Please note: This discussion topic is one of a set about species that are endemic or nearly endemic to the European Union (EU), and whose status in the EU therefore effectively determines their global status. To ensure consistency between the 2020 global and EU Red List assessments of these species, this set of topics is being fast-tracked through BirdLife’s Globally Threatened Bird Forums to inform decisions on the EU (and global) status of relevant species, which must be finalised and communicated to the European Commission by mid-April 2020. Topics on other species will be posted on the Forums shortly, for discussion later in the spring, as per usual. The results of the 2020 global Red List update for birds will be published by IUCN and BirdLife in early December.
Balearic Shearwater Puffinus mauretanicus is a Spanish breeding endemic seabird, nesting in multiple colonies on all five major island groups in the Balearic Islands: Menorca, Mallorca, Cabrera, Ibiza and Formentera. The relevant three generation length period for assessing trends in this species under IUCN Red List Criterion A is 42 years (based on an estimated generation length of 13.9 years; Bird et al. 2020).
The species has been assessed as Critically Endangered since 2004, on the basis of demographic modelling, which inferred and projected that an extremely rapid population reduction was taking place and would continue (Oro et al. 2004, Genovart et al. 2016). These models were parameterised using values measured during a long-term study at a single breeding colony (Sa Cella on Mallorca), and predict an annual decline of 14% (Genovart et al. 2016). Further demographic investigation at two islets off Ibiza between 2011 and 2019 also suggested a 14% annual decline (J.M. Arcos in litt. 2020). Mammalian predators do not affect the studied colonies, so it is suggested that declines might be even more severe at colonies where predators are present (Genovart et al. 2016).
As the modelled estimate of population decline forms the basis of the current assessment as Critically Endangered, it is necessary to carefully examine the assumptions within the complex multi-event model presented in Genovart et al. (2016). One concern is that the model is based on a closed population and estimates only ‘local’ survival, rather than taking into account possible emigration to (or immigration from) other colonies. The poor fit of the best global model is explained by the presence of juvenile transients (i.e. birds that leave this colony), which is addressed post-hoc by including age in the models and adding a correction for over-dispersion. However, if adult birds switch colonies (or are simply not observed, in this difficult-to-access sea cave colony), the model appears to treat them as equivalent to dead birds and assigns them the probability of being dead (from bycatch or other causes). As with any estimate of ‘local’ survival in an open population, this approach will underestimate true survival in the absence of any actual mortality. Although breeding site fidelity is likely to be high in this species, it is more likely to be imperfect, as indeed is detection.
A second, related, concern regarding the Genovart et al. (2016) paper is the derivation of the estimates of bycatch mortality, which are almost entirely responsible for the presented population trend projections. The input data for this parameter in the model needs clarification, as the number of bycatch events used is not readily apparent. There are a list of bycatch reports in the supplementary table, but this refers to a separate section. This is a critical parameter, but the standard error is >50%, and the 95% confidence interval is 0.119–0.837, which implies that the predicted rate of decline may be sensitive to this highly uncertain value. Given the scenarios in the discussion of the paper, it seems it is highly sensitive. Lowering the estimated bycatch mortality greatly improves the population persistence (Figure 2, scenarios 3-6 in Genovart et al. 2016). The statement that the estimate of bycatch rate “contains much uncertainty and should be treated with caution” is crucial and should be heeded.
For these reasons, it appears likely that the approach taken in Genovart et al. (2016) would overestimate the actual rate of global population decline. This does not mean that the species is unlikely to be declining. But it does mean that unless a bound of lower and upper plausible rates of estimated decline can be inferred from the modelling, it is not credible (nor in line with the IUCN Red List Guidelines; IUCN 2019) to continue to use an estimate predicting an immediate population crash to assess the extinction risk of this species, despite no evidence of such a crash for two decades.
There is good evidence of rapid, directly observed declines for certain parts of the population. On Formentera, more than 1,500 pairs were present in the early 1990s, declining to fewer than 1,000 pairs in 2001 and 692 pairs in 2003-2006 (Arcos 2011): a rate of decline exceeding 80% over three generations (an annual decline of c. 4% per year). Mammalian predators (cats Felis catus, genet Genetta genetta and Martes sp.) were considered likely to have caused significant adult mortality here, as well as driving local extinctions on Cabrera and causing recorded mortality of adults and nests in the cave colony at Mola de Maó on Menorca (Arcos 2011).
Also, despite the uncertainty highlighted above, there is good reason to believe that rates of bycatch, which principally impacts adults (Cortés et al. 2018), are higher than can be sustained by the species’s known population (Genovart et al. 2016, Tarzia et al. 2017). Using self-reporting logbook information, the estimate of annual bycatch was 622 Balearic Shearwaters in the small-scale fisheries where the risk was highest, concentrated in the pre-breeding and early breeding period (Tarzia et al. 2017). Such a rate of additional mortality is unlikely to be sustainable for a Procellariformspecies, for which adult survival has a high impact on population trajectory (Weimerskirch 2002, Arcos 2011).
The total breeding population size used in the modelling was 7,200 pairs, taking 23,780 individuals as an initial global population size (consistent with at-sea estimates from migration counts from the Gibraltar Strait bottleneck), and assuming demographic equilibrium (Arcos et al. 2012,Arroyo et al. 2016, Genovart et al. 2016). This differs from the estimate resulting from colony estimates (c. 3,000 breeding pairs), as some colonies are either difficult to access or inaccessible, and have only been censused rarely or indirectly, which probably explains the discrepancy in size (Arcos et al. 2017).
However, the post-breeding count data from the Spanish side of the Gibraltar Strait, collected annually since 2008, have recently been analysed and suggest the population is actually increasing (Martín et al. 2019). These counts cover a high proportion of the estimated global population (see Table 1 in Arroyo et al. 2016: over 12,500 individuals directly counted annually), almost all of which are known to leave the Mediterranean after breeding to winter in the NE Atlantic (Guilford et al. 2012). Birds missed due to gaps in monitoring were estimated using Generalised Additive Models trained on the actual count data, resulting in an estimation of effectively the entire global population based on present knowledge (Martín et al. 2019). These estimates of population size were initially supported separately by a boat-based transect method within the Iberian Mediterranean, with good congruence between the results of the two methods (Arcos et al. 2012).
Concerns have been expressed that the land-based survey methodology suffers from the risk of inflating the population size due to potential double-counting, if more localised foraging circuits by some birds result in repeated passes in the same westerly direction on multiple days. However, the similarity in estimates of the larger population size between the land- and boat-based methodologies indicates that there really are a large number of individuals present annually. Crucially, there is no evidence of the rapid declines predicted by the demographic model. It is likely that the land-based counts may lack the precision to detect small population changes, but the trends predicted by the model above are not small changes.
As such, this topic is a request for information to assist with determining the appropriate data to use in reassessing the species’s extinction risk. On one hand, there are annual count-based estimates of an apparently stable or increasing population in excess of 25,000 individuals (of unknown age) passing one location (Arroyo et al. 2016, Martín et al. 2019). One the other, there is a model based on demographic parameters measured at one colony indicating that time to global extinction is c. 61 years (Genovart et al. 2016), and evidence of similar levels of adult survival (0.81) and rates of decline (14% annually) from demographic studies on two islets off Ibiza (J.M. Arcos in litt. 2020).
Using the information from Genovart et al. (2016), and starting with 7,200 breeding pairs (14,400 mature individuals) in 2014, an exponential decline of 14% annually suggests that in 2020 there are now 5,828 mature individuals, in three generations’ time there will be 10 mature individuals, and three generations (42 years) ago there were 2,593,382 mature individuals. Obviously, this rate of decline (99.8% over three generations) would classify the species as Critically Endangered under A2, A3 and A4. As a side note, if we set 14,400 as the global population size in 2004, this rate of decline would mean current mature individuals number only 1,290, or 695 pairs.
Accepting that the species is stable or increasing, as indicated by the post-breeding counts (Martín et al. 2019), would mean that the species could only be assessed as globally Least Concern. The population far exceeds the threshold (1,000 mature individuals) for listing under Criterion D, which is the only criterion that applies in the absence of a decline or an extremely limited geographic range (IUCN 2012). However, this ignores the observed rate of loss on Formentera (>80% over three generations; Arcos 2011), and requires that this apparent mortality has been offset elsewhere. It also assumes that reported rates of bycatch are not sufficient to impact the overall population, which seems unlikely (Genovart et al. 2016).
Of course, both situations may be true. The breeding colonies studied in the Balearic Islands may be in steep decline, as evidenced by the loss of peripheral colonies and the decline of core colonies. And they may act as population sinks, with ‘floating’ mature individuals replacing others killed each year. But floaters have to come from somewhere, and in order to explain the large and apparently stable numbers counted on passage at the Gibraltar Strait, there must be at least one large additional breeding population which has either been overlooked in the Balearic Islands, or is potentially in a hitherto unsurveyed area, e.g. Algeria (Arcos et al. 2012). Either way, there must be a large number of additional breeding birds within colonies whose demographic rates are superior to those measured to date, such that the species’s overall survival rate has been underestimated.
We need to know what proportion of the global population the apparently rapidly declining studied colonies represent (Sa Cella holds c. 300 mature individuals), and what proportion of mature individuals there are in the post-migration counts (c. 25,000 individuals). Even if we assume that the fraction of mature individuals in the population estimate from Martín et al. (2019) is actually small, immature survival would have to be exceptionally high to offset such a rapid modelled decline (contra values in Genovart et al. 2016). And if immature survival is that high, these birds would then be recruited into the breeding population from three years of age, thus further reducing the estimated rate of decline considerably.
Given the best knowledge available at present, the Red List status of this species warrants reclassification. Without a means of linking the estimates from the demographic modelling to observed estimates of population size and trend, it is not credible to use the headline figure of projected rate of decline from Genovart et al. (2016). On the basis of the recent observed data, it does not appear that current projections could justify suspecting a rapid future decline exceeding 30% over three generations (let alone any higher rate), unless we see evidence of declines at the global population level. However, the uncertainty around the contrasting trends presented suggests that it may be reasonable to suspect a future decline. Taking a precautionary approach, it is therefore possible to suspect that the rate of decline in the next three generations due to exploitation (bycatch) and effects of introduced taxa may plausibly approach 30%, such that Balearic Shearwater could potentially be reclassified as Near Threatened under Criterion A3de.
Relevant comments and information on this fast-track topic are welcome by 8 April 2020, please.
Please note that this forum topic is not designed to be a general discussion about the ecology of the species, but rather a discussion of the species’s Red List status. Therefore, please ensure your comments are relevant to the species’s Red List status and the information requested. By submitting a comment, you confirm that you agree to the BirdLife Forums’ Comment Policy.
Arcos, J.M. (compiler) (2011) International species action plan for the Balearic Shearwater Puffinus mauretanicus. SEO/BirdLife & BirdLife 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. (2012). New estimates at sea suggest a larger global population of the Balearic Shearwater Puffinus mauretanicus. 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). Pp. 84-94.
Arcos, J.M., Alonso, J., López, I. & Mayol, J. (2017). Study, monitoring and conservation of the Balearic shearwater in Spain: an update. Fourth Meeting of the Population and Conservation Status Working Group, ACAP – PACSWG Inf 25 Rev 1.
Arroyo, G.M., Mateos-Rodriguez, M., Munoz, A.R., De la Cruz, A., Cuenca, D. and Onrubia, A. (2016) New population estimates of a critically endangered species, the Balearic Shearwater Puffinus mauretanicus, based on coastal migration counts. Bird Conservation International 26: 87-99.
Bird, J.P., Martin, R., Akçakaya, H.R., Gilroy, J., Burfield, I.J., Garnett, S., Symes, A., Taylor, J., Şekercioğlu, Ç.H. & Butchart, S.H. (2020). Generation lengths of the world’s birds and their implications for extinction risk. Conservation Biology. https://doi.org/10.1111/cobi.13486
Cortés V, García-Barcelona S, González-Solís J (2018) Sex- and age-biased mortality of three shearwater species in longline fisheries of the Mediterranean. Mar Ecol Prog Ser 588:229-241. https://doi.org/10.3354/meps12427
Genovart, M., Arcos, J.M., Álvarez, D., McMinn, M., Meier, R., B. Wynn, R., Guilford, T. and Oro, D., (2016) Demography of the critically endangered Balearic shearwater: the impact of fisheries and time to extinction. Journal of Applied Ecology 53: 1158-1168. https://besjournals.onlinelibrary.wiley.com/doi/full/10.1111/1365-2664.12622
Guilford, T., Wynn, R., McMinn, M., Rodríguez, A. Fayet, A. Maurice, L., Jones, A. and Meier, R. (2012) Geolocators reveal migration and pre-breeding behaviour of the critically endangered balearic shearwater Puffinus mauretanicus. PLoS One 7 : 1-8. https://doi.org/10.1371/journal.pone.0033753
IUCN (2012) IUCN Red List Categories and Criteria: Version 3.1. Second edition. Gland, Switzerland and Cambridge, UK: IUCN. iv + 32pp.
IUCN (2019) Guidelines for Using the IUCN Red List Categories and Criteria. Version 14. Prepared by the Standards and Petitions Committee. http://www.iucnredlist.org/documents/RedListGuidelines.pdf
Martín, B., Onrubia, A. and Ferrer, M. (2019) Endemic shearwaters are increasing in the Mediterranean in relation to factors that are closely related to human activities. Global Ecology and Conservation, 20, p.e00740.
Tarzia, M. (compiler), Arcos, P., Cama, A., Cortés, V., Crawford, R., Morkūnas, J., Oppel, S., Raudonikas, L., Tobella, C., Yates, O. (2017) Seabird Task Force: 2014-2017. Technical report. Available at www.seabirdbycatch.com.
Weimerskirch, H. (2002) Seabird demography and its relationship with the marine environment. In: Schreiber, E.A., Burger, J. (Eds.), Biology of Marine Birds. CRC Press, Boca Raton, Florida, pp. 115–135.