Skip to main content
Browse Subject Areas

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Isotopic Evidence of a Wide Spectrum of Feeding Strategies in Southern Hemisphere Humpback Whale Baleen Records

  • Pascale Eisenmann ,

    Contributed equally to this work with: Pascale Eisenmann, Brian Fry, Susan Bengtson Nash

    Affiliation Environmental Futures Research Institute, Griffith University, Brisbane QLD 4111, Australia

  • Brian Fry ,

    Contributed equally to this work with: Pascale Eisenmann, Brian Fry, Susan Bengtson Nash

    Affiliation Australian Rivers Institute, Griffith University, Brisbane QLD 4111, Australia

  • Carly Holyoake ,

    ‡ These authors also contributed equally to this work.

    Affiliation Murdoch University, Perth WA 6150, Australia

  • Douglas Coughran ,

    ‡ These authors also contributed equally to this work.

    Affiliation Department of Parks and Wildlife, Kensington WA 6151, Australia

  • Steve Nicol ,

    ‡ These authors also contributed equally to this work.

    Affiliation Institute for Marine and Antarctic Studies, University of Tasmania, Hobart TAS 7000, Australia

  • Susan Bengtson Nash

    Contributed equally to this work with: Pascale Eisenmann, Brian Fry, Susan Bengtson Nash

    Affiliation Environmental Futures Research Institute, Griffith University, Brisbane QLD 4111, Australia


Our current understanding of Southern hemisphere humpback whale (Megaptera novaeangliae) ecology assumes high-fidelity feeding on Antarctic krill in Antarctic waters during summer, followed by fasting during their annual migration to and from equatorial breeding grounds. An increase in the number of reported departures from this feeding/fasting model suggests that the current model may be oversimplified or, alternatively, undergoing contemporary change. Information about the feeding and fasting cycles of the two Australian breeding populations of humpback whales were obtained through stable isotope analysis of baleen plates from stranded adult individuals. Comparison of isotope profiles showed that individuals from the West Australian breeding population strongly adhered to the classical feeding model. By contrast, East Australian population individuals demonstrated greater heterogeneity in their feeding. On a spectrum from exclusive Antarctic feeding to exclusive feeding in temperate waters, three different strategies were assigned and discussed: classical feeders, supplemental feeders, and temperate zone feeders. Diversity in the inter-annual feeding strategies of humpback whales demonstrates the feeding plasticity of the species, but could also be indicative of changing dynamics within the Antarctic sea-ice ecosystem. This study presents the first investigation of trophodynamics in Southern hemisphere humpback whales derived from baleen plates, and further provides the first estimates of baleen plate elongation rates in the species.


Southern hemisphere humpback whales (SHHW) exhibit one of the longest mammalian migration on Earth, travelling close to 10 000 kilometres between their equatorial breeding grounds and Antarctic feeding grounds [13]. The classical feeding ecology model for these populations assumes high-fidelity summer feeding on Antarctic krill (Euphausia superba) followed by fasting throughout the entire migration, with feeding only resumed in Antarctic waters the following summer [47]. Because they are capital breeders, the humpback whales breed, calve and nurse during the migratory fast. As such, the SHHWs rely on the presence of high krill biomass to feed, and to replenish their blubber stores for a successful migration and reproduction.

There are 7 distinct breeding stocks (A-G) of SHHW recognised by the International Whaling Commission (IWC) [8]. The populations migrating along the western and eastern Australian coasts are classified as the D and E1 breeding stocks, respectively (Fig 1) [8]. While uncertainty still surrounds the extent and inter-annual consistency of the Antarctic feeding grounds associated with both of these populations, mark-recapture data evidence led the IWC to allocate feeding areas IV and V (Fig 1) to the D and E1 breeding stocks respectively. Tagging evidence corroborates the significance of area V, which include the Balleny Islands, as a summer feeding ground for the E1 population [9, 10].

Fig 1. Migration, feeding grounds and breeding grounds of the Australian humpback whale populations.

Approximate feeding grounds of the D and E1 populations are the historical stock boundaries Area IV and V, as defined by the International Whaling Commission. Approximate breeding grounds are represented by the dotted areas, and migration routes are indicated by arrows.

The narrow feeding niche of SHHWs places them at an elevated risk of exposure to the detrimental effects of climate-induced environmental change. In a warming climate, the Antarctic cryosphere is expected to undergo dramatic change [11, 12]. Sea-ice represents a critical nursing ground for Antarctic krill larvae, hence a decline in sea-ice cover and duration has been predicted to have a devastating impact on Antarctic krill biomass [1315]. One study has predicted that unmitigated CO2 emissions will result in a complete collapse of Antarctic krill stocks by the year 2300 [16]. Negative impacts to this keystone species carry repercussions for predators within the Antarctic sea-ice ecosystem, including SHHWs. Adverse impacts to the energetic health of capital breeders such as SHHWs have direct implications for population fecundity, survival, and toxicological risk of lipophilic contaminant burdens [1722].

In recent years, departures from the classical feeding model have been observed in several SHHW populations, with feeding along their migration routes reported in breeding populations from Western Africa, East Australia, and both Pacific and Atlantic sides of South America [8, 2329]. A rise in such observations may simply be an artefact of the rapid, post-whaling re-expansion of the Southern Ocean baleen whale stocks [30, 31]. Alternatively, these observations may signal changing dynamics within the Antarctic sea-ice ecosystem [13, 14, 32], or likely, a combination of both. Nonetheless, all scenarios place the classical feeding model of SHHW under scrutiny. New research is required to carefully evaluate whether the classical model is an accurate representation of present day feeding, or in fact an over-simplification of feeding by a highly adaptable species.

Stable isotopes (SI) are regularly used in trophodynamic studies because the isotopic composition of an organism closely reflects the composition of ingested food [33, 34]. Stable isotope measurements of δ13C and δ15N can show seasonal variations in diet and feeding location [35, 36], as exemplified in studies of migratory marine fauna such as shrimp [37, 38], southern right whales [39] and fin whales [40]. Carbon isotopes are typically used to distinguish two geographically distinct food webs. Altabet and Francois [41] showed that surface water δ13C values of particulate organic material are near -22‰ in temperate latitudes but decrease to -25‰ closer to Antarctica. Animals feeding in Antarctic food webs show correspondingly low carbon isotope values [4245] compared to temperate food webs [46, 47]. Nitrogen isotope values more strongly record trophic level and metabolic changes such as fasting [33, 48, 49]. Several studies have shown that fasting introduces variations in δ15N independently of food trophic level, although the variations appear to be species and tissue specific [39, 5053]. To our knowledge, no studies on humpback whale fasting exist, hence no assumptions were made regarding the possible effects of fasting on baleen δ15N values.

Tissues with incremental growth, such as bone, hair or nails retain long-term records of historical feeding patterns [50, 54]. Whale baleen plates are keratinous hair-like structures that grow continuously, although older sections are eventually worn-off. Because keratinous materials are formed from multiple metabolic pools with different turnover times [50, 54], shifts in feeding behaviour are not recorded instantly. The dominant pools in keratin, however, appear to have a fast turnover of one to seven days [50, 54] during which the organism equilibrates with the isotopic values of newly ingested food. Dietary changes have been found to be well resolved in baleen, with the most recently formed section of the plate reflecting the isotopic composition of the diet from the past two weeks [39, 50, 52].

In this study, we investigate the validity of the classical SHHW feeding model through analysis of baleen stable isotope records from twenty stranded adult individuals, with uncertain cause of deaths. We hypothesized that isotopic signals would indicate if and when a change to classical feeding had occurred. This study further estimates elongation rates for baleen plates in SHHW, and provides insight into metabolic patterns associated with the migratory life-history of SHHWs, highlighting isotopic changes associated with fasting.


Ethics statement

The samples were obtained from necropsied whales that stranded on the Australian coast between 1940 and 2015. They were preserved in museum collection until requested for analyses. In the case of the Southern Ocean Persistent Organic Pollutant Program (SOPOPP) collection, the necropsy samples were obtained under Scientific Purposes permit WISP14251214, granted by the QLD Department of Environment and Heritage Protection and permit ENV1710AEC granted by the Griffith University Animal Ethics Committee.

Sample collation

Baleen plates were collected from SHHW animals stranded between 1940 and 2015 (Table 1, S1 Table) along Australian shores. Young animals smaller than 6.4m in length were excluded from analysis as they were likely dependent on maternal provisioning at the time of death. Preliminary analysis of these younger animals has shown relatively constant isotope values (Eisenmann, unpublished data), as expected if calves fed on maternal reserves. If no size information was recorded at the time of stranding, plates with isotopic oscillations indicative of one or more migrational cycles were considered to be of adults. In total, plates from thirteen E1 population whales and seven D population whales were analysed (Table 1, S1 Table).

Baleen plates were collected opportunistically. These plates were of varying length, reflecting their position in the mouth. Population separation of the animals was done according to location of stranding. Baleen plates from the E1 breeding population were obtained from the Museum Victoria, the Australian Museum, the Queensland Museum, the Ceduna Trust museum, the South Australian Museum, and the Southern Ocean Persistent Organic Pollutant Program (SOPOPP) collection at Griffith University (S1 Table). Two plates were further obtained courtesy of the Tasmanian Department of Primary Industries, Parks, Water and Environment. Baleen plates from the D breeding population were provided by Dr. Carly Holyoake and by Mr. Douglas Coughran (S1 Table).

Stable Isotope Analysis

Baleen plates were first washed to remove sand, blood and other particulates that may have accumulated during collection, transportation and storage. The first wash consisted of deionized water and scrubbing with a steel brush to remove coarser contaminants. Complete removal of tissue such as the gum was done along the external edge to allow full access of the emerging baleen hair. The plate was then washed a second time in a 2:1 chloroform:methanol solution to remove any oils and lipids [52, 55]. Plates were oven-dried at 58°C to constant weight. On average, baleen plates stayed in the chloroform:methanol wash for 1 day, followed by 2 days in the drying oven. Those times varied somewhat depending on baleen plate size and collection date.

SIA required 1–2 mg of clean sample material. In the case of keratin-composed baleen plates, using an engraving tool for sampling was sufficient to create enough powder or slivers for analysis. Samples were taken every centimetre along the outer edge of the plate. The outer edge was preferable as the keratin filaments were unbroken and the edge was smooth compared to the inner side, where keratin filaments were loose to provide a “netting” network used to retain prey. Measurements started at the oldest point on the baleen (0 cm), and continued until under the gum line (newest part of the baleen). Powder or a sliver of baleen was weighed into tin capsules for isotope analysis [55].

All stable isotope abundances are calculated in ‰ using the following formula: (1) where X = 13C or 15N, and R = the respective ratio 13C/12C or 15N/14N. The international reference standards for carbon and nitrogen are respectively Vienna Pee Dee Belemnite and N2 in air. International standards IAEA-CH6 for carbon and IAEA N1 for nitrogen were used for calibration of laboratory standards sucrose and (NH4)2SO4. The preparation system was a Europa EA-GSL interfaced to a SERCON Hydra 20–20 isotope ratio mass-spectrometer (IRMS). Based on analysis of replicate standards, the standard deviation for δ13C and δ15N respectively averaged 0.1‰ and 0.15‰.

Source prediction and trophic fractionation

Measured baleen isotope values were compared to literature-derived estimates of three distinct prey sources (combining prey type and feeding location). The three estimates are presented in Table 2 together with corresponding references [4247]. Literature averages represent data accumulated across a range of years, so that temporal variability was at least partially accounted for. We included the standard deviation of all values in the defined prey isotope ranges to account for possible taxonomic and sub-regional variations in δ13C and δ15N (Table 2).

Table 2. Reported isotopic values for possible humpback whale prey items.

In all cases, literature-derived prey estimates were adjusted to account for trophic fractionation and represent the baleen isotopic values of whales consuming these prey (S2 Table). The adjusted estimates are calculated as follow: (2)

Where δX = δ13C or δ15N values, B is baleen, P is prey, X = 13C or 15N, TF is the trophic enrichment factor, and SD is the standard deviation. Two sets of TF estimates were initially compared (S2 Table, Fig 2). The first set corresponds to generic estimates of +3.4‰ for nitrogen and 0.5‰ for carbon. Although δ13C TEF values are known to range from 0.5–1.5‰ [48, 56], the generic estimates given here are the most commonly used for species with unknown TFs. Because a recent study on krill-feeding northern hemisphere fin whales reported TFs of +2.77‰ for nitrogen isotopes and +2.26‰ for carbon isotopes in baleen plates [57], this TF set was also included in the preliminary analysis (Fig 2).

Fig 2. Bivariate comparison of whale and prey isotope values corrected using two sets of TFs.

Each prey source is plotted twice depending on the set of TFs: rectangles a, c, and e represent the generic set of TFs: + 3.4‰ for δ15N and + 0.5‰ for δ13C, while rectangles b, d, and f represent the fin whale-specific set of TFs: +2.77‰ for δ15N and +2.26‰ for δ13C. Sources: Antarctic Krill (dashed line), Australian Krill (full line), average Australian fish species (dash-dot line). Each zone is therefore created using isotopic coordinates consistent with prey values and corrected for trophic fractionation. The whale isotope data should plot within the zone appropriate to the prey consumed by the individual.

The isotopic values from the known diet of SH humpback whales aligned more closely with fin whale-specific TF estimates (+2.77‰ per trophic level for nitrogen and +2.26‰ for carbon) than generic estimates (+3.4‰ per trophic level for nitrogen and +0.5‰ for carbon; Fig 2). The fin whale specific TF set was thus used in our interpretations. As the estimates do not overlap in δ13C or δ15N, they provide distinct isotopic zones specific to the consumption of that prey (Fig 2).

Statistical analyses

Statistical analyses were performed using SPSS for Windows. The whole dataset, comprising each individual observation from every baleen, was tested for normality using a Shapiro-Wilk test and equality of variances using a non-parametric Levene test. Because the distributions and variances for δ13C and δ15N across the populations were not normal (p<0.05), we used a Kruskal-Wallis non-parametric comparison of ranked means, followed by appropriate post-hoc tests to look for differences between feeding categories and between populations.

Feeding categories were assigned according to adherence of the observed individual isotopic profiles to the three distinct feeding estimates.

Results and Discussion

Baleen elongation rates, and annual isotopic features

It has previously been reported that baleen plate isotope records were consistent between plates within a single individual in bowhead whales [58, 59], and gray whales [60]. To confirm that SHHW plates contain the same isotopic patterns and grow at the same rate regardless of mouth location, we analysed paired plates from whales E14 and D01 (S1 Fig). There were minimal isotopic differences between the plates for each individual, with the main difference due to the length of the plates: a shorter plate covered a smaller temporal record than a longer one (D01, S1 Fig). This observation would suggest that the size differences between plates of a single individual are due to the rate at which there are eroded during feeding and not due to variations in elongation rate, as found in other cetacean species.

Recurring δ15N minima (e.g. S1 Fig, arrows) were visible in plates longer than 15cm. These minima were attributed to annual Antarctic summer feeding as justified by three separate lines of evidence. Firstly, the δ15N values for Antarctic krill were the lowest values of all the possible prey items identified (Table 2). It then follows that whales feeding on Antarctic krill during the summer would have low δ15N values in comparison to whales that fed on other prey. Secondly, SHHWs spend only the most productive months (November to March) feeding in Antarctica. By extension, the comparatively shorter periods in δ15N values could therefore be attributed to the intensive feeding periods with the longer periods of elevated isotope values being attributed to the extended fast. Finally, most of the whales show locally higher δ15N values in the most recently formed part of the baleen (closest to stranding time, e.g.: S1 Fig, D01 cm #40–42). As sampled individuals were collected in Australian waters, higher δ15N values can confidently be associated with migration and fasting in whales with low Antarctic δ13C, or in Australian feeding with high δ13C, such as the temperate zone feeders discussed further below.

Overall, the observed general annual pattern was that lower δ15N values were associated with Antarctic feeding, and moderately higher plateauing δ15N values were associated with fasting during the migration. These δ15N cycles were generally regular and interpreted as annual, in line with previous studies [39, 51, 52, 55, 58, 61]. With this understanding of the annual nature of δ15N in SHHWs, the length of baleen between two δ15N minima was used to estimate growth occurring in one year. The average adult humpback whale plate contained three years of data, although two plates from different individuals were in excess of 60 cm in length, corresponding to approximately five years of migration cycles. Both of these individuals were large, likely fully grown, animals (14 m and 17 m in length respectively) and had long baleen plates (Table 1). The baleen elongation rates for adult humpback baleen observed in this study varied from 12–20 cm per year, with rates averaging 16 cm/yr and decreasing to 12 cm/yr in larger and presumably older animals. These adult elongation rates are similar to those of southern right whales [39], fin whales [51] and gray whales [52], that have been shown to average 12–24 cm per year for adults.

The annual oscillations in δ15N likely indicate the alternating use of two differing nitrogen pools, or the use of two different nitrogen metabolism. Two or more isotopically distinct nitrogen pools have been identified in mammals [54], birds [62] and in marine animals such as shrimp and salmon [63, 64], and there is increasing evidence for multiple types of nitrogen metabolism affecting the isotopic values of marine animals [65, 66]. For SHHWs, the δ15N minima likely marks the end of summer Antarctic feeding, during which direct dietary input has supported whale growth and formation of blubber reserves. At the start of the migratory fast, δ15N values rise as whales switch nitrogen pool or metabolism. While the principal cost of fasting is met by the accumulated energy reserves of the blubber triglycerides, lipids lack nitrogen, and a non-critical protein source is required for tissue maintenance and regeneration. Although not yet explicitly identified, this second, non-essential protein pool is used during the normal migratory fast. Use of this pool results in the observed high δ15N plateau after the summer minimum. The decline in δ15N at the end of the plateau marks the resumption of Antarctic feeding.

Heterogeneity in individual feeding records

The baleen isotope records evaluated in this study did not support the null hypothesis that all SHHWs adhere to a high-fidelity Antarctic krill diet interspersed with extended periods of fasting. Instead, individual feeding records fell on a spectrum from those following the classical krill-feeding/fasting model to those shifting to permanent feeding within temperate waters. The three feeding categories, based upon literature-derived prey choices, were termed: 1) Classical feeders, 2) Supplementary feeders, and 3) Temperate zone feeders. Individuals were assigned a feeding category according to visual examination of their isotope profiles, and comparison to the prey consumption zone estimates (S2 Table, Table 3).

Table 3. Means, SD, minima/maxima for δ13C and δ15N for all whales.

We report the means, standard deviations, and minimum/maximum isotope values for each individual as well as assigned feeding categories in Table 3.

Classical feeders.

Classical feeders were defined as individuals with isotopic profiles consistent with continual provisioning (dietary provenance) from a low trophic level, Antarctic prey source. The literature-derived estimates of the isotopic range representing Antarctic krill feeding were calculated to account for possible temporal and geographical variations reflected in δ13C and δ15N. More localised feeding grounds with distinctive isotope values may exist within this broad Antarctic range. Typical classical feeder isotopic profiles are exemplified by annual oscillations in δ15N and stable δ13C values, both confined within the Antarctic range estimates (S2 Fig).

Among classical feeders of the E1 population, localised carbon maxima and minima were visible in E08 (Fig 3, cm #7–15), E24 (Fig 3, cm #15–25) and E26 (Fig 3, cm #13–16), as well as a slow shift from -26‰ to -23‰ and back down to -26‰ over the course of 4 years in whale E10. Similar examples of carbon anomalies in the classical feeders from D population can be seen in D01 (Fig 4, cm #35–42), D12 (Fig 4, cm #5–10, #23–28, #40–44 and #50–63).

Fig 3. Isotope profiles for E1 population classical feeders.

Each horizontal zone is created using isotopic coordinates consistent with prey values and corrected for trophic fractionation. The whale isotope data plots within the zone appropriate to the prey consumed by the individual. Food zones: Antarctic Krill (dashed line), Australian Krill (full line), average Australian fish species (dotted line). Time flows from left to right.

Fig 4. Isotope profiles for D population classical feeders.

Each horizontal zone is created using isotopic coordinates consistent with prey values and corrected for trophic fractionation. The whale isotope data plots within the zone appropriate to the prey consumed by the individual. Food zones: Antarctic Krill (dashed line), Australian Krill (full line), average Australian fish species (dotted line). Time flows from left to right.

An overall rise in δ15N unmatched in δ13C were visible in E10 (Fig 3, cm #45–63), E24 (Fig 3, cm #15–25) and E26 (Fig 3, cm #0–10), as well as D01 (Fig 4, cm #40–42) and D13 (Fig 4, cm #10–13). Increases in δ15N not paired with increases in δ13C may be representative of metabolic changes during fasting when the body shifts from diet-derived pools to body stores [67, 68]. In contrast, the δ13C remains the same, as the geographical location of the original food web from which body reserves were created stays the same. Another explanation for such increases in δ15N decoupled from δ13C could be a shift to a higher trophic level of Antarctic feeding, either by the krill or the whales, but in insufficient quantities to exit the Antarctic δ13C isotope range based on average krill-feeding values.

Unfortunately, plates for whales E08, D11, D13 and D14 were quite short and close to our estimated one-year growth, complicating the identification of annual variations consistent with migration. Finally, the comparative shortness of the last nitrogen isotope minima in whales E10 (Fig 3, cm #37–42) and D01 (Fig 4, cm #55–41) may be caused by a shorter migration period. This would imply that the whales spent less time that year feeding within Antarctic waters before returning to Australia. Such a scenario would result in a lower volume of accumulated energy stores. Other possibilities should not be dismissed however, such as feeding on atypical prey with different nitrogen composition than body reserves [42, 69]. Compound specific isotope analysis [49, 65, 7072] may help resolve these possible trophic-level shifts in future studies of baleen.

Supplementary feeders.

SHHW supplementary feeders are defined as animals that diverged slightly from the classical feeding model, as demonstrated by carbon and nitrogen isotope values occasionally exceeding the Antarctic krill-feeding range (S3 Fig). Principal provisioning could still be attributed to an Antarctic, low trophic level food source because the higher δ13C and δ15N values never reached the temperate water feeding ranges (S3 Fig). The comparatively higher δ13C and δ15N values instead displayed intermediate values between the Antarctic and Australian isotope ranges, and were therefore attributed to supplementary feeding.

Supplementary feeding can be split into two different behaviours, which result in slightly different isotope profiles. The first behaviour is supplementary feeding at a higher trophic level than krill (such as fish) but within Antarctica. This results in δ13C values within the defined carbon Antarctic range and increased δ15N values possibly outside the Antarctic range. Baleen profiles including this kind of supplementary feeding are E12 (Fig 5, cm #9–20), E13 (Fig 5, cm #0–15), E23 (Fig 5, cm #0–7) and E27 (Fig 5, cm #0–10).

Fig 5. Isotope profiles of supplementary feeders from both populations.

Each horizontal zone is created using isotopic coordinates consistent with prey values and corrected for trophic fractionation. The whale isotope data plots within the zone appropriate to the prey consumed by the individual. Food zones: Antarctic Krill (dashed line), Australian Krill (full line), average Australian fish species (dotted line). Time flows from left to right.

The second behaviour corresponds to discrete feeding events in temperate zone waters during the migration, causing isotope values to increase for both δ13C and δ15N at the same time. This occurs in whales D10 (Fig 5, cm #0–3), E03 (Fig 5, cm #5–10), E12 (Fig 5, cm #22–29), E13 (Fig 5, cm #15–22), E14 (Fig 5, cm #8–14) and E27 (Fig 5, cm #11–16), Because energy intake during migration is likely to represent only a fraction of the annual requirements, isotopic composition of the baleen shifts only slightly and may not turn-over completely to temperate zone values.

Temperate zone feeders.

The final category of whales, termed temperate zone feeders, are animals that undertook significant feeding outside of the Antarctic (S4 Fig). Two of the individuals exhibited a slow turnover from Antarctic feeding to Australian feeding (E01 and E18, Fig 6) while the isotopic record from the third individual did not contain any remnants of Antarctic provisioning (E05, Fig 6). These individuals showed the most surprising isotopic signals of the entire sample set with animals seemingly forgoing Antarctic feeding entirely, and possibly remaining most of the year prior to death in Australian waters, feeding on Australian prey.

Fig 6. Isotope profiles for temperate zone feeders.

Each horizontal zone is created using isotopic coordinates consistent with prey values and corrected for trophic fractionation. The whale isotope data plots within the zone appropriate to the prey consumed by the individual. Food zones: Antarctic Krill (dashed line), Australian Krill (full line), average Australian fish species (dotted line). Time flows from left to right.

Unlike the animals discussed earlier, E01 and E18 (Fig 6) show no large cyclic variations indicative of a seasonal shift in prey or feeding location. Instead, there is a gradual directional shift from the Antarctic feeding range to temperate prey (Fig 6). Because both E01 and E18 stranded in southern Victoria, it is possible that the year preceding death was spent around Tasmania instead of returning to Antarctic feeding grounds. Accounts of partial migration [73] in SHHWs populations exist [8, 74], so it is possible that these two whales are Australian examples of non- or partially migrating whales. This is further supported in whale E18 by its stranding record. Although the exact day is unclear, stranding occurred during the austral summer in Victoria (S1 Table), when the whales are expected to be in Antarctic waters replenishing their blubber stores.

E05 (Fig 6) has δ15N and δ13C signals reminiscent of the classical feeding cycles (Figs 3 and 4), but these signals are displaced into the Australian feeding zone. It is possible that this animal is a very extreme example of partial migration, with the baleen record no longer showing any Antarctic signal. Furthermore, the δ15N and δ13C values from whale E05 rise and fall at opposite times. This could be explained if this whale continued the feeding /fasting regime, but never travelled all the way to the Antarctic feeding grounds, instead remaining within Australian waters and feeding on Australian krill species.

It is important to acknowledge that supplementary feeding and partial migrations may not be recent behavioural changes in SHHWs, nor indeed the behaviour of healthy individuals. The 1940 collection date for E01 and the departure from the classical feeding model apparent in its isotope profile (Fig 6) provide evidence of historical feeding outside of Antarctica.

Changes in ocean biochemistry may also be reflected with the 1940 E01 profile. Studies have shown that ocean acidification [69, 75] and low levels of bio-available iron [76] may cause a decrease in δ13C values in the lowest trophic levels of the food web over time [69]. Because anthropogenic releases of CO2 in the atmosphere have been shown to contribute to ocean acidification [69, 75], and the possible effects of intensive whaling on iron fertilisation are being brought to light [76], a decrease in the carbon isotope baseline between 1940 and modern Antarctica is possible. This change would not be accounted for in our prey predictions, because no literature was found reporting 1940 prey isotope values. As a result, our interpretation of E01 as not having fed in Antarctica during the earliest parts of its baleen may be erroneous (Fig 6, cm #0–20).

In summary, of all the individual baleen measurements, 72% were located within the isotopic ranges corresponding to an Antarctic, low-trophic level feeding signal, as expected from the classical feeding model (Fig 2, zone B). For D population individuals, 96% of samples were contained within this isotopic range, while only 59% of E1 samples correspond to classical feeding (S2 Fig). The supplementary feeding category corresponded to 21% of E1 samples and 4% of D samples that were outside the Antarctic range for δ13C and/or δ15N (S3 Fig). Whales identified as temperate zone feeders accounted for 20% of the E1 population observations, and were not found in D population individuals (S4 Fig).

The Kruskal-Wallis test determined that there were significant differences in variance and distribution between the three feeding categories and between the two populations for both δ13C and δ15N (p<0.05). Scheffe’s comparison of means between the different feeding groups showed that there were significant differences in δ15N between all categories but that temperate zone feeders were the only category significantly different in δ13C (Table 4, Fig 7).

Fig 7. Boxplot representation of each individual whale according to feeding category.

Representations of the spread of values for each individual baleen plate, with colours assigned as per feeding category. Boxes represent the 25th and 75th quartiles and are sectioned by the mean. Whiskers represent the 95th quartiles, with outliers as dots.

Table 4. Scheffe comparison of means between the different feeding groups.

This is consistent with the way the groups were formed, because we expect classical and supplementary feeders to be feeding primarily in Antarctica (with the same δ13C baseline) and temperate zone feeders to have consistently higher δ13C values from non-Antarctic feeding. The most unexpected result was that the classical feeding SHHW from the D and E1 populations had significantly different δ15N values (Table 4). The mean δ15N for classical feeders of both populations fit within our Antarctic krill feeding range, hence this observation may indicate some spatial δ15N variability in the average δ15N of Antarctic prey used by the different SHHW populations. Indeed, δ15N values lower than the defined Antarctic range were observed in individuals of both populations (D12 and E10, Table 3). Further work is needed to establish more detailed prey isoscapes for Antarctic waters.

The observation that six out of seven D population animals were assigned to the classical feeding category, compared to four out of thirteen E1 individuals (Table 3) could indicate that the D population is less likely to feed outside the traditional Antarctic zone than the E1 population. Whales from the E1 population may have greater opportunities to engage in supplementary feeding along the migration route when they pass productive waters around New-Zealand and Tasmania [6, 46, 77, 78]. Seamounts have been shown to be of potential importance for the New-Caledonia humpback whale breeding stock [79] and such interactions may exist in the E1 population. It is however acknowledged that stranded individuals are typically biased towards old, young or sick individuals, and hence both sample type and number preclude further discussion of ecological differences between the two breeding stocks.


Stable isotope analysis of SHHW baleen plates from the two Australian breeding populations showed a greater diversity in feeding behaviour than expected. Evidence of feeding in temperate waters and partial migration was observed in half of the individuals studied, pointing to plasticity in both prey selection and migratory behaviour. This plasticity was more evident in the East coast population (nine out of thirteen E1 individuals) than in the West coast population (one out of seven D individuals). Further differences were observed between the classical feeders from the two populations, with significant differences in δ15N pointing at spatial variability between Antarctic feeding locations.

This study provides the first estimate of SHHW baleen elongation rate, which averaged 16 cm/year, and indication of a slowed elongation with age. Trophic fractionation in SHHWs corresponded to fin whale TFs reported by Borrell, Abad-Oliva [57] of +2.77‰ for δ15N and +2.26‰ for δ13C.

Supporting Information

S1 Fig. Isotopic profiles of paired plates from two individual whales.

Two baleen plates were collected from two individual humpback whales. The resulting isotopic profiles were superposed to study reproducibility between plates from a single individual. Approximate migratory cycles are shown, with arrows indicating annual δ15N lows indicative of Antarctic feeding. Time flows left to right. D01: Plates A and B, E14: Plates C and D.


S2 Fig. Bivariate plot of classical feeders.

Each zone is created using isotopic coordinates consistent with prey values and corrected for trophic fractionation. The whale isotope data plots within the zone appropriate to the prey consumed by the individual. Zones: Antarctic Krill (dashed line), Australian Krill (full line), and average Australian fish species (dotted line).


S3 Fig. Bivariate plot of supplementary feeders.

Each zone is created using isotopic coordinates consistent with prey values and corrected for trophic fractionation. The whale isotope data plots within the zone appropriate to the prey consumed by the individual. Zones: Antarctic Krill (dashed line), Australian Krill (full line), and average Australian fish species (dotted line).


S4 Fig. Bivariate plot of temperate zone feeders.

Each zone is created using isotopic coordinates consistent with prey values and corrected for trophic fractionation. The whale isotope data plots within the zone appropriate to the prey consumed by the individual. Zones: Antarctic Krill (dashed line), Australian Krill (full line), and average Australian fish species (dotted line).


S1 Table. Extended sample details.

“MN” is the shorthand for Megaptera novaeanglia, “STRND” for stranded and “EUT” for euthanized. “E1”/”E” and “D” are the respective populations. “QLD”, “WA”, “VIC”, “NSW”, “TAS” and “SA” respectively refer to Queensland, Western Australia, Victoria, New South Wales, Tasmania and South Australia. 1Samples from museum collections have an additional ID number reflecting the museum’s collection ID. All collections are Australian collections. 2Museum Victoria, Carlton VIC. 3Australian Museum, Sydney NSW. 4Southern Ocean Persistent Organic Pollutant Program (SOPOPP), Brisbane QLD. 5Tasmanian Department of Primary Industries, Parks, Water and Environment (DPIPWE), TAS. 6South Australian Museum, Adelaide SA. 7Ceduna National Trust Museum, Ceduna SA. 8Care of Dr Carlysle Holyoake, Murdoch University WA.


S2 Table. Expected isotopic values for prey items.

Average isotopic values (with standard deviation) for prey items liable to be consumed by Australian humpback whale populations. These have not been adjusted for trophic level.



We thank the Museum of Victoria, Australian Museum, Queensland Museum, Ceduna National Trust Museum and South Australian Museum for their loan of SHHW baleen plates from their collections. We are also grateful to the Tasmanian Department of Primary Industries, Parks, Water and Environment for providing us with SHHW baleen plates. We thank Rene Diocares and Rad Bak at the Griffith University Stable Isotope Laboratory for performing the stable isotope analyses. We also thank two anonymous reviewers for their comments and advice regarding this manuscript.

Author Contributions

Conceived and designed the experiments: PE BF SBN SN. Performed the experiments: PE. Analyzed the data: PE BF SBN. Contributed reagents/materials/analysis tools: CH DC. Wrote the paper: PE BF SBN SN CH DC.


  1. 1. Corkeron PJ, Connor RC. Why do baleen whales migrate? Marine Mammal Science. 1999;15(4):1228–45.
  2. 2. Stone G, Florez-Gonzalez L, Katona S. Whale migration record. Nature. 1990;346(6286):705–.
  3. 3. Stevick PT, Neves MC, Johansen F, Engel MH, Allen J, Marcondes MCC, et al. A quarter of a world away: female humpback whale moves 10 000 km between breeding areas. Biology Letters. 2011;7(2):299–302. pmid:20943678
  4. 4. Kawamura A. A review of baleen whale feeding in the Southern Ocean. Report of the International Whaling Commission. 1994;44:261–71.
  5. 5. Murase H, Matsuoka K, Ichii T, Nishiwaki S. Relationship between the distribution of euphausiids and baleen whales in the Antarctic (35°E–145°W). Polar Biology. 2002;25(2):135–45.
  6. 6. Chittleborough R. Dynamics of two populations of the humpback whale, Megaptera novaeangliae (Borowski). Marine and Freshwater Research. 1965;16(1):33–128.
  7. 7. Waugh CA, Nichols PD, Noad MC, Bengtson Nash S. Lipid and fatty acid profiles of migrating Southern hemisphere humpback whales Megaptera novaeangliae. Marine Ecology Progress Series. 2012;471:271–81.
  8. 8. Acevedo J, Haro D, Dalla Rosa L, Aguayo-Lobo A, Hucke-Gaete R, Secchi E, et al. Evidence of spatial structuring of Eastern South Pacific humpback whale feeding grounds. 2013.
  9. 9. Constantine R, Steel D, Allen J, Anderson M, Andrews O, Baker CS, et al. Remote antarctic feeding ground important for East Australian humpback whales. Marine Biology. 2014;161(5):1087–93.
  10. 10. Gales N, Double MC, Robinson S, Jenner C, King E, Gedamke J, et al. Satellite tracking of southbound East Australian humpback whales (Megaptera novaeangliae): challenging the feast or famine model for migrating whales. Journal of Cetacean Research and Management. 2009.
  11. 11. Pörtner H-O. Ecosystem effects of ocean acidification in times of ocean warming: a physiologist’s view. Mar Ecol Prog Ser. 2008;373:203–17.
  12. 12. Swart PK, Greer L, Rosenheim BE, Moses CS, Waite AJ, Winter A, et al. The 13C Suess effect in scleractinian corals mirror changes in the anthropogenic CO2 inventory of the surface oceans. Geophysical Research Letters. 2010;37(5).
  13. 13. Atkinson A, Siegel V, Pakhomov E, Rothery P. Long-term decline in krill stock and increase in salps within the Southern Ocean. Nature. 2004;432(7013):100–3. pmid:15525989
  14. 14. Flores H, Atkinson A, Kawaguchi S, Krafft BA, Milinevsky G, Nicol S, et al. Impact of climate change on Antarctic krill. Marine Ecology Progress Series. 2012;458:1–19.
  15. 15. Loeb V, Siegel V, Holm-Hansen O, Hewitt R, Fraser W, Trivelpiece W, et al. Effects of sea-ice extent and krill or salp dominance on the Antarctic food web. Nature. 1997;387(6636):897–900.
  16. 16. Kawaguchi S, Ishida A, King R, Raymond B, Waller N, Constable A, et al. Risk maps for Antarctic krill under projected Southern Ocean acidification. Nature Clim Change. 2013;3(9):843–7.
  17. 17. Waugh C, Schlabach M, Noad M, Bengtson Nash S. The effect of migration and fasting on organochlorine contaminant burdens in Antarctic humpback whales. Organohalogen Compounds. 2012;74:919–22.
  18. 18. Bengtson Nash S, Dawson A, Burkhard M, Waugh C, Huston W. Detoxification enzyme activities (CYP1A1 and GST) in the skin of humpback whales as a function of organochlorine burdens and migration status. Aquatic Toxicology. 2014;155:207–12. pmid:25058559
  19. 19. Bengtson Nash S, Waugh C, Schlabach M. Mobilisation of persistent organochlorine burdens in migrating Southern Hemisphere humpback whales. 2014.
  20. 20. Bengtson Nash SM, Waugh CA, Schlabach M. Metabolic concentration of lipid soluble organochlorine burdens in the blubber of Southern Hemisphere humpback whales through migration and fasting. Environmental Science & Technology. 2013;47(16):9404–13.
  21. 21. Christiansen F, Víkingsson GA, Rasmussen MH, Lusseau D. Female body condition affects foetal growth in a capital breeding mysticete. Funct Ecol. 2013:n/a-n/a.
  22. 22. Miller CA, Reeb D, Best PB, Knowlton AR, Brown MW, Moore MJ. Blubber thickness in right whales Eubalaena glacialis and Eubalaeba australis related with reproduction, life history status and prey abundance. Marine Ecology Progress Series. 2011;438:267–83.
  23. 23. Danilewicz D, Tavares M, Moreno IB, Ott PH, Trigo CC. Evidence of feeding by the humpback whale (Megaptera novaeangliae) in mid-latitude waters of the Western South Atlantic. Marine Biodiversity Records. 2009;2.
  24. 24. De Sá Alves LCP, Andriolo A, Zerbini AN, Pizzorno JLA, Clapham PJ. Record of feeding by humpback whales (Megaptera novaeangliae) in tropical waters off Brazil. Marine Mammal Science. 2009;25(2):416–9.
  25. 25. Stamation KA, Croft DB, Shaughnessy PD, Waples KA. Observations of humpback whales (Megaptera novaeangliae) feeding during their southward migration along the coast of Southeastern New South Wales, Australia: identification of a possible supplemental feeding ground. Aquatic Mammals. 2007;33(2):165–74.
  26. 26. Stockin KA, Burgess EA. Opportunistic feeding of an adult humpback whale (Megaptera novaeangliae) migrating along the coast of Southeastern Queensland, Australia. Aquatic Mammals. 2005;31(1):120–.
  27. 27. Barendse J, Best PB, Carvalho I, Pomilla C. Mother knows best: occurrence and associations of resighted humpback whales suggest maternally derived fidelity to a Southern Hemisphere coastal feeding ground. PLoS ONE. 2013;8(12).
  28. 28. Barendse J, Best PB, Thornton M, Pomilla C, Carvalho I, Rosenbaum HC. Migration redefined? Seasonality, movements and group composition of humpback whales Megaptera novaeangliae off the west Coast of South Africa. African Journal of Marine Science. 2010;32(1):1–22.
  29. 29. Owen K, Kavanagh AS, Warren JD, Noad MJ, Donnelly D, Goldizen AW, et al. Potential energy gain by whales outside of the Antarctic: prey preferences and consumption rates of migrating humpback whales (Megaptera novaeangliae). Polar Biology. 2016:1–13.
  30. 30. Bruce E, Albright L, Sheehan S, Blewitt M. Distribution patterns of migrating humpback whales (Megaptera novaeangliae) in Jervis Bay, Australia: a spatial analysis using geographical citizen science data. Applied Geography. 2014;54(0):83–95.
  31. 31. Zerbini AN, Clapham PJ, Wade PR. Assessing plausible rates of population growth in humpback whales from life-history data. Marine Biology. 2010;157(6):1225–36.
  32. 32. Gascon V, Werner R. Antarctic krill: A case study on the ecosystem implications of fishing. Puerto Madryn (Argentina): An article prepared for the Lighthouse Foundation, Antarctic and Southern Ocean Coalition, 2005.
  33. 33. DeNiro MJ, Epstein S. Influence of diet on the distribution of nitrogen isotopes in animals. Geochimica et Cosmochimica Acta. 1981;45(3):341–51.
  34. 34. West JB, Bowen GJ, Cerling TE, Ehleringer JR. Stable isotopes as one of nature's ecological recorders. Trends in Ecology & Evolution. 2006;21(7):408–14.
  35. 35. Hobson KA. Tracing origins and migration of wildlife using stable isotopes: a review. Oecologia. 1999;120(3):314–26.
  36. 36. Kelly JF. Stable isotopes of carbon and nitrogen in the study of avian and mammalian trophic ecology. Canadian Journal of Zoology. 2000;78(1):1–27.
  37. 37. Fry B, Mumford PL, Robblee MB. Stable isotope studies of pink shrimp (Farfantepenaeus duorarum Burkenroad) migrations on the southwestern Florida shelf. Bulletin of Marine Science. 1999;65(2):419–30.
  38. 38. Fry B. Natural stable carbon isotope tag traces Texas shrimp migrations. Fish Bull. 1981;79(2):337–45.
  39. 39. Best PB, Schell DM. Stable isotopes in Southern right whale (Eubalaena australis) baleen as indicators of seasonal movements, feeding and growth. Marine Biology. 1996;124(4):483–94.
  40. 40. Bentaleb I, Martin C, Vrac M, Mate B, Mayzaud P, Siret D, et al. Foraging ecology of mediterranean fin whales in a changing environment elucidated by satellite tracking and baleen plate stable isotopes. Marine Ecology Progress Series. 2011;438:285–302.
  41. 41. Altabet MA, Francois R. Sedimentary nitrogen isotopic ratio as a recorder for surface ocean nitrate utilization. Glob Biogeochem Cycle. 1994;8(1):103–16.
  42. 42. Wada E, Terazaki M, Kabaya Y, Nemoto T. 15N and 13C abundances in the Antartic Ocean with emphasis on the biogeochemical structure of the food web. Deep Sea Research Part A Oceanographic Research Papers. 1987;34(5–6):829–41.
  43. 43. Cherel Y. Isotopic niches of emperor and adélie penguins in Adélie Land, Antarctica. Marine Biology. 2008;154(5):813–21.
  44. 44. Hall-Aspland SA, Hall AP, Rogers TL. A new approach to the solution of the linear mixing model for a single isotope: application to the case of an opportunistic predator. Oecologia. 2005;143(1):143–7. pmid:15599768
  45. 45. Hodum PJ, Hobson KA. Trophic relationships among Antarctic fulmarine petrels: insights into dietary overlap and chick provisioning strategies inferred from stable-isotope (15N and 13C) analyses. Marine Ecology Progress Series. 2000;198:273–81.
  46. 46. Davenport SR, Bax NJ. A trophic study of a marine ecosystem off Southeastern Australia using stable isotopes of carbon and nitrogen. Canadian Journal of Fisheries and Aquatic Sciences. 2002;59(3):514–30.
  47. 47. Harris BP, Young JW, Revill AT, Taylor MD. Understanding diel-vertical feeding migrations in zooplankton using bulk carbon and nitrogen stable isotopes. J Plankton Res. 2014.
  48. 48. Fry B. Food web structure on Georges Bank from stable C, N and S isotopic compositions. Limnology and Oceanography. 1988;33(5):1182–90.
  49. 49. Chikaraishi Y, Ogawa NO, Kashiyama Y, Takano Y, Suga H, Tomitani A, et al. Determination of aquatic food-web structure based on compound specific nitrogen isotopic composition of amino acids. Limnology and Oceanography: methods. 2009;7:740–50.
  50. 50. Matthews CJ, Ferguson SH. Seasonal foraging behaviour of Eastern Canada-West Greenland bowhead whales: an assessment of isotopic cycles along baleen. Marine Ecology Progress Series. 2015;522:269–86.
  51. 51. Aguilar A, Giménez J, Gómez–Campos E, Cardona L, Borrell A. δ15N value does not reflect fasting in Mysticetes. PLoS ONE. 2014;9(3):e92288. pmid:24651388
  52. 52. Caraveo-Patiño J, Hobson K, Soto L. Feeding ecology of gray whales inferred from stable-carbon and nitrogen isotopic analysis of baleen plates. Hydrobiologia. 2007;586(1):17–25.
  53. 53. Gómez-Campos E, Borrell A, Aguilar A. Nitrogen and carbon stable isotopes do not reflect nutritional condition in the striped dolphin. Rapid Communications in Mass Spectrometry. 2011;25(9):1343–7. pmid:21488129
  54. 54. Ayliffe LK, Cerling TE, Robinson T, West AG, Sponheimer M, Passey BH, et al. Turnover of carbon isotopes in tail hair and breath CO2 of horses fed an isotopically varied diet. Oecologia. 2004;139(1):11–22. pmid:14730442
  55. 55. Hobson KA, Schell DM. Stable carbon and nitrogen isotope patterns in baleen from Eastern Arctic bowhead whales (Balaena mysticetus). Canadian Journal of Fisheries and Aquatic Sciences. 1998;55(12):2601–7.
  56. 56. Peterson BJ, Fry B. Stable isotopes in ecosystem studies. Annual Review of Ecology and Systematics. 1987;18(1):293–320.
  57. 57. Borrell A, Abad-Oliva N, Gómez-Campos E, Giménez J, Aguilar A. Discrimination of stable isotopes in fin whale tissues and application to diet assessment in cetaceans. Rapid Communications in Mass Spectrometry. 2012;26(14):1596–602. pmid:22693115
  58. 58. Lubetkin S, Zeh J, Rosa C, George J. Age estimation for young bowhead whales (Balaena mysticetus) using annual baleen growth increments. Canadian Journal of Zoology. 2008;86(6):525–38.
  59. 59. Schell DM, Saupe SM, Haubenstock N. Bowhead whale (Balaena mysticetus) growth and feeding as estimated by δ13C techniques. Marine Biology. 1989;103(4):433–43.
  60. 60. Caraveo-Patiño J, Soto LA. Stable carbon isotope ratios for the gray whale (Eschrichtius robustus) in the breeding grounds of Baja California Sur, Mexico. Hydrobiologia. 2005;539(1):99–104.
  61. 61. Schell DM, Saupe SM, Haubenstock N. Natural isotope abundances in bowhead whale (Balaena mysticetus) baleen: markers of aging and habitat usage. Stable isotopes in ecological research: Springer; 1989. p. 260–9.
  62. 62. Bearhop S, Waldron S, Votier Stephen C, Furness Robert W. Factors that influence assimilation rates and fractionation of nitrogen and carbon stable isotopes in avian blood and feathers. Physiological and Biochemical Zoology. 2002;75(5):451–8. pmid:12529846
  63. 63. Fry B, Baltz DM, Benfield MC, Fleeger JW, Gace A, Haas HL, et al. Stable isotope indicators of movement and residency for brown shrimp (Farfantepenaeus aztecus) in coastal Louisiana marshscapes. Estuaries. 2003;26(1):82–97.
  64. 64. Doucett RR, Booth RK, Power G, McKinley RS. Effects of spawning migration on the nutritional status of anadromous Atlantic salmon (Salmo salar): Insights from stable-isotope analysis. Canadian Journal of Fisheries and Aquatic Sciences. 1999;56(11):2172–80.
  65. 65. Germain LR, Koch PL, Harvey J, McCarthy MD. Nitrogen isotope fractionation in amino acids from harbor seals: implications for compound-specific trophic position calculations. Marine Ecology Progress Series. 2013;482:265–77.
  66. 66. Nielsen JM, Popp BN, Winder M. Meta-analysis of amino acid stable nitrogen isotope ratios for estimating trophic position in marine organisms. Oecologia. 2015;178(3):631–42. pmid:25843809
  67. 67. Bowes RE, Lafferty MH, Thorp JH. Less means more: nutrient stress leads to higher δ15N ratios in fish. Freshwater Biology. 2014;59(9):1926–31.
  68. 68. Cherel Y, Hobson KA, Bailleul F, Groscolas R. Nutrition, physiology and stable isotopes: new information from fasting and molting penguins. Ecology. 2005;86(11):2881–8.
  69. 69. Trull TW, Davies D, Casciotti K. Insights into nutrient assimilation and export in naturally iron-fertilized waters of the Southern Ocean from nitrogen, carbon and oxygen isotopes. Deep Sea Research Part II: Topical Studies in Oceanography. 2008;55(5–7):820–40.
  70. 70. McMahon KW, Thorrold SR, Houghton LA, Berumen ML. Tracing carbon flow through coral reef food webs using a compound-specific stable isotope approach. Oecologia. 2015:1–13.
  71. 71. Popp BN, Graham BS, Olson RJ, Hannides CCS, Lott MJ, López‐Ibarra GA, et al. Insight into the Trophic Ecology of Yellowfin Tuna, Thunnus albacares, from Compound Specific Nitrogen Isotope Analysis of Proteinaceous Amino Acids. In: Todd ED, Rolf TWS, editors. Terrestrial Ecology. Volume 1: Elsevier; 2007. p. 173–90.
  72. 72. Seminoff JA, Benson SR, Arthur KE, Eguchi T, Dutton PH, Tapilatu RF, et al. Stable isotope tracking of endangered sea turtles: validation with satellite telemetry and δ15N analysis of amino acids. PLoS ONE. 2012;7(5).
  73. 73. Chapman BB, Brönmark C, Nilsson J-Å, Hansson L-A. The ecology and evolution of partial migration. Oikos. 2011;120(12):1764–75.
  74. 74. Findlay KP, Best PB. Summer incidence of humpback whales on the west coast of South Africa. South African Journal of Marine Science. 1995;15(1):279–82.
  75. 75. Wei G, McCulloch MT, Mortimer G, Deng W, Xie L. Evidence for ocean acidification in the Great Barrier Reef of Australia. Geochimica et Cosmochimica Acta. 2009;73(8):2332–46.
  76. 76. Ratnarajah L, Bowie AR, Lannuzel D, Meiners KM, Nicol S. The biogeochemical role of baleen whales and krill in Southern Ocean nutrient cycling. PLoS ONE. 2014;9(12):e114067. pmid:25469984
  77. 77. O’Driscoll RL, McClatchie S. Spatial distribution of planktivorous fish schools in relation to krill abundance and local hydrography off Otago, New Zealand. Deep Sea Research Part II: Topical Studies in Oceanography. 1998;45(7):1295–325.
  78. 78. Young JW, Jordan AR, Bobbi C, Johannes RE, Haskard K, Pullen G. Seasonal and interannual variability in krill (Nyctiphanes australis) stocks and their relationship to the fishery for jack mackerel (Trachurus declivis) off Eastern Tasmania, Australia. Marine Biology. 1993;116(1):9–18.
  79. 79. Garrigue C, Clapham PJ, Geyer Y, Kennedy AS, Zerbini AN. Satellite tracking reveals novel migratory patterns and the importance of seamounts for endangered South Pacific humpback whales. Royal Society Open Science. 2015;2(11).