An examination of coherence of directional wave time series from bottom-mounted pressure and current sensors

Abstract

This paper examines the coherence between kinematic variables from different types of bottom mounted instruments used for coastal directional wave measurements. Long term field data from PPP, PEMCM and PADV for several locations and wave conditions are analyzed for wave direction and coherence. Directional wave measurements analyzed from arrays of sensors assume that the time series of wave kinematics are statistically homogenous. Sensor time series should differ only in amplitude and phase. For real-world field measurements sensor data contains noise components and systematic errors and biases. Coherence between sensor pairs is a common method of quantifying the noise and error. Our results show very high coherence between individual pressure time series in short base-line pressure arrays. Slope components computed from the pressure array are used for directional analysis. We find slope coherence less than the scalar pressures but generally high. Coherence between U and V components of horizontal velocity measured by current meters should be very high because they are Cartesian components of the current vector. Our data show this is generally true for common E-M current meters. For acoustic Doppler (PADV) velocity meters the coherence was significantly lower. Coherence is potentially useful as a metric for comparing different sensor systems accuracy for directional estimates. It also is frequently used as a quality control test for automated data analysis. Non-acoustic systems generally exhibit high coherence when operating correctly. For acoustic systems, the question arises as how good must coherence be for acceptable data. We present exploratory analyses that examine then relationship between coherence and the quality of directional estimates.