A SIMPLE METHOD FOR SPECIFYING SNOWPACK WATER EQUIVALENT IN THE NORTHEASTERN UNITED-STATES

Statistical regression models were developed to estimate snowpack wate r equivalent (SWE) using only meteorological variables available at Na tional Co-operative Observer Program (co-op) sites. These include the square root of snow depth, the number of previous consecutive days wit h the maximum temperature below freezing, the snowfall during the 24 h ours preceding the day of prediction, and the water equivalent of prec ipitation that fell during that period. Model development was performe d using data from the 15 National Weather Service offices (NWSOs) in N ew York and New England, where SWE is measured daily. In addition, geo graphically and seasonally grouped estimation models were derived, whi ch include a term for rain-on-snow events. Independent verification re sults showed that SWE at individual co-op stations in New York were we ll estimated by equations for at least one NWSO, with results comparab le to those from developmental sites. Model-described variation ranged from 60% to 81% for most January data, and from 54% to 90% for most F ebruary data. Relative confidence intervals decrease as the estimated SWE increases, with 67% of observed SWE values typically falling withi n approximately +/-50% and +/-15% of the specified SWE for 2.5 and 25 cm SWE estimated, respectively.