Wednesday, December 03, 2008

Bird Surveys and Background Noise

I recently came across an article (originally noted here) with some discouraging news about bird surveys. One of the key datasets used to measure North American bird population trends is the Breeding Bird Survey, in which observers conduct routine point counts along standardized routes. These point counts involve some visual observation, but most identifications are made by ear. A study found that even experienced observers tended to underestimate when identifying birds in the presence of background noise.

In order to explore these questions, Simons and others worked to develop "Bird Radio:" a series of remotely controlled playback devices that can be used to accurately mimic a population of singing birds. Researchers could then control variables, such as background noise, to see whether it affected birdwatchers' ability to estimate bird populations.

The study found that even small amounts of background noise, from rustling leaves or automobile traffic, led to a 40 percent decrease in the ability of observers to detect singing birds. What's more, said Simons, "we also learned that misidentification rates increased with the number of individuals and species encountered by observers at a census point." In other words, the researchers found that traditional means of estimating the abundance and diversity of bird species are flawed due to complications such as background noise and the accuracy of the data observers collect on surveys of breeding birds.

But the Bird Radio research also points the way toward possible solutions. Simons explains that the Bird Radio findings are helping researchers develop better sampling methods and statistical models that will provide more accurate bird population estimates. For example, researchers are attempting to identify data collection methods that will help account for background noise or other outside factors in estimating bird populations.
The result may call into question the numbers undergirding some recent State of the Birds reports, which rely heavily on data from the Breeding Bird Survey and Christmas Bird Counts. Those surveys are used for other reports and management decisions as well, so it is important that they be as accurate as possible, so that conservation money is going to the right place. I hope that the study authors find a better way of sampling, and that they share it with the rest of us who participate in such surveys.