DIRECT EVIDENCE FOR PERENNIAL ICE IN EASTERN NORTH AMERICAN COLD TRAPS

In the late 1800s, climatologist Edwin Balch visited 33 sites in eastern North America that were known or rumored to retain ice for longer periods of time than was usual for their surroundings. These cold traps, which Balch termed “glaciéres naturelles” or natural refrigerators, were found in ravines and gorges, taluses and boulder heaps, as well as in man-made wells, mines and tunnels from West Virginia northward into New England. Despite local reports to the contrary, no perennial ice was ever observed by Balch, although accumulated ice from the winter and spring clearly persisted for an unusually long time at sites where it was naturally or artificially protected from the sun and wind in the summer. Cold air coming from the base of talus slopes throughout the summer suggested the possible continued presence of ice deep in the talus, but this could not be confirmed. In the current study, we used temperature data loggers to monitor the thermal regime of cold traps in Pennsylvania and New York where ice exhibited unusual persistence: these sites included man-made mine shafts (“ice mines”), a talus slope, two “rock cities,” and a collapsed anthracite coal mine. Perennial ice was demonstrated at two of these sites for the first time during the 2014-2015 season. In both cases, deep man-made recesses allowed the accumulation of significant ice relative to the volume and configuration of the cold trap; in summer, thermal stratification of the air mass further minimized the introduction of warm air and limited the melting of ice. Ice did not persist at sites where the winter/spring accumulation was not sufficient to maintain thermal stratification of the air mass. Sufficient annual melting occurred at sites with perennial ice such that little, if any, continuous ice accumulation over longer periods of time would be expected in the current climate. Most locations in the northeastern United States where the mean annual temperature is lower than the surrounding region are unlikely to develop perennial ice in the absence of heavy winter snow accumulation and sub-freezing spring temperatures. Perennial ice development appears to be restricted to very specific sites with unique geographical features that protect accumulated ice and snow from sunlight and the incursion of warm air.