider that 3 out 4 estimates from this period have been proposed by the same team of experts and there is no qualified reason to reject them as the authors themselves made reference to those numbers even in their last published paper (2012). Numbers are fluctuating, as stressed by the authors, but constantly increasing. Furthermore, the consistency of their last estimate based on counts of rafting individuals is confirmed by an estimate based on nest density provided by the same authors. If we assume this last estimate (9991-13424 pairs in 2008, Baccetti et al. 2012) as reliable, a negative short term trend should result in biases of the previous estimates that are really difficult to accept. In a hypothesis of stability, the errors of previous estimates are listed below (the year of publications is indicated because sometimes the year of the estimate is not clear (at least to me now) or the estimate is based on data from different years).

    Year of publication 2002 (estimate 1000-2000 pairs, Brunner et al. 2002): error 85-90%
    Year of publication 2006 (estimate 500-5000 pairs, Zenatello et al. 2006): error 63-95%
    Year of publication 2009 (estimate 1200-7800 pairs, Baccetti et al. 2009): error 42-88%

    In a hypothesis of a slight decrease, let assume it as 10% in the last 10 years, the size of the error becomes really puzzling and put some concern on which information conservation policies for the species are built on.

    In the text we discussed the possibility that the Italian population might have decreased on the long term trend as we took in consideration comments received by the same team of expert cited above, who asked us to give some notes on the hypothesis that population numbers from the ‘80ies were not directly comparable with present estimates due to the development of new count methods (even if one might object on this point, as old and present estimates originate from observations of individuals in proximity of the breeding colony and there is really not that great technical advancement in visual observations).

    An indirect indication of a possible population decrease comes from the observed abandonment of some historical breeding sites in Sardinia and in the Tuscan Archipelago. However, even for these observations, there is a great uncertainty as there is evidence of this only for the short term (the contrast with population trend is only apparent, the sign of the trend is given by the Tavolara colony) and the lesson we learnt from the species is that we must be very cautious to declare local extinctions.

    As a last point, I would like to stress that the first available estimate of the Tavolara Archipelago population indicates 6,000-9,000 pairs in 1978-79 (Schenk and Torre 1986). Not such a large difference with recent estimates. In case of decline, even a species with a long longevity and a delayed sexual maturation as the Yelkouan Shearwater should have given a sign of decrease in 35 years, especially in its largest colony, which could be hardly sustained by small peripheral ones.

    Eugenio Dupre Ministero dell’Ambiente e della Tutela del Territorio e del Mare (16/6/15):
    We fully concur with Lorenzo’s thoughts and remarks; data that have been provided are sound and show without doubt, and regardless of periodic fluctuations, a constant increase, even if low, in the population of this species. These are facts from the best available data

    Nicola Baccetti – Istituto Superiore per la Protezione e la Ricerca Ambientale (ISPRA) – request for quotes from 2012 paper to be included in Forum:

    Introduction: continuous fluctuations of Sardinian population estimates Sardinia (24,000 km2) is the second largest island in the Mediterranean Sea. An update of the national breeding figures for Scopoli’ s Shearwater Calonectris diomedea and Yelkouan Shearwater Puffinus yelkouan highlighted the outstanding importance of this area for the conservation of both species (Baccetti et al. 2009). Concerning the latter species, the most recent estimates (Bourgeois & Vidal 2008, Baccetti et al. 2009) were suggesting that at least 18% (and possibly up to 61%) of its world population breed at 12 colonies located on Sardinia’s sheer cliffs and satellite islands. The large degree of uncertainty comes from the fluctuating estimates proposed so far for the largest breeding site, Tavolara island. Indeed, estimates for this crucial area were repeatedly changed since the late 1970s, although re-assessments were not always based on updated field data. Consequently, not all changes of national (and global) estimates for this species were associated to genuine improvements in knowledge. The first estimates for Tavolara date back from 1978-79 (6,000-9,000 pairs for the whole archipelago, i.e. Tavolara, Molara and Figarolo islands) and were based on counts of huge rafts that at that time were usually seen offshore (Schenk & Torre 1986). These figures were used as such for the following 15 years, until Brunner et al. (2002) dropped them to 1,000-2,000 pairs. In 2006, counts from land of shearwaters moving towards their colonies (Zenatello et al. 2006) increased again the population estimate. The values which were proposed for Tavolara since 2006 on (500-5,000 pairs: Bourgeois & Vidal 2008, then 1,200-7,800 pairs: Baccetti et al. 2009) were based on steadily improving estimates obtained through the latter method.

  2. Dilek Sahin says:

    We have been counting yelkouan shearwaters in the Bosporus since 2010. We conducted systematic coastal counts twice in a month throughout a year between 2010-2014. In 2014 we decided to cover only the first week of February as this is the peak period for yelkouan shearwater numbers in the area.

    We discovered this peak period by the high numbers that we obtained and realised that there is an extremely synchronised migration day in the Bosporus each year. So the peak number is coming from just one day, not from the whole week. In other days the numbers are close to the average of the month or sometimes very low. So, before we started to cover the whole week the high numbers we obtained doesn’t mean the highest numbers for that year. The difference between 73,000 and 90,000 yelkouans can only mean that we are learning when to count yelkouan shearwaters; it doesn’t indicate any population trend, any increase in population etc. It just shows the changing effort. We need many more counts from the same period in order to gain some insight into the total numbers of this “migration” day.

    Another point is I have been putting my effort on colony research for two years in Turkey. So far I have checked only 4 islands properly with night work and have 9 more islets and islands to visit. I have no positive result from this work so far and moreover, all the islands that I visited have obvious rat population. So in case we have breeding colonies in those islands that I have visited, the birds are probably not doing very well. But obviously, more effort should be put for colony research in Turkey.

    With Jose Pedro Tavares, we offered 0-? for the Turkey population size of Yelkouans in Developing Population Assessments for Coryʼs and Yelkouan Shearwaters Workshop in 2012. Because we didn’t have any data on the breeding status of yelkouan in Turkey in that period. The situation is exactly the same now. I saw data on Turkish population as 0-1000 breeding pairs, if this number is included in total (39,000 – 62,000) despite its poor quality then I strongly recommend to change it into 0-? again. Because to the best of my knowledge, no one is undertaking colony research in Turkey other than me and no one has discovered any breeding yelkouan shearwater pair in Turkey so far.

  3. Joe Sultana says:

    “In the case of Malta…it should not be down-listed to least concern, in spite of the successful project we had at the largest colony L-Ahrax tal-Madonna on mainland Malta, with the successful of a rat control programme. However elsewhere the islands are infested with rats. We have no knowledge what is happening in the cliffs where the breeding sites are not humanely accessible. So long as the situation remains as such…we should continue to regard it as vulnerable in Malta. Furthermore we still need to evaluate the negative effect of light pollution.”

  4. Nicola Baccetti says:

    A joint comment from N. Baccetti and L. Serra, Italy:
    Italian population estimates and trends of YS originated from a standard approach to available data applied by our Institute (ISPRA) for the reporting activity to the Birds Directive. We would like to make clear some points, after some disorientation followed the news release.
    1) In absence of national population trends specifically modelled, but in presence of population estimates, national trends were calculated from initial/final values of population size, both in the long term (1983-2012) and in the short term (2001-2012). This is the case of the YS.
    2) For YS in particular, the national population estimates and trends strongly depend on values that are recorded at the most important breeding site i.e. Tavolara.
    3) A recent overview (Zenatello et al 2012) provided a new, comparatively high local population estimate for Tavolara and resumed all former estimates, without clearly warning on the unsuitability of any of these values for comparisons and trends, this being often the case for many nocturnal burrowing petrels).
    4) The paper mentioned above didn’t deal with threats and evidences that had (not much earlier: Bourgeios & Vidal 2008) lead to YS up-listing, whose conclusions might well be considered as still valid on the global and national scale.
    5) A different approach for the Italian BD reporting could have theoretically ignored any published YS population estimate for trends. However, this would have had the only possible result of leading to an ‘Unknown’, and never to a Decreasing trend category.
    6) Considering that BD reporting dataset contained no mistakes in numbers and/or in (standard) procedures for the YS, we propose that its outcomes are interpreted by the extensors of the down-listing with our caveats in mind, which might hold true also for other petrel species and other countries, that is, the positive trend on the short term has a weak basis as already underpinned in the report by the low score assigned to data quality. A further indication of the low quality of the data is that the magnitude of the short term trend was not calculated in the report.
    7) We hope that the discussion on this conservation case will offer a stimulus to improve monitoring efforts and enhance monitoring techniques in time for the next report (2013- 2018).

  5. Dr. Nicky Petkov says:

    Some time ago I expressed my doubts when there was suggestion for uplisting the species indicating that there are large counts from Black sea region I indicated one personal observation when I counted over 15000 birds flying along Bulgarian N Balck sea coast towards Romania in April-May period. However then number of experts on the species breeding colonies expressed concerns over poor breeding success. Is there now enough data indicating that breeding success has improved and that there is actual increase in the population or we just now more about the species distribution and have better data?!?!

  6. Pierre Yésou says:

    As said when introducing my contribution to the previous (2011-2012) evaluation, although not working on Yelkouan Shearwater I am confident enough in my knowledge of the Mediterranean seabird scene to give an informed advice on this species.
    The most recent serious evaluation of the Yelkouan Shearwater status and threats still is that conducted in October 2011 when a BirdLife workshop on this topic was hosted by Medmaravis and Parco di Porto Conte in Alghero, Sardinia. Thirty-six ornithologists, both researchers and conservation managers from all over the species range, have contributed to this evaluation (Derhé 2012) which was a major contribution to the 2012 IUCN red-listing process.
    Relatively little new information as been collected between this evaluation and the preparation of the national contributions for N2000 reporting in 2014, on which the BirdLife for IUCN 2015 European Red List is mostly based. Such a lack of true improvement in our knowledge on the species status might have led to a statu quo, i.e. keeping on the 2012 evaluation of the Yelkouan Shearwater as VU. However the species has been down-listed as NT on the 2015 European Red List.
    Such a counterintuitive evaluation may result from misinterpretation of the available data, as suggested by the comments above from Dilek Sahin, Nicola Baccetti and Joe Sultana. The fact is that shearwaters numbers are particularly difficult to evaluate, and surely the evaluation of available data is better done by those who collect them than in any office by people without strong field expertise of the species (sorry for BirdLife evaluators).
    Derek Sahin told it simply: increasing numbers does not always mean increasing population, increasing numbers can result from improved practice. This is the case of the Bosphorus counts. Similarly, the expert advice of ornithologists conducting the most advanced surveys on Yelkouan Shearwater, on Italian and Maltese colonies, is that their data should not be misinterpreted: the species should not be down-listed.
    I concur: Yelkouan Shearwater should continue to be evaluated as a VU species, since there is no strong signal that its overall status, including demographic trend, has improved since the previous evaluation.

    Reference: Derhé, M. 2012. Developing a population assessment for Yelkouan Shearwater Puffinus yelkouan (Pp.65-73). In Yésou, P., Baccetti, N. & Sultana, J. (Eds.), Ecology and Conservation of Mediterranean Seabirds and other bird species under the Barcelona Convention – Proceedings of the 13th Medmaravis Pan- Mediterranean Symposium. Alghero (Sardinia) 14-17 Oct. 2011. Medmaravis, Alghero.

    • Nicky Petkov says:

      It seems my perception of the case is correct – we have better data and improved efforts on acquiring data, leading to higher figures and this seems to lead to false impression of population increase. We should not rush to downlisting species just because we have better data, which is sometime obscure as well.. Downlisting seems to me premature one..

  7. Karen Bourgeois says:

    The Yelkouan shearwater is proposed to be down-listed from Vulnerable to Least Concern based on a speculative population increase. In general, efforts dedicated to the species population census have dramatically increased in the past 20 years, leading inevitably to the discovery of “new” (most probably old but unknown) populations and to increased numbers. As an example, the French breeding population was estimated at 292-414 breeding pairs in the late 1990s while it was estimated at 627-1044 breeding pairs in 2007. This is due to a much higher effort in terms of the number of people involved in the census, means used for the census (boat, rope access, burrowscope…) and the time dedicated to the census, as well as to the improvement of the census methodology, but this definitely does not indicate an increase in the population size. By the way, the stable long term population trend in France is very questionable regarding the history of the species population census in France. I would agree to state the trend as unknown but it would even be more accurate to state it as decreasing because of the impact of feral cat predation (Bonnaud et al 2009, 2012, 2015). Moreover, even after feral cat population control, adult survival is too low to allow sustainable populations (Oppel et al 2011). Similarly, the extent to which the population really increase in Malta needs to be determined. Banding data in Malta from 1969 to 1994 and from 2007 to 2010 show low rates of survival at least for part of the population (74 ± 2.8 %, 85 ± 13 %) which means that theoretically the population is declining (Oppel et al 2011). It is difficult to monitor the breeding and to census the population in Malta as the Yelkouan shearwater nests in cracks/rocky places which are too deep for monitoring and along cliffs which are very difficult to access. However, data from banding and population models indicate declining populations despite conservation actions, probably because of at-sea mortality of adults.
    Population estimates and trends in Italy are mainly based on raft counts. Unfortunately, this method is not recognized worldwide as an accurate way to monitor shearwater populations. This method is generally regarded as weakly accurate to estimate population sizes because the number of rafting birds varies markedly according to weather conditions, moon conditions, the period of the breeding season and other unidentified factors. Thus, counts of rafts are rarely used to estimate seabird (shearwater) population sizes. Also, it is known from work carried out in Madeira and the Azores (by Mougin et al., Monteiro and Furness mainly, but also others) that number of birds in rafts can change quite largely during the breeding season, and among years. This is also underlined by the authors of the last counts around Tavolara (Zenatello et al 2012): “Inter-annual differences were large, and so were the differences of counts performed in consecutive weeks… Possible confounding factors (e.g. height of waves and visibility) were not quantified, hence their effect on count accuracy could not be evaluated. Both investigated years gave extremely fluctuating figures, even in similar dates” and “The large variability of counts from land showed that, besides seasonal variations, local weather conditions and distribution of prey might play an important role, affecting daytime distribution, flight paths, detectability, and possibly also the return rate of individuals; some of these factors are known to affect the nocturnal activity patterns at the colony”. Similarly, counts in the Bosporus are susceptible to vary markedly according to environmental conditions and behavioral factors. It should also be noticed that the first week of February can be considered as the breeding period. Indeed, Yelkouan shearwaters start to return to their breeding sites by late October-early November and all of them are back by January in the western and central Mediterranean (Péron et al 2013, Raine et al 2013). Thus, birds counted in the Bosporus in early February are most likely non-breeders (since much less than 90,000 birds breed in the neighborhood), whom behavior and migration are even less predictable. It is thus hard to tell if variation in numbers in the Bosporus is related to real variation in the population size or to changes in behavior or migration patterns…
    As highlighted by other experts, there is relatively little new information since the most recent serious evaluations of the Yelkouan shearwater status and threats (Bourgeois and Vidal 2008, Derhé 2012) which were major contributions to the 2008 and 2012 IUCN red-listing processes. Despite successful conservation actions, the most critical threats to the species, that is those impacting adult survival (predation by feral cats, at-sea mortality as fisheries bycatch), are still ongoing thus threatening the persistence of most populations. Regarding this information, I cannot understand the results of the evaluation report for European countries or the subsequent down-listing of the species on the European Red List of Birds. I definitely do not support the down-listing of the species which should continue to be regarded as Vulnerable.

    Bonnaud E., Berger G., Bourgeois K., Legrand J., Vidal E. 2012. Predation by cats could lead to the extinction of the Mediterranean endemic Yelkouan shearwater Puffinus yelkouan at a major breeding site. Ibis 154, 566–577.
    Bonnaud E., Bourgeois K., Vidal E., Legrand J., Le Corre M. 2009. How can the Yelkouan shearwater survive feral cat predation? A meta-population structure as a solution? Popul. Ecol. 51, 261–270.
    Bonnaud E., Palmas P., Bourgeois K., Ollier S., Zarzoso-Lacoste D., Vidal E. 2015. Island specificities matter: cat diet differs significantly between islands of a major breeding archipelago for a vulnerable endemic seabird. Biol. Invasions in press.
    Bonnaud E., Zarzoso-Lacoste D., Bourgeois K., Ruffino L., Legrand J., Vidal E. 2010. Top-predator control on islands boosts endemic prey but not mesopredator. Anim. Conserv. 13: 556–567.
    Bourgeois K., Vidal E., 2008. The endemic Mediterranean Yelkouan shearwater Puffinus yelkouan: distribution, threats and a plea for more data. Oryx 42, 187–194.
    Oppel S., Raine A.F., Borg J.J., Raine H., Bonnaud E., Bourgeois K., Breton A.R. 2011. Is the Yelkouan Shearwater Puffinus yelkouan threatened by low adult survival probabilities? Biol. Conserv. 144: 2255–2263.
    Péron C., Grémillet D., Prudor A., Pettex E., Saraux C., Soriano-Redondo A., Authier M., Fort J. 2013. Importance of coastal Marine Protected Areas for the conservation of pelagic seabirds: The case of Vulnerable yelkouan shearwaters in the Mediterranean Sea. Biol. Conserv. 168: 210–221.
    Raine A.F., Borg J.J., Raine H., Phillips R.A. 2013. Migration strategies of the Yelkouan shearwater Puffinus yelkouan. J. Ornithol. 154: 411–422.

  8. John J. Borg says:

    In a paper presented at Hammamet, Tunisia last February during a RAC/SPA and MEDMARAVIS meeting I stressed on the interpretation of numbers when estimating populations. One must consider carefully dates and time of counts as well as a strong knowledge of the bird`s behavior, at sea and on land. Telemetry studies are showing that rafting birds in front of breeding colonies may originate from other colonies from other countries, therefore such counts can only be regarded as “guesstimates”.

    Until definite proof of breeding in the Black Sea is confirmed, the thousands of birds recorded are to be regarded as young birds, non or failed breeders from Mediterranean colonies.

    The increase of breeding birds registered at Rdum tal-Madonna colony in Malta does not reflect an actual increase in population numbers but an increase in occupied sites formerly abandoned due to predation by rats. The situation is far from acceptable in the other colonies around the Maltese archipelago with a mortality rate in chicks as a result of rat predation. Light pollution and land encroachment are also major impact on the breeding colonies.

    Like my colleagues before me I concur that: Yelkouan Shearwater should continue to be evaluated as a VU species, since there is no strong signal that its overall status, including demographic trend, has improved since the previous evaluation.

  9. Ben Metzger, BirdLife Malta says:

    The latest population trends on P. yelkouan in Malta, published as part of the article 8 (1) of the MSFD and article 12 of the Birds Directive reporting requirements are interpreted by the Maltese authorities as follows:
    ‘These trends are indicative of a general increase in the number of breeding pairs at most localities throughout the years, however this increase may be mostly due to an increase in research efforts.’
    Indeed, a strong increase in research effort on Maltese seabird populations, mainly over the last 8 years (2007 – 2014) as part of two EU-Life funded projects carried out by BirdLife Malta, led to the discovery of previously unknown breeding sites and hence to a seemingly positive population trend of this cryptic species.
    In Malta, only one colony, hosting approximately 25% of the Maltese P. yelkouan population, has a predator management scheme implemented. This resulted locally in an increase in hatching and fledging success. However, all other breeding colonies are infested with R. rattus and are widely lacking any management measures. Nest monitoring (representative sub-sample) in one of the large unmanaged colonies revealed that more than 70% of eggs and chicks were predated by rats. To date, there is no evidence that young birds originating from the managed colony are recruiting into other colonies or establishing new ones.
    The last evaluation of adult survival rates of P. yelkouan in Malta (Oppel et al. 2011) concluded from capture mark recapture data (2007-2010) that mortality rates were too high to maintain a stable population. Since then, there has not been any re-assessment that would indicate an improvement of adult survival rates. To my knowledge, large scale by-catch assessment and mitigation programmes have not yet been implemented for the fishing fleets in the Central Mediterranean, the Aegean and the Black Sea, which would tackle the believed major cause of non-natural adult mortality.
    Over the last decade, other major threats and disturbances to Maltese P. yelkouan populations such as coastal zone development, light pollution and boat tourism close to cliffs and into sea caves where P. yelkouan nest have strongly increased.
    Overall, a true short-term increase in P. yelkouan breeding pair numbers in Malta by 20-30% appears unlikely and I recommend to carry out an in-depth trend assessment accounting for the research effort, before considering a down-listing of the species from Vulnerable to Least Concern.

    • Joe Sultana says:

      First of all I fully agree with what both John J. Borg and Ben Metzger has said above. Furthermore I wish to add with what I had written on July 16, 2015. The Yelkouan Shearwater, which used to breed on Filfla up to the late 1960s (pers. obs.), has not been recorded breeding there since the 1970s.

  10. W.R.P. Bourne says:

    Yelkouan Shearwaters are not rare in most of the Mediterranean, and the situationis complicated by the fact that many seem to migrate into the Black Sea. It seems likely that there are still many undiscovered breeding=places where the birds probably do bette4r than at those that aqre known. I see no reason to consider it vulnerable.

  11. Carles Carboneras says:

    This sequence of posts proves clearly that, despite living in the densely-populated Mediterranean Sea, Puffinus yelkouan is still a poorly known species. One of the least known aspects of its ecology is how much it is affected by mortality at sea. In a review presented at the ACAP 7th Advisory Ctte meeting in 2013 (Carboneras et al., 2013), we identified this as a potentially major factor in its conservation, and recommended Mediterranean countries that are Parties to ACAP to propose it for listing in Annex I and to address seabird bycatch in their fisheries.

    Unknown levels of bycatch are affecting Yelkouan shearwaters in several parts of its distribution range. Direct observation confirms that it is happening at least in western Mediterranean fisheries. Investigating the magnitude of that threat (e.g., through direct observation on board fishing vessels), and how it is affecting the species’ demography (e.g., through long-term capture-recapture studies), have been identified as conservation priorities. It would make sense to postpone any decision on the status of this species until enough robust evidence is available to predict with a level of confidence the fate of its population in the time of 3 generations.

    Carboneras C, Derhé M, Ramírez I. 2013. Update on the population status and distribution of Mediterranean shearwaters. 7th meeting of the Advisory Ctte of ACAP, La Rochelle. Available from:

  12. Andy Symes (BirdLife) says:

    Yoav Perlman and colleagues at Israel Ornithological Center / Society for the Protection of Nature in Israel have provided a summary of the status of Yelkouan Shearwater in Israel -see summary copied here:

    Substantial numbers winter off the Mediterranean coast of Israel, and are counted mainly during storms that push seabirds closer to shore. This effort is not standard, so we divided total birds seen in winter by number of observations. There is no significant change in numbers between 2002 and 2011.

  13. In virtually all the colonies where the reproductive success of PY has been examined, this has always been nil or very low, except for those were rats (or cats) had been eradicated or controlled (questo varrà per l’Italia, per Francia no). In 2011, while presenting the Tavolara LIFE project proposal, we estimated the number of pairs protected from predation. This number resulted in 1200-1500 protected pairs, of which over 500 only thanks to annual sessions of rodent control (therefore conservation dependent). From 2011 to date there have not been, as far as I know, significant changes.
    Then, only 5-10% of the population of PY seems to have a normal productivity, the remaining colonies presumably achieve some breeding success only in years in which the fluctuations of rats reduce the rate of predation. On Tavolara, where the largest known population breeds, only in 2015 was reproductive success greater than 0 (around 0.4) in areas not subject to rodent control, thanks to a natural collapse of the rat population that has presumably reached its minimum levels (capture rate = 1/10 of that of 2010).
    Considering also the other pressures and the fact that the supposed increase seems to be due to knowledge improvement, the species should still be considered as vulnerable.

  14. Andy Symes (BirdLife) says:

    Preliminary proposals

    Based on available information, our preliminary proposal for the 2015 Red List would be to list:

    Yelkouan Shearwater as Near Threatened.

    There is now a period for further comments until a final deadline (later for this species) of 11 September, after which the recommended categorisation will be put forward to IUCN.

    The final Red List categories will be published on the BirdLife website in late October and on the IUCN website in November, following further checking of information relevant to the assessments by both BirdLife and IUCN.

    • Nicola Baccetti says:

      It’s better than nothing… but we still cannot see which changes or improvements at the range level might support a modification of the very recent evaluation as Vulnerable. The rat problem has been removed on a few islands (not the most important ones for the population) and nothing was done to reduce adult mortality or just to assess it.
      Nicola Baccetti, Marco Zenatello

    • Karen Bourgeois says:

      There’s no new data suggesting that the overall population is not declining by >30% as stated in 2012 based on sound data (breeding success, adult survival, population viability analyses). Contrarily, the fact that “the European breeding population is now increasing overall, with stable or increasing trends in those countries with the largest populations (Italy, Greece and Malta, which together hold c. 95% of the European breeding population)” is not proven at all and national experts do not support this statement, indicating that increased numbers are related to an increase in research efforts rather than a real increase in population sizes. If the status of the species was seriously evaluated, it would continue to be evaluated as Vulnerable.

  15. Pep Arcos says:

    I did not have the chance to participate in this debate in its early phases, and now I see that most of the arguments that came to my mind against the downlisting of the Yelkouan shearwater (YS) have been already expressed by many other colleagues. I must say that these arguments seem very convincing, and should lead to keep the YS status as VU, at least for the sake of the precautionary principle. This status was reached 4 years ago through a thorough process that included a proper consultation to researchers and conservation technicians well experienced on the species, considering population trends and threats, with the available information at the time. On the other hand, the “new” information claimed to support the downlisting of the species to LC was produced only 2 years later, and was basically the result of a less thorough compilation of data by EU Member States (the process-and the degree of implication of experts on the species- varying from country to country). The current debate has showed that the compiled information was in many cases misinterpreted, as stressed by the same people that produced such “new” information. The main alleged points are the following:

    1. The European (i.e. almost the global) breeding population is increasing overall. This statement was based primarily on (a) a perceived increase in Tavolara, the major Italian colony; and (b) an increase of the Maltese population.

    (a) As stressed by several colleagues, the census of breeding shearwaters is complex due to their often inaccessible breeding sites and secretive (nocturnal) behaviour in the colonies. It is often relied on indirect methods, which can range from simple counts of birds around the colonies (as is the Tavolara case) to more sophisticated inferences out of well prospected nesting areas and taking into account habitat features. The count of birds at sea is particularly subject to very strong biases, and should not be used as a method to assess trends. It is even worst when different people have been in charge of the censuses at different times, using different approaches and having different perceptions. Therefore, inferring an increasing trend from such type of censuses is completely unreliable. As an example, it is worth to recall a paper by Perrins et al. (2012) who describe an “increase” of Mannx shearwaters in Skomer (Wales) from 101,000 to 316,000 breeding pairs in 13 years (1998-2011); the increase is likely attributed to a change of methodology.

    (b) Again as already noted, the perceived increase in Malta is limited to a single (though important) well-managed colony (Rdum Tal-Madonna), where nests previously abandoned due to predation were reoccupied due to rat control. But the situation is still bad for most of the Maltese population. And, moreover, a proper demographic study has not been conducted to assess if the Rdum Tal-Madonna population increase could be sustained without the recruitment of birds from other sites.

    Other related issues worth to be considered:

    (c) An additional way of perceiving “false” increases is through the discovery of new sites, or the better prospection of already known ones. This could have also played a role in Italy, as already stressed. On the other hand, the disappearance (or at least sharp decline) of some small colonies has not been considered seriously, but once a colony disappears, it’s difficult to get it recolonized, and this could be a sign of general population decline.

    (d) Claiming that a given population is simply subject to important fluctuations (as described for Tavolara) just confirms that population assessments are too weak. Shearwater populations are rather stable, due to their long lifespan, low breeding rates, and high phylopatry. Only through changing the rates of birds taking sabbaticals could a population fluctuate to some extent between years, and strong differences in sabbatical rates do not seem likely.

    (e) For species such as shearwaters, which are so difficult to count properly in the breeding grounds, population trends should rely whenever possible in sound demographic information and population viability analysis based on it. As noted, all colonies with demographic information show unusually low adult survival rates (around 0.74-0.82, when values over 0.90 should be expected; Oppel et al. 2011), and breeding success is also very low in most of the colonies, being close to 0 in many cases. These demographic traits would point to a severe declining trend (see below a comparison with the Balearic shearwater), and therefore they should receive more attention.

    2. The perceived increase in the Bosporus, away from the breeding areas, has also been considered as a sign of a healthy, increasing population. Again, as stressed by the responsible of these counts, this “increase” is not real, but rather the result of increasing and optimizing the survey effort. The figures provided, of up to 90,000 birds in a single day, suggest that the global population might be larger than previously expected, but this must be taken with caution. First, because we don’t know the origin of these birds, that are most likely non-breeders. And second, because relating global to breeding figures is subject to biases. See also the analogy with the Balearic shearwater right below.

    I think it’s worth now to make here an analogy with the case of the closely-related Balearic shearwater (BS), as the situation of both species is quite similar:
    The BS was catalogued in 2004 as CR as a result of a small population (around 2000 breeding pairs) subject to a severe decline (7.4% per year), that would lead to its extinction in an average time of 40 years (Oro et al. 2004), less than 3 generations (54 years).
    A few years ago, different studies at sea suggested that the global population of BS might be around 25,000 birds (Arcos et al. 2012), a figure that was far larger than expected (inferred estimates from the breeding population suggested around 10,000 birds). This prompted a reassessment of the breeding population, which was raised to 3,193 breeding pairs (Arcos 2011). This increase was not real, but the simple result of better prospection for some colonies, and simply a subjective reassessment for others. The figure is still low to explain 25,000 birds, which would be more consistent with a breeding population over 7,000 pairs. It is difficult to imagine that the actual population is indeed that high, but it’s not unimaginable. Alternatively, a huge “floating” population might be involved, although it is difficult to imagine how/when this large number of non-breeders was originated.
    The news of the increased global population soon raised concerns about the status of the species, which might be less extreme than previously considered and therefore needed revision. To this aim, a couple of years ago we asked Daniel Oro to run again the population viability analysis of the species, updating his model of 2004 with new information on breeding success and a few minor further changes on the assumptions. We run the model under two different assumptions regarding population size: 3,200 breeding pairs (official estimate of breeding pairs) and c. 7000 breeding pairs (likeliest figure inferred from a global population of 25,000). Information on survival was not updated, so we still had a low adult survival (0.78 – similar to the values estimated for YS) that resulted in a severe population decline. In both projections the population would reach the 10% of its current value in 23 years (i.e. far less than 3 generations, threshold to classify as CR). The average extinction time increased relative to the PVA of 2004, being of 76 years if the population was of 3200 bp, but only 8 years more (84) if the population was of 7000 bp. A recent revision, with further information (including new survival estimates) and more robust modelling approaches has led to relatively similar results, or even worse, with an average time of extinction of slightly over 60 years departing from a population of 7000 bp (Genovart et al. in prep).

    The similarities between BS and YS are too many, and considering one with the highest category of concern in the wild (CR – justified!) while downlisting the other to LC or even NT seems not appropriate. The YS population is considerably larger, and therefore this might certainly justify a lower concern; but the two species show similar demographic values (or at least this is what the little information available suggests) and face very similar threats, thus both running towards extinction, the difference being a mere question of time (and maybe not that much). In fact, both species seem equally affected by fisheries bycatch, but the pressure of predation by rats (resulting in very low breeding success) appears to be stronger in the YS.
    With all this said, I strongly recommend to adopt the precautionary principle and hold the status of the YS as VU, at least until a proper population viability analysis is conducted, taking into account information from a diversity of colonies throughout the species’ range.

    Quoted references:

    Arcos J.M. (compiler) (2011). International species action plan for the Balearic Shearwater, Puffinus mauretanicus. SEO/Bird-life & Bird-life International.

    Arcos J.M., Arroyo G.M., Bécares J., Mateos-Rodríguez M., Rodríguez B., Muñoz A.R., Ruiz A., de la Cruz A., Cuenca D., Onrubia A. & Oro D. (2012b). New estimates at sea suggest a larger global population of the Balearic Shearwater Puffinus mauretanicus. Pp. 84-94 in Yésou P., Bacceti N. & Sultana J. (Eds), Ecology and conservation of Mediterranean seabirds and other bird species under the Barcelona Convention. Proceedings of the 13th MEDMARAVIS Pan-Mediterranean Symposium, Alghero (Sardinia).

    Oro D., Aguilar J.S., Igual J.M. & Louzao M. (2004). Modelling demography and extinction risk in the endangered Balearic Shearwater. Biological Conservation) 116: 93-102.

    Oppel S., Raine A.F., Borg J.J., Raine H., Bonnaud E., Bourgeois K. and Breton A.R. 2011. Is the Yelkouan shearwater Puffinus yelkouan threatened by low adult survival probabilities? Biological Conservation 144(9): 2255-2263.

    Perrins, C.M., Wood, M.J., Garroway, C.J., Boyle, D., Oakes, N., Revera, R., Collins, P. & Taylor, C. 2012. A whole-island census of the Manx Shearwaters Puffinus puffinus breeding on Skomer Island in 2011. Seabird 25: 1-13.

  16. Ben Metzger, BirdLife Malta says:

    I share the opinion of Nicola Baccetti, Karen Bourgeois and Pep Arcos to keep the status of the YS as VU following the precautionary principle as long as no clearer picture on population viability, demography etc. from main colonies indicate a real improvement of the situation.

  17. Andy Symes (BirdLife) says:

    Many thanks to all who have taken the time to comment and provide data on this species.

    We accept the argument that the most prudent approach at this stage would be to precautionarily maintain Yelkouan Shearwater as Vulnerable, recognising that the various components of the 2012 study may represent a more accurate assessment of the species’ status than the European Red List data, given the difficulties associated with monitoring the species.

    If further study and monitoring fail to provide evidence of declines, the species should again be considered for downlisting in the future.

Comments are closed.