Attraction of petrels to artificial lights in the
Canary Islands: effects of the moon phase and age
class
AIRAM RODR ÍGUEZ 1 * & BENEHARO RODR ÍGUEZ 2
Estación Biológica de Doñana (CSIC), Avda. Américo Vespucio s ⁄ n, 41092 Seville, Spain
2
C ⁄ La Malecita s ⁄ n, Buenavista del Norte 38480, S ⁄ C de Tenerife, Canary Islands, Spain
1
The extent and intensity of artificial night lighting has increased with urban development
worldwide. The resulting light pollution is responsible for mortality among many Procellariiformes species which show nocturnal activity on their breeding grounds. Here, we
report light-induced mortality of Procellariiformes during a 9-year study (1998–2006) on
Tenerife, the largest island of the Canary archipelago. A total of 9880 birds from nine
species were found grounded, the majority being Cory’s Shearwaters Calonectris diomedea (93.4%). For this species the majority of grounded birds were fledglings (96.4%),
which fall apparently while leaving their nesting colony for the first time; for the smaller
species (storm-petrels) adult birds were more often grounded than fledglings. For almost
all species, grounding showed a seasonal pattern linked with their breeding cycle. Certain
phases of the moon influenced grounding of Cory’s Shearwater, with the extent of
grounding being reduced during phases of full moon. The percentage of fledglings
attracted to lights in relation to the fledglings produced annually varied between species
and years (0–1.3% for the Madeiran Storm-petrel Oceanodroma castro; 41–71% for
Cory’s Shearwater). Mean adult mortality rates also varied between species (from 0.4%
for the European Storm-petrel Hydrobates pelagicus and the Cory’s Shearwater, to 2.3%
for the Manx Shearwater Puffinus puffinus). Here we show that light-induced mortality
rates are of concern, at least for petrels and small shearwaters. Thanks to efforts involving
civil cooperation, 95% of grounded birds have been returned to the wild. To minimize
the impact of artificial lights on petrels we recommend several conservation measures:
continuing rescue campaigns, alteration of light signatures and reduction of light emissions during the fledging peaks. Furthermore, we recommend that a monitoring program
for petrel populations be implemented, as well as further studies to assess the fate of
released fledglings and continued research to address why petrels are attracted to lights.
Keywords: anthropogenic perturbation, Atlantic Ocean, light pollution, moonlight, seabirds.
Urban development has brought the need for artificial lighting of roadways, shopping centres, stadiums, homes and more. The use of artificial lights at
night has many physiological, epidemiological and
ecological consequences for wildlife (Verheijen
1985, Rich & Longcore 2006, Navara & Nelson
2007). For birds in the order Procellariiformes
(hereafter referred to as ‘petrels’) that commonly
*Corresponding author.
Email: airamrguez@ebd.csic.es
attend breeding colonies at night, artificial lighting
has been reported to cause behavioural anomalies
and increased predation rates by gulls (Ruiz &
Martí 2004, Oro et al. 2005). Moreover, artificial
lights can attract and disorientate birds (Verheijen
1981, Longcore & Rich 2004), particularly fledglings. Fledglings are attracted by lights during their
first flights to sea, either falling to the ground with
fatal injuries, being killed by predators or dying
from starvation (Imber 1975, Telfer et al. 1987, Le
Corre et al. 2002). Inexperienced birds would be
attracted to artificial lights because of incorrect
visual orientation, as visual cues for first-time
navigation to the sea at night would depend on the
extent of moon and star light (Telfer et al. 1987).
Alternatively, petrels may also tend to search for
lights to improve their chances of getting a meal,
because they feed mostly on bioluminescent squid
(Imber 1975, Klomp & Furness 1992, Montevecchi
2006).
Many petrels have undergone substantial
declines in recent times; more than 50% of the
species are threatened, mostly due to the presence
of introduced predators in their breeding grounds
and the impact of commercial fisheries at sea
(BirdLife International 2000). Although many of
these endangered species breed on islands inhabited by people, very few studies have focused on
the impacts of artificial lighting at night. In Hawaii,
where this problem was identified very early on,
lights have affected three species (Telfer et al.
1987). On Réunion Island, light-induced mortality
is known to affect four species, two of them endemic and classified as Critically Endangered (Le
Corre et al. 2003, Salamolard et al. 2007). In both
archipelagos, conservation measures have been
implemented to reduce the impact of artificial
lights, and action plans have begun to be developed
(Reed et al. 1985, Telfer et al. 1987, Ainley et al.
1997, Podolsky et al. 1998, Le Corre et al. 2002,
Salamolard et al. 2007). Apart from these two
locations,
light-induced mortality has been
reported on several other islands and for several
additional species, some of them Critically Endangered, but a precise quantification estimate of mortality is lacking (Carlile et al. 2003, Baccetti et al.
2005, Raine et al. 2007, see also Le Corre et al.
2002 and references therein).
In the Canary Islands, attraction of petrels to
artificial lights was first detected in 1990 with
the regional government subsequently responding
by carrying out awareness campaigns (Anonymous 1995). Nine petrel species have been
recorded as grounded near urban lights and
many thousands of birds have been released back
into the wild (Anonymous 1992, 1993, 1995,
Viceconsejería de Medio Ambiente 1995, Martín
& Lorenzo 2001).
The aim of this paper is to report the temporal
and spatial patterns of light-induced mortality and
its potential impact on petrel populations on Tenerife Island. Based on our results and a review of
relevant literature, recommendations and actions
are proposed to reduce attraction of these seabirds
to artificial lights.
METHODS
Study area and species involved
The Canary Islands are a volcanic archipelago
located 100–450 km off the Atlantic coast of
northwest Africa (27°37¢–29°25¢N and 13°20¢–
18°19¢W). The group comprises seven major
islands as well as some small islets and rocks. Tenerife is the largest island (2034 km2 and up to
3718 m asl), and is situated in the central part of
the archipelago (Fig. 1). Approximately 800 000
people live on Tenerife, the majority of whom are
concentrated along the coast (Morales & Pérez
2000).
Urbanization and infrastructure have
increased considerably during the last 40 years,
currently occupying 5.7% of the total area of the
island.
Seven petrel species breed regularly in the
Canarian archipelago, six of them in Tenerife: Bulwer’s Petrel Bulweria bulwerii, Cory’s Shearwater
Calonectris diomedea, Manx Shearwater Puffinus
puffinus, Macaronesian Shearwater Puffinus baroli,
Madeiran Storm-petrel Oceanodroma castro and
European Storm-petrel Hydrobates pelagicus
Figure 1. Geographic distribution and importance by municipality of light-induced mortality of Procellariiforms on Tenerife
Island during the period 1998–2006. Numbers show the percentage of petrels found during the study, lines mark the municipal limits, the dotted line marks 400 m asl, and the shaded
areas demarcate urban areas.
Table 1. Distribution, numbers, breeding success (proportion of eggs producing fledglings) and IUCN category of threat (Madroño
et al. 2004) of the grounded species found on Tenerife Island from January 1998 to December 2006.
Breeding pairs in
Distribution
IUCN
category
Canary
Islands
Tenerife
Breeding
success
L, Gc, T, P, G, H
All islands
T, P
Non-breeder
L, T, G, H
L, T
Non-breeder
L
EN
VU
EN
–
EN
VU
–
EN
1000a
30 000c
220f
–
400d
600d
–
60k
400a
2500d
20g
–
70d
> 100d
–
–
0.74b
0.75e
0.61h
–
0.59i
0.76j
–
–
L, F, Gc, T, G, H
VU
> 1000l
?
0.53m
Species
Bulwer’s Petrel Bulweria bulwerii
Cory’s Shearwater Calonectris diomedea borealis
Manx Shearwater Puffinus puffinus
Great Shearwater Puffinus gravis
Macaronesian Shearwater Puffinus baroli baroli
Madeiran Storm-petrel Oceanodroma castro
Leach’s Storm-petrel Oceanodroma leucorhoa
White-faced Storm-petrel Pelagodroma marina
hypoleuca
European Storm-petrel Hydrobates pelagicus
L, Lanzarote including the northern islets; F, Fuerteventura; Gc, Gran Canaria; T, Tenerife; P, La Palma; G, La Gomera; H, El
Hierro; EN, Endangered; VU, Vulnerable.
a
Herná ndez et al. (1990b), bNunes and Vicente (1998), cMartı́n and Lorenzo (2001), dLorenzo (2007), eMonteiro et al. (1996), fMartı́n
et al. (1989), gHernández et al. (1990a), hBrooke (1990), iOliveira and Moniz (1995), jBolton et al. (2004), kRodrı́guez et al. (2003),
l
Nogales et al. (1993), mCadiou (2001).
(Table 1). Other non-breeding petrel species are
regularly sighted in Canary waters (Martín &
Lorenzo 2001) and may be susceptible to attraction from artificial lights.
Collection of the grounded birds and
rescue campaigns
Awareness campaigns carried out by Cabildo Insular de Tenerife have involved local news media,
talks in primary schools and high schools, stickers,
posters, and public release of rescued birds, among
other actions. As part of the education initiative,
people were asked to collect fallen birds and telephone the ‘Centro de Recuperación de Fauna
Silvestre La Tahonilla’ (onwards CRFS-La Tahonilla), depending on the island government (Cabildo
Insular de Tenerife). Birds were collected by staff
of the CRFS-La Tahonilla, examined and identified.
Birds were ringed (except Cory’s Shearwater fledglings) and released at the coast during daylight to
avoid being attracted again by lights. Injured birds
were either held for rehabilitation or euthanized.
For each bird, date, location, age class, phase of the
moon and date and location of release were
registered. Age class (adults or fledglings) was
determined using a combination of feather coloration and wear, biometrics, and the presence of
down or an incubation patch. Analysis of the
geographic distribution of the light-induced
mortality was based on 31 municipalities of
island (Fig. 1). As a proxy for light pollution,
estimated the number of inhabitants and
urbanized area (ha) below 400 m of altitude
each municipality. The data analysed spanned
years 1998–2006.
the
we
the
of
the
Effects of the phase of the moon
The effect of the phase of the moon on the number of petrels was only assessed for Cory’s Shearwater due to the small numbers of birds found
grounded for the remaining species. To determine
the effect of the phase of the moon on numbers of
fledglings grounded, we assigned a value (0–14),
where 0 corresponded to new moon and 14 to full
moon. We ran a Spearman rank correlation
between this score and the number of fledglings
during the period 1998–2006. In addition, the
moon phase was estimated as a binary variable
with periods of full moon (the exact day of full
moon ± 2 days) and new moon plus intermediate
phases (all other days). This permitted us to test
for differences in the mean number of petrels
collected between these periods using a Mann–
Whitney U-test. To avoid background noise, we
only used the period between 25 October and
25 November, when more than 95% of Cory’s
Shearwater fledglings were grounded as a consequence of artificial lights.
Impact of light-induced mortality on the
petrel population
FP ¼ BS x BP
then
Without the aid of rescue campaigns, most of the
grounded petrels would either be killed by predators or die of starvation and injuries (Imber 1975,
Telfer et al. 1987, Le Corre et al. 2002). Even if
the birds are not injured, they are unable to take
off as many species need a long runway to take off.
Therefore, we assumed that all grounded birds
would die in the absence of intervention.
Petrel populations often include a large proportion of ‘floaters’, immature birds and adult nonbreeders (Warham 1990). The proportion of floaters with respect to the total population is not
known in the Canary Islands. We estimate annual
adult mortality as the number of adult birds
grounded relative to the size of the population,
assuming a rate of 37% for floaters (immature and
non-breeder adults; Ristow 1998) and number of
breeders given in Table 1. Furthermore, we estimate a minimum adult mortality rate based on the
total number of breeders (Table 1). To estimate
the impact of light-induced mortality at fledging
we had to determine the total number of fledglings
lost (FL) and the total number of fledglings produced annually by the population (FP). The proportion of fledglings lost (PL) is then
PL ð%Þ ¼ FL x 100=ðBS x BPÞ
The total annual number of fledglings attracted
by lights (FL) is hard to estimate and depends on
several factors, especially the efficiency of the
awareness campaign conducted each year before
the fledging season. For the purpose of our study
we used estimates available in the literature, and
we assumed BP was constant among years. BS has
not been estimated for these species in the Canary
Islands, therefore we used published information
from other Macaronesian archipelagos whenever
possible (Table 1). Despite all these sources of
constraints, estimates give us a rough idea of the
magnitude of the problem.
RESULTS
Species affected, age-related variation
in fallout, and success of rescue
campaigns
Over the 9 years of study, at least 9880 petrels
from nine species (Table 2) were found grounded
in urban areas of Tenerife. Cory’s Shearwater represented 93.4% of the birds recovered. In addition
to the local breeding species, three non-breeding
petrel species were also recovered: Great Shearwater Puffinus gravis, Leach’s Storm-petrel Oceanodroma leucorhoa and White-faced Storm-petrel
Pelagodroma marina hypoleuca. The mean number
of birds recovered annually was 1098 ± 212 (sd).
Most grounded birds were fledglings (93.9%)
which probably were leaving the nests for the first
PL ð%Þ ¼ FL=FP x 100
The number of fledglings produced annually
depends on the breeding success (BS, number of
birds which successfully fledge out of the number
of eggs laid per female), and on the number of
breeding pairs (BP). Therefore if the number of
fledglings produced each year is:
Table 2. Number of petrels found grounded on the island of Tenerife from January 1998 to December 2006.
Species
1998
1999
2000
2001
2002
2003
2004
2005
2006
Total
Frequency (%)
Bulweria bulwerii
Calonectris diomedea
Puffinus puffinus
Puffinus gravis
Puffinus baroli
Oceanodroma castro
Oceanodroma leucorhoa
Pelagodroma marina
Hydrobates pelagicus
Total
38
886
3
1
25
1
3
3
0
960
41
873
3
1
27
3
5
7
2
962
51
890
1
0
16
4
4
4
1
973
34
869
3
0
29
5
4
7
3
954
39
1311
2
2
10
4
3
3
3
1377
34
806
3
0
10
3
17
5
1
879
42
1252
3
0
12
1
14
3
1
1328
31
1359
1
0
7
7
6
4
0
1415
30
985
4
0
8
3
2
1
1
1035
340
9231
23
4
144
31
58
37
12
9880
3.4
93.4
0.2
0.0
1.5
0.3
0.6
0.4
0.1
Table 3. Total number (and %) of grounded petrels classified
by age group, and the proportion of individuals released
successfully during the study period 1998–2006 on the island
of Tenerife.
Species
Fledglings
Adults
Released (%)
Bulweria bulwerii
Calonectris
diomedea
Puffinus puffinus
Puffinus gravis
Puffinus baroli
Oceanodroma
castro
Oceanodroma
leucorhoa
Pelagodroma
marina
Hydrobates
pelagicus
Total
199 (58.6)
8932 (96.4)
141 (41.4)
299 (3.6)
88.2
95.2a
10
0
130 (90.3)
5
13
4
14 (9.7)
26
87.0
75.0
95.1
90.3
0
58
75.9
0
37
86.5
0
12
91.7
9276 (93.9)
604 (6.1)
94.7
In the case of Cory’s Shearwater Calonectris diomedea,
release data encompass a total of 8379 individuals.
a
time (Table 3). Fledglings of larger Procellariiformes (Bulwer’s Petrel and the shearwaters) were
more heavily affected than those of the smaller
species (storm-petrels). Fewer than 6% of the
birds were found dead, or dying and were euthanized. Thus rescue campaigns released more than
94% of the grounded birds.
Seasonal aspects and effects of the
phase of the moon
Pooling all species together, light-induced mortality
occurred in every month of the year (Fig. 2). The
highest numbers were recorded in October and
November (16.6% and 74.7%, respectively), coinciding with the first flights of Cory’s Shearwater
fledglings. The majority of species exhibited a clear
seasonal pattern, which was related to their breeding phenologies. For five of six breeding species in
Tenerife the peak of light-induced mortality
occurred during the period when fledglings undertook their first flights (Fig. 2).
A negative correlation was detected between
the phase of the moon and the number of Cory’s
Shearwater fledglings grounded through the period
25 October to 25 November of all studied years
(Spearman rank correlation, r = )0.26, P < 0.001,
n = 288). Fewer Cory’s Shearwater fledglings were
found daily during full moon than at any other
stage of the lunar cycle. Statistically significant differences were reached for years in which a full
moon did not coincide with the usual fledging peak
(Table 4). A lower number of fledglings were
found when the full moon coincided with the
mean date of first flight, but no statistical significant correlation was detected (see Fig. 3).
Geographic distribution of light-induced
mortality
Geographically, the grounding of birds was more
intensive in coastal cities and tourist resorts and
the majority of rescues were carried out at an altitude of less than 400 m asl. Thus, more than 65%
of birds were found in the municipalities of the
southwest coast (Fig. 1). The distribution of the
birds found was related to the extent of urban
areas: statistical significance and positive correlations were detected between the number of inhabitants and urban areas below 400 m asl, and the
number of birds grounded (log (x + 1) transformed
data; Pearson correlation coefficient, r = 0.63,
P < 0.001, n = 31 and r = 0.65, P < 0.001, n = 30;
respectively).
Estimation of the light-induced mortality
at the population level
The percentage of fledglings affected by artificial
lights varied highly between species and years, 95%
confidence limits ranging between 0% and 60.5%.
Approximately half of the Cory’s Shearwater fledglings are affected by ‘light dazzling’, followed in
importance by Macaronesian Shearwater fledglings
(33.9 ± 19.9%; Table 5). In general, annual adult
mortality was low, ranging between 0% and 5% of
the adult population. Adult Manx Shearwaters
were the most affected species, whereas adult
Cory’s Shearwaters and European Storm-petrels
were least affected. The estimates of adult mortality
for the two different scenarios, only assuming
breeding individuals and assuming that 37% of the
individuals are floaters, are presented in Table 6.
DISCUSSION
Magnitude, seasonality and distribution
of grounded birds
Nine petrel species were affected by light pollution
in the study area. This represents the highest
A. Rodrı´guez & B. Rodrı´guez
175
Bulweria bulwerii
6000
150
4500
125
3000
Calonectris diomedea
1500
100
100
75
75
50
50
25
0
25
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
12
0
Puffinus baroli
10
100
8
80
6
60
4
40
2
20
0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
120
Puffinus puffinus
Number of birds found grounded
304
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Oceanodroma castro
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
6
25
5
20
4
Oceanodroma leucorhoa
15
3
10
2
5
1
0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
0
6
Hydrobates pelagicus
Pelagodroma marina
5
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
3
4
2
3
2
1
1
0
0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Figure 2. Monthly distribution of the light-induced mortality of petrels in Tenerife Island during the period 1998–2006. Y-axis shows the
number of birds grounded: dark bars represent adults and lined bars fledglings. Horizontal bars indicate phenologies of petrels in the
Canary Islands (white: not present in the Canary Islands or present in very low numbers; black: colony attendance or at least present at
sea in the proximity of the islands; grey: incubation; lines: chick rearing; after Martı́n & Lorenzo 2001). Phenologies of Puffinus puffinus,
Puffinus baroli, Oceanodroma castro, Pelagodroma marina and Hydrobates pelagicus are not well known in the Canary Islands.
Petrel attraction to lights
305
Table 4. Mean number (± sd) of birds found daily during the fledging peak of Cory’s Shearwater (25 October–25 November) in
relation to the phase of the moon.
Year
1998
1999
2000
2001
2002
2003
2004
2005
2006
Total
Full moon (n)
Mean ± sd
12.8
1.9
19.8
16.6
5.0
22.0
7.43
5.6
36.0
13.12
±
±
±
±
±
±
±
±
±
±
Intermediate moon phases (n)
Mean ± sd
7.5 (5)
2.6 (8)
17.0 (5)
5.1 (5)
0.7 (5)
8.0 (5)
7.9 (7)
5.6 (5)
15.7 (5)
13.11 (50)
31.2
38.0
27.2
27.8
45.3
24.0
46.0
47.2
27.5
34.78
±
±
±
±
±
±
±
±
±
±
28.7 (27)
31.1 (24)
29.8 (27)
31.7 (27)
40.9 (27)
23.9 (27)
42.4 (25)
47.6 (27)
18.7 (27)
34.39 (238)
Mann–Whitney U test (P )
54.0 (0.51)
17.0 (< 0.001)***
66.0 (0.94)
54.5 (0.50)
27.0 (0.035)*
58.0 (0.62)
34.5 (0.016)*
32.5 (0.069)
52.0 (0.42)
3932.0 (< 0.001)***
The number of occurrences of full moon (the exact day of full moon ± 2 days) and intermediate moon phases (all other days) respectively in the period studied are indicated by (n).
* and *** significant at a = 0.05 and a = 0.001 levels respectively.
number of species affected by light pollution in
any studied archipelago to date. The absolute number of grounded birds is lower than in the Hawaiian archipelago (Telfer et al. 1987, Podolsky et al.
1998), but higher than in Réunion or the Malta
Islands (Le Corre et al. 2002, Raine et al. 2007).
The numbers of grounded Cory’s Shearwaters in
the Canary Islands are only comparable with the
estimates for Newell’s Shearwater Puffinus auricularis newelli on Kauai, Hawaii (Telfer et al. 1987,
Podolsky et al. 1998).
Although the species involved are different,
there are similarities between our results and those
from other studies. Mortality shows a seasonal pattern that varies according to the breeding schedules
of each species, and mainly coincides with the date
when fledglings undertake their first flights to the
sea (Telfer et al. 1987, Le Corre et al. 2002). Interspecific differences in the number of grounded
birds appear to be correlated with a differential
degree of attraction to lights for each species, even
when differences in the population size of each
species are taken into account. As in other studies,
the smaller species were least affected (Telfer et al.
1987, Le Corre et al. 2002).
The geographical pattern of light-induced mortality could be a consequence of the presence of
important colonies and mass tourism developments in the southwestern part of the island.
Lights from towns that are not near the coast or
those at higher altitudes (above c. 400 m asl) did
not attract petrels. This might be related to the distance to the breeding colonies, as reported for
Réunion Island, where 35% of grounded Barau’s
Petrels Pterodroma baraui were found in a little
town at 1500 m asl located just below an important breeding colony (Le Corre et al. 2002).
The effect of the phase of the moon on
grounding birds
The maximum number of grounded fledglings varies annually with the lunar phase, producing a
bimodal distribution around the date of full moon,
or shifting the peak of fledgling grounding. As a
rule, during full moon (± 2 days) a lower number
of grounded fledglings were recovered than at
other phases of the moon. This could be a consequence of two non-mutually exclusive factors. On
the one hand, full moon could inhibit the first
flight of young shearwaters. Indeed, during fullmoon nights, petrel activity at breeding colonies
decreases (Imber 1975, Bretagnolle 1990), apparently as an anti-predator mechanism (Watanuki
1986, Mougeot & Bretagnolle 2000, Oro et al.
2005). On the other hand, fledglings might fall in
lower numbers because a greater ambient light
reduces attraction towards artificial lights (Reed
et al. 1985, Montevecchi 2006). If the latter
hypothesis were correct, we would expect the
number of grounded fledglings to vary among years
according to the period elapsed between full moon
and the middle of the fledging peak. Lower
weights and longer wings of grounded fledglings
were documented on Réunion Island during the
last part of the fledging period in years when a full
306
A. Rodrı´guez & B. Rodrı´guez
A
100
80
1998
80
60
60
40
40
20
20
0
0
20 Oct 27 Oct
100
3 Nov 10 Nov 17 Nov 24 Nov
1999
20 Oct 27 Oct
80
3 Nov 10 Nov 17 Nov 24 Nov
2000
80
60
60
40
40
20
20
0
Number of fledglings found grounded
20 Oct 27 Oct
80
3 Nov 10 Nov 17 Nov 24 Nov
20 Oct 27 Oct
120
2001
3 Nov 10 Nov 17 Nov 24 Nov
2002
100
60
80
60
40
40
20
20
0
20 Oct 27 Oct
60
3 Nov 10 Nov 17 Nov 24 Nov
20 Oct 27 Oct
150
2003
3 Nov 10 Nov 17 Nov 24 Nov
2004
125
100
45
75
30
50
15
25
0
0
20 Oct 27 Oct
150
3 Nov 10 Nov 17 Nov 24 Nov
20 Oct 27 Oct
3 Nov 10 Nov 17 Nov 24 Nov
60
2005
2006
125
45
100
30
75
50
15
25
0
0
20 Oct 27 Oct
3 Nov 10 Nov 17 Nov 24 Nov
20 Oct 27 Oct
3 Nov 10 Nov 17 Nov 24 Nov
Figure 3. Frequency distribution of Cory’s Shearwater Calonectris diomedea fledglings found grounded during the study period
1998–2006 (‘A’ indicates all years; mean ± sd) and the different studied years. Shaded areas represent 5-day periods indicating the
exact day of full moon ± 2 days.
moon coincided with the peak of the fledging period. This pattern seems to support the first
hypothesis (inhibition of first flight by the full
ª 2009 The Authors
Journal compilation ª 2009 British Ornithologists’ Union
moon, Le Corre et al. 2002). We did not find any
evidence against this hypothesis in our study. Further studies to clarify these patterns are warranted.
Petrel attraction to lights
307
Table 5. Numbers of fledglings grounded (minimum and maximum values per annum, and mean ± sd) and estimates of the numbers
of fledglings affected by light pollution (95% confidence limits and mean ± sd) for each breeding species on Tenerife Island during the
period 1998–2006.
Fledglings
Species
Min
Max
Bulweria bulwerii
Calonectris diomedea
Puffinus puffinus
Puffinus baroli
Oceanodroma castro
Hydrobates pelagicus
16
775
0
4
0
0
31
1327
2
28
1
0
Percentage of affected fledglings (%)
Mean ± sd
22.1 ±
992.7 ±
1.1 ±
14.0 ±
0.4 ±
0
7.5
216.8
0.9
8.2
0.5
Lower limit
Upper limit
6.4
45.4
4.1
20.9
0.1
0
8.6
60.5
14.1
46.9
1.0
0
Mean ± sd
7.5 ±
52.9 ±
9.1 ±
33.9 ±
0.6 ±
0
1.7
11.6
7.6
19.9
0.7
Table 6. Estimates (95% confidence intervals and mean ± sd) of the proportion of adults affected by dazzling from lights for each
breeding species in Tenerife Island during the period 1998–2006, under two different scenarios.
Percentage of affected adults: only
breeding individuals
Percentage of affected adults:
breeding and floaters individuals
Species
95% CI
Mean ± sd
95% CI
Bulweria bulwerii
Calonectris diomedea
Puffinus puffinus
Puffinus baroli
Oceanodroma castro
Hydrobates pelagicus
(1.5, 2.4)
(0.6, 0.8)
(2.8, 4.5)
(0.6, 1.6)
(0.9, 1.9)
(0.3, 1.0)
1.96 ±
0.7 ±
3.6 ±
1.1 ±
1.4 ±
0.7 ±
(1.0, 1.5)
(0.4, 0.5)
(1.7, 2.8]
(0.4, 1.0)
(0.6, 1.2)
(0.2, 0.7)
0.7
0.1
1.3
0.7
0.8
0.6
Mean ± sd
1.2
0.4
2.3
0.7
0.9
0.4
±
±
±
±
±
±
0.4
0.1
0.8
0.5
0.5
0.4
In the first scenario, we only considered breeders. In the second one, we considered 37% of adult population to be composed of floaters (immatures and non-breeding adults, Ristow 1998).
Effects on population size
Estimates of the number of birds affected may be
biased because of several sources of error. First, we
assume that the number of breeding pairs did not
change during the course of our study period.
Secondly, the number of breeding pairs reported
in the literature may not be reliable due to the
secretive breeding behaviour of petrels (nocturnal
activity, underground nests, communal burrows
with only one entrance, etc.), and the rugged relief
of Tenerife which makes it very difficult to provide
an accurate estimate of population size (Gregory
et al. 2004). Thirdly, the breeding success of petrels from Tenerife may be different from other
populations studied and probably changes annually.
Also, given that not all birds were banded, we
could have recaptured some of them. However,
recaptures of banded birds were negligible (n = 18
of > 10 000) on Kauai (Telfer et al. 1987).
Although the number of grounded birds on Tenerife per year is striking, actual numbers impacted
are certainly higher than the numbers reported
here because it is assumed that not all grounded
birds are recovered.
Despite these biases, the rough estimates
reported in this paper provide valuable information relevant to the population dynamics of the
studied species (Le Corre et al. 2002, Day et al.
2003). Petrel species are long-lived organisms
(Warham 1990) and they are more vulnerable to
changes in adult mortality rates than changes in
fledgling mortality rates (Simons 1984, Jones
2002). Although adult mortality is very low, mortality rates reported on Tenerife could be high
enough to have severe effects on the population
dynamics of some species, especially for the rare or
threatened ones.
Populations of the Macaronesian
and Manx
Shearwaters deserve particular attention given
their rarity, their local threatened status and
because Tenerife lies at the southern extremity of
their breeding ranges. The Manx Shearwater
appears to be more severely affected than the
ª 2009 The Authors
Journal compilation ª 2009 British Ornithologists’ Union
308
A. Rodrı´guez & B. Rodrı´guez
Macaronesian Shearwater. We think it could be due
to their different breeding habitats, the former
being an inland breeder, where it may be more vulnerable to light pollution (Rodríguez et al. 2008).
The Macaronesian Shearwater breeds primarily in
coastal cliffs and, like the Wedge-tailed Shearwater
Puffinus pacificus at Réunion Island, fledglings do
not cross over the illuminated urban areas to reach
the sea, thus decreasing the risk of light attraction
(Le Corre et al. 2002). In addition, our data suggest
that the number of Macaronesian Shearwater
breeding pairs is probably slightly higher than that
stated in the literature (Lorenzo 2007), although
this number has probably decreased during the
study period if we take into account a decline in
the number of rescued birds by CRFS-La Tahonilla.
Assessment of rescue and awareness
campaigns
Thanks to rescue campaigns, 94.7% of grounded
birds were released during the study period 1998–
2006. These campaigns have been carried out over
16 years (Anonymous 1995), and assuming an
average of 1098 downed petrels per year, this
means that over 16 000 birds have been released.
However, the fate of released fledglings remains
unclear. Fledglings of Cory’s Shearwater have been
observed to be unable to fly at sea due to the lack
of waterproofing of their feathers (A. Rodríguez
and B. Rodríguez pers. obs.). It is also possible that
grounded birds are debilitated in some manner
after their disorientated flight or by the subsequent
stress of human handling.
Why are petrels attracted by lights?
Very little is known about why birds are attracted
to artificial lights (Verheijen 1985), but some
hypotheses have been proposed. Nocturnal seabirds often have larger eyes than diurnal ones, and
their retinas have a preponderance of rods, thus
enabling these species to be more sensitive to light
(McNeil et al. 1993). Furthermore, species that
consume bioluminescent prey appear to be
affected to a greater extent
by artificial light
(Montevecchi 2006).
Inexperience may also play an important role.
Petrel behaviour to reach the ocean from their
nests must be innate, as they are abandoned by
their parents some days before fledging (Warham
1990). They may be attracted to lights because this
ª 2009 The Authors
Journal compilation ª 2009 British Ornithologists’ Union
behaviour could increase their chances of finding
food under natural conditions (Imber 1975, Klomp
& Furness 1992). However, a more plausible explanation may be related to the fact that visual cues
to reach the ocean during their first flight depend
on moon or star light in a natural environment
(Telfer et al. 1987). Young birds are attracted to
illuminated areas because they need visual cues for
navigation. Supporting this idea is the fact that
many young petrels do not fly directly into lights,
but rather circle illuminated areas, becoming disoriented over time, and finally landing stunned or
exhausted (Telfer et al. 1987, A. Rodríguez and
B. Rodríguez pers. obs.).
Open to question are why some adults or individuals of non-breeding species are also found
grounded as a consequence of artificial lights. In
Tenerife, Great Shearwater, Leach’s Storm-petrel
and White-faced Storm-petrel have been found
grounded under artificial lights in coastal areas.
The closest breeding colonies are on the Selvagem
Islands
for
the
White-faced
Storm-petrel
(> 170 km away) and thousands of kilometres
away for the other two species. One possibility is
that birds with no previous experience of light
pollution are involved in these cases.
Conservation measures
Although research is urgently needed to develop
effective mitigation measures for light attraction in
birds (Verheijen 1985, Montevecchi 2006), several
measures could be implemented immediately.
These include reducing artificial lighting sources
and changing light signatures. The current location
of artificial light sources must be revised and some
could be switched off if they are unnecessary. Furthermore, an amount of unused artificial lighting is
projected into space due to inadequate orientation
of light sources, and this should be eliminated. The
peak concentrations of grounded birds are around
the fledgling periods of the different species (Reed
et al. 1985, Telfer et al. 1987, Le Corre et al. 2002,
this study), so intensive light reduction could be
undertaken during these periods. Some changes in
light signature could reduce bird attraction to artificial lights (Reed 1987, Jones & Francis 2003,
Baccetti et al. 2005, van de Laar 2007). Shielding
lights to avoid upward radiation decreased the
number of grounded petrels by > 40% on Kauai,
Hawaii (Reed et al. 1985). More experimental
studies testing light signatures and bird attraction
Petrel attraction to lights
to lights are necessary to develop evidence-based
management measures (Reed 1987, van de Laar
2007). On the other hand, awareness campaigns
are necessary to help the public to become aware
of light pollution and demand solutions from local
and national governments. At least until light pollution is mitigated, rescue campaigns should be
continued indefinitely. Great efforts must be made
during the peak fledging periods of the most
endangered species such as Macaronesian Shearwater and Manx Shearwater. In addition to these
considerations, a monitoring program needs to be
established to detect population changes on the
Canary Islands.
Special thanks to all the anonymous people who
kindly helped rescue grounded birds and to the staff
of Centro de Recuperación de Fauna Silvestre ‘La
Tahonilla’ (Cabildo Insular de Tenerife), who provided
us with data of dazzled birds. Fernando Pacios helped
us to make the Tenerife map. Thanks to Luis Cadahía, Alejandro Martínez-Abraín, Matthieu Le Corre,
Andrea Erichsen, Rauri Bowie and four anonymous
reviewers for comments on the early drafts of this
manuscript.
RE FE R E NC E S
Ainley, D., Podolsky, R., Deforest, L. & Spencer, G. 1997.
New insights into the status of the Hawaiian petrel on
Kauai. Colon. Waterbirds 20: 24–30.
Anonymous 1992. Protección de pardelas en Canarias. La
Garcilla 85: 2.
Anonymous 1993. Recuperadas 60 pardelas chicas. Quercus
93: 39.
Anonymous 1995. Rescate de pardelas en Canarias. La Garcilla 92: 6.
Baccetti, N., Sposimo, P. & Giannini, F. 2005. Artificial lights
and mortality of Cory’s Shearwater Calonectris diomedea
on a Mediterranean island. Avocetta 29: 89–91.
BirdLife International 2000. Threatened Birds of the World.
Barcelona-Cambridge: Lynx Editions-BirdLife International.
Bolton, M., Medeiros, R., Hothersall, B. & Campos, A.
2004. The use of artificial breeding chambers as a conservation measure for cavity-nesting procellariifom seabirds: a
case study of the Madeiran Storm-petrel (Oceanodroma
castro). Biol. Conserv. 116: 73–80.
Bretagnolle, V. 1990. Effet de la lune sur l’activité des pétrels
(classe Aves) aux ı̂les Salvages (Portugal). Can. J. Zool.
68: 1404–1409.
Brooke, M. 1990. The Manx Shearwater. London: T. & A.D.
Poyser.
Cadiou, B. 2001. The breeding biology of the European
Storm-petrel Hydrobates pelagicus in Brittany, France. Atl.
Seabirds 3: 149–164.
Carlile, N., Priddel, D., Zino, F., Natividad, C. & Wingate,
D.B. 2003. A review of four successful recovery programmes
309
for threatened, sub-tropical petrels. Mar. Ornithol. 31: 185–
192.
Day, R.H., Cooper, B.A. & Telfer, T.C. 2003. Decline of
Townsend’s (Newell’s) Shearwaters (Puffinus auricularis
newelli) on Kauai, Hawai. Auk 120: 669–679.
Gregory, R.D., Gibbons, D.W. & Donald, P.F. 2004. Bird
census and survey techniques. In Sutherland, W.J., Newton, I. & Green, R.E. (eds) Bird Ecology and Conservation;
A Handbook of Techniques: 17–56. Oxford: Oxford University Press.
Herná ndez, E., Nogales, M., Quilis, V. & Delgado, G. 1990a.
Nesting of the Manx Shearwater (Puffinus puffinus Brü nnich, 1764) on the Island of Tenerife (Canary Islands).
Bonn. Zool. Beitr. 41: 59–62.
Herná ndez, E., Martı́n, A., Nogales, M., Quilis, V., Delgado,
G. & Trujillo, O. 1990b. Distribution and status of Bulwer’s
Petrel (Bulweria bulwerii Jardine & Selby, 1828) in the
Canary Islands. Bol. Mus. Mun. Funchal 42: 5–14.
Imber, M.J. 1975. Behaviour of petrels in relation to the moon
and artificial lights. Notornis 22: 302–306.
Jones, C. 2002. A model for the conservation management of
a ‘secondary’ prey: Sooty Shearwater (Puffinus griseus)
colonies on mainland New Zealand as a case study. Biol.
Conserv. 108: 1–12.
Jones, J. & Francis, C.M. 2003. The effects of light characteristics on avian mortality at lighthouses. J. Avian Biol. 34:
328–333.
Klomp, N.I. & Furness, R.W. 1992. Patterns of chick feeding
in Cory’s Shearwaters and the associations with ambient
light. Colon. Waterbirds 15: 95–102.
van de Laar, F.J.T. 2007. Green Light to Birds. Investigation
into the Effect of Bird-Friendly Lighting. Report NAM locatie
L15-FA-1. Assen, The Netherlands: Nederlandse Aardolie
Maatschappij.
Le Corre, M., Ollivier, A., Ribes, S. & Jouventin, P. 2002.
Light-induced mortality of petrels: a 4-year study from
Réunion Island (Indian Ocean). Biol. Conserv. 105: 93–102.
Le Corre, M., Ghestemme, T., Salamolard, M. & Couzi, F.-X.
2003. Rescue of the Mascarene Petrel, a critically endangered seabird of Réunion Island, Indian Ocean. Condor
105: 387–391.
Longcore, T. & Rich, C. 2004. Ecological light pollution. Front.
Ecol. Environ. 2: 191–198.
Lorenzo, J.A. (ed.) 2007. Atlas de las aves nidificantes en el
archipiélago canario (1993–2003). Madrid: Direcció n General de Conservación de la Naturaleza-SEO ⁄ BirdLife.
Madroñ o, A., Gonzá lez, C. & Atienza, J.C. (eds) 2004. Libro
Rojo de las Aves de España. Madrid: Direcció n General
para la Conservación de la Biodiversidad-SEO ⁄ BirdLife.
Martı́n, A. & Lorenzo, J.A. 2001. Aves del Archipiélago
Canario. La Laguna: Lemus.
Martı́n, A., Delgado, G., Nogales, M., Quilis, V., Trujillo, O.,
Hená ndez, E. & Santana, F. 1989. Premiè res donné es sur
la nidification du Puffin des Anglais (Puffinus puffinus), du
Pé trel-Fré gate (Pelagodroma marina) et de la Sterne de
Dougall (Sterna dougallii) aux ı̂les Canaries. Oiseau Rev.
Fr. Ornithol. 59: 73–83.
McNeil, R., Drapeau, P. & Pierotti, R. 1993. Nocturnality in
colonial waterbirds: occurrence, special adaptations, and
suspected benefits. In Power, D.M. (ed.) Current Ornithology: 187–245. New York, NY: Plenum Press.
ª 2009 The Authors
Journal compilation ª 2009 British Ornithologists’ Union
310
A. Rodrı´guez & B. Rodrı´guez
Monteiro, L.R., Ramos, J.A., Furness, R.W. & Nevo, A.J.
1996. Movements, morphology, breeding, molt and feeding
of seabirds in the Azores. Colon. Waterbirds 19: 82–97.
Montevecchi, W.A. 2006. Influences of artificial light on marine birds. In Rich, C. & Longcore, T. (eds) Ecological Consequences of Artificial Night Lighting: 94–113. Washington,
D.C.: Island Press.
Morales, G. & Pérez, R. 2000. Gran Atlas temático de Canarias. Santa Cruz de Tenerife: Editorial Interinsular Canaria.
Mougeot, F. & Bretagnolle, V. 2000. Predation risk and
moonlight avoidance in nocturnal seabirds. J. Avian Biol.
31: 376–386.
Navara, K.J. & Nelson, R.J. 2007. The dark side of light at
night: physiological, epidemiological and ecological consequences. J. Pineal Res. 43: 215–224.
Nogales, M., Martı́n, A., Quilis, V., Herná ndez, E., Delgado,
G. & Trujillo, O. 1993. Estatus y distribució n del Paı́ñ o
Comú n (Hydrobates pelagicus) en las Islas Canarias. In
Aguilar, J.S., Monbailliu, X. & Paterson, A.M. (eds) Estatus
y Conservación de Aves Marinas: 15–24. Madrid:
SEO ⁄ BirdLife-Medmaravis.
Nunes, M. & Vicente, L. 1998. Breeding cycle and nestling
growth of Bulwer’s Petrel on the Desertas Islands, Portugal.
Colon. Waterbirds 21: 198–204.
Oliveira, P. & Moniz, P. 1995. Population size, breeding chronology, annual cycle and effects of inter-specific competition on the reproduction success of Little Shearwater
Puffinus assimilis baroli in Selvagem Grande. In Threats to
Seabirds: Proceedings of the 5th International Seabird
Group conference, 24–26 March 1995: 35–36. Glasgow.
Oro, D., De Leó n, A., Minguez, E. & Furness, R.W. 2005.
Estimating predation on breeding European Storm-petrels
(Hydrobates pelagicus) by Yellow-legged Gulls (Larus
michahellis). J. Zool. Lond. 265: 421–429.
Podolsky, R., Ainley, D., Spencer, G., Deforest, L. &
Nur, N. 1998. Mortality of Newell’s Shearwaters caused
by collisions with urban structures on Kauai. Colon.
Waterbirds 21: 20–34.
Raine, H., Borg, J.J., Raine, A., Bariner, S. & Cardona, M.B.
2007. Light Pollution and Its Effect on Yelkouan Shearwaters
in Malta; Causes and Solutions. Malta: Life Project Yelkouan
Shearwater. BirdLife Malta. Available at: http://www.
lifeshearwaterproject.org.mt (accessed 19 August 2008).
Reed, J.R. 1987. Polarizing filters fail to reduce light attraction
in Newell Shearwaters. Wildl. Soc. Bull. 15: 596–598.
Reed, J.R., Sincock, J.L. & Hailman, J.P. 1985. Light attraction in endangered Procellariiform birds: reduction by
shielding upward radiation. Auk 102: 377–383.
ª 2009 The Authors
Journal compilation ª 2009 British Ornithologists’ Union
Rich, C. & Longcore, T. (eds) 2006. Ecological Consequences
of Artificial Night Lighting. Washington: Island Press.
Ristow, D. 1998. The prospectors in a colony of Cory’s Shearwater Calonectris diomedea. In Walmsley, J., Gounter, V.,
El Hili, A. & Sultana, J. (eds) Ecologie des oiseaux marins
et gestion intégrée du littoral en Méditerranée: 70–93.
Tunisia: Les amis des oiseaux et Medmaravis.
Rodrı́guez, B., de Leó n, L., Martı́n, A., Alonso, J. &
Nogales, M. 2003. Status and distribution of breeding seabirds in the northern islets of Lanzarote (Canary Islands).
Atl. Seabirds 5: 41–56.
Rodrı́guez, A., Rodrı́guez, B., Barone, B., Pérez, B. &
Herná ndez, A. 2008. Status and conservation requirements
of Manx Shearwaters Puffinus puffinus on Tenerife (Canary
Islands). Alauda 76: 72–74.
Ruiz, A. & Martı́, R. (eds) 2004. La Pardela Balear. Madrid:
SEO ⁄ BirdLife-Conselleria de Medi Ambient del Govern de
les Illes Balears.
Salamolard, M., Ghestemme, T., Couzi, F.-X., Minatchy, N.
& Le Corre, M. 2007. Impacts des é clairages urbains sur
les Pétrels de Barau, Pterodroma baraui sur I’lle de la
Réunion et mesures pour réduire ces impacts. Ostrich 78:
449–452.
Simons, T.R. 1984. A population model of the endangered
Hawaiian Dark-rumped Petrel. J. Wildl. Manage. 48: 1065–
1076.
Telfer, T.C., Sincock, J.L., Byrd, G.V. & Reed, J.R. 1987.
Attraction of Hawaiian seabirds to lights: conservation
efforts and effects of moon phase. Wildl. Soc. Bull. 15:
406–413.
Verheijen, F.J. 1981. Bird kills at tall lighted structures in the
USA in the period 1935–1973 and kills at a Dutch lighthouse in the period 1924–1928 show similar lunar periodicity. Ardea 69: 199–203.
Verheijen, F.J. 1985. Photopollution: artificial light optic spatial
control systems fail to cope with. Incidents, causations,
remedies. Exp. Biol. 44: 1–18.
Viceconsejerı́a de Medio Ambiente 1995. Rescatadas más
de 200 pardelas chicas desde 1992. Quercus 117: 48–49.
Warham, J. 1990. The Petrels: Their Ecology and Breeding
Systems. London: Academic Press.
Watanuki, Y. 1986. Moonlight avoidance behavior in Leach’s
Storm-petrels as a defense against Slaty-backed Gulls. Auk
103: 14–22.
Received 3 April 2008;
revision accepted 11 February 2009.