How to detect newly introduced species

By: Dr John Gollan, Category: Science, Date: 03 Dec 2009

Museum scientists compared two strategies to predict the whereabouts of an introduced bee – one using climate suitability data and the other habitat preference.

Find out which strategy worked best.

Halictus survey sites

Halictus survey sites
Photographer:  © Australian Museum

In 2004, Museum researchers found an unfamiliar bee species during a series of surveys along the Upper Hunter River in NSW. The species was later identified as the Emerald Furrow Bee (Halictus smaragdulus), native to the Mediterranean and Middle East. One question that immediately came to mind was: how widespread was the species?

While further survey work could address this question, it was important to choose the most efficient survey strategy because newly introduced species often occur in low densities, making it difficult to determine their distribution. A grant from the Winifred Violet Scott Charitable Trust allowed Museum scientists Dr John Gollan, Dr Michael Ashcroft, and Dr Michael Batley to compare two approaches to find some answers.

In the first approach, specimen records from the bee’s native range were overlain with climate data to determine the most favourable climatic conditions for the bee. This information was then used to produce a climate suitability map for Halictus in Australia. Almost 300 sites were subsequently surveyed to include locations that were either in close proximity to the Hunter Valley (where they were first found), or had high climatic suitability. The result? Halictus was found at only 3% of sites, with only one new location outside the Hunter Valley. Nonetheless, these surveys extended the known occurrence of the species 170km westward to Wellington, and 20km south and east to the lower Hunter region.

In the second approach, to improve their encounter rate with the introduced species, the researchers reviewed the data and field notes of all previous surveys. This exercise provided new insights on the ecology of the species, which suggested that Halictus favoured habitat patches with exposed sandy deposits near creeks and rivers, and an abundance of flowering weed species. The team conducted further surveys to target areas with apparently favourable habitat. Surveys using this approach detected Halictus at 22% of sites, extending the known range to between Sydney, Tamworth, and Wellington, and providing a more complete picture of the distribution by detecting the species at 12 sites outside the Hunter Valley.

A comparison between the two approaches showed that climate matching provides an indication of the range that is climatically suitable for a species, but is of limited use in determining where a species exists within that range. Surveying areas with favourable habitat, however, significantly increases the chance of detecting a species.

Our recommendation is to adapt and modify survey strategies as new field data about suitable habitat become available. The importance of field notes, expert knowledge and more generally ‘gut-instinct’ in surveys should not be forgotten.


Natalie Sullivan assisted with field work and sorting and Alain Pauly from the Royal Belgian Institute of Natural Sciences provided distribution data in Europe.