We caught up with Katie Leonard, a ﬁfth-year graduate student in Lamont-Doherty’s Ocean and Climate Physics Division, at the start of an eight-week research cruise to the South Polar Amundsen Sea aboard the RVIB Nathaniel B. Palmer. Communication was problematic since, as Katie explained, transmission disruptions could occur when penguins stood on ice ﬂoes to the north of the ship and interfered with the ship’s satellite access. Nevertheless, Katie was able to describe her current dissertation research to us.
Editor: You mention that you are interested in the way wind patterns redistribute fallen snow, and, in particular, how this interaction reshapes the surface of the Antarctic ice sheet. Can you tell us what initially drew you to the topic and in what direction your thesis (“Antarctic Snow Drift Processes”) is headed?
Sure. Drifting snow, which occurs at moderate to low wind speeds, is a very common process in Antarctica (people often comment that they can never see their shoes when they’re doing ﬁeldwork, since snow is perpetually skimming the ice’s surface), but it is very hard to quantify, since it is difficult to measure snow particles moving in the wind without either inﬂuencing the wind or melting the snow.
On short timescales, signiﬁcant mass transport of snow only happens at high wind speeds, so low wind speed transport—referred to as “drifting snow”—has largely been ignored. However, for sites such as Lake Vostok or the South Pole, on the central Antarctic Plateau, the wind is rarely strong enough to cause “blowing snow” conditions, whereas drifting snow is extremely common.
I became interested in this while I was on staff at Lamont-Doherty, working for Robin Bell and Michael Studinger on their Lake Vostok radar dataset. We determined that although wind speeds at the Antarctic Vostok station were not strong enough to cause much blowing snow, there was a significant amount of wind redistribution of snow, inﬂuencing the climate records in the Vostok ice core. I entered the Department of Earth and Environmental Sciences Ph.D. program and began working with Bruno Tremblay and Stan Jacobs to reﬁne a snow model to better represent low-wind conditions. These low winds cause deposition and snow erosion and they form snow dunes, sometimes carving features in these dunes called sastrugi, which are ridges or grooves made of snow.
I began collaborating with some researchers at the Swiss Avalanche Research Institute who were measuring and modeling snow saltation (from the Latin saltare “to leap”), one of the mechanisms through which snow moves close to the ground. I spent part of a summer in Davos working on their model, learning about snow characterization and how these researchers make measurements in a cold drifting snow wind tunnel in the Swiss Alps.
I joined a field team studying the giant icebergs near McMurdo Station, Antarctica, in 2006, bringing a number of snow measurement devices with me so as to compare the various instruments against one another. I was able to learn several new things about the process by noting the differences in results from instruments with various characteristics and sensitivities. I conducted another ﬁeld study in 2007 on a sea ice ﬂoe in the Bellingshausen Sea, looking at the spatial variability of drift under carefully monitored atmospheric conditions, including my own measurements of precipitation rates, using a new method that I am continuing to reﬁne during this current research cruise.
Editor: Could you explain to our readers why it is important to create accurate predictions of precipitation at a place like the Antarctic?
In some areas in Antarctica, the average snow accumulation is only a few centimeters per year. If winds consistently erode that snow, our estimates of how much snow precipitated from the sky will be too low. Accurately modeling precipitation over Antarctica is crucial to forecasting future precipitation responses to climate change.
Editor: If your goal is a precise precipitation rate, would more sophisticated means of measuring and modeling snow drift let you determine the amount of fallen snow at the Antarctic that was later redistributed by these wind conditions?
Editor: Do more accurate precipitation measurements help one predict the way in which glacial melting will affect sea levels in the future? I imagine that a higher rate of precipitation at the Antarctic (which, as you are discovering, has been partly masked by wind redistribution) would offset some effects of glacial melting.
That’s right. Glacial melting can raise sea levels when it enters the ocean. Another component of this same question is estimating the amount of snow that is lost from the continent to the Southern Ocean via coastal blowing snow each year. There is a general hope that precipitation over Antarctica will increase with global atmospheric temperature, leading to increased storage of water on the ice sheet, which would help mitigate sea level rise (but there is no deﬁnitive evidence yet that this is happening).
Editor: Tell us about the cruise you are currently on. Who are you collaborating with and what are you investigating?
Stan Jacobs, whom I work with at Lamont-Doherty, is the chief scientist aboard the RVIB Nathaniel B. Palmer. We are surveying the oceanographic properties of the Amundsen Sea, with a particular focus on water column freshening and ice shelf melting in the Pine Island Glacier region. There are several other oceanographers here on the ship, many of whom have been afﬁliated with Lamont at some point.
There is also a group of researchers from the British Antarctic Survey, who brought an unmanned submarine (called “autosub”), which will be mapping the cavity underneath the Pine Island Glacier’s ice shelf.
We’ll be studying the water properties at speciﬁc points with CTD stations, and we’ll also be deploying moorings to measure changes in the water properties at individual sites over the next two years. (CTD stations are occasions when the ship stops and we lower instruments over the side to measure conductivity [which corresponds to salinity], temperature, dissolved oxygen content and other properties as a function of water depth.)
Editor: Just now you referred to the concept of “water column freshening.” Is this a way of speaking about the increase in freshwater that occurs in the ocean due to glacial melting?
In a word, yes.... Mixing freshwater into the ocean leads to a decline in the salinity of that ocean water. The reduction in salinity near the coast of Antarctica has been so large since the 1950s that the only obvious source for so much freshwater appears to be the ice sheet. That is not the only possible source though: changes in sea ice, precipitation or the rate at which the wind blows snow off the ice sheet into the ocean may also contribute. A paper in Nature a few years ago suggested that the source of Southern Ocean freshening was a rise in precipitation over the ocean near the Antarctic coastline, which is part of the motivation for the ship-based precipitation measurements I’ve been making. Having a dataset of current precipitation at known locations on the Southern Ocean allows us to compare the snowfall I measure on the ship with various atmospheric models. Knowing how well those models are doing at predicting current precipitation allows us to evaluate how well they might be doing at times when we lack measurements.
Editor: I’ve read that shifting wind patterns are driving warmer water underneath the Amundsen Sea’s frozen ice sheets, causing some ice to melt. Is this in any way connected to global warming and should we be at all concerned?
I cannot comment on whether shifting wind patterns are responsible, but there does appear to be significant and possibly accelerating melting occurring underneath the Amundsen Sea ice shelves. It may or may not be directly associated with anthropogenic climate change (the ﬁrst deﬁnitive attribution of anthropogenic climate change to Antarctic atmospheric warming was just published a few months ago) but is certainly cause for concern. The West Antarctic Ice Sheet is a sizable mass of ice, and indications that it is melting at an accelerating rate have implications for sea level rise, though the timescale on that is very uncertain. We desperately need more observational data (such as the ocean water column properties measured during this cruise) to understand what is happening and why.