Records Database Methods

Grids or Points?
Many good bird atlases are based on grids with the objective to survey every grid square, visiting all the habitats present. When done thoroughly this technique should deliver reliable distribution maps and reporting rates. However, grid searches need highly dedicated, experienced and mobile observers with time. There are not enough resident bird watchers in Vanuatu.

We expect most of the submissions to our bird atlas will come from visiting birders. We have chosen to use point based surveys because this is how we expect people will make bird lists. Casual birders tend to visit specific sites. The opportunistic nature of point-based searches suits them. Also, access across a wide area of land may be difficult and the best we can do is take a few sample point searches. Our primary objective is to encourage people to make bird lists and contribute them. Using point surveys increases the potential for bias but we don't feel grid searches are practical in Vanuatu today. A simple and undemanding survey procedure is needed.

Surveys
This project applies much of the survey methodology from Birds Australia's The new atlas of Australian birds (Barrett et al, 2003) with some simplifications.

Our database is point based, there are no grid squares and each survey is ultimately located by latitude and longitude. Squares and whatever other way we wish to visualize the data are an analysis choice. Presently we are using 6 second squares (0.1 minute) to generate maps (because some of the islands are quite small) but we only need to change some code to change the presentation grid size.

The recommended survey (a point survey) is a short duration (20 to 30 minutes) search of a small, accurately located area (around 2 hectares). It's a quick and easy presence/absence survey that can be performed by any birdwatcher. These surveys should be fairly consistent and comparable because observer behaviour is restricted by the time and area limits. The main influence on the survey results is expected to be observer experience (not to be confused with skill) - people who are familiar with the local birds and their calls are expected to detect more birds than newcomers in a short duration survey. Accurate topographic maps are not readily available for Vanuatu and a handheld GPS receiver is recommended for locating point surveys. A project to develop modern, accurate maps is ongoing.

Area surveys, with longer durations and larger areas are more flexible and more thorough but less controlled. If the search covers more than one atlas grid square then useful resolution is lost. We recommend search areas be no bigger than about 8 km x 8 km (6 second squares).

Breeding records must be proven. Acceptable breeding records include a nest with contents seen, adults making repeated visits to a nest or hollow carrying food or newly fledged young with a parent near the nest site (Barrett et al, 2003).

Records of new and unusual species are more likely to be accepted if the bird has been photographed, video taped and/or recorded. An Unusual Record Report Form can be downloaded here.

The database aims to collect new and recent data on the distribution and abundance of Vanuatu's birds. Records preceding 1990 generally do not serve this purpose and this earlier knowledge has already been documented by Bregulla (1992).

More details on bird atlassing are in our observer's guide which can be downloaded here.

Submission
Don't be discouraged by the rules and suggestions above! They are guidelines recommended for future surveys and since we need to improve coverage at this early stage we can accommodate island lists, incidental lists and other imprecise and/or incomplete surveys.

New observers can download a survey form and observer's guide here.