The systematic under-estimation of solid precipitation due to wind is the most significant error in the measurement of winter precipitation and can be as large as 100%. This bias is highly dependent on the precipitation gauge and wind shield configuration used for the observations resulting in implications for climate data homogenization, hydrological modelling and water resource forecasting. Several national agencies including those in Canada, the United States, and Sweden have been using the Geonor T-200B accumulating precipitation gauge to measure winter precipitation for more than a decade in their national observation networks and research programs. This necessitates continued development of bias adjustment and homogenization procedures for this instrument.

During the 2003-2006 winter periods, 21 snowfall events greater than 2mm (as measured by the Double Fence Intercomparison Reference or DFIR) were observed at the Bratt's Lake intercomparison site located in the centre of the Canadian Prairies. The average wind speed during these snowfall events was greater than 5 m/s. Only 36% of the total snowfall was measured by the Alter-shielded Geonor T-200B precipitation gauge as compared to the DFIR. The catch efficiency of the Alter-shielded Geonor gauge is shown to decrease exponentially with increasing wind speed such that the catch efficiency is reduced to less than 50% at a wind speed of 5 m/s. Adjustments using this exponential relationship result in a corrected accumulation that is very similar to that measured by the DFIR.