 |
| Titel |
Freak waves in crossing directional seas: a laboratory experiment |
| VerfasserIn |
Suzana Ilic, Jamie Luxmoore, Peter McClintock, Victor Efimov, Aneta Stefanovska, Juana Fortes, João Santos, Rui Capitao, German Kolmakov, Csaba Pakodzi, Carl Trygve Stansberg, Ivar Nygaard, Nobuhito Mori |
| Konferenz |
EGU General Assembly 2013
|
| Medientyp |
Artikel
|
| Sprache |
Englisch
|
| Digitales Dokument |
PDF |
| Erschienen |
In: GRA - Volume 15 (2013) |
| Datensatznummer |
250081094
|
|
|
|
|
|
| Zusammenfassung |
| The statistical properties of a random wave field generated by the crossing of two directional
random waves were studied in wave basin experiments in the MARINTEK laboratory. The
overall aim was to understand whether interactions between swell and wind waves can
increase the probability that freak waves will be created.
The MARINTEK facility is 70 x 50m. Directional waves could be generated by paddles
on the 70m side. Tests were conducted using different degrees of directional spreading and
angle of crossing. The water depth was kept at 3m for all tests. Here we report
the results of tests to investigate the effect of the crossing angle and directional
spreading.
The input spectrum in the frequency domain was composed of two JONSWAP spectra
with identical peak periods (Tp=1s) and significant wave heights (Hs=0.058m), and peak
enhancement factors (γ) equal to 3 and 6 respectively. The distribution of energy in the
directional domain was generated by using a cosine-type function with directional spreadings
N=50 and N=200 in the first series of tests. The angles between the two propagating wave
fields were α =10o, 20o, 30o and 40o. In the second series, the angle was kept constant
(40o) and the directional spreading of N=50 changed to N=200 and N=840. For
comparison, an irregular (long-crested) and a random directional sea field were also
generated.
Four realisations of the random wave field were measured using the same input
spectrum with different sets of random amplitudes and phases. There were more
than 1300 individual waves in each 20-min time series (more than 5000 in total).
Measurements of surface elevation were taken every 5m along the main axis of the
basin.
Freak waves, i.e. events with crests larger than or equal to five times the standard
deviation or wave heights larger than or equal to twice the significant wave height, were
observed in each test. The number increased with increasing angle between the wave fields,
and with reduction of directional spreading. This is also reflected in the tails of the wave
height and wave crest distributions, which deviate increasingly from a Rayleigh distribution
with growing distance from the wave paddle. The fourth order moment of the probability
density function of the surface elevation, or kurtosis, was calculated from each surface
elevation time series. The kurtosis is believed to be influenced by the nonlinear dynamics of
free waves responsible for the formation of extreme events. The maximum observed kurtoses
were above 3.0 (the Gaussian expectation) for each test. The observed kurtoses
were compared with an estimate obtained by use of a second-order theory, which
includes only the contribution of bound waves. There was little difference between the
observed and estimated values, indicating that the contribution of free waves is very
small.
Unlike previously published findings for the crossing of irregular waves, the crossing
angle does not appear to exert a significant influence on the kurtosis. Rather, it
seems that the degree of directional spreading is the most important parameter. |
|
|
|
|
|
|