The caves of the Yorkshire Dales National Park are potentially an important assemblage of underground bat sites, but prior to this study nothing was known about the ecology of cave use. Between 2002 and 2005,53 cave entrances were logged for swarming activity using automated echolocation call loggers. A subset was chosen for continuous monitoring outside the hibernation period and another for winter logging. Logging was followed up by trapping at caves with significant autumn bat activity. All bats caught were ringed, identified to species and sexed. Biopsies (3 mm diameter) were taken from each wing for genetic analysis.
A large proportion of the caves surveyed were used by bats, but activity was concentrated in a small number of hotspots. Peak activity occurred during August and September, but significant activity was recorded in all months between March and October. All five swarming species were caught at most sites: Myotis brandtii, M. daubentonii, M. mystacinus, M. nattereri and Plecotus auritus. M. nattereri was the most abundant species late in September, but earlier other species made up a significant proportion of those caught. The sex ratio of swarming bats was heavily skewed towards males.
Swarming activity was positively correlated with the degree of cave chamber development, cave length and shelter at the entrance, and negatively correlated with the amount of water the cave carried. Entrance size, altitude and the degree of connectivity between cave and summer habitat were not correlated with activity. In a multiple regression model, chamber development and hydrology were the best predictors of activity. Entrance orientation and shelter contributed towards the model but were less important. The regression model could explain 55% of the variance in bat activity.
Genetic diversity of Myotis nattereri, based on an analysis of six microsatellite loci, was high in swarming populations suggesting high gene flow and large effective populations. Most of the genetic variation was explained by variation among individuals within populations, but some was due to differences between swarming areas and by differences between populations within swarming areas. Isolation by distance was found between sites in six swarming areas between Cumbria and Sussex, but not between sites in the Yorkshire Dales. The differentiation observed was more logically explained on the basis of local geography and probable catchment areas.
Preliminary experiments suggest that Time Domain Signal Coding has potential as a sound analysis system for real time bat call identification.