numerical broadband modelling of ocean waves, from 1 to 300 s: implications for seismic wave sources and wave climate studies

Thursday, 18 December 2014: 4:00 PM
Fabrice Ardhuin1,2, Eleonore Stutzmann3 and Lucia Gualtieri3, (1)IFREMER, LOS, Plouzané, France, (2)IUEM Institut Universitaire Européen de la Mer, Laboratoire de Physique des Océans, Plouzané, France, (3)Institut de Physique du Globe de Paris, Paris, France
Ocean waves provide most of the energy that feeds the continuous vertical oscillations of the solid Earth. Three period bands are usually identified. The hum contains periods longer than 30 s, and the primary and secondary peaks are usually centered around 15 and 5 s, respectively. Motions in all three bands are recorded everywhere on our planet and can provide information on both the solid Earth structure and the ocean wave climate over the past century. Here we describe recent efforts to extend the range of validity of ocean wave models to cover periods from 1 to 300 s (Ardhuin et al., Ocean Modelling 2014), and the resulting public database of ocean wave spectra (http://tinyurl.com/iowagaftp/HINDCAST/ ).

We particularly discuss the sources of uncertainty for building a numerical model of acoustic and seismic noise on this knowledge of ocean wave spectra. For acoustic periods shorter than 3 seconds, the main uncertainties are the directional distributions of wave energy (Ardhuin et al., J. Acoust. Soc. Amer. 2013). For intermediate periods (3 to 25 s), the propagation properties of seismic waves are probably the main source of error when producing synthetic spectra of Rayleigh waves (Ardhuin et al. JGR 2011, Stutzmann et al. GJI 2012). For the longer periods (25 to 300 s), the poor knowledge of the bottom topography details may be the limiting factor for estimating hum spectra or inverting hum measurements in properties of the infragravity wave field. All in all, the space and time variability of recorded seismic and acoustic spectra is generally well reproduced in the band 3 to 300 s, and work on shorter periods is under way. This direct model can be used to search for missing noise sources, such as wave scattering in the marginal ice zone, find events relevant for solid earth studies (e.g. Obrebski et al. JGR 2013) or invert wave climate properties from microseismic records.

The figure shows measured spectra of the vertical ground acceleration, and modeled result for the primary and secondary mechanisms using our numerical wave model. (a) Median ground acceleration power spectra (LHZ channel) at the SSB seismic station (Geoscope Network), for the month of January 2008. (b) Spectrogram of modeled ground displacement and (c) measured spectrogram.