- Email: [email protected]
Competitive interactions between artificial lighting and natural cues during seafinding by hatchling marine turtles
BIOLOGICAL CONSERVATION
Biological Conservation 121 (2005) 311–316 www.elsevier.com/locate/biocon
Short communication
Competitive interactions between artificial lighting and natural cues during seafinding by hatchling marine turtles Susan M. Tuxbury, Michael Salmon
*
Department of Biological Sciences, Florida Atlantic University, P.O. Box 3071, 777 Glades Road, Boca Raton, FL 33431, USA Received 3 September 2003; received in revised form 26 January 2004; accepted 30 April 2004
Abstract Artificial lighting disrupts the nocturnal orientation of sea turtle hatchlings as they crawl from their nest to the ocean. Laboratory experiments in an arena were used to simultaneously present artificial light (that attracted the turtles toward ‘‘land’’) and natural cues (a dark silhouette of the dune behind the beach) that promoted ‘‘seaward’’ orientation. Artificial lighting disrupted seaward crawling in the presence of low silhouettes, but not high silhouettes. Low silhouettes provided adequate cues for seaward crawling when the apparent brightness of artificial light was reduced. Based upon these results, we postulate that artificial light disrupts orientation by competing with natural cues. Current restoration practices at nesting beaches emphasize light reduction. However at many sites some lights cannot be modified. Our results suggest that pairing dune restoration (to enhance natural cues) with light reduction (to the extent possible) should significantly improve hatchling orientation, even at nesting beaches where lighting cannot be entirely eliminated. Ó 2004 Elsevier Ltd. All rights reserved. Keywords: Sea turtle; Orientation; Photopollution; Habitat restoration
1. Introduction Hatchling marine turtles emerge from underground nests at night and, within minutes, crawl to the sea. This ‘‘seafinding’’ is directed by two visual cues; light intensity and horizon elevation. Light intensity is an orientation cue mediated by a positive phototaxis (Daniel and Smith, 1947; Mrosovsky, 1972; Verheijen and Wildschut, 1973). Light is reflected from the ocean and absorbed by vegetation behind the beach. Hatchlings crawl away from the dimmer landward horizon and crawl toward the brighter seaward horizon (Mrosovsky, 1967; Mrosovsky and Shettleworth, 1968; van Rhijn and van Gorkom, 1983). Horizon elevation is a second cue (Limpus, 1971; Witherington, 1992a; Salmon et al., 1992). Turtles crawl away from a higher dune and its associated shrubbery, and toward the lower, oceanic
*
Corresponding author. Tel.: +1-115612972747; 115612972749. E-mail address: [email protected] (M. Salmon).
fax:
+1-
0006-3207/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.biocon.2004.04.022
horizon (Parker, 1915; Mrosovsky and Shettleworth, 1968; van Rhijn and van Gorkom, 1983). Hatchling orientation is often abnormal when beaches are exposed to artificial lighting (Verheijen, 1985). Instead of moving toward the sea, turtles may crawl on circuitous paths (‘‘disorientation’’), or they may crawl landward, apparently attracted to the lights (‘‘misorientation’’; Witherington and Martin, 1996). On Florida’s nesting beaches, artificial lighting (hereafter, ‘‘lighting’’) poses a threat to the survival of marine turtles. Thousand of hatchlings that fail to locate the sea perish annually as a consequence of exhaustion, dehydration, or capture by predators (Witherington and Martin, 1996). Abnormalities in seafinding are positively correlated with luminaire ‘‘directivity’’: the contrast in irradiance between light sources and background (Verheijen, 1958). In Florida, habitat restoration is accomplished through ‘‘light management’’ that reduces directivity by turning off unnecessary lights, reducing wattage, and lowering and/or shielding luminaires (Witherington and Martin, 1996). Management reduces the incidence of abnormal hatchling orientation and its consequences. However,
312
S.M. Tuxbury, M. Salmon / Biological Conservation 121 (2005) 311–316
the procedure is most effective where development is modest and few lights are located near the beach. At sites where development is more extensive (many lights, some located inland), it is impossible to modify all of the luminaires. An alternative strategy at such sites might be light management where possible, coupled with dune restoration to increase horizon elevation. This strategy assumes that hatchlings will respond favorably to enhanced natural cues, even when substantial lighting remains. That hypothesis has not been formally tested. On the basis of preliminary observations, we hypothesized that hatchling behavior at illuminated beaches might be the outcome of interactions between lighting and natural orientation cues (the ‘‘cue competition’’ hypothesis). We predicted that the three responses resulting from such a comparison (seafinding, disorientation, or misorientation) might grade, one into another, depending upon different perceived ratios of the ‘‘competing’’ stimuli. If true, then increasing horizon elevation, reducing light directivity, or combining both manipulations might improve seafinding accuracy. To test this hypothesis, we performed experiments in an arena where we could change horizon elevation and lighting directivity while hatchlings were exposed to simulated street lighting.
2. Methods 2.1. Hatchlings Hatchlings loggerheads (Caretta caretta L.) and green turtles (Chelonia mydas L.) were obtained from nests relocated to a ‘‘hatchery’’ at Hillsboro, Broward County, FL, USA (26°180 N, 80°050 W). Hatchlings that would have emerged that evening were collected in the late afternoon and stored separately by nest in light-
proof StyrofoamÒ containers. Containers were placed in a dark, windowless laboratory at 27–30 °C until evening, when tests began. About 15 min before experiments began, hatchlings were removed from their coolers, placed in a large, empty sink, and exposed to lower ambient room temperatures (15–18 °C) and dim lighting to induce locomotor activity (Mrosovsky, 1968). Once the hatchlings became active, experiments began. When they were completed 2–3 h later, the turtles were released at a nearby dark beach. 2.2. Laboratory arena Experiments were performed within a 1.2 m diameter arena (Salmon et al., 1992; Fig. 1). The arena floor was made of rough-textured plywood painted flat black (1.1 m in diameter). The inside wall of the arena was made of a double layer of white styrene screening (1 mm thick, 41 cm high). One half of the arena wall was designated as a landward (180° wide) horizon while the opposite side was designated as a seaward horizon. The center of the landward horizon was designated as 0°; the center of the seaward horizon was 180°. The arena floor was flat except for a center peg, used to attach one end of an 8 cm long monofilament tether. The opposite end was attached to the posterior of the turtle’s carapace with a mini-alligator clip (Fig. 1). The tether allowed each turtle to crawl in any direction, but prevented it from making forward progress. Hatchling orientation to the nearest 5° was estimated using marks on top of the arena. 2.3. Street lighting Two miniature book lights (Model 10050, Zeico, Mt. Vernon, NY) were used as streetlight surrogates (Fig. 1).
Fig. 1. The arena. Left, overhead view showing a tethered hatchling during its crawl. Marks on top of the arena are used to estimate turtle orientation. Right, set-up used to test the effect of background (overhead) illumination.
S.M. Tuxbury, M. Salmon / Biological Conservation 121 (2005) 311–316
Lights were placed against the landward wall, 45° to either side of center. They were adjusted to match typical elevations (35°) and spacing (range: 27–70 m) of streetlights on coastal roadways as viewed from a turtle nest centered between the lights on the beach. Light radiance at the position of a tethered turtle’s head was adjusted with rheostats to duplicate average streetlight radiance (4.6–4.9 1011 photons/cm2 /s) measured from nests at the beach. A radiometer with a uniform response between 400 and 700 nm (model 351A, UDT Instruments, Baltimore, MD) was used to make all light measurements. Each book light was shielded by an opaque cover that directed illumination downward through an amber filter. Identical filters are currently installed on coastal roadway streetlights in Florida that also possess downward-directed (‘‘cut-off’’) shielding. The filter (#2422, General Electric Lighting Systems, Inc., Hendersonville, NC) excludes most but not all of the wavelengths that disrupt hatchling seafinding. 2.4. Natural (silhouette) cues and background illumination Crescent-shaped silhouettes made of flat-black paper were centered against the landward wall of the arena (Fig. 1). From the perspective of a hatchling emerging from its nest, these silhouettes resembled a vegetated dune behind the beach. Four silhouettes were used that differed in center elevation (2°, 4°, 8° and 15°, measured in degrees above the arena platform from the position of a tethered hatchling’s eyes). The higher silhouettes (8° and 15°) were slightly lower than the highest horizon elevations measured from nests at local beaches (up to 20°). A 7.5 W incandescent lamp, suspended above the arena, was used to provide background illumination and to reduce street lighting directivity (Fig. 1). This lighting passed through a 4.0 mm thick 1.22 m2 white plastic sheet supported by a frame. The sheet acted as a diffuser and was suspended horizontally 1 m below the luminaire and 70 cm above the arena platform (Fig. 1). Overhead light intensity was adjusted with a rheostat to match full moon radiance levels at the beach on a clear evening (lunar azimuth between 70° and 80°; 2.6 and 2.7 1010 photons/cm2 /s). 2.5. Procedures and experiments To start each trial, a tethered hatchling was gently placed in the center of the arena and allowed 1 min to select a direction. Orientation was then recorded every 10 s for 100 s before the hatchling was removed and replaced by another turtle. Each turtle began its trial facing 90° to the right of the previous hatchling’s release direction. No turtle was used for more than one trial. Equal numbers of turtles from two or more nests were used in each test.
313
Three arena experiments were conducted. In the first, both book lights were turned on to determine whether hatchlings of each species were attracted to simulated street lighting. In these experiments, no silhouette was present. Experiments concluded when sample sizes were n P 24 hatchlings from two or more nests. In the second experiment, turtles were simultaneously exposed to book lighting and to a landward silhouette. After 3–6 turtles were tested, the silhouette was changed to one of a different elevation so that approximately equal numbers of hatchlings were exposed to each silhouette. Experiments were concluded when sample sizes were equal for all groups (n P 24 hatchlings). In the third experiment, orientation by both species was measured in the presence or absence of background illumination, coupled with a silhouette (4° or 8° in elevation) and street lighting. Responses were measured under three treatment conditions: (i) book lighting and background illumination turned on, (ii) book lighting on and background illumination turned off, and (iii) book lighting off and background illumination turned on. Over several evenings of testing, equal numbers of hatchlings from several nests were exposed to each treatment until sample size (n P 24 hatchlings) was achieved for each treatment. 2.6. Data analysis The 10 vectors recorded for each hatchling were used to calculate its average orientation. Averages for each turtle were then used to determine a second-order group mean angle and r-vector (measure of dispersion) for all of the hatchlings in a single experiment (using standard procedures in circular statistics; Zar, 1999). Rayleigh tests were performed to determine if groups of turtles showed significant orientation (at p 6 0:01). Significant ‘‘landward’’ orientation (movement toward the book lights) was interpreted as misorientation. Significant ‘‘seaward’’ orientation was interpreted as seafinding. The absence of significant orientation was considered disorientation.
3. Results When presented only with street lighting, both loggerheads and green turtles were significantly oriented (Fig. 2a and b). Group mean angles fell between the two book lights. When presented with both street lighting and silhouettes, loggerheads were significant oriented only when exposed to a 15° high silhouette (Fig. 2f). The group mean angle (193°) was toward the center of the seaward horizon. Loggerheads exposed to lower silhouettes (2°, 4°, and 8°) failed to show significant group orientation (Fig. 2(c)–(e).
314
S.M. Tuxbury, M. Salmon / Biological Conservation 121 (2005) 311–316
Fig. 2. Arena orientation. Each filled dot shows a turtle’s mean angle of orientation. Arrow inside the diagram shows the group’s mean angle and 95% confidence interval when significantly oriented. A pair of lights is positioned outside the arena. Top, green turtles (a) and loggerheads (b) exposed to book lighting. (c)–(f), Loggerheads exposed to book lighting and silhouettes elevated between 2° (c) and 15° (f).
In the presence of background (overhead) illumination, both species showed significant seaward orientation (Fig. 3, top and bottom rows). Orientation occurred when the books lights were either turned on (top row) or turned off (bottom row). In the absence of background illumination, neither silhouette (4° for loggerheads; 8°for green turtles) was high enough to induce significant group orientation (middle row, Fig. 3).
4. Discussion Our experiments demonstrate that in the laboratory, we can create the conditions that induce hatchling sea turtles to orient normally (toward the sea) or show the orientation abnormalities that occur when beaches are exposed to artificial lighting. By separately varying horizon elevation (Fig. 2), background illumination, and lighting ‘‘directivity’’ (Fig. 3), we showed that each variable influenced how the turtles respond to artificial lighting. Our results also duplicated some of the changes that occur in nature. For example, hatchlings were unaffected by artificial lighting when exposed to background illumination. Similarly, few hatchlings are affect by artificial
Fig. 3. Orientation shown by loggerheads (left column) and of green turtles (right column) when background illumination was turned on (top and bottom row) or off (middle row). The ‘‘streetlights’’ were turned off in the bottom row.
lighting during full moon (Salmon and Witherington, 1995), an observation also noted for other wildlife such as migrating birds, flying insects, and fishes (Verheijen, 1980, 1981). 4.1. Artificial lighting ‘‘theory’’ and the cue competition hypothesis Why does artificial lighting result in an inability to maintain direction under some conditions, and orientation toward light sources under other conditions? To our knowledge, there have been no studies designed to determine how artificial lighting affects either visual receptors or visual processing areas in the brain. In the absence of these studies, no explanation based upon sensory (or higher-order) mechanisms is available. Verheijen (1982) hypothesized that the extreme directivity of artificial lighting was beyond the range that
S.M. Tuxbury, M. Salmon / Biological Conservation 121 (2005) 311–316
315
Fig. 4. The cue competition hypothesis. Hatchling responses (seafinding, disorientation, or misorientation) depend upon the perceived magnitude of artificial lighting and the natural cues used to locate the sea. These stimuli are represented by paired triangles whose width is proportional to perceived magnitude. Seafinding is normal in the absence of lunar illumination and artificial lighting (lower left), or when the beach is exposed to both (right) because moonlight reduces lighting directivity. Misorientation occurs when the perceived magnitude of artificial lighting is strong compared to the natural cues (upper left). Disorientation occurs when the competing stimuli are similar in perceived magnitude.
could be processed by higher neural circuits. By making detailed, 3-dimensional maps of light fields, he demonstrated that irradiance gradients generated by artificial lighting exceeded those of natural celestial lighting, thereby confirming that the physical differences in directivity were present (Verheijen, 1958). However, Verheijen never specified what properties of artificial lighting were responsible for disorientation or for misorientation, nor did he investigate whether interactions between natural cues and artificial lighting might affect these responses. Our experiments with marine turtles indicate that normal orientation (seafinding), disorientation, and misorientation are graded responses correlated with different magnitudes of co-occurring natural and artificial visual stimuli. On that basis, we postulate that the interaction between them is competitive (Fig. 4). However, the model we present is preliminary and will require further study to confirm and refine. For example, the model suggests that disorientation will occur when lighting and natural cues are perceived as similar in relative magnitude, even when they differ over a wide range in absolute magnitude (Fig. 4). That prediction could be easily tested either in our arena, or in another apparatus where each element of the stimulus complex can be precisely controlled. 4.2. Management implications Artificial lighting continues to pose a worldwide threat to the survival of nocturnally active wildlife, and
to marine turtles (Witherington, 1997; Salmon et al., 2000). In Florida today, conservation efforts center on protecting the few remaining dark beaches where most nesting occurs. However, managing lighting at sites with modest development is also an important goal, not only because such sites are much more common but also because darkening a patch of beach can lead to an increase in nesting at that site (Witherington, 1992b; Salmon et al., 1995b). Loggerheads nesting on Florida’s East Coast may have historically ‘‘spread their risk’’ by dispersing their clutches over a relatively long coastline. But presently, about 90% of all nests are deposited at four beaches that differ from adjacent sites only in relative darkness (Salmon et al., 2000). Habitat restoration is essential to reverse that trend. Our arena experiments demonstrate that if the cues promoting normal orientation are enhanced, seafinding will occur even in the presence of artificial lighting. Thus, restoration may not require control over all sources of artificial lighting. Rather, it might be accomplished by pairing light management (controlling as many lights as possible) with dune restoration to increases silhouette darkness and/or elevation. Thus far, there have been no systematic attempts to investigate how well such a management strategy might work. In an arena study, Salmon et al. (1995a) found that higher, wider, and complete silhouettes placed against an illuminated landward horizon were more effective than lower, narrower, and broken silhouettes at
316
S.M. Tuxbury, M. Salmon / Biological Conservation 121 (2005) 311–316
directing the crawls of hatchlings seaward. In a field study, Witherington et al. (1994) determined that artificial silhouettes placed on an illuminated beach also improved the orientation accuracy of crawling hatchlings. However, neither study combined silhouette enhancement with light reduction. In summary, our data suggest that hatchling orientation at illuminated beaches depends upon interactions between lighting and the cues used naturally by the turtles to locate the sea. These results suggest a new approach to beach restoration that involves both light management and dune modification. While we believe this approach has promise, tests at nesting beaches will be required to confirm its efficacy. In the interim, dark sites must be protected, and continued vigilance will be required to prevent any further degradation of nesting beaches at already developed sites. Acknowledgements This study was completed by S.M.T. as part of the requirements for a Masters degree at Florida Atlantic University. Financial support came from the National Save-the-Sea-Turtle Foundation of Ft. Lauderdale, Florida, the Florida Department of Transportation, the Boca Raton Garden Club and Project A.W.A.R.E. We thank Donna Weyrich Franzmathes and John Aguirre for their help in the laboratory. Bill Irwin’s comments on earlier drafts improved the manuscript. References Daniel, R.S., Smith, K.U., 1947. The sea-approach behavior of the neonate loggerhead turtle (Caretta caretta). Journal of Comparative Physiology and Psychology 40, 413–420. Limpus, C.J., 1971. Sea turtle ocean finding behaviour. Search 2, 385– 387. Mrosovsky, N., 1967. How turtles find the sea. Science Journal 3, 52–57. Mrosovsky, N., 1968. Nocturnal emergence of hatchling sea turtles: control by thermal inhibition of activity. Nature 220, 1338–1339. Mrosovsky, N., 1972. The water-finding ability of sea turtles. Brain Behavior and Evolution 5, 202–225. Mrosovsky, N., Shettleworth, S.J., 1968. Wavelength preferences and brightness cues in the water finding behaviour of sea turtles. Behaviour 32, 211–257. Parker, G.H., 1915. The crawling of young loggerhead turtles toward the sea. Journal of Experimental Zoology 6, 323–331. Salmon, M., Wyneken, J., Fritz, E., Lucas, M., 1992. Seafinding by hatchling sea turtles: role of brightness, silhouette and beach slope as orientation cues. Behaviour 122, 56–77.
Salmon, M., Witherington, B.E., 1995. Artificial lighting and seafinding by loggerhead hatchlings: evidence for lunar modulation. Copeia 4, 931–938. Salmon, M., Reiners, R., Lavin, C., Wyneken, J., 1995a. Behavior of loggerhead sea turtles on an urban beach. I. Correlates of nest placement. Journal of Herpetology 29, 560–567. Salmon, M., Tolbert, M., Pender, D., Goff, M., Reiners, R., 1995b. Behavior of loggerhead sea turtles on an urban beach. II. Hatchling orientation. Journal of Herpetology 29, 568–576. Salmon, M., Witherington, B.E., Elvidge, C., 2000. Artificial lighting and the recovery of sea turtles. In: Pilcher, N., Ismail, G. (Eds.), Sea Turtles of the Indo-Pacific, Research Management and Conservation. Asean Academic Press, Sarawak, Malaysia, pp. 25–34. van Rhijn, F.A., van Gorkom, J.C., 1983. Optic orientation in hatchlings of the sea turtle Chelonia mydas. III. Sea-finding behaviour: the role of photic and visual orientation in animals walking on the spot under laboratory conditions. Marine Behaviour and Physiology 9, 211–228. Verheijen, F.J., 1958. The mechanisms of the trapping effect of artificial light sources upon animals. Les Archives Neerlandaises de Zoologie 13, 1–107. Verheijen, F.J., 1980. The moon: a neglected factor in studies on collisions of nocturnal migrant birds with tall-lighted structures and with aircraft. Die Vogelwarte 30, 305–330. Verheijen, F.J., 1981. Bird kills at lighted man-made structures: not on nights close to a full moon. American Birds 35, 251–254. Verheijen, F.J., 1982. Environmental features and orientation. In: Addink, A.D.F., Spronk, N. (Eds.), Exogenous and Endogenous Influences on Metabolic and Neural Control, vol. I. Pergamon Press, Oxford, pp. 371–382. Verheijen, F.J., 1985. Photopollution: artificial light optic spatial control systems fail to cope with. Incidents, causation, remedies. Experimental Biology 44, 1–18. Verheijen, F.J., Wildschut, J.T., 1973. The photic orientation of hatchling sea turtles during water finding behaviour. Netherlands Journal of Sea Research 7, 53–67. Witherington, B.E., 1992a. Sea-finding behavior and the use of photic orientation cues by hatchling sea turtles. Doctoral Dissertation, University of Florida, Gainesville, FL. Witherington, B.E., 1992b. Behavioral responses of nesting sea turtles to artificial lighting. Herpetologica 48, 31–39. Witherington, B.E., 1997. The problem of photopollution for sea turtles and other nocturnal animals. In: Clemmons, J.R., Buchholz, R. (Eds.), Behavioral Approaches to Conservation in the Wild. Cambridge University Press, Cambridge, England, pp. 303– 328. Witherington, B.E., Bjorndal, K.A., Bolten, A., 1994. An evaluation of the use of dune structure to reduce effects of artificial lighting on hatchling sea turtle sea-finding orientation and offshore orientation. Report submitted to US Air Force, Cape Canaveral Air Force Station and Florida Cooperative Fish and Wildlife Research Unit. Witherington, B.E., Martin, R.E., 1996. Understanding, assessing, and resolving light-pollution problems on sea turtle nesting beaches. Florida Marine Research Institute Technical Report TR-2. Zar, J.H., 1999. Biostatistical Analysis, fourth ed. Prentice-Hall, New Jersey.
Biological Conservation 121 (2005) 311–316 www.elsevier.com/locate/biocon
Short communication
Competitive interactions between artificial lighting and natural cues during seafinding by hatchling marine turtles Susan M. Tuxbury, Michael Salmon
*
Department of Biological Sciences, Florida Atlantic University, P.O. Box 3071, 777 Glades Road, Boca Raton, FL 33431, USA Received 3 September 2003; received in revised form 26 January 2004; accepted 30 April 2004
Abstract Artificial lighting disrupts the nocturnal orientation of sea turtle hatchlings as they crawl from their nest to the ocean. Laboratory experiments in an arena were used to simultaneously present artificial light (that attracted the turtles toward ‘‘land’’) and natural cues (a dark silhouette of the dune behind the beach) that promoted ‘‘seaward’’ orientation. Artificial lighting disrupted seaward crawling in the presence of low silhouettes, but not high silhouettes. Low silhouettes provided adequate cues for seaward crawling when the apparent brightness of artificial light was reduced. Based upon these results, we postulate that artificial light disrupts orientation by competing with natural cues. Current restoration practices at nesting beaches emphasize light reduction. However at many sites some lights cannot be modified. Our results suggest that pairing dune restoration (to enhance natural cues) with light reduction (to the extent possible) should significantly improve hatchling orientation, even at nesting beaches where lighting cannot be entirely eliminated. Ó 2004 Elsevier Ltd. All rights reserved. Keywords: Sea turtle; Orientation; Photopollution; Habitat restoration
1. Introduction Hatchling marine turtles emerge from underground nests at night and, within minutes, crawl to the sea. This ‘‘seafinding’’ is directed by two visual cues; light intensity and horizon elevation. Light intensity is an orientation cue mediated by a positive phototaxis (Daniel and Smith, 1947; Mrosovsky, 1972; Verheijen and Wildschut, 1973). Light is reflected from the ocean and absorbed by vegetation behind the beach. Hatchlings crawl away from the dimmer landward horizon and crawl toward the brighter seaward horizon (Mrosovsky, 1967; Mrosovsky and Shettleworth, 1968; van Rhijn and van Gorkom, 1983). Horizon elevation is a second cue (Limpus, 1971; Witherington, 1992a; Salmon et al., 1992). Turtles crawl away from a higher dune and its associated shrubbery, and toward the lower, oceanic
*
Corresponding author. Tel.: +1-115612972747; 115612972749. E-mail address: [email protected] (M. Salmon).
fax:
+1-
0006-3207/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.biocon.2004.04.022
horizon (Parker, 1915; Mrosovsky and Shettleworth, 1968; van Rhijn and van Gorkom, 1983). Hatchling orientation is often abnormal when beaches are exposed to artificial lighting (Verheijen, 1985). Instead of moving toward the sea, turtles may crawl on circuitous paths (‘‘disorientation’’), or they may crawl landward, apparently attracted to the lights (‘‘misorientation’’; Witherington and Martin, 1996). On Florida’s nesting beaches, artificial lighting (hereafter, ‘‘lighting’’) poses a threat to the survival of marine turtles. Thousand of hatchlings that fail to locate the sea perish annually as a consequence of exhaustion, dehydration, or capture by predators (Witherington and Martin, 1996). Abnormalities in seafinding are positively correlated with luminaire ‘‘directivity’’: the contrast in irradiance between light sources and background (Verheijen, 1958). In Florida, habitat restoration is accomplished through ‘‘light management’’ that reduces directivity by turning off unnecessary lights, reducing wattage, and lowering and/or shielding luminaires (Witherington and Martin, 1996). Management reduces the incidence of abnormal hatchling orientation and its consequences. However,
312
S.M. Tuxbury, M. Salmon / Biological Conservation 121 (2005) 311–316
the procedure is most effective where development is modest and few lights are located near the beach. At sites where development is more extensive (many lights, some located inland), it is impossible to modify all of the luminaires. An alternative strategy at such sites might be light management where possible, coupled with dune restoration to increase horizon elevation. This strategy assumes that hatchlings will respond favorably to enhanced natural cues, even when substantial lighting remains. That hypothesis has not been formally tested. On the basis of preliminary observations, we hypothesized that hatchling behavior at illuminated beaches might be the outcome of interactions between lighting and natural orientation cues (the ‘‘cue competition’’ hypothesis). We predicted that the three responses resulting from such a comparison (seafinding, disorientation, or misorientation) might grade, one into another, depending upon different perceived ratios of the ‘‘competing’’ stimuli. If true, then increasing horizon elevation, reducing light directivity, or combining both manipulations might improve seafinding accuracy. To test this hypothesis, we performed experiments in an arena where we could change horizon elevation and lighting directivity while hatchlings were exposed to simulated street lighting.
2. Methods 2.1. Hatchlings Hatchlings loggerheads (Caretta caretta L.) and green turtles (Chelonia mydas L.) were obtained from nests relocated to a ‘‘hatchery’’ at Hillsboro, Broward County, FL, USA (26°180 N, 80°050 W). Hatchlings that would have emerged that evening were collected in the late afternoon and stored separately by nest in light-
proof StyrofoamÒ containers. Containers were placed in a dark, windowless laboratory at 27–30 °C until evening, when tests began. About 15 min before experiments began, hatchlings were removed from their coolers, placed in a large, empty sink, and exposed to lower ambient room temperatures (15–18 °C) and dim lighting to induce locomotor activity (Mrosovsky, 1968). Once the hatchlings became active, experiments began. When they were completed 2–3 h later, the turtles were released at a nearby dark beach. 2.2. Laboratory arena Experiments were performed within a 1.2 m diameter arena (Salmon et al., 1992; Fig. 1). The arena floor was made of rough-textured plywood painted flat black (1.1 m in diameter). The inside wall of the arena was made of a double layer of white styrene screening (1 mm thick, 41 cm high). One half of the arena wall was designated as a landward (180° wide) horizon while the opposite side was designated as a seaward horizon. The center of the landward horizon was designated as 0°; the center of the seaward horizon was 180°. The arena floor was flat except for a center peg, used to attach one end of an 8 cm long monofilament tether. The opposite end was attached to the posterior of the turtle’s carapace with a mini-alligator clip (Fig. 1). The tether allowed each turtle to crawl in any direction, but prevented it from making forward progress. Hatchling orientation to the nearest 5° was estimated using marks on top of the arena. 2.3. Street lighting Two miniature book lights (Model 10050, Zeico, Mt. Vernon, NY) were used as streetlight surrogates (Fig. 1).
Fig. 1. The arena. Left, overhead view showing a tethered hatchling during its crawl. Marks on top of the arena are used to estimate turtle orientation. Right, set-up used to test the effect of background (overhead) illumination.
S.M. Tuxbury, M. Salmon / Biological Conservation 121 (2005) 311–316
Lights were placed against the landward wall, 45° to either side of center. They were adjusted to match typical elevations (35°) and spacing (range: 27–70 m) of streetlights on coastal roadways as viewed from a turtle nest centered between the lights on the beach. Light radiance at the position of a tethered turtle’s head was adjusted with rheostats to duplicate average streetlight radiance (4.6–4.9 1011 photons/cm2 /s) measured from nests at the beach. A radiometer with a uniform response between 400 and 700 nm (model 351A, UDT Instruments, Baltimore, MD) was used to make all light measurements. Each book light was shielded by an opaque cover that directed illumination downward through an amber filter. Identical filters are currently installed on coastal roadway streetlights in Florida that also possess downward-directed (‘‘cut-off’’) shielding. The filter (#2422, General Electric Lighting Systems, Inc., Hendersonville, NC) excludes most but not all of the wavelengths that disrupt hatchling seafinding. 2.4. Natural (silhouette) cues and background illumination Crescent-shaped silhouettes made of flat-black paper were centered against the landward wall of the arena (Fig. 1). From the perspective of a hatchling emerging from its nest, these silhouettes resembled a vegetated dune behind the beach. Four silhouettes were used that differed in center elevation (2°, 4°, 8° and 15°, measured in degrees above the arena platform from the position of a tethered hatchling’s eyes). The higher silhouettes (8° and 15°) were slightly lower than the highest horizon elevations measured from nests at local beaches (up to 20°). A 7.5 W incandescent lamp, suspended above the arena, was used to provide background illumination and to reduce street lighting directivity (Fig. 1). This lighting passed through a 4.0 mm thick 1.22 m2 white plastic sheet supported by a frame. The sheet acted as a diffuser and was suspended horizontally 1 m below the luminaire and 70 cm above the arena platform (Fig. 1). Overhead light intensity was adjusted with a rheostat to match full moon radiance levels at the beach on a clear evening (lunar azimuth between 70° and 80°; 2.6 and 2.7 1010 photons/cm2 /s). 2.5. Procedures and experiments To start each trial, a tethered hatchling was gently placed in the center of the arena and allowed 1 min to select a direction. Orientation was then recorded every 10 s for 100 s before the hatchling was removed and replaced by another turtle. Each turtle began its trial facing 90° to the right of the previous hatchling’s release direction. No turtle was used for more than one trial. Equal numbers of turtles from two or more nests were used in each test.
313
Three arena experiments were conducted. In the first, both book lights were turned on to determine whether hatchlings of each species were attracted to simulated street lighting. In these experiments, no silhouette was present. Experiments concluded when sample sizes were n P 24 hatchlings from two or more nests. In the second experiment, turtles were simultaneously exposed to book lighting and to a landward silhouette. After 3–6 turtles were tested, the silhouette was changed to one of a different elevation so that approximately equal numbers of hatchlings were exposed to each silhouette. Experiments were concluded when sample sizes were equal for all groups (n P 24 hatchlings). In the third experiment, orientation by both species was measured in the presence or absence of background illumination, coupled with a silhouette (4° or 8° in elevation) and street lighting. Responses were measured under three treatment conditions: (i) book lighting and background illumination turned on, (ii) book lighting on and background illumination turned off, and (iii) book lighting off and background illumination turned on. Over several evenings of testing, equal numbers of hatchlings from several nests were exposed to each treatment until sample size (n P 24 hatchlings) was achieved for each treatment. 2.6. Data analysis The 10 vectors recorded for each hatchling were used to calculate its average orientation. Averages for each turtle were then used to determine a second-order group mean angle and r-vector (measure of dispersion) for all of the hatchlings in a single experiment (using standard procedures in circular statistics; Zar, 1999). Rayleigh tests were performed to determine if groups of turtles showed significant orientation (at p 6 0:01). Significant ‘‘landward’’ orientation (movement toward the book lights) was interpreted as misorientation. Significant ‘‘seaward’’ orientation was interpreted as seafinding. The absence of significant orientation was considered disorientation.
3. Results When presented only with street lighting, both loggerheads and green turtles were significantly oriented (Fig. 2a and b). Group mean angles fell between the two book lights. When presented with both street lighting and silhouettes, loggerheads were significant oriented only when exposed to a 15° high silhouette (Fig. 2f). The group mean angle (193°) was toward the center of the seaward horizon. Loggerheads exposed to lower silhouettes (2°, 4°, and 8°) failed to show significant group orientation (Fig. 2(c)–(e).
314
S.M. Tuxbury, M. Salmon / Biological Conservation 121 (2005) 311–316
Fig. 2. Arena orientation. Each filled dot shows a turtle’s mean angle of orientation. Arrow inside the diagram shows the group’s mean angle and 95% confidence interval when significantly oriented. A pair of lights is positioned outside the arena. Top, green turtles (a) and loggerheads (b) exposed to book lighting. (c)–(f), Loggerheads exposed to book lighting and silhouettes elevated between 2° (c) and 15° (f).
In the presence of background (overhead) illumination, both species showed significant seaward orientation (Fig. 3, top and bottom rows). Orientation occurred when the books lights were either turned on (top row) or turned off (bottom row). In the absence of background illumination, neither silhouette (4° for loggerheads; 8°for green turtles) was high enough to induce significant group orientation (middle row, Fig. 3).
4. Discussion Our experiments demonstrate that in the laboratory, we can create the conditions that induce hatchling sea turtles to orient normally (toward the sea) or show the orientation abnormalities that occur when beaches are exposed to artificial lighting. By separately varying horizon elevation (Fig. 2), background illumination, and lighting ‘‘directivity’’ (Fig. 3), we showed that each variable influenced how the turtles respond to artificial lighting. Our results also duplicated some of the changes that occur in nature. For example, hatchlings were unaffected by artificial lighting when exposed to background illumination. Similarly, few hatchlings are affect by artificial
Fig. 3. Orientation shown by loggerheads (left column) and of green turtles (right column) when background illumination was turned on (top and bottom row) or off (middle row). The ‘‘streetlights’’ were turned off in the bottom row.
lighting during full moon (Salmon and Witherington, 1995), an observation also noted for other wildlife such as migrating birds, flying insects, and fishes (Verheijen, 1980, 1981). 4.1. Artificial lighting ‘‘theory’’ and the cue competition hypothesis Why does artificial lighting result in an inability to maintain direction under some conditions, and orientation toward light sources under other conditions? To our knowledge, there have been no studies designed to determine how artificial lighting affects either visual receptors or visual processing areas in the brain. In the absence of these studies, no explanation based upon sensory (or higher-order) mechanisms is available. Verheijen (1982) hypothesized that the extreme directivity of artificial lighting was beyond the range that
S.M. Tuxbury, M. Salmon / Biological Conservation 121 (2005) 311–316
315
Fig. 4. The cue competition hypothesis. Hatchling responses (seafinding, disorientation, or misorientation) depend upon the perceived magnitude of artificial lighting and the natural cues used to locate the sea. These stimuli are represented by paired triangles whose width is proportional to perceived magnitude. Seafinding is normal in the absence of lunar illumination and artificial lighting (lower left), or when the beach is exposed to both (right) because moonlight reduces lighting directivity. Misorientation occurs when the perceived magnitude of artificial lighting is strong compared to the natural cues (upper left). Disorientation occurs when the competing stimuli are similar in perceived magnitude.
could be processed by higher neural circuits. By making detailed, 3-dimensional maps of light fields, he demonstrated that irradiance gradients generated by artificial lighting exceeded those of natural celestial lighting, thereby confirming that the physical differences in directivity were present (Verheijen, 1958). However, Verheijen never specified what properties of artificial lighting were responsible for disorientation or for misorientation, nor did he investigate whether interactions between natural cues and artificial lighting might affect these responses. Our experiments with marine turtles indicate that normal orientation (seafinding), disorientation, and misorientation are graded responses correlated with different magnitudes of co-occurring natural and artificial visual stimuli. On that basis, we postulate that the interaction between them is competitive (Fig. 4). However, the model we present is preliminary and will require further study to confirm and refine. For example, the model suggests that disorientation will occur when lighting and natural cues are perceived as similar in relative magnitude, even when they differ over a wide range in absolute magnitude (Fig. 4). That prediction could be easily tested either in our arena, or in another apparatus where each element of the stimulus complex can be precisely controlled. 4.2. Management implications Artificial lighting continues to pose a worldwide threat to the survival of nocturnally active wildlife, and
to marine turtles (Witherington, 1997; Salmon et al., 2000). In Florida today, conservation efforts center on protecting the few remaining dark beaches where most nesting occurs. However, managing lighting at sites with modest development is also an important goal, not only because such sites are much more common but also because darkening a patch of beach can lead to an increase in nesting at that site (Witherington, 1992b; Salmon et al., 1995b). Loggerheads nesting on Florida’s East Coast may have historically ‘‘spread their risk’’ by dispersing their clutches over a relatively long coastline. But presently, about 90% of all nests are deposited at four beaches that differ from adjacent sites only in relative darkness (Salmon et al., 2000). Habitat restoration is essential to reverse that trend. Our arena experiments demonstrate that if the cues promoting normal orientation are enhanced, seafinding will occur even in the presence of artificial lighting. Thus, restoration may not require control over all sources of artificial lighting. Rather, it might be accomplished by pairing light management (controlling as many lights as possible) with dune restoration to increases silhouette darkness and/or elevation. Thus far, there have been no systematic attempts to investigate how well such a management strategy might work. In an arena study, Salmon et al. (1995a) found that higher, wider, and complete silhouettes placed against an illuminated landward horizon were more effective than lower, narrower, and broken silhouettes at
316
S.M. Tuxbury, M. Salmon / Biological Conservation 121 (2005) 311–316
directing the crawls of hatchlings seaward. In a field study, Witherington et al. (1994) determined that artificial silhouettes placed on an illuminated beach also improved the orientation accuracy of crawling hatchlings. However, neither study combined silhouette enhancement with light reduction. In summary, our data suggest that hatchling orientation at illuminated beaches depends upon interactions between lighting and the cues used naturally by the turtles to locate the sea. These results suggest a new approach to beach restoration that involves both light management and dune modification. While we believe this approach has promise, tests at nesting beaches will be required to confirm its efficacy. In the interim, dark sites must be protected, and continued vigilance will be required to prevent any further degradation of nesting beaches at already developed sites. Acknowledgements This study was completed by S.M.T. as part of the requirements for a Masters degree at Florida Atlantic University. Financial support came from the National Save-the-Sea-Turtle Foundation of Ft. Lauderdale, Florida, the Florida Department of Transportation, the Boca Raton Garden Club and Project A.W.A.R.E. We thank Donna Weyrich Franzmathes and John Aguirre for their help in the laboratory. Bill Irwin’s comments on earlier drafts improved the manuscript. References Daniel, R.S., Smith, K.U., 1947. The sea-approach behavior of the neonate loggerhead turtle (Caretta caretta). Journal of Comparative Physiology and Psychology 40, 413–420. Limpus, C.J., 1971. Sea turtle ocean finding behaviour. Search 2, 385– 387. Mrosovsky, N., 1967. How turtles find the sea. Science Journal 3, 52–57. Mrosovsky, N., 1968. Nocturnal emergence of hatchling sea turtles: control by thermal inhibition of activity. Nature 220, 1338–1339. Mrosovsky, N., 1972. The water-finding ability of sea turtles. Brain Behavior and Evolution 5, 202–225. Mrosovsky, N., Shettleworth, S.J., 1968. Wavelength preferences and brightness cues in the water finding behaviour of sea turtles. Behaviour 32, 211–257. Parker, G.H., 1915. The crawling of young loggerhead turtles toward the sea. Journal of Experimental Zoology 6, 323–331. Salmon, M., Wyneken, J., Fritz, E., Lucas, M., 1992. Seafinding by hatchling sea turtles: role of brightness, silhouette and beach slope as orientation cues. Behaviour 122, 56–77.
Salmon, M., Witherington, B.E., 1995. Artificial lighting and seafinding by loggerhead hatchlings: evidence for lunar modulation. Copeia 4, 931–938. Salmon, M., Reiners, R., Lavin, C., Wyneken, J., 1995a. Behavior of loggerhead sea turtles on an urban beach. I. Correlates of nest placement. Journal of Herpetology 29, 560–567. Salmon, M., Tolbert, M., Pender, D., Goff, M., Reiners, R., 1995b. Behavior of loggerhead sea turtles on an urban beach. II. Hatchling orientation. Journal of Herpetology 29, 568–576. Salmon, M., Witherington, B.E., Elvidge, C., 2000. Artificial lighting and the recovery of sea turtles. In: Pilcher, N., Ismail, G. (Eds.), Sea Turtles of the Indo-Pacific, Research Management and Conservation. Asean Academic Press, Sarawak, Malaysia, pp. 25–34. van Rhijn, F.A., van Gorkom, J.C., 1983. Optic orientation in hatchlings of the sea turtle Chelonia mydas. III. Sea-finding behaviour: the role of photic and visual orientation in animals walking on the spot under laboratory conditions. Marine Behaviour and Physiology 9, 211–228. Verheijen, F.J., 1958. The mechanisms of the trapping effect of artificial light sources upon animals. Les Archives Neerlandaises de Zoologie 13, 1–107. Verheijen, F.J., 1980. The moon: a neglected factor in studies on collisions of nocturnal migrant birds with tall-lighted structures and with aircraft. Die Vogelwarte 30, 305–330. Verheijen, F.J., 1981. Bird kills at lighted man-made structures: not on nights close to a full moon. American Birds 35, 251–254. Verheijen, F.J., 1982. Environmental features and orientation. In: Addink, A.D.F., Spronk, N. (Eds.), Exogenous and Endogenous Influences on Metabolic and Neural Control, vol. I. Pergamon Press, Oxford, pp. 371–382. Verheijen, F.J., 1985. Photopollution: artificial light optic spatial control systems fail to cope with. Incidents, causation, remedies. Experimental Biology 44, 1–18. Verheijen, F.J., Wildschut, J.T., 1973. The photic orientation of hatchling sea turtles during water finding behaviour. Netherlands Journal of Sea Research 7, 53–67. Witherington, B.E., 1992a. Sea-finding behavior and the use of photic orientation cues by hatchling sea turtles. Doctoral Dissertation, University of Florida, Gainesville, FL. Witherington, B.E., 1992b. Behavioral responses of nesting sea turtles to artificial lighting. Herpetologica 48, 31–39. Witherington, B.E., 1997. The problem of photopollution for sea turtles and other nocturnal animals. In: Clemmons, J.R., Buchholz, R. (Eds.), Behavioral Approaches to Conservation in the Wild. Cambridge University Press, Cambridge, England, pp. 303– 328. Witherington, B.E., Bjorndal, K.A., Bolten, A., 1994. An evaluation of the use of dune structure to reduce effects of artificial lighting on hatchling sea turtle sea-finding orientation and offshore orientation. Report submitted to US Air Force, Cape Canaveral Air Force Station and Florida Cooperative Fish and Wildlife Research Unit. Witherington, B.E., Martin, R.E., 1996. Understanding, assessing, and resolving light-pollution problems on sea turtle nesting beaches. Florida Marine Research Institute Technical Report TR-2. Zar, J.H., 1999. Biostatistical Analysis, fourth ed. Prentice-Hall, New Jersey.