Snow Water Equivalent Intercomparisons during the WMO Solid Precipitation Intercomparison Experiment

During the WMO Solid Precipitation Intercomparison Experiment (SPICE), automated measurements of snow water equivalent (SWE) were made at the Sodankylä (Finland) and Caribou Creek (Canada) SPICE intercomparison sites during the northern hemisphere winters of 2013/2014 and 2014/2015. Both Sodankylä and Caribou Creek hosted a Campbell Scientific CS725 passive gamma radiation SWE sensor and Sodankylä hosted a Sommer SSG1000 snow scale. The CS725 measurement principle is based on the attenuation of soil emitted gamma radiation by the snow pack and this attenuation is directly related to the SWE. The SSG1000 measures the mass of the overlying snow pack directly by using a weighing platform and load cell. Each of these instruments has advantages and disadvantages related to installation and measurement. Manual SWE measurements were obtained at both sites on a bi-weekly basis over the accumulation/melt periods using bulk density samplers and these manual measurements are considered to be the reference for the intercomparison. Results from both SPICE sites showed that the CS725 generally overestimated SWE with the deviation from the reference increasing after the occurrence of melt in both seasons. The SSG1000 snow scale agreed quite closely with the manual measurements throughout the intercomparison periods and when compared to the CS725 measurements, highlights the post-melting deviation in the CS725 measurements. Since both intercomparison sites have sandy, very well-drained soil, it is hypothesized that water from the snow pack is infiltrating into the frozen or unfrozen soil where it can no longer be measured by a manual snow tube. However, this water still has an impact on the attenuation of gamma radiation and hence creates an instrument overestimation. To supplement the SPICE intercomparison, concurrent CS725 and manual SWE measurements were also made at the University of Saskatchewan Fortress Mountain Snow Observatory located in the Canadian Rocky Mountains during the 2012/2013 and 2013/2014 winters. Results show a much closer agreement between the instrument and the manual measurements, even during snow melt periods, at this non-sandy site. The closer agreement at Fortress Mountain confirms that the sandy soils at Sodankylä and Caribou Creek are contributing to the measurement bias of the CS725 and suggests that this instrument should be used with caution in these soil conditions.