Climate effects and shifting Arctic coastlines

Welcome to Science Sessions, the podcast of the proceedings of the National Academy of Sciences, where we connect you with Academy members, researchers, and policymakers. Join us as we explore the stories behind the science. I'm Paul Gabrielson. The Arctic is at the forefront of climate change effects, where temperatures are rising faster than in other parts of the Earth. Melting of the perennially frozen permafrost reshapes the Arctic landscape. In a recent PNAS study, Roger Creel of the Woods Hole Oceanographic Institution and colleagues modeled the compound effects of sea level rise, coastal erosion, and ground subsidence due to permafrost melting on the North Coast of Alaska. They found that these processes combine to erode the Arctic coastline much faster than would be expected by erosion alone, and that communities and infrastructure in that region face significant risks. Roger, why is ground subsidence happening in the Arctic? And why is it an important phenomenon in relation to erosion? Most of the time, the Arctic is very cold, so the entire landscape is frozen for most of the year, and only a thin layer of soil, thaws every summer, though this is what we call the active layer, thaws, and then freezes again. But underneath that is permafrost, this ground that's frozen all year round, and much of that ground, which has been frozen for more than 50,000 years, is full of ice. So as we are warming the Arctic, that freezing of thawing is eating away into the old, old permafrost underneath it, much of which is ice rich, which means that incrementally, year by year, the ice in the soil in the Arctic is melting, and that water isn't sticking around. So every year, the entire landscape subsides just a little bit because the ice that was there melts and goes away. How did you see the effects of erosion, subsidence, and sea level rise playing out in the Alaskan communities during your field work? I was up in northern Alaska in August and early September, and the town that I was in is called Utkiavic up until recently, it was named Barrow. So this is the northernmost point in Alaska. And what you see is that there are folks in bouldersers out every day building up berms along the coast to stop erosion from happening. This is an area where the coastline is eroding by 10 or more meters per year, very fast erosion, but local leaders also recognize that permafrost is subsiding. This is a problem that they have to deal with every day. Erosion is something that we all think about when you're standing on the coast because you can see chunks of the coastline falling into the ocean. Local rise is harder to see until it floods your community, but when I was in Utkiavic, we had the rare opportunity to climb down into an ice cellar. This is where community members store things like whale meat and because the ice cellars are in the permafrost zone, you would expect that they would be frozen, but along the coastlines, as sea level rises, saltwater starts to intrude into the permafrost that lowers the freezing point and that means that these ice cellars start to melt. How did you construct your model of how the Alaskan Arctic coast may be reshaped by the end of the century? Thealval projections are already available for Alaska, which is where we focus our study, but those sea level projections don't include what they call vertical land motion. So that's where we came in. There right now is no pan-Arctic or even regional estimate for how much permafrost is thawing, so we had to come up with one. And what we did was to say there are a lot of point estimates for permafrost subsidence. Thealval geos out with a GPS sensor every year for 10 years and measures the position of the land relative to a satellite. So we took as many of those as we could find and related them to the type of landscape that they were measuring. Landscape type, there are maps made from satellites on a pan-Arctic scale. So then we have the mapping of subsidence to landscape and then we projected that for it through time, assuming the same amount of subsidence, which is actually a conservative assumption because probably subsidence will accelerate as the Arctic warms. Okay, we have sea level, we have permafrost subsidence, now we need erosion. There's also no physics-based model for how coastlines in the Arctic may erode over the next century. However, here we could leverage work that was published a few years ago relating other quantities along the coast like how much wave action is hitting the coast, how warm is the coast. The quantities that you can easily measure relating that to present day rates of erosion. And that gave us at least a first order estimate for coastal erosion over the next century. What did your model show? By 2100, erosion and in addition taken together may transform Alaska's north slope, the Arctic coastal plain, causing 6 to 8 times more land loss than would occur if you only think about erosion. When you erode a coastline, you're putting sediment in the ocean, it can get chewed up by microbes. When you flood permafrost, you're giving the ocean access to a lot of organic carbon that then can get metabolized. Taking the north slope as a whole, we estimate that 8 to 11 times more organic carbon may get mobilized or otherwise disturbed over the next century with all these processes. The number being about half a pedogram, which is a significant amount. What are the infrastructure and community risks from erosion, subsidence and sea level rise? We took these projections of landscape change and compared them to existing infrastructure on the north slope of Alaska, everything from buildings and roads associated with communities to oil field, buildings to pipelines to legacy government infrastructure and assessed what fraction of this infrastructure would intersect with the ocean over the next 100 years. This has some important simplifications. First, infrastructure gets damaged before it's intersecting with mean sea level. A lot of the big damages that we should be concerned about over the next century in Alaska come from dorm searches and extreme sea level, which pushes much higher than mean sea level. Second, we're not taking into account any adaptation measures. But we find that without mitigating measures, 40 to 65% of infrastructure in present-day Arctic coastal plain, coastal cities and towns, stands to be damaged by rising sea levels plus erosion plus permafrost, subsidence, 10 to 20% of oil field infrastructure. But on the bright side, perhaps, oil field pipelines should be fine. What are the caveats or limitations of this study? The purpose of this study was not to say here's exactly what the coastline may look like, but rather to say the processes that we haven't been thinking about may be the most important ones. I wouldn't want somebody to look at our study and focus on a figure where we show outlines of the coast for the next century and to say, "Uh oh, my house looks like it's going to get damaged or not. Therefore, I should have some level of concern because I think there's so much that we don't know about the next 100 years that our results should mostly stand as a first-order wake-up call." How applicable are these results to other parts of the Arctic besides Alaska? In Canada, which used to be covered by the North American ice sheet, the history of that glaciation means that Northern Canada is still rising out of the ocean because of the ice weight that used to be there. Alaska, however, Russia places that weren't covered by ice sheets, particularly Siberia, these are places where this compounding effect will be particularly important. What are the policy or societal takeaways from your study? We need to think about climate hazards as adding on top of each other. And this is a different paradigm than sometimes we as scientists want to live in. We want to focus on a process and say, "We can quantify this process, but when you add processes on top of each other, one process plus one process plus one process doesn't equal three, it equals ten." And that's what we're saying. and showing here. Science is a team sport. This science happened because of my scientific team and because of the funding agencies that supported us, the National Science Foundation, local Alaska organizations that supported us doing field work in Alaska and native tribal corporations like UIC in Northern Alaska. This study is you could call it basic science looking at processes, but it matters for the communities and their voices don't get heard enough. Thanks for tuning into science sessions. You can subscribe to science sessions on iTunes, Spotify, or wherever you get your podcasts. If you like this episode, please consider leaving a review and helping us spread the word.