open TELEMAC-MASCARET The mathematically superior suite of solvers

Hello everybody

I am kind of confused about JONSWAP parameters in TOMAWAC

If you look at page 64/144 of the Tomawac’s manual (6.1), there is a relationship between peak parameter (Gamma) and Phillips constant as follows:

a_phil=0.0624/(.23+.0336*Gamma-0.185/(Gamma+1.9))

If you replace Gamma=3.3 (recommended) , then is: a_phil=0.2044

There are two questions,

1-Assuming that these parameters (Peak parameter and Phillips constant) are related, why a user can specify both parameters? Please refer to Table 8.1 in page 66/144. I don’t know which one will be finally used if I specify both?!
(e.g., Case 4 of that Table)


2- In one of the test cases of TOMAWAC (turning wind), in the cas file,

CONSTANTE DE PHILLIPS INITIALE = 0.024

Which means that Phillips constant is 0.024 for initial condition. Referring to above equation, the appropriate range of Phillips constant corresponding to Gamma (1.6~6) would be 0.1529~0.2702 and the recommended value would be 0.2044. So why that value has been selected?

I will appreciate any comments.
Thank you

Hello,

thank you for your comments and questions. Actually, you are right this point is not very clear in the TOMAWAC documentation. We will ensure to clarify this for the next version of the documentation.

In fact, you have different ways to specify your initial JONSWAP spectrum :
- you can specify alpha_phil, the Phillips constant (and fp, gamma...) : this corresponds to cases 2 without wind, 3 with wind and 4 of Table 8.1 page 66/144
- or you can specify Hm0, the significant wave height (and also fp, gamma...) : cases 5 with wind and 6 of Table 8.1 page 66/144

The formula at page 64/144 is exactly what is implemented in the code when the JONSWAP spectrum is parameterized by Hm0, and fp, gamma... but not by the Phillips constant. So the alpha_phil you find in this formula is not the real Phillips constant.

The general formula for the JONSWAP spectrum can be written as:
E(f) = alpha_phil * g^2 * (2*Pi)^(-4) * f^(-5) * exp (...) * gamma^(...)
Here alpha_phil really is the Phillips constant, that you can specify in your parameter file. So you see that you can either specifiy alpha_phil directly, or if you give Hm0, the Phillips constant will be calculated in the code as :
alpha_phil = 0.0624/(.23+.0336*Gamma-0.185/(Gamma+1.9)) * Hm0^2 * (2*Pi*fp)^4 / g^2
Then you fall back on the documentation formula.

I hope this answers your question. Please don't hesitate to ask if you need more explanations.

Best regards,

Elodie

Hi Elodie

I did some numerical experiments which confirms your reply. Many thanks for clarifying this.