Could the predation risk hypothesis explain large-scale spatial sexual segregation in fallow deer (Dama dama)?

Large-scale spatial segregation was assessed by means of radio-telemetry in 48 fallow deer studied for 4 years. Three hypotheses have traditionally been used to explain sexual segregation: (1) the predation risk hypothesis, (2) the forage selection hypothesis, (3) the activity budget hypothesis. The first of these seems to be a valid explanation of large-scale segregation in fallow deer at San Rossore, where the use by the two sexes of areas characterized by intense anthropic disturbance, during the day, was compared with other areas not affected by human pressure. Males showed a high use of disturbed areas, both during the day and the night, with the exception of the rutting period, when they reached more remote areas to mate. Females frequented disturbed areas only during the night, with the exception of the birth period, when sexual segregation peaked because females never used these areas, not even during the night. The forage selection hypothesis was invalid on a large scale, considering that no differences between the degree of day and night sexual segregation were to be expected. However, the predation risk hypothesis seems not to be a valid explanation of small-scale sexual segregation, when further subdivisions of disturbed areas are considered, because sexes proved to be segregated also during the night in the area they both used. This emphasizes both the importance of scale in understanding ecological processes, since a combination of many different factors may be responsible for the evolution of sexual segregation in ungulates, and the importance of human pressure in influencing deer behaviour.