On the application of spectral corrections to particle flux measurements
To study the environmental fate of nanoparticles, reliable measurements of particle fluxes in the atmosphere are of importance. The eddy-covariance (EC) technique can be used to calculate surface fluxes. In this study, the EC technique has been applied to calculate particle fluxes in Helsinki (Finland) and Cabauw (the Netherlands). For reliable estimations of the surface fluxes, particle flux measurements need to be corrected for attenuation at the highest frequencies. This attenuation is caused by the relatively long response time of scalar sensors and measurement set-up. The attenuation can be estimated using a theoretical or an empirical approach. Horst [Horst, Boundary-Layer Meteorology, 1997, 82(2)] developed a simple formula to estimate the attenuation, based on the empirical approach. The empirical approach relies on the assumption of spectral similarity between two scalars. In this paper the effect of the spectral similarity assumption is investigated. It is shown that in order to apply the Horst formula reliably, a decent estimate of the sensor response time is required. Furthermore, it is shown that in order to apply the empirical method, a fast sensor response is required. It is concluded that theory does not predict the position of the peak in the cospectrum well, which means that for measurement set-ups not yet operational, this requirement is not easily evaluated. The assumption of spectral similarity seems reasonable for particle fluxes and heat fluxes, when compared to similarity between fluxes of other scalars. An altered assumption of spectral similarity has been applied, where similarity is assumed only at frequencies higher than the peak frequency in the cospectrum. This assumption leads to a better estimate for the attenuation, when applied to the Helsinki data. It does not lead to an improvement for the Cabauw data set, due to the large response time of the measurement set-up.
- This article is part of the themed collection: ICEENN