Andrew J. Tyre

At a broad scale, I am interested in using theoretical models to understand real ecological systems at scales larger and longer than can be addressed with traditional experimental and observational studies. At the same time, I am committed to connecting my models to observations of real systems. Exploring methods for challenging models with real data (Tyre and Tenhumberg, 2000 Austral Ecology; McCarthy et al., 2001 Conservation Biology) has lead me to model the observation process itself, and to work on modern statistical methods to quantitatively fit ecological models to empirical data (Tyre et al. in press Ecological Applications).

My work with survey data leads me to use very simple models that can be fit directly to ecological data. My other interest is in identifying the limits of what simple models can tell us about ecological systems (e.g. Tyre et al. 2001, Ecological Applications). In this work I have used spatially explicit simulation models to create artificial realities sampled by virtual ecologists. In this way I am able to directly link the kinds of data collected by ecologists with the underlying dynamic processes.

In pretty much everything I do I work together with Dr. Brigitte Tenhumberg.

Estimating false negative rates in presence/absence surveys
A presence/absence survey conducted at two different times can have four possible outcomes; in two cases it is possible that there was an ecological change, either the species has disappeared locally, or it has recolonised the area. The other possibility is that in one survey the species was simply missed, although it was still present. This is a ‘false negative’ observation. I developed a method based on fitting mixtures of binomial distributions to repeated visits to estimate the false negative rate. My collaborators and I have used the method to quantify false negative errors in bird, frog, and aquatic invertebrate surveys (Tyre et al. in press, Ecological Applications).
More >>

Population regulation in ticks on reptiles
I'm collaborating with Prof. Mike Bull of Flinder's University on understanding the population dynamics of two ixodid tick species that share the same lizard host in Southern Australia. The biogeographic distributions of the two species abut along a narrow parapatric boundary thousands of kilometres long. I am using individual based simulation models, mark recapture statistics, and maximum likelihood methods in an effort to extract more information from the existing data, and reveal the mechanisms responsible for the boundary.
More >>

Optimal monitoring in the presence of survey errors
I'm working with Dr. Scott Field, Prof. Hugh Possingham, and doctoral students Jonathan Rhodes and Liana Joseph at The Ecology Centre on designing ecological surveys that account for observation error. Scott is working on the woodland birds of the Mt. Lofty Ranges in South Australia, Jonathan is working on Koala populations in northern New South Wales, and Liana is studying the interaction between observation errors and rarity in bird communities of the wet tropics in Queensland.
More >>

Metapopulation dynamics of mound spring invertebrates
Around the edge of the Great Artesian Basin in the Austrlian outback are a large number of natural artesian springs. These permanent wetlands in the middle of one of the most arid regions on earth are home to an astonishing diversity of endemic invertebrates. I am collaborating with Dr. Chris Wilcox, Prof. Hugh Possingham, and Darren Niejalke of Western Mining Corporation on a project to model the dynamics of, and develop survey methods to accurately assess the status and trend of these populations.

Marine reserve design
I have a passing interest in using optimisation methods developed by Dr. Ian Ball (Australian Antarctic Division) and Prof Hugh Possingham for marine reserve design. In collaboration with Dr. Rodrigo Bustamante and Ian Mcleod of CSIRO Marine Research I employed the software package MARXAN to explore the distribution of biodiversity in deep waters in the southeast marine region of Australia.