Nest box monitoring data

Full description

Data was collected from Sept 2014 to June 2016. It includes records of animals recorded in nest boxes across southeast Queensland and northern New South Wales. The study area included the central coast region of New South Wales (NSW) and SE Queensland (QLD; Figure 1). These regions were selected because they have comparative patterns of urbanization and common myna invasion histories [24]. We defined 'source' sites in each region as long-invaded areas where mynas were reported before 1970 (here termed "source" sites). Recently-invaded 'front' regions were areas in which mynas have been reportedly established (i.e. breeding) since 1990 (termed "front" sites; Figure 1). Source regions were generally located in larger cities and front regions in smaller cit-ies/towns (Table 1), except in Gatton in Queensland, a more rural source site. The com-mon myna was brought to southeast QLD and Newcastle in NSW via independent in-troduction events [24]. As a result, the birds have expanded northwards and southwards along the east coast in QLD and westward towards the inland in NSW [15]. In NSW, we selected source sites in the greater Newcastle area, specifically in the suburbs of New Lambton and Glendale (Table 1). In addition, we selected front sites within the larger Hunter region approximately 100 km north-west of Newcastle, in the towns of Gloucester and Krambach (Table 1). In Queensland, we selected source sites at the University of Queensland Gatton campus near Lawes and in the Brisbane suburbs of Oxley and New Farm. Front sites were approximately 100 km north of Brisbane in the rural towns of Dayboro and Landsborough (Figure 1). 2.2. Nest box monitoring Within each site, we set up 24 nest boxes across three sub-sites (8 boxes in each) selected to sample variation in urban habitat characteristics. The three sub-sites were all within one km of each other and included areas with remnant vegetation, open parks, and an urbanized area (ex., schools, suburban streets, caravan parks). The nest box dimensions were constructed out of plywood and had dimensions of 400 mm (height) x 170 mm (width) x 170 mm (depth). Boxes had an entrance hole that was 65 mm in diameter. We chose these boxes as they have been shown to be suitable for the common myna and several native species [16]. Nest boxes were placed at heights between three and five meters and at least 10 meters from the closest box. We monitored the nesting boxes using a Signet fiber optic inspection camera mounted on a pole. We would insert the camera into the entrance hole in the nest box and take a photo or video if there was an animal in the box or if there was evidence of nesting. We had animal ethics for all fieldwork activities approved through the University of Queensland (Re-search and Integrity office, Animal Welfare Unit 454/13) and the University of Newcastle (Animal Research Authority A-2014-424). Boxes were set up in September 2014 and monitored until June 2016. We checked boxes weekly during the breeding season of the common myna and monthly during the non-breeding season. Boxes were not checked during heavy rain to minimize disturb-ance to potential box occupants; heavy rain occurred on nine QLD visits and seven NSW visits. Monitoring was done by a team of research students and volunteers familiar with the species likely to be in the boxes. Mammals were identified to species in QLD, but not in NSW. Both common brushtail possums and common ringtail possums (Pseudocheirus peregrinus) occur in both regions. If an animal or nest was observed in the box, that box was considered occupied or ‘in use’ during that week. A nesting attempt was recorded if eggs were found in a box. We identified common myna nests by their use of nesting material (including feathers and plastic) and light blue eggs. To identify other nests, we recorded egg color and shape, we watched nests until the parent birds returned, or emerging adult feathers on chicks allowed identification of the species [25]. For each nesting attempt, after eggs hatched, the general age of chicks was recorded (i.e. not feathered, feather pins, fully feathered, and fledgling). A nesting attempt was considered a failure if eggs disappeared or chicks disappeared before they had their adult feathers. We considered a nesting attempt successful if chicks developed adult feathers and left the box [26]. Nesting success per species was calculated as the sum of successful nesting attempts over the two years. Nesting failure per species was the sum of unsuccessful nesting attempts over the two years. 2.3. Environmental characteristics To identify vegetation diversity and structure around each nest box to a 15-m ra-dius, local habitat characteristics were measured using circular surveys. Within each plot, we counted individual trees, measured the diameter at breast height, and identified na-tive trees to a genus where possible. Additionally, the per cent ground cover of shrub, turf, and sealed ground (i.e., asphalt and concrete) was estimated from the centre of the plot. Finally, we counted the number of potential natural hollows in trees within each plot. Due to the difficulty in assessing whether a cavity in a tree was a suitable nesting hollow, hollow occurrence at each site provided only a general measure of potential nesting site availability [27]. In order to measure variation in the remaining natural habitat area around study sites, we used remotely sensed satellite data of normalized difference vegetation index (NDVI) sensu Bino et al. (2010). We used NDVI data from NASA’s LandSat satellite, which produces images with a 30 x 30m pixel resolution. NDVI images for all sites were downloaded for the years 2009-2014. We used only images that had less than 10% cloud cover, resulting in 14 cloud-free images included in the analysis. We averaged NDVI values for pixels within a 100 m2 area centered on the center of each sub-site to create a single NDVI value for each sub-site.