Arctic Workshop Logo 2002       INSTAAR logo


View Abstracts


Registration Submit Abstract Accommodations
Home Student Support Local Info

32nd Annual Arctic Workshop Abstracts
March 14-16, 2002
INSTAAR, University of Colorado at Boulder

Previous | Abstract Index | Next



PFEFFER, W TAD . INSTAAR - University of Colorado.
Meier, Mark F. INSTAAR - University of Colorado.
Krimmel, Robert M. USGS Tacoma.
Cohn, Josh . INSTAAR - University of Colorado.

Columbia Glacier is a ca. 1000 km^2 temperate tidewater glacier terminating in Alaska's Prince William Sound. Since 1982 it has retreated up its marine-grounded channel 12 km, thinned as much as 400 m in the terminus region (lowest 10 km), and maintained sustained annually-averaged flow speeds in excess of 25 m/day (9 km/year), making it presently the world's fastest glacier. Calving flux has increased along with flow velocity, rising from 1 km/yr prior to the onset of retreat to a present value of 8 km/yr. Retreat has been accompanied by activation of fast flow far upglacier, beyond the limits of the marine-grounded channel.

The retreat of Columbia Glacier has been documented by aerial photography and photogrammetric analysis, starting in 1974 and continuing to the present, with 124 photo flights made to date. Photogrammetric analysis of the photography gives 3-D positions of trackable features to 0.3 m accuracy and highly detailed sub-annual velocity and strain rate fields.

Following the pattern of Alaska's other major tidewater glaciers, retreat can be expected to continue until the terminus retreats to a position in the channel grounded at or above sea level - a point approximately 20 km upglacier from the present terminus. Retreat to this point requires removal of ca. 80 km^3 of ice. The present ice flux at the terminus is approximately 20 km^3/yr; this flux will be diminished in the future by declining thickness, although declining thickness may be partly offset by increasing flow velocity. Three simple estimates of retreat predict that the terminus will retreat to the head of the marine-based channel in 9 - 12 years.

Present observations and uncertainties raise several important and unanswered questions:

1. What better estimate can be made for future retreat? The present estimate of 9 - 12 years is based only on bounding assumptions about flow velocity and calving. Accurate numerical modeling is difficult in this situation because of the critical and essentially unknown basal sliding boundary condition and the likely important role of longitudinal coupling.

2. What will be the fate of calved ice? Most calved ice is presently held in the channel by the moraine shoal created by the advance of Columbia Glacier to its extended position (attained in the early 19th century and held until 1982). This shoal prevents most ice from floating out into Prince William Sound and into tanker shipping lanes. Erosion of the shoal by icebergs could allow larger and substantially greater numbers of icebergs into the Sound, threatening shipping. Conversely, the impounded floating ice is known to support significant horizontal stresses, and increased iceberg densities resulting from increased iceberg flux may create backstresses which help stabilize the glacier terminus.

3. How are calving and flow related? This critical question remains unanswered. Neither the calving process itself nor the cause-effect relation (if any) between calving and flow are at known with any degree of confidence; present viewpoints range from calving controlling near-terminus dynamics and terminus position to calving being simply dictated by conditions on unsupported ice cliff height at the terminus. A knowledge of calving/flow interactions is needed not only for predictive understanding of tidewater glacier retreat but also for insight into the dynamics of marine-based ice sheets such as West Antarctica.

4. How were upglacier regions of Columbia Glacier activated by changes in conditions at the terminus at the start of retreat? Whether calving controls dynamics or vice versa, it is at least evident that accelerated flow appeared first in the vicinity of the calving front and subsequently propagated far upglacier, even activating a previously stagnant tributary. This propagating influence extended far beyond the reach of changes in local thickness and slope, and was presumably accomplished through long-range coupling in subglacial hydrology. This question relates directly to the larger - and also poorly understood - problem of glacier sliding, and like calving, has a bearing on ice sheet as well as temperate glacier dynamics.


Previous | Abstract Index | Next

Copyright © 2001 INSTAAR, Univ. of Colorado