Questions:

1. Is it possible that complex velocity fields induced by the change from flow through the narrow valley to the open coastal plain could create contoured moraines without surging? 
2. What do the velocity and strain fields look like for a Malaspina-type glacier? 
3. How does the model compare to the real Malaspina?

Flubber:

We used "Flubber" to mimic ice in the Malaspina Glacier as it flows through the Seward Glacier valley onto the Alaskan continental plain. Flubber is an elasto-plastic substance that flows plastically under low stresses but fails brittley under high stresses, much like ice.

Model setup (click to enlarge)	Our model consisted of a 75 cm-long, 12 cm-wide, 15 cm-high plexiglass "valley" inclined at an angle of 20º, supported by threaded poles attached to a plywood base.  At the base of the valley was a flat 65 x 65 cm plexiglass sheet to mimic the relatively flat continental plain along the Gulf of Alaska coast.  Above the "valley", a plexiglass "reservoir" was placed at a lower angle to slowly feed Flubber into the valley.  To constrain the flow of the "glacier" on the "continental plain", we constructed two 6 cm-high walls out of aluminum sheeting and lined them with 150-grit sandpaper. The walls were arranged such that the "glacier" could spread out from a width of 12 cm at the valley mouth to a width of ~36 cm at a distance of ~30 cm from the valley mouth.  To mimic the basic geometry of the Malaspina Glacier, the wall on the left (if viewed from the continental plain looking up-valley) was placed at an angle nearly parallel to the valley wall representing the contact between the Seward lobe and the Agassiz lobe. Bricks were used to hold the walls in place.
Model setup (click to enlarge)	We created four batches of Flubber, two green (using food coloring) and two white (the natural color of Flubber), and placed alternating green and white blocks of Flubber into the top of the valley. The alternating colors visually accentuate velocity gradients during flow. We propped-up the valley to an angle of 0º to prevent Flubber flow during setup (click to view starting conditions).  We created a grid of flow markers using both small flecks of colored chalk and magic marker dots (Sharpie pens work well) so that numerical anlyses could be completed (click to jump to analysis).  These markers had to be visible enough to allow them to be identified in digital photographs, yet small enough to minimize errors in our numerical anlyses. We found that black magic marker dots work best on the white Flubber, while light colored chalk works best on the green Flubber.
Model setup (click to enlarge)	Two digital cameras were mounted on tripods and placed in front of the valley at an elevated location to allow photos of the complete "valley" and "continental plain" to be taken. Two cameras were used so that we could take pictures at different magnifications.  During the model runs, digital photographs were taken every 30 seconds for the first 15 minutes of the run, every 1 minute for minutes 15-30, and every 2 minutes thereafter.  Each model run lasted approximately 1 hour.  These digital images were compiled into movies of the model runs (click to view movies), and gridded to produce velocty, strain rate, dilatation and vorticity fields for analysis (click to jump to analysis). We used an overhead projector as a light source to ensure constant, flat lighting of the Flubber for the best digital photographs.