Witherington 1992 Behavioral responses of nesting sea turtles to artificial lightingHerpetologists' League
Behavioral Responses of Nesting Sea Turtles to Artificial Lighting
Author(s): Blair E. Witherington
Source: Herpetologica, Vol. 48, No. 1 (Mar., 1992), pp. 31-39
Published by: Herpetologists' League
Stable URL: http://www.jstor.org/stable/3892916 .
Accessed: 23/06/2014 09:29
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp
.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact support@jstor.org.
.
Herpetologists' League is collaborating with JSTOR to digitize, preserve and extend access to Herpetologica.
http://www.jstor.org
This content downloaded from 164.51.49.151 on Mon, 23 Jun 2014 09:29:09 AM
All use subject to JSTOR Terms and Conditions
Herpetologica, 48(1), 1992, 31-39
? 1992 by The Herpetologists' League, Inc.
BEHAVIORAL RESPONSES OF NESTING SEA TURTLES TO
ARTIFICIAL LIGHTING
BLAIR E. WITHERINGTON
Archie Carr Center for Sea Turtle Research and Department of Zoology,
University of Florida, Gainesville, FL 32611, USA
ABSTRACT: Effects of artificial lighting on loggerhead (Caretta caretta) and green turtle (Che-
lonia mydas) nesting behavior were determined experimentally at major nesting beaches: Melbourne
Beach, Florida, USA (loggerheads) and Tortuguero, Costa Rica (green turtles). I conducted exper-
iments in which a portion of each nesting beach remained dark, or was illuminated with white,
mercury vapor (MV) or yellow, low pressure sodium vapor (LPS) luminaires of equal luminance.
Lighting beaches with MV luminaires significantly reduced the numbers of green turtles and
loggerheads emerging and nesting within lighted study areas. Lighting beaches with LPS luminaires
had no significant effect on nesting in either species. Some turtles were misdirected by lighted
luminaires (primarily mercury vapor) on their return to the ocean following nesting attempts.
Lighted luminaires did not significantly affect the stages at which nesting attempts were abandoned
nor the positioning of nests relative to dune vegetation. Results suggest that MV luminaires and
other broad-spectrum lighting types have the potential to disrupt the nesting of loggerheads and
green turtles. LPS luminaires may be an acceptable alternative where lighting on nesting beaches
cannot be completely extinguished.
Key words: Caretta caretta; Chelonia mydas; Light; Nesting behavior; Photopollution; Sea
turtle
SEA turtles emerge from the ocean onto
sand beaches where they nest. Except for
subtle modifications, all species of sea tur-
tles have in common a series of stereotyped
nesting behaviors or modal action patterns
(Bustard, 1972; Carr et al., 1966; Ehren-
feld, 1979). Although modal action pat-
terns and their sequence during sea turtle
nesting are largely constant, some ele-
ments of the behavior may vary. Variable
elements include where the turtle emerg-
es, where on the beach it begins nest con-
struction, whether or not the nesting at-
tempt is abandoned, what stage is reached
before abandonment, and the directness
of orientation on the return crawl. This
variation in nesting behavior by sea turtles
affects successes in egg deposition, in
hatchling production, and in the seaward
return of the adult (Witherington, 1986).
The conditions that influence variation
in sea turtle nesting behavior, such as nest-
placement, are poorly known. Two studies
correlate nest-site choice with environ-
mental conditions; Mortimer (1982) found
that green turtles (Chelonia mydas) with
a choice of nesting beaches selected those
with more open, uncluttered approaches,
and Stoneburner and Richardson (1981)
found that loggerhead turtles (Caretta ca-
retta) moving up a beach tended to nest
as they encountered rises in sand temper-
ature. Other environmental cues used by
sea turtles to discriminate nesting sites may
be visual, as are the cues used by adult sea
turtles in relocating the sea (Ehrenfeld and
Carr, 1967). Because most species of sea
turtles are nocturnal nesters, artificial
lighting of nesting beaches may present an
environmental modification that disrupts
visual cues. The disruptive effect of beach
photopollution on the seaward orientation
of hatchling sea turtles is well documented
(for a review, see Verheijen, 1985).
Increasing human development adja-
cent to sea turtle nesting beaches world-
wide has brought with it increasing levels
of artificial illumination. Some authors have
observed correlations between lighted, de-
veloped beaches and lower nesting activity
by sea turtles (Mortimer, 1982; Proffitt et
al., 1986; Worth and Smith, 1976). These
correlations, however, do not directly im-
plicate lighting as the cause of decreased
nesting.
The purpose of this study was to deter-
mine experimentally the effects that two
forms of artificial lighting have on the noc-
31
This content downloaded from 164.51.49.151 on Mon, 23 Jun 2014 09:29:09 AM
All use subject to JSTOR Terms and Conditions
32 HERPETOLOGICA [Vol. 48, No. 1
turnal nesting behavior of green turtles and
loggerhead turtles. Aspects of nesting be-
havior examined included nest-site choice,
nest-site abandonment, and seaward ori-
entation of adult females on the nesting
beaches.
METHODS
Study Areas
Melbourne Beach.-I chose a 1300 m
stretch of east-facing beach, approximate-
ly 6.5 km south of the town of Melbourne
Beach, Florida, USA, because of the high
density (500-700 nests/km/season: With-
erington, 1986) of loggerhead turtles nest-
ing there. Work at Melbourne Beach took
place between 15 May and 27 June 1989,
during the nesting season but prior to the
emergence of hatchlings from nests. The
study beach was backed by dune scrub and
was undeveloped except for a small (50 m
length) building with no lighting on the
beach side of the building. Other than the
single building, and subtle variation in
beach width (15-20 m, high water mark
to dune vegetation), the study beach ap-
peared homogeneous. The entire moni-
tored study beach remained dark at night,
although light from scattered luminaires
was visible to the north and south of the
study beach boundaries. The primary dune
at the site was eroded from a recent storm
into a vertical escarpment 3-4 m high. This
did not restrict access of turtles to suitable
nesting sites. Human presence at night was
minimal.
Tortuguero.-I chose a 1450 m stretch
of east-facing beach within the Parque Na-
cional Tortuguero, 15 km south of the Vil-
lage of Tortuguero, Costa Rica, for the
high density (1300-3500 nests/km/season:
K. Horikoshi, unpublished data) nesting of
green turtles there. Work at Tortuguero
took place between 14 July and 16 August
1989, during the nesting season but prior
to the emergence of hatchlings from nests.
There was no human development within
15 km of the study beach, which was
backed by lowland rainforest. The pri-
mary dune was low, and the line separat-
ing vegetation from beach was roughly de-
fined. Beach width was 3-15 m and varied
both according to area of study beach and
time of study-period. I maintained a camp
at the site, which was not visible from the
beach. No other humans were present at
the site.
Experimental Design
Melbourne Beach. -I affixed five wood-
en poles, set 50 m apart, at the edge of the
primary dune escarpment in the center of
the study beach. Luminaires were mount-
ed on the poles 2 m above the dune and
5-6 m above the berm. Three 650 W por-
table generators (Honda EG-650) powered
the luminaires. Each generator powered
one or two luminaires and was housed in
a plywood hut positioned 25 m from the
nearest luminaire. Additional gasoline res-
ervoirs allowed the generators to run for
12 h (throughout the night). Due to dis-
tance, surf-noise, and a predominantly on-
shore breeze, the sound and smell of run-
ning generators were undetectable to a
human observer on the lower berm. The
study beach was marked every 50 m for
550 m both north and south of the north-
ernmost and southernmost light poles.
I used two experimental treatments,
which differed in the type of luminaire
that lighted the study beach during a night
of nesting. The two experimental treat-
ments were MV (mercury vapor lumi-
naires used, 80 W Innovative Controls Lite
Light) and LPS (low pressure sodium va-
por luminaires used, 35 W VL Lighting
SOX DD-35). In a control treatment, lu-
minaires remained darkened and the gen-
erators were not run.
Both MV and LPS luminaires were
"dusk-to-dawn" types, each having a cir-
cular light distribution with the center of
the light field beneath the lamp. I selected
wattages of the luminaires so that the MV
and LPS types had similar luminance lev-
els-i.e., similar brightness with respect to
human photopic vision. Illuminance val-
ues at varied distances from single lumi-
naires were similar for the two luminaire
types (Table 1). The MV luminaires emit-
ted broad-spectrum "white" light with
spectral line peaks near 400, 440, 550, 580,
620, and 700 nm (Licor LI-1800 spectro-
radiometer). The LPS luminaires ap-
peared yellow, and between 400 and 700
This content downloaded from 164.51.49.151 on Mon, 23 Jun 2014 09:29:09 AM
All use subject to JSTOR Terms and Conditions
March 1992] HERPETOLOGICA 33
nm, emitted a single spectral line at 590
nm.
I conducted each of the three treatments
once, from sunset until sunrise, during each
of 14 consecutive three-night blocks be-
tween 16 May and 27 June 1989. The order
of treatment was randomized within each
three-night block.
Tortuguero.-Methods and materials
used at Tortuguero were identical to those
used at Melbourne Beach except for the
following. (1) Six rather than five lumi-
naires per treatment were used and were
mounted approximately 50 m apart in the
center of the study beach on poles or trees
available on site. (2) Luminaires were
mounted approximately 2 m above a low
primary dune. (3) Generators were housed
in thatch huts 4-6 m into the vegetation;
each was undetectable to observers on the
beach. (4) The study beach was marked
each 50 m for 600 m north and south of
the northernmost and southernmost lu-
minaires. (5) MV, LPS, and control treat-
ments were conducted in each of 11, three-
night blocks from 14 July-16 August 1989.
(6) Due to failure of equipment, only the
four northernmost luminaires were lighted
during MV and LPS treatments in the last
three-night block.
Data and Analysis
Assessments of the nightly nesting be-
havior of green turtles and loggerheads
were made through appraisals of crawls
left on the beach. Crawls include tracks
(marks left in the sand by emerging and
returning turtles) and nest sign (marks left
by the excavation of body pits and/or egg
chambers, or by nest-site obliteration).
Surveys of crawls took place shortly after
dawn. No nesting turtles emerged during
daylight. Because crawls below the recent
high tide line could be erased by tides prior
to inspection, these crawls were not in-
cluded in the analysis. During the exper-
iments, only non-nesting crawls fell into
this category. The following data were
taken for each crawl. (1) Crawl location
was measured as the distance of the crawl
vertex (point farthest from the surf) from
the nearest study area marker. (2) Species
of turtle responsible for the crawl was de-
TABLE 1.-Illuminance measurements (in lux) taken
at varying distances from single luminaires (mercury
vapor = MV, or low pressure sodium vapor = LPS)
at the Melbourne Beach study area. Measurements
were made with a Minolta T-1 illuminance meter at
the given lateral distances from the globe of the lu-
minaire. Ambient illuminance with luminaires off
was <0.01 lux.
Distance (m)
Luminaire 1 5 10 20
LPS 470 13.0 2.03 0.68
MV 479 6.6 1.36 0.57
termined by characteristic sign in the beach
substrate due to differences in turtle mor-
phology and gait (Pritchard et al., 1983).
Sea turtles other than the target species
nested at the study beaches in numbers too
low for study. (3) Type of crawl was spec-
ified as nesting or non-nesting, and if non-
nesting, as abandoned during emergence,
body pit excavation, or egg chamber ex-
cavation (Pritchard et al., 1983). (4) Dis-
tance to vegetation was measured as the
distance from the crawl vertex (if non-
nesting) or the nearest edge of the nest site
(if nesting) to the nearest vegetation. (5)
Straightness index was measured as the
length of the straightest route seaward to
the most recent high tide mark, divided
by the length of the return track path.
Straightness indices <1.0 represent devi-
ations from straightness. Path length was
measured from the crawl vertex of crawls
with no digging activity and from the most
seaward edge of the nest or excavation site
of other crawls. Direction of the return
(north or south) and a simplified diagram
of the crawl also were recorded.
Comparisons among treatments were
made using nonparametric statistics (Gib-
bons, 1985), with a = 0.05.
RESULTS
There were 516 loggerhead emergences
within the study area at Melbourne Beach,
of which 47.1% resulted in nests. At the
study beach at Tortuguero, there were 798
green turtle emergences, of which 58.0%
resulted in nests. After an exploratory anal-
ysis of the spatial distribution of crawl-
data, I determined that the area affected
maximally by lighted luminaires extended
This content downloaded from 164.51.49.151 on Mon, 23 Jun 2014 09:29:09 AM
All use subject to JSTOR Terms and Conditions
34 HERPETOLOGICA [Vol. 48, No. 1
20 0 NON-NESTING
CONTROL U NESlNG
15
10
5
0
20
LPS
w 15
z
20 5
LU
20
MV
15
10
5
0
N L , S
LOCATION
FIG. 1.-Distribution of nesting and non-nesting
emergences of loggerhead turtles within the 1300 m
study beach at Melbourne Beach, Florida, USA. Lo-
cation divisions represent 50 m sections. Bracketed
sections represent the centrally lighted portion of the
study area, which remained dark (control) or was
illuminated with either mercury vapor (MV) or low
pressure sodium vapor (LPS) luminaires of equal lu-
minance. Peripheral areas include all locations out-
side the centrally lighted area.
approximately 50 m from each luminaire
(Figs. 1, 2). For further analysis, I divided
each study beach into three areas: a central
lighted area-including the beach within
50 m of any luminaire -and two periph-
eral areas-areas of the study beach north
and south of the central lighted area. Cen-
tral lighted areas for Melbourne Beach and
Tortuguero were 300 m and 350 m in
length, respectively.
There were significantly fewer logger-
head nesting and non-nesting crawls, and
green turtle nesting crawls, within the cen-
tral lighted areas during MV treatment
nights, compared with control treatment
30 -
<, 25- ~~LPS
25
C.) 20
G 15
W 10
5
ILl
30 LP
U) 25
C) 20
15
10
LU 5
0
N ' . 'S
LOCATION
FIG. 2.-Distribution of nesting and non-nesting
emergences of green turtles within the 1450 m study
beach at Tortuguero, Costa Rica. See legend to Fig. 1.
nights (Table 2). The numbers of nesting
and non-nesting loggerhead and green
turtle crawls during LPS treatment nights
were not significantly different from those
numbers during control treatment nights
(Table 2). Numbers of nesting and non-
nesting crawls within peripheral areas at
each study beach did not differ among
treatments (Table 2). In a complementary
analysis, central lighted areas and periph-
eral areas during MV treatments for both
species were significantly different in the
density of nesting and non-nesting crawls
(Table 3). Density of nesting and non-nest-
ing crawls were not different between cen-
tral and peripheral areas in LPS or control
treatments for either species (Table 3).
I found no relationship between exper-
imental treatments and the stages at which
loggerhead non-nesting emergences were
This content downloaded from 164.51.49.151 on Mon, 23 Jun 2014 09:29:09 AM
All use subject to JSTOR Terms and Conditions
March 1992] HERPETOLOGICA 35
TABLE 2.-Numbers of nesting and non-nesting emergences of loggerheads and green turtles each night
during control (dark beach), low pressure sodium vapor luminaire (LPS), and mercury vapor luminaire (MV)
treatments. Emergence locations are either central (within the central study area where luminaires were
placed) or peripheral (outside the central area). Statistical values (H and P) are for Kruskal-Wallis tests among
treatments. Different letters within rows indicate significant differences among treatments as detected with
a nonparametric multiple comparison test (a = 0.05). Sample size (nights) is constant among treatments for
each species (loggerhead, n = 14; green turtle, n = 11).
i emergences + 1 SD
Control LPS MV H P
Loggerheads
Central
Nesting 1.4 ? 0.9A 1.1 ? ilA 0.07 ? 0.27B 17 0.01
Non-nesting 1.9 ? 1.9A 1.4 ? 1.2AB 0.50 ? 1.2B 7.5 0.02
Peripheral
Nesting 5.0 ? 2.3 5.9 ? 3.5 4.1 ? 2.3 1.7 0.43
Non-nesting 5.8 ? 5.5 5.6 ? 4.0 4.4 ? 2.7 0.3 0.87
Green turtles
Central
Nesting 4.2 ? 3.OA 2.1 ? 1.8AB 0.36 ? 0.67B 14 0.01
Non-nesting 3.4 ? 2.5 1.6 ? 1.2 1.3 ? 1.1 4.4 0.11
Peripheral
Nesting 10 ? 4.3 12 ? 4.7 13 ? 4.7 1.4 0.49
Non-nesting 8.3 ? 4.9 7.6 ? 4.2 8.6 ? 5.2 0.1 0.93
abandoned, either within the entire mon-
itored study beach (X2 = 2.12, df = 2, P =
0.35) or within the central lighted area
only (X2 = 0.041, df = 1, P = 0.84). Nor
was such a relationship found for green
turtle crawls either within the entire mon-
itored study area (x2 = 1.07, df = 2, P =
0.90) or within the central lighted area
only (X2 = 0.94, df = 2, P = 0.62). No
significant difference was found among
TABLE 3.-Density of loggerhead and green turtle nesting and non-nesting emergences among treatments.
Statistical values (z and P) are for Mann-Whitney tests between central and peripheral groups. Definition of
groups and sample sizes are in Table 2.
x density (emergences/km/night) ? 1 SD
Central Peripheral z P
Loggerheads
Control
Nesting 4.5 ? 3.1 4.9 ? 2.3 0.37 0.36
Non-nesting 6.2 ? 6.2 5.8 ? 5.5 0.09 0.46
LPS
Nesting 3.6 ? 3.6 5.9 ? 3.7 1.1 0.13
Non-nesting 4.8 ? 4.1 5.6 ? 4.0 0.46 0.32
MV
Nesting 0.20 ? 0.91 4.1 ? 2.3 4.3 0.0001
Non-nesting 1.7 ? 3.9 4.4 ? 2.7 3.1 0.001
Green turtles
Control
Nesting 11.9 ? 8.4 9.6 ? 3.8 0.76 0.22
Non-nesting 9.6 ? 7.1 7.2 ? 4.3 0.66 0.25
LPS
Nesting 6.0 ? 5.0 10.8 ? 4.3 1.5 0.062
Non-nesting 4.7 ? 3.4 6.9 ? 3.8 0.92 0.18
MV
Nesting 1.0 ? 1.9 11.6 ? 4.2 4.0 0.0001
Non-nesting 3.6 ? 3.2 7.9 ? 4.6 2.1 0.022
This content downloaded from 164.51.49.151 on Mon, 23 Jun 2014 09:29:09 AM
All use subject to JSTOR Terms and Conditions
36 HERPETOLOGICA [Vol. 48, No. 1
TABLE 4.-Distance to dune vegetation for nesting and non-nesting emergences of loggerheads and green
turtles. Definition of groups is in Table 2.
x distance (m) ? 1 SD (n)
Control LPS MV H P
Loggerheads
Central
Nesting 9.6 ? 8.0 (19) 11 ? 6.1 (15) 14.7 (1) *0.59 0.27
Non-nesting 16 ? 7.0 (26) 18 ? 6.5 (20) 16 ? 2.5 (6) 1.5 0.48
Peripheral
Nesting 8.5 ? 6.2 (70) 8.9 ? 6.6 (82) 8.0 ? 7.2 (58) 1.1 0.58
Non-nesting 14 ? 7.4 (80) 15 ? 7.4 (78) 15 ? 7.7 (61) 1.1 0.59
Green turtles
Central
Nesting 0.8 ? 3.9 (46) 0.4 ? 2.5 (23) 1.8 ? 6.0 (4) 0.04 0.98
Non-nesting 4.5 ? 7.5 (36) 3.6 ? 6.8 (24) 4.2 ? 9.9 (15) 1.2 0.54
Peripheral
Nesting 1.6 ? 5.0 (114) 1.4 ? 4.7 (139) 2.5 ? 4.9 (141) 4.9 0.09
Non-nesting 3.1 ? 6.6 (90) 2.9 ? 5.4 (82) 4.3 ? 7.4 (96) 1.4 0.49
* A Mann-Whitney test was conducted between control and LPS groups. The statistic presented is z.
treatments in the distances of nesting and
non-nesting emergences from the dune
vegetation (Table 4).
All emerging turtles were able to reenter
the ocean without assistance. Most turtles
reentered the ocean with path deviations
no more than 300 (straightness index ap-
proximately 0.8) from the most direct
route. However, three green turtles and
one loggerhead during MV treatment
nights, and one green turtle during a LPS
treatment night were misdirected after
nesting. Return tracks of these turtles ex-
tended laterally toward a lighted lumi-
naire and circled beneath it before either
extending toward another luminaire or ap-
proaching the ocean. Straightness of return
tracks differed among treatments in only
two groups: loggerhead nesting emer-
gences and green turtle non-nesting crawls
located peripheral to the central lighted
areas (Table 5).
DIscUSSION
Lighting Effects on Nesting and
Non-nesting Emergences
MV lighting.-The presence of white,
MV luminaires sharply reduced the num-
bers of green turtles and loggerheads nest-
ing within central lighted areas (Figs. 1,
2); MV lighting also reduced numbers of
non-nesting crawls within central lighted
areas, although this trend was significant
only for loggerheads (Table 2). For both
species, MV lighting reduced densities of
nesting and non-nesting crawls within cen-
trally lighted areas as compared with pe-
ripheral areas (Table 3). Reduced nesting
within centrally lighted areas during MV
treatment nights was primarily due to low-
er numbers of turtles emerging onto the
beach. This suggests that most of the de-
cisions to abandon nesting due to the pres-
ence of MV lighting were made prior to
emerging onto the dry beach. The nesting-
inhibition effects at artificially lighted
beaches, therefore, might not be revealed
by a preponderance of non-nesting crawls.
Numbers of emergences in MV treatments
were too small to test statistically for dif-
ferences in nesting abandonment between
MV and control treatments. If increased
abandonment of nesting occurred, it con-
tributed little to reducing numbers of nests
in MV treatments, because few turtles at-
tempted nesting during MV treatments
(Figs. 1, 2; Tables 2, 3).
A major component of nest-site selection
by sea turtles probably involves assess-
ments of visual cues. Although the artificial
lighting of beaches is a recent phenome-
non in the evolutionary history of sea tur-
tles, it is possible that nesting on "brighter"
beaches may have been selected against
historically. Beaches having bleached tree
trunks or exposed white sand on the dune
may constitute a confusing photic envi-
This content downloaded from 164.51.49.151 on Mon, 23 Jun 2014 09:29:09 AM
All use subject to JSTOR Terms and Conditions
March 1992] HERPETOLOGICA 37
TABLE 5.-Straightness indices of return paths from nesting and non-nesting emergences of loggerheads and
green turtles. Indices <1.0 indicate deviations from straightness. Definition of groups is in Table 2.
x straightness index (range; n)
Control LPS MV H P
Loggerheads
Central
Nesting 0.82 0.77 1.0 *1.5 0.07
(0.26-1.0; 19) (0.36-1.0; 15) (1.0; 1)
Non-nesting 0.95 0.91 0.82 5.5 0.07
(0.82-1.0; 26) (0.70-1.0; 19) (0.72-1.0; 6)
Peripheral
Nesting 0.88A 0.77AB 0.64B 9.2 0.01
(0.55-1.0; 70) (0.20-1.0; 82) (0.05-1.0; 58)
Non-nesting 0.91 0.86 0.90 2.2 0.34
(0.50-1.0; 80) (0.25-1.0; 77) (0.61-1.0; 61)
Green turtles
Central
Nesting 0.93 0.63 0.94 4.9 0.08
(0.64-1.0; 46) (0.14-1.0; 23) (0.80-1.0; 4)
Non-nesting 0.81 0.84 0.37 1.2 0.55
(0.24-1.0; 36) (0.50-1.0; 24) (0.08-1.0; 15)
Peripheral
Nesting 0.93 0.86 0.81 3.5 0.17
(0.38-1.0; 114) (0.17-1.0; 139) (0.07-1.0; 141)
Non-nesting 0.92A 0.87AB 0.84B 6.7 0.04
(0.55-1.0; 90) (0.32-1.0; 82) (0.40-1.0; 96)
* A Mann-Whitney test was conducted between control and LPS groups. The statistic presented is z.
ronment to hatchlings attempting to locate
the ocean from nests. During sea-finding,
hatchling green turtles (Carr, 1962) and
loggerheads (Witherington, unpublished
data) may become distracted by reflective
objects on the beach. Artificial lighting also
may misrepresent the time of day to turtles
attempting to nest. Loggerheads and green
turtles are nocturnal nesters. To a turtle
that has not yet stranded to nest, a brightly
lighted beach may signify daylight, and
thus inhibit nesting.
The area of beach where MV lighting
affected loggerhead and green turtle nest-
ing was limited to the most brightly lighted
portion of the beach (Figs. 1, 2). Numbers
of nesting and non-nesting emergences in
peripheral areas did not vary significantly
among treatments (Table 2). Levels of light
intensity in peripheral areas may be too
low to discourage the emergence of nest-
ing turtles.
Other broad-spectrum luminaires may
affect loggerhead and green turtle nesting
as MV luminaires do. Incandescent, metal
halide, and conventional fluorescent lamps
all emit wavelengths similar to MV lamps
(Witherington, unpublished data). High
pressure sodium vapor lamps are not a
broad-spectrum, white source, but unlike
LPS lamps, they do emit short wavelengths
(Witherington and Bjorndal, 1991) that
may disrupt nesting.
LPS lighting.-Light emitted by LPS
luminaires had no significant effect on the
numbers of green turtles and loggerheads
emerging and nesting (Tables 2, 3). LPS
luminaires emit light near the peak of
spectral sensitivity for green turtles (Gran-
da and O'Shea, 1972). Rather than ap-
pearing dim to turtles, LPS lighting may
appear as an innocuous color to emerging
turtles or as a color not interpreted as day-
light. The complete spectral sensitivity of
the loggerhead has not been determined.
Although I found no direct effect of LPS
lighting on nesting, I am not able to rule
out indirect effects that LPS lighting may
have. Artificial lighting not affecting nest-
ing turtles directly may indirectly enhance
human interference on nesting beaches
utilized by sea turtles. Turtles nesting in
lighted areas are more conspicuous and,
therefore, more likely to be approached
This content downloaded from 164.51.49.151 on Mon, 23 Jun 2014 09:29:09 AM
All use subject to JSTOR Terms and Conditions
38 HERPETOLOGICA [Vol. 48, No. 1
by humans visiting the beach. Lighting in
turn may make approaching humans more
conspicuous to nesting turtles. The pres-
ence of humans moving within sight of a
female loggerhead or green turtle prior to
oviposition prompts abandonment of nest-
ing in most instances (Witherington, un-
published data).
Lighting Effects on Nest Placement and
Sea-finding
Neither MV nor LPS lighting affected
placement of nests by adult females once
they emerged onto the beach to nest (Table
4), the stages at which nesting behavior
sequences were abandoned, or the distance
traversed before nesting was abandoned
(Table 4). The effect that either type of
lighting had on turtles returning to the sea
was small in most instances. Statistical dif-
ferences in path-straightness among treat-
ments were detected only for turtles that
emerged within peripheral areas of the
study beaches (Table 5). The small effect
(averaged overall) that lighted luminaires
had on sea-finding may only be detectable
in the larger numbers of turtles emerging
in the peripheral areas. Although there was
little effect on seaward orientation in the
majority of emergences, some individuals
(n = 5) apparently spent a large portion
of the night wandering in search of the
ocean. Because misdirected female turtles
may become unable to reenter the ocean
due to topography, the biological effect of
lighting on the sea-finding ability of nest-
ing turtles should not be considered neg-
ligible.
CONCLUSIONS
Reduction of sea turtle nesting observed
on developed beaches is explained by the
effects that many types of artificial light-
ing have on the nest-site-selection behavior
of sea turtles. Light emitted by LPS lu-
minaires, however, had no significant ef-
fect on the numbers of turtles emerging
and nesting or on their behavior. Previous
work has shown that light from LPS lu-
minaires and light of similar spectral qual-
ity have a much smaller effect on sea-find-
ing in loggerhead and green turtle
hatchlings as compared with light from
other sources (Witherington and Bjorndal,
1991, in press). Given this evidence, LPS
luminaires show promise as an alternative
to other types of lighting on loggerhead
and green turtle nesting beaches, but should
be considered only as a compromise. Elim-
inating beach-lighting remains the most
complete way to protect sea turtle hatch-
lings and preserve nesting on historical
nesting beaches.
RESUMEN
Los efectos de la luz artificial en el com-
portamiento de las especies caguama (Ca-
retta caretta) y la tortuga verde (Chelonia
mydas) al hacer sus nidos fueron deter-
minados experimentalmente en las playas
principales de nidada: Melbourne Beach,
Florida EUA (caguamas) y Tortuguero,
Costa Rica (Tortugas verdes). Lleve a cabo
experimentos en que una porcion de cada
playa de nidada permanecia oscura, o era
iluminada con vapor de mercurio (MV) o
con lumbreras de igual luminiscencia de
vapor amarillo de sodio a baja presion
(LPS). La iluminacion de las playas con
lumbreras MV redujo significativamente el
numero de tortugas verdes y caguamas sa-
liendo y anidando dentro de las 'areas de
estudio iluminadas. La iluminacion con
lumbreras LPS no tuvo un efecto signifi-
cativo en la nidada de ninguna de las es-
pecies. Algunas tortugas fueron mal diri-
gidas por las lumbreras encendidas
(principalmente vapor de mercurio) en su
regreso al oceano despues de sus intentos
de anidar. Las lumbreras encendidas no
afectaron significativamente las etapas en
que los intentos de anidar fueron suspen-
didos ni tampoco la posicion de los nidos
en relacion con la vegetacion de las dunas.
Las resultados sugieren que las lumbreras
MV y otros tipos de iluminacion de amplio
espectro tienen el potencial de trastornar
la nidada de las caguamas y las tortugas
verdes. Las lumbreras LPS pueden ser una
alternativa aceptable cuando la ilumina-
cion en las playas de nidada no puede ser
completamente extinguida.
Acknowledgments. -This study was funded by
grants to K. Bjorndal and B. E. Witherington from
Florida Power and Light Company, the U.S. Fish and
Wildlife Service, the Caribbean Conservation Cor-
This content downloaded from 164.51.49.151 on Mon, 23 Jun 2014 09:29:09 AM
All use subject to JSTOR Terms and Conditions
March 1992] HERPETOLOGICA 39
poration, Greenpeace, and the National Fish and
Wildlife Foundation. For their aid and technical ad-
vice on the project, I thank K. Bjorndal, A. Bolten,
C. Diez, K. Horikoshi, J. Johnson, J. Mortimer, E.
Possardt, J. Provancha, P. Ross, R. Wheeler, and R.
Wilcox. I appreciate access to the study area of L.
Ehrhart and the aid provided by his students. Ad-
ditional cooperation was received from Bionetics Co.,
the Brevard County Commission, the Florida De-
partment of Natural Resources, and the Servicio de
Parques Nacionales de Costa Rica.
LITERATURE CITED
BUSTARD, H. R. 1972. Sea Turtles. Natural History
and Conservation. Collins, London.
CARR, A. 1962. Orientation problems in the high
seas travel and terrestrial movements of marine
turtles. Am. Sci. 50:359-374.
CARR, A., H. HIRTH, AND L. OGREN. 1966. The
ecology and migrations of sea turtles, 6. The hawks-
bill turtle in the Caribbean Sea. Am. Mus. Novitates
2248:1-29.
EHRENFELD, D. W. 1979. Behavior associated with
nesting. Pp. 417-434. In M. Harless and H. Morlock
(Eds.), Turtles: Perspectives and Research. Wiley
and Sons, New York.
EHRENFELD, D. W., AND A. CARR. 1967. The role
of vision in the sea-finding orientation of the green
turtle (Chelonia mydas). Anim. Behav. 15:25-36.
GIBBONS, J. D. 1985. Nonparametric Methods for
Quantitative Analysis, 2nd ed. American Sciences
Press, Columbus, Ohio.
GRANDA, A. M., AND P. J. O'SHEA. 1972. Spectral
sensitivity of the green turtle (Chelonia mydas my-
das) determined by electrical responses to hetero-
chromatic light. Brain Behav. Evol. 5:143-154.
MORTIMER, J. A. 1982. Factors affecting beach se-
lection by nesting sea turtles. Pp. 45-51. In K. A.
Bjorndal (Ed.), Biology and Conservation of Sea
Turtles. Smithsonian Institution Press, Washington,
D.C.
PRITCHARD, P., P. BACON, F. BERRY, A. CARR, J.
FLETEMEYER, R. GALLAGHER, S. HOPKINS, R.
LANKFORD, R. MARQUEZ, L. OGREN, W. PRINGLE,
JR., H. REICHART, AND R. WITHAM. 1983. Man-
ual of Sea Turtle Research and Conservation Tech-
niques, 2nd ed. K. Bjorndal and G. Balazs (eds.).
Center for Environmental Education, Washington,
D.C.
PROFFITT, C. E., R. E. MARTIN, R. G. ERNEST, B. J.
GRAUNKE, S. E. LECROY, K. A. MULDOON, B. D.
PEERY, J. R. WILCOX, AND N. WILLIAMS-WALLS.
1986. Effects of power plant construction and op-
eration on the nesting of the loggerhead sea turtle
(Caretta caretta): 1971-84. Copeia 1986:813-816.
STONEBURNER, D. L., AND J. I. RICHARDSON. 1981.
Observations on the role of temperature in logger-
head turtle nest site selection. Copeia 1981:238-
241.
VERHEIJEN, F. J. 1985. Photopollution: Artificial
light optic spatial control systems fail to cope with.
Incidents, causations, remedies. Exp. Biol. 44:1-18.
WITHERINGTON, B. E. 1986. Human and Natural
Causes of Marine Turtle Clutch and Hatchling
Mortality and Their Relationship to Hatchling Pro-
duction on an Important Florida Nesting Beach.
M.S. Thesis, University of Central Florida, Orlan-
do, Florida.
WITHERINGTON, B. E., AND K. A. BJORNDAL. 1991.
Influences of artificial lighting on the seaward ori-
entation of hatchling loggerhead turtles (Caretta
caretta). Biol. Conserv. 55:139-149.
Spectral characteristics of hatchling sea turtle
phototaxis with implications for sea-finding behav-
iour. Copeia:In press.
WORTH, D. F., AND J. B. SMITH. 1976. Marine turtle
nesting on Hutchinson Island, Florida, in 1973.
Florida Mar. Res. Publ. 18:1-17.
Accepted: 18 March 1991
Associate Editor: William Cooper, Jr.
This content downloaded from 164.51.49.151 on Mon, 23 Jun 2014 09:29:09 AM
All use subject to JSTOR Terms and Conditions