STRAINMETER MEASUREMENTS OF POSSIBLE SEICHES IN LAKE YELLOWSTONE

In July 2009, unexpected signals were recorded on the strainmeter in Grant Village on Lake Yellowstone. The signals occurred in periods of 11 minutes, which differentiated them from the much larger, lower frequency signal of Earth tides. A large wind event had also occurred earlier that same day, and the signals were additionally being picked up by strainmeters over 20 km away. Because changes in local water levels have been found to affect the local strain fields and wind is a typical forcing mechanism for seiches, a seiche was eventually considered the most likely cause for those signals. Seiches are standing waves in an enclosed or partially enclosed body of water. The motion of seiche waves are different from tidal waves because they don’t curl over themselves; they produce more of a sloshing movement consistent with the smooth oscillatory nature of the signal. Seiches have a very specific period depending on the shape of the basin in which they are contained.

Crustal strain modeling shows that to create a strain such as that observed at Grant Village, the amplitude of the seiche would have to be 3-5 meters. However, the actual wave at Lake Yellowstone is not detectable by direct observation. Seiches normally dampen off after about 5-6 cycles, and this signal can last for hundreds of cycles indicating an unusually high q-factor (a dimensionless quantity describing the loss of energy compared to stored energy). The expected period for this wave was calculated using Merian’s formula T = (2L⁄√(gh)) to be about 11 minutes, consistent with the actual period of the Grant Village signal, supporting the hypothesis that the signal is caused by a seiche.

Two years later, the signal is still being recorded frequently. In August 2011, two pressure gauges will be installed at Grant Village to measure the pressure changes of the lake. One will be placed above below water level and one above to correct for varying barometric pressure. The data will be collected at a frequency interval of 1 second and low pass filtered at 1 minute to remove wave noise. Verifying the cause of this signal will help to remove it from tectonic and other signals. Results will be given in the forthcoming presentation.