Acrocephalus 21 (98-99): 3 – 7, 2000 Winter diet of the Long-eared Owl Asio otus in Slovenia Zimska prehrana male uharice Asio otus v Sloveniji Davorin Tome National Institute of Biology, Ve~na pot 111, SI-1000 Ljubljana, e-mail: davorin.tome@uni-lj.si Winter diet of the Long-eared Owl Asio otus was investigated at 15 localities across Slovenia, belonging to seven geographical regions (Fig. 1). Of 3,712 prey items found in pellets, small mammals represented more than 98%, with voles Arvicolinae being the most important (Tab. 1). The proportions of Common Vole Microtus arvalis were the only with a significant negative cor-relation with food niche-breadth (Fig 2.). Common Vole was therefore the main prey for the owl. The most important alternate prey species were: Field Vole Microtus agrestis, mice of the genus Apodemus, Common-pine Vole Microtus subterraneus, and Water Vole Arvicola terrestris. Which of them occurred in higher percentages in the diet was dependent on their geographi-cal distribution and ecological requirements. Field Voles, for example, were present with more than 10% of all prey items only at Ljubljansko barje and Ribni{ko podolje. This was probably due to the high ground-water level and regular floods in these two regions, which gain some ecological advantages to Field Voles against other small mammals. In the areas with high degree of field cover, on the other hand, the community of small mammals is often dominated by mice. In such areas (Bela krajina, Dravsko podolje, Mursko podolje), the most important alternate prey species for the Long-eared Owl was Wood Mouse Apodemus sylvaticus (Fig. 3). Key words: Long-eared Owl, Asio otus, winter diet, prey use Klju~ne besede: mala uharica, Asio otus, zimska prehrana, izbor plena 1. Introduction Most of the studies on the Long-eared Owl Asio otus in Slovenia deal with its diet (Kry{tufek 1980, Tome 1991, Tome 1994, [orgo & Jan`ekovi~ 1996, Kata-lini~ 1997). These contributions, however, are geo-graphically very narrowly limited only to certain parts of the country. So far, there was no attempt to make a survey of the owls diet across the entire area of Slovenia. The aim of this article is to investigate the Long-eared Owl’s winter diet in different geographi-cal regions of the country, as well as to establish which are the main and which alternate prey species for the owl. 2. Materials and methods Pellets were collected in different parts of Slovenia between 1986 and 1991. Some of the samples were obtained from my associates. For this study only sam-ples found between November and April were taken into consideration. For the presentation of the Long-eared Owl’s diet, the samples were combined accord-ing to the existing mezoregions (hereinafter referred to as “the regions”; Perko & Oro`en–Adami~ 1998). All prey species reaching 10% in the diet were con-sidered as important for the Long-eared Owl. During the more detailed analysis of the feeding habits, the samples were not combined, and only those with more than 20 prey items were taken into considera-tion. For this analysis the data from Mursko polje (Katalini~ 1997) were added to my personal results. Food niche breadth (FNB) was estimated according to Levins (1968). The optimal diet theory (Pyke 1984) foresees that the food niche widens when the population of the main prey species decreases, and vice versa, which means that FNB correlates negative with the proportion of the main prey species in the diet. 3 D. Tome: Winter diet of the Long-eared Owl Asio otus in Slovenia Small mammals were identified with the determination key of Kry{tufek (1985), while birds and insects were not identified to lower taxonomic units. Biomasses were calculated from average masses of small mammals and birds obtained from individual skull (humerus in birds) length (Tome in print) or taken from literature (Tome 1991). The distribution of small mammals in Slovenia is derived from Kry{tufek (1991). The proportions of field areas in the open habitat were estimated in a circle of 100 m around the place where pellets were found: low proportion denotes that there were no or almost no fields in its vicinity, intermediate proportion indicates that there were fields present, while high proportion denotes that at least half of the area was covered with fields. For the pellets from Dravsko polje and Mursko polje that were not collected by me, the proportion of fields were estimated on the basis of statistical values for separate mezoregions (Perko & Oro`en–Adami~ 1998). Significance of differences in the proportion of mice in the diet with regard to the proportion of fields in a certain surrounding were calculated using ANOVA. Prior to the calculations, the data were arc-sin transformed. Figure 1: Study area (location and date of pellet collection in brackets) Slika 1: Obmo~je raziskave (v oklepaju lokacija in datum pobiranja izbljuvkov): A=Piv{ko podolje (Koritnice – 7.2.1991; Kne`ak – 17.3.1991); B=Notranjsko podolje (Vrhnika pri Starem trgu – 25.2.1990; 17.3.1991); C=Ljubljansko barje (Mah, Kozlarjeva go{~a, ^rna vas, Ljubljana, Bevke, Gri~ica; vsa leta od 1986 do 1991); D=Savska ravan (Predoslje – 30.1.1986; Hrastje – 15.11.1990; 25.12.1990; 1.3.1991); E=Bela krajina (Krasinec – 7.4.1991); F=Dravska ravan (Bukovci – 2.2.1986; 21.2.1987; 26.2.1989; Hajdo{e – 23.4.1989); G=Murska ravan (Katalini~ 1997); H=Ribni{ko podolje (Dolenja vas – 8.4.1989; 10.3.1990; 24.3.1991) 4 4- x 3.5- x 3 x g 2.5 - x "- x "° 1.5 - x 1- * 0.5- o---------------1---------------1---------------1---------------1---------------r 0 20 40 60 80 100 % Microtus arvalis Figure 2: Change of food niche breadth (FNB) according to the proportions of Common Vole Microtus arvalis in the diet of the Long-eared Owl (r=-0.961; p<0.001; N=8) Slika 2: Spreminjanje {irine prehranjevalne ni{e ([PN) glede na dele` poljske voluharice Microtus arvalis v prehrani male uharice (r=-0,961; p<0,001; N=8) 3. Results Pellets were collected at 15 localities in seven different geographical regions (Fig. 1). 3,712 prey units belonging to 17 species of small mammals and to unidentified number of bird species were isolated from them. The most important prey species was Common Vole Microtus arvalis with 51% by number, followed by Field Vole Microtus agrestis with 17% and mice of the genus Apodemus with 13%, respectively. The proportion of other species was lower than 10%. The proportions by mass did not essentially differ from the proportions by number. Fifty-five birds represented a little more than 1% in the diet in view of the number of prey units as well as biomass (Tab. 1). Common Vole was the most important prey species in the Long-eared Owl’s diet in all seven regions. The highest share was in the pellets from the localities in Savska ravan and Piv{ko podolje, the lowest at Ljubljansko barje and Bela krajina. Mice played an important part in the owl’s diet in four regions, Common-pine Vole Microtus subterraneus and Field Vole in two, and Water Vole Arvicola terrestris in a single region. In two regions, Common Voles were the only important prey species for the Long-eared Owl (Tab. 1). The most diverse diet was recorded at Ljubljansko barje and Dravska ravan, the least diverse at Savska Acrocephalus 21 (98-99): 3 – 7, 2000 ravan and Piv{ko podolje (Tab. 1). FNB correlates negative only with changes in the proportion of the Common Vole (r=-0.962; p<0.001; N=8; Fig. 2). The proportion of Wood Mouse Apodemus sylvaticus in the pellets changed significantly with the proportion of fields around the locations were pellets were found (F=10.08; p=0.002; Fig. 3). 4. Discussion The Long-eared Owl is specialized in hunting small mammals in open countryside, among which it chooses opportunistically according to their occur-rence (Tome 1991). A large proportion of small mam-mals in the diet, in the first place voles Arvicolinae, is thus not surprising and in accordance with the findings of other authors (Kry{tufek 1980, Tome 1994, [orgo & Jan`ekovi~ 1996, Katalini~ 1997). Due to the five to fifty times greater sample from Ljubljansko barje in comparison with other regions, two to three times more prey species have been dis-covered. This fact, however, has no essential effect on the FNB value of this location. The FNB correlates negative only with the proportion of Common Vole in the diet, due to which this species was regarded as the only main prey species for the owl. When Common Voles were less ready avail-able, the owl captured more mice, Field Voles, Common Pine Voles and Water Voles. In different geografical regions these species were the most impor-tant alternate prey, respectively. In fields, Wood Mice are often the dominant small mammal species (Kry{tufek 1991). It is therefore not surprising that their proportion in the Long-eared Owl’s diet depended on the area covered by fields (Fig. 3). The main difference between Field Vole and Common Vole is that the latter spends most of its life underground, where it also builds its nest, while the first makes its nest and connecting tracks above the ground as well, due to which it is less sensitive to high groundwater level (Kry{tufek 1991). Height of the groundwater level together with flood frequency is thus an important ecological factor, which deter-mines the population density of both species and their proportion in the Long-eared Owl’s diet. The most frequently flooded areas dealt with in this study are Ljubljansko barje and the surroundings of Dolenja vas in Ribni{ko podolje, and it was only in these areas where Field Voles were found to be an important alternate prey species. On the basis of the collected data I was unable to establish the characteristic ecological entities, due to which Common Pine Vole and Water Vole were an 35 (O 30 3 O ati 25