Like any other science, ecology depends on a range of methods and, during the course of my research, I have developed several new approaches or refinements to existing ones.
While some of these advances are conceptual or quantitative, some rely on advances in technology. Just as the use of remotely-triggered cameras have revolutionaized the way we survey for various shy and elusive animals, acoustic monitoring is an emerging field with great potential for surveying for all those animals which vocalize—revealing which animals occur where, when and .
Frustrated with existing methods to survey terrestrial bird communities that rely on fixed efforts, I developed a new approach using results-based stopping rules to allow areas of greatly differing size, complexity and diversity to be sampled to the same degree. Called the Standardised Search, this approach uses repeat samples to estimate those species present but not detected, generating robust completeness measures and comparable species richness estimates.
Measuring reproductive effort is a fundamental aspect of autecological research, and the basis of many comparative approaches to understanding the evolutionary ecology of life history variation. Studies of birds have traditionally relied on clutch size (the number of eggs laid in a single reproductive bout) as the sole estimate of reproductive effort. While of great biological significance (more eggs = more chicks = greater fitness) simply counting the number of eggs ignores potentially informative variation in eggs size. We have pioneered the use of clutch volume as a more inclusive measure of reproductive effort, calling into question some long-held beliefs about bird reproduction and demonstrating that clutch size may not be representative of underlying variation in reproductive effort.
One of the most fundamental aspects of any parasite is host range—the number of different organisms which can act as a host. Using mistletoes and other parasitic plants as a model system, I am developing novel approaches to estimate host range, evaluating whether results-based stopping rules can determine how many samples are required to generate robust estimates.