In July 1991, the freighter Tuo Hai collided with the fishing vessel Tenyo Maru off the northern coast of Washington State, resulting in an oil spill that quickly spread to cover hundreds of square kilometers. At the time, I was working on a seabird colony in the path of the spill, and I realized, as the oiled birds began to wash in, that we had no baseline against which to compare the mounting body count. How many wash in normally?
In years of coastal fieldwork, I had already noticed a seasonal uptick in the beaching of my focal species, the common murre, during the summer-fall transition. These deaths were a natural result of the end of the breeding season when exhausted parents and inept chicks are more likely to die as the winter storm season begins, but I wondered to what degree this normal signal might be influenced by environmental forcing. To address either question, I needed a much larger, longer-term data set on the beaching rate.
I decided to create that data set and in the process founded the Coastal Observation and Seabird Survey Team (COASST). At the time, I was a young faculty member looking for a method of expanding data collection beyond the physical abilities of myself and my team. That mission has evolved over the course of 2 decades into a passion for promoting citizen science as a rigorous method of data collection, fully worthy of being part of the toolbox of mainstream science.
Where seabirds are found across the seascape is directly influenced by the quantity, quality, and predictability of their prey—a table that is set by oceanographic and atmospheric processes operating locally to globally, over days to decades. Seasonal upwelling, decadal climate patterns like the Pacific Decadal Oscillation, and relatively sudden shifts such as an El Niño or a marine heat wave will all influence prey availability, causing birds to choose to move—rightly or wrongly—in search of dinner. Wrong choices result in death, and that drifting carcass will succumb to the wind, currents, and tides. In the Pacific Northwest, a bird that dies within 125 kilometers of shore has a chance of reaching the beach before sinking. Those few carcasses washing ashore have a short shelf life before they are scavenged, buried by the wind, or washed back out during the next high tide.
Despite these sources of variability, there is a discernible cadence to beaching over time, an annual rhythm that can be tracked by monthly surveys. The most important thing that COASST does is statistically document this regionally specific pattern—the right place, right time, right species baseline—of what birds are expected where and when.
I started COASST with the help of a postdoc who knew far more about ornithology than I did, a grant from the David and Lucile Packard Foundation, and a question: Could we find locals enthusiastic about collecting data on the identity, abundance, and condition of beached birds on an ongoing basis and in a standardized manner? We also needed to design a data collection program that allowed for expert verification of species identity. If these data were ever to be used in mainstream science or in a legal proceeding following an oil spill, they needed to be beyond doubt.
What we came up with is one of the simplest tenets of science: evidence first, deduction second. In our case, the evidence includes classification of the foot type, three specific body measurements, and two photographs with a standard scale. The deduction is lowest taxonomic classification, which can be made with our custom field key, Beached Birds.
We started our data collection corps with 12 residents of Ocean Shores, along the southern coast of Washington. COASST today includes around 800 people collecting monthly data on beached birds and another 200 collecting data on marine debris. Our footprint stretches from Mendocino, Calif., to the Canadian border and throughout Alaska. We now work with partners in California (BeachCOMBERS and Beach Watch) and in western Canada (British Columbia Beached Bird Survey). That means our data collection currently spans three large marine ecosystems: the California Current, Gulf of Alaska, and eastern Bering Sea, as well as parts of the western Bering Sea and Chukchi Sea.
Unlike citizen science programs that use online training and occasionally rely on self-taught hobbyists, COASST commits to hands-on, in-community trainings designed for beginners. If you can tell it’s a bird and it’s dead, we can teach you how to “get to species” in a single 5-hour session. And the citizens in our citizen science? They are citizens of the planet, citizens of the ecosystem. We start in new communities when we’re invited, and we work with local partners to recruit trainees without questioning their knowledge about birds or their politics. I’ve trained people in bars, churches, ferries, elementary schools, libraries, malls, tribal headquarters, and senior centers. We currently visit about 80 small coastal communities in a 2-year cycle. And we commit to returning. We constantly communicate with our participants, offering feedback about whether that last carcass they found really was a northern fulmar or sending out the latest information we have about how and why beaching patterns are changing.
So, yes, we travel a lot, and, no, citizen science isn’t free, nor is it easy. We write grant proposals to support our science just like everybody else, and we push ourselves and our collaborators to get that next paper submitted. We think about what we do and how we can make it better every day.
From Data to People
In the beginning, I was obsessed with data quality and making sure our data collectors were being as accurate as possible. One way we solved that problem was by breaking down the science into component tasks, things like foot type classification and body measurements. Although this approach worked with species identification, our efforts to control the sampling design weren’t as successful.
Initially, we created a specific list of sites to be surveyed on the basis of the substrate, orientation, and inclination of the beach. But people wanted to choose “their place” whether our sampling design indicated it was needed or not. It seems impossibly arrogant to me now, thinking back on it, that my starting expectation was that I could waltz into a coastal community and tell people what to do and that they would do it without question. That works with undergraduate students and technicians, so why not with everyone else? We now know that what brings people to a training in the first place—a strong attachment to a specific beach and a desire to learn more about that place—was also the reason they politely but firmly refused to be assigned a survey location. COASSTers now choose the beach they want, and over 90% of training attendees sign on to participate.
What I’ve come to realize is that COASSTers are first and foremost people. If COASST can pique their interests, support their sense of place, provide them with proper training, respond to their questions and concerns, and thus offer them an authentic role on our science team, we are rewarded with a cadre of highly devoted, rigorous, long-term data collectors. The average COASSTer is able to identify the species of a carcass 87% of the time and maintains near-monthly survey frequency for about 3 years.
COASSTers understand their role as both scientific and social: They perform rigorous data collection, communicate to others about COASST, and recruit community members into the program. They talk to friends and family about their experiences, and they connect with influencers such as resource managers, politicians, and the media to tell them about our research findings. COASSTers embody the saying “Bear witness, take action.”
Big Data Citizen Science
Documenting environmental change requires long-term data on where and when natural things happen: earthquakes, extreme weather, the first flowers of spring, dead birds on beaches. At its scientific best, citizen science can create huge, detailed data sets that capture these patterns at local, regional, and even global scales. In COASST, thousands of participants have created a highly accurate, highly rigorous data set that has gone directly into science and resource management.
Our science stories are sobering. We’ve documented the largest marine bird die-off on record anywhere in the world due to a harmful algal bloom. We’ve shown that the impact of the largest and longest-lasting marine heat wave the planet has yet experienced included multiple, massive seabird mortality events from California to Alaska. Working with resource management partners, we’ve cocreated a series of annual ecosystem indicators that inform everything from the California Current Integrated Ecosystem Assessment to the annual report to the North Pacific Fishery Management Council. COASST is regularly asked to assist in decision-making on the basis of our data and our expertise: Should the hunting season for marine ducks be opened next week? Do these carcasses present a disease risk to coastal peoples? Should the beaches be closed to tourists? We do science that matters.
Today is Earth Day, and the 2019 campaign website kicks off with Rachel Carson’s quote, “In nature, nothing exists alone.” It’s a statement about ecosystem connectedness and fragility, but I believe it’s also a mission statement for how we conduct science within that ecosystem. If the past century was about expansion of science through technology, this century had better be about expansion through people. Closing the doors of the ivory tower and cloaking ourselves in a language that few can understand won’t save science, and it certainly won’t save the world.
Expanding our science teams to include everyone with an interest or a need is a scary but exciting thought because as the face of science changes, so will the practice. With a larger and more diverse team, we will ask and answer questions differently. Citizen science is one strand of that braided river of change. And now that I’m in it, I honestly can’t imagine why I would ever do things any other way.
—Julia K. Parrish ([email protected]), University of Washington, Seattle