Aspirated and non‐aspirated automatic weather station Stevenson screen intercomparison

Abstract

In this study six wooden Stevenson screen configurations (five aspirated, one non‐aspirated) were evaluated at the Centre for Atmospheric Research Experiments in Egbert, Ontario, Canada. The field experiment was performed over a 1‐year period to evaluate aspirated and non‐aspirated Stevenson screen configurations used at the Meteorological Service of Canada’s Automated Weather Stations. The results show the non‐aspirated screen is warmer than the aspirated screen duct location by 0.11 °C on average with a significant increase in the maximum daily temperature of 0.22 °C due to radiant heating effects. Temperature differences up to 2.1 °C were observed between the aspirated and non‐aspirated screens with a mean increase of 0.46 °C following 30‐min periods of mean global radiation greater than 200 W/m2 and mean wind speed less than 2 m/s. Temperature differences between the two screens are more variable with standard deviations up to 0.3 °C for wind speeds below 2 m/s. The maximum daily temperature for the historic minimum and maximum thermometer locations and upper regions of the aspirated screen is also found to be significantly warmer during radiant heating conditions. The three locations within the aspiration duct provided very similar results with mean differences of −0.02 and −0.01 °C below the probe uncertainty. The aspirated Stevenson screen with motor off produced low overall bias with lower maximum daily temperatures and higher minimum daily temperatures compared to the aspirated screen. Comparison between the historic wooden frame and new aluminium mounting is not significant with the wooden frame cooler by 0.02 °C overall. Differences between the new aluminium and new plastic ducts showed similar results of 0.02 and 0.01 °C overall. The aspirated Stevenson screen configuration is recommended at new Automatic Weather Station installations where AC power is available.