Discussion and reassessment of the method used for accepting or rejecting data observed by a Bowen ratio system

Abstract

The Bowen ratio energy balance method often produces extremely inaccurate magnitudes of the flux due to resolution limits of the instruments. We analysed the criteria used for rejecting inaccurate data observed using a Bowen ratio system and the resolution limit of the sensors to analytically determine the reliable values of the Bowen ratio (β) and the latent and sensible heat fluxes. The formula used to calculate the error limit of the Bowen ratio (β) was corrected based on the theory of error analysis. An example was proposed for the common case with 0.2 °C resolution limit of temperature measurement and 0.08 kPa resolution limit of water vapour pressure measurement, to show the steps of accepting or rejecting data observed by a Bowen ratio system. The acceptance or rejection of data observed by a Bowen ratio system is a dynamic process, which should be performed based on the excluded interval of the Bowen ratio and the qualitative relationships among the data observed by a Bowen ratio system. The excluded interval of the Bowen ratio can be structured based on the accuracy of the sensors used. Data are excluded first if they do not satisfy the qualitative relationships between the vapour pressure difference, the temperature difference, Bowen ratio, and the available energy, whereas the data in the rejection region range of the Bowen ratio are excluded second. It is necessary to improve the accuracy of the temperature and humidity probes to improve the acceptance rate for data collected using the Bowen ratio system, apart from improving the observed precision of available energy. Copyright © 2013 John Wiley & Sons, Ltd.