GRADIENT MEASUREMENTS WITH FIXED AND REVERSING TEMPERATURE AND HUMIDITY SENSORS ABOVE A THIN FOREST

Recent studies have indicated that the universal Monin-Obukhov flux-gradient relationships may not be applicable close to an aerodynamically rough forest surface. The ability to discern the limits of this theory in such situations largely depends on the ability to measure the very small gradients of temperature and humidity commonly occurring. The aim of this paper was to show how such gradients could be measured with sufficient accuracy. This was achieved by using a traditional fixed sensor system and a reversing sensor system. Systematic errors were quantified and minimized for the reversing sensor system. Measurements were carried out over a thin pine forest in Jädraås, central Sweden during summer 1987. Gradients were very small during daytime, with 90% of the potential temperature gradients in the range -0.014 to 0.021 K m-1 and a median value of -0.0064 K m-1. The corresponding specific humidity gradients fell between -9.0 × 10-6 m-1 and 1.1 × 10-6 m-1 with a median value of -3.1 × 10-6 m-1. Temperature and humidity differences over a height interval of 5 m had to be measured with an accuracy of 0.003 K and 0.0016 g kg-1 to obtain an accuracy of 10% in the gradients during typical daytime conditions. This accuracy could not be met with the fixed sensor system whereas the accuracy of the reversing sensor system was normally better than this. Errors in humidity gradients caused by the natural variation of wind speed with height could induce errors larger than 10% under conditions of large vapour pressure deficits, small humidity gradients and large wind speed gradients. It was concluded that, for studies of flux-gradient relationships above forests, gradients must be measured with reversing sensor systems and that, in addition, explicit account must be given of the errors caused by ambient conditions that vary systematically with height. © 1990.