NOTE / NOTE
The effect of artificial light on male breeding-
season behaviour in green frogs, Rana clamitans
B.J. Baker and J.M.L. Richardson
Abstract: Artificial night lighting (or ecological light pollution) is only now gaining attention as a source of long-term ef-
fects on the ecology of both diurnal and nocturnal animals. The limited data available clearly indicate that artificial light
can affect physiology and behaviour of animals, leading to ecological consequences at the population, community, and
ecosystem levels. Aquatic ecosystems may be particularly vulnerable to such effects, and nocturnally breeding animals
such as frogs may be especially affected. To address this potential, we quantify the effects of artificial light on calling and
movement behaviour in a rural population of male green frogs (Rana clamitans melanota (Rafinesque, 1820)) during the
breeding season. When exposed to artificial light, frogs produced fewer advertisement calls and moved more frequently
than under ambient light conditions. Results clearly demonstrate that male green frog behaviour is affected by the presence
of artificial light in a manner that has the potential to reduce recruitment rates and thus affect population dynamics.
´clairement artificiel de nuit (la pollution lumineuse e
´cologique) commence tout juste a
ˆtre reconnue comme
une source d’effets a
`long terme sur l’e
´cologie tant des animaux diurnes que nocturnes. Les rares donne
´es disponibles indi-
quent clairement que la lumie
`re artificielle peut affecter la physiologie et le comportement des animaux, avec des conse
quences sur l’e
´cologie de la population, de la communaute
´et de l’e
`me. Les e
`mes aquatiques pourraient e
`ces effets, surtout les animaux a
`reproduction nocturne comme les grenouilles. Afin de ve
fier ces conse
´quences potentielles, nous mesurons les effets de la lumie
`re artificielle sur les comportements d’appel et de
´placement chez une population rurale de grenouilles vertes (Rana clamitans melanota (Rafinesque, 1820)) ma
la saison de reproduction. Expose
`re artificielle, les grenouilles e
´mettent moins d’appels de signalisation et se
´placent plus fre
´quemment que sous un re
´gime de lumie
`re ambiante. Ces re
´montrent clairement que le com-
portement des grenouilles vertes ma
ˆles est affecte
´par la pre
´sence de lumie
`re artificielle d’une fac¸on qui pourrait re
les taux de recrutement et ainsi affecter la dynamique de la population.
[Traduit par la Re
While nocturnal animals experience natural variance in
light levels through changes in moon phase and cloud cover,
artificial lights are typically brighter and less diffuse than
moonlight illumination, generating very different patterns of
light change. A recently published collection of papers from
the first North American conference on the ecological ef-
fects of artificial night lighting clearly indicates both the
enormous potential for large effects of artificial light on all
taxa and the dearth of data regarding the consequences of
these anthropogenic habitat changes (Rich and Longcore
2006). Aquatic ecosystems in particular can be especially
sensitive to the effects of ecological light pollution (Rich
and Longcore 2006). The purpose of this study is to quantify
the effects of change in light environment on the breeding-
season behaviour of male green frogs, Rana clamitans mela-
nota (Rafinesque, 1820).
While astronomers began sounding an alarm regarding
light pollution 30 years ago (Berry 1976), little has been
done to quell the problem (Tyson 2002). Zoologists have
only recently begun to consider the potential effect of artifi-
cial lights on nocturnal animals (Rich and Longcore 2006).
Two aspects of artificial lighting may have important and
different consequences on animal populations: chronic in-
creases in mean light level and increased variation in light
level created by spot lighting (Buchanan 2006). Recent liter-
ature suggests that both nocturnal and diurnal animals can
Received 1 March 2006. Accepted 28 August 2006. Published on the NRC Research Press Web site at http://cjz.nrc.ca on 5 December
B.J. Baker and J.M.L. Richardson.1Department of Biological Sciences, 500 Glenridge Avenue, Brock University, St. Catharines, ON
L2S 3A1, Canada.
1Corresponding author (e-mail: firstname.lastname@example.org).
Can. J. Zool. 84: 1528–1532 (2006) doi:10.1139/Z06-142 #2006 NRC Canada
be affected by artificial lighting. For example, sea turtle
hatchlings are disoriented by artificial street lighting near
beaches (Tuxbury and Salmon 2005). Beach mice reduce
foraging in the presence of artificial lighting (Bird et al.
2004). Conversely, many insectivorous frogs, bats, and birds
take advantage of streetlamps for the large aggregation of
insects they typically attract, thereby increasing foraging
success rates (Buchanan 2006; Longcore and Rich 2006;
Rydell 2006). Robins, diurnal in nature, begin their dawn
chorus earlier in sites that have increased night lighting,
starting their chorus during true night in highly lit areas,
with no apparent relationship to actual sunrise (Miller 2006).
Anurans may be especially vulnerable to effects of artifi-
cial lighting. First, their nocturnal habits require dark-
adapted eyes, and a sudden increase of light not only dis-
rupts this but, depending on the brightness of the light, can
require a long recovery period before the eyes return to a
dark-adapted state (Cornell and Hailman 1984; Fain et al.
2001). Frogs are particularly likely to experience such dy-
namic light change because they tend to be in vegetated
areas where spotlights and vegetation are likely to co-occur,
creating a habitat that has extremes of light and dark within
a small area. Further, wind leads to movement of branches
that makes these light and dark areas unpredictable in loca-
tion. Another reason dynamic light change is likely to be
particularly relevant to frog populations is that frogs are
often in the proximity of roads (either when breeding in
ditches or when migrating to ponds across roads), and car
headlights create strong, sudden, and transient light level
Studies of the behaviour of nocturnal frogs have been fur-
ther complicated by the tendency for researchers to use
flashlights or headlamps to light their own way and to locate
animals. Several researchers have noted anecdotally that
frog calling seems unaffected by the use of a flashlight
(e.g., Martof 1953a, 1953b, 1956; Howard 1978), although
Howard (1978) also noted that his light disrupted mating in
bullfrogs (Rana catesbeiana Shaw, 1802). Further, many
frog species are known to respond to light in a variety of
ways (Jaeger and Hailman 1973). Light levels affect both
male calling behaviour and female mate choice behaviour
´ngara frogs (Physalaemus pustulosus (Cope, 1864)), as
well as calling and anti-predator behaviour in the treefrog
Smilisca sila Duellman and Trueb, 1966 (Tuttle and Ryan
1982; Tuttle et al. 1982; Rand et al. 1997). Buchanan
(1998) studied the effects of artificial light use on foraging
behaviour of the gray treefrog (Hyla chrysoscelis Cope,
1880) and observed that rapid increases in light intensity
(such as those generated by use of headlamps or flashlights)
led to decreased foraging performance.
In this study, we quantify the effects of artificial light on
the behaviour of territorial male green frogs (R. clamitans)
in the field during the breeding season, when males call to
attract mates. Data were collected from a population of frogs
in a rural area to minimize the potential of confounding ur-
ban human disturbance effects (Longcore and Rich 2004).
We used a paired design to consider the change in calling
and movement behaviour of frogs under ambient light con-
ditions (observed with night vision goggles) compared with
artificial light conditions. The purpose of our study is to as-
sess the potential for ecological consequences of artificial
lighting on anuran populations by comparing male green
frog behaviour in the presence and absence of artificial light.
Rana clamitans males were observed in a swampy area
of Wainfleet Bog, Niagara Region, Ontario, Canada
(42853’N, 79820’W). The site had thick low brush inter-
spersed with areas of open water of depths ranging from
approximately 0.2 to 1 m. Observations were taken be-
tween 2000 and 0100 from 1 July 2003 to 4 August 2003.
Cloud cover conditions were noted during observations and
the effects of lunar phase and cloud cover were considered
in subsequent analyses. Frogs were located by calls and
without artificial light (the ‘‘ATN Viper’’ IR generation 1
night vision scope was used by the researcher to see).
Once a frog was located and the researcher was positioned
for observations (approximately 2 m away), frogs were
given a 5 min habituation period (under the appropriate
light treatment). Observations were then recorded for
15 min. A 5 min habituation period is long enough to al-
low the frog’s eyes to become mostly adapted to the artifi-
cial light (Fain et al. 1996; Calvert et al. 2002), although
subtle pupillary responses to light change are known to
continue for many hours (Cornell and Hailman 1984).
We used a paired design to test for the effect of light on
behaviour while controlling for differences among males.
Frogs were observed from the same position for each treat-
ment (the position was marked with flagging tape during the
first observation to allow the observer to find the same posi-
tion for the second observation) and frogs were left undis-
turbed for 1 h between observation periods, allowing frogs’
eyes to return to a dark-adapted state (Leibrock et al. 1994)
for those frogs that received the light treatment first. Treat-
ment order was arbitrarily assigned for each frog; 9 frogs re-
ceived the ambient light treatment first and 11 frogs
received the artificial light treatment first. The calling site
of each frog was mapped and each site was used only once
for each treatment. While it is possible that some males
switched sites during the 4 weeks of observations, R. clami-
tans males typically use the same calling site for extended
periods (Martof 1953b; Wells 1977), so the probability that
the same male was sampled more than once is low.
In the ambient light treatment, no artificial light was used
and frog behaviour was observed with the same night vision
goggles used to locate frogs initially. Frogs do not detect IR
light (Jaeger and Hailman 1973; Buchanan 1998).
The artificial white light treatment was created using a
Maglite1flashlight (3 D battery size). The flashlight was
used to illuminate the frog and a circular area with a diame-
ter of approximately 1 m surrounding the frog. Throughout
the trial the frog was kept in the centre of the light. Flash-
light batteries were replaced regularly and before noticeable
dimming of the flashlight occurred. Regrettably, we were
unable to measure actual light levels in the field for each
observation. However, we recreated the same beam width
and distance, in the laboratory, subsequent to this study. We
measured light energy output using an LI-189 Quantum/
Radiometer/Photometer (LI-COR Inc., Lincoln, Nebraska);
Baker and Richardson 1529
#2006 NRC Canada
output was 0.73 mmol/m2s–1 with 4-week-old batteries and
1.66 mmol/m2s–1 with new batteries. Standard conversion,
based on the predominant wavelength energy of visible
light (Sager and McFarlane 1997), leads us to estimate
that the light output we used ranged from approximately
52 lx to 120 lx.
During each observation period, each call and movement
of the frog was recorded. Only type I advertisement calls
(Wells 1978) were observed; these calls typically consist of
a single note but can also be multi-note calls (Wells 1978;
Ramer et al. 1983). Multi-note calls were identified as dou-
ble, triple, or four-plus calls. A movement was defined as
movement of any body part, whether it resulted in displace-
ment of the male from its initial location or not. Movements
that resulted in any displacement were noted separately but
occurred too seldom to analyse separately.
Differences in mean numbers of calls and movements be-
tween artificial light and ambient light treatments were ana-
lysed using a repeated measures MANOVA, with light
treatment (artificial versus ambient light) as one repeated
measure and behaviour (movement versus calling) as a sec-
ond repeated measure. Independent factors included in the
model were moon (moonlit night versus no moon visible)
and treatment order (observations done in light first or sec-
ond). Both the number of calls and the number of move-
ments were log-transformed prior to analysis to meet
parametric assumptions. Number of calls was transformed
using ln(calls + 1) and number of movements was trans-
formed using ln(movements + 0.1), as movements occurred
at a frequency of roughly 10% of that of calls. The effect of
treatment on complex calls was further analysed separately
from single calls, using a Wilcoxon’s signed-ranks test be-
cause a high number of zeros in the artificial light data vio-
lated assumptions for a parametric test.
Frogs observed under ambient light conditions called
more (total calls = 453 in ambient light versus 256 in artifi-
cial light) and moved less (total moves = 12 in ambient light
versus 81 in artificial light) (Wilks’ statistic = 0.337,
F[1,16] = 31.42, P= 0.001). A significant interaction was
also present between behaviour (calling versus movement)
and the artificial light treatment along with significant main
effects for behaviour and artificial light (Wilks’ statistic =
0.212, F[1,16] = 59.53, P= 0.001). The significant interaction
effect occurs because movement increases in the light, while
calling decreases (Fig. 1). No other interaction terms were
statistically significant. Notably, further consideration of
only the relatively rare complex calls shows strong treatment
effects. Double calls occurred 2.9 (±1.6) times, on average,
in ambient light compared with 0.15 (±0.109) times in artifi-
cial light (Wilcoxon’s signed-ranks test; P= 0.008). The tri-
ple and four-plus calls occurred only under ambient light
(3 times and 1 time, respectively).
Treatment order had no effect on calling or movement be-
haviour (repeated measures ANOVA; F[1,16] < 0.005, P>
0.99), nor did it interact with light effects (Wilks’ statistic =
0.21, F[1,16] = 0.13, P= 0.72). Six of the 20 frogs were
observed on nights that had both a full moon and a clear
sky (observations done 13, 14, and 16 July). The behaviour
of individuals tested on these nights did not differ from
that of the other 14 frogs (repeated measures ANOVA;
F[1,16] = 0.06, P= 0.81).
Male green frogs modify behaviour in the presence of ar-
tificial light. In particular, frogs called less, performed fewer
multi-note calls, and moved more frequently in the presence
of artificial light. This is in contrast to previously published
anecdotes suggesting that light does not affect frog territorial
behaviour (Martof 1953a; Howard 1978), but in agreement
with more recent data suggesting that high levels of ambient
light may prevent frogs from chorusing (Buchanan 2006).
Many nocturnal animals modify activity and foraging
rates with changing light levels (moonlight or artificial light;
reviewed in Kronfeld-Schor and Dayan 2003). Moonlight
has long been known to affect frog reproductive physiology
and behaviour, either through changes in female receptivity
(Church 1960, 1961) or through changes in activity and
movement during the full moon (Ferguson 1960; Fitzgerald
and Bider 1974). Decreased calling with increased light may
also reflect an anti-predator response, similar to the de-
creased foraging of many nocturnal rodents during a full
moon to minimize predation risk (Kotler 1984; Wolfe and
Summerlin 1989; Topping et al. 1999; Kramer and Birney
Number of Moves
Number of Calls
Moon No Moon
Fig. 1. Calling and movement of Rana clamitans males on terri-
tories. Manipulated treatments were dark (no artificial light) and
light (flashlight-illuminated frog). All frogs (n= 20) were observed
under both treatments. Data are further divided by ambient light
conditions: moon = clear, moonlit night (n= 6); no moon = cloudy
or new-moon night (n= 14).
1530 Can. J. Zool. Vol. 84, 2006
#2006 NRC Canada
The results of this study reveal a potential for artificial
lighting to negatively affect breeding success, particularly
dynamic light changes such as those generated by car head-
lights, landscape lighting that is partially occluded by vege-
tation (leading to dark and light areas in close spatial
proximity, such that the frog moving a short distance may
encounter large differences in light level), or lighting that is
activated by a motion detector and then stays on for some
pre-set time period. The light stimulus in our study differed
somewhat from that predicted to come from cars, etc., as the
frog was exposed to the light stimulus for a 5 min acclima-
tion period prior to data collection. We suggest that our data
are conservative relative to higher frequency light changes
because the sudden onset of light leads quickly to desensiti-
zation of light receptors in the vertebrate eye, and the subse-
quent return to dark adaptation by the eye occurs far more
slowly (Fain et al. 2001). Further, frogs have a greatly re-
duced ability, compared with birds or mammals, to control
light levels reaching the retina through pupillary responses
(Cornell and Hailman 1984).
Individuals in this population were in an unlit rural loca-
tion, far from any houses, and we expect our results to re-
flect those of a population naı
¨ve to light effects. Further
work on whether populations adjust to static changes in light
levels remains to be done. Regardless of any chronic in-
crease in background light levels, dynamic changes in light
levels are still likely to occur. If such dynamic light reduces
the amount of time a frog is calling by even a small percent-
age, the effects on mating success may be large (depending
on the relationship between time spent calling and the prob-
ability of attracting a female) and hence may lead to a sig-
nificant decrease in population recruitment rates. This
potential is especially distressing given observed global am-
Thanks to M.H. Richards for comments on multiple ver-
sions of this work. This research was funded by the Depart-
ment of Biological Sciences, Brock University and the
Natural Sciences and Engineering Research Council of Can-
ada (Discovery Grant No. 261587-03 to J.M.L.R.).
Berry, R.L. 1976. Light pollution in southern Ontario. J. R. Astr.
Soc. Can. 70: 97–115.
Bird, B.L., Branch, L.C., and Miller, D.L. 2004. Effects of coastal
lighting on foraging behavior of beach mice. Conserv. Biol. 18:
Buchanan, B.W. 1998. Low-illumination prey detection by squirrel
treefrogs. J. Herpetol. 32: 270–274. doi:10.2307/1565308.
Buchanan, B.W. 2006. Observed and potential effects of artificial
night lighting on anuran amphibians. In Ecological conse-
quences of artificial night lighting. Edited by C. Rich and T.
Longcore. Island Press, Washington, D.C. pp. 192–220.
Calvert, P.D., Govardovskii, V.I., Arshavsky, V.Y., and Makino,
C.L. 2002. Two temporal phases of light adaptation in retinal
rods. J. Gen. Physiol. 119: 129–145. doi:10.1085/jgp.119.2.129.
Church, G. 1960. The effects of seasonal and lunar changes on the
breeding pattern of the edible Javanese frog, Rana cancrivora
(Gravenhurst). Treubia, 25: 215–233.
Church, G. 1961. Seasonal and lunar variation in the numbers of
mating toads in Bandung (Java). Herpetologica, 17: 122–126.
Cornell, E.A., and Hailman, J.P. 1984. Pupillary responses of two
Rana pipiens-complex anuran species. Herpetologica, 40: 356–
Fain, G.L., Matthews, H.R., and Cornwall, M.C. 1996. Dark adap-
tation in vertebrate photorecepters. Trends Neurosci. 19: 502–
507. doi:10.1016/S0166-2236(96)10056-4. PMID:8931277.
Fain, G.L., Matthews, H.R., Cornwall, M.C., and Koutalos, Y.
2001. Adaptation in vertebrate photoreceptors. Physiol. Rev. 81:
Ferguson, D.E. 1960. Observations on movements and behavior of
Bufo fowleri in residential areas. Herpetologica, 16: 112–115.
Fitzgerald, G.J., and Bider, J.R. 1974. Influence of moon phase and
weather factors on locomotory activity in Bufo americanus. Oi-
kos, 25: 338–340.
Howard, R.D. 1978. Influence of male-defended oviposition sites
on early embryo mortality in bullfrogs. Ecology, 59: 789–798.
Jaeger, R.G., and Hailman, J.P. 1973. Effects of intensity on the
phototactic responses of adult anuran amphibians: a comparative
survey. Z. Tierpsychol. 33: 352–407. PMID:4206432.
Kotler, B.P. 1984. Risk of predation and the structure of desert ro-
dent communities. Ecology, 65: 689–701. doi:10.2307/1938041.
Kramer, K.M., and Birney, E.C. 2001. Effect of light intensity on
activity patterns of patagonian leaf-eared mice, Phyllotis xantho-
pygus. J. Mammal. 82: 535–544. doi:10.1644/1545-1542(2001)
Kronfeld-Schor, N., and Dayan, T. 2003. Partitioning of time as an
ecological resource. Annu. Rev. Ecol. Syst. 34: 153–181.
Leibrock, C.S., Reuter, T., and Lamb, T.D. 1994. Dark adaptation
of toad rod receptors following small bleaches. Vision Res. 34:
2787–2800. doi:10.1016/0042-6989(94)90048-5. PMID:7975314.
Longcore, T., and Rich, C. 2004. Ecological light pollution. Front.
Ecol. Environ. 2: 191–198.
Longcore, T., and Rich, C. 2006. Synthesis. In Ecological conse-
quences of artificial night lighting. Edited by C. Rich and T.
Longcore. Island Press, Washington, D.C. pp. 413–430.
Martof, B. 1953a. Home range and movements of the green frog,
Rana clamitans. Ecology, 34: 529–543. doi:10.2307/1929725.
Martof, B. 1953b. Territoriality in the green frog, Rana clamitans.
Ecology, 34: 165–174. doi:10.2307/1930316.
Martof, B. 1956. Factors influencing size and composition of popu-
lations of Rana clamitans. Am. Midl. Nat. 56: 224–245. doi:10.
Miller, M.W. 2006. Apparent effects of light pollution on singing
behavior of American robins. Condor, 108: 130–139. doi:10.
Ramer, J.D., Jenssen, T.A., and Hurst, C.J. 1983. Size-related var-
iation in the advertisement call of Rana clamitans (Anura: Rani-
dae), and its effect on conspecific males. Copeia, 1983(1): 141–
Rand, A.S., Bridarolli, M.E., Dries, L., and Ryan, M.J. 1997. Light
levels influence female choice in Tungara frogs: predation risk
assessment? Copeia, 1997(2): 447–450. doi:10.2307/1447770.
Rich, C., and Longcore, T. (Editors). 2006. Ecological consequences
of artificial night lighting. Island Press, Washington, D.C.
Rydell, J. 2006. Bats and their insect prey at streetlights. In Ecolo-
gical consequences of artificial night lighting. Edited by C. Rich
and T. Longcore. Island Press, Washington, D.C. pp. 43–60.
Sager, J.C., and McFarlane, J.C. 1997. Radiation. In Plant growth
chamber handbook. Edited by R.W. Langhans and T.W. Tib-
bitts. Iowa Agric. Home Econ. Exp. Stn. Spec. Rep. No. 99.
Baker and Richardson 1531
#2006 NRC Canada
Topping, M.G., Millar, J.S., and Goddard, J.A. 1999. The effects of
moonlight on nocturnal activity in bushy-tailed wood rats (Neo-
toma cinerea). Can. J. Zool. 77: 480–485. doi:10.1139/cjz-77-3-
Tuttle, M.D., and Ryan, M.J. 1982. The role of synchronized call-
ing, ambient light, and ambient noise, in anti-bat predator beha-
vior of a treefrog. Behav. Ecol. Sociobiol. 11: 125–131. doi:10.
Tuttle, M.D., Taft, L.K., and Ryan, M.J. 1982. Evasive behavior of
a frog in response to bat predation. Anim. Behav. 30: 393–397.
Tuxbury, S.M., and Salmon, M. 2005. Competitive interactions be-
tween artificial lighting and natural cues during seafinding by
hatchling marine turtles. Biol. Conserv. 121: 311–316. doi:10.
Tyson, N.D. 2002. Let there be dark. Nat. Hist. 111: 34–37.
Wells, K.D. 1977. Territoriality and male mating success in green
frog (Rana clamitans). Ecology, 58: 750–762. doi:10.2307/
Wells, K.D. 1978. Territoriality in the green frog (Rana clamitans)—
vocalizations and agonistic behavior. Anim. Behav. 26: 1051–
Wolfe, J.L., and Summerlin, C.T. 1989. The influence of lunar
light on nocturnal activity of the old-field mouse. Anim. Behav.
37: 410–414. doi:10.1016/0003-3472(89)90088-2.
1532 Can. J. Zool. Vol. 84, 2006
#2006 NRC Canada