Niveograph Interpolation to Estimate Peak Accumulation of Snow Water Equivalent in Rocky Mountain National Park

Recent studies of trends in peak snow water equivalent (SWE) have underscored the value of long-term series of peak SWE. Since the late 1970s the US Department of Agriculture’s Natural Resources Conservation Service (NRCS) has used snow pillows at snowpack telemetry (SNOTEL) stations to measure the pressure of the overlying snowpack to provide a real-time daily record of SWE. The SNOTEL stations therefore provide direct observations of peak SWE whenever it occurs. However, snow-course SWE measurements recorded manually by the NRCS and its predecessor, the Soil Conservation Service, date back as far as the 1930s, but typically include only four SWE measurements per year, on the first of February, March, April, and May. While the April 1 measurement is typically assumed to represent annual peak SWE, this has been shown to underestimate the actual peak by as much as 12%.

To more accurately estimate peak SWE for snow-course sites, a procedure was developed to estimate daily niveographs (graphs of SWE versus time) for these sites using median daily niveographs derived from period-of-record daily SWE values for selected nearby SNOTEL sites. Recognizing that different physical processes drive the shape of the niveograph during the accumulation, peak, and melt phases, separate techniques were used to adjust the shape of the median niveographs for each of these phases. The procedure was tested by selecting two SNOTEL sites in Rocky Mountain National Park, Colorado, and treating them as though they were Snow course sites with just 4 SWE measurements per year. The estimated peak SWE values were then compared with the actual observed values. Results showed good agreement between estimated and observed peak SWE values, and an improvement over the assumption that April 1 SWE represents the peak.