Database of containing abundance and location information for vertebrates of the Australia Wet Tropics rainforests. Records contained here are sourced from miscellaneous records & standardised surveys of the Centre for Tropical Biodiversity & Climate Change, supplemented by an array of external personal and institutional datasets.
Detailed Methodology for Occurrences Distributional data for the rain forest vertebrates of the AWT was collected during field intensive surveys and collated from the literature and institutional databases (as per Williams 2006). Major sources of species occurrences included: field intensive surveys by various researchers in the School of Marine and Tropical Biology, James Cook University, from which the data is now maintained and continued monitoring standard survey sites is done by the Centre for Tropical Biodiversity and Climate Change, James Cook University; Birds Australia Atlas of Australian Birds; QPWS Wildnet fauna database; and individual biologists (see special reference section of Williams et al. 1996). All occurrences were vetted for positional and taxonomic accuracy prior to use in modelling. REFERENCES
Williams, S. E. 2006. Vertebrates of the wet tropics rainforests of Australia: species distributions and biodiversity. Cooperative Research Centre for Tropical Rainforest Ecology and Management. Rainforest CRC, Cairns, Australia. Williams, S. E., R. G. Pearson, and P. J. Walsh. 1996. Distributions and biodiversity of the terrestrial vertebrates of Australia's Wet Tropics: a review of current knowledge. Pacific Conservation Biology 2:327–362.
Detailed Methodology for Abundance Estimates (adapted from as Appendix B from J. VanDerWal et al., “Abundance and the Environmental Niche: Environmental Suitability Estimated from Niche Models Predicts the Upper Limit of Local Abundance” (Am. Nat., vol. 174, no. 2, p. 282) General Overview of Methodology Used to Estimate Relative Abundances
Most locations were sampled once using a particular survey technique, although a number of locations were sampled repeatedly on up to 21 occasions. Where multiple counts were conducted on the same transect, an average abundance for each species was used for that location. These data represent a combination of many years of sampling effort (August 1992–present) for several separate studies that aimed at understanding the determinants of spatial patterns of vertebrate biodiversity but that all utilized identical standardized methodology. Although abundances and detection probabilities are not expected to be static through time, obtaining a definitive estimate of abundance at any one location was not an objective of the study. Rather, we were interested in the pattern of relative abundance across environmental space. In the context of our study, interpretation of these patterns is only problematic where (1) there is a spatial/temporal bias in the sampling of transects or (2) detection probabilities vary greatly between multiple observers. Neither of these problems was applicable here.
The vast majority of sampling was conducted during the late dry and the wet seasons (September–April; table B2), that is, the time of year when the majority of the wet tropics rain forest species are known to breed. Most species in our study are considered to be year-round residents within the region. The sampling approach, then, maximized the likelihood that surveys were conducted during periods of peak activity. Environmental gap analyses using generalized dissimilarity models (Ferrier et al. 2002) were performed to guide sampling effort and to maximize the coverage of environmental space across the region. There was no obvious seasonal bias in the distribution of survey effort for any of the techniques employed (here shown as the major environmental gradients of annual mean temperature and precipitation; fig. B1). Detection probabilities were maximized, false negatives were minimized (i.e., a zero count when the species is actually present), and observer detection probability was standardized in a number of respects. We used appropriate methodologies that were suitable for sampling the vast majority of vertebrate species utilizing rain forest habitats within the region. Exceptions included nocturnally active birds (e.g., sooty owl, rufous owl, Papuan frogmouth, tawny frogmouth), waterbirds, highly fossorial species (e.g., burrowing reptiles), and “flyover” species active above the rain forest canopy (e.g., raptors). However, none of these species was included in our analysis. Sampling techniques were employed only at appropriate times of day and under suitable weather conditions (see individual techniques for details). Difference in detection probability between observers is expected to be low. All data collectors were trained in the field by researchers with considerable experience with the sampling techniques and the fauna. Only species that were readily recorded on surveys were used and, thus, we do not expect that detection would vary significantly among observers. Difference in detectability between species was not relevant in this study, as all analyses are based on relative abundances across sites for each individual species. This only makes the assumption that detectability of a species does not vary significantly across locations a reasonable one, given that all surveys are within rain forests of relatively similar structure, density, and conditions.
This research has been funded in full or part by the below funding agencies: - Skyrail Rainforest Foundation - Birds Australia - Earthwatch Institute - Marine and Tropical Sciences Research Facility (MTSRF) - National Environmental Research Program (NERP) - Rainforest CRC - Australian Research Council - James Cook University Research Advancement Program (RAP) - Wet Tropics Management Authority (WTMA) - National Geographic - National Science Foundation - Queensland Smart State Program
Disturbance is not considered to be a significant factor affecting abundance patterns in this study. Although considerable habitat clearing has occurred in the lowlands, there are still large tracts of relatively intact rain forest at all elevations, and there is no evidence to suggest that any species have disappeared locally due to anthropogenic disturbance, except in smaller fragments (J. Moloney, personal communication). Disturbance is probably most relevant at the local scale (<1 km), where disturbance events such as tropical cyclones damage vegetation structure, thereby influencing assemblage structure. Disturbed sites were avoided during site selection, and all sites included were in intact forest.
Protocols for Individual Sampling Techniques
Amphibians Amphibians were sampled using either microhylid ( n p 3 species) or stream survey ( n p 2 species) techniques. Microhylid surveys consisted of a nocturnal, slow-paced walk (∼10 min duration) along a 50-m transect through rain forest. All calls were identiﬁed to species, and counts were made of number of individuals calling within 10 m on either side of the transect. Because detection probability was dependent on calling males, surveys were timed to coincide with known months of breeding. In all, 25 of the 27 dated breeding records compiled for microhylid species in the region are from the late dry to the early wet seasons (Hoskin 2004). Surveys for microhylid species were thus exclusively conﬁned to this time period. Surveys were also conducted only on wet, humid nights (180% relative humidity). These conditions are known to be conducive to calling activity. Subsets of abundance counts using the same standardized technique have been used in analyses elsewhere (Shoo and Williams 2004). Stream surveys consisted of a 200-m nocturnal transect along rain forest streams. All individuals were located and identiﬁed visually (usually by eye-shine reﬂection) and/or acoustically (by call). Counts were made of any individual located within the stream itself or along adjacent streamside embankments. This methodology is consistent with those previously employed within the region (e.g., Hodgkison and Hero 2002).
Nocturnal visual counts of mammals were made in 1,000-m spotlighting transects along old unused logging tracks, with 1 h of search effort, using a combination of one observer with a 30-W handheld spotlight and another observer with a low-power head torch. This combination maximizes efﬁciency, as the powerful spotlight is best for detecting arboreal mammals and owls, while the low-power head torch is better for detecting geckos and frogs (Williams 1995). Reptiles Reptiles were counted during 1 person-hour, and the search was restricted to a 50-m radius of active diurnal examination of the forest ﬂoor and vegetation, as well as shelter sites, such as under logs. Surveys were conducted only under warm (air temperature, 122°C) or sunny conditions and never during rain. Sampling criteria were based on regression tree analysis of reptile abundance and environmental conditions (S. E. Williams, unpublished data). Birds Bird surveys consisted of walking for exactly 30 min along a 150-m transect (measured using a hip chain) through the rain forest, using both visual observations and bird calls to identify species. Only calls made within ∼50 m of the transect line were recorded. As much care as possible was taken to avoid double-counting of calling individuals. Surveys were conducted only if environmental conditions on the sampling day were suitable enough to ensure that daily conditions did not bias the results; for example, no surveys were conducted in rain or strong winds. Subsets of abundance counts using the same standardized techniques have been used in analyses elsewhere (Shoo et al. 2005a, 2005b; Williams and Middleton 2008).
REFERENCES Ferrier, S., M. Drielsma, G. Manion, and G. Watson. 2002. Extended statistical approaches to modelling spatial pattern in biodiversity in northeast New South Wales. II. Community-level modeling. Biodiversity and Conservation 11:2309–2338.
Hodgkison, S. C., and J.-M. Hero. 2002. Seasonal behaviour of Litoria nannotis, Litoria rheocola and Nyctimystes dayi in Tully Gorge, North Queensland, Australia. Pages 29–39 in A. E. O. Nattrass, ed. Frogs in the community: Proceedings of the Brisbane Symposium, February 13–14, 1999. Queensland Frog Society, Brisbane. Hoskin, C. J. 2004. Australian microhylid frogs (Cophixalus and Austrochaperina): phylogeny, taxonomy, calls, distributions and breeding biology. Australian Journal of Zoology 52:237–269.
Shoo, L. P., and Y. Williams. 2004. Altitudinal distribution and abundance of microhylid frogs (Cophixalus and Austrochaperina) of northeastern Australia: baseline data for detecting biological responses to future climate change. Australian Journal of Zoology 52:667–676. Shoo, L. P., S. E. Williams, and J.-M. Hero. 2005a. Climate warming and the rainforest birds of the Australian wet tropics: using abundance data as a sensitive predictor of change in total population size. Biological Conservation 125:335–343. ---. 2005b. Potential decoupling of trends in distribution area and population size of species with climate change. Global Change Biology 11:1469–1476. Williams, S. E. 1995. Measuring and monitoring wildlife communities: the problem of bias. Pages 140–144 in G.C. Grigg, P. T. Hale, and D. Lunney, eds. Conservation through sustainable use of wildlife. Centre for Conservation Biology, University of Queensland, Brisbane. Williams, S. E., and J. Middleton. 2008. Climatic seasonality, resource bottlenecks, and abundance of rainforest birds: implications for global climate change. Diversity and Distributions 14:69-77.
Coinvestigators: None Related JCU Research Themes: Tropical Ecosystems, Conservation and Climate Change