Episodic energy transfer process within a developing wave spectrum
The evolution of measured and modeled directional wave field spectra were examined and contrasted. Measured wave data was derived from buoy motions. A state-of-the-science model for deep water wave evolution was found to be inadequate in addressing statistically significant wave growth events that occurred after an atmospheric frontal passage near Mobile, Alabama on March 10, 1994. Following the frontal passage, two distinct wave fields existed, separated in frequency and direction-space. The post-frontal wave field had a developing component aligned with the wind's direction, the other wave field was a relict of the pre-frontal wind. A new technique, based upon Efron's (1987) bootstrapping technique, was developed to demonstrate that the spectral peaks associated with the two wave fields were statistically different. During the three-hour research period, there were 13 growth events in the relict wave field, while it was under the influence of an opposing wind field, which were anomalous under current wave growth theory. Using one-dimensional directional wave spectra, in conjunction with a principal component analysis, the differences between the measured and modeled spectral evolution were used to establish a pattern for a process to explain the 13 growth events. Analysis of the relationship between the changes in the wind's friction velocity and the percentage increase in the wave energy in the post-frontal wave field resulted in the development of a transfer function for the process. The new process was found to be aperiodic, but highly predictable. The combination of the spectral pattern and the transfer function fully defined the new process for this research period. The process was termed the Episodic Energy Transfer (EET) process. The addition of the EET process improved the wave model's capability to hindcast the relict wave field's growth and significantly improved the directionality of the wave field's evolution. The EET spectral pattern may be only applicable to this particular field data, the formulation of the transfer function is expected to have broader application in wind energy transfer applications. The procedures outlined in this research can form a basis to analyze other frontal passage events and determine if there is a similar episodic process occurring.