Wave processing parameters

This article provides an overview of the main parameters used in Nortek wave processing. These parameters control how the wave spectrum is estimated and how directional properties are calculated. A basic understanding of these settings helps ensure that the processed wave data are representative of the measured sea state.

Spectrum

When using Nortek processing software, you can select which measurement type is used to estimate the one-dimensional frequency spectrum. Depending on the available data, the spectrum can be based on pressure, velocity, AST, or an optimized combination.

When the optimized option is selected, the software uses the AST-based spectrum for bursts with less than 10% bad detects. If the number of bad detects exceeds 10%, the processing switches to the pressure-based spectrum in deployments shallower than 20 m, or to the velocity-based spectrum in deeper deployments.

The detection of bad detects is described in detail in AST time series quality control.

When using onboard wave processing on Nortek instruments, the spectrum type is automatically selected using the optimized approach described above.

Method

This setting specifies the method used for directional wave processing, i.e., how the directional wave spectrum is estimated from the measurements.

A detailed description of the available methods is provided in Principles of Operation - Waves.

Smoothing

This parameter controls the degree of averaging applied across frequencies in the spectrum. Specifically, it defines the number of FFT bins included in the averaging at each frequency.

Increasing the number of bins results in a smoother spectrum, while a lower number produces a more irregular (“spiky”) appearance. The total energy is conserved, but the distribution of energy across frequencies may be slightly affected.

Frequency range

This parameter defines the frequency range of the non-directional (energy) spectrum and therefore also limits the directional spectrum.

If the selected upper frequency exceeds the resolvable range, the software automatically applies the Nyquist frequency (sampling frequency divided by two) as the upper limit.

Updated