Long-term glacial erosion of active mountain belts: Example of the
Chugach–St. Elias Range, Alaska
James A. Spotila*
Jamie T. Buscher
Andrew J. Meigs
Peter W. Reiners
 Department of Geosciences, 4044 Derring Hall, Virginia Polytechnic Institute and State University,

 Blacksburg, Virginia 24061, USA
Department of Geosciences, 104 Wilkinson Hall, Oregon State University, Corvallis, Oregon 97331, USA
Department of Geology and Geophysics, P.O. Box 208109, Yale University, New Haven, Connecticut 06520,
USA
ABSTRACT
An emerging paradigm that equates glaciers to ‘‘buzz saws’’
of exceptional erosional efficiency has been strengthened by shortterm (,102 yr) sediment yields from southern Alaska. New lowtemperature cooling ages from this area, the glaciated Chugach–
St. Elias Mountains, constrain long-term (106 yr) exhumation rates.
Vertically averaged exhumation rates reach ;3 mm/yr, but are an
order of magnitude lower than rates based on short-term sediment
yields. Whereas these exhumation rates are not exceptional for orogenic belts, denudation patterns are strongly correlated with the
distribution of glaciers, and erosion appears to keep pace with convergence and uplift. These findings imply a coupling between glacially dominated erosion and tectonics.
Keywords: mountain building, denudation, glacial erosion, (U-Th)/He
dating, tectonics, southeast Alaska.
INTRODUCTION
Short-term (102 yr) rates of denudation based on sediment yields
from temperate, glaciated catchments into fjords of southern Alaska
exceed 1 cm/yr and may be the fastest in the world (Hallet et al., 1996).
This observation suggests that glaciers have a powerful impact on
mountain building and landscape evolution and has helped fuel ideas
for the role of glaciers in orogenesis. Glacial erosion is generally considered more effective than fluvial erosion (Montgomery, 2002) and is
thought to be capable of keeping pace with rock uplift in active orogens.
Glaciers have been considered ‘‘buzz saws,’’ in which erosion matches
rock uplift and limits topography via balanced negative feedbacks (Brozović et al., 1997; Meigs and Sauber, 2000; Montgomery et al., 2001).
Rapid erosion by glaciers is also a major facet of the proposition that
global climate change has shaped mountain belts. Expansion of alpine
glaciers in the late Cenozoic may have produced incision, isostatic uplift
of mountain peaks, and widespread acceleration of exhumation and sediment production (Molnar and England, 1990; Small and Anderson,
1995).
These short-term erosion rates, however, are not representative of
erosion averaged over many glacial cycles at orogenic time scales (105–
107 yr). Transitory, high sediment fluxes are a reflection of extreme ice
discharge, loss of sediment storage, and a disequilibrium response to
post–Little Ice Age deglaciation (Koppes and Hallet, 2002). Estimates
of basin-wide, mean Holocene erosion rates based on sediment yields
derived from seismic reflection profiles are slower (#5 mm/yr, 650%
error) for the same southern Alaska mountains (Jaeger et al., 1998;
Sheaf et al., 2003). Glacial erosion is thought to be a function of icesliding velocity and is understood best at the scale of the ice-bed interface (Humphrey and Raymond, 1994; Alley et al., 2003), but the
physical processes of glacial erosion and its interaction with mountain
building are not completely known (Brocklehurst and Whipple, 2002).
There are also glaciated mountain belts where glaciers are not the primary influence on denudation pattern (Tippett and Kamp, 1995). Given
the role of erosion for strain partitioning, thermal structure, and topog*E-mail: [email protected].
raphy in orogenic belts (Koons, 1995), knowing how glaciers erode over
long time scales is a key to understanding the interaction of tectonics
and surface processes.
To better understand the role of glaciers in active mountain building, we have quantified the pattern and history of long-term exhumation
of the active, glaciated Chugach–St. Elias Mountains of southern Alaska
(Fig. 1). This range is the greatest source of sediment into the Pacific
Ocean from the Americas, has one of the top 20 (worldwide) riverine
sediment fluxes (the Copper River), and contains the greatest coastal
relief on Earth (Jaeger et al., 1998). Unlike lower-latitude ranges, this
range has been heavily glaciated during most of the past ;5 m.y. of
active orogenesis (Lagoe et al., 1993) and currently has tidewater glaciers because of high latitude and ;3 m/yr orographic precipitation
along the Gulf of Alaska (Jaeger et al., 1998; Meigs and Sauber, 2000).
The topography of the range is apparently consistent with the glacial
buzz-saw hypothesis, as mean elevation and slope-altitude distribution
are correlated with glacier equilibrium line altitude (ELA) (Meigs and
Sauber, 2000). Rapid rock uplift is a function of the collision and partial
accretion of the composite, oceanic-continental Yakutat microplate since
the middle Miocene at the syntaxial bend between the Aleutian subduction zone on the west and the dextral Fairweather fault on the east
(Plafker et al., 1994; Bruhn et al., 2004) (Fig. 1). The Yakutat block is
partially fixed to the Pacific plate, which is being subducted northwestward beneath North America at ;4.5 cm/yr (Plafker et al., 1994).
The microplate moves roughly parallel to the Fairweather transform and
converges into North America, forming a fold-and-thrust belt that absorbs ;3.5 cm/yr of northwest-southeast shortening (Sauber et al.,
1997; Fletcher and Freymueller, 1999).
METHODOLOGY
We combined radiogenic helium thermochronometry [(U-Th)/He]
with existing fission-track data to constrain the low-temperature cooling
history and exhumation pattern of the range. These techniques provide
estimates of cooling that are typically associated with exhumation from
several kilometers depth (Ehlers and Farley, 2003). Helium closure temperatures average ;70 8C for apatite and ;180 8C for zircon (Farley,
2000; Reiners et al., 2004), as compared to ;90–110 8C for apatite
fission-track dating (AFT). Multigrain apatite helium ages (AHe) were
measured at Virginia Polytechnic (610% error, 2s), and single-grain
zircon helium ages (ZHe) were measured at Yale University (68% error, 2s). Uncertainties are greater for ages measured on small crystals,
which are common in some lithologies present in the range (e.g., schist).
Ages were measured on 20 samples collected by helicopter along 2
transects across the orogen (Fig. 1). Four additional ages were measured
northwest of the collision zone and above the Aleutian Trench, as a
control for regional background exhumation.
RESULTS
Measured ages indicate variability in low-temperature cooling history (Fig. 1; Tables DR1 and DR21). The oldest AHe ages (37–28 Ma)
1GSA Data Repository item 2004084, Tables DR1 and DR2 and Figures
DR1 and DR2, is available online at www.geosociety.org/pubs/ft2004.htm, or
on request from [email protected] or Documents Secretary, GSA, P.O.
Box 9140, Boulder, CO 80301-9140, USA.
q 2004 Geological Society of America. For permission to copy, contact Copyright Permissions, GSA, or [email protected].
Geology; June 2004; v. 32; no. 6; p. 501–504; doi: 10.1130/G20343.1; 4 figures; Data Repository item 2004084.
501
Figure 1. Chugach–St.
Elias Mountains are at
syntaxial bend between
dextral Fairweather fault
and Aleutian subduction
zone in southeast Alaska
(shown in inset). Sample
locations are shown as
circles (apatite helium,
AHe age) and boxes (previously measured apatite
fission-track, AFT age).
Distribution of glaciers is
shown in white. Major
faults and plate motion of
Pacific and Yakutat plates
relative to North America
are also illustrated (Plafker et al., 1994). Rocks
of Yakutat plate are
mainly Cenozoic sedimentary deposits, whereas older terranes of North
American margin consist of metamorphosed
flysch and isolated plutons. CR—Copper River,
CSEF—Chugach–St. Elias fault, DF—Denali fault,
DRZ—Dangerous River
zone, FF—Fairweather fault, TF—Transition fault; NR—nonreset age.
are from samples in the northwest, outside of the immediate collision
zone. These samples have been cooler than ;70 8C since prior to onset
of the Yakutat collision in the middle Miocene (Plafker et al., 1994).
AHe ages from the northern part of transect A–A9 are younger (10–18
Ma), but also indicate considerable time since cooling through closure
temperatures. Age versus elevation relationships for samples from the
northern part of transect A–A9 and the control group are comparable
and imply 0.15 mm/yr exhumation during Oligocene through Miocene
(Fig. DR1; see footnote 1). Previously reported AFT ages from the
northern parts of transects A–A9 and B–B9 (spanning ;4 km relief at
Mount Logan) are older than AHe ages (K. Long and S. Roeske, 1992,
personal commun.; O’Sullivan and Currie, 1996). Each transect displays
a group of old AFT ages (26–40 Ma), which on transect B–B9 is in-
Figure 2. Apatite helium (AHe) and apatite fission-track (AFT) ages
vs. distance north of coast. Age vs. distance relationship is particularly well defined for transect A–A9 (see inset). Boxes show age
groups used for age vs. elevation plots in Figure DR1 (see text footnote one). NR—nonreset; age consists of amalgamated detrital ages
of individual grains that exceed depositional age of sedimentary
rock.
502
terpreted to represent a partial annealing zone. If the comparable ages
from transect A–A9 represent the same annealing zone, the ridges of
both regions have undergone similar magnitudes of exhumation since
the Eocene, despite their difference in mean elevation. Younger AFT
ages (ca. 4 Ma) occur below the partial annealing zone on B–B9.
Cooling ages vary systematically with distance from the coast (Fig.
2). Detrital AHe ages that are older than depositional age (unreset)
occur in samples from the syntectonic Yakataga Formation within 25
km of the coast. AHe ages from the southern parts of transects A–A9
and B–B9 are ca. 4–1 Ma, defining a zone of relatively rapid, recent
cooling within the core of the collision zone. The youngest AHe age
comes from Icy Bay to the south of Mount St. Elias (;5500 m) and
near fjords in which some of the most rapid sediment yields have been
measured (Hallet et al., 1996). Young ages are also observed near the
outlet of the Bering Glacier to the west (Fig. 1). AFT ages versus distance from the coast mirror the AHe pattern (Fig. 2). ZHe ages range
from Paleocene to early Miocene (63–26 Ma) and are generally not
reset. All samples dated with ZHe, including those that produced the
youngest AHe ages, had cooled below closure temperature prior to the
middle Miocene onset of collision (Plafker et al., 1994).
These data reveal clues for the exhumation of the Chugach–St.
Elias Mountains. The pattern of cooling ages implies that synorogenic
exhumation has been greater near the coast (Fig. 3). We infer rates of
vertical exhumation from these cooling ages, with the assumption of a
geothermal gradient of 30 8C/km (O’Sullivan and Currie, 1996). Accordingly, closure depths for AHe and ZHe are ;2 km and ;6 km,
although the AHe closure temperature is also adjusted for grain-size
dependence and cooling history (Farley, 2000). A closure temperature
of 110 8C is used for AFT ages. Given that ZHe ages are older than
the onset of collision and in many cases have not been reset, the average
Cenozoic exhumation rate for the range has been ,;0.5 mm/yr. Exhumation rates based on AHe and AFT ages are faster. AFT ages from
samples at the base of Mount Logan indicate exhumation at ;1 mm/
yr based on closure temperature, although the age versus elevation gradient predicts a slightly slower rate (O’Sullivan and Currie, 1996). Rates
based on AHe ages are variable, but are as high as ;3 mm/yr for the
youngest sample at Icy Bay. Average exhumation rates along transect
GEOLOGY, June 2004
Figure 4. Erosion (exhumation) rate vs. time scale of observation for
glacially dominated Chugach–St. Elias orogen. Rates inferred from
100 yr sediment yield are >1 cm/yr and possibly higher (Hallet et al.,
1996). Average, orogen-scale rates from Holocene sediment yield
range from 2 to 5 mm/yr (Jaeger et al., 1998; Sheaf et al., 2003).
Average exhumation rates based on apatite helium (AHe) ages on
transects A–A9 and B–B9 are ~1 mm/yr, but could be slightly higher
because of offscraping of nonreset Yakutat sediments and are thus
plotted as 1–2 mm/yr. Exhumation rates from zircon helium (ZHe)
are lower (<0.5 mm/yr), given that ZHe ages are older than timing of
collision.
Figure 3. A: Inferred exhumation for samples along transects A–A9
and B–B9. Vertical lines represent exhumation rate based on apatite
helium (AHe) sample age and 30 8C/km geothermal gradient, except
that one exhumation rate is based on apatite fission-track (AFT)
ages at Mount Logan, as indicated. Exhumation rate between samples is interpolated, as shown by dark gray shading. Interpolated
area is transformed into total exhumation volume per distance (or
area along profiles; km2) summed over 1 m.y. Shaded boxes represent total volume of tectonic influx per distance, based on convergence rate of 3.5 cm/yr and 10 km thickness of Yakutat continental
crust. Percentages of tectonic influx accommodated by erosional
outflux based on this rough calculation are indicated on both profiles. Elevation profiles along transects are also illustrated. B: Mean
annual precipitation and modern equilibrium line altitude (ELA)
along transect A–A9 (Péwé, 1975). These climate trends coincide
with age vs. distance (Fig. 2) and exhumation vs. distance (A) trends.
A–A9 and B–B9 are 0.4 and 1.2 mm/yr, respectively. These averages
are more than an order of magnitude lower than erosion rates calculated
from 100 yr sediment yields and are several times lower than (although
overlapping within error bars) erosion rates based on Holocene sediment yields.
Exhumation rates increase toward the coast, where glacier coverage is dense and low-metamorphic-grade crust is being scraped off the
colliding Yakutat plate (Fig. 1). The exhumation pattern thus correlates
with a southward increase in orographic precipitation, decrease in ELA,
increase in ELA range between glacial and interglacial times (Péwé,
1975), and decrease in topography (Fig. 3). Exhumation pattern does
not bear a simple relationship to the position or orientation of major
thrust faults. Exhumation, mean elevation, and/or rock uplift also inGEOLOGY, June 2004
crease from west to east. The greatest relief and youngest cooling ages
occur at the syntaxial bend south of Mounts Logan and St. Elias. Shortening may be focused in this area, given that the Yakutat crust east of
the Dangerous River zone consists of a thicker continental component.
This area also displays higher mean elevation and more extensive glacier coverage.
Variations in geothermal gradient associated with topography or
isotherm advection are not factored into these exhumation-rate determinations. Samples are almost exclusively from ridges and should have
been similarly affected by topographic effects on isotherm shape (Safran, 2003). Estimated exhumation rates would match short-term sediment yields only if the effective geothermal gradient beneath ridges is
,7 8C/km to unreasonable depths of .10 km. Advection of isotherms,
in contrast, would selectively affect rapidly exhumed samples, resulting
in slight overestimation of the exhumation rate. Inferred exhumation
rates could be slightly underestimated if relief has increased, given that
the cooling history of valleys is unconstrained. Nonvertical particle trajectories during exhumation are another source of uncertainty. If rock
from shallow crustal levels is preferentially scraped off the Yakutat
plate, fast exhumation could occur undetected by cooling ages. The rate
at which rock with unreset helium ages could be eroded is controlled
by convergence rate and geothermal gradient. We estimate the maximum possible undetected exhumation for the deforming edge of the
Yakutat plate to be 1.2 mm/yr for AHe and 3.5 mm/yr for ZHe (Fig.
DR2; see footnote 1), still lower than short-term rates.
We thus conclude that glacially influenced denudation rates measured over the ;106 yr time scale are significantly slower than ;102
yr time-scale erosion rates, which have previously been shown as transient and not sustained through glacial-interglacial cycles (Koppes and
Hallet, 2002). Erosion rates based on ;104 yr sediment yields are of
an intermediate magnitude, such that apparent denudation rates increase
and become more variable toward the present (Fig. 4). Additional data
are required to test this trend.
DISCUSSION
This long-term exhumation pattern has implications for the coupling between glacially influenced erosion and crustal deformation. Exhumation rates are no faster in this range than in other collisional mountain belts, including those dominated by fluvial or mixed glacial-fluvial
erosion (Tippett and Kamp, 1995; Burbank et al., 1996). The denudation
503
rates do not support the buzz-saw hypothesis or argue that glacial erosion is more effective than fluvial erosion. However, there are apparent
links between glacial climate and orogenic denudation pattern. The most
rapid exhumation and highest topography occur in the south, where the
Yakutat plate is actively deforming and accreting and where precipitation is greatest. This association makes the Yakutat collision similar to
other accretionary wedge complexes, in which crustal trajectories and
patterns of mass transfer are influenced by local erosive power (Koons,
1995; Willett and Brandon, 2002). High topography at the orographic
divide within the Chugach–St. Elias Mountains traps precipitation,
which contributes to the extensive glacial coverage of the windward
flank. Greater ice flux and temporally variable ELA presumably translate to greater erosion, which may be sufficient to keep pace with tectonic rock uplift and impose a limit to mean topography. If such a
threshold of erosion exists, denudation is coupled to deformation, and
predictions of the buzz-saw hypothesis are met (Brozović et al., 1997).
A test of whether denudation keeps pace with collision is whether
erosional outflux equals tectonic influx in steady state (Willett and Brandon, 2002). On the basis of the Yakutat terrane’s #10 km thickness of
nonoceanic crust (Plafker et al., 1994) and ;35 km/m.y. convergence
rate, denudational outflux is 17%–48% of tectonic influx (Fig. 3). This
estimate does not include exhumation of unreset rock, does not take
into account convergence due to the terrane’s thick oceanic crust, and
ignores that some of the tectonic influx is recycled, syntectonic sediment. Although poorly constrained, the analysis implies that nearly half
of the inbound crust is accommodated by erosional mass transfer. Remaining tectonic influx may be accommodated by exhumation and penetrative deformation outside the immediate collision zone, lateral motion along the orogen’s axis (Bruhn et al., 2004), or subduction and/or
permanent accretion of the Yakutat plate. However, given the similar
magnitudes of denudational and tectonic flux, it is plausible that glacial
erosion helps keep pace with collision.
Glaciers may thus act as buzz saws in the southern part of the
Chugach–St. Elias Mountains, despite the lack of extreme exhumation
rates that had been predicted by short-term sediment yields (Hallet et
al., 1996). This result illustrates the efficiency of long-term, glacially
influenced erosion and highlights its importance as an agent of crustal
mass transport in mountain belts. If late Cenozoic cooling increased the
area dominated by such efficient glacial erosion, climate change could
have globally accelerated denudation (Molnar and England, 1990).
However, within mountain belts dominated by glacial erosion, the maximum rates of denudation are limited by rates of tectonic rock uplift.
This relationship is illustrated in southern Alaska and is explained by
negative feedbacks; ELA exerts a threshold on surface uplift, whereas
denudation that exceeds rock uplift results in lowering of topography,
reduction in glacier coverage, and slower rates of glacial erosion (Brozović et al., 1997). Thus, a logical coupling exists between glacial erosion and tectonic uplift, such that geographic and climatic setting (latitude, altitude, precipitation) dictate the landscape evolution and
erosional mass transfer in glaciated orogenic belts.
ACKNOWLEDGMENTS
We thank the reviewers, including John Jaeger, for their effort and insight.
Paul O’Sullivan, Ray Donelick, Greg Bank, and Fritz and Stan are thanked for
assistance. This work was funded by National Science Foundation grant EAR0001239.
REFERENCES CITED
Alley, R.B., Lawson, D.E., Larson, G.J., Evenson, E.B., and Baker, G.S., 2003,
Stabilizing feedbacks in glacier-bed erosion: Nature, v. 424, p. 758–760.
Brocklehurst, S.H., and Whipple, K.X., 2002, Glacial erosion and relief production in the eastern Sierra Nevada, California: Geomorphology, v. 42,
p. 1–24.
Brozović, N., Burbank, D.W., and Meigs, A.J., 1997, Climatic limits on land-
504
scape development in the northwestern Himalaya: Science, v. 276,
p. 571–574.
Bruhn, R.L., Pavlis, T., Plafker, G., and Serpa, L., 2004, Deformation during
terrane accretion in the Saint Elias orogen, Alaska: Geological Society of
America Bulletin (in press).
Burbank, D.W., Leland, J., Fielding, E., Anderson, R.S., Brozović, N., Reid,
M.R., and Duncan, C., 1996, Bedrock incision, rock uplift, and threshold
hillslopes in the northwestern Himalayas: Nature, v. 379, p. 505–510.
Ehlers, T.A., and Farley, K.A., 2003, Apatite (U-Th)/He thermochronometry:
Methods and applications to problems in tectonic and surface processes:
Earth and Planetary Science Letters, v. 206, p. 1–14.
Farley, K.A., 2000, Helium diffusion from apatite: General behavior as illustrated by Durango fluorapatite: Journal of Geophysical Research, v. 105,
p. 2903–2914.
Fletcher, H.J., and Freymueller, J.T., 1999, New GPS constraints on the motion
of the Yakutat block: Geophysical Research Letters, v. 26, p. 3029–3032.
Hallet, B., Hunter, L., and Bogen, J., 1996, Rates of erosion and sediment
evacuation by glaciers: A review of field data and their implications: Global and Planetary Change, v. 12, p. 213–235.
Humphrey, N.F., and Raymond, C.F., 1994, Hydrology, erosion and sediment
production in a surging glacier: Variegated Glacier: Journal of Glaciology,
v. 40, p. 539–552.
Jaeger, J.M., Nittrouer, C.A., Scott, N.D., and Milliman, J.D., 1998, Sediment
accumulation along a glacially impacted mountainous coastline: Northeast
Gulf of Alaska: Basin Research, v. 10, p. 155–173.
Koons, P.O., 1995, Modeling the topographic evolution of collisional belts: Annual Review of Earth and Planetary Sciences, v. 23, p. 375–408.
Koppes, M.N., and Hallet, B., 2002, Influence of rapid glacial retreat on the
rate of erosion by tidewater glaciers: Geology, v. 30, p. 47–50.
Lagoe, M.B., Eyles, C.H., Eyles, N., and Hale, C., 1993, Timing of late Cenozoic tidewater glaciation in the far North Pacific: Geological Society of
America Bulletin, v. 105, p. 1542–1560.
Meigs, A., and Sauber, J., 2000, Southern Alaska as an example of the longterm consequences of mountain building under the influence of glaciers:
Quaternary Science Reviews, v. 19, p. 1543–1562.
Molnar, P., and England, P., 1990, Late Cenozoic uplift of mountain ranges and
global climate change: Chicken or egg?: Nature, v. 346, p. 29–34.
Montgomery, D.R., 2002, Valley formation by fluvial and glacial erosion: Geology, v. 30, p. 1047–1050.
Montgomery, D.R., Balco, G., and Willett, S.D., 2001, Climate, tectonics, and
the morphology of the Andes: Geology, v. 29, p. 579–582.
O’Sullivan, P.B., and Currie, L.D., 1996, Thermotectonic history of Mt. Logan,
Yukon Territory, Canada: Implications of multiple episodes of middle to
late Cenozoic denudation: Earth and Planetary Science Letters, v. 144,
p. 251–261.
Péwé, T.L., 1975, Quaternary geology of Alaska: U.S. Geological Survey Professional Paper 835, 145 p.
Plafker, G., Moore, J.C., and Winkler, G.R., 1994, Geology of the southern
Alaska margin, in Plafker, G., and Berg, H.C., eds., The geology of Alaska: Boulder, Colorado, Geological Society of America, Geology of North
America, v. G-1, p. 389–449.
Reiners, P.W., Spell, T.L., Nicolescu, S., and Zanetti, K.A., 2004, Zircon (UTh)/He thermochronometry: He diffusion and comparisons with 40Ar/39Ar
dating: Geochimica et Cosmochimica Acta (in press).
Safran, E.B., 2003, Geomorphic interpretation of low-temperature thermochronologic data: Insights from two-dimensional thermal modeling: Journal
of Geophysical Research, v. 108, no. B4, 2189, doi: 10.1029/
2002JB001870.
Sauber, J., McClusky, S., and King, R., 1997, Relation of ongoing deformation
rates to the subduction zone process in southern Alaska: Geophysical Research Letters, v. 24, p. 2853–2856.
Sheaf, M.A., Serpa, L., and Pavlis, T.L., 2003, Exhumation rates in the St. Elias
Mountains, Alaska: Tectonophysics, v. 367, p. 1–11.
Small, E.E., and Anderson, R.S., 1995, Geomorphically driven late Cenozoic
rock uplift in the Sierra Nevada, California: Science, v. 270, p. 277–280.
Tippett, J.M., and Kamp, P.J.J., 1995, Geomorphic evolution of the Southern
Alps, New Zealand: Earth Surface Processes and Landforms, v. 20,
p. 177–192.
Willett, S.D., and Brandon, M.T., 2002, On steady state in mountain belts:
Geology, v. 30, p. 175–178.
Manuscript received 2 December 2003
Revised manuscript received 4 February 2004
Manuscript accepted 9 February 2004
Printed in USA
GEOLOGY, June 2004