The Most Accurate Count of Emperor Penguin Populations. From Space!

Good news for penguin-enthusiasts everywhere (that's all of us, right? Good... just checking): scientists studying emperor penguin populations in Antarctica have tallied almost 600,000 of the birds waddling around the icy continent. That's around twice as many as they were expecting to find — but what's really impressive is how they went about counting them.

According to SciAm, they actually did it from spa(aaaaaa)ce:

In 2009, [geographer Peter Fretwell and his colleagues] took satellite images of emperor penguins during their September-through-December breeding season. Using a sharpening technique... they adjusted their images to differentiate between adult penguins, their droppings and shadows, details that had confounded earlier efforts to survey the species. Using ground-based counts for reference, they developed an algorithm to identify which pixels in an image represented penguins, as opposed to the surrounding environment, and counted the hundreds of thousands of birds.

That's exciting for at least three reasons. Number one: it means there are more penguins than we thought there were, and that's always adorable news.

Number two: the team's results, which are published in today's issue of PLoS ONE, constitute the first-ever total population count for an entire species. Which leads us to number three: population studies in inhospitable (or inaccessible) places like Antarctica are notorious for being too expensive or too technically difficult to carry out on a large scale, so the team's findings really showcase the utility that satellite data offers ecologists and conservationists looking for efficient, cost-effective, and accurate methods of population analysis.

All told, it's just the latest in a recent string of examples of how satellite imagery is changing the way we approach long-standing scientific problems.

Read more about this study over at Scientific American, or check out the original paper (no subscription required) over on PLoS ONE.