Pine Island Glacier on the Move

Pine Island Glacier, Antarctica, is the focus of a large observational effort to better understand how glaciers and floating ice shelves interact with the ocean.

Pine Island Glacier (view is to the north, ocean in the top left) with crevasses and large crack extending from the east (right) to the west (left) as seen from aboard NASA's DC-8 research aircraft in October 2011. Credit: Michael Studinger/NASA

Scientists, pilots, technicians, and students working with NASA’s IceBridge and NSF’s Antarctic programmes tried hard for several years now to reach this glacier, set up a base, and drill through the 400-600 m thick ice shelf to reach the ocean. The data from these gargantuan efforts will reveal physics of ice-ocean interactions. This process is poorly represented in the climate models that are used to project past and present climates into the future. Harsh and hostile conditions cut these efforts short today, again, as reported by OurAmazingPlanet.

The expedition leader, NASA’s Dr. Bindschadler wrote today, that

A decision had been made by NSF the day we left McMurdo that if the helos were not able to be flown to PIG by Saturday, January 7, this year’s field work would be cancelled … We worked through our cargo—some had not been seen for two years when we tested our equipment at Windless Bight—preparing for either helos or the Twin Otter to start moving us onto the ice shelf. Neither came. Weather worsened.

Despite this dramatic turn of events, skies were clear over Pine Island Glacier today as they on New Year Jan.1, 2012. Two MODIS images show detailed features at 250-m resolution. I here show the near infra-“red” signals that the satellite receives (865 nm). The dark ocean reflects little of red (low reflectance) as it is all absorbed while the bright snow and ice reflects lots of red (high reflectance). Recall that the color “white” looks white, because it reflects all colors into our eyes including red, while “black” absorbs all colors, so none are left to reach our eyes.

Pine Island Glacier and Bay, Antarctica on Jan.-1, 2012 as seen by MODIS Terra, notice the whitish crack near the center of the image.

I show lots of the near infra-“red” as, well, red, and I color little red as blue. I chose the colors of the “crayons” to do the coloring. The technical term for this is contouring. Formally, I am depicting a function f=f(x,y) where f is the amount of red and x and y are locations east and north, respectively.

Pine Island Glacier and Bay, Antarctica on Jan.-12, 2012 as seen by MODIS Terra, notice the whitish crack near the center of the image.

They almost look the same, don’t they? If they were identical, then the difference would get zero. Except, glaciers move, especially this one. It is also about to spawn a large ice island. A crack was first reported in Oct.-2011 by scientists aboard a DC-8 of a NASA Icebridge flight. This crack is also widening as, I speculate, the front moves faster seaward of the crack than it does landward. My question is if I can see movements in these easily accessible public MODIS images. And my first answer, to be refined later, is 80 meters per day plus or minus 50%:

Difference of reflectance by subtracting Jan.-1 reflectances from those on Jan.-12, 2012. Very dark red colors show large positive numbers, meaning that the ice occupies a place on Jan.-12 that was water on Jan.1.

I am neither a glaciologist nor a remote sensing person, so I may be running a few red lights differencing two images and assign meaning to it. For example, I estimate the speed at which the front of the glacier moves by dividing the width of the very dark thick red line (about 1 km wide) by 12 days to get 80 meters per day or 3.5 meters per hour. The error here is at least 2 pixels (500-m), about half the estimated speed. My assumption here is that the high reflectance on Jan.-12 at a location with a low reflectance on Jan.1 means that the “bright” glacier has moved to a place that was “dark” ocean before. There is more to this, but I have to start somewhere.

Incidentally, Dr. Bindschadler, the leader of the current Pine Island field project who had to leave the base camp near Pine Island Glacier today, is the very person who wrote a wonderful peer-reviewed paper in 2010 with the title “Ice Sheet Change Detection by Satellite Image Differencing.” I will need to study it more closely … along with the vagarities of field work in polar regions.

It is difficult to get data from the field as opposed to data from remote sensing or modeling. This is especially true for remote and hostile locations the ice and the oceans interact. It is frustrating to be sent home early because of inclement weather and the very narrow window of opportunity when the few available helicopters and planes can fly or the ships can sail near Antarctica and Greenland.

EDIT Jan.-13: The National Snow and Ice Center estimated speeds of Pine Island Glacier as determined from two LandSat images from 1986 and 1988:

Contours of glacier speeds in meter per year of Pine Island Glacier from 1986 and 1988 LandSat Imagery, National Snow and Ice Center

These speeds are very different, 2-3 km per year versus 1 km in 12 days. The former estimate is made from 2 carefully geolocated images 2 years apart without a crack across the floating glacier, while my estimate yesterday is more noisy, but it is for a segment of the glacier that is barely connected to it. Perhaps we should consider the segment seaward fo the crack a separate ice island that is moving with the ocean rather than the glacier?

One response to “Pine Island Glacier on the Move

  1. Pingback: Pine Island Glacier Ice Island 2012 Shoving Off | Icy Seas

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s