Abstract
Upper-air variables have been correlated with near-simultaneous precipitation observations from 35 winter storms that produced heavy snowfall in the eastern coastal region of the United States. Standard radiosonde observations (raob’s) were used to calculate upper-air variables at mandatory levels, while liquid-equivalent precipitation amounts were gathered from surface reports at five stations within the region. This study seeks to identify those synoptic-scale variables, individually and in combination with other variables that may be useful to estimate short-term precipitation amount. Using both simple and multiple correlation analyses, the selected variables (e.g., thermal and vorticity advections, wind speeds, moisture convergence and advection, divergence, etc.) were correlated with precipitation amounts occurring over 1-, 3-, and 6-h periods near raob times.
The best correlated upper-air variables (using the 6-h duration) are 850-mb temperature advection (0.48), 850-mb equivalent potential temperature advection (0.45), 250-mb temperature advection (0.43), and 300-mb vorticity advection (0.41). Adding a second variable increased the explained variance by 7%, but only a very slight additional explained variance was obtained by adding a third variable. The relative importance of upper-air dynamics to precipitation amount also varies by storm stage. For instance, vorticity advection is apparently more important at the beginning stage, while temperature advection at lower- and upper-tropospheric levels and upper-tropospheric divergence are better correlated near the conclusion of a storm.
Corresponding author address: Dr. Robert P. Harnack, Cook College, Rutgers—The State University of New Jersey, P.O. Box 231, New Brunswick, NJ 08903.
Email: harnack@rci.rutgers.edu
