Population densities and habitat associations of introduced muntjac Muntiacus reevesi and native roe deer Capreolus capreolus in a lowland pine forest

Introduced Chinese muntjac Muntiacus reevesi are increasing in numbers and range within England and are colonising habitats occupied by native roe deer. Effective management of impacts requires knowledge of deer numbers and habitat associations. Few published estimates of muntjac numbers exist and habitat associations of this species are poorly known. Numbers of both muntjac and roe deer were estimated in 185 km2 of lowland pine forest in eastern England, by standing crop pellet-group counts conducted in 217 plantation stands in February-March 2002. Estimated population density of muntjac (mean ± 95% CI: 64 ± 13 km−2) exceeded that of roe deer (28 ± 6 km−2), giving forest population estimates of 11,900 ± 2370 and 5200 ± 1070, respectively. Sensitivity analysis incorporating different published values of defaecation rate of each species, gave mean density estimates ranging from 59 to 76 muntjac km−2 and 25 to 33 roe deer km−2. The introduced species outnumber native roe deer two- to three-fold in this landscape. For both species, estimated numbers were approximately double subjective population assessments from deer managers, based on frequency of sightings. Both species attained high densities in pre-thicket and mature stands and used open re-stocked stands (aged 0-4 years) less than other growth stages. Muntjac also occurred at substantial densities in thicket and pole stages. Roe deer aggregated on bramble across all habitats, while muntjac were positively associated with bramble in older stands (≥25 years). The two species showed substantial overlap in their use of growth stages (Pianka's index of overlap, O = 0. 93) and individual stands (O = 0.63). This supports proposals that inter-specific competition may occur between these two species. At a landscape scale, muntjac density was higher in forest blocks with a greater ratio of open habitat perimeter to forest area; roe deer did not show this relationship. For both species, pellet-groups disappeared much faster than published estimates of decay rates previously used in designing pellet accumulation studies.