Cassowaries and Camera Traps
by Wren. R. McLean,
A cassowary research project was conducted in the Daintree throughout 2014-15 by ex Rainforest Rescue employee and on-going giver Wren McLean.
Wren has a long history with the Daintree and spent her 21st birthday (some 20 years previous) exploring the region on a gearless second hand bike riding up and down steep washed out 4wd tracks and through flooded rivers during the wet season of 1995, sleeping under only a tarp. Surprisingly she did not spot any cassowaries during this adventure but she is sure, after more recent experiences that they would have spotted her.
Now a post-grad wildlife researcher with Southern Cross University she has had enough one-on-one encounters with these formidable birds understand that they have a sharp awareness of what’s what and who’s where in their dense rainforest territories which are estimated to be approximately 80ha in size. Wren established 31 study sites from the Daintree river in the south to Melissa Creek in the north of the Daintree which included four Rainforest Rescue Nature Refuges purchased and protected forever as part of the Daintree Buyback program.
Each site was surveyed 4 times for any sign of cassowaries (sighting, vocalisation, scat or footprint) with their fruity dung (or scats) being collected for dietary analysis. An an experimental survey technique was also trialled whereby fake fruits were placed in front of camera traps at half of the sites and not at the other half. Rainforest Rescue loaned camera traps for this research, which were an invaluable tool which contributed to the first strategic use of remote camera traps to survey cassowaries and the first use of visual lures (representing large red and blue fruits) used in conjunction with camera traps for any wildlife species.
The ‘camera trap visual lure technique’ was shown to significantly was reduce the amount of time required to get the first photographs of cassowaries at a site from an average 11.4 days without the use of lures to 6.5 days when using lures. This allows for considerable savings in time and costs of surveying this species that is known as notoriously difficult to study due to them being cryptic (solitary, shy and silent) and inhabiting remote dense terrain. Cameras with lures were also found to detect more than twice as many cassowaries as those without. As the cameras were randomly allocated across all sites this may suggest that cameras without lures failed to detect around half the cassowaries potentially present at those sites. Cassowaries were also twice as likely to stop in front of cameras with lures and spent a significantly longer time in front of them which provided a significantly more images of all three angles of the birds both of these factors increased the ability see unique features that allow for individual identification. All camera traps combined provided a pool of 466 visual records from which we were able to identify 45 individual cassowaries of which 64% were adults, 18% sub-adults and 18% chicks. Cameras provided other useful information on cassowary predators (domestic, pig-hunting and wild dogs), feral pigs and photos of three Bennetts tree kangaroos at two different locations.
The Rainforest Recuse reserves surveyed were;
- Rainforest Rescue Nature Reserve at Forest Creek where an adult and two stripy chicks were videoed on one occasion and several scats were found.
- Milky Pine Wildlife Refuge where two courting adults were videoed on numerous occasions along with one independent sub-adult. Scats, footprints and one sighting of this sub-adult were recorded here also.
- Baralba Corridor Nature Refuge is prime cassowary habitat with a 2 single adults and one male with accompanying brown chick photographed. Numerous sightings and vocalisations were recorded along with many scats.
- Kulki anga Nature Reserve, only one scat was found here on the edge of the reserve but the property protects important cassowary habitat that is undoubtedly used throughout the year but a number of resident birds in the area.
Valuable information about the diversity and pattern of habitat use can be derived from dietary analysis; however, no such studies have been undertaken on the Daintree coast. To date, the Daintree cassowaries are mentioned in one published work (Webber and Woodrow, 2004) for their role in dispersing the seed of a rare rainforest tree Ryparosa sp.
The lean season diet was of particular interest to Wren as this often corresponds with the juvenile dispersal period and adds the additional risk of starvation to dispersing chicks. Wren considered the lean season to be from April- July but did not get as much lean season fieldwork opportunity as hoped due the arrival of tropical cyclone Ita which produced 764mm of rain over her first 17 days fieldwork. Many attempts to to establish forest transects during this time were fraught by impenetrable thickets of calamus spp, cyclone damaged vegetation or impassable flooded waterways. A follow up study lean season diet is recommended to add to our knowledge of lean season cassowary food species for habitat restoration and enrichment projects.
The dietary analysis and fruiting study compared fleshy rainforest fruits found on transects both in the scat and on the ground. 71 scats were analysed and contained 28 identified and 11 unidentified species and two exotic palms. On the forest floor, 201 occurrences of 39 species were encountered that represented 23 plant families.
The estimated biomass of 18 identified fruits found common to the scat and the ground were compared between sites, seasons and species as well as by family, colour, weight range and fruiting pattern. The results suggest a preference for some species and an avoidance of other species. Five species; Syzygium Kuranda, Syzygium graveolens, Cerbera floribunda, Elaeocarpus augustifolius, Beilschmiedia castrisinensis made up 87.2% of the total estimated biomass of fruits eaten.
Syzygium kuranda was the most readily utilized species over both seasons. Whilst none was found on the forest floor during the abundant season a 3.4 fold increase of this species is seen in the scat between the lean and abundant season suggesting abundant season selective foraging. The only species suggesting preferential foraging during the lean season is Cerbera floribunda, whereby 56.5% of the total biomass of this species was consumed during this time
These 5 species, along with Syzygium hemilampra, are all substantial contributors to the lean season diet and should be given special attention by seed collectors, native nurseries and rainforest regenerators.