Quantized eddy accumulation with error diffusion: a new direct micrometeorological technique with minimal requirements

Atmospheric fluxes near the surface are key metrics for understanding the interactions between the biosphere and the atmosphere. There is an increasing demand for highly accurate flux measurements for species where fast-response analytical techniques are not available. This includes, among others, stable isotopes, oxygen, ammonia, nitrogen compounds, and bio-aerosols.

Here we introduce quantized eddy accumulation with error diffusion, a new easy-to-implement, high-accuracy eddy accumulation method that is compatible with slow-response analytical techniques. Similar to relaxed eddy accumulation, this method involves sampling air at a constant flow rate and directing it into one of two containers, depending on the vertical wind velocity. The flux is then calculated based on accumulated concentration averages over the flux averaging interval. However, unlike relaxed eddy accumulation, the new method is a direct method that does not require the empirical coefficient β. These developments were made possible by developing a new representation of conditional sampling at a constant flow rate as a quantization process of vertical wind velocity. Fluxes estimated with relaxed eddy accumulation were found to be biased due to sub-optimal quantization. To account for these errors, an error diffusion algorithm was developed, which made it possible to minimize the biases inherent in the quantization process, thereby allowing for accurate and direct flux estimates.

Quantized eddy accumulation with error diffusion is shown to achieve direct flux measurements with errors smaller than 0.1% of the reference eddy covariance flux. Additionally, this method enables an increase in the concentration difference in accumulated samples between updrafts and downdrafts without compromising accuracy, making it especially suitable for detecting smaller fluxes. It also provides improved accumulation volume dynamics, flexible accumulation intervals, and is less prone to errors from non-zero vertical wind velocities.

These new developments are especially useful for measuring small fluxes of elusive atmospheric constituents, particularly in the presence of measurement challenges such as instrument drift or frequency attenuation. A notable application is the accurate measurement of water stable isotopes, which enables the tracing of biological processes and the accurate partitioning of measured fluxes.