Etude de l'écologie des micromammifères à partir des pelotes de réjection de Tyto alba (Scopoli, 1769). Application au Bas-Chablais (Haute-Savoie, France)

Abstract

A method of investigating the ecology of small mammals in open habitats is proposed, based on the author’s own investigations carried out in Bas Chablais, Haute-Savoie, France. Pellets of Barn Owl (Tyto alba) were collected at 15 sites within a defined study area. The remains of 14,056 small mammals found in these pellets were identified to the species level. The environmental descriptors of each of the 15 sites were identified, and a correlation matrix between the percentage occurence of each prey category and the environmental variables was calculated. The occurence of each prey species was subjected to a correspondence analysis, and the environmental parameters were projected onto the axes thus produced. The overall results confirm the view that Barn Owl captures its prey at random in open environments ; the analysis thus suggests that the pellets provide a goad random sample of the prey populations. The ecology of the 16 species of small mammals captured by the Barn Owl is discussed. For the species living in open environments the habitat is defined by the first three factors of the correspondence analysis, i.e. rainfall, area under cereal cultivation, and grassland type (Mesobrometum or Arrhenatheretum). A strong positive relationship is suggested between the percentage occurence of Apodemus sylvaticus and the total area under cereal cultivation. Interspecific competition is shown by the opposite position of species on the axis of the correspondence analysis. Such is the case for Sorex coronatus versus Crocidura russula, and Microtus arvalis versus Arvicola terrestris. The results also provide information on the movements and dispersal of several woodland species : Clethrionomys glareolus, Sorex minutus, Neomys anomalus and N. fodiens mostly move within the understorey ; N. fodiens would appear to have higher power of dispersal than N. anomalus. Such a method might be useful in other regions, and provide data on regional variations in habitat preferences of small mammals.