The spatial and trophic ecology of small seabird species is far less known when compared to their larger relatives.
There is a sexual foraging and trophic segregation in Monteiro’s storm petrels.
Longer wings of females when compared to males might be a first driver of spatial and trophic segregation on the species.
Sexual isotopic segregation occurred both during the breeding and the non-breeding periods.
Females exhibited a narrower isotopic niche when compared to males.
At-sea distribution and trophic ecology of small seabird species (i.e. < 100 g) is far less known when compared to their larger relatives. We studied the habitat use (spatial ecology) and isotopic niches (trophic ecology) of the endangered Monteiro’s storm-petrel Hydrobates monteiroi during the incubation and chick-rearing periods of 2013. There was a sexual foraging segregation of Monteiro’s storm-petrels during the breeding period (tracking data) but also during the non-breeding stage (stable isotope analysis). Females took advantage of their longer wings to forage over the shallower Mid-Atlantic ridge (MAR) north of Azores, under colder and windier regimes when compared to males, who mostly exploited northern deep waters comparatively closer to the breeding colony. Between-sex differences in the spatial distribution were more obvious during the incubation period, with the overlap in their distribution increasing during the chick-rearing phase. There was also an isotopic segregation between sexes both during the previous breeding and the non-breeding stages, with females exhibiting a narrower, lower level isotopic niche when compared to males. Though the distribution patterns reported here should be useful for the at-sea conservation of this endangered species, future research should focus on (1) performing year-round tracking to map the species’ distribution during the non-breeding period and (2) gathering multi-year tracking information to understand the effect of inter-annual environmental stochasticity on the foraging choices and trophic habits of the species.
Seabirds are top marine predators increasingly used as indicators of changes in the marine environment (e.g. Parsons et al., 2008). Their at-sea distribution is strongly linked with the distribution of their prey, which in turn is influenced by oceanographic characteristics such as ocean currents, lower sea-surface temperature (SST) and higher chlorophyll a concentration (CHL) (Nogueira et al., 2012; Paiva et al., 2010). Linkages with ocean productivity might be reduced during the breeding season, when seabirds are forced to become ‘central-place foragers’ (Orians and Pearson, 1979), thus constrained to obtaining prey in comparatively less productive areas nearer the colony to provision their chick (Burke and Montevecchi, 2009). Especially when searching for food over less productive areas or constrained to forage closer to the colony, a pattern of sexual segregation in foraging might arise. This may cause the competitive exclusion of one or other of the sexes to allow partitioning of resources. In male-biased, sexual size dimorphic species like black-browed albatrossesThalassarche melanophris (Phillips et al., 2004) or northern giant petrels Macronectes halli(González-Solís et al., 2000), females are excluded by larger males, while the opposite effect–inverse sexual dimorphism–has been noted in masked boobies Sula dactylatra(Weimerskirch et al., 2009). Sexual size dimorphism in body size and wing morphology should play a functional role in flight performance and is used to explain inter-sexual differences in at-sea distribution an effect used to explain inter-sexual differences in the at-sea distribution of black-browed albatrosses (Phillips et al., 2004). However, sexual differences in size and shape might not be the only predictors of sex-specific foraging behaviour, and sexual segregation might be environmentally driven (Cleasby et al., 2015).
Stable isotope analyses (SIA) have also been used to study the at-sea distribution (Rubenstein and Hobson, 2004) and trophic choices of marine apex predators (Paiva et al., 2016). Specifically, the carbon isotopic signature of consumers is similar to basal primary productivity with only a slight enrichment across trophic levels and tends to be depleted towards the poles (Quillfeldt et al., 2005) and from coastal to pelagic environments(Rubenstein and Hobson, 2004). Thus, this biomarker assigns a broad geographical location of the birds’ feeding activity, both in terms of latitude and distance from the coast. Nitrogen isotope signatures of animal tissues reflect the predators’ trophic position, with a stepwise increase at each trophic level, and therefore can be used to infer the trophic position at which birds fed (Kelly, 2000).
Monteiro’s storm-petrel Hydrobates monteiroi is a vulnerable (BirdLifeInternational, 2016) species endemic to the Azores archipelago (north-eastern subtropical Atlantic) for which the at-sea distribution is virtually unknown (but see Meirinho et al., 2014) with habitat suitability maps from at-sea census surveys. Although the species leave their colonies by the end of the breeding period, stable isotope signatures (Bolton et al., 2008) and at-sea observations (Ramírez et al., 2008) indicate that the species may remain in the Azores waters throughout the year. However, the at-sea foraging areas used by the species remain unknown, making it impossible to assess which oceanographic profiles characterize those areas as well as to which potential threats the species might be exposed to (e.g. competition with fisheries). The species has suffered a dramatic population decline since the introduction of mammals (rats Rattus rattus, cats Felis catus and rabbits Oryctolagus cuniculus by the Portuguese in the 16th century (Monteiro et al., 1996a), and is currently known to breed only on two rodent-free islets, with breeding numbers not exceeding 300 pairs until 2008 (Monteiro and Furness, 1998; Bolton et al., 2008). More recently, Oppel et al. (2013) estimated a population of 471 pairs for Praia Islet, as a result of both the eradication of rabbits in 1997, which accelerated soil erosion through overgrazing and destroyed seabird nets, and the construction of artificial nest boxes in early 2000 (Bolton et al., 2004; Bried et al., 2009; Bried and Neves, 2015). Yet, the size of the non-breeding population remains unknown for the entire archipelago, and therefore, globally. Climate stochasticity also plays a pivotal role in influencing breeding performance and thus the population dynamics of the species, with years of low marine productivity (i.e. low CHL and high SST) and likely decrease in food availability, translating into a decrease in both the overall breeding success (Robert et al., 2014) and the survival rates of breeders (Robert et al., 2012).
We report the first at-sea tracking information of the endangered Monteiro’s storm-petrel during the incubation and chick-rearing periods of 2013 in relation to environmental variables (e.g. SST). Specifically, we studied the habitat use (spatial ecology) and isotopic niches (trophic ecology) of male and female birds during incubation and chick-rearing. We expect that (1) birds will forage in areas with lower SST and bathymetry (BAT) and higher CHL and wind speed (WSP), i.e. to the north of the breeding colony, as previous studies with Cory’s shearwater at the same archipelago (and in Praia Islet) showed that most individuals head northwards to forage in cooler waters (Paiva et al., 2010); (2) individuals will forage farther from their colony during incubation, when parents do not have to make frequent trips back to the colony to feed their chick (Magalhães et al., 2008); (3) both sexes should display a wide isotopic niche, since the species feeds on plankton (low trophic level prey) but also opportunistically scavenges on fishery discards (high trophic level prey) (Monteiro et al., 1996b; Bolton et al., 2008); (4) females will forage farthest from the breeding colony, taking more advantage of the wind regimes with their significantly longer wings when compared to males (Nava et al., 2017). The better use of sea winds by seabirds with longer wings is well documented in the literature, as a characteristic assisting a more effortless flight to reach distant foraging regions (e.g. Spear and Ainley, 1997). Ultimately, spatial sexual segregation could lead to sexual differences in trophic ecology, isotopic niche and therefore, the environmental drivers of habitat use (Cleasby et al., 2015; Paiva et al., 2017).
2.1. Study area and field methods
Fieldwork was carried out on Praia Islet (39.059N, 27.954W), an uninhabited 12-ha volcanic islet lying 1 km east of Graciosa Island in the Azores archipelago, from May to mid-August 2013. Monteiro’s storm-petrels breed between late April and September. Incubation extends until early August. The first chicks hatch in early June and the latest-hatched chicks fledge in late September (Bolton et al. 2008).
Twenty-five adults were captured overnight by hand at their nests, and their body mass, tarsus and wing length were measured. A global location sensor (GLS loggers; Intigeo W65A9RJ, Migrate Technology, UK) weighing 0.65 g (i.e. 1.5% of body mass) was attached to the tail feathers using waterproof tape (Tesa AG, Germany). Tag attachment took less than 10 minutes and thereafter birds were returned immediately to the nest. This tracking procedure was performed during the incubation and chick-rearing phase, with similar tracking duration and number of birds tracked between phases (see Table 1). After device recovery, birds were weighed to 0.5 g using a Pesola spring balance and a tip (∼1 cm) of the innermost primary (P1) and eighth secondary (S8) feathers were collected for stable isotopes analyses. Sexing of all individuals was already known from previous work with the species, and was determined following molecular methods. (e.g. Bolton et al., 2008; Robert et al., 2012).
|Foraging trip characteristics||Females||Males||Females||Males|
|Tracking period||21 Jun–4 Jul||21 Jun–8 Jul||24 Jul–2 Aug||9 Jul–2 Aug|
|N birds tracked||7||8||5||5|
|Deployment duration (d)||9 ± 3||12 ± 2||12 ± 5||15 ± 3|
|Maximum distance from colony (km)||479 ± 201||428 ± 254||493 ± 227||599 ± 287|
|Maximum longitude (°)||-31.2 ± 17.4||-29.9 ± 11.3||-24.8 ± 10.9||-24.3 ± 16.1|
|Maximum latitude (°)||41.6 ± 4.2||39.1 ± 7.0||42.9 ± 2.9||38.9 ± 4.3|
|FR overlaps between sexes and within the same breeding stage||41.7 ± 4.6||62.9 ± 5.9|
|FR overlaps within the same sex and between breeding stages||33.1 ± 4.4|
|Habitat of foraging regions (within FR)|
|Bathymetry (BAT; m)||1194 ± 336||2199 ± 392||2166 ± 801||2412 ± 882|
|Sea surface temperature (SST; °C)||18.9 ± 1.0||21.2 ± 1.2||18.4 ± 0.6||17.1 ± 0.9|
|Chlorophyll a concentration (CHL; g m−3)||0.4 ± 0.11||0.2 ± 0.08||0.5 ± 0.18||0.7 ± 0.1|
|Wind speed (WSP; m s−1)||3.3 ± 0.3||1.6 ± 0.4||3.1 ± 0.5||1.3 ± 0.6|
2.2. Data processing and spatial analysis
Light data were analysed using IntiProc v1.02 (Migrate Technology, UK) and the Geolight R package (R Core Team, 2016) to generate latitudes and longitudes from pairs of dawn/dusk events. We assumed a sun elevation angle of −3.5° based on known positions obtained during calibration of the loggers carried out at the colony before and after each deployment. All estimated locations were examined visually in a geographical information system (Arc GIS 10.1) and any unrealistic positions (i.e. positions estimated from sunrise or sunset curves with some shady portions, N = 8) were excluded from further analysis. Retained geolocations of each bird were examined using the adehabitatHR R package (Calenge, 2006) generating kernel utilization distribution (kernel UD) estimates with a smoothing parameter (h) of 2° and a cell size of 1°. Both the h value and the cell size of the grid were chosen based on the mean accuracy of the devices; i.e. ∼0.93° during incubation and ∼0.77° during chick-rearing, estimated from GLS calibration during the 48 hours preceding and after deployment. We considered the 50% and 95% kernel UD contours to represent the core foraging areas (FR) and the home range (HR), respectively (Hamer et al., 2007). The overlap between kernel FRs (50% kernel UDs) of different (1) sexes and (2) breeding stages were computed to study the spatial segregation within and among groups with the kerneloverlap function and VI method of the adehabitatHR library (Calenge, 2006).
2.3. Environmental variables
To characterize the oceanographic conditions in areas used by the tracked individuals we extracted: (1) bathymetry (BAT, blended ETOPO1 product, 0.03° spatial resolution, m), (2) sea surface temperature (SST, Aqua MODIS NPP, 0.04°, °C), (3) sea surface chlorophyll aconcentration (CHL, Aqua MODIS NPP, 0.04°, mgm−3), and (4) wind speed (WSP, QuickSCAT, 0.12°, ms−1). BAT was downloaded from http://ngdc.noaa.gov/mgg/global/global.html, SST and CHL were extracted from http://oceancolor.gsfc.nasa.gov, while WSP was downloaded from http://winds.jpl.nasa.gov. Monthly averages were used for the dynamic variables (variables 2–4). All environmental predictors were gathered to a 1° grid cell to match the GLS accuracy.
2.4. Trophic ecology
Feather samples (collected upon logger removal) were processed to determine the carbon (δ13C, 13C/12C) and nitrogen (δ15N, 15N/14N) stable isotopic values. Feather samples were stored in polythene bags until stable isotope analysis (SIA). Isotopic ratios were determined by continuous-flow isotope-ratio mass spectrometry (CF-IRMS) using a Finnigan conflo II interface to a Thermo Delta V S mass spectrometer (Thermo Scientific, Bremen, Germany) coupled to a Flash EA1112 Series elemental analyser. Replicate measurements of secondary isotopic reference material (acetanilide STD, Thermo scientific-PN 338 36700) in every batch, indicating precision < 0.25‰ for both δ13C and δ15N values. In Monteiro’s storm-petrels, the moult of the innermost primaries remiges starts in early August (chick-rearing; Bolton et al., 2008), thus isotopic ratios on the innermost primary feather (P1) should reflect the diet when it is grown at the end of the previous breeding period. The eighth secondary feather (S8), grown at the end of the previous non-breeding period should reflect the trophic choices at that stage (Monteiro et al., 1996b).
2.5. Isotopic niche
The stable isotope bayesian ellipses in R (SIBER; siar package; Parnell et al., 2010) were used to establish the isotopic niche width of each tracked individual (Jackson et al., 2011), based on the isotopic signatures of P1 (representative of the breeding period) and S8 (representative of the non-breeding period) feathers. The Standard Ellipse Area after small sample size correction (SEAC) was used to compare estimated isotopic niches between sexes and the two annual phases (breeding and non-breeding). SEAC, an ellipse that has 40% probability of containing a subsequently sampled datum regardless of sample size, was used to quantify niche width. A Bayesian estimate of the standard ellipse and its area (SEAB) to test whether group 1 is smaller than group 2 (e.g. p, the proportion of ellipses of females that were lower than in males, for 104 replicates; see (Jackson et al., 2011) for more details), and to measure the overlap of the isotopic niches between sexes and the two annual phases (Jackson et al., 2011; Parnell et al., 2013). All the metrics were calculated using standard.ellipse and convexhull functions from the siar package (stable isotope analysis in R; Parnell et al., 2010).
2.6. Statistical analysis
Student’s t-tests were used to assess differences in 1) body mass, 2) tarsus length and 3) wing length between sexes of tracked individuals. The following parameters (response variables) were calculated for each individual during each breeding stage (incubation and chick-rearing periods): (1) mean max. longitude (°), (2) mean max. latitude (°), (3) mean max. distance to colony (m), (4) bathymetry (BAT; m), (5) sea surface temperature (SST; °C), (6) chlorophyll a concentration (CHL; g m−3) and (7) wind speed (WSP; m s−1). To have a better insight on the spatial segregation of each individual we also computed the (8) 50% kernel UD area (km2) and (9) 50% kernel UD overlap (%) (9) among individuals and (10) between breeding stages. Generalized linear models (GLMs), fitted to a quasi-Poisson error distribution were used to test the effect of (1) sex and (2) breeding stage (incubation vschick-rearing) as independent variables on the formerly mentioned 10 dependent variables (Zuur et al., 2007).
We tested the differences of the δ13C and δ15N values for the innermost primary (P1) and secondary eight (S8) feathers of male and female Monteiro’s storm-petrels. For such comparisons, we performed MANOVAs (Wilk’s lambda), followed by factorial ANOVAs, for carbon and nitrogen stable isotope values, with post-hoc Bonferroni pairwise comparisons to identify significant differences between the two feather types and sexes. Linear regressions were used to 1) explore the relationship between both carbon and nitrogen stable isotope values of different feather types of the same individual, and 2) obtain an estimate of long-term consistency in carbon source and trophic level. To estimate consistency in carbon we used the residuals of the relationship between δ13C and δ15N in the same feather, because δ13C has a trophic component (Votier et al., 2010; Ceia et al., 2012).
Throughout the results, all values are presented as the mean ± SD, unless otherwise stated. All statistical analyses were carried out in R (Version 3.01; R Core Team, 2016) using different functions of the MASS package (Venables and Ripley, 2002). Response variables were tested for normality (Q-Q plots) and homogeneity (Cleveland dotplots) before each statistical test and transformed when needed (Zuur et al., 2010). All analyses were performed assuming a significance level of α < 0.05.
3.1. Habitats used by male and female Monteiro’s storm-petrels during breeding
In the birds we tracked, females (N = 12) had significantly longer wings than males (N = 12; 162.3 ± 2.3 mm vs 157.2 ± 2.2 mm; t1,23 = 2.50, p = 0.01). Tarsus length and body mass did not show significant differences between males and females (tarsus length: 23.3 ± 1.7 mm vs 23.3 ± 1.3 mm; body mass: 52.0 ± 12.4 g vs 51.1 ± 12.2 g); t test, t1,23, both p ≥ 0.19). Females foraged at significantly higher latitudes (F3,21 = 4.87, p = 0.01), over colder (F3,21 = 4.10, p = 0.02) and windier (F3,21 = 4.93, p = 0.01) regions during both incubation and chick-rearing when compared to males. During incubation, females also foraged over the shallower Mid-Atlantic ridge (MAR) when compared to males foraging over deeper waters north of the breeding colony (F3,21 = 3.62, p = 0.03; Table 1). The mean (1) maximum distance from colony, (2) maximum longitude, (3) bathymetry and (4) chlorophyll aconcentration did not vary between breeding phases (all p > 0.15), sex (all p > 0.21) and breeding phase x sex interaction (all p > 0.22; Table 1). The between sexes overlap increased nearly 20% from incubation to chick-rearing, with a rather small overall inter-sexual spatial overlap between breeding phases (Table 1).
3.2. Isotopic niche of males and females during breeding
The isotopic signature of Monteiro’s storm-petrels differed between the two feather types (P1, representative of the trophic choices during the previous breeding period and S8, the previous non-breeding stage; MANOVA, Wilk’s λ, F2,81 = 7.2, p = 0.001), sex (MANOVA, Wilk’s λ, F2,81 = 7.2, p = 0.001) and interaction feather x sex (MANOVA, Wilk’s λ, F2,81 = 7.2, p = 0.001). A factorial ANOVA for each stable isotope revealed a significant effect of sex (F3,46 = 6.30, p = 0.001) and interaction feather x sex (F3,46 = 4.27, p = 0.01), but no significant effect of type of feather (F3,46 = 3.27, p = 0.03) on both δ13C and δ15N values. Post-hoc Bonferroni pairwise comparisons (p < 0.05) indicated that the (1) carbon and (2) nitrogen isotopic values of male feathers (P1: δ13C ± SD, −18.1 ± 0.2 and S8: −18.0 ± 0.2) were significantly enriched when compared, respectively, to the carbon and nitrogen isotopic values of female feathers (P1: −18.2 ± 0.1 and S8: −18.2 ± 0.1), and (3) the stable nitrogen isotopic values of males’ S8 feathers (δ15N ± SD, 14.4 ± 0.1) was significantly enriched when compared to the other categories, i.e. females’ P1 (12.9 ± 0.1), females’ S8 (12.9 ± 0.2) and males’ P1 (13.8 ± 0.2) feathers. The overlap on the isotopic niche was very low between the feather P1 of males and feather S8 of females (6.4%), but much higher between the P1 and S8 feathers of males (67.3%) and females (96.9%). The (Fig. 1) niche width of the isotopic signatures of S8 feathers of males was significantly larger than those of both P1 and S8 feathers of females (SEAB: p = 0.01 and p = 0.03, respectively; Fig. 2). There was a significant relationship between P1 and S8 feather carbon (r = 0.65, p = 0.01) and nitrogen (r = 0.69, p = 0.01) stable isotopic values of females. No significant relationship was found for the same regressions applied to the P1 and S8 feathers of male individuals (r < 0.16, p > 0.72).
Our study documented sexual foraging segregation of Monteiro’s storm-petrels during the breeding (tracking data) and non-breeding (stable isotope values) periods. Females may have taken advantage of their longer wings (result confirmed using a larger sample from the same population; Nava et al., 2017) to forage over the shallower Mid-Atlantic ridge (MAR) north of Azores, under colder and windier regimes when compared to males, who mostly exploited northern deep waters comparatively closer to the breeding colony. This divergence could be confirmed through (1) tracking a larger number of male and female birds and (2) extending such tracking to the non-breeding period, to determine if the differences persist outside the breeding stage. Between-sex differences in the spatial distribution were more obvious during the incubation period, with the overlap in their distribution increasing during the chick-rearing phase. There was also an isotopic segregation between sexes both during the previous breeding and the non-breeding stages, with females exhibiting a narrower isotopic niche when compared to males. Females also seemed to be much more consistent in their trophic ecology between their breeding (isotopic ratios of P1 feathers) and non-breeding (isotopic ratios of S8 feathers) periods, preying on more pelagic, lower trophic levelprey than males.
4.1. Environmental driven spatial sexual segregation
Females and males overlapped comparatively more in their foraging distribution during the chick-rearing than during the incubation period. This pattern is likely due to the comparatively more pronounced ‘central-place foraging behaviour’ of breeding birds, which have to commute between patches of enhanced prey availability and their colony, to successfully provision their growing chick (Orians and Pearson, 1979). Despite the increase in the spatial overlap between sexes, females foraged farther from the colony during both incubation and chick rearing, exploiting the colder and windier regions north of Azores, whereas males remained comparatively closer to the breeding colony during the entire breeding period and concentrated their foraging effort (i.e. 50% kernel UD) on warmer and less productive waters surrounding the colony (Paiva et al., 2010). Seven populations of Leach’s storm-petrels Hydrobates leucorhous, breeding on the coast of Canada, exhibited similar foraging ranges, although no significant sexual segregation was reported (Pollet et al., 2014b).
Sexual segregation in foraging may be a strategy to reduce inter-sexual competition through resource partitioning, this has been seen in other sexually dimorphic seabird species, such as northern giant petrels Macronectes halli (González-Solís et al., 2000), black-browed and grey-headed Thalassarche chrysostoma albatrosses (Phillips et al., 2004), and in some sexually monomorphic species, such as Barau’s petrel Pterodroma baraui (Pinet et al., 2012). In Monteiro’s storm-petrels, the longer wings of the females may have produced sexual segregation by facilitating their foraging in the windier areas to the north of the colony. There, females likely took advantage of the wind strength and direction to reduce foraging effort, similar to the behaviour exhibited by albatrosses (Weimerskirch et al., 2012), shearwaters (Yonehara et al., 2016) or gadfly petrel (R. Ramos et al., 2016) species. The distinctive foraging pattern of females might also be a result of the nutrient demands following costs incurred by egg production (Monaghan et al., 1998). Such costs naturally translate into females being in poorer condition than males at the onset of incubation, in line with the Energetic Constraint Hypothesis (ECH), which suggests that relative investment by females and males might differ along the breeding stages (Elliott et al., 2010). Thus, to restore their energy budgets and endure in breeding duties, females might allocate more time to self-provisioning in productive, sometimes remoter, oceanic regions (Ricklefs, 1983; Warham, 1996).
4.2. Isotopic niche and trophic choices of Monteiro’s storm-petrels
In the ocean, the distribution of carbon and nitrogen isotopes varies geographically (Graham et al., 2010; Somes et al., 2010), which shapes the trophic niche of prey inhabiting a specific location (Navarro et al., 2013) and predators feeding on those prey (Quillfeldt et al., 2005). As opposed to large and medium sized seabird species, which feed mostly on cephalopodsand pelagic fish (Alonso et al., 2014; Ramos et al., 2015), storm-petrels were expected to feed mainly on planktonic crustaceans and pelagic fish larvae, i.e. lower trophic level prey (e.g. D’Elbee and Hemery, 1998). However, Monteiro’s storm-petrels exhibited high δ15N values, in line with those reported for Madeiran storm-petrels from Iceland, North Ireland or Canary Islands (Roscales et al., 2011) or Leach’s storm-petrels breeding on the Atlantic coast of Canada (Pollet et al., 2014b). For Leach storm-petrels such range of isotopic values corresponded to a diet based on Myctophidae fish and zooplankton (Hedd and Montevecchi, 2006).
Stable isotopic signatures of feathers from the previous breeding (innermost primary) and non-breeding (eight secondary) periods documented an evident isotopic niche segregation between sexes, with males exhibiting higher δ13C and δ15N values, and larger isotopic niches when compared to females. By contrast, no sexual isotopic segregation was reported in Leach’s storm-petrels breeding in Canada, during either the breeding or non-breeding periods (Hedd and Montevecchi, 2006). In reality, the isotopic sexual segregation we found in Monteiro storm-petrels for the breeding and non-breeding periods, could be mostly related to the spatial sexual segregation and exploitation of areas of contrasting environmental conditions. Indeed, years of poor environmental conditions were responsible for a strong sexual segregation on the spatial and trophic ecology of Cory’s shearwaters, which dissipated during years of ameliorating environmental conditions (Paiva et al., 2017).
4.3. Final considerations
We report seminal information on the at-sea foraging distribution of the Monteiro’s storm-petrel during the breeding period, which constitutes key information for the at-sea conservation of this threatened species. Nonetheless, future research efforts should be focused on: (1) Performing year-round tracking, to map the distribution of both sexes during the non-breeding period. The species might perform trans-equatorial migration, like Leach’s storm-petrels (Pollet et al., 2014a), or remain in the north Atlantic like Audubon’s shearwaters (Paiva et al., 2016). This has obvious implications for the at-sea conservation of this endangered species within national and international waters (Lascelles et al., 2014); (2) Gathering multi-year tracking information to understand the effect of inter-annual environmental variability on the foraging habits and trophic choices of the species and on the sexual segregation in foraging. We know that inter-annual oceanographic stochasticity has a strong effect on nest fidelity (Robert et al., 2014), reproductive performance and survival of the species (Robert et al., 2012, Robert et al., 2015); (3) Tracking the other known population of the species, breeding in Baixo Islet (still Graciosa Island) just 5.5 km south of Praia Islet, to clarify possible inter-colony foraging segregation (Pollet et al., 2014b); (4) Conducting a better assessment of the feeding and trophic ecology of the species using pyrosequencing techniques on faecal samples (Mirra, 2010), to better understand the sexual segregation in the trophic ecology of this species.
V.H.P., V.N., J.B. and M.M. acknowledge the support provided by the Portuguese Foundation for Science and Technology (SFRH/BPD/85024/2012, SFRH/BPD/88914/2012, SFRH/BPD/20291/2004 and SFRH/BPD/47047/2008, respectively). This study benefited from the strategic program of MARE, financed by FCT (MARE–UID/MAR/04292/2013). We thank the ‘Secretaria Regional do Ambiente’ and ‘Parque Natural da Graciosa’ for the unconditional logistical support and permissions to visit Praia Islet and work with Monteiro’s storm-petrel. We thank our colleague Jessica Hey for proof-reading an early version of the manuscript. We also acknowledge the detailed and very useful comments provided by Matt Rayner and two anonymous reviewers, which greatly improved the quality of the manuscript.
Read more:- https://www.microbiologyjournal.org