Intercomparison of Automated Snow Depth Sensors at CARE During WMO-SPICE

Four types of automated snow depth sensors were tested at the Centre for Atmospheric Research Experiments (CARE) test site in Egbert, Ontario, during the World Meteorological Organization Solid Precipitation Intercomparison Experiment (WMO-SPICE). Testing was conducted over two winter seasons, and employed both ultrasonic and optical sensors. The intercomparison utilized a novel field configuration, with three sets of four sensors (three ultrasonic, one optical) evaluated at three separate measurement platforms. Artificial surface targets were positioned below the sensors at each platform.

Each sensor’s accuracy was assessed relative to co-located manual observations. The influence of environmental conditions was factored into the analysis. For ultrasonic sensors, temperature measurements are required to compensate for variations in the speed of sound with temperature. For applicable test sensors, the impacts of temperature measurement uncertainty, radiation shielding, and proximity to the test sensors were investigated. The response times for each sensor to incident snowfall were characterized for precipitation events over the duration of the assessment period. Specific events were identified within the dataset to assess the sensitivity of sensors to snowfall on bare (snow-free) targets.

The results of this study summarize the effects of sensor type, configuration, and environmental conditions on the availability, quality, and accuracy of automated snow depth measurements. Response times to incident snowfall were determined primarily by the measurement resolution of sensors as configured for the assessment. The results demonstrate that snow depth measurements varied across each surface target and measurement platform. This variation illustrates the confounding role of spatial heterogeneity when assessing sensor performance, and the broader challenge of obtaining representative snow depth measurements at a given location. Based on the assessment results, recommendations for the operational implementation of snow depth sensors are presented.