At the King Street ferry dock in downtown Burlington, the lake temperature the morning of the 15th was 33oF (0.6C). The air temperature over the southern portion of Lake Champlain was around 0oF (-18C). Thus, the vertical temperature difference in the first few meters above the lake was near 19C. This contributed to the arctic sea smoke or steam fog, and the steam devil activity. Based on the 15/1255z (7:55am EST) ACARS sounding, the lake to 850mb temperature difference was 23C. The 1-h Rapid Update Cycle (RUC) forecast (Figure IV-7) valid at 15/15z indicated a lake-to-850mb temperature difference closer to 25C. The lake to 700mb temperature difference was 32C based on the ACARS profile and around 31C based on the 1-h RUC forecast. This resulted in extremely unstable low-level conditions over the lake. The convective overturning resulted in ascent over the lake, and associated convergence from adjacent land areas to the east and the west. The deeper-layer instability and convective updrafts likely contributed to waterspout formation.
Figure IV-7. The (a) 850mb and (b) 700mb analysis valid at 15z on 15 January 2009, based on the 1-hr forecast from the Rapid Update Cycle (RUC) model. Analyses from the NCEP/Storm Prediction Center mesoanalysis archive.
The Rapid Update Cycle (RUC) and Global Forecast System (GFS) model vertical profiles (Figure IV-8) showed west-northwesterly winds of 15-30 kts from 2 to 5 kft. Winds below 2 kft were generally less than 10 kts. Assuming lifted parcels over the lake are roughly the average of the lake and near-surface temperature, lake induced convective available potential energy (CAPE) values were around 500 J/kg. The model profiles suggested equilibrium levels between 7000-9000 ft. Because the observed echo tops were only around 2500 ft, it appears this methodology for determining the lifted parcel temperature results in the parcel being too warm compared to reality. The 0h RUC sounding at BTV is also available for viewing here. If we assume a temperature over the lake of 0oF and a dewpoint of -5oF, the lifting condensation level is around 1100 ft. This seems reasonable, and would suggest cloud depths of 1000-1500 ft within the waterspout producing lake cloud band.
More explanation here:
http://www.erh.noaa.gov/btv/events/15Jan2009/wx.shtml