We have analyzed the flow of polythermal McCall Glacier in Arctic Alas
ka. Using measurements of surface Velocity from the 1970s and 1990s, t
ogether with measurements of ice thickness and surface slope, we have
investigated both the present flow, and seasonal and long-term flow va
riations. Our analysis of the present flow reveals that (i) longitudin
al stress coupling is important along the entire length of the glacier
; and (ii) there is significant basal sliding beneath a 2 km long sect
ion of the lower glacier, This sliding exists year-round and it accoun
ts for more than 70% of the total motion there, We have developed a nu
merical model which shows that such a sliding anomaly causes an asymme
tric decrease in ice thickness. Accompanying this decrease in thicknes
s is a decrease in surface slope at the center of the anomaly and art
increase in slope up-glacier from it. Both effects are reflected in th
e observed surface profile of McCall Glacier. The longitudinal stress-
coupling length of McCall Glacier is three times the ice thickness, al
most twice that typical of temperate glaciers. This is a direct effect
of lower strain rates, which themselves are associated with the small
er mass-balance gradients of Arctic and continental glaciers. Long-ter
m variations in surface velocity between the 1970s and 1990s are expla
ined solely by the effects of changes in glacier geometry on the defor
mational flow contribution. This means that long-term variations in th
e spatial patterns of longitudinal stresses and basal sliding must hav
e been small, Seasonally, velocities reach their annual minimum in spr
ing and increase during the short summer melt season by UP to 75% abov
e mean winter values. However, the extra motion associated with the pe
riod of elevated velocities is only about 5% of the total annual motio
n. The speed-up, is due to an increase in basal sliding. This implies
that most of the glacier bed is at the melting point. The zone affecte
d by the melt-season speed-up extends well up-glacier of any moulins o
r other obvious sources for meltwater at the bed.