ArticlePDF Available

Dietary response of Barn Owls (Tyto alba) to large variations in populations of common voles (Microtus arvalis) and European water voles (Arvicola terrestris)

Authors:

Abstract and Figures

The diet of the Barn Owl (Tyto alba (Scopoli, 1769)) was studied over an 8-year period in the Jura mountains of France, during two population surges of its main rodent prey (common voles, Microtus arvalis (Pallas, 1778), and European water voles, Arvicola terrestris (L.,1758)), allowing us to test whether T. alba is an opportunistic predator as is often cited in the literature or exhibits more complex patterns of prey selection as is reported in arid environments. Small mammals were sampled by trapping and index methods. We observed (i) significant correlations between the proportions of A. terrestris, M. arvalis, and woodland rodents in the diet and their respective densities in the field; (ii) interactions between populations of A. terrestris and M. arvalis, indicating that the proportion of each species in diet was affected by the density of the other; (iii) proportions of red-toothed shrews (genus Sorex (L., 1758)) in the diet did not correlate with their abundance in the field, indicating that those species were likely to be preyed upon when others were no longer available. This confirms that T. alba is generally opportunistic; however, prey selection of a focal species (e.g., Sorex spp., grassland species) can be affected by the density or availability of the other prey species.
Content may be subject to copyright.
Dietary response of Barn Owls (Tyto alba) to large
variations in populations of common voles
(Microtus arvalis) and European wate r voles
(Arvicola terrestris)
Nadine Bernard, Dominique Michelat, Francis Raoul, Jean-Pierre Que
´
re
´
,
Pierre Delattre, and Patrick Giraudoux
Abstract: The diet of the Barn Owl (Tyto alba (Scopoli, 1769)) was studied over an 8-year period in the Jura mountains
of France, during two population surges of its main rodent prey (common voles, Microtus arvalis (Pallas, 1778), and Euro-
pean water voles, Arvicola terrestris (L.,1758)), allowing us to test whether T. alba is an opportunistic predator as is often
cited in the literature or exhibits more complex patterns of prey selection as is reported in arid environments. Small mam-
mals were sampled by trapping and index methods. We observed (i) significant correlations between the proportions of A.
terrestris, M. arvalis, and woodland rodents in the diet and their respective densities in the field; (ii) interactions between
populations of A. terrestris and M. arvalis, indicating that the proportion of each species in diet was affected by the den-
sity of the other; (iii) proportions of red-toothed shrews (genus Sorex (L., 1758)) in the diet did not correlate with their
abundance in the field, indicating that those species were likely to be preyed upon when others were no longer available.
This confirms that T. alba is generally opportunistic; however, prey selection of a focal species (e.g., Sorex spp., grassland
species) can be affected by the density or availability of the other prey species.
Re
´
sume
´
: Nous avons e
´
tudie
´
le re
´
gime alimentaire de l’effraie des clochers (Tyto alba (Scopoli, 1769)), sur une pe
´
riode
de 8 anne
´
es dans les montagnes du Jura en France, durant deux pics de
´
mographiques d’abondance des principaux rongeurs
qui leur servent de proies (les campagnols des champs, Microtus arvalis (Pallas, 1778), et les campagnols terrestres, Arvi-
cola terrestris (L., 1758)). Nous avons ainsi pu tester si T. alba est un pre
´
dateur opportuniste comme l’indique souvent la
litte
´
rature ou si elle montre des patrons plus complexes de se
´
lection des proies comme il a e
´
te
´
observe
´
dans les milieux
arides. Les petits mammife
`
res ont e
´
te
´
e
´
chantillonne
´
s par trappage et par me
´
thodes indiciaires. Nous observons (i) des cor-
re
´
lations significatives entre les proportions d’A. terrestris,deM. arvalis
et des rongeurs forestiers dans le re
´
gime alimen-
taire et leurs densite
´
s respectives en nature, (ii) des interactions entre les populations d’A. terrestris et de M. arvalis,ce
qui indique que la proportion de chacune des espe
`
ces dans le re
´
gime est affecte
´
e par la densite
´
de l’autre espe
`
ce et
(iii) des proportions de musaraignes a
`
dents rouges (genre Sorex (L., 1758)) dans le re
´
gime qui ne correspondent pas a
`
leur abondance en nature, ce qui laisse croire que ces espe
`
ces sont utilise
´
es lorsque les autres ne sont plus disponibles.
Cela confirme que T. alba est ge
´
ne
´
ralement une opportuniste, mais que la se
´
lection des proies d’une espe
`
ce cible
´
e particu-
lie
`
re (par ex. les Sorex spp., des espe
`
ces de prairies) peut e
ˆ
tre affecte
´
e par la densite
´
ou disponibilite
´
des autres espe
`
ces de
proies.
[Traduit par la Re
´
daction]
Introduction
The Barn Owl (Tyto alba (Scopoli, 1769)) is a model
largely used either to estimate the composition of a small-
mammal community at various temporal and spatial scales
(Love et al. 2000; Le Louarn and Que
´
re
´
2003), or to study
its response to variations in prey population. For instance,
Taylor (1994) reviewed patterns of variations in abundance
of common voles (Microtus arvalis (Pallas, 1778)) and field
voles (Microtus agrestis (L., 1761)) and their effects on
life-history traits of Barn Owl. Higher prey density posi-
tively affects molt initiation, hatching, clutch size, fledgling
success, and recruitment, but negatively affects dispersal
distance. This corroborates a numerical response of the
Barn Owl to variations in prey resource, which is now
largely accepted.
In addition to a number of studies dealing with breeding
ecology and population dynamics of owls in response to
Received 26 July 2009. Accepted 4 February 2010. Published on the NRC Research Press Web site at cjz.nrc.ca on 26 March 2010.
N. Bernard,
1
D. Michelat, F. Raoul, and P. Giraudoux. Unite
´
Mixte de Recherche Centre National de la Recherche Scientifique 6249
Chrono Environnement, Universite
´
de Franche-Comte
´
, Place Leclerc, 25030 Besanc¸on CEDEX, France.
J.-P. Que
´
re
´
and P. Delattre. Unite
´
Mixte de Recherche 1062, Centre de Biologie et Gestion des Populations, Institut National de la
Recherche Agronomique, CS 30016, 34988 Montferrier/Lez CEDEX, France.
1
Corresponding author (e-mail: nadine.bernard@univ-fcomte.fr).
416
Can. J. Zool. 88: 416–426 (2010) doi:10.1139/Z10-011 Published by NRC Research Press
prey availability (Boreal Owl (Aegolius funereus (L., 1758)):
Za
´
rybnicka
´
et al. 2009; Tawny Owl (Strix aluco L., 1758):
Kekkonen et al. 2008; Karell et al. 2009a, 2009b), vast litera-
ture is available on food habits of various species of owls
worldwide, e.g., T. alba (Taylor 1994; Bellocq 2000; Rasoma
and Goodman 2007), Northern Spotted Owls (Strix occidenta-
lis caurina (Merriam, 1898)) (Hamer et al. 2001; Forsman et
al. 2004), or Barred Owls (Strix varia Barton, 1799) (Livezey
2007). These studies report variation in owl diet according
to season, landscape, location, or biogeographical context.
Surprisingly few surveys have been carried out that have exam-
ined changes in owl diet as it relates to local variation in prey
availability in the field. For instance, studies on S. aluco have
shown its opportunistic behaviour towards bats in urban
areas (Lesin
´
ski et al. 2008, 2009). In Scotland, S. aluco
responded functionally to the 3- to 4-year cycles of abun-
dance of M. agrestis, and over the long term, fluctuations
in rodent prey reflected changes in rodent guilds (Petty
1999). To our knowledge no such studies are available for
the well-studied S. o. caurina, the California Spotted Owls
(Strix occidentalis occidentalis (Xantus de Vesey, 1860)), or
S. varia.
Tyto alba is generally assumed to be an opportunistic
predator of small mammals, taking its prey proportionally
to their availability in the field (Vernon 1972; Cramp
1985). However, both theory (Yom-Tov and Wool 1997)
and the lack of empirical data have made this issue contro-
versial, especially when conclusions are drawn from studies
where owl diet and behaviour are not monitored in the same
time and place as the prey communities in the field (Glue
1967; Webster 1973; Goszczynski 1981; De Bruijn 1994).
Dietary response of T. alba to variations in prey has been
investigated only in arid environments. Gubanyi et al.
(1992) in western Nebraska and Jorgensen et al. (1998) in
the foothills of the Chihuahuan desert analyzed the diet of
T. alba and their nest success in relation to proportions of
prey in the diet and prey density estimated by trapping.
They found prey preference governed by a number of
factors such as habitat-patch quality and individual foraging
ability. Lima et al. (2002) in semiarid Chile monitored small
mammals through mark–recapture methods to obtain a time
series of species abundance. They conducted parallel time-
series analysis of regurgitated pellets from two resident
owls (T. alba and the Lesser Horned Owls (Bubo magellani-
cus (Lesson, 1828))), enabling them to combine time series
and mathematical models to estimate food-web structure and
climate effects on the dynamics of small mammals and
owls. In a Sahelian ecosystem in Mali, T. alba tended to
prey on smaller-than-average individuals (Granjon and
Traore
´
2007). In arid Australia, T. alba selected larger prey
during a rodent outbreak (Pavey et al. 2008).
To our knowledge, no studies of this type on T. alba have
been reported yet from temperate areas. However, in tem-
perate regions, various patterns of population dynamics of
small mammals can be found, from stable densities with
only seasonal variations to multiannual fluctuations, deter-
mined by a number of factors that are still debated (preda-
tion, diseases, density-dependence, competition, resource
limitation, etc.) (Lidicker 2000). In mid-altitude mountain-
ous regions of eastern France, grassland small-mammal spe-
cies (e.g., M. arvalis and European water voles (Arvicola
terrestris (L., 1758))) have been shown to respond to land-
scape composition. Empirical data support the idea that the
percentage of optimal habitats in a landscape influences the
probability that those populations would undergo multi-
annual cycles, as a combined effect of dispersal and preda-
tion (Delattre et al. 1992; Giraudoux et al. 1997; Lidicker
2000). In regions with large proportions of permanent grass-
land, during the high-density peaks of grassland rodent pop-
ulations (hundred to thousands of animals per hectares), the
proportion of grassland small mammals increases in the diet
of predators and constitutes up to 90% of the food items
identified, e.g., in fox faeces (Giraudoux et al. 2002; Dupuy
et al. 2009). This pattern of population fluctuation has indi-
rect consequences that are still not fully understood on the
whole community of small-mammal species, including those
living in hedges, forests, and marshes. For instance, popula-
tion declines of grassland small mammals and those of other
species (bank voles (Myodes glareolus (Schreber, 1780)),
Old World mice (genus Apodemus
Kaup, 1829), red-toothed
shrews (genus Sorex (L.,1758)), etc.) have been shown to be
synchronous (Michelat and Giraudoux 2006).
Such systems offer a unique opportunity for studying how
the diet of T. alba responds to variation in prey density, as
long as variations in the prey community can be monitored
in parallel. Because a variety of prey species is present at
various densities over time, it is possible to observe prey
switches of predators when the relative proportion of their
prey changes.
Here we monitored variations in prey population and the
dietary response of T. alba over 8 years in a landscape where
two dominant grassland species (M. arvalis and A. terrestris)
experienced population cycles. The aim of the study was to
assess whether T. alba is an opportunistic predator as
reported in the literature or exhibit more complex patterns
of prey selection as reported in arid environments. We spe-
cifically correlated the relative population density of prey es-
timated by trapping to prey occurrence in the diet of T. alba
using logistic regression to characterize the dietary response
of T. alba in a temperate system where the small-mammal
community is regularly dominated by two grassland species.
Materials and methods
Study area
The study site was located in the canton of Levier, France
(Doubs department; 6812E, 46858N), at a mean altitude of
800 m. Housing is mainly clustered in villages and virtually
all farmland is grassland. Dietary response of T. alba was
monitored in three church towers situated as close as possible
(2.5, 2, and 1.5 km) to the small-mammal sampling area
(Fig. 1). The towers were regularly occupied by T. alba
from April 1987 to 1995 as resting, roosting, nesting, and
breeding places. Small mammals were monitored in habitat
types generally foraged by T. alba (open grassland, hedges),
thus excluding forests (Michelat and Giraudoux 1991, 1993).
The relationship between small-mammal species and habitats
in the study site has been described in Giraudoux et al.
(1994) and Raoul et al. (2001). Eurasian shrews (Sorex ara-
neus (L., 1758)) and Millet’s shrews (Sorex coronatus Millet,
1828) cannot be reliably identified without karyotype, which
Bernard et al. 417
Published by NRC Research Press
could not been done here. Thus, they were reported as
Sorex spp. (S. araneus and S. coronatus complex).
Small-mammal sampling
Sampling was carried out from 1987 to 1995 in April,
June (only in 1987), July (from 1987 to 1993), and October.
Sampling sessions included livetrapping and standard index-
ing according to species (see below) and were completed
within 1 week. Sampling and handling protocols followed
the recommendations of the Panel on Euthanasia of the
American Veterinary Medical Association and were ap-
proved by the Groupe Naturaliste de Franche-Comte
´
and
the Institut National de Recherche Agronomique (INRA).
Traplines of 100 m long consisted of 34 traps placed
every 3 m (Spitz 1963, 1964). Sampling was performed in
the hedgerow network and in grassland. On average, 9 trap-
lines were set in hedges and forest edges and 23 traplines in
grasslands at each session (see Appendix A, Table A1). Tra-
pline locations were changed from one season to another to
avoid overtrapping. Distribution map of traplines and more
details can be found in Giraudoux et al. (1994). The traps
were set for 3 days; small mammals were collected every
day and humanely killed for further study on population
structure and parasites. Small species (<50 g) were sampled
with INRA live traps. Median mass of the different species
in the study area is given in Table 1. Variations in small-
mammal abundance were calculated for each species and by
sampling period (t; season). Each species exhibited strong
habitat specificity (e.g., M. glareolus was trapped only in
hedges and forest edges) except Sorex spp., which was
trapped in grasslands, hedges, and forest edges. Therefore,
for each species an instant abundance index (i
t
) was calcu-
lated as the mean number of individuals captured per tra-
pline in the species habitat at a time t, except for Sorex spp.
which had i
t
calculated for both habitat types. When trap-
ping was impossible (e.g., winter, during the seasonal de-
cline of populations), i
t
was interpolated as the mean
between the preceding and the following seasons. Thus, the
winter abundance index was the mean between October and
April indices. Spitz et al. (1974) found that population den-
sities could be estimated from such an index when species
movement capability and home-range size were taken into
account. Based on this, they proposed conversion coeffi-
cients for M. arvalis, M. glareolus, and Apodemus spp. that
provided a gross estimate of absolute densities (d
t
; no. of
individuals/ha). This method permits cautious comparison
between species when density differences are very large.
Yellow-necked mice (Apodemus flavicollis (Melchior,
1834)) and M. glareolus were not abundant in the diet of
T. alba and so were pooled as woodland rodents’ when an-
alyzing variations in the population of the prey resource in
the field.
Arvicola terrestris cannot be trapped using INRA live
traps because its median mass is 62 g. Therefore, the varia-
tion in abundance of this species was assessed as follows.
The Fe
´
de
´
ration Re
´
gionale de De
´
fense contre les Organismes
Nuisibles (FREDON; a farmers’ crop protection organiza-
tion) surveys the entire district area each year and monitors
the population dynamics of A. terrestris to control its out-
breaks. Each district was scored on a semiquantitative scale
using the following index (i
t
) adapted from Giraudoux et al.
(1995): 0, no colony observed; 1, some isolated colonies; 2,
colonies present in many pastures and meadows; 3, numer-
ous colonies and serious damage to grassland; 4, most grass-
land area covered by the outbreak and heavy damage; and 5,
total grassland area covered with vole tumuli, no hay pro-
duction. Category 5 corresponds with the highest density
peak (>400 individuals/ha), whereas category 0 corresponds
with the lowest measurable density (<100 individuals/ha).
The i
t
and d
t
values are estimates of abundances and den-
sities, respectively, of small mammals at a given time (the time
when small mammals were trapped). For this reason they were
termed the instant index and instant density, respectively. How-
ever, the pellets collected were produced by owls during the 3-
month period between two pellet collections. As prey densities
changed over time, the quantity of prey available from time t
1
to time t
2
was
Z
t
2
t
1
f ðtÞdt
With a linear approximation of the increase or decrease of
prey populations for the 3-month period of pellet accumula-
tion, this corresponds to the following indices: i = 1.5
(i
t1
+ i
t2
) and d = 1.5 (d
t1
+ d
t2
). These indices were used
as indices of population abundance and density over a 3-
month period (see below).
Fig. 1. Location of the study area in eastern France.
418 Can. J. Zool. Vol. 88, 2010
Published by NRC Research Press
Diet of T. alba
Pellets were collected from church towers in three villages
(Levier, Chapelle d’Huin, and Le Souillot) in January, April,
July, and October. The towers were cleaned after each pellet
collection. These pellets came from an unknown number of
T. alba visiting the church tower during the study period. In
general, for each church tower, there was one or two birds in
winter and a pair and their offspring during the breeding sea-
son.
Vertebrate prey were identified using teeth and skulls,
whereas the other prey were identified using any recogniz-
able remains (Le Louarn and Que
´
re
´
2003). To reduce error
in proportions estimated from small samples, pellet collec-
tions containing fewer than 17 prey were not kept for further
analysis. Transposing prey frequency into biomass can be
biased, in part because of variations in digestibility and
body mass within and between prey categories (Goszczynski
1976). Thus, this kind of conversion was not used here. The
number of specimens of M. agrestis, greater white-toothed
shrews (Crocidura russula (Hermann, 1780)), and bicolored
shrews (Crocidura leucodon (Hermann, 1780)) collected in
pellets was very low. Thus, these species were not included
in the analysis.
Statistics
Pearson’s correlation coefficient (r) was used to assess
synchronies between the population densities of small mam-
mals in the field. Because the population dynamics of
Sorex spp. were highly synchronous among grasslands,
hedges, and forest boundaries, data were pooled for analysis
of population synchrony and dietary response of T. alba. For
a given species and time, a probability of prey occurrence
was computed as the number of prey items of the species
over the total number of prey items. This probability, ob-
tained from count data of various sample sizes, has a bino-
mial distribution. To take account of differences in sample
size and because a probability varies between 0 and 1, the
probability was modeled against the relative abundance of
prey in the fields using a general linear model with logit
link, also called a logistic regression. An analysis of devi-
ance was performed on the model to test the null hypothesis
of coefficients being equal to zero. Residuals were plotted
against time and examined to detect any temporal correla-
tion following Venables and Ripley (2003). Significant auto-
correlation patterns would have indicated that the addition of
a correlation structure was necessary. Examining the corre-
sponding autocorrelation function failed to detect significant
correlation for any time lag at a 0.05 risk. Thus, an addi-
tional structure was not required. The Kolmogorov–Smirnov
(KS) test of goodness of fit was then used to assess the nor-
mality of residuals. Statistics were computed with R version
2.9.0 (R Development Core Team 2005).
Results
Population dynamics of small mammals
The small-mammal population exhibited higher densities
in autumn and a decrease in winter, as well as multiannual
fluctuations (Figs. 2a–2d). Population dynamics were signif-
icantly synchronous for M. arvalis and Sorex spp. (r = 0.58,
p = 0.002), M. arvalis and woodland rodents (r = 0.36, p =
0.05), and A. terrestris and woodland rodents (r = 0.47, p =
0.02). The biomasses of M. arvalis and A. terrestris were es-
timated to vary from approximately zero kilogram per hec-
tare to several kilograms per hectare, and at maximum
densities, reached biomasses that were more than 100 times
greater than those of woodland rodents (A. flavicollis and
Table 1. Diet of Barn Owls (Tyto alba) at pellet collection sites during the study
period from 1987 to 1995.
Ma* At
{
AfMg
{
So
§
Other Total
Levier
Number of items 2935 1809 364 1882 1385 8375
Ratio in the diet 0.35 0.22 0.04 0.22 0.17 1
Minimum ratio 0.05 0.02 0 0.03 0.03
Maximum ratio 0.69 0.63 0.30 0.76 0.53
Chapelle d’Huin
Number of items 1444 1047 493 1855 615 5454
Ratio in the diet 0.26 0.19 0.09 0.34 0.11 1
Minimum ratio 0.03 0.02 0 0.13 0.02
Maximum ratio 0.57 0.63 0.38 0.62 0.33
Souillot
Number of items 1478 634 392 1187 430 4121
Ratio in the diet 0.36 0.15 0.10 0.29 0.10 1
Minimum ratio 0.08 0.01 0 0 0
Maximum ratio 0.56 0.75 0.33 0.55 0.35 .
*Common voles (Microtus arvalis); median mass = 16 g.
{
European water voles (Arvicola terrestris); median mass = 62 g.
{
Yellow-necked mice (Apodemus flavicollis) and bank vole (Myodes glareolus); median
mass = 20 and 18 g, respectively.
§
Eurasian shrews (Sorex araneus) and Millet’s shrews (Sorex coronatus) combined under
Sorex spp.; median mass = 8 g.
||
Other prey; median mass £ 18 g.
Bernard et al. 419
Published by NRC Research Press
M. glareolus). In terms of biomass, the community was
largely dominated by grassland species.
Dietary response of T. alba
A total of 17 950 prey items was identified. Each pellet
collection included between 17 and 1500 prey items, with a
Fig. 3. Prey frequency in the diet of Barn Owls (Tyto alba). Fre-
quencies are expressed in percentages of prey items. The numbers
and characters in the figure indicate year and season (e.g., Spr 1987
refers to Spring 1987).
Fig. 2. Density variations of small mammals based on several in-
dices. The numbers along the time axis indicate months and years
(e.g., 04-1987 refers to April 1987).
420 Can. J. Zool. Vol. 88, 2010
Published by NRC Research Press
mean of 251 prey items. Figure 3 shows variations in diets
of T. alba by site and by season. The category others’ in-
cludes 17 taxa, most of them seldom recorded, such as
European moles (Talpa europaea L., 1758), shrews, bats,
birds, amphibians, and insects. The frequency of each prey
item in the diet was highly variable among sites (Table 1).
The frequency of a given prey item in the diet was mod-
eled against the relative density of the corresponding species
in the field using logistic regression. Table 2 shows that all
models were significant and that coefficients were positive
(e.g., frequency in the diet increases with population den-
sities in the field), except for Sorex spp. The frequency of
Sorex spp. correlated negatively with its density in Levier
and no correlation was observed in Chapelle d’Huin and Le
Souillot. Furthermore the frequency of Sorex spp. in the diet
correlated negatively with the abundance of M. arvalis and
A. terrestris in the field at Chapelle d’Huin and Levier (p <
0.001).
Arvicola terrestris and M. arvalis were the two dominant
species (in terms of biomass) in the field and obviously in
the diet. Table 3 and Fig. 4 show the interaction between
A. terrestris and M. arvalis in the field and the frequency of
each species in the diet. In most cases, the population density
increase of M. arvalis in the field led to the increase of both
M. arvalis and A. terrestris in the diet, irrespective of popu-
lation densities of A. terrestris. Moreover, the increase in
population density of A. terrestris in the field mostly led to
an increase in frequency of M. arvalis in the diet.
Discussion
Sampling methods
The median of 38 home ranges of T. alba reported by
Taylor (1994) was 3.6 km
2
(2.5% and 97.5% quantiles were
0.68 and 19.3 km
2
, respectively). The three church towers
where pellets were collected were less than 3 km from the
sampling site of small mammals. This site was assumed to
be close to, if not within, the actual home range of T. alba.
We cannot prove that T. alba under study foraged in the
area where small mammals were sampled. However, we can
reasonably assume that the population dynamics of small
mammals in this area were representative of the populations
of small mammals in the home ranges of T. alba. Indeed,
demographic changes in rodents occur at a regional scale
(Hansson and Henttonen 1985), and this was shown for
M. arvalis (Delattre et al. 1992) and A. terrestris (Giraudoux
et al. 1997) in our study area.
Capture–recapture methods for small mammals are ex-
tremely time-consuming and for this reason cannot be applied
at the scale of a population of T. alba (Lima and Jaksic 1999;
Huitu et al. 2004). Although alternative index methods such
as landscape transects based on mammal indices (Hansson
1979; Que
´
re
´
et al. 2000) or damage scoring (Giraudoux et al.
1997) are open to large sampling error, they have proven effi-
cient and have lead to convincing results in monitoring the
demographic patterns of small mammals (Hansson and Hent-
tonen 1985; Hanski and Tiainen 1988; Duhamel et al. 2000;
Michelat and Giraudoux 2006).
Variations in prey density
Two complete population cycles of M. arvalis and one of
A. terrestris were observed. All population declines were
synchronous. This pattern was consistent with those gener-
ally observed in the French Jura mountains (Saucy 1988;
Delattre et al. 1992; Giraudoux et al. 1997).
Table 2. Proportion of total deviance explained by the logistic regression between
prey frequency in the diet of Barn Owls (Tyto alba) and abundance in the field.
Ma/Maf* At/Atf
{
AfMg/AfMgf
{
So/Sof
§
Levier
Percent total deviance 36 61 7 2
Coefficient 1.96 2.40 1.37 –0.52
Standard error 0.10 0.08 0.23 0.13
p <2.10
–16
<2.10
–16
<1.10
–9
<9.10
–5
Chapelle d’Huin
Percent total deviance 8 70 30 1
Coefficient 0.003 0.211 0.015 –0.018
Standard error 0.0003 0.0078 0.0013 0.0118
p <2.10
–16
<2.10
–16
<2.10
–16
ns
Le Souillot
Percent total deviance 17 54 32 3
Coefficient 0.002 0.186 0.016 –0.029
Standard error 0.0003 0.0100 0.0015 0.0086
p <2.10
–16
<2.10
–16
<2.10
–16
<7.10
–4
Note: ns, not significant.
*Ratio of common voles (Microtus arvalis) in the diet of T. alba to M. arvalis in the field.
{
Ratio of European water voles (Arvicola terrestris) in the diet of T. alba to A. terrestris in the
field.
{
Ratio of yellow-necked mice (Apodemus flavicollis) and bank voles (Myodes glareolus) in the
diet of T. alba to A. flavicollis and M. glareolus in the field.
§
Ratio of Eurasian shrews (Sorex araneus) and Millet’s shrews (Sorex coronatus) combined
under Sorex spp. in the diet of T. alba to Sorex spp. in the field.
Bernard et al. 421
Published by NRC Research Press
Diet and dietary response of T. alba to variations in prey
density
Diets of T. alba consists mainly of small mammals, representing
about 90% of the prey in 79% of the studies reviewed by Tay-
lor (1984). Among small mammals, rodents constitute more
than one-half of the prey items in 90% of these studies. In
the current study, the frequency of rodents in the diet ranged
from 54% to 61%. Grassland species (M. arvalis and A. ter-
restris) were the dominant prey in the diet of T. alba. The
proportion of Sorex spp. was relatively high (mean between
22% and 34% of prey items). Woodland rodents were much
less represented. However, their frequency in the diet in-
creased temporarily (e.g., from winter 1989 to autumn 1990
at the three sites) despite their relatively low and stable popu-
lation densities. The dietary dominance of grassland rodents
and the contingent importance of shrews and of woodland ro-
dents are consistent with results already reported by other au-
thors (Goszczynski 1981; Catalisano and Massa 1987; Marti
1988). In the current study, differences in dietary composition
were recorded among sites. This may indicate local variations
in prey distribution and accessibility owing to landscape dif-
ferences, as well as to differences in the individual behaviour
of T. alba.
Previous studies of dietary response of T. alba generally
were based on pellet contents without assessing availability
of prey (e.g., Goutner and Alivizatos 2003; Pardinas and
Teta 2005; Leonardi and Dell’Arte 2006; Charter et al.
2009; Platt et al. 2009). It was sometimes assumed that pel-
let analysis provided a reliable indication of the proportion
of vertebrates in the field (Taylor 1994; Cooke et al. 1996).
This implies that prey diversity in the diet of T. alba directly
reflects the community structure and composition of small
mammals, and so T. alba capture prey randomly (Ba et al.
2000). The present study is the first one to our knowledge
to document the dietary response of T. alba to variations in
prey density in a system where two dominant species fluctu-
ate greatly. It is also the first study in temperate ecosystems
to take variations in densities of prey populations into
account, considering the community of prey species exten-
sively. The proportions of A. terrestris, M. arvalis, and
Table 3. Logistic regression and interactions between the two grassland species (common voles (Microtus arvalis) and European water
voles (Arvicola terrestris)) in Levier, Chapelle d’Huin, and Le Souillot.
Residual
Coefficient Deviance df Deviance p(c
2
)
Levier (pellet collection site)
Response: frequency of A. terrestris in the diet (p
KS
(normality of residuals) = 0.02; percent total deviance = 65%)
Intercept –2.8 28 1567.7
A. terrestris 0.002 6.1 26 600.6 0.01
M. arvalis 0.29 959.0 27 606.7 <0.000001
A. terrestris M. arvalis –0.00005 53.5 25 547.1 <0.000001
Response: frequency of M. arvalis in the diet (p
KS
(normality of residuals) = 0.5; percent total deviance = 46%)
Intercept –1.04 28 1223.9
M. arvalis –0.01 99.7 26 680.9 <0.000001
A. terrestris 0.004 443.4 27 780.6 <0.000001
M. arvalis A. terrestris 0.0004 27.2 25 653.7 <0.000001
Chapelle d’Huin (pellet collection site)
Response: frequency of A. terrestris in the diet (p
KS
(normality of residuals) = 0.02; percent total deviance = 72%)
Intercept –3.1 27 1195.5
A. terrestris –0.0001 18.0 25 342.0 0.00002
M. arvalis 0.28 835.5 26 360.0 <0.00001
A. terrestris M. arvalis –0.0003 6.7 24 335.3 0.01
Response: frequency of M. arvalis in the diet (p
KS
(normality of residuals) = 0.04; percent total deviance = 65%)
Intercept –0.86 27 892.9
M. arvalis –0.17 511.9 25 311.8 <0.00001
A. terrestris 0.005 69.3 26 823.6 <0.00001
M. arvalis A. terrestris 0.00005 0.3 24 311.4 0.56
Le Souillot (pellet collection site)
Response: frequency of A. terrestris in the diet (p
KS
(normality of residuals) = 0.50; percent total deviance = 55%)
Intercept –2.3 26 625.4
A. terrestris –0.001 1.9 24 284.8 0.17
M. arvalis 0.15 338.8 25 286.7 <0.000001
A. terrestris M. arvalis 0.0002 1.7 23 283.1 0.2
Response: frequency of M. arvalis in the diet (p
KS
(normality of residuals) = 0.50; percent total deviance = 31%)
Intercept –1.06 26 440.2
M. arvalis 0.04 50.3 24 315.7 <0.0001
A. terrestris 0.004 74.2 25 366.1 <0.000001
M. arvalis A. terrestris –0.0005 14.2 23 301.5 <0.00001
Note: p
KS
is the probability resulting from a Kolmogorov–Smirnov test of goodness of fit.
422 Can. J. Zool. Vol. 88, 2010
Published by NRC Research Press
woodland rodents in the diet were found to vary in propor-
tion to their relative densities in the field at all three study
sites. This result is consistent with the common belief that
T. alba is an opportunistic predator. However, despite its
low body mass (mean = 8 g), which might have made it a
nonprofitable prey for T. alba, the proportion of Sorex spp.
in the diet was generally high and was not correlated with
its density in the field in Chapelle d’Huin and was nega-
tively correlated in Levier and Le Souillot. The common
shrew was recorded in every habitat of the study area, in-
cluding grasslands (Giraudoux et al. 1994). This may indi-
cate that the species was considered an alternate prey when
more profitable rodent prey species were at low densities.
Furthermore, we found that the proportions of M. arvalis
and A. terrestris in the diet of T. alba also were determined
by significant interactions between populations of
A. terrestris and M. arvalis in the field. This means that the
frequency of A. terrestris in the diet of T. alba depended not
only on its own population density in the field, but also on
the population density of M. arvalis, and a reciprocal pattern
applies to the frequency of M. arvalis in the diet. Our study
provides a limited range of comparisons; for instance, it
does not include a period when high densities of
A. terrestris could be observed during a low-density phase
of M. arvalis. However, it clearly shows that the response
of T. alba to variations in prey densities is more complex
than believed in a number of earlier studies that assumed
random selection by T. alba or which were based on the
monitoring of a single prey species.
Dietary studies undertaken in captivity emphasize the ex-
istence of food preferences (e.g., Ille 1991) in T. alba.
Mikkola (1983) mentioned a preference for Eurasian water
voles (Arvicola amphibius (L., 1758)), a species of similar
size to A. terrestris. To our knowledge, such a preference
has not been recorded previously in Europe for A. terrestris
(Bunn et al. 1982; Mikkola 1983; Cramp 1985; Taylor
1994). In the current study, A. terrestris was the biggest spe-
cies among those captured by T. alba: 60–120 g, compared
with 10–30 g for other voles and mice, and 5–15 g for
shrews. Studies of T. alba in other countries report a food
preference related to prey size. A theoretical model by
Yom-Tov and Wool (1997) indicated that optimal prey
mass for T. alba is between 80 and 100 g. Janes and Barss
(1985) reported a preference for medium-sized prey (33–
64 g) like the northern pocket gopher (Thomomys talpoides
(Richardson, 1828)), rather than for small prey (10–33 g).
The high reproductive success recorded in plantations of
palm tree where T. alba feed almost exclusively on medium-
sized rodents (80 g) is another example of the ability of
T. alba to benefit from medium-sized prey (Lenton 1984).
Luthy et al. (1985) suggested that the multiannual variations
in populations of T. alba were related to variations in
Fig. 4. Dietary response and interactions between the two grassland rodent species. (a) Frequency of European water voles (Arvicola ter-
restris) in the diet of Barn Owls (Tyto alba) and (b) frequency of common voles (Microtus arvalis) in the diet of T. alba.
Bernard et al. 423
Published by NRC Research Press
population density of A. terrestris in the Swiss Jura moun-
tains. All those studies provide indications that T. alba pre-
fer larger small-mammal prey when available. However,
Granjon and Traore
´
(2007) reported a preference for younger
or smaller animals than the mean population of Hubert’s mas-
tomys (Mastomys huberti (Wroughton, 1909)) in Mali. Gran-
jon and Traore
´
(2007) hypothesize that M. huberti may
occupy less-favourable habitats in which they are more vul-
nerable to predation. Our study shows (i) significant correla-
tions between the proportion of A. terrestris, M. arvalis,and
woodland rodents in the diet of T. alba, as well as their re-
spective densities in the field; (ii) interactions between popu-
lations of A. terrestris and M. arvalis, indicating that the
proportion of either of these species in the diet of T. alba
was affected by the density of the other; (iii) Sorex spp. in
the diet of T. alba did not correlate or correlated negatively
with their abundance in the field, making those species alter-
nate preys when others were no longer available. This status
of alternate prey was already observed by Balciauskiene and
Narusevicius (2006) for S. aluco in Lithuania.
Optimal foraging theory predicts that T. alba concentrate
on the largest and presumably most profitable prey, and that
inclusion of other prey types in the diet depends not on their
own abundance but on the abundance of the most profitable
prey species (Schluter 1981; Steenhof and Kochert 1988).
Our results are consistent with this prediction for rodent spe-
cies prey versus Sorex spp. However, a simple preference
for larger prey would imply that A. terrestris would be pre-
ferred over M. arvalis when available, which would translate
into a dietary response that was independent of densities of
M. arvalis, a pattern that was not observed. In fact the two
grassland species were consumed in proportion to their
abundance in the field, with interactions, as noted above.
This indicates that the proportion of one species in the diet
of T. alba was dependent not only on its own population
density in the field but also on the population density of the
other species, the effect being mathematically multiplicative
and not only additive. The concept of profitability may be
complex to evaluate in the field in real conditions. The
present study provides evidence that for T. alba, profitability
does not depend on the mass of prey alone. One can
hypothesize that in our case, profitability may also depend
on prey accessibility (see also Granjon and Traore
´
2007).
Arvicola terrestris is a much more subterranean species
than M. arvalis and so may be less accessible to T. alba.In
periods of lower density, this may offset, at least in part, the
lower mass of M. arvalis in terms of its profitability to
T. alba and may explain the observed result.
Acknowledgements
This study was carried out as part of the Rodents and
Rural Planning’ 1992–1996 contract and the Plan d’action
campagnol 2001–2007’’, with financial support of the Con-
seil re
´
gional de Franche-Comte
´
(France). We are extremely
grateful to Kent Livezey for his kind guidance and support
in the revision of the earlier version of the manuscript.
References
Ba, K., Granjon, L., Hutterer, R., and Duplantier, J.M. 2000. Les
micromammife
`
res du Djoudj (Delta du Se
´
ne
´
gal) par l’analyse
du re
´
gime alimentaire de la chouette effraie, Tyto alba. Bonn.
Zool. Beitr. 1–4: 31–38.
Balciauskiene, L., and Narusevicius, V. 2006. Coincidence of small
mammal trapping data with their share in the Tawny Owl diet.
Acta Zool. Lituanica, 1(2): 93–101.
Bellocq, M.I. 2000. A review of the trophic ecology of the Barn
Owl in Argentina. J. Raptor Res. 34(2): 108–119.
Bunn, D.S., Warburton, A.B., and Wilson, R.D.S. 1982. The Barn
Owl. T & AD Poyser, Calton, UK.
Catalisano, A., and Massa, B. 1987. Considerations on the structure
of the diet of the Barn Owl (Tyto alba) in Sicily (Italy). Ital. J.
Zool. 54: 69–73. doi:10.1080/11250008709355559 .
Charter, M., Izhaki, I., Meyrom, K., Motro, Y., and Leshem, Y.
2009. Diets of Barn Owls differ in the same agricultural region.
Wilson J. Ornithol. 121(2): 378–383. doi:10.1676/08-083.1.
Cooke, D., Nagle, A., Smiddy, P., Fairley, J., and Muircheartaigh,
I.O. 1996. The diet of the Barn Owl Tyto alba in County Cork
in relation to land use. Biol. Environ. 96B(2): 97–111.
Cramp, S. 1985. Handbook of the Birds of Europe, the Middle East
and North Africa. Oxford University Press. Oxford.
De Bruijn, O. 1994. Population ecology and conservation of the
Barn Owl Tyto alba in farmland habitats in Liemers and Achter-
hoek (The Netherlands). Ardea, 82(1): 1–109.
Delattre, P., Giraudoux, P., Baudry, J., Musard, P., Toussaint, M.,
Truchetet, D., Stahl, P., Poule, M.L., Artois, M., Damange,
J.-P., and Que
´
re
´
, J.-P. 1992. Land use patterns and types of
common vole (Microtus arvalis) population kinetics. Agric.
Ecosyst. Environ. 39(3–4): 153–168. doi:10.1016/0167-8809(92)
90051-C.
Duhamel, R., Que
´
re
´
, J.-P., Delattre, P., and Giraudoux, P. 2000.
Landscape effects on the population dynamics of the fossorial
form of the water vole (Arvicola terrestris sherman). Landsc.
Ecol. 15(2): 89–98. doi:10.1023/A:1008173508817.
Dupuy, G., Giraudoux, P., and Delattre, P. 2009. Numerical and
dietary responses of a predator community in a temperate zone
of Europe. Ecography, 32(2): 277–290. doi:10.1111/j.1600-
0587.2008.04930.x.
Forsman, E.D., Anthony, R.C., Meslow, E.C., and Zabel, C.J. 2004.
Diets and foraging behavior of Northern Spotted Owls in Ore-
gon. J. Raptor Res. 38(3): 214–230.
Giraudoux, P., Delattre, P., Que
´
re
´
, J.-P., and Damange, J.-P. 1994.
Distribution and kinetics of rodent populations in a region under
agricultural land abandonment. Acta Oecol. 15(4): 385–400.
Giraudoux, P., Pradier, B., Delattre, P., Deblay, S., Salvi, D., and
Defaut, R. 1995. Estimation of water vole abundance by using
surface indices. Acta Theriol. 40(1): 77–96.
Giraudoux, P., Delattre, P., Habert, M., Que
´
re
´
, J.-P., Deblay, S.,
Defaut, R., Duhamel, R., Moissenet, M.F., Salvi, D., and Tru-
chetet, D. 1997. Population dynamics of fossorial water vole
(Arvicola terrestris scherman): a land usage and landscape per-
pective. Agric. Ecosyst. Environ. 66(1): 47–60. doi:10.1016/
S0167-8809(97)80706-2.
Giraudoux, P., Delattre, P., Takahashi, K., Raoul, F., Que
´
re
´
, J.-P.,
Craig, P., and Vuitton, D. 2002. Transmission ecology of Echino-
coccus multilocularis in wildlife: what can be learned from com-
parative studies and multiscale approaches? In Cestode zoonoses:
echinococcosis and cysticercosis. Edited by P. Craig and Z. Paw-
lowski. IOS Press, Amsterdam, the Netherlands. pp. 251–265.
Glue, D.E. 1967. Prey taken by the Barn Owl in England and Wales.
Bird Study, 14(3): 169–183. doi:10.1080/00063656709476160.
Goszczynski, J. 1976. Estimation of daily food ratio of Tyto alba
Scopoli under natural conditions. Pol. Ecol. Stud. 2: 95–102.
Goszczynski, J. 1981. Comparative analysis of food of owl in agro-
cenoses. Ekol. Polska, 29: 413–439.
424 Can. J. Zool. Vol. 88, 2010
Published by NRC Research Press
Goutner, V., and Alivizatos, H. 2003. Diet of the Barn Owl (Tyto
alba) and Little Owl (Athene noctua) in wetlands of northeastern
Greece. Belg. J. Zool. 133(1): 15–22.
Granjon, L., and Traore
´
, M. 2007. Prey selection by ban owls in
relation to small-mammal community and population structure
in a Sahelian agro-ecosystem. J. Trop. Ecol. 23(02): 199–208.
doi:10.1017/S026646740600383X.
Gubanyi, J., Case, R., and Wingfield, G. 1992. Diet and nesting
success of Barn Owls breeding in western Nebraska. Am. Midl.
Nat. 127(2): 224–232. doi:10.2307/2426528.
Hamer, T., Hays, D., Senger, C., and Forsman, E. 2001. Diets of
Northern Barred Owls and Northern Spotted Owls in an area of
sympatry. J. Raptor Res. 35(3): 221–227.
Hanski, I., and Tiainen, J. 1988. Populations and communities in
changing agro-ecosystems in Finland. Ecol. Bull. 39: 159–168.
Hansson, L. 1979. Field signs as indicators of vole abundance. J.
Appl. Ecol. 16(2): 339–347. doi:10.2307/2402512.
Hansson, L., and Henttonen, H. 1985. Gradients in density varia-
tions of small rodents: the importance of latitude and snow
cover. Oecologia (Berl.), 67(3): 394–402. doi:10.1007/
BF00384946.
Huitu, O., Norrdahl, K., and Korpima
¨
ki, E. 2004. Competition, pre-
dation and interspecific synchrony in cyclic small mammal com-
munities. Ecography, 27(2): 197–206. doi:10.1111/j.0906-7590.
2003.03684.x.
Ille, R. 1991. Preference of prey size and profitability in Barn Owls
Tyto alba guttata. Behaviour, 116(3–4): 180–189. doi:10.1163/
156853991X00021.
Janes, S.W., and Barss, J.M. 1985. Predation by three owl species
on northern pocket gophers of different body mass. Oecologia
(Berl.), 67(1): 76–81. doi:10.1007/BF00378454.
Jorgensen, E.E., Sell, S.M., and Demarais, S. 1998. Barn Owl prey
use in Chihuahuan Desert foothills. Southwest. Nat. 43(1): 53–
56.
Karell, P., Ahola, K., Karstinen, T., Zolei, A., and Brommer, J.E.
2009a. Population dynamics in a cyclic environment: conse-
quences of cyclic food abundance on tawny owl reproduction
and survival. J. Anim. Ecol. 78(5): 1050–1062. doi:10.1111/j.
1365-2656.2009.01563.x. PMID:19493132.
Karell, P., Pietiainen, H., Siitari, H., Pihlaja, T., Kontiainen, P., and
Brommer, J.E. 2009b. Parental allocation of additional food to
own health and offspring growth in a variable environment.
Can. J. Zool. 87(1): 8–19. doi:10.1139/Z08-133.
Kekkonen, J., Kolunen, H., Pietiainen, H., Karell, P., and Brom-
mer, J.E. 2008. Tawny owl reproduction and offspring sex ratios
under variable food conditions. J. Ornithol. 149(1): 59–66.
doi:10.1007/s10336-007-0212-7.
Le Louarn, H., and Que
´
re
´
, J.-P. 2003. Les Rongeurs de France:
Faunistique et Biologie. Institut National de la Recherche Agro-
nomique Editions, Paris.
Lenton, G.M. 1984. The feeding and breeding ecology of Barn
Owls Tyto alba, in peninsular Malaysia. Ibis, 126(4): 551–575.
doi:10.1111/j.1474-919X.1984.tb02080.x.
Leonardi, G., and Dell’Arte, G.L. 2006. Food habits of the Barn
Owl (Tyto alba) in a steppe area of Tunisia. J. Arid Environ.
65(4): 677–681. doi:10.1016/j.jaridenv.2005.10.007.
Lesin
´
ski, G., Gryz, J., and Kowalski, M. 2008. Does the diet of an
opportunistic raptor, the tawny owl Strix aluco, reflect long-term
changes in bat abundance? A test in central Poland. Folia Zool.
(Brno), 57: 258–263.
Lesin
´
ski, G., Ignaczak, M., and Manias, J. 2009. Opportunistic pre-
dation on bats by the tawny owl Strix aluco. Anim. Biol. 59(3):
283–288. doi:10.1163/157075609X454917.
Lidicker, W.Z., Jr. 2000. A food web/landscape interaction model
for microtine rodent density cycles. Oikos, 91(3): 435–445.
doi:10.1034/j.1600-0706.2000.910304.x.
Lima, M., and Jaksic, F.M. 1999. Population rate of change in the
leaf-eared mouse: The role of density-dependence, seasonality
and rainfall. Aust. J. Ecol. 24(2): 110–116. doi:10.1046/j.1442-
9993.1999.241951.x.
Lima, M., Stenseth, N.C., and Jaksic, F.M. 2002. Food web struc-
ture and climate effects on the dynamics of small mammals and
owls in semi-arid Chile. Ecol. Lett. 5(2): 273–284. doi:10.1046/
j.1461-0248.2002.00312.x.
Livezey, K.B. 2007. Barred Owl habitat and prey: a review and
synthesis of the literature. J. Raptor Res. 41(3): 177–201.
doi:10.3356/0892-1016(2007)41[177:BOHAPA]2.0.CO;2.
Love, R.A., Webbon, C., Glue, D.E., and Harris, S. 2000. Changes
in the food of British Barn Owls (Tyto alba) between 1974 and
1997. Mammal Rev. 30(2): 107–129. doi:10.1046/j.1365-2907.
2000.00060.x.
Luthy, P., Shaller, J.C., Monnerat, M., and Georgy, A. 1985. La re-
production de la Chouette effraie, Tyto alba, dans le district de
De
´
le
´
mont (Canton du Jura, Suisse) en 1982 et 1983, et les an-
ne
´
es suivantes. Nos Oiseaux, 38(4): 161–178.
Marti, C.D. 1988. A long-term study of the food-niche dynamics in
the Common Barn Owl: comparisons within and between popu-
lations. Can. J. Zool. 66(8): 1803–1812. doi:10.1139/z88-261.
Michelat, D., and Giraudoux, P. 1991. Dimension du domaine vital
de la Chouette effraie (Tyto alba) pendant la nidification.
Alauda, 59(3): 137–142.
Michelat, D., and Giraudoux, P. 1993. Relations proies–pre
´
dateur–
paysage chez la Chouette effraie (Tyto alba), pendant l’e
´
levage
des jeunes. Alauda, 61(2): 65–72.
Michelat, D., and Giraudoux, P. 2006. Synchrony between small
mammal population dynamics in marshes and adjacent grassland
in a landscape of the Jura plateau, France: a ten year investiga-
tion. Acta Theriol. 51: 155–162.
Mikkola, H. 1983. Owls of Europe. T & AD Poyser, Calton, UK.
Pardinas, U.F.J., and Teta, P. 2005. Vertebrate prey of the Barn
Owl (Tyto alba
) in subtropical wetlands of northeastern Argen-
tina and eastern Paraguay. J. Raptor Res. 39(1): 65–69.
Pavey, C.R., Gorman, J., and Heywood, M. 2008. Dietary overlap
between the nocturnal letter-winged kite Elanus scriptus and
barn owl Tyto alba during a rodent outbreak in arid Australia.
J. Arid Environ. 72(12): 2282–2286. doi:10.1016/j.jaridenv.
2008.07.013.
Petty, S.J. 1999. Diet of tawny owls (Strix aluco) in relation to
field vole (Microtus agrestis) abundance in a conifer forest in
northern England. J. Zool. (Lond.), 248(4): 451–465. doi:10.
1111/j.1469-7998.1999.tb01045.x.
Platt, S.G., Rainwater, T.R., Leavitt, D.J., and Miller, S.M. 2009.
Diet of Barn Owls (tyto alba) in northern Belize. Southwest.
Nat. 54(1): 104–107. doi:10.1894/MH-43.1.
Que
´
re
´
, J.-P., Raoul, F., Giraudoux, P., and Delattre, P. 2000. An
index method applicable at landscape scale to estimate relative
population densities of the common vole (Microtus arvalis).
Rev. Ecol. (Terre Vie), 55: 25–32.
R Development Core Team. 2005. R: a language and environment
for statistical computing. R Foundation for Statistical Comput-
ing, Vienna, Austria. Available from http://www.r-project.org/.
Raoul, F., Defaut, R., Michelat, D., Montadert, M., Pe
´
pin, D.,
Que
´
re
´
, J.P., Tissot, B., Delattre, P., and Giraudoux, P. 2001.
Landscape effects on the populations dynamics of small mam-
mal communities: a preliminary analysis of prey-resource varia-
tions. Rev. Ecol. (Terre Vie), 56: 339–352.
Rasoma, J., and Goodman, S.M. 2007. Food habits of the Barn Owl
(Tyto alba) in spiny bush habitat of arid southwestern Madaga-
Bernard et al. 425
Published by NRC Research Press
sar. J. Arid Environ. 69(3): 537–543. doi:10.1016/j.jaridenv.
2006.10.004.
Saucy, F. 1988. Dynamique de population, dispersion et organisa-
tion sociale de la forme fouisseuse du Campagnol terrestre
(Arvicola terrestris scherman). Ph.D. thesis, University of Neu-
cha
ˆ
tel, Neucha
ˆ
tel, Switzerland.
Schluter, D. 1981. Does the theory of optimal diets apply in complex
environments? Am. Nat. 118(1): 139–147. doi:10.1086/283810.
Spitz, F. 1963. Les techniques d’e
´
chantillonnage utilise
´
es dans l’e
´
tude
des populations de petits mammife
`
res. Rev. Ecol. (Terre Vie), 2:
203–237.
Spitz, F. 1964. Etude des densite
´
s de population de Microtus arva-
lis Pall. a
`
Saint-Michel-en-L’Herm (Vende
´
e). Mammalia, 28(1):
40–75. doi:10.1515/mamm.1964.28.1.40.
Spitz, F., Le Louarn, H., Poulet, A., and Dassonville, B. 1974.
Standardisation des pie
´
geages en ligne pour quelques espe
`
ces
de rongeurs. Rev. Ecol. (Terre Vie), 38: 564–578.
Steenhof, K., and Kochert, M.N. 1988. Dietary responses of three
raptors species to changing prey densities in a natural environ-
ment. J. Anim. Ecol. 57(1): 37–48. doi:10.2307/4761.
Taylor, R.H. 1984. Distribution and interactions of introduced ro-
dents and carnivores in New Zealand. Acta Zool. Fenn. 172:
103–105.
Taylor, I. 1994. Barn Owls: predator–prey relationships and conser-
vation. Cambridge University Press, Cambridge.
Venables, W.N., and Ripley, B.D. 2003. Modern applied statistics
with S. Springer-Verlag, Berlin, Germany.
Vernon, C.J. 1972. An analysis of owl pellets collected in southern
Africa. Ostrich, 43: 109–124.
Webster, J.A. 1973. Seasonal variation in mammal contents of
Barn Owl castings. Bird Study, 20(3): 185–196. doi:10.1080/
00063657309476380.
Yom-Tov, Y., and Wool, D. 1997. Do the contents of Barn Owl
pellets accurately represent the proportion of prey species in the
field? Condor, 99(4): 972–976. doi:10.2307/1370149.
Za
´
rybnicka
´
, M., Sedla
´
c
ˇ
ek, O., and Korpima
¨
ki, E. 2009. Do Teng-
malm’s Owls alter parental feeding effort under varying condi-
tions of main prey availability? J. Ornithol. 150(1): 231–237.
doi:10.1007/s10336-008-0342-6.
Appendix A
Table A1. Sampling sessions, habitats, and traplines set
up from 1987 to 1994.
Month and year
Hedges and
forest edges Grasslands
April 1987 5 29
June 1987 9 15
July 1987 5 17
October 1987 8 19
April 1988 21 42
July 1988 5 20
October 1988 10 24
April 1989 10 33
July 1989 5 22
October 1989 10 26
April 1990 13 47
July 1990 5 22
October 1990 10 25
April 1991 9 20
July 1991 5 22
October 1991 10 19
April 1992 8 21
July 1992 5 21
October 1992 9 20
April 1993 18 29
July 1993 5 21
October 1993 9 20
April 1994 8 19
October 1994 7 18
April 1995 8 18
October 1995 7 18
Total number of traplines 224 607
426 Can. J. Zool. Vol. 88, 2010
Published by NRC Research Press
... The species typically inhabits agricultural landscapes where it hunts mostly small mammals such as rodents and shrews (Taylor 1994;Scherzinger and Mebs 2020). Local variations in diet usually correspond to the composition of small-mammal communities in a hunting territory and follow the fluctuations of availability and accessibility of prey populations (Tores et al. 2005;Miltschev and Georgiev 2009;Bernard et al. 2010;Paspali et al. 2013;Horváth et al. 2018Horváth et al. , 2020Milana et al. 2019;Szép et al. 2019;Romano et al. 2020). Therefore, Common Barn-owls are important regulators of populations of mammalian agricultural pests (Wood and Fee 2003;Peleg et al. 2018). ...
... The lack of data on the populations of small mammals in the area and in its diverse microhabitats did not permit assessing how selectively the owls hunted. The study suggested the plastic hunting strategy of Common Barn-owl as an opportunistic generalist (Bernard et al. 2010;Veselovský et al. 2017;Saufi et al. 2020), allowing it to inhabit a remarkable variety of different habitats combinations in southern Bulgaria. ...
Article
Full-text available
The diet of the Common Barn-owl in a forest- and shrub-dominated hunting area in the Strandzha Mountains, southern Bulgaria, was identified from 516 prey specimens. Shrews (52.9% by number, 26.7% by biomass) and rodents (42.1% N, 71.5% B) were prevalent. Among them, White-toothed shrews, Crocidura sp., (45.3% N, 21.4% B) was the most numerous prey genus. Mice, Apodemus sp., (15.7% N, 29% B) contributed with the largest share to the food biomass due to high predation of Striped field mice, A. agrarius, (12.2% B). The proportions of forest species in diet (Apodemus sp, Sorex sp., and dormice Gliridae) increased with the higher proportion of forest habitats (forests and shrublands cover more than 25% of the area) in most Barn Owl hunting territories in southern Bulgaria.
... The Common Barn-owl is very sensitive to the configuration of landscape elements and the changes in landscape composition (Andries et al. 1994, Bond et al. 2005, Frey et al. 2011, Hindmarch et al. 2012. Thus, several studies emphasised the impact of change in agricultural practices and intensification on its foraging pattern, which depends on the most frequent prey species, especially different herbivore microtine voles that are considered to be agricultural pests (Taylor 2004, Marti 1998, Bernard et al. 2010, Kross et al. 2016). On the other hand, several studies examined the hypothesis that variation in habitat features surrounding the nest sites determine the reproductive success of Common Barn-owls (Meek et al. 2009, Frey et al. 2011, Charter et al. 2012. ...
... Considering the small mammals' abundance distribution at the species level and their association to landscape features, the Common Vole was the most hunted prey, which was also described in several studies (e.g. Bernard et al. 2010, Frey et al. 2011, Veselovsky et al. 2017, Horváth et al. 2018. In both landscape types, this species accounted as the main predominant prey in the barn owl's food composition and appeared in a higher proportion in the landscape dominated by agricultural areas, reflecting that the Common Vole is a typical species of open lowlands and farmlands (Delattre et al. 1996 Heroldová et al. 2007, Arlettaz et al. 2010, Fischer et al. 2011. ...
Article
Full-text available
As an opportunistic predator, the Common Barn-owl (Tyto alba) proved to be an appropriate model organism to survey the composition of small mammal assemblages. This study analysed barn owls’ pellet samples from 14 localities containing 34 animal taxa and 4,088 prey items in two years (2015–2016). Two groups of samples (7–7 localities) were separated based on the dominance of semi-natural habitats and agricultural lands. Rarefaction analysis proved that the species richness and diversity of barn owls’ diet were significantly higher in semi-natural landscapes. The multiple regression analysis between PCA scores showed that in the agrarian landscape the abundance of generalist species was influenced by the proportion of forests, while the value of the trophic level index was determined by the size of arable fields. In the case of semi-natural landscapes, the abundance of the synantrop guild and generalist species, especially S. araneus and A. agrarius, was influenced by the proportion of urban areas, the number of habitats and the size of arable fields. The results of this study suggested that the small mammal consumption of the Common Barn-owl is significantly different in the two landscapes, which reflects the impact of habitat heterogeneity and agricultural activity on prey availability.
... Because T. alba is a cosmopolite raptor, known to prey predominantly on small mammals within their home range, the analysis of owl pellet samples is considered an extremely valuable tool for mammalogists, particularly in broad-scale studies (Avenant, 2005;Bonvicino & Bezerra, 2003;Massa et al., 2020;Teta et al., 2010). Moreover, it has been proposed that the relative number of type of prey found in these pellets can be taken, with caution and considering some limitations, as a proxy to the availability of micromammals present in a particular area (Bernard et al., 2010;Bonvicino & Bezerra, 2003;Errington, 1930;Lyman, 2012). This area is defined by the home range of the owl, which is considered to vary between 0.25 and 3 km in radius (Taberlet, 1983;Taylor, 1994). ...
Article
Full-text available
Hantavirus pulmonary syndrome (HPS) is a zoonotic emerging infectious disease caused by New World orthohantaviruses (family Hantaviridae) hosted by rodents of the family Cricetidae. In Argentina, one of its main hosts is the sigmodontine rodent Oligoryzomys flavescens, a widely distributed mouse of the Pampas, Delta and Espinal ecoregions of central‐east Argentina. Because the abundance of the reservoir and its proportion in the rodent community affects both virus prevalence and human exposure risk, its estimation throughout its known geographical distribution is of key importance for the design of public health strategies to prevent HPS. The aim of this study was therefore to model the relative abundance of O. flavescens in most of the Pampas ecoregion within Buenos Aires Province, Argentina, where hantavirus pulmonary syndrome is endemic. To do this we used owl‐pellet samples collected between 2006 and 2008 from 51 sites distributed throughout most of Buenos Aires province. Mammalian prey in each pellet was identified to the lowest possible taxonomic level by examination of the skulls, dentaries and molars. We modelled the frequency of O. flavescens found in each sample as a function of climatic, environmental, and topographic data of each site. The two best models were applied to a Geo referential Information System to build maps of estimated frequency (as a proxy of relative abundance) within Buenos Aires province. Estimated relative abundance of O. flavescens in Buenos Aires province was significantly associated with annual mean temperature, annual precipitation and presence of freshwater bodies, and varied among sub‐regions, with the Inland and Rolling Pampas being the regions with highest frequencies. Knowing in which areas O. flavescens abundance is expected to be higher can be used to concentrate limited sanitary efforts in those areas that are most needed in order to reduce transmission and increase detection.
... Owing good preservation of the remains, accumulations of pellets of birds of prey are most suitable for research of teeth wear, with the Barn owl (Tyto alba) and the Eagle owl (Bubo bubo) belonging to the most frequent accumulators in Northern Eurasia (Andrews 1990). Voles from the genus Microtus are constitute one of the main prey of these avian predators and mostly large number in deposits (Bernard et al 2010;Penteriani and del Mar Delgado 2019;Horváth et al 2020). ...
Article
Evaluation of macro- and mesowear of the cheek teeth of Microtus voles as indicators of diet in paleoecological reconstructions are by now only at the level of pilot studies but already show promising results. Microtus voles exhibit food plasticity, and the structure of their molars determines a subtle response to changes in food composition. Dental remains of voles are abundant in Late Pleistocene and Holocene localities. However, before estimating fossil material, it is necessary to study the range of variation in extant samples. The aim of the current study is to evaluate whether or not seasonal variability of tooth wear is visible in the common vole M. arvalis from the Ural region (Russia). This is the first in vivo study of vole tooth wear using intravital impressions method. The process of individual macro- and mesowear of the first lower molar of wild-caught common voles has been investigated and quantified. The resulting data show a tendency of decreasing occlusal angle and increasing crown height from April to June/July. Moreover, the differentiation of wear of the dense and soft dentine of the occlusal surface of the enamel prisms decreased, and the depth of the relief increased. All these tendencies are obviously correlated with the change from less abrasive to more abrasive green feed components. From July to August, a reverse trend can be observed returning to a similar as during the spring months.
... Shrews are more typical prey for the Barn Owl (Tyto alba) 66-68 than for the Long-eared Owl. 22 These functional responses in diet to a changing availability of the main prey are wellknown in Long-eared Owls 29,30,69 and also in other owl species, such as the Barn Owl, 68,70,71 and other raptors which prefer an agricultural landscape. 72,73 Other prey species whose abundance increased only in some winters were a temporary supplementary prey. ...
Article
Full-text available
Owl diets undergo qualitative changes across the different regions of their area of distribution. During the four winters (from 2014–15 to 2017–18), Long-eared Owls’ pellets were collected at three winterroosts located at the southern part of Pannonian Plain, in the Serbian province of Vojvodina. In 8070 prey items from pellets, we identified 16 mammal and 32 bird species. The Common Vole was the dominant prey species with a proportion in a range from 27.4% to 71.6%. The Muridae family formed a supplementary part of the diet: Mus sp., wood mouse and harvest mouse, during all winters. Birds were also a major supplementary prey during winter 2014–15, comprising 10.6%. A comparison of our results with the diet of Long-eared Owls wintering at the northern Pannonian plain (southwestern Slovakia) indicated an increase the proportion of some species in the southern part. How different land uses in agriculture and environmental conditions may be reflected in the food supply are discussed in relation to the diet composition of Long-eared Owls and an environment whose is richer provides both growing diversity in the diet of these owls and an expansion of their food niche. Our study described the Long-eared Owl as opportunistic predators expanding their food niche in the presence of diversified prey.
... The feeding habits of Barn Owls are highly influenced by population fluctuations of the common and water voles, which appeared to be more specialists in the selection of food items. This shows a highly complex correlation among common and water voles with forest rodents that favored the permanent establishment of roosts of the Barn Owl in woodland (Bernard et al. 2010). In South Australia, Barn Owls intermittently predated on a variety of rodents in the plague-affected area and played a significant role in lowering the rodent populations and ultimately decreased the incidence of the disease (Janžekovič & Klenovšek 2020). ...
Article
Full-text available
Owls are considered as dominant predators for control of rats and mice population in agricultural fields and presently their populations are continuously declining in Punjab, Pakistan. The present study was aimed to assess the knowledge and attitude of people of rural and urban areas about the declining trend of owls. During this survey, more than 1600 people were asked to collect information regarding the owl populations from six localities including Faisalabad, Sialkot, Jhang, Lahore and Bahawalnagar. Four parameters were the major part of the questionnaire: familiarity, misconceptions, awareness about owls and their acceptance as a biological controlling agent. The results revealed that about 95% of people were familiar with owls in the agro-ecosystem. Only 15.6% of people thought that owls should be eliminated; 23.0% were of the opinion that owl’s presence in a locality leads to ruination; 33.3% agreed that the owls presence was a sign of bad omen; 41.8% considered them as signs of foolishness; 47.0% believed that owl’s body parts were used for black magic purposes. In contrast, 50% of people acknowledged that owls are beneficial to humans; 60.3% knew that owls are the enemies of rodents, 67.7% agreed that they are suppressors of rats and mice and 63.8% agreed that artificial nest boxes can serve as their nests and roosts. It was encouraging to know that 74.8% showed their willingness to enhance the owl’s population on their farms, and 74.0% were willing to permit the installation of nest boxes in or near the villages. The study of attitudes of respondents towards owl will help to develop an effective conservation strategy and to boost owl’s population in croplands for biological control of rats and mice.
Chapter
Rodent outbreaks have, in historical memory and at more or less regular intervals, massively affected crops and stored goods locally and sometimes even regionally. The Jura Massif is known, among other things, for its cheese specialties with several emblematic protected designations of origin (PDO), such as Comté, Morbier, Mont‐d'or and Bleu de Gex. In the 1950s, in this region of medium‐altitude mountains, grassland covered between 20% and almost 75% of farmland, regardless of altitude. From the early 1980s to the mid‐1990s, population demographic monitoring of small mammals was carried out in the Septfontaines and Le Souillot area, covering nearly 20,000 hectares. It has been suggested that predation is a major driver of fluctuations in rodent populations. It has taken about 30 years to describe and understand the factors that determine vole outbreaks in the Jura Massif in PDO grassland areas and to establish the basis for controlling them.
Article
Full-text available
La recolección y análisis de egagrópilas resulta un método adecuado en el estudio de presas pequeñas, debido a que permite el estudio de áreas extensas, genera gran cantidad y variedad de individuos por lote. El orden Strigiformes cuenta con el mayor número de estudios de dieta de aves rapaces a nivel mundial, destacándose los referidos a Tyto furcata con gran cantidad de contribuciones sobre el análisis de sus regurgitados. El estudio de las regurgitaciones de esta especie ha ayudado a realizar nuevas interpretaciones sobre la distribución, taxonomía y conservación de muchas especies de roedores. Aquí, se describe la presencia de Holochilus chacarius en el sur de la provincia de Santa Fe a partir de la recolección y análisis de egagrópilas de la lechuza del campanario (Tyto furcata). Estos registros amplían la distribución de H. chacarius en la provincia y contribuyen a actualizar los mapas de distribución existentes.
Article
Full-text available
Damage associated with rodent's outbreaks is common in the actual intensive agriculture. In order to reduce these economic losses, the placement of nest boxes for raptors could be a useful alternative to the chemical control programs. Barn owl (Tyto alba) and common kestrel (Falco tinnunculus) are two of the most commonly used species as rodent biological controls. The increase of the density of these two raptors is able to reduce the rodent's abundance without the huge negative impact caused by the use of chemical products in surrounding biodiversity. However, other prey species could be affected by these raptors too. In order to evaluate the different impacts of barn owl and common kestrel on non-pest species, we studied the diet in 50 paired nest boxes during 3 consecutive years, based on 1249 pellets. The proportion of small mammals considered pests for agriculture (voles and mice) consumed by barn owl (80%) and common kestrel (75.6%) were quite similar. Based on nutritional requirements, nest-box occupancy probability and productivity, we estimated a consumption of 5.24 rodents per day per common kestrel nest-box placed and of 4.95 rodents per barn owl nest-box placed. However, shrews were significantly more captured by barn owl and birds were significantly more captured by common kestrel. Our results showed that both species of raptors can be effective in controlling rodents that are harmful to agriculture, but the common kestrel seemed to be slightly more efficient and might also prey on insect taxa that could be important agricultural pests.
Article
Full-text available
The hypothesis tested concerned whether the diet of the tawny owl Strix aluco, as an opportunistic predator, does reflect increases in the density of bat populations in the owl’s hunting areas. In our study area, there was a mass use of toxic pesticides during which numbers of bats declined drastically, after which recoveries in the populations of most European species occurred. Thus, in Poland, numbers of bats reached their lowest levels in the 1980s. We examined the diets of tawny owls in Warsaw and the adjacent Kampinos Forest of central Poland, based on the remains of 9142 prey items. Bat specimens were found to comprise the following percentages of all vertebrate prey items: 1976–1989: 0.03–0.14%, 1990–1999: 0.32–0.40%, and 2000–2007: 0.54–1.71%. If the share taken by bats among mammalian prey is in turn considered, the analogous figures are 0.09–0.17%, 0.45–0.99% and 0.92–3.26%. Patterns in owl diets were consistent with trends in bat numbers at 15 large winter roosts located some 10–50 km from the study area in 1989–2006.
Book
A guide to using S environments to perform statistical analyses providing both an introduction to the use of S and a course in modern statistical methods. The emphasis is on presenting practical problems and full analyses of real data sets.
Article
Prey density was the same both inside and outside the home range; within it, prey density was more than twice as high in open fields than in areas with hedgerows. However, the barn owl hunted more often in the hedgerow habitat. This leads to a discussion on the influence of landscape structure on foraging strategy. -from English summary
Article
Examines the effect of different densities of vole Microtus arvalis on the diets of Strix aluco, Asio otus and Tyto alba. Seasonal variability of the diet and the variability due to environment are also discussed. Specialization of owls in capture of voles and its consequences are discussed. -from Author
Article
In Doubs, France the male had a home range size >750 ha. Average area forged in any one night was 250 ha. Rain during the night decreases the amount of time spend flying but not the size of the area covered. In some areas the females' home range partially overlapped that of the male, in other areas not at all. -from English summary
Article
From January to May, 1993 we collected 725 skulls from barn owl (Tyto alba) pellets at three active nests in foothills of the Sacramento Mountains, southern New Mexico. To estimate rodent abundance, we concurrently live-trapped 1,555 rodents, from 48 trapping grids distributed in six habitat types near the nests, and marked 1,236 of these with monel ear tags. Prey taken by owls was not concordant with any possible combination of availability. Barn owls demonstrated a considerable selectivity for Sigmodon and Perognathus, while failing to capture Chaetodipus and Peromyscus in numbers representative of their abundance. Sigmodon was abundant in only one uncommon habitat type that was not close to nests. Barn owl selection for Sigmodon indicates that they forage in favored habitat patches. Further, because Reithrodontomys and Peromyscus also occurred in this habitat but were not taken in great numbers, we also conclude that barn owls selected favored prey species within favored habitat patches.