Stream use by the large-footed myotis myotis macropus in relation to environmental variables in northern New South Wales

The large-footed myotis (Myotis macropus) feeds by trawling aquatic invertebrates and small fish from water surfaces, but little is known about what types of streams are used. We investigated habitat use by M. macropus in northern New South Wales (NSW) by analysing data-base records, conducting field surveys and assessing environmental variables on or near waterways. A GIS comparison of 222 data-base records of M. macropus with broad environmental patterns indicated that most records were near large and permanent waterways at low elevations, in flat or undulating terrain, usually surrounded by vegetation. Three different field surveys were carried-out. First, 25 waterways on the NSW mid-north coast were stratified into five size classes and sampled (December 1996-January 1997) using harp traps and ultrasonic detectors. Just two captures and 20 passes were recorded. No M. macropus were detected on either the smallest or largest waterways, although the latter may have been undersampled due to their broad expanse. Records of M. macropus were significantly associated with larger, more permanent waterways (when the largest size class was omitted). Non-metric multi-dimensional scaling of sites, described by their environmental attributes, failed to separate sites on the basis of presence/absence of M. macropus. The second survey (February 1998) targeted 25 sites on waterways within 10 km of a known roost in Kerewong State Forest. Twenty-eight passes were recorded, all confined to just three sites on Upsalls Creek, a large 4th order stream. The third survey (February 2000) investigated the use of adjacent riparian vegetation in the vicinity of the Kerewong roost. At each of 12 sites, one detector was placed on the stream, one 10 m from it and a third 20 m into adjacent vegetation. An average of 11 M. macropus passes per night was recorded over streams (total passes = 273), whereas no passes were recorded in adjacent vegetation. We conclude that M. macropus is sparsely dispersed across streams in forests of north-east NSW and that they are most likely to be recorded on large streams in the lower end of catchments. Management implications for timber production forests include redirecting effort from pre-logging surveys to monitoring the changing status of M. macropus over time, especially at day roosts where local populations are concentrated.