A long-term increase in eggshell thickness of Greenlandic
Peregrine Falcons Falco peregrinus tundrius
Knud Falka
, Sbren Mbllerb,T, William G. Mattoxc
a
Teglstrupvej 6a, DK-2100 Copenhagen, Denmark
b
Roskilde University Library, P.O. Box 258, DK-4000 Roskilde, Denmark
c
Conservation Research Foundation, 8300 Gantz Ave., Boise, ID 83709, USA
Received 11 October 2004; accepted 18 February 2005
Available online 10 May 2005
Abstract
Thickness of eggshell fragments and whole eggs from the Peregrine Falcon Falco peregrinus collected in South and West
Greenland between 1972 and 2003 was measured and compared to shell thickness of pre-DDT eggs, also collected in
Greenland. Linear regression yields a significant increase in the average thickness of eggshells over the period of 0.19% per
year, corresponding to a change in eggshell thinning from 13.9% in 1972 to 7.8% in 2003. Backwards extrapolation of the data,
suggests that the Greenlandic Peregrine population probably was never critically affected by DDT-induced eggshell thinning.
By sampling eggshell fragments in many nests the spatial and temporal sample distribution was enlarged, allowing the detection
of a significant long-term decrease in pollutant-induced eggshell thinning--a trend that could not have been identified if only
the rarer whole, addled eggs had been sampled.
D 2005 Elsevier B.V. All rights reserved.
Keywords: Peregrine Falcon; Falco peregrinus; Eggshell thickness; Greenland; Arctic
1. Introduction
The effects of persistent organic pollutants (POP)
on the eggshell thickness and breeding success in
high-trophic level birds have been widely documented.
Especially DDT and its derivatives have
been identified as a key group of POPs responsible
for the widespread reduction in breeding success and
subsequent population decline in the Peregrine
Falcon Falco peregrinus (Hickey, 1969; Hickey
and Anderson, 1968; Newton, 1979; Peakall et al.,
1975, 1976; Peakall and Kiff, 1979; Peakall and
Lincer, 1996; Ratcliffe, 1970, 1993; Walker II et al.,
1973).
The Greenland Peregrine Falcon (F.p. tundrius)
population has been the subject of long-term studies
in West Greenland since 1972 (e.g. Burnham and
0048-9697/$ - see front matter D 2005 Elsevier B.V. All rights reserved.
doi:10.1016/j.scitotenv.2005.02.024
T Corresponding author. Tel.: +45 46742493; fax: +45 46743090.
E-mail addresses: kf@vandrefalk.dk (K. Falk)8 moller@ruc.dk
(S. Mbller), wgmattox2@earthlink.net (W.G. Mattox).
Science of the Total Environment 355 (2006) 127­134
www.elsevier.com/locate/scitotenv
Mattox, 1984; Mattox and Seegar, 1988; Restani and
Mattox, 2000) and South Greenland since 1981 (Falk
et al., 1986; Falk and Mbller, 1988). The Greenland
population was conservatively estimated at 500­1000
pairs (Falk and Mller, 1988). Autumn migration
counts in the eastern US, where Peregrines from
Greenland and the eastern Canada mix during the
migration, suggest a slight increase in Arctic populations
(Titus and Fuller, 1990). It is thus to be
expected that a population of Arctic Peregrines
possibly recovering from the effects of POP-exposure
will show an increase in eggshell thickness over time.
Eggshell thickness analyses have often relied on
measurements of whole dead eggs, which, however, i)
are relatively rare occurrences, ii) are not present in
the most successful nests (all eggs hatch), and iii) may
potentially be biased since very thin eggs might break
(Odsjo¨, 1982). By using eggshell fragments from all
nests, a more representative and comprehensive
survey of the wild population can be obtained. Based
on the intra- and inter-clutch variation in shell
thickness, Falk and Mbller (1990) recommended that
studies of eggshell thinning should include fragments
from as many nests as possible.
Although studies of Arctic Peregrine Falcon
populations have shown that thinning in recent eggs
is less severe than older samples (Ambrose et al.,
2000; Johnstone et al., 1996), no gradual time trends
have been documented, probably due to relatively
short duration of most field studies. This paper
presents the results of a 32-year study of changes in
eggshell thickness in Greenlandic Peregrines, in order
to test the assumption that shell-thinning is continuously
improving following the general reduction in
use of the contaminants causing the effect.
2. Materials and methods
2.1. Study area and sampling
The study combines samples from two study areas
(Fig. 1): South Greenland (samples from 1981­2003),
and West Greenland (samples from 1972­1988).
The South Greenland study area covers the inner
parts of the three southernmost municipalities of
Southwest Greenland, Nanortalik, Qaqortoq and
Narsaq, approx. within 608­618 N and 458­468 W.
The area is low Arctic, with tundra vegetation, and
willow and birch shrub in the warm, sub-arctic areas
far from the cool outer coast. A substantial part of the
area is grazed by sheep. Field surveys of the Peregrine
Falcon population have been conducted here annually
between 1981 and 2003 by KF and SM.
The West Greenland study area covers the lowarctic
inland areas around Kangerlussuaq (Sbndre
Strbmfjord), approx. within 668 45V­678 15V N and
508­528 W. The area is hilly/mountainous and dotted
with lakes, and is divided by the large fjord Sbndre
Strbmfjord. The main vegetation is treeless tundra
(Burnham and Mattox, 1984). Studies were initiated
in 1972 by WGM and Richard Graham (see Burnham
and Mattox, 1984), and the present paper includes
eggshell fragments from 1972­1988.
In both survey areas, active nests were visited at
least once post-hatching, and, when conditions
allowed, the nest scrape carefully searched for eggshell
fragments deriving from the hatched eggs. In
addition, any whole dead eggs present were collected
for contaminant analyses; whole eggs are included in
this paper for comparison with fragments, while
contaminant studies of the egg contents from South
Greenland will be published elsewhere.
2.2. Measurements and analyses
The shell fragments were measured with a computer-connected
Mitutoyo Digital Micrometer (type
293-521-30) with a small stainless steel ball glued to
the rotating jaw in order to fit the inner curved surface
of the eggshell fragments.
Each fragment was scrutinised to determine
whether any membrane was still adhering to the inner
surface. Measurements were performed only on (parts
of) fragments without any membrane because on
fragments it is difficult to be certain if both
membranes (shell and egg membrane) are present.
Samples from a total of 93 clutches from South
Greenland were measured and provided from 3 to 91
membrane-free measurements. However, since eggshell
thickness varies within the egg there is a risk that
too few samples may bias results. Hence, we chose to
include only 79 clutches that provided 20 or more
measurable fragments--the same threshold selected
by Odsjo¨ (1982) in a study of Swedish Ospreys--and
assumed they represented the thickness of the entire
K. Falk et al. / Science of the Total Environment 355 (2006) 127­134128
clutch. Similarly, among 58 samples from West
Greenland, 40 clutches provided at least 20 measurable
fragments.
Whole, addled eggs were also collected and kept
frozen, and used for eggshell measurements: in South
Greenland 38 eggs from 28 clutches (1986­2003),
Fig. 1. Sampling sites in South and West Greenland; all nest sites sampled in this study are located inside the hatched areas. White inland areas
are ice/glaciers.
K. Falk et al. / Science of the Total Environment 355 (2006) 127­134 129
and in West Greenland 8 eggs from 8 clutches (1986­
88). When opened in the laboratory for contaminant
studies (Srensen et al., 2004), the eggs were cut
along the equator and the empty half shells washed
with water before being left to dry for 3 months at
room temperature. The half shells were measured
along the equator with a modified Mitutoyo Micrometer
(type 147-301)--the same device used to
measure 16 Greenlandic pre-DDT clutches (48 eggs)
in the collection at Zoological Museum, Copenhagen
(Falk and Mbller, 1990). There was no significant
difference in measurements taken with the two
different tools. The egg opening method left parts of
the shell with membranes intact and other parts
without membranes. Consequently, most measurements
include membranes, but some are without. To
compare measurements with and without membranes,
a membrane correction factor was independently
determined by measuring adjacent points with and
without membranes on 26 whole eggs (the cut
halves): 0.071 mm (n=83, SD=0.013). This compares
well to the 0.07 mm used by Nyga°rd (1983) and 0.069
mm reported by Court et al. (1990).
The sampling unit used in all analyses is dclutchT--
i.e., each nest and year provides one mean eggshell
thickness value.
3. Results
Mean shell thickness for fragments from South
Greenland and West Greenland did not differ when
comparing data from 1981­1989, the period during
which samples from both areas were available (t-test;
a
b
0.20
0.25
0.30
0.35
0.40
1970 1975 1980 1985 1990 1995 2000
Year
Thickness(mm),fragments
0.20
0.25
0.30
0.35
0.40
1970 1975 1980 1985 1990 1995 2000
Year
Thickness(mm),wholeeggs
Fig. 2. Eggshell thickness vs. sampling year of Peregrine Falcon eggs from South (circles) and West Greenland (triangles) based on: a) eggshell
fragments 1972­2003 (linear regression: y=0.000655 x -1.0018), and b) whole addled eggs 1986­2003 ( y=0.000525 x -0.74296, ns).
K. Falk et al. / Science of the Total Environment 355 (2006) 127­134130
df=20, 22; P=0.123). Hence, further analyses are
based on pooled data samples from the two survey
areas.
During the period 1972­2003 there was a weak but
highly significant increase in the average thickness of
eggshell fragments (Fig. 2a) ( P=0.00223, N=119).
The slope of the linear regression shows an average
increase of 0.19% per year. This would correspond to
a change in eggshell thinning from 13.9% in 1972 to
7.8% in 2003 when compared to pre-DDT eggs
collected in Greenland (48 eggs from 16 clutches,
1881­1930, mean 0.336 mm, SD=0.026; Falk and
Mbller 1990). When data from the two sampling areas
were treated separately, the increases were still
significant (Table 1). An analogous analysis of data
from whole egg samples revealed a similar, but
insignificant, increase, although the regression slopes
(SAS proc glm, separate slopes model) of the
fragment and whole egg samples did not differ
significantly ( P=0.8224).
When the fragment samples were grouped into two
equal time periods, the oldest samples were on
average significantly thinner than the more recent
(1972­1987: 0.294 mm; 1988­2003: 0.304 mm; ttest,
df=37, 66; P=0.0064).
Mean shell thickness for whole eggs and fragments
did not differ when comparing data from the period
(Fig. 2b) where both types of samples were collected
(1986­2003; t-test, df=80, 35; P=0.536), so we also
pooled data for the two sample types for a combined
analysis.
Several females with known identity provided
samples for more than one year. However, no trend
could be detected in eggshell thickness with age of the
laying bird.
4. Discussion
Although the embryo may extract minerals from
the eggshell, it has been shown that while shell
density (measured as shell index, Ratcliffe, 1967) is
affected, the shell thickness per se does not change
significantly during incubation (Bennett, 1995; Bunck
et al., 1985). Hence, stage of incubation is not
considered in this study.
Walker II et al. (1973) compared shell thickness of
material from 9 eggs (2 whole eggs and fragments
from 7 eggs) collected in West Greenland to 42 preDDT
eggs from Greenland and measured a 14%
thinning. Similarly, in a previous assessment based on
our samples from South Greenland 1981­1985
compared to the same collection of Greenlandic preDDT
eggs, we arrived at 14% eggshell thinning (Falk
and Mller, 1988). Based on the regression equation
from this study, the shell thinning in 1983 was 12%,
which is in reasonably good agreement with the
previous results.
In an early review Peakall and Kiff (1988) reported
post-DDT eggshell thinning in Peregrine populations
from 30 different study areas across the world ranging
from below 5% and up to 25%. Among these,
northern migrating populations had eggshell thinning
values between 13% and 23% in the 1960s and 70s
(Berger et al., 1970; Burnham and Mattox, 1984;
Cade et al., 1971; Nelson and Myres, 1975; Nyga°rd,
1983; Odsjo¨ and Lindberg, 1977; Peakall et al., 1975;
White and Cade, 1977). More recent studies of
eggshell thinning in Arctic migrating Peregrine
populations show values of 10.6% (Alaska, 1991­
95, Ambrose et al., 2000) and 15% (Arctic Canada,
1990­1994, Johnstone et al., 1996). No time trend
Table 1
Results of linear regression analysis of thickness data of eggshell fragments and whole Peregrine Falcon eggs sampled in South and West
Greenland; sample size, N, is number of clutches
Sample Years % increase
per yearT
N PTT % reduction
in 1950TTT
Fragments, pooled 1972­2003 0.19 119 0.0022 18.1
Fragments, South Greenland 1981­2003 0.21 79 0.0253 19.4
Fragments, West Greenland 1972­1989 0.44 40 0.0063 24.9
Whole eggs, pooled 1986­2003 0.16 34 0.4368 16.6
T Based on linear regression.
TT Significance level of the slope; statistically significant values in bold.
TTT Extrapolation, based on linear regression.
K. Falk et al. / Science of the Total Environment 355 (2006) 127­134 131
could be detected in any of these studies, probably
due to the relatively short time span covered.
We have found a highly significant long-term
gradual increase in the shell thickness over the 32
year survey period, supported by the grouped comparison
of equal time periods. If we assume the trend
to be linear, the regression line can be extrapolated
backwards (Table 1, Fig. 3) for a rough assessment of
when the shell-thinning would have exceeded the
critical empirical bthresholdQ of about 17% (Peakall
and Kiff, 1988). This rough assessment suggests that
the shell-thinning might have exceeded the critical
limit until in the early 1950s only. That was probably
too short after DDT became widespread (introduced
1947) for the pesticide to have had a marked effect on
the Greenlandic Peregrine Falcon population. This is
supported by evidence of a strong population since the
1970s (Burnham and Mattox, 1984), despite the fact
that the Arctic subspecies in Greenland migrates
through and/or to areas in Latin America where
phasing out of the pesticides has been slower than
in North America due to the need to fight Malaria
(Raloff, 2000). With the slow rate of increase, it will
take several decades before the eggshell thickness in
Greenlandic Peregrines can be expected to reach preDDT
levels.
Compared to other populations, the Greenlandic
Peregrines are in a unique position, because they
spend the summer on a relatively isolated island
colonised by few long-distance migrants, which could
carry pollutants with them. The majority of the
Peregrines' prey in Greenland consists of resident
(Rock Ptarmigan Lagopus mutus) or short-distance
migrant passerines wintering in the border area of
Canada and USA. Only the Northern Wheatear
(Oenanthe oenanthe) winters in potential DDT-usage
areas of West Africa (Lyngs, 2003). Pollutant levels of
passerine prey samples in Greenland were low already
in the 1970s (Burnham and Mattox, 1984), so the
falcons have a clean summer diet.
To our knowledge, this is the first time a
significant, gradual increase in eggshell thickness
over time has been detected in a Peregrine Falcon
population. No improvement in shell thickness
between decades (1982­86 vs. 1991­1994) was found
in F.p. tundrius from Arctic Canada (Johnstone et al.,
1996). However, Ambrose et al. (2000) reported that
egg thickness increased slightly, though not signifiFig.
3. Shell thickness (mean per clutch and year) of Peregrine Falcon eggs from Greenland; pre-DDT Peregrine eggs (1881­1930) from
museum collections, and samples from 1972­2003 from South and West Greenland; filled circles represent whole eggs, and open circles
indicate eggshell fragments. The reference lines indicate pre-DDT mean thickness (0.336 mm), and the approximate empiric b17% thresholdQ
(0.279 mm) for population declines (Peakall and Kiff, 1988). The combined regression line of the recent change in thickness is extrapolated
backwards.
K. Falk et al. / Science of the Total Environment 355 (2006) 127­134132
cantly, over time in Peregrine eggs collected in Alaska
1979-1995.
As a supplement to monitoring the long-term POP
loads and effects in Peregrines based on collection of
whole addled eggs only (e.g. USFWS 2003) it is
worth also to collect and measure the fragments from
hatched eggs. In our study it would not have been
possible to detect any long-term change in eggshell
thickness based on measurements of the whole addled
eggs alone; they occur too rarely to provide an
adequate sample. By using fragments from hatched
eggs--which are easier to obtain and analyse, and
probably more representative since they include all
breeding attempts--it was possible to expand the
temporal and spatial sampling. The larger sampling
coverage allowed us to verify that the Greenlandic
Peregrine Falcon population is slowly responding to
the expected gradually reduced exposure to shellthinning
POPs such as DDT and metabolites.
Acknowledgements
The field samples have been collected during more
than three decades, with financial support from public
funds: Danish Natural Science Research Council
(SNF), Commission for Scientific Research in Greenland
(KVUG), US Army Edgewood Research Development
and Engineering Center, The Peregrine Fund
Inc., and a wide range of private funds in the USA
and Denmark. In both areas, fieldwork was carried
out by tens of dedicated birders on a voluntary basis.
Especially, we thank Kaj Nielsen, Qaqortoq, for
providing our boat and various logistic support in
South Greenland every year. Managers and staff at
Fjeldstationen, Narsarsuaq, allowed us to use the
station as an informal base. At Kangerlussuaq, PICO
staff and KISS personnel assisted our teams, as did
numerous small plane pilots and boat owners. The
Danish Meteorological Institute allowed us to use
their facility outside Kangerlussuaq for many years,
The Danish Defence Command, NY Air National
Guard (109th TAG), and US Air Force provided
aerial support to/from Greenland. In South Greenland
the Ice Patrol assisted with ad hoc helicopter flights
for several years. Greenland Home Rule and Danish
Polar Center assisted in providing field research
permissions, and Danish Environmental Protection
Agency, Ministry of Environment, supported the
analyses.
References
Ambrose RE, Matz A, Swem T, Bente P. Environmental contaminants
in American and Arctic Peregrine Falcon eggs in
Alaska, 1979­1995. Ecological Services Fairbanks, AK, U.S.
Fish and Wildlife Service, Technical Report NAES-TR-00-02;
2000. 67pp.
Bennett RS. Relative sensitivity of several measures of eggshell
quality to the stage of embryonic development. Bull Environ
Contam Toxicol 1995;54:428­31.
Berger DD, Anderson DW, Weaver JD, Risebrough RW. Shell
thinning in eggs of Ungava Peregrines. Can Field-Nat 1970;
84:265­7.
Bunck CM, Spann JW, Pattee OH, Fleming WJ. Changes in
eggshell thickness during incubation: implications for evaluating
the impact of organochlorine contaminants on productivity.
Bull Environ Contam Toxicol 1985;35:173­82.
Burnham WA, Mattox WG. Biology of the Peregrine and Gyrfalcon
in Greenland. Meddr Grbnland 1984;14:1­25.
Cade TJ, Lincer JL, White CM, Roseneau DG, Swartz LG. DDE
residues and eggshell changes in Alaskan falcons and hawks.
Science 1971;172:955­7.
Court GS, Gates CC, Boag DA, MacNeil JD, Bradley DM, Fesser
AC, et al. A toxicological assessment of Peregrine Falcons,
Falco peregrinus tundrius, breeding in the Keewatin, District of
the Northwest Territories, Canada. Can Field-Nat 1990;
104:255­72.
Falk K, Mbller S. Clutch size effects on eggshell thickness in the
Peregrine Falcon and European Kestrel. Ornis Scand
1990;21:265­79.
Falk K, Mbller S. Status of the Peregrine Falcon in South
Greenland: population density and reproduction. In: Cade TJ,
Enderson JH, Thelander CG, White CM, editors. Peregrine
Falcon populations Their management and recovery. Boise7 The
Peregrine Fund, Inc; 1988. p. 37­43.
Falk K, Mbller S, Burnham WA. The Peregrine Falcon Falco
peregrinus in South Greenland: nesting requirements, phenology
and prey selection. Dansk Orn Foren Tidsskr 1986;80:113­20.
Hickey JJ. Peregrine Falcon populations, their biology and decline.
Madison7 Univ. Wisconsin Press; 1969.
Hickey JJ, Anderson DW. Chlorinated hydrocarbons and eggshell
changes in raptorial and fish-eating birds. Science 1968;
162:271­3.
Johnstone RM, Court GC, Fesser AC, Bradley DM, Oliphant LW,
MacNeil JD. Long-term trends and sources of organochlorine
contamination in Canadian tundra Peregrine Falcons, Falco
peregrinus tundrius. Environ Pollut 1996;93:109­20.
Lyngs P. Migration and winter ranges of birds in Greenland. Dansk
Orn Foren Tidsskr 2003;97:1­167.
Mattox WG, Seegar WS. The Greenland Peregrine Falcon survey,
1972­1985, with emphasis on recent population status. In: Cade
JH, Enderson JH, Thelander CG, White CM, editors. Peregrine
K. Falk et al. / Science of the Total Environment 355 (2006) 127­134 133
Falcon populations. Their management and recovery. Boise7
The Peregrine Fund, Inc; 1988. p. 27­36.
Nelson RW, Myres MT. Changes in the Peregrine population and its
seabird prey at Langara Island, British Columbia. Raptor Res
Rep 1975;3:13­31.
Newton I. Population ecology of Raptors. Berkhamstead7 T. and A.
D. Poyser; 1979.
Nyg3rd T. Pesticide residues and shell thinning in eggs of
Peregrines in Norway. Ornis Scand 1983;14:161­6.
Odsjf T. Eggshell thickness and levels of DDT, PCB and mercury in
eggs of Osprey (Pandion haliaetus L.) and Marsh Harrier
(Circus aeruginosus L.) in relation to their breeding success and
population status in Sweden. Stockholm7 University of Stockholm;
1982.
Odsjf T, Lindberg P. Reduction of eggshell thickness of Peregrines
in Sweden. In: Lindberg P, editor. Pilgrimsfalk, Report from a
Peregrine Conference held at Grimsf Wildlife Research Station,
Sweden 1­2 April 1977. Stockholm7 Swedish Society for
Nature Conservation; 1977. p. 61­4.
Peakall DB, Kiff LF. Eggshell thinning and DDE residue levels
among peregrine falcons: a global perspective. Ibis
1979;121:200­4.
Peakall DB, Kiff LF. DDE contamination in Peregrines and
American Kestrels and its effect on reproduction. In: Cade TJ,
Enderson JH, Thelander CG, White CM, editors. Peregrine
Falcon populations. Their management and recovery. Boise7
The Peregrine Fund Inc; 1988. p. 337­50.
Peakall DB, Lincer JL. Do PCBs cause eggshell thinning? Environ
Pollut 1996;91:127­9.
Peakall DB, Cade TJ, White CM, Haugh JR. Organochlorine
residues in Alaskan Peregrines. Pesticide Mon Journ
1975;8:255­60.
Peakall DB, Reynolds LM, French MC. DDE in eggs of the
Peregrine Falcon. Bird Study 1976;23:183­6.
Raloff J. The case for DDT: what do you do when a dreaded
environmental pollutant saves lives? Sci News 2000;158:12­4.
Ratcliffe DA. Decrease in eggshell weight in certain birds of prey.
Nature 1967;215:208­10.
Ratcliffe DA. Changes attributable to pesticides in egg breakage
frequency and eggshell thickness in some british birds. J Appl
Ecol 1970;7:67­115.
Ratcliffe D. The Peregrine Falcon. London7 T. and A.D. Poyser;
1993.
Restani M, Mattox WG. Natal dispersal of Peregrine Falcons in
Greenland. Auk 2000;117:500­4.
Sbrensen PB, Vorkamp K, Thomsen M, Falk K, Mbller, S.
Persistent organic pollutants (POPs) in the Greenland
environment-long-term temporal changes and effects on eggs
of a bird of prey. National Environmental Research Institute,
Denmark. NERI Technical report no 509; 2004. 126pp.
Titus K, Fuller MR. Recent trends in counts of migrant hawks from
northeastern North America. J Wildl Manage 1990;54:463­70.
USFWS. Monitoring plan for the American Peregrine Falcon,
a species recovered under the Endangered Species Act. U.S.
Fish and Wildlife Service, Division of Endangered Species and
Migratory Birds and State Programs, Pacific Region, Portland,
OR; 2003.
Walker II W, Mattox WG, Risebrough RW. Pollutant and shell
thickness determinations of Peregrine eggs from West Greenland.
Arctic 1973;26:256­8.
White CM, Cade TJ. Long term trends of Peregrine populations in
Alaska. In: Chancellor RD, editor. Proceedings of the ICBP
World Conference on Birds of Prey. London7 Intern. Council for
Bird Preservation; 1977. p. 63­71.
K. Falk et al. / Science of the Total Environment 355 (2006) 127­134134