The Long-eared Owl (Asio otus) is a Holarctic, nocturnal bird of prey whose diet has been extensively studied in North America and Europe. The literature on its diet indicates that it feeds mainly on rodents, especially voles, and other small mammals, complemented by other animals, including birds

and invertebrates (Cramp and Simmons 1985). It forages in the open, but also hunts near and below trees (Cramp and Simmons 1985) where it feeds on birds that roost in trees. In Israel there are resident, migratory, and wintering populations of Long-eared Owls (Paz 1987, Shirihai 1996). The species inhabits most low-lying areas in Israel, chiefly in the Mediterranean region, but also in agricultural settlements in the desert during the last three decades. This species prefers semi-open areas in Israel such as agricultural settlements, plantations, and patches or lines of trees (Shirihai 1996). The diet of the Long-eared Owl has not been studied in Israel. We collected data on the diet of the Long-eared Owl and report on the composition of the diet in the northern and central Negev desert, Israel.

METHODS

We studied Long-eared Owls in the northern and central Negev, Israel, a relatively arid region where rain occurs only during winter (Nov-Apr). Annual precipitation ranges between 300 mm in the north to 100 mm in the south and varies greatly from year to year. Mean monthly temperature ranges between 26[degrees]C in July and 11[degrees]C in January (Jaffe 1986).

We located communal winter roosts of Long-eared Owls and nesting sites during the breeding season in or near agricultural settlements; these were visited once or twice every month between May 2002 and December 2003. Long-eared Owls roost in dense vegetation which, in our study area, occurs only in settlements. All pellets were collected during each visit and the area was cleaned of pellets and remains of prey; each collection was composed of "fresh" pellets accumulated since the last visit. Each pellet was placed in a separate bag and its date and locality were recorded. All pellets were allocated to one of four seasons: winter (Dec-Feb), spring (Mar-May), summer (Jun-Aug), and autumn (Sep-Nov). Pellets were identified by their gray or light-black color and relatively (in comparison with those of the sympatric Barn Owl [Tyto alba]) narrower width, and with duller color. Pellets where species identification was not clear were not used. Each pellet was either separated to its components in the laboratory using tweezers or soaked in water until the remains, and cranial and post-cranial elements could be separated. Species identification was by comparison with identified specimens in the collection of the Zoological Museum of Tel Aviv University and the collection of The Hebrew University of Jerusalem.

We treated each single pellet as a unit containing the remains of the complete portion of food eaten. Bones of one prey item under field conditions usually appear in one pellet and only rarely are they discarded in two or more pellets (Raczyfiski and Ruprecht 1974). However, there is the possibility that owls will cache large prey items and return to them later. In these cases, over estimation of the large prey items would occur using our pellet analysis. Our data suggests that over estimation could have occurred for only one prey species (Rattus rattus) in one locality (Nirim).

The main identifiable elements were crania, mandibles, and femura for mammals, skulls and humeri for birds, mandibles for reptiles, and exoskeleton pieces for invertebrates. The minimum number of individuals (MNI) was calculated from the most common element for every species. All remains in the pellets were identified to the lowest possible taxon. Differences between species were often small and most remains were identified to genus. A small proportion (2-11%) of post cranial remains could not be identified to genus. These remains belonged to either Meriones or Gerbillus (Gerbilidae) and we divided them between the two genera by their relative amount actually found in the identifiable remains. Identified bird species were categorized by their status in Israel (residents or migrants). However, some bird genera and even species have both resident and migratory populations in Israel. The remains in these cases were categorized to the status of the most common species in the genus in the study area.

The seven settlements from which we collected pellets were (number of pellets collected in parentheses): Sde Boqer 30[degrees] 52' N, 34[degrees] 48' E (280 pellets), Revivim 31[degrees] 03' N, 34[degrees] 44' E (380), Gevulot 31[degrees] 12' N, 34[degrees] 28' E (340), Tse'elim 31[degrees] 12' N, 34[degrees] 32' E (97), Tzohar settlements (a group of 6 settlements within 1-2 km) 31[degrees] 14' N, 34[degrees] 25' E (480), Omer 31[degrees] 17' N, 34[degrees] 50'E (157), and Nirim 31[degrees] 19' N, 34[degrees] 22' E (1,300).

All but one (Omer) of the sites were agricultural settlements, and all had ornamental trees, bushes, and grass lawns. The surrounding fields contained agricultural crops (potatoes, peanuts, carrots, etc.) as well as plantations (olives, avocado) and had lines of or scattered ornamental trees (various genera of palms, Tamarisk ssp., Acacia ssp.). The natural small mammalian fauna consists of several species of rodents and shrews with a strong psammophilous element where loess soil was mixed with sand. Commensal rodents (mice and rats) were also present.

We used two-way ANOVA tests for each of the three food categories (rodents, invertebrates, and birds) to test if there were differences in diet composition among localities and seasons. We used the Shannon-Wiener Index to calculate a value for diet diversity for each locality and two-way ANOVAs to test differences in diet diversity among localities and seasons.

RESULTS

We collected 3,034 pellets of Long-eared Owls, comprising 4,668 prey items with a composition of 71.3% small mammals, 26.5% birds, 2.0% invertebrates, and 0.1% reptiles (Table 1). There were no significant differences among localities or among seasons in percent rodents (P-values were 0.331 and 0.076, respectively) nor in percent invertebrates (P = 0.513 and P = 0.473, respectively) in the diet. Percent birds in the diet did not differ among localities (P = 0.144), but differed among seasons (P = 0.037). There were no significant differences in diet diversity (Shannon-Wiener Index) among localities (P = 0.290) or among seasons (P = 0.803).

There were no differences in percent rodents (as a group) in the diet among localities, but rodent genera composition differed among localities: percent Gerbillus among rodents changed from 63-70% in Gevulot, Revivim, and Tse'elim to 43% in Tzohar, and 16% in Sde Boqer, and only 1-2% in Nirim and Omer, while the proportion of Meriones was highest (42-46%) in Nirim and Omer, 17-20% in Revivim, Tse'elim, Gevulot, and Tzohar and 10% in Sde Boqer. Jaculus composed 9, 6, and 1% in Revivim, Sde Boqer, and Gevulot, respectively. Commensal species comprised 67% (Mus only) of the rodents in Sde Boqer, 56% (30% Mus and 26% Rattus) in Nirim (with a possible risk of over estimation of Rattus in the pellets), 52% (42% Mus and 10% Rattus) in Omer, and 37, 18, and 11% in Tzohar, Tse'elim, and Gevulot, respectively.

We found a total of 1,236 bird remains (Table 1) and identified 881 of them. These belonged to at least 23 species, and between 31 and 85% of which (depending upon location) were House Sparrows (Passer domesticus). Percent birds in the diet differed significantly among seasons. The number of remains of resident birds was 156, 92, 203, and 295 in autumn, winter, spring, and summer, respectively, while the respective numbers for migrants were 33, 11, 73, and 18. The number of resident bird remains was similar during the migratory (spring and autumn) and non-migratory (winter and summer) seasons (359 and 387, respectively), but the number of migrant remains, which consisted mainly of small passerines, was 3.7 times larger during the migratory seasons than during the non-migratory period (106 and 29, respectively). Thus, migratory birds comprised a significantly larger part (of the total birds consumed) during migration than during the non-migratory months (Fisher's exact test, P = 2 x [10.sup.-11]).

DISCUSSION

The diet of the Long-eared Owl in North America, Europe, and Japan is composed mainly of voles (Glue and Hammond 1974; Marti 1974, 1976; Nilsson 1981; Village 1981; Marks 1984; Bosakowski and Smith 1991; Tome 1994; Capizzi et al. 1998; Alivizatos and Goutner 1999; Navarro et al. 2003; Chiba et al. 2005). The one species of vole in the Mediterranean region in Israel (Microtus guentheri) is not found in the desert and occurs only in the northern part of our study area. We found only three voles in the diet of the Long-eared Owl (at Nirim, the most northern of our localities). The main species of prey in our study area were gerbils, jirds, mice, and rats. We believe the proportions of rodent species in the diet of the Long-eared Owl reflects the composition of the rodent communities where the pellets were collected. The soil in our study area is either sand or loess (wind deposited) or a mixture of different proportions of these two types. There are four species of psammophile Gerbillus (G. pyramidum, G. allenbyi, G. gerbillus, and G. henleyi) and one non-psammophile (G. dasyurus) that occurs mainly in rocky habits. The diet of the Long-eared Owl in areas where the soil was sandy or had a large proportion of sand (Revivim, Gevulot, and Tse'elim) was mainly Gerbillus (63-70% of the rodents in the diet), while in loess-dominated areas (Nirim, Omer, Sde Boqer) it was below 17%. The proportion of Gerbillus constituted 43% of the rodents in Tzohar, where the soil is sandy-loess. Gerbillus was replaced by other rodents, mainly Meriones, where the soil was not sandy. Commensal rodents (house mouse [Mus musculus] and black rat [Rattus rattus]) comprised >50% of the diet in Sde Boqer, Nirim, and Omer. The high proportion of house mice in the diet in Sde Boqer is explained by a mice plague that occurred there in 2003, while the high proportion of commensal species in Omer is probably the result of this being an urban suburb of the city of Beer Sheva. Jerboas (Jaculus jaculus) occur on both types of soil, and occurred in a low proportion in Revivim (sand) and Sde Boqer (loess).

The numbers of resident bird remains were similar during the migratory (spring and autumn) and non-migratory (winter and summer) seasons (359 and 387, respectively). The number of remains of migrant birds was 3.7 times larger during the migratory season than during the non-migratory period (106 and 29, respectively).

The Long-eared Owl has been characterized as having a more specialized diet than other sympatric owls (Marti 1976, Andrews 1990). Other studies (Marks 1984, Bertolino et al. 2001) demonstrated that Long-eared Owls feed opportunistically. Our study indicates the diet of the Long-eared Owl in the northern and central Negev desert consists mainly of rodents (71.3%), but also of birds (26.5%). The proportion of psammophilious rodents within the diet was large in settlements where the soil was sand or sandy-loess, and small where the soil was loess or rocky, and the proportion of migratory birds was 3.7 times larger during migratory seasons than during nonmigratory periods. These variations reflect the availability of the different prey types, and suggest a noticeable plasticity and an opportunistic feeding behavior of the Long-eared Owl.

ACKNOWLEDGMENTS

We are grateful to Yoav Motro, Miriam Belmaker, Noam Leader, Mali Tores, Eran Levin, Leonid Friedman, Asaf Tsoar, Boaz Shacham, Igor Gavrilov, and three reviewers for their help, advice, and comments.

Received 15 August 2007. Accepted 24 December 2007.

LITERATURE CITED

ALIVIZATOS, H. AND V. GOUTNER. 1999. Winter diet of the Barn Owl (Tyto alba) and Long-eared Owl (Asio otus) in northeastern Greece: a comparison. Journal of Raptor Research 33:160-163.

ANDREWS, P. 1990. Owls, caves and fossils. Natural History Museum Publications, London, United Kingdom.

BERTOLINO, S., E. GHIBERTI, AND A. PERRONE. 2001. Feeding ecology of the Long-eared Owl (Asio otus) in northern Italy: is it a dietary specialist? Canadian Journal of Zoology 79:2192-2198.

BOSAKOWSKI, T. AND D. G. SMITH. 1991. Comparative diets of sympatric nesting raptors in the eastern deciduous forest biome. Canadian Journal of Zoology 70:984-992.

CAPIZZI, D., L. CAROLI, AND P. VARUZZA. 1998. Feeding habits of sympatric Long-eared Owl Asio otus, Tawny Owl Strix aluco and Barn Owl Tyto alba in a Mediterranean coastal woodland. Acta Ornithologica 33:85-92.

CHIBA, A., M. ONOJIMA, AND T. KINOSHITA. 2005. Prey of the Long-eared Owl (Asio otus) in the suburbs of Niijata City, central Japan, as revealed by pellet analysis. Ornithological Science 4:169-172.

CRAMP, S. AND K. E. L. SIMMONS. 1985. Handbook of the birds of Europe, the Middle East, and North Africa. The birds of the Western Palearctic. Volume IV. Terns to woodpeckers. Oxford University Press, Oxford, United Kingdom.

GLUE, D. E. AND G. J. HAMMOND. 1974. Feeding ecology of the Long-eared Owl in Britain and Ireland. British Birds 67:361-369.

JAFFE, S. 1986. Climate of Israel. Pages 79-94 in The zoogeography of Israel (Y. Yom-Tov and E. Tchernov, Editors). W. Junk, Dordrecht, The Netherlands.

MARKS, J. S. 1984. Feeding ecology of breeding Long-eared Owls in southwestern Idaho. Canadian Journal of Zoology 62:1528-1533.

MARTI, C. D. 1974. Feeding ecology of four sympatric owls. Condor 76:45-61.

MARTI, C. D. 1976. A review of prey selection by the Long-eared Owl. Condor 78:331-336.

NAVARRO, J., J. A. SANCHEZ-ZAPATA, M. CARRETE, F. BOTELLA, A. GAVRILOV, S. SKLYARENOKO, J. A. DONAZAR, O. CEBALLOS, AND F. HIRALDO. 2003. Diet of three sympatric owls in steppe habitats of eastern Kazakhstan. Journal of Raptor Research 37:256-258.

NILSSON, I. N. 1981. Seasonal changes in food of the Long-eared Owl in southern Sweden. Ornis Scandinavica 12:216-223.

PAZ, U. 1987. Birds of Israel. S. Steimatzky, Tel Aviv, Israel.

RACZYNSKI, J. AND A. L. RUPRECHT. 1974. The effect of digestion on the osteological composition of owl pellets. Acta Ornithologica 14:25-37.

SHIRIHAI, H. 1996. The birds of Israel. Academic Press, London, United Kingdom.

TOME, D. 1994. Diet composition of the Long-eared Owl in central Slovenia: seasonal variation in prey use. Journal of Raptor Research 28:253-258.

VILLAGE, A. 1981. The diet and breeding of Longeared Owls in relation to vole numbers. Bird Study 28:215-224.

Zohar Leader, (1) Yoram Yom-Tov, (2,3) and Uzi Motro (1)

(1) Department of Evolution, Systematics and Ecology, The Hebrew University of Jerusalem, Jerusalem 91904, Israel.

(2) Department of Zoology, Tel Aviv University, Tel Aviv 69978, Israel.

(3) Corresponding author; e-mail: yomtov@post.tau.ac.il

TABLE 1. Genera and species identified in Long-eared
Owl pellets. (n = number of individuals in each
taxonomic class; % = percent of these individuals
among the total number of individuals in the pellets).

Mammalia (n = 3,327; 71.3%)
  Rodentia
    Acomys spp.
    Gerbillus spp.
    Jaculus jaculus
    Meriones spp.
    Microtus guentheri
    Mus musculus
    Rattus rattus
    Spalax ehrenbergi
  Soricomorpha
    Crocidura spp.
  Chiroptera
    Otonycteri hemprichii
    Pipistrellus spp.
Aves (n = 1,236; 26.5%)
  Passeriformes
    Alaudidae sp.
    Carduelis carduelis
    C. chloris
    Cercomela melanura
    Emberiza spp.
    Fringilla coelebs
    Hippolais spp.
    Hirundo spp.
    Lanius spp.
    Motacilla spp.
    Nectarinia osea
    Oenanthe spp.
    Passer domesticus
  Coraciiformes
    Merops apiaster
  Galliformes
    Coturnix coturnix
Reptilia (n = 3; 0.1%)
  Squamata
    Gekkonidae sp.
    Scincidae sp.
Insecta (n = 93; 2.0%)
  Blattodea
    Blattidae sp.
  Coleoptera
    Scarabaeus spp.
    Curculionidae sp.
  Hymenoptera
    Dorylus fulvus
  Mantodea sp.
  Orthoptera
    Gryllidae sp.
    Gryllotalpa gryllotalpa
  Arachnida (n = 2; 0.0%)
    Solifugae sp.
Unidentified (n = 7; 0.1%)