7 May 2014
In vicious cycle, waves found to grow as Arctic sea ice declines
Posted by Nanci Bompey
By Jim Thomson
What makes an ocean? Water and salt, some would say. Vast expanses of blue horizon, others would claim. For me, it’s waves. That’s not just a career choice, although I am a physical oceanographer; it is an ingrained view of a water world. Call it a bias– I won’t argue.
In this context, the Arctic Ocean has hardly been an ocean. That is, until recently. The Arctic Ocean is covered in a few meters of sea ice most of the year which largely keeps the water from developing waves. The ice cover melts a bit in the summer at the southern edges, retreating back from the shores of Alaska, for example, and allowing a limited amount of vessel traffic in the area. It also allows some waves to form. That’s where it gets interesting (again, a bias).
Recently, I’ve analyzed the first known measurements of wave action in a portion of the Arctic Ocean called the Beaufort Sea, and what I’ve found is striking. The loss of Arctic sea ice from global warming is allowing for bigger waves which, in turn, are likely to further accelerate sea-ice loss and hasten the disintegration of Arctic Ocean shorelines.
In the past decade, the seasonal ice retreat has expanded dramatically, culminating in well over 1,000-kilometer (600-mile) stretches of ice-free ocean in the summer of 2012. That remarkable summer was presaged by the Office of Naval Research, which launched new programs in 2011 to study the marginal ice zone and the state of the Arctic Ocean. I proposed and was funded within these programs to study the relationship of retreating ice and advancing waves. The advancing waves part was my hypothesis at the time.
The major field measurement efforts in these programs are still to come in 2014 and 2015, but an opportunity came up earlier. Colleagues from the Woods Hole Oceanographic Institution (Mass., USA) were planning to redeploy a mooring in the Beaufort Sea in 2012, and they offered to add my wave gauge onto the mooring. This was a dicey proposition, since ice is not very forgiving with instruments. This wave gauge, however, could sit 50 meters (160 feet) below the surface (safely below most of the sea ice) and measure the waves using sonar beams, once the ice melted. I jumped at the opportunity.
A year later, the mooring was recovered, and a very precious data set was safe on dry land. I began with the basic processing I do on every wave dataset: quality control of individual data points, estimating wave heights and periods, and spectral analysis. Right away, something jumped out– wave heights of almost five meters (16 feet) during a day of September 2012. That’s a lot bigger than anything previously recorded up there. Of course, the ice had melted back a lot farther than anything previously recorded too. I checked the raw data points by hand, and I convinced myself that it was real.
Then I wrote my colleague Erick Rogers at the Naval Research Lab at the Stennis Space Center in Miss., who specializes in wave forecasts using computational models. I asked if five-meter waves were in the forecast– or rather, retrospectively, in the hindcast. They were, it turns out. At least, for that big-wave September day and many others, the model matched the data.
But, digging around, we found that the model did not match whenever ice was near the mooring, according to satellite data. That made sense; ice damps waves and changes the dynamics. Without ice, the wave dynamics should be the same as any other open ocean, where we already know that the wave model works well. However, the “open ocean” portion of the Arctic Ocean is a strange thing, because it constantly changes in size.
The changing size became the central point of our study. We analyzed our data for how wave sizes depended on the length of open ocean available and the strength of the winds, finding a mathematical relationship, called a “power law.” That’s very similar to the known dependence of waves in a bay or a lake on the “fetch” distance, which is how far the winds can blow over open water. The dependence we found included not only the short waves (sea), but also the long waves (swell).
The presence of swell in the Arctic was surprising at first, since generation of swell requires large distances, but that’s where it all came together. We found that waves in the Arctic Ocean are controlled by open water distances, not only for generation of short waves, but also for evolution of long waves.
Erick and I wrote up a paper on these new wave findings that was published online Monday in Geophysical Research Letters, a journal of the American Geophysical Union. It’s called “Swell and sea in the emerging Arctic Ocean.”
The interpretation and projection for the future is simple: less ice in the Arctic Ocean means more waves, including the kind of swells usually only seen in open oceans. Because those swells carry more energy and travel further through the ice than smaller waves, this potential build-up of wave activity could accelerate the breakup of Arctic sea ice, and may play a major role in driving the Arctic to become ice free in coming years.
Bigger, more powerful waves could also accelerate erosion of the Arctic Ocean coasts, which are already breaking down rapidly from the effects of climate change and melting of permafrost.
Now, it’s time to dig deeper on the processes of waves and ice interacting directly. We’re heading up north this year to get more data. I’m pretty sure there will be waves, both this year and many more to come.
Guest blogger Jim Thomson, PhD, is an AGU member and the Principal Oceanographer for the University of Washington’s Applied Physics Laboratory in Seattle.