Abstract
The eddy covariance (EC) technique is a powerful tool for measuring
atmospheric exchange rates that was recently adapted by biogeochemists
to measure aquatic oxygen fluxes. A review of aquatic biogeochemical EC
literature revealed that the majority of studies were conducted in
shallow waters where waves were present, and that waves biased sensor
and turbulence measurements. This review identified that larger
measurement heights shifted turbulence to lower frequencies, producing a
spectral gap between turbulence and wave frequencies. However, most
studies sampled too close to the boundary to allow for a spectral
turbulence-wave gap, and will require a paradigm shift in how EC
measurements are conducted to remove wave-bias. EC fluxes have only been
derived from the time-averaged product of vertical velocity and oxygen,
often resulting in wave-biased fluxes. Presented here is a new analysis
framework for removing wave-bias by accumulation of cross-power spectral
densities below wave frequencies. This analysis framework also includes
new measurement guidelines based on wave period, currents, and
measurement heights. This framework is applied to sand, seagrass, and
reef environments where traditional EC analysis resulted in wave-bias of
7.2 ± 5.8% error in biogeochemical (oxygen and H) fluxes, while more
variable and higher error was evident in momentum fluxes (10.4 ± 20.5%
error). It is anticipated that this framework will lead to significant
changes in how EC measurements are conducted and evaluated, and help
overcome the major limitations caused by wave-sensitive and
slow-response sensors, potentially expanding new chemical tracer
applications and more widespread use of the EC technique.