The effects of landscape context and foliage chemisty on the distribution and abundance of arboreal marsupials

A project undertaken at the Centre for Resource and Environmental Studies, Australian National University, and supervised by David Lindenmayer and Kara Youngentob


The levels of available nutrients and chemical defences in plant foliage are both genetically and environmentally determined. This results in a patchy distribution of forage quality within a landscape, even if it appears homogenous in terms of species composition and density of cover. Herbivores must make decisions about what and where to eat depending on the quality and availability of food resources. Studies on animal foraging have found a significant and even primary relationship between individual foraging decisions and both plant nutrient availability and plant chemical defences. Until recently, assessing plant chemistry on a landscape scale has been impractical because it required collecting thousands of leaf samples in the field for lengthy laboratory analyses. Recent technological advances in near-infrared spectrometry and hyperspectral remote sensing now allow for rapid assessment of leaf chemical composition in the lab and across whole forest canopies. Using these new techniques, we plan to investigate the relationship between animal distributions and foliage chemistry on a large scale. We predict that foliage chemistry distributions are a significant factor in explaining the presence of folivores and patterns of herbivory across landscapes and this knowledge could inform conservation and natural resources management decisions in new and important ways.

Study Site

The site, 100km west of Canberra near Tumut, NSW, Australia, is the location of a rigorous, large-scale ecological study initiated by David Lindenmayer and Ross Cunningham in 1995 to explore the effects of habitat fragmentation on animal populations (additional information at Data on the presence and abundance of mammals, birds and reptiles has been collected in both eucalypt forest remnants surrounded by pine plantations and in continuous blocks of neighbouring State and National Forests over several years. The close proximity to ANU facilities, the existing records on wildlife distributions, and the inclusion of protected forest reserves makes this a particularly attractive location to conduct this research.

Scientific Objectives

Firstly, we plan to collect up-to-date presence and abundance data for all arboreal marsupials in our study area, including the threatened yellow-bellied glider (Petaurus australis). The new data will be used in conjunction with previous surveys of the area to help assess arboreal marsupial response to habitat fragmentation over time. This will improve our understanding of the current status and conservation requirements of arboreal marsupials in eucalyptus forest fragments within the Tumut pine plantations and neighbouring Kosciusko National and Brindabella State Parks.

Our second main objective is to map the forest canopy distributions of nitrogen, condensed tannins and total polyphenols across our site using HyMap hyperspectral remote sensing data. This will allow us to test hypotheses relating to interactions between foliage chemistry and folivore distributions. We will identify the best predictors/models for explaining the distribution of ringtail possums (Pseudocheirus peregrinus) and greater gliders (Petauroides volans) based on patterns of these forage qualities, the presence of potential competitors and physical site characteristics. We have chosen to focus on these two species in a eucalyptus ecosystem to build upon existing laboratory research that has demonstrated that the foraging behaviours of these folivores are influenced by specific plant chemicals. This is an exciting step towards being able to predict herbivore distributions, movements and abundance using plant chemical compositions and to test theories of optimal foraging on a landscape scale. This information will also add to limited knowledge of foliage chemical distribution in montane ash eucalyptus forests and the effects of fragmentation and pine plantations on nutrient dynamics in a montane ecosystem.

Anticipated Outcomes

We anticipate that this research will allow us to test theories of optimal foraging and explore community interactions in new and exciting ways, owing to the scale of our investigation, the incorporation of recently identified foraging responses and the use of rapidly developing hyperspectral remote sensing technologies. The collaboration between researchers from several fields will hopefully provide a successful model for interdisciplinary studies and strengthen ties between the various faculty and research institutions involved. This study will also contribute to the understanding of how habitat fragmentation effects arboreal marsupial populations in plantation landscapes. This research should enable us to provide improved management recommendations for arboreal marsupial to State and National Parks managers in our study region.


Kara Youngentob, Professor David Lindenmayer and Ross Cunningham (Centre for Resource and Environmental Studies, Australian National University), Dr. Alex Held (Director, CSIRO Office of Space Science and Applications), Dr. Xiuping Jia (Australian Defence Force Academy, Dept. of Electrical Engineering), Professor William Foley and Dr. Ian Wallis (Dept. of Botany and Zoology, Australian National University), Dr. Jelle Ferwerda (Royal Melbourne Institute of Technology and Oxford University), and ANU WildCountry Research and Policy Hub.

The Tumut Fragmentation Experiment was initiated by David Lindenmayer and Ross Cunningham in 1995 to explore the effects of habitat fragmentation and landscape context on animal populations.

PhD student, Kara Youngentob, re-marking transects in the study area. The transects will be spotlit at night to count arboreal marsupials by their reflective eye-shine. Leaf samples will also be collected along transect lines for laboratory analyses and HyMap calibration.

A brushtail possum, Trichosurus vulpecular, feeding on eucalyptus leaves. Specific leaf chemicals have been shown to play an important role in forage selection and feeding behaviours. Photo by Karen Marsh.

Vegetation spectra extracted from AVRIS data. Chemicals of interest, such as nitrogen and condensed tannins absorb light at specific bands along the spectrum.

Reflectance spectra will be collected from the forest canopy at selected locations across our site, using HyMap hyperspecral remote sensing. This data will be used to map the variation in specific foliage chemicals that have been shown to influence herbivory.

The greater glider, Petauroides volans, a eucalyptus folivore specialist. Photo by Ester Beaton.

A ringtail possum, Pseudocheirus peregrinus. Photo by Ester Beaton.

This is a Landsat satellite image of the research area showing the original Tumut study transect locations (in red), and the planned HyMap flight-lines (in yellow). Each flight-line is 20 km long and 1.6 km wide with a resolution of 3-3.5 meter pixels. HyMap hyperspectral sensor captures 126 spectral bands, ranging from 450nm-2500nm.