Effect of snow-covered ground albedo on the accuracy of air temperature measurements

Abstract

Abstract. Solar radiation is one of the main factors introducing significant deviations between thermometers reading and true air temperature value. Techniques to protect the sensors from direct radiative influence have been adopted almost since the beginning of meteorological observations. Reflected radiation from snow-covered surface can also cause extra warming to thermometers hosted in solar shields, not always optimized to protect the sensors from this further backward radiative heat transfer. This phenomenon can cause errors in near-surface temperature data series, with relevant impact on the quality of data records. The study here presented experimentally evaluates the effect of albedo radiation from snow-covered surface, on the accuracy of air temperature measurements. The investigation is based on evaluating temperature differences between couples of identical instruments positioned above ground covered by natural vegetation, being one in snow-free conditions and the other above snow-covered surface, at the same time in the same site in close vicinity. The work involved a representative number of different typologies of sensors and shields from different manufactures. A mountain site with appropriate field conditions, offering long-lasting snow presence to maximize data availability, was selected to host the experiment. Quantities of influence such as relative humidity, wind speed and direction, solar radiation (direct and reflected) were constantly measured. The effect was evaluated to range up to more than 3 °C for some typologies of sensors. Full data analysis is here reported, together with complete results. This main scope of this work is to report on an experimental estimation and method to evaluate and include this effect as a component of uncertainty in temperature data series for near-surface stations above snowy areas.

Non-Contact Thermometer for Improved Air Temperature Measurements

“…A number of experiments and studies have already investigated specific aspects of atmospheric air temperature measurements, quantities of influence, and uncertainty components. During the MeteoMet projects [ 13 , 14 ] some of these effects were evaluated mainly in terms of single components in the overall measurement uncertainty budget: sensor dynamics, direct and reflected radiation [ 15 , 16 ], self-heating of meteorological sensors [ 17 ], and wind and rain.…”