(C) PLOS One
This story was originally published by PLOS One and is unaltered.
 . . . . . . . . . .



Penguins exploit tidal currents for efficient navigation and opportunistic foraging [1]

['Richard M. Gunner', 'Department For The Ecology Of Animal Societies', 'Max Planck Institute Of Animal Behaviour', 'Konstanz', 'Flavio Quintana', 'Instituto De Biología De Organismos Marinos', 'Ibiomar', 'Conicet', 'Puerto Madryn', 'Mariano H. Tonini']

Date: 2025-07 

Animals navigating in fluid environments often face forces from wind or water currents that challenge travel efficiency and route accuracy. We investigated how 27 Magellanic penguins (Spheniscus magellanicus) adapt their navigation strategies to return to their colony amid regional tidal ocean currents. Using GPS-enhanced dead-reckoning loggers and high-resolution ocean current data, we reconstructed penguin travel vectors during foraging trips to assess their responses to variable currents during their colony-bound movements. By integrating estimates of energy costs and prey pursuits, we found that birds balanced direct navigation with current-driven drift: in calm currents, they maintained precise line-of-sight headings to their colony. In stronger currents, they aligned their return with lateral flows, which increased travel distance, but at reduced energy costs, and provided them with increased foraging opportunities. Since the lateral tidal currents always reversed direction over the course of return paths, the penguins’ return paths were consistently S-shaped but still resulted in the birds returning efficiently to their colonies. These findings suggest that Magellanic penguins can sense current drift and use it to enhance energy efficiency by maintaining overall directional accuracy while capitalizing on foraging opportunities.

We used GPS-enabled dead-reckoning on breeding Magellanic Penguins Spheniscus magellanicus, central-place foragers operating out of sight of land during the breeding season [ 21 – 24 ], to study their abilities to return to their nests after foraging while exposed to considerable tide-driven current drift [ 25 ]. Dead-reckoning integrates compass-derived headings and modeled swimming speeds to reconstruct detailed movement paths of the penguins [ 26 ]. This approach provided explicit information on how penguins adjusted their navigation strategies in response to variable ocean current conditions. By integrating estimated travel vectors (pre-current integration) with a hydrodynamic model of ocean currents, we quantified the complexity of their navigational challenges and reconstructed their responses to both minimal currents (slack water) and strong opposing flows. Additionally, we estimated energy expenditure using established power-speed relationships and analyzed dive profiles to assess prey pursuit behavior during their return journeys. These methods allowed us to explore how penguins balance efficient navigation with opportunistic foraging in a dynamic marine environment.

Animals moving within water or air are subject to external fluid forces in the form of currents that can affect their locomotion, energy expenditure, and ability to navigate (see [ 1 ] for review). These forces lead animals to drift away from heading-based trajectories, which poses navigational challenges for any extensive movements, such as during migration and foraging [ 1 – 5 ]. There is thus strong selective pressure for animals to detect and respond effectively to these environmental vectors [ 6 , 7 ]. Several species have evolved strategies to mitigate the impact of currents by, e.g., timing movements to coincide with favorable conditions [ 8 , 9 ], adjusting speed [ 10 , 11 ], or altering headings to compensate for drift [ 12 – 15 ]. For instance, young frigatebirds have been shown to develop drift-compensation strategies over time [ 16 ]. Although marine animal navigation has been extensively investigated, the specific corrective mechanisms that enable these animals to maintain their course in the absence of visual landmarks or seabed references remain insufficiently characterized [ 17 ]. A standard approach to assessing the effect of drift on animals is to study movements during goal-orientated traveling because the detrimental effects of fluid flow vectors can be examined directly. Such work shows, for example, that sea turtles have appreciable track tortuosity during migrations to their breeding islands [ 5 ]. Similar deviations from supposed optimal (in terms of distance and time) trajectories have been observed in birds [ 18 – 20 ], specifically, that these animals have suboptimal trajectories based on extra distance and prolonged travel time due to drift-induced effects of currents. Although previous studies have evaluated animal travel vectors relative to fluid and ground movement, our understanding of how animals maintain a goal-oriented trajectory under environmental drift has been limited by the lack of high-resolution, simultaneous data on animal headings, external fluid vectors, and travel speed.

A bootstrapped KS test was used to compare penguin heading deviations between outbound and inbound journeys under varying current strengths (‘Slack’ < 0.3 m/s versus ‘Appreciable’ ≥ 0.3 m/s), accounting for sample variability and intra-individual effects ( S3 Text ). Approximately 30% of current data fell under the ‘Slack’ threshold, which was defined as less than 0.3 m/s—about half the mean current strength across all tracks. This threshold is below the penguins’ surface swimming speed, making it a logical boundary relative to their lowest observed travel speeds.

At each point along the DR track during the penguins’ inbound paths, coordinates were advanced based on the speed and direction of either the real penguin travel vector relative to the ground or the fully-compensated travel vector relative to the ground. For each vector, we then computed two variants of ease of transport (cf. Box 1 ):

To evaluate the energetic consequences of real versus fully-compensated travel vectors, we used an inverse of the cost of transport (COT) [cf. 35 ], defined as ‘ease of transport’ (the number of meters travelled per joule of energy (m/J)). This metric—analogous to fuel efficiency (‘miles per gallon’ or ‘km per litter’) used by the vehicle industry—was used to emphasize how much the pathways utilized by penguins facilitated their progression within their moving oceanic landscape. Power requirements for the penguins’ chosen speed were calculated using the formula from Luna-Jorquera and Culik [ 36 ], relating mass-specific power to swim speed for the morphologically and physiologically similar congeneric Humboldt penguins (Spheniscus humboldti) [ 33 ].

This diagram illustrates the primary vectors used in analyzing the penguin’s journey back to its colony. In panel (a) , the ocean current vector (solid blue) combines with either the real penguin travel vector relative to the water (solid red) or the fully-compensated travel vector relative to the water (solid green) at position ‘n.’ These interactions result in the projected real penguin travel vector relative to the ground (dashed purple) or the projected fully-compensated travel vector relative to the ground (dashed cyan) at position ‘n + 1’. The dotted black arrow represents the direct line-of-sight path to the colony. The fully-compensated travel vector relative to the ground is aligned with this line-of-sight path unless the penguin’s speed is insufficient to fully counteract the current. Panel (b) shows a sample of these vectors recalculated every 5 min along a penguin’s southward return journey, based on its dead-reckoned track. The fully-compensated travel vector relative to the water (solid green) sometimes counteracts the ocean current at angles greater than 90° to help the penguin stay on the most efficient path back to the colony.

Penguin horizontal swimming speed was estimated based on their rate of change of depth and body pitch [ 32 ] using speed thresholds for surface swimming, low pitch underwater swimming, and high pitch diving ( S1 Text ). Penguins were dead-reckoned (DR) at 1 Hz resolution using the Gundogs.Tracks() function in R [ 33 ], with corrections applied based on GPS fixes obtained when the birds surfaced. At 1-s intervals, the U and V current components were bilinearly interpolated at nearest hourly timestamps according to the birds’ DR proximity using the interpp() function from the akima package in R [ 34 ]. Outbound and inbound phases of each foraging trip were identified by cumulative changes in the shortest distance between the penguin and the colony, with the start of a continuous downward gradient marking the homing trajectory ( S3 Fig ). Additional details on the ocean current model, horizontal swimming speed estimation, and adjustments for pressure sensor drift and tag orientation discrepancies are provided in S1 Text .

Fieldwork was conducted between 22 November and 1 December 2019 at the San Lorenzo Magellanic penguin colony, Peninsula Valdés, Argentina (42.08° S, 63.86° W). We selected 27 adult Magellanic penguins (Spheniscus magellanicus) brooding small chicks for this study. Each penguin was equipped with a GPS logger (AxyTrek, Technosmart, Italy) and a Daily Diary (DD) logger [ 27 ]. The GPS units recorded positions at 1 Hz, while the DD loggers recorded tri-axial acceleration at 40 Hz and tri-axial magnetometry at 13 Hz, and pressure (indicating depth) at 4 Hz. Both devices were housed in hydrodynamic casings designed to minimize drag [ 28 ]. Combined, the devices weigh 85.7 g, which corresponds to approximately 2.1% of an average 4-kg bird, with published body masses typically ranging from 3.5 to 4.5 kg [ 24 ]. Devices were secured on the dorsal midline with Tesa tape [ 29 ] in a process taking less than 5 min. Penguins completed a single foraging trip before recapture.

Penguins continued to dive to depths exceeding 10 m—presumably capturing prey [ 37 ]—for up to approximately 80% of their return journey ( Figs 4b , S8 , and S10 ). After this point, dive depths decreased steadily, coinciding with fewer prey captures ( Fig 4b ) and more consistent, directed travel toward the colony ( Figs 3a and S7 ). The GAM revealed that deviation of bird vector headings relative to the water from the line-of-sight heading was significantly influenced by individual differences, trip progression, and environmental factors. Penguins adjusted their navigation strategies along the return journey, with significant effects observed for the angular difference from the ocean current, maximum dive depth, and resultant speed after current integration. Additionally, prey pursuits influenced heading deviation, and a complex interaction among ocean current speed, alignment with ocean current direction, and resultant bird vector speed differences also impacted heading deviation ( S3 Text ). Notably, higher prey pursuit rates were positively associated with heading deviation, indicating that birds engaged in more foraging activity tended to deviate further from the colony line-of-sight ( S11 Fig ). Furthermore, birds exhibited more variable headings when aligning their travel vectors with stronger currents, ultimately yielding higher resultant travel speeds ( S11 Fig ).

(a) Mean ease of e of transport (±1 SE, in m/J) is shown across the proportion of the return distance to the colony for the real penguin travel vector in any direction towards the colony (orange), the real penguin travel vector along the line-of-sight to the colony (purple), and the fully-compensated travel vector (cyan). Higher values indicate greater energy efficiency of movement. Significant pairwise differences (p < 0.05) between the real and fully-compensated vectors are marked with solid triangles at 0.05 intervals of the return journey, based on GAM-predicted ease of transport values, which account for individual variability and smooth trends over distance (see e S3 Text for details details). Upward-facing triangles indicate that the real travel vector (orange or purple) had significantly higher ease of transport than the fully-compensated vector (cyan), while downward-facing triangles indicate the opposite. The absence of a triangle indicates no significant difference at that interval. (b) Mean maximum dive depth (±1 SE, in m) and mean prey pursuit rate (±1 SESE, pursuits s − ¹), plotted at 0.05 intervals of the return journey’s proportion of distance travelled. Note that the overall depth decreases over time showing that birds allocate increasing time to horizontal travel. Dive depths decrease to less than 10 m when ca. Eighty percent of the journey is complete (marked by the dashed grey vertical line). This pattern corresponds with substantive changes in the ease of transport and a corresponding reduction in prey pursuit. Importantly, this reduction in dive depth over time was not influenced by the surrounding seafloor depth ( S8 Fig ). The data underlying this figure can be found in https://doi.org/10.6084/m9.figshare.28517873 .

We next investigated whether real versus fully-compensated travel vector heading choices translate into differences in energy efficiency ( Fig 4a ). Penguins achieved higher ease of transport values by moving generally, but not directly, toward the colony, following their real travel vector headings relative to the ground, compared to the theoretical fully-compensated travel vector scenario ( Fig 4a ). For movement along the line-of-sight path, ease of transport was generally slightly lower at the start of the return journey, aligning more closely with the fully-compensated travel vectors in the middle section, a pattern that coincided with increased time spent swimming with ocean currents during this part of the journey ( Fig 4b ).

(a) Mean (±1 SE) angular difference between real penguin travel vector headings and the line-of-sight direction to the colony. Headings are shown relative to the water (before accounting for ocean currents; red) and relative to the ground (after accounting for ocean currents; purple), with fitted loess smooth lines. For a complementary measure that factors in speed (the “Deviation Index ”), see S7 Fig . . . (b) Mean (±1 SE) proportion of time penguins spent swimming against the ocean current. Blue represents periods with “slack water” (<0.3 m/s), while red indicates periods with “appreciable current” (≥0.3 m/s). Time spent swimming against the current is defined as instances where heading angles exceeded 90° in absolute terms. The dashed horizontal line at y = 0.5 marks the 50% threshold, indicating equal time spent swimming with and against the current. In both panels, the x-axis is divided into 0.05 increments of straight-line distance travelled. The data underlying this figure can be found in https://doi.org/10.6084/m9.figshare.28517873 .

We also examined how heading strategies varied along the journey using a distance-based perspective ( Fig 3 ). Variation in penguin travel vector headings around the line-of-sight heading differed notably among individuals and changed non-linearly with distance to the colony ( Fig 3a ). Penguins showed a general trend toward more goal-oriented behavior at the start and near the end of their return journeys, with greater deviations in the middle. This pattern coincided with variations in the proportion of time spent swimming against ocean currents, especially under appreciable current strength ( Fig 3b ). Specifically, the penguins’ travel vector headings relative to the water often deviated from the line-of-sight—indicating active swimming at angles to counteract stronger currents—so that their net travel vector headings relative to the ground remained more closely aligned with the colony’s line-of-sight trajectory ( S7 Fig ).

Before accounting for ocean current effects, the real penguin travel vector headings relative to the water (represented by the red arrow in Fig 1 ) sometimes required the penguins to swim against the ocean current to maintain their line-of-sight heading (dotted black arrow in Fig 1 ) (see S6a Fig ). When ocean current vectors were included, the resultant penguin travel vector headings relative to the ground (purple arrow in Fig 1 ) showed more pronounced deviations from this direct line-of-sight path (see S6b Fig ).

To assess how penguins orient to the colony, we compared their travel vector heading (relative to the water) with the line-of-sight heading during both outbound and inbound phases, under ‘slack water’ and ‘appreciable current’ conditions ( Fig 2c ). A bootstrapped KS test showed that during the inbound phase, penguin travel vector headings relative to the water were more tightly distributed around their line-of-sight heading compared to the outbound phase (D = 0.232, 95% CI: 0.228–0.241; Fig 2c ). The variability of heading strategies was significantly influenced by current strength and trip phase ( Table 1 ). During inbound movement, on average, penguins directed themselves within 25° of the line-of-sight heading 60% of the time and within 45° 80% of the time, rarely deviating more than 90° regardless of current conditions ( Figs 2c and S2 ). Although these results illustrate a strong alignment with the colony’s direction, a separate analysis of the birds’ absolute headings ( S5 Fig ) reveals a broadly bimodal distribution, indicating that penguins do not adhere to a single compass heading per se.

(a) The GPS-corrected dead-reckoned tracks of 27 penguins at sea, colored to distinguish between the outbound (red) and inbound (blue) phases of their foraging trips. The base map was constructed using OpenStreetMap (OSM) tiles—licensed under the Open Data Commons ODbL. (b) A ‘snapshot’ of the current conditions within the grid area, showing peak tidal strength during a single day as a function of the tidal cycle, including the outgoing (‘ebb’ – left panel) and incoming (‘flood’ – right panel) tides. Penguins are often subjected to strong cross-currents during these phases. In both (a and b) , the colony location is marked by a small black filled circle with a white outline. (c) Distribution of heading differences between the penguin’s travel vector heading (relative to water) and the direct line-of-sight heading, shown for Slack Water (<0.3 m/s, blue) and Appreciable Current (≥0.3 m/s, red), and for outbound versus inbound phases. A heading difference of 0° indicates direct movement toward (inbound) or away from (outbound) the colony, whereas ±90° is perpendicular movement. In both ‘Slack Water’ (<0.3 m/s, blue) and ‘Appreciable Current’ (≥0.3 m/s, red) conditions, penguins align more closely with the line-of-sight heading during the return (inbound) phase, suggesting a more focused return to the colony compared to the departure (outbound) phase. The data underlying this figure can be found in https://doi.org/10.6084/m9.figshare.28517873 .

To characterize the overall foraging trajectories, we first examined the paths and durations of the penguins’ trips ( Fig 2 ). Twenty three penguins exhibited ‘looping pathways’ [cf. 38 ], heading approximately northward to forage and returning southward to their colony. Return tracks often displayed an S-shape ( Fig 2a ), being influenced by the prevailing ocean currents which predominantly ran East–West or West–East depending on the phase of the tidal cycle ( Fig 2b ). On average (mean ± SD), the return journeys began 50 ± 15 km (range: 20–75 km) from the colony and took 12 ± 4 hrs to complete, whereas outbound journeys lasted 20 ± 9 hrs. Penguins exposed to cross-currents during both the outbound and inbound phases spent approximately equal time navigating eastward and westward currents. During the outbound phase, they spent 50.7% ± 10.5% of the time in eastward currents and 49.3% ± 10.5% in westward currents [(paired t test: t = −0.366, p = 0.717)]. Similarly, during the inbound phase, they spent 49.3% ± 14.6% of the time in eastward currents and 50.7% ± 14.6% in westward currents [(paired t test: t = 0.237, p = 0.814)]. The average current strength across all tracks was similar between the two phases, averaging 0.59 ± 0.23 m/s during the outbound phase and 0.60 ± 0.20 m/s (max 2.0 m/s) during the inbound phase (cf. S4 Fig ). These represent a significant fraction (mean 29%, max 95%) of normal penguin travelling speed of ca. 2.1 m/s [ 39 , 40 ] and should have a corresponding influence on their trajectories. Birds landed on the coast at minimal distance from their departure point at the colony (0.17 ± 0.18 km, range: 0.01–0.72 km), with 85% of birds returning within 0.3 km. Eight of 27 penguins returned at night (10 PM to 4 AM).

Discussion

Adaptation to current conditions and goal-oriented navigation Magellanic penguins must navigate efficiently over long distances to return to their nests and feed their chicks. Our findings show that they often orient themselves toward the colony by following a general line-of-sight heading (Figs 2, S6 and S12), especially under calm ocean current conditions (Fig 2c). These observations point to a robust, goal-oriented navigation capacity that remains effective even without visible landmarks [41–43]. Given the penguin’s low profile and minimal visibility at the sea surface, visual contact with the colony is unlikely for most of the journey, implying reliance on alternative sensory mechanisms. Indeed, seabirds flexibly use a wide range of cues—including olfaction, magnetoreception, infrasound, sun/star compasses—to orient over medium and long distances [44–48]. In Procellariiform seabirds, olfaction is well documented; for instance, anosmic individuals show impaired homing orientation but unchanged foraging behavior [44], suggesting that smell may be especially critical during homeward phases. Although penguins are not airborne foragers, they also possess a functional sense of smell and may detect odors (e.g., dimethyl sulphide) or colony scents to navigate over open ocean and near land [e.g., 49]. Notably, because Magellanic penguins in our study rarely reach the seabed (S8 Fig), they are unlikely to use benthic cues for perceiving current drift. This contrasts with findings in yellow-eyed penguins [50,51], where seabed features can promote consistent foraging routes. Instead, Magellanic penguins likely rely on other sensory cues to navigate in dynamic marine environments. While geomagnetic cues remain a likely orientation mechanism for many birds [52], mechanoreception—directly sensing water flow or pressure changes—is another possibility [cf. 53]. However, our results indicate that Magellanic penguins may perceive currents indirectly, noticing discrepancies between their intended path (e.g., via compass orientation) and actual displacement over ground, then adjusting accordingly. This interpretation is supported by the differences in heading distributions under slack versus stronger currents (Fig 2c). While penguins still aim broadly toward the colony under strong currents (mean heading remains similar), they exhibit a more dispersed heading distribution, potentially reflecting repeated or fine-scale corrections to compensate for the drift. Such behavior is consistent with effective navigation even when out of sight of land. We also examined how heading strategies varied across the return journey using a distance-based perspective (Fig 3). Penguins generally showed more goal-oriented behavior (relative to the ground) at the beginning and end of their trips, with greater deviations in the middle (Fig 3a). This pattern coincided with the penguins swimming at angles ≥90° to the ocean current, reflecting active drift compensation (Fig 3b). Specifically, while their water-referenced travel vector headings often deviated from the line-of-sight (to counteract lateral currents), the ground-referenced headings generally remained more closely aligned with the colony’s line-of-sight trajectory. In other words, by angling themselves against the current, the penguins minimized net drift and ultimately maintained a more direct path home. This early precision reduces error accumulation over long distances—important because even small deviations at the start of a long return journey could otherwise compound into large errors. In contrast, mid-journey deviations appear when birds are influenced by strong currents although these are subsequently partially corrected as they near the colony. Closer to the colony, penguins again show a strong tendency to follow the line-of-sight, potentially aided by visual cues in the final approach—similar to behaviors observed in shearwaters [54]. However, since approximately 30% of penguins returned at night, visual cues alone may not be essential for homing [cf. 54] although lack of visual cues at this time this may explain the slight deviation from proper line of sight over the last few meters (Fig 3a). Nevertheless, as birds near their nesting site, any deviation from a direct trajectory becomes increasingly costly, reinforcing the importance of precise navigation during these final stages of the journey. Impressively, 85% of penguins returned within 0.3 km of their departure points, matching the accuracy reported by Quintana and colleagues [48]. To illustrate, for an average 50 km journey, a 300 m deviation represents a 99.4% direct return efficiency. Magellanic penguins exhibit strong parallels in their marine and terrestrial navigation strategies, suggesting they may rely on similar orientation mechanisms in both environments. Quintana and colleagues [48] observed that upon returning from the sea, penguins initially traverse a steep, featureless pebble beach with minimal landmarks, following direct paths toward their nests—akin to the line-of-sight headings we observed at sea. A similar pattern occurs in the final 6–7 m before reaching the nest, while mid-journey deviations were more pronounced, likely due to obstacles such as dense vegetation.

Flow-assisted navigation and adaptive heading adjustment for energy efficiency The tidal cycles in the San Matías Gulf expose penguins to alternating east-west currents in a semi-diurnal cycle (approximately 12.4 hrs, S1 Fig), causing lateral drift but also presenting opportunities for current-assisted movement (Figs 2, S2, S13 and S14). Although there are some differences in current speeds—for example, during inbound journeys, eastward currents averaged 0.69 ± 0.24 m/s compared to 0.49 ± 0.18 m/s for westward currents—the tidal cycle ensures that penguins encounter opposing currents in roughly equal measure (S4 Fig). This balance in exposure, even if the speeds are not identical, allows the overall current regime to partially cancel out lateral drift over the course of the journey, reducing the need for continuous active correction by the birds. This mean, that birds in this region can facilitate for lateral drift because it can later be corrected by reversed currents (e.g., Figs S13 and S14). Specifically, they adjust their headings to avoid opposing currents and take advantage of favorable flows, when possible, sometimes angling their path relative to the line-of-sight to maintain forward progress while conserving energy (e.g., Figs 3, S13, S14). This flow-assisted movement behavior [1] is evident in the ease of transport values: penguins’ real travel vectors achieved higher efficiency relative to the ground than the theoretical fully-compensated vectors (which prioritize the most direct movement back to the colony) with birds nonetheless decreasing the distance between themselves and the colony (orange line in Fig 4). By harnessing current strength (Figs 3, S2, S6 and S12), penguins enhanced their travel efficiency in a manner similar to birds using tailwinds during migration [7,55]. Swimming with the current during mid-journey stages reduces drag-related energy costs and conserves energy compared to swimming directly against currents [56–58], which is crucial given the exponential increase in power costs at higher swimming speeds (see (Eq 1)). We suggest that this strategy also allows penguins to search for prey opportunistically [59] because dive data reveal frequent deep dives during the return trip where penguins captured prey (Fig 4b, S10 and S14), even though surface swimming typically offers the most energy-efficient travel for air-breathing marine animals [60,61]. Although these prey pursuits are associated with deviations from a direct line-of-sight heading, the observed heading changes may primarily reflect an opportunistic response to encountering prey rather than a calculated trade-off between energy conservation and foraging gain. Notably, the ease of transport along the line-of-sight to the colony was only slightly higher overall in the theoretical fully-compensated scenario (cyan versus purple lines in Fig 4a), with penguins’ real movements achieving comparable efficiency despite deviating from a direct line-of-sight heading (Figs S6b and S12). In other words, by adjusting their headings to take advantage of favorable currents, penguins can maintain efficient progress toward their goal, even if their path deviates from the most direct route. Whether penguins rely solely on directional cues or also possess an exact distance-based cognitive map remains uncertain and warrants further investigation.

[END]
---
[1] Url: https://journals.plos.org/plosbiology/article?id=10.1371/journal.pbio.3002981

Published and (C) by PLOS One
Content appears here under this condition or license: Creative Commons - Attribution BY 4.0.

via Magical.Fish Gopher News Feeds:
gopher://magical.fish/1/feeds/news/plosone/