1998 agu poster

Flow Rates, Age Constraints, and Accumulation Zone Observations From the Galena Creek Rock Glacier, Wyoming

Sarah K. Konrad, Neil F. Humphrey, Eric J. Steig, Doug C. Clark, Noel Potter, Jr., W. Tad Pfeffer, and Daniel Costello

The Galena Creek rock glacier, Absaroka Mountains, Wyoming, is the site of ongoing research on the flow dynamics of a valley-type rock glacier. Surface velocities, established through a variety of methods including surveying of long-term markers, aerial photography, and ground photography, vary from 0-0.85 m/yr. Removal of the overlying debris layer (1-2 meters thick on the upper rock glacier) allowed ice cores of up to 9.5 meters to be retrieved. The ice is stratified in apparently annual layers. Thin debris layers are likely associated with summer ablation. Several organic samples including pine needles were recovered from the cores, and provide AMS radiocarbon ages of the debris horizons. These ages help to determine both past accumulation and flow rates of the rock glacier.

The accumulation zone of the rock glacier is a relatively small snowfield at the head of the rock glacier. The snowfield is separated from the cirque headwall by a bergshrund. Coring in the lower accumulation zone reveals that snow from the 96-97 season had turned to ice by the summer of 1998. Presumably, the snow became supersaturated with water during the summer of 1997, and froze during the 97-98 winter. Snow from the 97-98 winter was wet, although not supersaturated. In the upper, steeper reaches of the accumulation zone, snow from the last three winters is still preserved as snow. Presumably, this snow does not transform to ice until either deeper burial or transport into the lower, saturated portion of the accumulation zone.

Introduction

Rock glaciers have untapped potential as paleoclimate indicators. The debris cover of a rock glacier insulates underlying ice, reducing ablation rates. Consequently, rock glaciers have greater longevity than similarly sized glaciers, and may preserve relatively old ice. In polar climates, isotopic records from ice cores provide tens of thousands of years of precision climate records. The ephemeral nature of glaciers within temperate climates makes such old ice difficult to find at lower latitudes. Although rock glaciers have been the subject of much research, most workers have focused upon surficial features. The lack of data from inside rock glaciers has led to much speculation and disagreement within the geologic community about their interior structure and genesis. Until the interior of a rock glacier is understood more fully, paleoclimatic interpretation of ice cores from rock glaciers can be done only in a limited fashion. A more precise understanding of how rock glaciers form, flow, and modify over time in response to climate fluctuations is necessary to be able to interpret the data. The ultimate goal of our research is to establish a model of rock glacier rheology and dynamics based on long-term surface and interior observations at the Galena Creek rock glacier. This particular rock glacier was chosen for this study because it is the site of a relatively large amount of past research (Potter, 1972; Barsch, 1987; and Fitzpatrick et al., 1995) and because it is accessible by four-wheel-drive vehicle. It is formed of debris of a single lithology (Tertiary volcanics), the small size of which makes excavation down to ice relatively straightforward.

Surface Velocity

We drilled over one hundred bolts into boulders on the surface of the rock glacier in 1997, and surveyed their locations with a theodolite from two survey stations to the east of the rock glacier. We resurveyed the bolt locations a year later (1998), resulting in a 3-D velocity field accurate to approximately .01 cm/yr.

Digital Elevation Map

We constructed the DEM by supplementing the 1997 bolt survey data with approximately 200 additional points across the rock glacier. Surface slopes, which vary from 10 to 35 degrees, were calculated using this data set.

Vertical Ice Thickness

We calculated the vertical ice thickness by using a plane-strain approximation and standard ice flow parameters. Input data includes measured velocity and surface slope in the direction of flow. Ice thickness reflects the thickness of deforming ice, which, according to our working model, is a layer of relatively clean ice overlying non-deforming debris within an ice matrix. This interpretation is supported by the fact that the calculated ice depth at the location of the 1995 ice core matches the depth that we were able to drill before hitting rock (15 meters).

Ice Flux

We calculated the flux per unit width of the rock glacier by using a plane strain approximation to calculate velocity with depth, and assumed no basal sliding. The flux is significantly reduced in the lower third of the rock glacier, especially where the rock glacier is narrower. This drop in downglacier ice flux implies that the rock glacier is not in steady-state, and possibly that the Galena Creek rock glacier actually consists of two independent, active, lobes.

Surface Lowering

In order to obtain the absolute vertical velocity of the surface, the change in elevation due to horizontal translation downslope must be subtracted from the surveyed vertical velocity. The result shows a lowering of the glacier surface varying from insignificant at the snout to almost 20 cm/yr near the cirque. This increase in ablation with elevation is opposite of what is commonly observed with regular glaciers, and reflects the insulating effect of the debris cover, which thickens downglacier. Thus the rock glacier has a proportionally very small accumulation area, which is confined to the debris-free cirque.

Flow Divergence

Flux divergence is calculated using the gradient of flux per unit width in the northing and easting directions, and identifies areas of thinning and thickening due to ice flow. In several locations, thickening or thinning of up to 2 cm/yr is predicted, identifying regions of compression or extension, respectively. This is an order of magnitude less than the observed vertical surface velocity, suggesting that ablation, rather than ice flow factors, is the dominant control of surface elevation.

We recovered small organic samples including pine needles and bark fragments from several locations. A needle removed from the top 20 centimeters of ice adjacent to the 1995 ice core site has a radiocarbon age of 1680 ± 70 yr BP (1556 ± 150 cal yr BP). Another sample taken from ice just below the cirque has the much younger age of 200 ± 40 14C yr BP (AD 1671 to 1953). The 1995 core location is approximately 750 meters from the cirque. If the needle were transported downglacier at a constant rate (steady state), the ice velocity would have been 0.5 m/yr, which is similar to the ice velocity currently measured at the 95-core location. However, since the surface of the ice is currently ablating (about 0.1 m/yr at the 95 core location), the needle and the ice encompassing it must have only recently arrived at the surface. Therefore, the calculated paleo-velocity of the rock glacier is not for the relatively fast moving surface, but rather that of some deeper, slower, flow path. Consequently, the surface of the rock glacier must have moved faster than it currently is moving. Another important consequence is that the rock glacier must have formed before the Little Ice Age. We will have dates for several other organic samples shortly.

What is the composition of the rock glacier?

The sketch to the left is our working model, which in large part is based on the stratigraphy reported by Haeberli et al. (1988) from a core through the Murtel I rock glacier in Switzerland. We are currently designing a drill with which we hope to drill through the rock glacier next summer.

What is the vertical temperature gradient of the rock glacier?

Estimates of surface temperature and geothermal gradient show that the basal temperature is near 0 degrees C. If the basal temperature is above zero, the rock glacier is most likely supported by ice-free debris. If the temperature is below zero, meltwater arriving from the surface could potentially freeze onto the base of the rock glacier or freeze interstitially in the basal debris. Last year, we placed a 10-meter vertical string of thermistors within the ice layer, and also set up data loggers to record surface temperature at various locations. Data from these instruments will be retrieved next summer.

How does the past behavior of the rock glacier differ from the modern?

We now have a good idea of modern rock glacier velocities and ablation rates, but very little knowledge of how the rock glacier behaved in the past. We can make some inferences by reconstructing past average annual temperatures (for the most part, colder during the last 3000 yr). Ablation rates consequently would be reduced, the rock glacier would have been thicker, and thus would have flowed more rapidly. Additional radiocarbon dates, together with ice-core stratigraphy, will allow us to develop a flow model to predict the past and future behavior of the rock glacier.