Miriam Bobkoff is nervous. "It could be pretty bad out there today," she says, as we drive from her home in Port Angeles, Wash., to Kalaloch, an ocean beach on the Olympic Peninsula. Miriam volunteers with the Coastal Observation and Seabird Survey Team (COASST), a citizen-science program based out of the University of Washington in Seattle. At least once a month, she and other COASST volunteers survey Pacific beaches from Northern California to Washington, looking for dead seabirds.

Bobkoff likes birds well enough, at least live ones, but she doesn't go out of her way to see them. What she loves is the ocean. When she moved to Washington two years ago from New Mexico, she wanted reasons to go to the coast as often as possible, and if social good could be done in the process, so much the better. COASST had a training session that fit her schedule, so she went, and soon she was assembling her "citizen-science backpack" and learning to stanch her squeamishness about "dead bird cooties." The person leading the training, though, said she needn't worry.

"I was told I could walk my beaches for years and probably never see a (dead) bird," Bobkoff says. Which would have been just fine with her. But then September rolled in, and with it came the scoters.

The surf scoter (Melanitta perspicillata) is a large sea duck that breeds in Alaska and northern Canada. At the end of the summer, they migrate down the coast and spend the winter floating just offshore in enormous rafts dense with thousands -- sometimes tens of thousands –– of birds. Identifying a surf scoter can be a little tricky if, like Bobkoff, you aren't that into birds. The most helpful field marks, according to The Sibley Guide to Birds, are the adult male's multicolored bill and white head-patches; for females, look for white cheeks and a dark cap. The book Bobkoff used in September, however -- Beached Birds: A COASST Field Guide -- had a more macabre precision: The bill of the surf scoter is 33-41 millimeters long; the tarsus, or leg, 48-59 millimeters; the wing chord, or the distance between the wrist to the tip of the longest primary feather, 21-25 centimeters. These are not details you're likely to notice when the bird is swimming, but they're easy to determine when it is sprawled on the sand, dead.

The co-author of Beached Birds, and the director of COASST, is Julia Parrish. In mid-September, Parrish started to get reports of hundreds of dead and dying scoters along the Olympic peninsula. Volunteers told of beaches thick with a stiff brownish foam that was, says Parrish, "somewhere between a latte and meringue." To Parrish, who is also a professor of fisheries at the University of Washington, the numbers indicated that a major seabird mortality event, or wreck, was underway. She mobilized COASST volunteers and sent them out. 

On Sept. 14, Bobkoff and her COASST partner surveyed Rialto Jetty and Ellen Creek, a few miles north of Kalaloch. By the end of the day, the two had measured and tagged 10 birds -- one common murre, two gulls of indeterminate species and seven scoters. In one day, she had tripled the number of dead birds she'd found in two years.

"I usually do pretty well at staying detached with data," Bobkoff says. But as the afternoon dragged on, bird after bird, she came upon still another surf scoter, this one a male, limp, freshly dead. "Its bill was a sunset of colors," she remembers, "and I thought, This isn't just #756, or whatever -- it was a beautiful sleek swimming flying eating critter." She pauses. "I sort of lost my composure then."

The next weekend, she went to Third Beach to help two other COASST volunteers. It was, she says, "the heart of darkness." They measured 62 birds: one pelagic cormorant, one red-throated loon, and 60 scoters. At nearby beaches, other volunteers recorded more than 100 dead scoters per kilometer.

Back in Seattle, Parrish tried to figure out what was killing the birds. She initially thought that a toxic algae might be responsible, but researchers with the Northwest Fisheries Science Center had recently read a paper describing a 2007 wreck off Southern California, and the current conditions in the Northwest sounded suspiciously similar. A nontoxic plankton, Akashiwo sanguinea, had been the cause then. Such was the case now, the researchers found -- concentrations of A. sanguinea were thousands of times higher than normal. From the air, the algal bloom looked like a giant brown smear, covering hundreds of square miles. And then the weather turned bad.

"The timing couldn't have been worse," Parrish says. "The phytoplankton had gone to town, and then we had huge storms a couple of weeks earlier in the season than we would normally get here." Turbulence ruptured the phytoplankton cells and whipped them into foam. Into this foam came the migrating scoters, which molt after breeding, losing all their flight feathers at once. Normally, this isn't a problem -- scoters are perfectly at home in high waves -- but as they swam, the foam stuck to their feathers, collapsing them into droopy wet blobs. No longer insulated against the cold, the scoters were driven to the beaches, where they stood, unable to re-enter the water. Most died of hypothermia or starvation.

By the end of October, when the bloom finally dispersed, Parrish estimated that 10,000 surf and white-winged scoters had died, along with thousands of common murres, western grebes, and common and red-throated loons. Wildlife rehabilitation centers in Washington and Oregon were overwhelmed. The U.S. Coast Guard airlifted a few hundred birds to a rescue center in Fairfield, Calif. It was the largest wreck ever recorded in Washington.

In the aftermath, Vera Trainer, a scientist with the Northwest Fisheries Science Center, told the Seattle Times, "We need to look at these organisms as indicators of some kind of change, whether climate change or other factors. The ocean is trying to tell us something." But what exactly that is can be difficult to interpret. The ocean gives many hints, some of them spectacular: An oxygen-depleted "dead zone" bigger than Connecticut has appeared off the Oregon coast every summer for the past eight years; in southwest Washington, oysters fail to reproduce and changes in ocean chemistry are blamed; in September, salmon runs in British Columbia collapse, and the grizzlies and brown bears that eat them starve; now, thousands of seabirds wash up here, dying or dead. But all of these are still, for now, just hints.

Seabird wrecks are not uncommon events. Kristine Bovy, an archaeologist at the University of Rhode Island, has examined a midden in southwest Washington that dates back to the late Holocene. The midden was for a time scattered with the bones of sooty shearwaters, a wayfaring petrel that spends its time in North America, New Zealand and places between. At some point a little over a thousand years ago, the shearwater bones stopped appearing. (Bovy speculates that changes in oceanic conditions and/or overharvesting were responsible.) Curiously, the shearwater bones were replaced by those of Cassin's auklets and common murres, two relatively common local seabirds.

"The humans living there aren't known to have hunted far from shore," Bovy says. "And the Cassin's auklet especially is so small that I can't imagine anyone going to a lot of trouble to catch them. I think a plausible explanation is that the auklets and murres washed ashore in wrecks and were eaten."

Weaving such hints into a larger narrative is partly why Parrish started COASST in 1998. She wanted to determine baseline rates of seabird mortality so that, in the event of an oil spill or some other catastrophe, the magnitude of the death spike could be quantified. But when the numbers started rolling in from volunteers, Parrish saw something else: Seabirds die all the time, and for all sorts of reasons.

One of the first massive mortality events that COASST studied intensively was in the spring of 2005. The reproductive success of Cassin's auklets at a colony on Vancouver Island plummeted; at Protection Island in the Strait of Juan de Fuca, a colony of glaucous-winged gulls that normally fledged over 8,000 chicks produced only 88; at Tatoosh Island, also off the Olympic Peninsula, the nesting success at a large colony of common murres was just 20 percent; and all along the Oregon and Washington coast, tens of thousands of common murres and Brandt's cormorants washed ashore, starved and emaciated. Biologists concluded that an annual upwelling event in the Pacific, the spring transition, had been delayed for unknown reasons by about 50 days. A food web that depends on the influx of nutrient-rich waters was put on hold, in effect. But the birds, of course, didn't know, and so the various local catastrophes resulted.

Later, in a 2007 paper, Parrish and several colleagues compared data from those events against longer-term datasets of beached birds, in search of broader patterns. Did mortality rates correlate with factors like ocean temperature and current? The effects, they found, were highly localized. The timing and magnitudes of the various wrecks and failures were, again, suggestive, but underlying explanations for the marine upheaval were harder to tease out. "Are beached birds good indicators of ocean conditions?" the authors asked at the end of the discussion. "The present paper has clearly demonstrated that, although the answer can be yes, it is not necessarily so."

The rain has stopped by the time Miriam Bobkoff and I reach the Kalaloch campground. We get out of the car and bundle up, peering warily out at the Pacific. The tide is high and the ocean is foamy, but not obscenely so.

"Maybe it won't be so bad this time," Bobkoff says as we head down to the beach. Our one-and-a-half kilometer route skirts the edge of a huge mess of driftwood logs, slippery from the rain. We clamber over and under them, peeking in crevices to see what dead birds might be caught, but we don't find any.

Bobkoff is encouraged. She pulls out a garbage bag and starts filling it with pieces of paper and fraying rope and the odd plastic shovel. Heartened by a beach now free of garbage as well as dead birds, we walk far past the survey's appointed end. Even with this extra search effort (which we couldn't have used since it would have messed up the data), we don't find a single dead bird. By this time we're hauling more than 20 pounds of trash, so we decide to start back. As we near the walkway, we see a couple of large, dark shapes, riding the breaking waves.

"Can you make out what they are?" Bobkoff asks. "I can't tell from here."

I look at the field marks: heavy-bodied, white head patches, and some -- the males -- have wildly colored bills. I tell her that they are surf scoters.

"Surf scoters!" she says. We watch as two become eight, then 10, then a small raft of maybe 20 or more, rolling in the heave of the swell. Bobkoff rocks back and forth on her heels. "Look at them all, bobbing around out there!" she exclaims suddenly, happily. "Bless their hearts!" 

Eric Wagner writes from Seattle, Washington.