9 July, 2000

July 9, 2000

The lake is now 15 feet higher than when it broke out last year. The ice on the dam is fracturing. Yet, there is no sign of outburst. When considering this system from a very simplistic approach, one would think that the formula would be simple: The lake fills to a certain elevation, and in so doing floats the adjacent glacier dam, and drains beneath the glacier. There is some periodicity - outburst occurs every summer. It would make sense for a correlation to exist between the elevation of the lake surface and break out - i.e. the outburst should occur when the lake gets to a certain consistent height, which would be controlled by the rate of discharge of water into the lake from the drainage basin. Not so. PSU grad student Don Lindsay used a meteorological model to calculate the discharge of water into the lake for a given year using the precipitation amount in the drainage basin for that same year. He found that he was able to calculate pretty closely the expected amount of water that would drain into the lake. For example, if he used the data from a year that had a lot of snow in the surrounding mountains, then according to his model more water would drain from the surrounding slopes, and the lake had a larger volume of water. He was also able to validate the model from the measured results of last year. The upshot of all this is that he used this model to calculate the lake volume (and therefore surface elevation) from the precipitation data of a number of years. The date of outburst is known for each year for the past 35 years. There is no correlation between modeled discharge rate into the lake basin (and subsequent surface elevation) for a given year and breakout time. More complicated dynamics are involved than a simple analysis of ice flotation. These dynamics primarily involve what happens beneath the ice of the Kennicott Glacier.

The leading hypothesis now is that a central drainage conduit develops beneath the main stem of the Kennicott Glacier. Subglacial drainage in this manner was originally proposed by Hans Rothlisberger in 1972. It was later substantiated by John Nye in 1976, and Garry Clarke and Joe Walder in 1982. Conduit flow beneath glaciers is based on a physical and mathematical analysis of the interactions between water and ice under the influence of various pressure regimes. Conduits have never been observed directly - being beneath hundreds of feet of ice makes them difficult to see. When the conduit that is proposed for the Kennicott Glacier connects with the wedge of water beneath the ice dam, catastrophic drainage occurs. With this hypothesis, water does not leave the lake until the conduit system has tapped it. This implies that a hydraulic seal exists between the grounded ice of the ice dam and the subglacial drainage system (conduit system) of the main glacier.

There is a big question that still remains with this hypothesis: Does leakage occur? According to Rothlisberger's model it should not. However, some field observations suggest that leakage could be occurring. PSU grad student Don Lindsay suggests that leakage occurs with Hidden Creek Lake. Again, he used his meteorological model to make this argument. For last year, he used the model to show that about twice as much water flowed into the lake basin than was shown to be present at outburst. This means that when outburst occurred, there was an equal volume of water that was unaccounted for. Don hypothesizes that it leaked beneath the ice dam and into either a layer of gravel beneath the glacier, or into a cavity system beneath the glacier. He further argues that similar data has been shown for other outburst lakes that have been studied. He also suggests that catastrophic drainage has not occurred yet this year because leakage is relieving some of the high hydraulic pressure that must exist along the interface between the lake-bed and the glacier bed. However, as with most computer models, there is likely error in the model he has been using. The actual hypsometry (shape) of Hidden Creek Lake is not know very well, and as a result an accurate volume for the lake basin has been difficult to determine. It is not known how much water enters the lake from the adjacent glacier. Nor is it known how much water from the lake "wedges" beneath the glacier and permeates the glacier. So, the two PIs most interested in knowing the answer to this question aren't buying into leakage yet - they are sticking with the conduit theory (Andrew Fountain and Joe Walder). Hopefully the question will be answered this year. The stream team is measuring the discharge of Hidden Creek into the lake, so the expected volume will be measured and not modeled (which Don has been doing). Further, Suzanne Anderson will hopefully find a natural tracer in the water of Hidden Creek Lake that is different enough from the glacial water of the main body of the Kennicott Glacier to either prove or disprove leakage in this case.

The lake continues to rise…………..

Gathering data.....this photo shows a number of instruments that have been deployed to collect data for the various aspects of the project. The pole on the left (looks like a "T") is a GPS (Global Positioning System) antenna that is collecting coordinate data from a satellite that will ultimately give a very precise position of the pole. The pole with the orange square on top is a reflector stake - it is aimed at a bluff where surveyor Dennis Trabant will be recording any changes in its movement over the next couple of weeks. The borehole is the white area next to the green tarp. The PVC pipe holds electrical wiring that translates voltages from a pressure transducer in the hole to a data logger in the white plastic bucket. The pressure transducer is suspended in water that has filled the hole and will accurate record any changes in the water level.