Tidying up the cluttered understorey: Foraging strategy mediates bat activity responses to invasive rhododendron

Biological invasions pose a significant threat to biodiversity, yet the severity and direction of impacts differs between taxa. The non-native evergreen shrub Rhododendron ponticum has colonized large areas of woodland understorey in the UK and is currently undergoing widespread clearance due to its negative impacts on native flora and as a potential host reservoir for harmful pathogens. Presence of rhododendron significantly increases the amount of structural clutter in the understorey and, hence, modifies microhabitat structure, which might have an impact on bat foraging activity. Here, for the first time, we investigated the effects of rhododendron on activity of British bat species by performing autonomous acoustic surveys at twelve sites in Richmond Park, London, where rhododendron was either present, absent or recently removed. The effects of other microhabitat characteristics likely to influence bat activity (e.g. tree diversity, canopy cover, distance to water and the presence/absence of deer) were also considered. We predicted that the increased understorey clutter present in areas invaded by rhododendron would reduce the activity of larger, less agile bats Nyctalus noctula, Eptesicus serotinus and Nyctalus leisleri (NSL), whereas the activity of bat species able to forage within, or at the edges of, clutter (e.g. Myotis spp. and Pipistrellus spp.) would not be negatively affected. While our results indicated a significant reduction in the activity of larger, open space foragers (NSL) in rhododendron-invaded areas, deer presence had a stronger negative effect than woody understory density and canopy cover on these bat species. Activity of other bats recorded was not negatively impacted by the presence of rhododendron. However, sites where rhododendron had been removed demonstrated higher bat activity levels than sites where rhododendron was present and absent with the exception of Pipistrellus pygmaeus, which displayed higher activity levels in the presence of woody understorey. NSL and P. pipistrellus also demonstrated more frequent commuting/roosting activity in sites without rhododendron, although P. pygmaeus and Myotis spp. appeared to be regularly utilising at least some rhododendron sites for this purpose. To mitigate potential negative effects of rhododendron removal on some bat species, we therefore suggest restriction of deer access to sites to enable recovery of native woody understorey.