About a decade ago, biologists set out to count river dolphins along a murky, meandering stretch of the upper Amazon River between Colombia and Peru. Although these distant freshwater cousins of seagoing whales face growing threats from pollution, dams, and hunting, the results seemed to suggest they were doing well. By the end of the 2007 survey, the team had spotted 116 groups of pink river dolphins (Inia geoffrensis), as well as 220 groups of a second species, the gray river dolphin or Tucuxi (Sotalia fluvia). Those numbers are higher than in two prior tallies done in 1993 and 2002.
But because the tallies had all been done differently, conservation scientists could not say for sure whether the dolphin populations were growing, merely stable, or already in dangerous decline. Now, researchers have used a 250-year-old statistical approach to draw firmer—and potentially worrisome—conclusions about the dolphin population trends.
Researchers count the dolphins by standing in a boat and scanning the vast river—dozens of kilometers wide in places. But techniques varied among the three surveys: One used just a single observer, for example, while others used two. The surveys were also done at different times of the year and at varying water levels. That’s problematic because the dolphins might bunch up during some seasons, making them easier to count, but swim deep into the Amazon’s flooded forests when the water is high, making them nearly impossible to find.
“It’s a pretty frustrating situation, because you really need to know how the animals are doing if you want to do effective conservation,” says marine mammal scientist Rob Williams, who got involved in studying the dolphins in the early 2000s, while at the University of St. Andrews in the United Kingdom. But Williams, who has expertise in statistics, saw a potential solution: Call on the talents of the Reverend Thomas Bayes, an English Presbyterian minister who in the 1750s outlined a flexible statistical approach that enables researchers to combine and analyze disparate types of information—such as surveys done in different ways. It wasn’t until the advent of modern computers, however, that Bayesian methods became widely practical; since then, they’ve swept through the sciences.
Now, Bayes has helped Williams and his colleagues extract some potentially meaningful trends from the messy dolphin data. The message is decidedly mixed, they report in this month’s Biological Conservation: Although there is a 75% probability that the gray dolphin population was stable or increased between 1993 and 2007, there’s an equal probability that pink dolphin numbers declined.
The study represents “a really nice” example of how conservation scientists can cope with problematic survey data, says fisheries biologist Trevor Branch of the University of Washington, Seattle, who was not involved in the research. The researchers “were quite careful” in their analysis, he says, but notes that there are other possible interpretations: for instance, some dolphin populations may have declined and then rebounded, rather than simply increased.
The authors agree that more work is needed to confirm the decrease in pink dolphins. But the trend “could be a warning sign of a significant problem that we need to understand better,” says co-author Fernando Trujillo, the scientific director of the Fundación Omacha, a conservation group in Bogotá. It could signal, for instance, that pink dolphins are being killed in substantial numbers by Amazon fishers, who sometimes see the animals as competition for fish, or use their flesh as catfish bait. (Brazil recently took steps to end that practice.)
“Once you have trend data, you are able to look for the causes of the trend,” Trujillo says. But without data, he adds, “it is hard to persuade people to make changes or take needed action.” (Click here to see a video on Amazon dolphin research and conservation efforts.)
Cheaper, better, faster?
Designing and fielding relatively cheap and easy surveys that can help shape conservation strategies is a challenge. “It’s a really common problem in conservation biology,” says Williams, now a Pew Fellow in Marine Conservation and co-leader of the Oceans Initiative in Seattle, Washington. “There’s rarely enough money to rigorously monitor populations … especially in developing nations, so you are often dealing with very sparse data.”
It would be prohibitively expensive to survey and monitor South America’s four species of freshwater dolphins using traditional “gold standard” methods, Williams notes. “So if we aren’t going to see a dramatic increase in spending for that kind of monitoring—and it is likely we won’t—we need to decide what a ‘silver’ or ‘bronze’ standard looks like.”
To that end, the new study lays out a potentially doable strategy for keeping tabs on the dolphins. It calls for a single, streamlined abundance survey every 5 to 10 years, with smaller low-cost surveys every few years that focus on other issues, such as how the animals move seasonally. And it urges the use of Bayesian methods to strengthen the statistical power. “The math can be hard,” Williams says, “but that isn’t an excuse for ignoring the problem.”
Trujillo notes that more than 150 dams are now proposed for the Amazon, and that coming up with workable methods for understanding how these and other projects are affecting the mammals will be critical. “The dolphins are a good indicator of the health of the whole river system,” he says. “So we need to assess how they are responding—before it is too late.”