Frosty-Free November: Trends and Variability of November Snowfall and Temperature in the Lake Michigan Region

Previously, a climatological data set of wintertime temperature and snowfall since 1950 was composited for the Lake Michigan region. Data were examined for available long-term stations in order to explore the daily, monthly, and seasonal wintertime characteristics within six sub-regions to west and east of Lake Michigan, respectively. A clear feature of the evaluation was a marked decrease of November snowfall in recent decades. A similar decrease in snowfall has not occurred during the core winter months, but has been to some extent in the spring, resulting in an increasing fraction of the seasonal snowfall which occurs from December to February. While all six sub-regions have experienced a reduction in the November snow frequency, the change and impact are greater in the southern and eastern zones.

In order to contextualize the decrease in November snowfall, the relationships of snow frequency with monthly temperature, daily temperature thresholds, precipitation frequency, and large-scale tele-connection patterns have been examined. The November snowfall depends strongly on monthly temperature and the number of days with high temperature thresholds of 0 and 4.5 °C, respectively. The frequency of days with snowfall is also sensitive to temperature, with a decreasing sensitivity for the largest snowfall threshold of 15 cm. The frequency of precipitation doesn't correlate well with temperature, but the fraction of snow days to precipitation days depends very strongly on temperature and has decreased in recent decades.

The large-scale patterns of coldest and warmest Novembers illustrate clear W-E wave-train patterns, but aren't positioned geographically to project strongly onto the North Atlantic Oscillation or Pacific North American indices. This is much different for the core winter months, during which these indices and ENSO variability correlate pretty well with Lake Michigan temperature and snowfall anomalies. For November, however, the indices simply cannot explain the warming and recent decrease in snow frequency.