Hostname: page-component-7b7567c8c-7pwfn Total loading time: 0 Render date: 2026-11-16T06:29:47.804Z Has data issue: false hasContentIssue false

Spectral evolution of weakly nonlinear random waves: kinetic description versus direct numerical simulations

Published online by Cambridge University Press:  12 April 2018

Sergei Y. Annenkov*
Affiliation:
School of Computing and Mathematics, Keele University, Keele ST5 5BG, UK
Victor I. Shrira
Affiliation:
School of Computing and Mathematics, Keele University, Keele ST5 5BG, UK
*
†Email address for correspondence: s.annenkov@keele.ac.uk

Abstract

Kinetic equations are widely used in many branches of science to describe the evolution of random wave spectra. To examine the validity of these equations, we study numerically the long-term evolution of water wave spectra without wind input using three different models. The first model is the classical kinetic (Hasselmann) equation (KE). The second model is the generalised kinetic equation (gKE), derived employing the same statistical closure as the KE but without the assumption of quasistationarity. The third model, which we refer to as the DNS-ZE, is a direct numerical simulation algorithm based on the Zakharov integrodifferential equation, which plays the role of the primitive equation for a weakly nonlinear wave field. It does not employ any statistical assumptions. We perform a comparison of the spectral evolution of the same initial distributions without forcing, with/without a statistical closure and with/without the quasistationarity assumption. For the initial conditions, we choose two narrow-banded spectra with the same frequency distribution and different degrees of directionality. The short-term evolution ($O(10^{2})$ wave periods) of both spectra has been previously thoroughly studied experimentally and numerically using a variety of approaches. Our DNS-ZE results are validated both with existing short-term DNS by other methods and with available laboratory observations of higher-order moment (kurtosis) evolution. All three models demonstrate very close evolution of integral characteristics of the spectra, approaching with time the theoretical asymptotes of the self-similar stage of evolution. Both kinetic equations give almost identical spectral evolution, unless the spectrum is initially too narrow in angle. However, there are major differences between the DNS-ZE and gKE/KE predictions. First, the rate of angular broadening of initially narrow angular distributions is much larger for the gKE and KE than for the DNS-ZE, although the angular width does appear to tend to the same universal value at large times. Second, the shapes of the frequency spectra differ substantially (even when the nonlinearity is decreased), the DNS-ZE spectra being wider than the KE/gKE ones and having much lower spectral peaks. Third, the maximal rates of change of the spectra obtained with the DNS-ZE scale as the fourth power of nonlinearity, which corresponds to the dynamical time scale of evolution, rather than the sixth power of nonlinearity typical of the kinetic time scale exhibited by the KE. The gKE predictions fall in between. While the long-term DNS show excellent agreement with the KE predictions for integral characteristics of evolving wave spectra, the striking systematic discrepancies for a number of specific spectral characteristics call for revision of the fundamentals of the wave kinetic description.

Information

Type
JFM Papers
Copyright
© 2018 Cambridge University Press 
Figure 0

Table 1.Parameters for the numerical algorithms.

Figure 1

Figure 1.The short-term evolution of the initial spectra (a) A and (b) B, with a direct comparison of five different numerical approaches. Initial (black solid curve) and final (after 150 wave periods) omnidirectional energy spectra $S(k)$ are plotted. Evolution is simulated with the HOS; the Dysthe equation (BMNLS; both from Xiao et al.2013, cf. their figure 7); DNS-ZE; KE, WRT algorithm; and gKE.

Figure 2

Figure 2.Evolution of the directional spread $\unicode[STIX]{x1D703}_{m}$ (averaged second moment of directional distribution) of two initial spectra (a) A and (b) B. As in figure 1(a,b), a direct comparison of five different numerical approaches is shown.

Figure 3

Figure 3.Growth rates $\text{d}S(k)/\text{d}t$ versus wavenumber $k$, normalised by the initial spectral peak wavenumber $k_{p}$, over the first 50 periods of evolution for spectrum (a) A and (b) B, with a comparison of five different numerical approaches, as in figure 1(a,b).

Figure 4

Figure 4.Evolution of the total kurtosis for spectra (a) A and (b) B versus fetch with different numerical approaches, and comparison with observations. Black dashed curve, HOS (Xiao et al.2013); red solid curve, DNS-ZE; blue dashed curve, gKE. The dotted blue curve shows the bound harmonics kurtosis, calculated from the DNS spectra. The black dots show observations of the total kurtosis (Toffoli et al.2010).

Figure 5

Figure 5.Second moment of directional distribution $\unicode[STIX]{x1D703}_{2}(k,t)$ for spectrum B at time (a) $t=0$ (black solid line), 50, 150 and 300 periods, (b) $t=500$, 1000 and 2000 periods, with different numerical methods. In all cases, the width of the directional distribution increases with time.

Figure 6

Figure 6.Long-term evolution of initial spectra A and B, with comparison of three different numerical approaches. (a) Evolution of spectrum A at time $t=0$ (black solid curve), 50, 150, 300, 500, 800 and 1200 periods (DNS-ZE, gKE and KE), and $t=2000$ and 3200 periods (DNS-ZE and KE only). (b) Evolution of spectrum B at time $t=0$ (black solid curve), 50, 150, 300, 500, 800, 1200 and 2000 periods (DNS-ZE, gKE and KE), and $t=3200$ and 4750 periods (DNS-ZE and KE only).

Figure 7

Figure 7.Long-term evolution of mean directional spread $\unicode[STIX]{x1D703}_{m}$ of initial spectra (a) A and (b) B, with comparison of three different numerical approaches (DNS-ZE, KE and gKE). The colour codes are as in figure 2(a,b), which shows the first 150 periods of the same evolution.

Figure 8

Figure 8.Evolution in time of various characteristics for spectra A (a,c,e) and B (b,d,f) with theoretical asymptotes for swell. (a,b) Wavenumber of spectral peak $k_{p}$ for wave action spectrum $n(k)$. (c,d) Steepness $\unicode[STIX]{x1D700}$. (e,f) Amplitude of wave action $n(k)$ at the spectral peak.

Figure 9

Figure 9.Absolute growth rate $\text{d}S(k)/\text{d}t$ versus normalised wavenumber $k/k_{p}$ over the first 50 periods of evolution for spectra (a) A and (b) B with initial amplitudes multiplied by 1, $\sqrt{2}$ and $1/\sqrt{2}$. The middle curves correspond to the curves shown in figure 3(a,b).

Figure 10

Figure 10.The exponent $\unicode[STIX]{x1D708}$ for the scaling of the maximum absolute growth rate $\max \text{d}n/\text{d}t\sim \unicode[STIX]{x1D700}^{\unicode[STIX]{x1D708}}$ versus the normalised wavenumber $k/k_{p}$ for spectrum A: KE (black circles), gKE (grey circles), DNS-ZE (empty circles). The value of the exponent $\unicode[STIX]{x1D708}$ is obtained for each $k$ as a first-degree polynomial fit $(\log \text{d}n/\text{d}t=\unicode[STIX]{x1D708}\log \unicode[STIX]{x1D700}+\text{const.})$, for seven values of $\unicode[STIX]{x1D700}$ (initial amplitudes are multiplied by $2^{j/2}$, $j=-4,-3,\ldots ,2$). For each obtained curve $\unicode[STIX]{x1D708}(k)$, 95 % confidence bounds are also shown.

Figure 11

Figure 11.Evolution of spectrum B with amplitude multiplied by 0.5, with three numerical approaches. Spectra are plotted for time $t=0$ (black solid curve), 1000, 2000 and 3500 periods.