An assessment of the Geonor T-200B used with a large octagonal double fence wind shield as an automated reference for the gauge measurement of solid precipitation

The homogenization of precipitation data observed by various national monitoring programs using a wide range of precipitation gauge/shield configurations is a difficult task. The problem is exacerbated for the gauge measurement of snowfall because the environment and the gauge/shield configuration have a high degree of influence on the accuracy of the snowfall measurement. The most significant systematic error in the gauge measurement of snowfall is wind induced under-catch. In 1985, the WMO initiated a solid precipitation measurement intercomparison in which 16 countries participated and over 20 gauge/shield configurations were tested. The study took place between 1986/87 and 1992/93 and predominately involved manually observed gauges. The reference standard for this intercomparison was the Double Fence Intercomparison Reference (DFIR) consisting of a large octagonal double wind fence and a manually observed Tretyakov precipitation gauge. Although the DFIR is a well suited reference for manual gauges, it does not work as well as a reference for automated gauges that have extended capabilities for high temporal resolution observations. For this reason, and in anticipation of further intercomparisons of automated gauges, an automated reference is required.

This study assesses the Geonor-T200B all-weather precipitation gauge installed inside a large octagonal double fence wind fence for use as a reference for the automated measurement of solid precipitation. This is accomplished by comparing the relative snowfall catch of Geonor gauges in large double fences to the DFIRs at Bratt's Lake SK Canada and Jokioinen Finland. Wind adjustment curves for the Geonor in the large double fence are developed based on the intercomparisons with the DFIR. The utility of this automated reference is examined for gauge intercomparisons at temporal scales ranging from 1 to 24 hours.