J. Rasomaa, b and S.M. Goodmana, c,
, 
, Received 2 June 2006;
revised 21 September 2006;
accepted 12 October 2006.
Available online 28 November 2006.
Abstract
The food habits of the Barn Owl (Tyto alba) are relatively well documented across its Madagascar range based on pellet contents, with the exception of the arid zone of the extreme southwest. New data are presented herein to fill this void. The diet of this owl consists largely of introduced rodents and there are some marked seasonal differences in the other types of prey taken. Evidence is presented to support the hypothesis that this owl has been able to expand its distribution in the wake of human habitat disturbance and the associated trophic proliferation of non-native rodents in disturbed habitats.Keywords: Tyto alba; Prey choice; Introduced small mammals; Range expansion; Madagascar
Article Outline
- 1.
- Introduction
1. Introduction
Tyto alba (Tytonidae: Strigiformes) or Barn Owl is a nocturnal raptor with a broad distribution across much of the temperate, dry and tropical portions of the World (Taylor, 1994). On Madagascar, it is known from different regions, almost exclusively outside of relatively intact natural forest formations and often associated with synanthropic settings, tree plantations or heavily degraded forests (Thorstrom et al., 2003). Analyses of regurgitated pellets of this species have been conducted across a considerable portion of its range on Madagascar. These studies indicate that the vast majority of prey taken is of small mammals, particularly introduced species, and to a much lesser extent birds, amphibians, reptiles and insects (Goodman and Langrand, 1993; Goodman et al., 1993; Rasoloarison et al., 1995). Little data is available on this owl's diet in the arid regions of the extreme southwest, a zone of radical fluctuations in climate and with irregular meagre rainfall, as well as a unique and highly endemic biota.Although the 19th-century information on the birds of Madagascar is relatively extensive, Tyto alba was known during this period from a limited number of sites (e.g. Grandidier, 1879), and was notably rare or absent in major towns and villages where they would have been evident and are now common. Using these shifts in distribution and commonality through time as a point of extrapolation, particularly from the eastern humid portion of the island, Goodman and Langrand (1993) presented the hypothesis that Tyto alba has been able to greatly expand their distribution in less than a century. This was largely correlated with human transformation of the natural forests and associated trophic proliferation of introduced rodents in disturbed habitats. At certain eastern sites, particularly those not near native forests, introduced rodents approach 100% of the diet of this owl (Goodman and Langrand, 1993). Given the extreme arid nature of the southwest and very different ecology and prey base than the eastern mesic zone, it is uncertain if the same pattern holds there. Using new data presented here on the dietary regime of this owl in southwestern Madagascar, we are able to fill a void in information on the food habits of this owl and test the range expansion hypothesis of Goodman and Langrand (1993) associated with the proliferation of non-native rodents. Further, data is presented on seasonal variation in the dietary regime of this owl in the southwest.
2. Study area and methods
This study was conducted in an area of gallery forest along the Onilahy River in extreme southwestern Madagascar in a zone with a series of lakes known as the Sept Lacs (23°29′S, 44°04′E, 70 m a.s.l.). This zone is along the limestone Mahafaly Plateau, which rises abruptly from the Onilahy River, and is dominated by xerophytic bush forest along the cliff face and upper plateau (Emmett et al., 2003). In the narrow band along the banks of the river, there is a riparian forest dominated by Tamarindus indica trees, and other vegetation composed of the families Euphorbiaceae and Sapindaceae (Volomanjaka, 2001). The annual precipitation in the Sept Lacs region is 400–500 mm (Battistini, 1964) and the annual average temperature ranges from 20 to 25 °C (Chaperon et al., 1993). This region has two pronounced seasons: the rainy season, which can last up to 4 months, from December to March, and the dry season, normally of at least 8 months, from April to November.The Tyto alba pellets were collected each morning from May 2001 to July 2002 at three different sites in the immediate region of Sept Lacs. The perimeter of the study zones were delineated with continuous GPS mapping, transferred to an aerial photograph, and subsequently the surface area calculated in hectares; we do not have measures of the home range of this owl species. The sites, their surface areas, and number of localities at each where pellets were collected include: Andamilamy (20 ha) with three roost sites, Antombokamboa (28 ha) with two roost sites, and Mandarano (15 ha) with one roost site. All of the six owl roost sites were located at the entry of rock crevices or along cliff faces.
The different prey items recovered from the pellets were identified based on skeletal remains for vertebrates and exoskeleton remains for invertebrates. Dental characters were of considerable utility for specific identification of mammals. The minimum number of individuals (MNI) for each pellet was established by counting the most represented element of any taxon; in cases of bilateral bones, the greatest count of the right or left side was used. Across all identified prey types, the percent total individuals was obtained by dividing the number of individuals per taxonomic group, in most cases species, by the total summed number of individuals across invertebrate and vertebrate taxonomic groups. Data on the mass of prey animals was taken from the literature and in a few cases estimated. The biomass calculations were made by multiplying the MNI of a given taxon by their average mass and then expressing this as a percentage of the total biomass consumed. Results are presented separately by season.
3. Results and discussion
Based on the 415 recovered pellets of Tyto alba (249 during the rainy season and 166 during the dry season), 2800 individual prey items were identified from a variety of vertebrate groups and 225 invertebrates from three different insect groups (Table 1). Mammals were the most commonly recovered prey, by both percent total individuals and biomass, and were predominantly represented by introduced members of the order Rodentia (rodents). Other groups of mammals are also taken, including Afrosoricida (tenrecs), Soricomorpha (shrews), Chiroptera (bats), and Primata (small lemurs). The most common prey item taken by this owl at Sept Lacs was the introduced Mus musculus.Table 1. Dietary regime of Tyto alba in the Sept Lac region of southwestern Madagascar
Data are presented as minimum number of individuals (MNI), and percent total individuals and biomass during the dry and rainy season.
a Data on the mass of different identified prey animals is derived from Carleton (2003) and Carleton and Goodman (2003) for rodents, Peterson et al. (1995) for bats, Rasoloarison et al. (2000) for Microcebus spp. mouse-lemurs and Ravokatra et al. (2003) for birds.
b Introduced to the island.
| Rainy season | Dry season | Massa (g) | |||||
|---|---|---|---|---|---|---|---|
| MNI | % Total individuals | % Total biomass | MNI | % Total individuals | % Total biomass | ||
| Rodentia | |||||||
| Rattus rattusb | 135 | 5.4 | 35.4 | 28 | 5.1 | 32.4 | 109.9 |
| Mus musculusb | 1272 | 51.3 | 34.9 | 178 | 32.5 | 21.6 | 11.5 |
| Eliurus myoxinus | 1 | 0.04 | 0.2 | 0 | — | — | 82 |
| Macrotarsomys bastardi | 1 | 0.04 | 0.6 | 0 | — | — | 24.5 |
| Sub-total | 1409 | 56.8 | 71.1 | 206 | 37.6 | 54.0 | |
| Afrosoricida and Soricomorpha | |||||||
| Suncus madagascariensis | 41 | 1.7 | 0.2 | 8 | 1.5 | 0.1 | 1.7 |
| Geogale aurita | 26 | 1.0 | 0.4 | 3 | 0.5 | 0.2 | 7 |
| Echinops telfairi | 4 | 0.2 | 1.3 | 1 | 0.2 | 1.5 | 140.6 |
| Sub-total | 71 | 2.9 | 1.9 | 12 | 2.2 | 1.8 | |
| Chiroptera | |||||||
| Mormopterus jugularis | 162 | 6.5 | 4.3 | 95 | 17.4 | 11.2 | 11.2 |
| Chaerephon leucogaster | 9 | 0.4 | 0.2 | 2 | 0.4 | 0.2 | 10 |
| Mops leucostigma | 3 | 0.1 | 0.1 | 2 | 0.4 | 0.4 | 19 |
| Otomops madagascariensis | 0 | — | — | 1 | 0.2 | 0.3 | 26 |
| Molossidae indet. | 4 | 0.2 | 0.1 | 5 | 0.9 | 0.9 | 18 |
| Sub-total | 178 | 7.2 | 4.7 | 105 | 19.3 | 13.0 | |
| Primata | |||||||
| Microcebus murinus | 21 | 0.8 | 3.1 | 11 | 2.0 | 7.7 | 62 |
| Microcebus griseorufus | 11 | 0.4 | 1.7 | 4 | 0.7 | 2.9 | 63 |
| Microcebus sp. indet. | 7 | 0.3 | 1.0 | 0 | — | — | 61 |
| Sub-total | 39 | 1.5 | 5.8 | 15 | 2.7 | 10.6 | |
| Reptilia | |||||||
| Zonosaurus sp. | 5 | 0.2 | 0.3 | 2 | 0.4 | 0.0 | 29 |
| Small snake indet. | 0 | — | — | 1 | 0.2 | 0.0 | 20 |
| Sub-total | 5 | 0.2 | 0.3 | 3 | 0.6 | 0.0 | |
| Amphiba | |||||||
| Ptychadena mascareniensis | 566 | 22.8 | 6.6 | 56 | 10.2 | 3.1 | 4.9 |
| Aves | |||||||
| Foudia madagascariensis | 50 | 2.0 | 1.5 | 7 | 1.2 | 1.2 | 16.3 |
| Small-sized passerine | 24 | 1.0 | 0.6 | 2 | 0.4 | 0.2 | est. 10 |
| Medium-sized passerine | 28 | 1.1 | 3.3 | 4 | 0.7 | 2.1 | est. 50 |
| Large-sized passerine | 4 | 0.2 | 1.0 | 0 | — | — | est. 100 |
| Leptopterus viridis | 4 | 0.2 | 0.4 | 0 | — | — | 47 |
| Dicruris forficatus | 3 | 0.1 | 0.3 | 0 | — | — | 47.1 |
| Hypsipetes madagascariensis | 1 | 0.04 | 0.1 | 1 | 0.2 | 0.3 | 24.7 |
| Turnix nigricollis | 3 | 0.1 | 0.4 | 1 | 0.2 | 0.6 | 61.3 |
| Thamnornis chloropetoides | 0 | — | — | 1 | 0.2 | 0.2 | 14.3 |
| Newtonia brunneicauda | 0 | — | — | 1 | 0.2 | 0.1 | 10.1 |
| Ploceus sakalava | 1 | 0.04 | 0.1 | 0 | — | — | 23.8 |
| Sub-total | 118 | 4.8 | 7.7 | 17 | 3.1 | 4.7 | |
| Insecta | |||||||
| Acrididae | 55 | 2.2 | 0.7 | 114 | 20.8 | 6 | est. 5 |
| Blattidae | 2 | 0.1 | 0.02 | 1 | 0.2 | 0.1 | est. 5 |
| Coleoptera | 35 | 1.4 | 0.4 | 18 | 3.3 | 0.9 | est. 5 |
| Sub-total | 92 | 3.7 | 1.1 | 133 | 24.3 | 7.0 | |
| Total | 2478 | 547 | |||||
a Data on the mass of different identified prey animals is derived from Carleton (2003) and Carleton and Goodman (2003) for rodents, Peterson et al. (1995) for bats, Rasoloarison et al. (2000) for Microcebus spp. mouse-lemurs and Ravokatra et al. (2003) for birds.
b Introduced to the island.
Although the number of pellets recovered during the two different seasons is not equivalent, there are apparent differences in the proportion of vertebrate prey taken by Tyto alba. During the rainy season, the percent number of individuals and biomass for the five most commonly taken prey were, respectively, Mus musculus (51.3%; 34.9%), Ptychadena mascareniensis (22.8%; 6.6%), Mormopterus jugularis (6.5%; 4.3%), Rattus rattus (5.4%; 35.4%) and Foudia madagascariensis (2.0%; 1.5%), whereas, the percent number of individuals and biomass for the five most commonly taken prey during the dry season were, respectively, Mus musculus (32.5%; 21.6%), Mormopterus jugularis (17.4%; 11.2%), Ptychadena mascareniensis (10.2%; 3.1%), Rattus rattus (5.1%; 32.4%) and Microcebus spp. (2.7%; 10.6%). Thus, the principal prey types taken during the two seasons remain largely similar, although small birds (F. madagascariensis) are replaced by mouse lemurs (Microcebus spp.) and some aspects of percent representation change. These differences in vertebrate prey, particularly small mammals, show direct parallels to another site in the arid region of the island, where pellets of Tyto alba were collected on a regular basis during the course of one complete year (Goodman et al., 1993). Hence, based on available data, it would appear that annual seasonal fluctuations across the arid portions of Madagascar have an influence on prey population cycling, which has a relationship with their representation in the diet of this owl.
Among the invertebrates, three different groups were identified from the pellets (Table 1) and in some cases represented an important percentage of the total individuals taken, but largely negligible with regards to biomass. Although the number of pellets recovered during the dry season was notably less than during the rainy season, the calculated MNI for arthropods during the dry season was 133 and during the rainy season 192. Further, during the dry season the percentage of total individuals of Acrididae (grasshoppers) reached 20.8% of the total diet.
The relatively high percentage of Ptychadena mascareniensis frogs in the pellet remains during the rainy season and their rarity during the dry season is related to the reproductive biology of this animal. Ptychadena are active during the day and night, are seasonally common in marshlands and along stagnant and slow-moving waters, and vocalize intensively at the start of the rainy season (Vences et al., 2003). Another important seasonal difference is in the number of diurnally active birds represented in the samples, with a MNI of 118 during the rainy season and 17 during the dry season. However, when evaluated by percentage of total individuals and biomass (respectively) there is not a marked seasonal difference—rainy season (4.8%; 7.7%) and dry season (3.1%; 4.7%). It is presumed that diurnal active birds are taken from night roosts, rather than Tyto hunting them during the day.
Among the bats, which have the same activity period as Tyto alba, there was a notable seasonal shift in the relative importance of this group in its diet. During the rainy season, bats represented 7.2% of the total individuals and 4.7% of the biomass, while during the dry season these values are 19.3% and 13.0% (respectively). This seasonal increase in a prey type that has a relatively small body mass may be associated with compensation during the dry season for a number of other prey types that are not particularly active (e.g. frogs) or at the low end of population cycling (e.g. rodents). The same aspect seems to hold for nocturnal species of the genus Microcebus, which are proportionately more commonly taken during the dry season than the rainy season. This would imply that these primates are not in a constant state of torpor during the dry season and occasionally venture away from their tree holes and nests.
The two species of introduced rodents, Mus musculus and Rattus rattus, comprise the most important prey items of this owl, approaching or surpassing 50% of the diet as measured by total individuals and biomass during the two seasons. In comparison, native rodents of the genera Eliurus and Macrotarsomys, make up a negligible proportion of the diet during the rainy season and are not represented in the pellet remains during the dry season. This pattern is in parallel to Tyto alba pellet samples collected in the east within close proximity of native forest, such as Andasibe, where non-native mammals made up 64.7% of total individuals and 83.7% of total biomass and native rodents 4.0% and 4.2%, respectively (Goodman and Langrand, 1993).
The region of Sept Lacs with its limestone cliff faces and rock fissures contains excellent sites for Tyto alba roosts. The zone has recently experienced considerable levels of habitat degradation and deforestation (Emmett et al., 2003; Volomanjaka, 2001). These vicissitudes, combined with the presence of local permanent water sources, provide excellent habitat for introduced rodents. The dietary regime of this owl at Sept Lacs comprises a large percentage of non-native rodent species and it is presumed that local human habitat modification has provided the means for these rodents to colonize this zone and allow the expansion of local Tyto populations. Thus, there is a direct parallel to areas of the east, and the general hypothesis of Goodman and Langrand (1993) is supported for the arid areas of the southwest. The main difference is that in the east the larger body introduced rodent R. rattus is taken with about the same frequency as the smaller bodied M. musculus, while in the extreme southwest Mus is distinctly more common. Whether in degraded and open forested habitats Tyto alba is actually choosing introduced rodents over native small mammals or these differences are proportional to prey availability needs to be addressed with further field studies.
References
Battistini, 1964 R. Battistini, Etude géomorphologique de l’extrême Sud de Madagascar, tome I: Le relief de l’intérieure, Cujas, Toulouse (1964).Carleton, 2003 M.D. Carleton, Eliurus, tufted tailed rats. In: S.M. Goodman and J.P. Benstead, Editors, The Natural History of Madagascar, The University of Chicago Press, Chicago (2003), pp. 1373–1380.
Carleton and Goodman, 2003 M.D. Carleton and S.M. Goodman, Macrotarsomys, big footed mice. In: S.M. Goodman and J.P. Benstead, Editors, The Natural History of Madagascar, The University of Chicago Press, Chicago (2003), pp. 1386–1388.
Chaperon et al., 1993 P. Chaperon, J. Danloux and L. Ferry, Fleuves et rivières de Madagascar, ORSTOM, Paris (1993).
Emmett et al., 2003 Emmett, D.A., Fanning, E., Olsson, A. (Eds.), 2003. The proposed Parc Regional de Belomotse: biodiversity survey and conservation evaluation. Frontier Madagascar Environmental Research Report 6.
Goodman and Langrand, 1993 S.M. Goodman and O. Langrand, Food habits of the Barn Owl (Tyto alba) and the Madagascar Long-eared Owl (Asio madagascariensis) on Madagascar: adaptation to a changing environment, Annales Musée Royal de l’Afrique Centrale (Zoologie) 268 (1993), pp. 147–153.
Goodman et al., 1993 S.M. Goodman, O. Langrand and C.J. Raxworthy, The food habits of the Barn Owl Tyto alba at three sites on Madagascar, Ostrich 64 (1993), pp. 160–171. Full Text via CrossRef | View Record in Scopus | Cited By in Scopus (21)
Grandidier, 1879 A. Grandidier, Histoire physique, naturelle et politique de Madagascar: Histoire naturelle des oiseaux. Volume I—texte, Imprimerie Nationale, Paris (1879).
Peterson et al., 1995 R.L. Peterson, J.L. Eger and L. Mitchell, Faune de Madagascar, 84, Chiroptères, Muséum national d’Histoire naturelle, Paris (1995).
Rasoloarison et al., 1995 R.M. Rasoloarison, B.P.N. Rasolonandrasana, J.U. Ganzhorn and S.M. Goodman, Predation on vertebrates in the Kirindy Forest, western Madagascar, Ecotropica 1 (1995), pp. 59–65.
Rasoloarison et al., 2000 R. Rasoloarison, S.M. Goodman and J.U. Ganzhorn, Taxonomic revision of mouse lemurs (Microcebus) in the western portions of Madagascar, International Journal of Primatology 21 (2000), pp. 963–1019.
Ravokatra et al., 2003 M. Ravokatra, L. Wilmé and S.M. Goodman, Bird weights. In: S.M. Goodman and J.P. Benstead, Editors, The Natural History of Madagascar, The University of Chicago Press, Chicago (2003), pp. 1059–1063.
Taylor, 1994 I. Taylor, Barn Owls: Predator–Prey Relationships and Conservation, Cambridge University Press, Cambridge (1994).
Thorstrom et al., 2003 R. Thorstrom, L.A. Rene de Roland and R.T. Watson, Falconiformes and Strigiformes. In: S.M. Goodman and J.P. Benstead, Editors, The Natural History of Madagascar, The University of Chicago Press, Chicago (2003), pp. 1080–1085.
Vences et al., 2003 M. Vences, F. Glaw and A.P. Raselimanana, Ptychadena, Mascarene Grass Frog. In: S.M. Goodman and J.P. Benstead, Editors, The Natural History of Madagascar, The University of Chicago Press, Chicago (2003), pp. 927–928.
Volomanjaka, 2001 Volomanjaka, N.L., 2001. Etude phytosociologique du plateau calcaire Mahafaly, cas d’Antainosy (Ambohimahavelona). Mémoire de D.E.A. en Biodiversité et Environnement, Option Biologie Végétale, Université de Toliary.
Corresponding author. WWF, BP 738, Antananarivo (101), Madagascar. Tel.: +261 20 22 34885; fax: +261 20 22 34888. 
