| description abstract | Monthly data from 206 stations for the period 1947?93 are used to examine characteristics of snowfall over the eastern half of the United States and relationships with precipitation and the maximum temperature on precipitation days. Linkages between snowfall and modes of low-frequency circulation variability are diagnosed through composite analyses, based on results from a rotated Principal Component Analysis (PCA) of monthly 500-hPa geopotential height. Results are examined for the 2-month windows of November?December, January?February, and March?April. The three dominant PCAs for each window capture regional components of the Pacific?North American (PNA), Tropical-Northern Hemisphere (TNH), and east Pacific (EP) teleconnection patterns. Two general snowfall regimes are identified: 1) the dry and cold upper midwest, Nebraska and Kansas, where snowfall is strongly a function of precipitation; and 2) the Midwest, southeast, and northeast, where snowfall is more closely tied to the mean maximum temperature on precipitation days. The PNA (the dominant circulation mode) and the EP pattern are both associated with strong snowfall signals, best expressed for November?December and January?February. Snowfall for the PNA over the southeast, midwest, and mid-Atlantic states increases (decreases) under positive (negative) extremes, when the eastern United States is dominated by a strong 500-hPa trough (zonal flow or weak ridge) with associated lower (higher) precipitation-day temperatures. Snowfall signals are more extensive under positive PNA extremes where the lower temperatures have a greater impact on precipitation phase. An opposing precipitation-controlled snowfall signal is found over the upper Midwest. The positive phase of the EP pattern, describing a western ridge?eastern trough, is associated with negative snowfall signals clustered over the midwest and upper midwest. Opposing signals are found under the midwestern trough?eastern ridge pattern of the negative mode. These signals are primarily precipitation controlled, which for the Midwest are counter to the climatological control by temperature. TNH snowfall signals are fairly weak except for March?April, when significant differences are found for the upper Midwest and from Missouri northeast into New England. No coherent trends are observed in snowfall or in the strength of the circulation patterns derived from the PCA. | |
