Loading...
Witherington and Bjorndal 1991b Influences of artificial lighting on hatchling orientationBiological Conservation 55 (1991) 139-149 Influences of Artificial Lighting on the Seaward Orientation of Hatchling Loggerhead Turtles Caretta caretta Blair E. Witherington & Karen A. Bjorndal Archie Carr Center for Sea Turtle Research, Department of Zoology, University of Florida, Gainesville, Florida 32611, USA (Received 11 March 1989; revised version received 29 November 1989; accepted 10 April 1990) ABSTRACT The seaward orientation behavior of hatchling loggerhead turtles Caretta caretta when exposed to five different artificial light sources (high-pressure and low-pressure sodium vapor, and yellow, red, and white incandescent lamps) was examined. Each light source affected hatchling sea -finding performance either in direction of orientation or width of dispersion. Hatchlings were attracted to light sources emitting short -wavelength visible light and long -wavelength sources that excluded intermediate wavelengths. A negative response was observed toward sources emitting predominately yellow light. For this reason, low-pressure sodium vapor (LPS) luminaires, which emit only yellow light, are expected to affect loggerhead hatchling sea - finding minimally, if positioned behind the primary dune. LPS luminaires positioned between emerging hatchlings and the ocean, however, will disrupt - hatchling orientation. INTRODUCTION There are two terrestrial stages in the life history of all sea tutles. Breeding females come ashore on oceanic beaches where they deposit their eggs and subsequently return_ to the ocean. Hatchlings emerge from these sandy nests primarily at night and immediately move toward the sea. In both adult and hatchling sea turtles, correct seaward orientation is critical. 139 Biol. Conserv. 0006-3207/90/$03.50 ©1990 Elsevier Science Publishers Ltd, England. Printed in Great Britain 140 Blair E. Witherington, Karen A. Bjorndal The cues that guide hatchling sea turtles to the ocean are strongly visual (for a review see Mrosovsky & Kingsmill, 1985). One demonstration of the dependence of hatchling sea turles on photic cues may be the disabling effect that artifical lighting has on hatchling seaward orientation. Hatchling sea turtles emerging at night are commonly misdirected toward artificial light. In the midst of numerous artifical light sources, hatchlings move in apparently disoriented circular paths. By thwarting expeditious seaward movement, artificial beachfront lighting can cause substantial hatchling mortality. Hatchling mortality due to beach lighting has been reported for loggerhead turtles Caretta caretta (McFarlane, 1963; Mann, 1978), green turtles Chelonia mydas (Mortimer, 1979) and hawksbill turtles Eretmochelys imbricata (Philibosian, 1976). With increased shoreline development, disruption of hatchling orient- ation due to artificial lighting—and the mortality associated with it—is increasing. In Florida, USA, some communities adjacent to sea turtle nesting beaches have promulgated ordinances restricting the use of beachfront lighting. Preliminary evidence suggests that prohibition of beach lighting during the nesting season substantially reduces disruption of sea turtle hatchling orientation (Witherington, 1986). Voluntary programs in Queensland, Australia, have had similar effects (Limpus et al., 1981). Although the response of the public to these lighting restrictions has been encouraging, the large-scale elimination of artificial light visible from the beach remains a formidable task on densely developed beaches. One concern is that reduced lighting compromises the safety of people using the beach. If beachfront lighting could be designed to emit light visible to humans, but outside the spectral range that significantly affects the seaward orientation of sea turtles, the sea turtles could be protected without jeopardizing human safety. Such specialized lighting would constitute an acceptable compromise to beach communities otherwise unwilling to mitigate the effects of their - - lighting on sea turtles. The spectral sensitivity (action spectrum) of the green turtle is shifted away from longer wavelengths (i.e. redder light), compared with human spectral sensitivity (Granda & Dvorak, 1977). Loggerhead turtles are also believed to see poorly in the red region of the spectrum (Hooker, 1911) and have been observed to retain a seaward orientation despite the presence of luminaires emitting primarily long -wavelength light (Dickerson & Nelson, 1988). In this paper, the effects that five types of commercially available light sources have on the seaward orientation of loggerhead turtle hatchlings are examined. The light sources chosen varied in the spectral properties of light emitted. Our aims are to present the implications of this work for the conservation of loggerhead turtles on developed beaches and to contribute to the elucidation of orientation cues that guide these animals to the sea. Artificial lighting and turtle orientation 141 METHODS The study area was a section of dark, undeveloped beach in Indian River County, eastern Florida, USA, where there_ is substantial nesting of loggerhead turtles. Hatchling loggerheads for the experiments were taken from clutches transplanted into a secured hatchery located near the study area. Hatchlings were collected within one hour of sunset, transported to the study area in darkened buckets and released on the beach immediately after their respective trials. We used a circular pitfall arena similar to that described by Mrosovsky & Carr (1967) to determine the orientation direction of loggerhead hatchlings exposed to artificial lighting. The arena consisted of a circular trench, 30 cm deep and 8 m in diameter, dug into the berm (Fig. 1). The circular trench was divided into 32 compartments separated by numbered wooden dividers. The light source for the experimental and control treatments was positioned at a right angle to the direst ocean route from the arena center, 7.5 m from that center. Each of the five luminaires in the experimental treatments was mounted within a 0.5 m3 wooden box supported on a portable stand 75 cm above the beach. Each box had a 0.5 ml opening fitted with diffusing acrylic. This acrylic was found not to bias the transmission of light between 350 and _ ARENA 4m RELEASE MECHANISM--- HUI TIDE WRACK HATCHLING RELEASE POINT LIGHT SOURCE Fig. 1. Beach study area showing the placement of the pitfall arena and light source. 142 Blair E. Witherington, Karen A. Bjorndal 750 nm (measurement made with a LICOR LI -1800 spectroradiometer). Luminaires were powered with a 1000 W generator, from which the AC voltage did not vary more than one percent. Illuminance of the light sources was set at two intensities by inserting apertures of differing size into the boxes. Apertures were inserted so that light coming from each box filled the entire acrylic window. The two illuminance levels, 1.9lux(I) and 6.2lux(II) (±0.051ux, Minolta T-1 illuminance meter, accuracy ±0.01 lux) measured at the arena center, are the levels of illumination typically used for outside security and public walkways, respectively (Philips Corporation, no date). The five luminaires used as light sources in the experiments were high-pressure sodium vapor (HPS), low-pressure sodium vapor (LPS), yellow -tinted incandescent `bug' lamps (BUG), red -tinted incandescent lamps (RED), and white, untinted quartz lamps (WHITE). Figure 2 illustrates the spectral energy emission for each of these sources. The five light sources at two intensity levels each constituted ten experimental treatments. In the control treatment, light sources were stationed as in the experimental treatments, with the generator operating, but were not turned on. Electric discharge luminaires (LPS and HPS) were run for at least 20 min prior to each trial so that both spectral emission and illuminance remained constant. Experiments were conducted at night between 2200 and 0330 h, a time when loggerhead hatchlings normally emerge from nests and move to the sea (Witherington et al., 1990). At the beginning of each experimental trial, individual hatchlings were transferred in darkness to a black cloth bag and released into the centrer of the areana using a remote release mechanism, LIGHT SOURCE 1001 X --•'•"-- BUG %. J � LL LPS > 60 ¢ HPS /•• LU Z RED i i 60 --------- WHITE X 40 LL O►- W 20 U �j• � ¢ i j 01 350 400 450 500 550 600 650 700 WAVELENGTH (Nth Fig. 2. Spectral emission of the five commercial light sources, low- (LPS) and high-pressure sodium vapor (HPS), yellow- (BUG) and red -(RED) tinted incandescent, and white (WHITE) quartz lamps. Measurements were made with a LICOR LI -1800 spectroradiometer. Artificial lighting and turtle orientation 143 operated from a point 3 m from the arena perimeter in the dune vegetation (Fig. 1). Initial directions in which hatchlings pointed after exiting the bag appeared to be random. Hatchlings that did not move to the outer trench of the arena within 5 min were discarded as subjects (less than 5% of those used). The behavior of hatchlings during the release was observed using night -vision goggles. The pitfall compartment into which a hatchling fell at the end of its trial determined the orientation vector of that hatchling. The release of a single hatchling and. its subsequent capture in the surrounding pitfall constituted a trial. There were 30 trials for each of the 11 treatments. Trials for any given treatment were conducted with hatchlings from different clutches (30 clutches used). Blocks of 11 trials (one trial of each treatment per block) were conducted within a 1-h period and under similar meteorological conditions. Hatchlings were used only once and released. An additional experiment was conducted in which the LPS light source was placed in line with the most direct path to the ocean, 1 m outside the arena perimeter and pointing toward the center. We compared the results from two treatments, LPS source on (illuminance 29.2 lux at arena center) and LPS source off. Apart from the positioning and illuminance of the LPS light source, conditions in these treatments were similar.to those previously described. We made paired comparisons of the orientation vector data from each experimental treatment and the corresponding control treatment. Nonpara- metric tests for direction and dispersion specified for circular data (Batchelet, 1981) were used in these comparisons. A Bonferroni adjustment of P was made so that the significance level for all tests was less than 0.05 (for each test individually, p < 0.005). RESULTS Each of the luminaires tested affected the seaward orientation of hatchlings significantly, either in the direction they traveled or in the angular width at which they dispersed (Table 1, Fig. 3). Only the BUGII source did not significantly affect direction and only the LPS-I source did not significantly affect dispersion. Not all of the sources affected the direction of orienting hatchlings similarly (Fig. 3). In trials where HPS, RED and WHITE sources were presented, the mean hatchling orientation vectors bisected the angle between the direct ocean route and the light source. The orientation of hatchlings presented BUG -I and both LPS sources, however, angled toward the ocean and away from the light source. Treatments with the highest proportion of hatchlings. orienting seaward (±34°) were those with BUG -II, and LPS-I and 11 sources (Fig. 3). 144 Blair E. Witherington, Karen A. Bjorndal ° not significant. Hatchling loggerheads were released at the center of a circular arena. Light sources used in the treatments were presented at 90° with the most ocean -direct path at 0°. The light sources were low-(LPS) and high-pressure sodium vapor (HPS), yellow- (BUG) and red- (RED) tinted incandescent, and white (WHITE) quartz lamps, designated I or II for high or low illuminance. In the control, light sources remained off. Hatchlings in the BUG and LPS treatments were least attracted to the light source. Hatchlings in treatments with WHITE sources exhibited the poorest seaward orientation and the greatest attraction to the light source. Among treatments with HPS, BUG, RED and WHITE sources, the lower illuminance levels (I) corresponded with the widest angular range of hatchling orientation vectors (Table 1). - - In the experiment in which the light source was positioned at 90° (Fig. 1), the hatchlings released in each treatment maintained relatively straight paths to the outside of the arena. Circling was rare ( < I%) and only occurred . immediately after a hatchling fell from the release bag. In contrast, many of the hatchlings confronted with an LPS source illuminating the arena from a position between the hatchling release point and the ocean changed their orientation directions (n = 24 of 30, change > 90°) and circled (n =11 of 30) before reaching the arena perimeter. The distribution of hatchlings orienting with the LPS source turned on differed significantly from that of the control (LPS source positioned similarly but off) both in direction (p <0.001) and dispersion (p < 0.001) (Fig. 4). Whereas hatchlings in the control treatment oriented toward the ocean with little dispersion (r = 0.93, where r is the mean TABLE 1 Results of Nonparametric Tests for Direction and Dispersion (Modifications of the Mann- Whitney U -test, Batchelet, 1981) between Experimental and Control Treatment Groups Treatment Mean angle (°) Angular range (°) Control vs experimental treatment (P) Direction Dispersion Control 0 56 — — LPS-I 340 68 <0.001 0.081° LPS-11 334 101 <0-001 0.004 HPS-I 21 146 <0.001 <0.001 HPS-II 33 124 <0.001 <0.001 BUG -1 343 203 0.001 <0.001 BUG -11 355 124 0.106° <0-001 RED -1 21 135 0.001 <0.001 RED -II 37 90 <0.001 <0.001 WHITE -I 57 101 <0.001 <0.001 WHITE -II 62 68 <0.001 <0.001 ° not significant. Hatchling loggerheads were released at the center of a circular arena. Light sources used in the treatments were presented at 90° with the most ocean -direct path at 0°. The light sources were low-(LPS) and high-pressure sodium vapor (HPS), yellow- (BUG) and red- (RED) tinted incandescent, and white (WHITE) quartz lamps, designated I or II for high or low illuminance. In the control, light sources remained off. Hatchlings in the BUG and LPS treatments were least attracted to the light source. Hatchlings in treatments with WHITE sources exhibited the poorest seaward orientation and the greatest attraction to the light source. Among treatments with HPS, BUG, RED and WHITE sources, the lower illuminance levels (I) corresponded with the widest angular range of hatchling orientation vectors (Table 1). - - In the experiment in which the light source was positioned at 90° (Fig. 1), the hatchlings released in each treatment maintained relatively straight paths to the outside of the arena. Circling was rare ( < I%) and only occurred . immediately after a hatchling fell from the release bag. In contrast, many of the hatchlings confronted with an LPS source illuminating the arena from a position between the hatchling release point and the ocean changed their orientation directions (n = 24 of 30, change > 90°) and circled (n =11 of 30) before reaching the arena perimeter. The distribution of hatchlings orienting with the LPS source turned on differed significantly from that of the control (LPS source positioned similarly but off) both in direction (p <0.001) and dispersion (p < 0.001) (Fig. 4). Whereas hatchlings in the control treatment oriented toward the ocean with little dispersion (r = 0.93, where r is the mean Artificial lighting and turtle orientation 145 Fig. 3. Circular histograms showing the distribution of orientation vectors of loggerhead hatchlings exposed to five different commercial light sources. For each treatment, n = 30. I and Ii designate low and high illuminance levels, respectively. Arrows indicate mean vector direction. The ocean direction is indicated at azimuth 0° and the light source at azimuth 90°. LPS OFF LPs ON Fig. 4. Circular histograms showing the distribution of orientation vectors of loggerhead hatchlings exposed to light from a low-pressure sodium vapor luminaire (LPS on) and to the luminaire turned off (LPS off). Arrows indicate mean vector direction. Both the ocean direction and the light source are at azimuth 0°. 146 Blair E. Witherington, Karen A. Bjorndal vector length), hatchlings illuminated with the LPS source dispersed widely (r = 0.18) away from the ocean and the LPS source. DISCUSSION Responses of loggerhead hatchlings to artificial light Visible light (human spectral range) can disrupt the sea -finding behavior of hatchling loggerhead turtles. Visible light of differing spectral quality, however, affects hatchling sea -finding ability in different ways. Other experimenters have examined the behavioural responses of sea turtle hatchlings to colored light. Although blue light is reported to elicit a stronger orientation response than red light in loggerheads (Hooker, 1911) and green turtles (Mrosovsky & Shettleworth, 1968), red light will elicit an equal or greater response in green turtles if sufficiently intense (Mrosovsky & Shettleworth, 1968). Granda & O'Shea's (1972) examination of the green turtle action spectrum demonstrated a precipitous attenuation in retinal response from 600 to 700 nm. Mrosovsky (1972) compared the behavioral responses of green turtle hatchlings to blue (400-470 nm) and red (590-640 nm) light, with the action spectrum provided by Granda & O'Shea (1972) and surmised that a true preference existed for short -wavelength light. Because each of the light sources we tested affected the sea -finding behavior of hatchling loggerheads to some degree, we conclude that none emits light exclusively beyond the sensory range of the hatchlings. The light source emitting the greatest proportion of short -wavelength light (WHITE, Fig. 2) attracted loggerhead hatchlings to the greatest extent (Fig. 3). We found sources emitting intermediate and long -wavelength light (BUG and LPS, Fig. 2) to -be less attractive to orienting hatchlings than the source emitting primarily long -wavelength light (RED, Fig. 2; Fig. 3). A surprising, negative-phototactic response seen in hatchlings presented pure, intermediate -wavelength (590 nm, yellow, LPS) light may explain the varied response of hatchlings to the mixed, intermediate to long -wavelength source (BUG, and to some extent RED) and the intermediate to short -wavelength source (HPS) (Fig. 3). Sea turtle hatchlings have been over -generalized as being positively phototactic. The study introduces an additional response of sea turtles to light, a `xanthophobic' response, characterizing the negatively phototactic behavior that loggerhead hatchlings display toward yellow light. A robust demonstration of this phenomenon is the negative phototaxis of hatchlings presented yellow LPS light from the ocean direction (Fig. 4). We have also observed this trait in loggerhead hatchlings presented narrow bandwidth Artificial lighting and turtle orientation 147 spectral sources under controlled conditions in the laboratory (With- erington & Bjorndal, in prep.). The new evidence does not support the contention that artificial light sources initiate in sea turtle hatchlings a compulsion for point -source orientation or `trapping effect' as hypothesized for other animals (Verheijen, 1985). It is most likely that the behavior exhibited by hatchling loggerheads moving toward artificial lighting is structurally identical to the behavior of hatchlings moving toward the sea. In at least some circumstances, this behavior involves some spectral assessment on the part of the hatchlings. - Xanthophobia and seaward orientation A negative response to a select portion of the spectrum would benefit a hatchling engaged in sea -finding if that portion of the spectrum characterized the dune direction, or predominated in natural light sources that might otherwise misdirect hatchlings. The orientation directions that hatchlings might avoid to their benefit are those competing with the ocean direction. Rising or setting celestial bodies, such as the sun or moon, must substantially affect the photic environment perceived at the horizon and, depending on the alignment of the beach, compete photically with the ocean direction. The attenuation of blue light due to short -wavelength scattering in the atmosphere causes celestial light sources at the horizon to appear yellow to red depending on conditions. In observing the orientation of loggerhead and green turtle hatchlings on beaches during a rising and setting sun, Mrosovsky & Kingsmill (1985) found a definite effect from the direction of the sun, although the mean orientation vector of the hatchlings pointed seaward regardless of solar position. If the xanthophobic response of loggerhead hatchlings does function as a qualitative assessment of these competing photic cues, hypotheses of sea -finding mechanisms that predict `brightest' direction orientation (Verheijen & Wildschut, 1973; van Rhijn, 1979; Mrosovsky & Kingsmill, 1985) may not be adequate models. The xanthophobic response of loggerhead hatchlings merits additional study. Implications for conservation Verheijen (1985) termed the introduction of detrimental light into_ the environment `photopollution'. Like other forms of pollution, light on sea v turtle nesting beaches can be eliminated or reduced to lessen its threat. Because the elimination of artificial lighting visible from sea turtle nesting beaches is the most guaranteed approach to solving the problem of hatchling mortality from misdirection, we hestitate to suggest alternatives to this ideal. As mentioned previously, however, complete attainment of this goal may be unrealistic. 148 Blair E. Witherington, Karen A. Bjorndal Although some artificial lighting may be present on nesting beaches without significantly affecting sea turtles, we were unable to quantify this level. The level of illumination necessary to impede hatchling sea -finding apparently varies greatly with the spectral qualities of the source(s) (this study) and, very likely, the level and quality of ambient illumination as well (Verheijen, 1985). The use of luminaires that have certain spectral characteristics can be addressed. Broad-spectrum luminaires, such as the WHITE source tested in this study, apparently disrupt hatchling loggerhead sea -finding to the greatest extent. Although all luminaires affected hatchling sea -finding to some degree, the LPS luminaire showed the smallest biological effect. The justification for this conclusion is two -fold: (1) as lighting is conventionally placed on beaches, hatchlings orienting away from LPS luminaires will move in the seaward direction; and (2) LPS lighting prompts hatchlings to move so that light from the LPS source reaching the hatchling (irradiance) lessens with time. LPS luminaires positioned further from the sea will have the smallest effect on the orientation of hatchling loggerheads. If LPS luminaires are positioned between hatchlings and the ocean, however, disruption of hatchling sea -finding can result (Fig. 4). Although LPS luminaires may be used on nesting beaches in such a way as to minimally affect hatchlings, important questions must be answered with regard to other sea turtle species and nesting adults before the use of such lighting is recommended. Because the light emitted by LPS luminaires is visible to the loggerhead, the potential exists for this light to disrupt the nesting and nest -site choice behavior of adult females. Hatchlings and adults of other sea turtle species may also behave differently in the presence of LPS lighting. We are now investigating the relationship between beach lighting and the nesting behavior of adult loggerhead and green turtles with specific regard to LPS and other luminaires. ACKNOWLEDGEMENTS We thank Florida Power and Light Co., the National Fish and Wildlife Foundations and the US Fish and Wildlife Service for funding this study. We are grateful to E. Possardt and R. Wilcox for their support, J. Provancha and R. Wheeler of Bionetics Co. for the use of the spectroradiometer, and L. Ehrhart and M. Horton for graciously allowing us access to their study areas. A. Bolten provided invaluable advice throughout the study, and T. Holderfield and W. Hancock aided in the field work. Artificial lighting and turtle orientation 149 REFERENCES Batchelet, E. (1981). Circular Statistics in Biology. Academic Press, New York. Dickerson, D. D. & Nelson, D. A. (1988). Use of long wavelength lights to prevent disorientation of hatchling sea turtles. NOAA Tech. Memo. NMFS-SEFC- 214, 19-21. Granda, A. M. & Dvorak, C. A. (1977). Vision in turtles. In Handbook of Sensory Physiology, VII(5). The Visual System in Vertebrates, ed. F. Crescitelli. Springer -Verlag, Berlin, pp. 451-95. Granda, A. M. & O'Shea, P. J. (1972). Spectral sensitivity of the green turtle Chelonia mydas mydas determined by electrical responses to heterochromatic light. Brain Behay. Evol., 5, 143-54. Hooker, D. (1911). Certain reactions to color in the young loggerhead turtle. Publs Carnegie Inst. Wash., 132, 71-6. Limpus, C., Carter, R. W. & McLean, S. (1981). Lights and hatchling turtles: an education program. Mar. Turtle Newslett., 19, 11. Mann, T. M. (1978). impact of developed coastline on nesting and hatchling sea turtles in southeastern Florida. Fla Mar. Res. Publ., 33, 53-5. McFarlane, R. W. (1963). Disorientation of loggerhead hatchlings by artificial road lighting. Copeia, 1963, 153. Mortimer, J. A. (1979). Ascension Island: British jeopardize 45 years of conservation. Mar. Turtle Newslett., 10, 7-8. Mrosovsky, N. (1972). The water -finding ability of sea turtles. Brain, Behay. Evol., 5, 202-25. Mrosovsky, N. & Carr, A. (1967). Preference for light of short wavelengths in hatchling green sea turtles, Chelonia mydas, tested on their natural nesting beaches. Behaviour, 28, 217-31. Mrosovsky, N. & Kingsmill, S. F. (1985). How turtles find the sea. Z. Tierpsychol, 67, 237-56. Mrosovsky, N. & Shettleworth, S. J. (1968). Wavelength preferences and brightness cues in the water finding behaviour of sea turtles. Behaviour, 32, 211-57. Philibosian, R. (1976). Disorientation of hawksbill turtle hatchlings, Eretmochelys imbricata, by stadium lights. Copeia, 1976, 824. Philips Corporation (no date). Outdoor lighting made easy. Hightstown, New Jersey, North American Philips Lighting Corporation. van Rhijn, F. A. (1979). Optic orientation in hatchlings of the sea turtle, Chelonia mydas, 1. Brightness: not the only optic cue in sea -finding orientation. Mar. Behay. Physiol., 6, 105-21. Verheijen, F. J. (1985). Photopollution: Artificial light optic spatial control systems fail to cope with. Incidents, causations, remedies. Exp. Biol., 44, 1-18. Verheijen, F. J. & Wildschut, J. T. (1973). The photic orientation of sea turtles during water finding behaviour. Neth. J. Sea Res., 7, 53-67. Witherington, B. E. (1986). Human and natural causes of marine turtle clutch and hatchling mortality and their relationship to hatchling production on an important Florida nesting beach. Masters thesis, University of Central Florida, Orlando. Witherington, B. E., Bjorndal, K. A. & McCabe, C. M. (1990). Temporal pattern of nocturnal emergence of loggerhead turtle hatchlings from natural nests. Copeia, 1990(4), 1135-8.