Results and discussion
Navigating thousands of kilometers to a target location through a three-dimensional ocean is among the most impressive feats in nature and has important implications for the evolution, ecology, and conservation of many marine species.
Sharks, skates, and rays, from the subclass Elasmobranchii, are among the most ecologically important groups of marine fishes. Many species of elasmobranchs are highly mobile and their habitats can span thousands of kilometers,
,with some migratory species exhibiting site fidelity, in which individuals return to specific locations.
,Researchers have long known that elasmobranchs are sensitive to electromagnetic fields, and the possibility that sharks use their electrosensory organs in some capacity to glean information from Earth’s magnetic field (hereby referred to as the geomagnetic field [GMF]) for navigational purposes has been widely discussed.
,The GMF provides animals with both map and compass information.
, ,The map allows animals to garner spatial information relative to their location,
while the compass allows animals to maintain a directed heading,
and together, these facilitate successful migrations toward targeted locations.
, ,Elasmobranchs appear capable of discriminating between different components of the GMF
and have also been trained to respond to geomagnetic polarity and intensity.
, , ,Tracking studies of wild sharks have revealed striking associations between swimming trajectory and local magnetic maxima and minima extending from seamounts to feeding grounds;
however, whether sharks use geomagnetic cues for navigation remains unresolved. Our first aim was to experimentally determine whether sharks use magnetic cues to derive spatial information for orientation. Our second goal was to determine whether map-like use of the GMF could help explain spatial patterns of genetic variation in sharks. For both, we studied the bonnethead (
Sphyrna tiburo), a widely distributed, coastal shark that displays site fidelity to particular estuaries, bays, and sounds.
We captured 20 juvenile bonnetheads from Turkey Point Shoal off the coast of Florida, USA, in the Gulf of Mexico (29.887°N, 84.511°W;
Table S1). Sharks were transported to the Florida State University Coastal and Marine lab for experimentation (29.916°N, 84.511°W). We used “magnetic displacements” to expose animals to magnetic conditions representing locations hundreds of kilometers away from their capture location. The experimental approach is straightforward and allows specific predictions to be tested about how magnetic map information is used in orientation.
, ,The manipulation of local magnetic fields was accomplished with Merritt coils, organized as two orthogonal series of horizontal and vertical lumber frames
(
Figure 1). An experimental tank was positioned in the center of the coils and a GoPro camera recorded shark movements from above. Each shark was tested in three fields, presented in randomized order: (1) the field at the capture site as a control, (2) a field that exists ∼600 km south of the capture site within the Gulf of Mexico (weaker magnetic intensity and decreased inclination versus control), and (3) a field that exists ∼600 km north of the capture within the continental United States (stronger magnetic intensity and increased inclination versus control;
Table 1). If sharks derive positional information from the GMF, then we predicted northward orientation in the southern magnetic field and southward orientation in the northern magnetic field (in each case to compensate for the perceived displacement), but no orientation preference in the magnetic field at the capture site. This design was chosen in part because of the geographic constraints of the study area, but also to explore whether sharks respond more robustly to changes in magnetic field conditions that are relevant from an ecological/evolutionary perspective (i.e., the southern field) or whether they are equally adept at extrapolating magnetic information in an unnatural situation (i.e., the northern field).
Figure 1Merritt coil systems and shark tracking procedure
(A) Our series of Merritt coils with the experimental tank in the center.
(B) A sample from our video analysis in which this shark has been tracked through 4 s. The O2 aeration can be seen at the tank’s center.
Table 1Target location where sharks were magnetically displaced, and the associated synthetic magnetic fields created in our laboratory trials during the 2 study years
Due to the gradual drift of the geomagnetic field, we slightly modified intensity and inclination between study years to maintain continuity in relative differences between treatments. A statistically significant homeward orientation was observed for the southern treatment. n = 20 for all treatments.
The orientation of sharks tested in the capture site field (control) was indistinguishable from random (mean bearing = 186°; Rayleigh test:
r= 0.055, p = 0.942, n = 20), indicating that the testing procedure did not cause an obvious directional bias in the sharks (
Figure 2). When sharks were exposed to the southern magnetic field, orientation was significantly northward (mean bearing = 347°; Rayleigh test:
r= 0.406, p = 0.035, n = 20;
Figure 2). A paired Hotelling’s test indicated a significant difference between the orientation of sharks in the control and the southern field (
F= 5.835, p = 0.011). By contrast, when exposed to the northern field that exists in the continental United States, sharks were not oriented (mean bearing = 322°; Rayleigh test:
r= 0.221, p = 0.380, n = 20), and a paired Hotelling’s test found no difference between the control and northern field (
F= 1.055, p = 0.369). Details on the behavioral analysis can be found in the
STAR Methodssection.
Figure 2Orientation of bonnetheads to magnetic displacements
Orientation of bonnetheads in the 3 magnetic treatments (white stars). Adjacent rose diagrams detail individual headings of each shark (gray circles, n = 20) in the corresponding magnetic fields. Significant, homeward orientation was elicited by the southern magnetic field. The shaded area represents the 95% confidence interval (CI) and the outer triangle represents mean bearing (347°). No orientation preference was elicited in the control field (capture site) or the northern field that exists outside the range of the bonnethead. The intensity of the total magnetic field is represented by the color bar, and the inclination is represented by 2° contours.
These results suggest that sharks can differentiate geographic locations using map information from the GMF. Bonnetheads appeared to perceive the southern magnetic field as different from the field at the capture site and responded to the magnetic displacement with homeward orientation. It is tempting to speculate that the northern field did not elicit different orientation from the field at the capture site because the sharks had no experience with such strong magnetic fields and that their magnetic map is “learned.” Sharks in the Gulf of Mexico could learn that fields weaker than those at the capture site indicate more southward locations but would never experience stronger fields than the capture site and thus may not know how to respond to such conditions. However, the lack of response to the northern treatment is also consistent with findings in animals with innate magnetic maps; hatchling loggerhead sea turtles (
Caretta caretta) failed to orient in magnetic fields far outside of their normal migratory route, but were strongly oriented within the typical population range.
While our experiment suggests that magnetic fields that are more familiar (either from individual experience or evolutionary history) elicit more robust orientation responses, further study is required to conclude how bonnetheads derive and extrapolate magnetic map information. Regardless, the lack of response to a northern field does not disqualify the bonnethead from having a magnetic map, as information within a map may be tailored to the specific needs of an organism, and therefore maps may be unique to the spatial ecology of each species.
In this geographic setting, the map of bonnetheads may primarily be used to infer whether or how far south they are from their foraging site. It would be interesting to compare our findings with bonnetheads that are restricted in southward movements (e.g., populations along the Bay of Campeche in the southern Gulf of Mexico) and those not restricted in north-south movements (e.g., populations along the US Atlantic coast). Conducting longitudinal magnetic displacements would further inform what representation of space these sharks derive from the GMF.
, ,Our finding that bonnetheads derive spatial information from geomagnetic cues may have important implications for understanding their current migrations and biogeographic patterns.
, ,One such example is that genetic differences between populations of sharks may be predicted by spatial variation in the GMF.
Population structure can be a function of geographic distance, and in an isolation by distance model, populations will be more diverged if separated by greater distances. Likewise, environmental conditions can affect components of genetic variation, with organisms in disparate habitats experiencing reductions in gene flow.
In addition, if sharks use magnetic maps to home on particular locations, then magnetic differences between sites may be a better predicator of divergence than geographic distance. We explored this hypothesis by comparing genetic distances between bonnetheads sampled at discrete geographic locations in the northwest Atlantic (estimated as
FSTin one published nDNA dataset and Φ
STin three mtDNA datasets) to the percentage of difference in magnetic field values, coastal distance, and difference in mean sea surface temperature between sites (a proxy for environmental distance). Multiple regression and variance partitioning analyses indicated that the combination of these three variables accounted for 42.86% of the variance genetic distance inferred from nuclear DNA (nDNA) and 42.94% of the variance genetic distance inferred from mtDNA. For nDNA, variation partitioning analyses uniquely ascribed 12.58% of the variation to magnetic differences, 17.03% to temperature differences, and 7.74% to the coastal distances between sites. For mtDNA, 15.83% of the variation ascribed to magnetic differences, 1.10% to temperature differences, and −0.60% to coastal distances (
Figure 3;
Table S2).
Figure 3Relationships between genetic structure of bonnetheads in the northwest Atlantic Ocean relative to the geomagnetic field
(A) Sites where genetic samples were obtained for bonnetheads. Circles with crosses, nDNA; circles with x’s, mtDNA. Map conventions as in
Figure 2.
(B) Results of variation partitioning procedures for multiple linear regression in predicting FST values for nDNA (light gray bars) and ΦST values for mtDNA (dark gray bars) based on the maximum percentage of magnetic difference, the mean annual sea surface temperature difference, and the coastal distance between sites.
These findings provide an important test of the hypothesis that genetic structure in populations may be shaped by magnetic-based navigation. Brothers and Lohmann
developed this hypothesis from genetic patterns in mtDNA of female loggerhead sea turtles nesting across the peninsula of Florida. Our study extends this initial work by analyzing nDNA and mtDNA from both sexes of a shark species across a wider geographic area (
Figure 3A). We find that magnetic differences account for more variation in mtDNA than temperature differences or coastal distance, but in nDNA, a similar amount of variation is explained by each of these variables. This result is consistent with an earlier study assessing the genomic diversity of bonnetheads in the eastern Gulf of Mexico and US Atlantic.
Genetic markers putatively under selection were associated with latitude, while neutral markers were more correlated to distance.
The authors speculated that patterns observed at loci under selection were due to site fidelity of breeding females and localized adaptation, while patterns observed in the neutral markers reflected gene flow resultant of nomadic males. It is likely that our observations are due to the magnetic similarities between locations that were originally colonized by females, with philopatry to these locations driving the observed patterns in maternally inherited mtDNA. As with sea turtles, this effect results from Florida’s peninsula, where geographically distant sites can be more magnetically similar than those that are closer.
Over evolutionary timescales, the nomadic tendencies of males, which contributes to patterns of genetic variance in bi-parentally inherited nDNA, likely accounts for coastal distance and temperature difference contributing similarly to magnetic differences in the microsatellite dataset. It is important to note that all three variables are correlated and distinguishing the relative importance of each factor is difficult. We encourage future studies in which geographic sites are not simply sampled opportunistically but specifically chosen so that these three variables show different trends across locations, and their relative contributions to observed genetic population structure can be more clearly assessed.
Even so, our experiment provides evidence that sharks have a magnetic map that is used for orientation and that this ability may contribute to population-level processes. These findings complement recent research that has shown elasmobranchs likely have a polarity-based magnetic compass.
The combination of magnetic map and compass senses would likely be highly adaptive and allow the evolution of complex movement patterns that are a hallmark of elasmobranch life histories. Our results are significant because for 50 years researchers have highlighted the importance of determining whether sharks and rays use the GMF to aid in orientation and navigation.
, , ,Multiple species of elasmobranchs have been shown capable of detecting various components of the magnetic field,
, , , ,and this research provides ecologic context for how these abilities may be used.
The use of magnetic maps appears to be a fundamental tactic of how marine animals migrate,
and we have added evidence that this is also the case for an ecologically important taxonomic group. To date, most studies on magnetic-based navigation in marine animals have relied on species with either a terrestrial or freshwater component of their life cycles (e.g., sea turtles, salmonids, anguillid eels). Our findings suggest that the same sensory basis for navigation extends to fully marine taxa as well. This work points to a solution for a major puzzle in biogeography: how are migratory routes and population structure maintained in marine environments, where few physical barriers limit movements of vagile species? The ability for marine animals to discriminate different oceanic regions using geomagnetic cues is a possible answer.
,Moreover, the importance of magnetic maps in the spatial ecology of animals likely extends well beyond migratory marine taxa. The use of magnetic maps appears to be a widely shared trait in species that occupy a variety of habitats, possess divergent life history strategies, and move over a wide range of spatial scales.
, ,Our work adds to the growing body of literature that the map-like use of the GMF is an evolutionary underpinning for how animals across a variety of taxa successfully derive spatial information from diverse habitats.
Acknowledgments
A special thanks to Tim Tricas and two anonymous reviewers for their constructive feedback. We thank Kyle Newton and James Anderson for their assistance with experimental design. We also thank Pete Klimley for providing valuable feedback in the preparation of the manuscript. Without the assistance of the faculty and staff of the Florida State University Coastal and Marine Laboratory, this work would not have been possible, and we would like to extend our sincerest gratitude. Finally, we thank all of those who assisted with field work, construction of the Merritt coil system, and husbandry of the sharks. This research was funded by Save Our Seas Foundation Small Grant no. 392 , the Aylesworth Foundation Scholarship , the Guy Harvey Foundation Scholarship , and the FSUCML research grant . We also received in-kind donations from Encore Wire Corporation , KORAD Technology , and AlphaLab Inc.
Author contributions
B.A.K., N.F.P., R.D.G., and T.P.M. conceived the experiment and designed the project. All of the authors assisted with the formal analysis and contributed written sections. B.A.K. secured the funding and conducted the field work. N.F.P., B.A.K., D.S.P., and R.D.G. conducted the population structure analyses. The authors are listed in the order of their contributions to the study.
Declaration of interests
The authors declare no competing interests. The scientific results and conclusions, as well as any views or opinions expressed herein, are those of the author(s) and do not necessarily reflect the views of NOAA or the US Department of Commerce.
from Hacker News https://ift.tt/2S2rw4C
No comments:
Post a Comment
Note: Only a member of this blog may post a comment.