Chronology of the Northern Yellowstone Ice Cap
Prepared for WHOI Geodynamics Field Trip, June 20-25, 2001
J.M. Licciardi (a), P.U. Clark (b), E.J. Brook (c), K.L. Pierce (d), M.D. Kurz (a), D. Elmore (e), P. Sharma (e)
The sensitivity and short response times of alpine glaciers (Paterson, 1994) make records of their fluctuations among the best terrestrial proxies available for identifying spatial and temporal climate variability. Previous work on records of alpine glacier fluctuations in the western US suggested synchronous millennial-scale paleoclimate oscillations in western North America and the North Atlantic region (Clark and Bartlein, 1995; Phillips et al., 1996). These studies were limited by the accuracy, resolution, and geographic coverage of dated terrestrial paleoclimate records.
The late Pleistocene Yellowstone Ice Cap was the largest independent glacier system in the western U.S. (Fig. 1). In the past 25 years, many innovative geochronological methods have been used to develop a record of fluctuations of the Yellowstone Ice Cap, and these have formed a foundation for understanding the timing of the last (Pinedale) glaciation for much of the Rocky Mountains. The initial chronology of Yellowstone glaciation was based on bracketing radiocarbon data from nearby localities in the greater Yellowstone region, and supplemented by calibrated obsidian hydration and weathering rind data on deposits in the West Yellowstone area (Pierce et al., 1976; Pierce, 1979; Colman and Pierce, 1981). A detailed U-series chronology of glaciation in the northern Yellowstone region was subsequently developed by Sturchio et al. (1994).
Most recently, our research group has applied cosmogenic 3He and 10Be to directly date a well-preserved moraine sequence deposited by the large outlet glacier that drained the northern Yellowstone Ice Cap (Licciardi et al., submitted). Our exposure ages establish a high-resolution record of late-Pleistocene fluctuations of this outlet glacier. The newly developed 3He and 10Be chronologies of Yellowstone Ice Cap are broadly synchronous with abrupt climate events in the North Atlantic region during the last deglaciation, reinforcing previous suggestions of climatic linkages between these two regions. These responses are consistent with our understanding of the influence of the Laurentide Ice Sheet as well as changes in North Atlantic climate on the climate of the western US during the last deglaciation.
SURFACE EXPOSURE DATING
The basic principal of surface exposure dating with cosmogenic nuclides (3He, 10Be, 14C, 21Ne, 26Al, and 36Cl) is simple (cf. reviews in Kurz and Brook, 1994; Cerling and Craig, 1994; Gosse and Phillips, in press). Cosmogenic nuclides are produced in surface materials primarily by cosmic ray neutron-induced spallation reactions (Fig. 2). The concentration of accumulated cosmogenic nuclides within surficial rocks is directly related to the time the surface has been exposed. The exposure age of a sample is simply the concentration of the cosmogenic nuclide divided by the effective production rate of the nuclide at the sample site, accounting for any loss due to radioactive decay.
Cosmogenic 3He is most commonly measured in olivine because this mineral is known to be helium-retentive. For measurement of 10Be, quartz is the preferred mineral phase because it is ubiquitous, chemically resistant, and has a simple target chemistry. Glacial deposits at Yellowstone contain both olivine-bearing basaltic lithologies and quartz-bearing granitic lithologies, providing the opportunity to measure both isotopes, in some cases on the same landform.
In the past decade, surface exposure dating using cosmogenic nuclides has proven highly successful in developing chronologies for glacial records. The new 3He and 10Be chronologies presented here show that surface exposure dating has a ~500 to ~1000 year resolution, which is sufficient to allow the use of alpine glacier records as indicators of regional and global climate forcing.
We sampled basaltic and granitic boulders from the outermost set of end moraines (Eightmile terminal moraines) that mark the Pinedale maximum position of the northern Yellowstone outlet glacier (Pierce, 1979) (Figs. 3 and 4), providing a rare opportunity to measure cosmogenic 3He and 10Be concentrations on the same landform. We also sampled granitic boulders from the recessional Chico moraines that are 7 km upvalley (Fig. 4), and from the recessional Deckard Flats moraines that were deposited another 47 km upvalley (Pierce, 1979) (Fig. 5). Finally, we sampled granitic boulders from a late-glacial flood deposit that occurs in the Yellowstone River valley within the limit of the Deckard Flats event (Fig. 5).
SUMMARY OF COSMOGENIC CHRONOLOGY
The Eightmile terminal moraines are dated independently by 3He and 10Be at 16.5 ± 0.4 3He ka and 16.2 ± 0.3 10Be ka (Fig. 6). The weighted mean 10Be age of the Eightmile moraines is ~2% lower than that obtained from the 3He ages, and the age ranges from the two isotopes are nearly identical. These observations indicate that the 3He and 10Be ages are concordant within the combined error of production rates, scaling uncertainties, and measurement error. The concordance of the 3He and 10Be ages is a highly significant finding, because it provides strong evidence that the production rates and scaling methodologies used in the age calculations from each isotope are reasonably accurate.
Although the weighted mean age of the recessional Chico moraines (15.7 ± 0.5 10Be ka) is slightly younger than that of the Eightmile terminal moraines, consistent with stratigraphic requirements, the difference in mean age of the two moraine complexes is not statistically significant. The ages therefore support field evidence that these two moraine-building events occurred within a short time period (Pierce,1979).
Cosmogenic 10Be ages obtained from boulders on the Deckard Flats moraines yield a weighted mean of 14.0 ± 0.4 10Be ka. Formation of the Deckard Flats moraines is interpreted to signal deglaciation of the Yellowstone Plateau, and a concomitant shift in the source region of the northern Yellowstone outlet glacier to ice caps in the adjacent Beartooth uplift and Gallatin Range (Fig. 3) (Pierce, 1979). Cosmogenic ages from the Deckard Flats moraines therefore suggest that disintegration of the Yellowstone Ice Cap occurred by ~14 10Be ka. The valley in the Deckard Flats region must have been largely ice-free during deposition of the late-glacial flood deposit, which is dated at 13.7 ± 0.5 10Be ka (Fig. 6), hence these data imply a significant and rapid retreat of the Deckard Flats ice front shortly after ~14 10Be ka.
CORRELATION WITH OTHER WESTERN U.S. RECORDS, AND IMPLICATIONS FOR MECHANISMS OF CLIMATE CHANGE
The newly-developed cosmogenic chronologies of the northern Yellowstone ice cap (Licciardi et al., submitted) and the Wallowa Mountains (Licciardi et al., 2000), together with previous results from the Wind River Mountains (Gosse et al., 1995a, 1995b), enable reconstruction of a composite history of alpine glacier fluctuations across a northern transect of the western United States (Fig. 7). These particular ranges provide some of the best-dated and most detailed chronologies of alpine glaciation available in the region. Moreover, these three ranges comprise a long transect that enables characterization of climatic oscillations over a broad region (Fig. 1). Finally, the cosmogenic 10Be-derived chronologies can be directly compared with minimal concern for potential inequalities between various isotopic timescales.
Last Glacial Maximum (LGM)
The 21.1 ± 0.4 10Be ka maximum Pinedale advance in the Wallowa Mountains corresponds with that of glaciers in the Wind River Mountains at 20.4 ± 1.0 10Be ka (Gosse et al., 1995a). The timing of the Pinedale maximum in the Wind River Mountains is supported by similar 36Cl ages on glacial deposits elsewhere in the range (Phillips et al., 1997). In the Olympic Mountains (Fig. 1), Thackray (2001) recently documented a radiocarbon-dated advance at ~19.1-18.3 14C ka (~22.6-21.7 cal ka). These events are correlative with each other and coeval with the LGM (~21 cal ka). In contrast, U-series data suggest that the margin of the large outlet glacier that drained the northern Yellowstone ice cap was restricted to less than 50% of its maximum extent between 22.5 and 19.5 ka (Sturchio et al., 1994).
Simulations with atmospheric general circulation models offer insight into this non-uniform response of the western US to climate forcing identified from the glacier records. A recurring feature of simulated LGM climate, for example, is a strong glacial anticyclone over the LIS in the winter, and a deflection of intensified winter storm tracks along the southern margin of the ice sheet (Broccoli and Manabe, 1987; Kutzbach et al., 1998). In the zone immediately adjacent to the southern margin of the LIS, the simulated climate is strongly influenced by the cold and arid, and hence glacier-inhibiting anticyclone. We suggest that the margins of the Yellowstone ice cap and southern Cordilleran ice sheet were retracted at the LGM because they were under the influence of the cold dry air mass of the glacial anticyclone downwind of the LIS. The maximum advance of glaciers in the Wind River Mountains during the LGM, ~150 km south of Yellowstone, indicates that the Wind River Mountains were just beyond the influence of the anticyclone, thus identifying the presence of a sharp climatic gradient between these two adjacent ranges. The LGM advance of glaciers in the Wallowa and Olympic Mountains suggests that the Pacific Northwest also lay outside the region dominated by the anticyclone, and received sufficient moisture to promote extensive glaciation at this time.
Heinrich Event 1 (H1)
The near full-glacial advance in the Wallowa Mountains at 17.0 ± 0.3 10Be ka correlates with, or closely predates, the maximum Pinedale advance of the northern Yellowstone outlet glacier, which is dated independently by 3He and 10Be at 16.5 ± 0.4 3He ka / 16.2 ± 0.3 10Be ka (Licciardi et al., submitted) (Fig. 6). Four recessional moraines (mean age ~17.6 ± 0.7 10Be ka) behind the Pinedale terminal moraine in the Wind River Mountains record the timing of readvances, or possibly stabilizations, of this ice cap (Gosse et al., 1995a). We note also that the Puget Lobe of the Cordilleran Ice Sheet (Fig. 1) reached its maximum extent sometime between ~17.4-16.4 cal ka (Porter and Swanson, 1998). These records provide evidence for broadly correlative periods of extensive alpine glaciation across a northern transect of the western US. Within dating uncertainties, all these events are coeval with the North Atlantic's Heinrich event 1 (H1), whose age is centered on ~17 cal ka (Bond et al., 1997).
Climate simulations suggest that the significant lowering of the LIS that accompanied a Heinrich event would result in a weakening of the glacial anticyclone over the ice sheet (Hostetler and Bartlein, 1999), thus allowing increased advection of moist Pacific air masses and cold northern air masses into the continental interior, promoting glaciation over much of the northern portion of the western US. The geologic evidence of widespread glaciation in the western US is remarkably consistent with model results (Hostetler and Bartlein, 1999), indicating that alpine glacier advances were a response to changes in atmospheric circulation due to Heinrich-related lowering of the LIS.
As noted previously, cosmogenic ages from the Deckard Flats moraines suggest that disintegration of the Yellowstone Ice Cap was complete by 14.0 ± 0.4 10Be ka (Licciardi et al., submitted). Glaciers in the Wallowa and Wind River Mountains were experiencing large-scale retreat at this time. These contemporaneous periods of significant deglaciation all occurred during the Bølling-Allerød warm interval (~14.7-12.9 cal ka; Hughen et al., 2000).
Model simulations also isolate the response of alpine glaciers in the western US to a warming of North Atlantic sea surface temperatures in the presence of a lowered LIS, such as occurred at the onset of Bølling (Hostetler and Bartlein, 1999). The simulated Wallowa, Yellowstone, and Wind River ice caps all exhibit negative mass balances relative to the LGM, consistent with the geologic records of significant retreat in these ranges. However, where dating is available, glacier retreat was underway shortly after H1 and before the onset of the Bølling, suggesting that the climatic influence of H1 was short-lived with respect to other controls affecting the climate of the western US during deglaciation (increasing insolation and concentrations of greenhouse gases).
Younger Dryas (YD)
The Titcomb Lakes moraines record a minor advance or standstill of cirque glaciers in the Wind River Mountains at 12.2 ± 0.5 10Be ka (Gosse et al., 1995b). Gosse et al. (1995b) noted previously that this event is coeval with the Younger Dryas (YD) cold interval (~12.9-11.6 cal ka; Hughen et al., 2000). In the Wallowa Mountains, the tentatively dated Glacier Lake moraine (11.2 ± 1.1 10Be ka) may have also formed during the YD. The Yellowstone region contains recessional moraines that may record events around this time period, but these deposits remain undated (Pierce, 1979).
Assuming that cold and dry conditions were transmitted to the western US during the YD, we would expect temperature-sensitive glaciers to advance, while those limited primarily by moisture availability should retreat. The relative sensitivity of glaciers to these climatic controls probably varied substantially across the western US (Hostetler and Clark, 1997), which would imply a spatially complex pattern of responses to YD cooling. The ranges that contain geologic evidence of possible YD-age advances (Wind River Mountains, Colorado Front Range, Wallowa Mountains) (Gosse et al., 1995b; Menounos and Reasoner, 1997) are among those controlled primarily by temperature, whereas those ranges that apparently did not experience YD-age advances (Sierra Nevada, Mount Rainier) (Clark and Gillespie, 1997; Heine, 1998) are comparatively more sensitive to precipitation. Available geologic evidence is therefore consistent with the expected responses of alpine glaciers to cold and dry conditions, and supports the hypothesis of transmission of a YD climate signal over an extensive reach of western North America.
The new data reported here, particularly combined with other data from the western US, show that surface exposure dating has sufficient resolution to evaluate the timing of alpine glacier fluctuations, and hence test the various mechanisms that influence climate change. We conclude that the record of alpine glacier fluctuations in the western US provides compelling evidence for broad synchronicity with abrupt climate events in the North Atlantic region during the last deglaciation, reinforcing previous suggestions of climatic linkages between these two regions. Modeling studies provide a range of plausible mechanisms that explain the inferred patterns of climate response, and suggest that the relative importance of individual controls on climate in western North America evolved through time during the last deglaciation. The LIS, however, probably played a central role at each stage in this evolution, and may have modulated the hemispheric-scale climate change originating in the North Atlantic through its gradual disappearance during the deglaciation.
Bond, G., Showers, W., Cheseby, M., Lotti, R., Almasi, P., deMenocal, P., Priore, P., Cullen, H., Hajdas, I., and Bonani, G., 1997, A pervasive millennial-scale cycle in North Atlantic Holocene and glacial climates: Science, v. 278, p. 1257-1266.
Broccoli, A.J., and Manabe, S., 1987, The influence of continental ice, atmospheric CO2, and land albedo on the climate of the last glacial maximum: Climate Dynamics, v. 1, p. 87-99.
Cerling, T.E., and Craig, H., 1994, Geomorphology and in-situ cosmogenic isotopes: Annual Review of Earth and Planetary Science, v. 22, p. 273-317.
Clark, D.H., and Gillespie, A.R., 1997, Timing and significance of late-glacial and Holocene cirque glaciation in the Sierra Nevada, California: Quaternary International, v. 38/39, p. 21-38.
Clark, P.U., and Bartlein, P.J., 1995, Correlation of late-Pleistocene glaciation in the western United States with North Atlantic Heinrich events: Geology, v. 23, p. 483-486.
Colman, S.M., and Pierce, K.L., 1981, Weathering rinds on andesitic and basaltic stones as a Quaternary age indicator, western United States: U.S. Geological Survey Professional Paper 1210, 56 p.
Gosse, J.C., Klein, J., Evenson, E.B., Lawn, B., and Middleton, R., 1995a, Beryllium-10 dating of the duration and retreat of the last Pinedale glacial sequence: Science, v. 268, p. 1329-1333.
Gosse, J.C., Evenson, E.B., Klein, J., Lawn, B., and Middleton, R., 1995b, Precise cosmogenic 10Be measurements in western North America: Support for a global Younger Dryas cooling event: Geology, v. 23, p. 877-880.
Gosse, J.C., and Phillips, F.M., in press, Terrestrial in situ cosmogenic nuclides: theory and application: Quaternary Science Reviews.
Heine, J.T., 1998, Extent, timing, and climatic implications of glacier advances Mount Rainier, Washington, U.S.A., at the Pleistocene/Holocene transition: Quaternary Science Reviews, v. 17, p. 1139-1148.
Hostetler, S.W., and Clark, P.U., 1997, Climatic controls of western U.S. glaciers at the last glacial maximum: Quaternary Science Reviews, v. 16, p. 505-511.
Hostetler, S.W., and Bartlein, P.J., 1999, Simulation of the potential responses of regional climate and surface processes in western North America to a canonical Heinrich event, in Clark, P.U., Webb, R. S., and Keigwin, L.D., eds., Mechanisms of Global Climate Change at Millennial Time Scales: Washington, D.C., American Geophysical Union, p. 313-327.
Hughen, K.A., Southon, J.R., Lehman, S.J., and Overpeck, J.T., 2000, Synchronous radiocarbon and climate shifts during the last deglaciation: Science, v. 290, p. 1951-1954.
Kurz, M.D., and Brook, E.J., 1994, Surface exposure dating with cosmogenic nuclides, in Beck, C., ed., Dating in Exposed and Surface Contexts: Albuquerque, University of New Mexico Press, p. 139-159.
Kutzbach, J., Gallimore, R., Harrison, S., Behling, P., Selin, R., and Laarif, F., 1998, Climate and biome simulations for the past 21,000 years: Quaternary Science Reviews, v. 17, p. 473-506.
Licciardi, J.M., Clark, P.U., Brook, E.J., Elmore, D., and Sharma, P., 2000, Cosmogenic 10Be chronology of late-Pleistocene glaciation in the Wallowa Mountains, Oregon, USA: Eos, Transactions, American Geophysical Union, v. 81, no. 48, p. F25.
Licciardi, J.M., Clark, P.U., Brook, E.J., Pierce, K.L., Kurz, M.D., Elmore, D., and Sharma, P., submitted, Cosmogenic 3He and 10Be chronologies of the late Pinedale northern Yellowstone ice cap, Montana, USA: to Geology.
Menounos, B., and Reasoner, M.A., 1997, Evidence for cirque glaciation in the Colorado Front Range during the Younger Dryas chronozone: Quaternary Research, v. 48, p. 38-47.
Paterson, W.S.B., 1994, The physics of glaciers: Oxford, Pergamon, 480 p.
Phillips, F.M., Zreda, M.G., Benson, L.V., Plummer, M.A., Elmore, D., and Sharma, P., 1996, Chronology for fluctuations in late Pleistocene Sierra Nevada glaciers and lakes: Science, v. 274, p. 749-751.
Phillips, F.M., Zreda, M.G., Gosse, J.C., Klein, J., Evenson, E.B., Hall, R.D., Chadwick, O.A., and Sharma, P., 1997, Cosmogenic 36Cl and 10Be ages of Quaternary glacial and fluvial deposits of the Wind River Range, Wyoming: Geological Society of America Bulletin, v. 109, p. 1453-1463.
Pierce, K.L., 1979, History and Dynamics of Glaciation in the Northern Yellowstone Park Area: U.S. Geological Survey Professional Paper 729-F, p. 90 p.
Pierce, K.L., Obradovich, J.D., and Friedman, I., 1976, Obsidian hydration dating and correlation of Bull Lake and Pinedale glaciations near West Yellowstone, Montana: Geological Society of America Bulletin, v. 87, p. 703-710.
Porter, S.C., Pierce, K.L., and Hamilton, T.D., 1983, Late Wisconsin mountain glaciation in the western United States, in Porter, S.C., ed., Late-Quaternary Environments of the United States. Volume 1. The Late Pleistocene: Minneapolis, University of Minnesota Press, p. 71-111.
Porter, S.C., and Swanson, T.W., 1998, Radiocarbon age constraints on rates of advance and retreat of the Puget Lobe of the Cordilleran Ice Sheet during the last glaciation: Quaternary Research, v. 50, p. 205-213.
Stone, J.O., 2000, Air pressure and cosmogenic isotope production: Journal of Geophysical Research, v. 105, p. 23,753-23,759.
Sturchio, N.C., Pierce, K.L., Murrell, M.T., and Sorey, M.L., 1994, Uranium-series ages of travertines and timing of the last glaciation in the northern Yellowstone area, Wyoming-Montana: Quaternary Research, v. 41, p. 265-277.
Thackray, G.D., 2001, Extensive early and middle Wisconsin glaciation on the western Olympic Peninsula, Washington, and the variability of Pacific moisture delivery to the northwestern United States: Quaternary Research, v. 55, p. 257-270.