Dr. Terence J. Hughes
Department of Earth Sciences/Climate Change Institute/University of Maine
Orono, ME 04469-5790, USA
1 April 2006

1992-2002
In 1992, my paper, "On the pulling power of ice streams," was published in the Journal of Glaciology. It introduced a geometrical solution to the force balance, instead of the traditional analytical solutions of the Navier-Stokes equilibrium or momentum equations. This allowed me to combine the longitudinal tensile "pulling" stress with basal shear and side shear to obtain a smooth transition from slow sheet flow to fast stream flow to buttressing shelf flow along any flowband of an ice sheet, allowing the ice surface to progress from convex for sheet flow to concave for stream flow to flat for shelf flow. This is what I had observed for the West Antarctic Ice Sheet in 1970, which led to my four ISCAP bulletins. My controlling variable was the ratio of basal water pressure to the ice overburden pressure, which quantified ice-bed coupling and ice-shelf buttressing, and controlled “life cycles” of ice streams. This ratio varies from zero to one as the ice overburden progresses from resting on bedrock to floating above bedrock in a pattern linked to bed topography. Ice streams that pull ice out of ice sheets is topsy-turvy from the conventional view that ice is pushed out because gravity pulls ice downward.
Fourteen years and several publications later, in 2006, I'm the only one using this method. In the conventional method, gravitational forcing tied to the ice surface slope is resisted only by the basal shear stress, which vanishes where my water-to-ice pressure ratio is one, such as under a floating ice shelf. However, ice shelves spread faster as their surface slope gets smaller, or even vanishes. This would make ice shelves more efficient than perpetual motion machines, being able to run on an empty gravitational "gas" tank. My geometrical force balance provides gravitational forcing from the height of ice floating above water, with ice-shelf spreading beyond its grounding line resisted by my tensile “pulling” stress, which is never zero in ice shelves. My gravitational “gas” tank is full.

In 1994, Palaeo-3 published a topsy-turvy paper by my wife Bev and me called, "Transgressions: Re-thinking Beringian glaciation." A big problem was to explain why East Asians couldn't enter North America until after 12,000 years ago, when they supposedly had a land bridge 1000 kilometers wide, teeming with game, and had the West Wind at their backs for 80,000 years. Aboriginies had crossed a stormy shark-infested sea to Australia over 60,000 years ago. Bev and I argued that a lobe of the Arctic Ice Sheet shot through Bering Strait during the last glaciation, blocked the land bridge, and pushed down the shallow continental shelf of the Bering Sea, which didn't rebound above sea level until 12,000 years ago, long after the ice lobe was gone.

I was then invited to the 1994 International Conference on Arctic Margins, held in the Russian port city of Magadan in the Sea of Okhotsk, to defend Misha Grosswald's (and my) ideas about former Siberian glaciation. By then, Misha and I were arguing for an ice sheet from Japan to Washington State on the North Pacific rim, as an extension of our Arctic Ice Sheet. I even co-chaired a workshop on this subject. Misha wasn't there, so I stood alone. The paper and the workshop deliberations were published in the Conference Proceedings in 1995. I riled enough people to spur field activities aimed at proving our topsy-turvy ideas were wrong. Julie Brigham-Grette’s work in Siberia is one example.

In 1995, Misha Grosswald and I argued in the Journal of Glaciology that the Arctic Ice Sheet was "Paleoglaciology's grand unsolved problem." We borrowed our title from Hans Weertman's letter to Nature in 1976 stating that the marine West Antarctic Ice Sheet, also grounded below sea level on a continental shelf, was "Glaciology's grand unsolved problem." These problems are “unsolved” because the notion that ice sheets can ground in the sea and be “marine” seems topsy-turvy. Yet they existed then and one exists now.

In 1996, I asked in Arctic and Alpine Research, "Can Ice Sheets Trigger Abrupt Climate Change?" The conventional view, one incorporated in CLIMAP, was that ice sheets were primarily passive components of Earth's climate machine. That's why CLIMAP chose to reconstruct global climates during the last glacial and interglacial maxima. Those were the maximum "cold" and "hot" reversible perturbations of a climate that was assumed to be fundamentally stable. I asked the topsy-turvy question, “What if Earth's climate is fundamentally unstable, always seeking but never finding some stable equilibrium?” Then we should investigate times of most rapid climate change because that is when the instability mechanisms are most dominant. Ice sheets are the components of Earth's climate machine that are the most unstable and the least permanent, so perhaps the fundamental instability mechanisms reside in them. When their changes are big enough and fast enough, perhaps they can trigger abrupt climate change. Since 1996, detailed climate records extracted from ice cores to bedrock at the summit of the Greenland Ice Sheet show rapid climate changes that are compatible with these ice-sheet instabilities.

In 1998, Oxford University Press published my book, Ice Sheets. I wrote the book to rescue glaciologists from a trap set by Dick Peltier. He argued in 1992 in Science that climatology didn’t need glaciology. Given the shrinking areal extent of ice sheets during the last deglaciation, he could reconstruct the vertical extent of ice from the changing loads of ice and water on Earth's surface computed from his model of mantle rheology and the global sea-level curves. Just as glaciologists had gained access to the center ring under the Big Top in the circus of global climate change, Dick would banish us back to the Midway where we would resume our customary act biting the heads off chickens and snakes along with the rest of the sideshow geeks. I pointed to Dick's Achilles heel. His model, based on the most sluggish component of Earth's climate machine, couldn't produce abrupt climate change. Glaciologists could, once we got over the mental block of tying gravitational forcing to the basal shear stress that resists slow sheet flow with its sluggish response to any kind of forcing, and tied forcing to the “topsy-turvy” pulling stress that resists fast flow in ice streams, with an immediate response to even minor perturbations in forcing. NASA glaciologist Bob Thomas demonstrated this admirably in 2004 to account for the nearly doubled velocity of Jakobshavn Isbrae in only one year. The bad news: Oxford published Ice Sheets using my preliminary disc, which had many misprints. Bev and I sent Oxford the final corrected disc after Ice Sheets was in-press.

A central theme in Ice Sheets was that all ice sheets have a relatively stable inner core surrounded by a relatively unstable outer periphery, where ice streams proliferate and manifest the instabilities. I proposed steady-state mechanisms of glacial erosion and deposition in the core that made a stronger first-order imprint on the landscape with each glaciation cycle, and transient mechanisms in the outer periphery that made a temporary second-order imprint that was either erased or reoriented during each advance and retreat of the unstable periphery. Allowing transient effects to override steady-state processes seems topsy-turvy. Yet, stadials and interstadials during a glaciation cycle correlate with unstable fluctuations of the periphery. Terminations occurred when these unstable transient mechanisms were able to penetrate the “stable” core and cause gravitational collapse of the whole ice sheet, something Peltier’s approach cannot simulate. Unstable mechanisms operated in ice streams and ice shelves that buttressed marine ice streams.
The bottom-up approach to reconstructing former ice sheets that I introduced in The Last Great Ice Sheets in 1981, and expanded upon in Ice Sheets in 1998 after I introduced the pulling power of ice streams in 1992, was used in diagnostic applications using basal thermal regimes deduced from glacial geology. This approach can also be used in prognostic applications that determine basal thermal regimes beneath the Greenland and Antarctic ice sheets today in order to predict their future behavior. These are topsy-turvy.

Diagnostic applications for reconstructing former ice sheets were used for slow sheet flow in the stable core and fast stream flow in the unstable periphery. First-order glacial geology produced in the core can be interpreted as revealing where the bed had been frozen, thawed, and freezing or melting, with freezing and melting zones consisting of a mosaic of frozen and thawed patches. The thawed fraction is then used to compute former ice elevations, based on bed traction variations linked to basal thermal zones. First-order glacial geology in the unstable periphery locates former ice steams in glaciated valleys and straits. Their profiles are reconstructed based on what fraction of their basal ice was supported by basal water pressure, instead of by bedrock or till.

Prognostic applications to assess the future behavior of present-day ice sheets calculate the thawed fraction under slow sheet flow and the floating fraction under fast stream flow, based the known bed topography, ice thickness, and ice surface slope along flowbands. Ellen Wilch and I did this for sheet flow in much of Antarctica, where these data were available. Our work was published in the Journal of Glaciology in 2000. Doug Reusch and I did it for stream flow in Byrd Glacier along one radio-echo flightline where the data were available. Antarctic Science published our work in 2003.