Photo by Ethan Parrish

Slip experiments

We have constructed a custom ring shear device to empirically study the mechanics of glacier slip and erosion. This new device, which resides in a large freezer at UW-Madison, can apply up to 400 kPa effective stress at effectively infinite displacements. The sample chamber walls are constructed from transparent acrylic so that physical processes at the ice-bed interface can be observed directly throughout the experiments. The device operates in either stress- or velocity-controlled mode, and the effective stress can be precisely modulated or held constant. We also use a direct shear device housed in the same freezer to test rate and state friction of both ice and sediment.

Using these shear devices, we slide ice over beds of unconsolidated sediment or rigid geometries to simulate the slip mechanics present at the base of soft- or hard-bedded glaciers, like those along the Siple Coast of Antarctica or Thwaites glacier, Antarctica. From these data, we constrain the fundamental physics controlling at motion the ice-bed interface and better understand the mechanics of basal slip, till deformation, sediment flux, and erosion.

Schematic of glacial erosion from Zoet et al., 2013
White light interferometer scan of lab generated glacial striation

Glacial Erosion

Glacial erosion is one the most powerful geomorphic processes on Earth’s surface. The rates at which glaciers erode are orders of magnitudes larger than rivers when compared by basin size. Glaciers primarily erode through two processes: 1) abrasion and 2) quarrying. Abrasion is analogous to sandpaper rubbing against the underlying rock; whereas quarrying occurs when a block of rock is detached from the bedrock in one segment. The mechanics of these processes as they occur beneath sliding glaciers have been studied for over 30 years but continue to be improved upon today.

We model subglacial quarrying using COMSOL by solving J integrals.  We also have built rigid beds that can be inserted into the ring shear to simulate quarrying in the lab. Using both the direct shear and the ring shear, we have slid debris-laden ice over both flat and shaped beds to constrain the abrasion law. We couple these abrasion measurements with high-resolution scans collected with a white light interferometer to estimate eroded volume to directly link independent parameters (water pressure, sliding velocity, vertical velocity, rock properties) for the first time solve the glacial abrasion law. We also have made new abrasion energy measurements using these experimental devices.

Glacial earthquakes from David Glacier, Antarctica from Zoet et al., 2012
Glacial earthquakes from NEGIS Greenland from McBrearty et al., 2020

Glacier seismicity

Glaciers are some of the noisiest environments in the world, at least if you listen with a seismometer.  One reason for this large amount of seismic energy is that many glaciers continually move at high speeds, effectively creating a constantly slipping fault between the glacier sole and the underlying bed.  The rate and magnitude of this seismicity is tied to many aspects of glacier dynamics, including ice-bed properties, tides, and subglacial hydrology.  Through the use of passive seismic techniques, we can learn about the properties that regulate glacier slip and deformation.

We install passive seismic instruments on glaciers around the world to better understand glacial slip by precisely monitoring seismic behavior.  We are currently analyzing passive datasets we collected from Thwaites Glacier in Antarctica, Saskatchewan Glacier in Banff National Park, Canada and Bench Glacier, Alaska, and we are developing a new detection method to better constrain the temporal and spatial patterns of slip.

Map of Green Bay Lobe drumlins from Zoet et al., 2021

Remote sensing of landforms

Using GIS to study glacial systems has been a popular tool for the past 20+ years, but only recently has wide spread high resolution LiDAR data become freely and publicly available for regions that were once glaciated.  Simply identifying the 2d geometry of glacial features is often insufficient for understanding the processes that formed them.  Because of this were are using various datasets with advanced spatial analysis scripts we’ve written in programs such as Python and MATLAB to scan across 10’s of thousands of identified features to test the spatial and 3D significance.  We are looking for ways in which the landforms vary in statistically significant manners to help us better understand what might be controlling their distribution.

We have mapped in detail the 10,000+ drumlins and other elongated features that fall in the footprint of the Green Bay Lobe (GBL), so that we can systematically interrogate their 3D morphology in relation to parameters such as flow history of the GBL, local bed rock types, and composition.  Using advanced spatial analysis techniques of our own development we have tested various drumlin formation mechanism to understand the relationship between subglacial slip and landform development.

Active and passive seismic profile from Green Bay Lobe Tunnel Channel from Zoet et al., 2019
GPR echogram from Icelandic drumlin from Woodard et al., 2019

Glacial Landforms

Glacial landforms are widespread in areas that were once occupied by glaciers.  The shape and form of the landforms reflects the interplay between glaciology and the geology of the region.  We use a variety of geophysical, remote sensing and engineering techniques to investigate affects glacier have on the landscape

The Green Bay Lobe has numerous glacial geomorphic feature formed by subglacial deformation and subglacial drainage events. One of the most striking features are the large Tunnel Channels that drained the Western side of the Green Bay Lobe. These large linear features are remnants of huge drainage events that vacated vast amounts of subglacially stored water from the base of the Laurentide Ice Sheet. These channels can be 100’s of meters wide and 10’s of meters deep.

To studies features such as these we construct glacial hydraulic models that indicate regions which are likely to have stored the subglacial water, and conducted a geophysical survey in Waushara county using active and passive seismic as well at GPR to better characterize the cross-cutting relationship between the Tunnel Channels and the local stratigraphy and bedrock (Figure). This work is and done in an attempt to better constrain the timing and mechanics of the discharge events. Large drainage events like this are likely analogs to current processes that are draining regions of Greenland.

We also use advanced statistical techniques such as Bayesian analysis to analyze high resolution terrain models we have created of glacier forefields to understand the affect geology has on glacial terrains.

Thin section of ice grains under cross polarized light

Ice rheology

Ice deforms at a given strain rate in response to an applied stress, but this deformation becomes quite complicated when ice resides near its pressure melting point or contains debris. Most of the ice near the base of glaciers contains debris (up to 40% by volume), but the mechanics of the debris laden ice’s deformation are poorly constrained. This is in part due to the fact that once ice is near 0 C liquid water lies at the grain boundaries of the ice and next to debris particles.  This liquid water along with the debris complicates the dislocation mechanics of the ice.

Because ice near the bottom of the glacier undergoes the largest deformation and because it is in contact with the base it is likely the most important ice to know its mechanical properties. Due to the importance of debris laden basal ice we use uniaxial ice deformation rigs to study the stress vs. strain-rate properties of debris laden ice as a function of stress, temperature, and debris content.

To aid our study of ice deformation UW-Madison has one of the worlds only Ice EBSD that allow micro-mechanical properties to be investigated.  In addition we also study the c-axis fabric of ice to understand its deformation on a courser scale.

Cryowave tank with ice block in place on the right side of the image

Cryowave processes

To study the processes active on cold coasts and within ice areas we have constructed a 3x1x1 m wave tank within the walk in freezer.  A plunging style wave maker connected to a variable speed and amplitude wave generator is used to create a range of waves. Temperature is controlled in the water and room using a series of heating and cooling sources. On the other end of the wave tank can either be placed a ramp to simulate ice berg erosion via waves.  Several cameras including a subaqueous camera are mounted to capture evolution of the sample.  Pressure transducers are mounted to record wave heights.