20.07.2013 Views

3+1 formalism and bases of numerical relativity - LUTh ...

3+1 formalism and bases of numerical relativity - LUTh ...

3+1 formalism and bases of numerical relativity - LUTh ...

SHOW MORE
SHOW LESS

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

8.3 Conformal thin s<strong>and</strong>wich method 141<br />

8.3.2 Extended conformal thin s<strong>and</strong>wich method<br />

An input <strong>of</strong> the above method is the conformal lapse Ñ. Considering the astrophysical problem<br />

stated in Sec. 8.1.1, it is not clear how to pick a relevant value for Ñ. Instead <strong>of</strong> choosing an<br />

arbitrary value, Pfeiffer <strong>and</strong> York [202] have suggested to compute Ñ from the Einstein equation<br />

giving the time derivative <strong>of</strong> the trace K <strong>of</strong> the extrinsic curvature, i.e. Eq. (6.107):<br />

<br />

∂<br />

− Lβ K = −Ψ<br />

∂t −4 Di<br />

˜ ˜ D i N + 2 ˜ Di ln Ψ ˜ D i <br />

N + N 4π(E + S) + Ãij Ãij + K2<br />

<br />

. (8.94)<br />

3<br />

This amounts to add this equation to the initial data system. More precisely, Pfeiffer <strong>and</strong> York<br />

[202] suggested to combine Eq. (8.94) with the Hamiltonian constraint to get an equation involving<br />

the quantity NΨ = ÑΨ7 <strong>and</strong> containing no scalar products <strong>of</strong> gradients as the ˜ Di ln Ψ ˜ DiN term in Eq. (8.94), thanks to the identity<br />

˜Di ˜ D i N + 2 ˜ Di ln Ψ ˜ D i N = Ψ −1 Di<br />

˜ ˜ D i (NΨ) + N ˜ Di ˜ D i <br />

Ψ . (8.95)<br />

Expressing the left-h<strong>and</strong> side <strong>of</strong> the above equation in terms <strong>of</strong> Eq. (8.94) <strong>and</strong> substituting<br />

˜Di ˜ D i Ψ in the right-h<strong>and</strong> side by its expression deduced from Eq. (8.92), we get<br />

˜Di ˜ D i ( ÑΨ7 )−( ÑΨ7 )<br />

1<br />

8 ˜ R + 5<br />

12 K2 Ψ 4 + 7<br />

8 Âij Âij Ψ −8 + 2π( ˜ E + 2 ˜ S)Ψ −4<br />

where we have used the short-h<strong>and</strong> notation<br />

<strong>and</strong> have set<br />

˙K := ∂K<br />

∂t<br />

<br />

+ ˙K i<br />

− β DiK ˜ Ψ 5 = 0,<br />

(8.96)<br />

(8.97)<br />

˜S := Ψ 8 S. (8.98)<br />

Adding Eq. (8.96) to Eqs. (8.92) <strong>and</strong> (8.93), the initial data system becomes<br />

˜Di ˜ D i Ψ − ˜ R 1<br />

Ψ +<br />

8<br />

<br />

1<br />

˜Dj<br />

Ñ (˜ Lβ) ij<br />

<br />

+ ˜ <br />

1<br />

Dj<br />

Ñ<br />

˜Di ˜ D i ( ÑΨ7 ) − ( ÑΨ7 <br />

˜R<br />

)<br />

8<br />

8 Âij Âij Ψ −7 + 2π ˜ EΨ −3 − K2<br />

˙˜γ ij<br />

<br />

− 4<br />

3 Ψ6D˜ i i<br />

K = 16π˜p<br />

+ 5<br />

12 K2 Ψ 4 + 7<br />

12 Ψ5 = 0 (8.99)<br />

8 Âij ÂijΨ −8 + 2π( ˜ E + 2 ˜ S)Ψ −4<br />

<br />

<br />

+ ˙K i<br />

− β DiK ˜ Ψ 5 = 0<br />

(8.100)<br />

, (8.101)<br />

where Âij is the function <strong>of</strong> Ñ, βi , ˜γij <strong>and</strong> ˙˜γ ij defined by Eq. (8.90). Equations (8.99)-(8.101)<br />

constitute the extended conformal thin s<strong>and</strong>wich (XCTS) system for the initial data<br />

problem. The free data are the conformal metric ˜γ, its coordinate time derivative ˙˜γ, the extrinsic<br />

curvature trace K, its coordinate time derivative ˙ K, <strong>and</strong> the rescaled matter variables ˜ E, ˜ S <strong>and</strong><br />

˜p i . The constrained data are the conformal factor Ψ, the conformal lapse Ñ <strong>and</strong> the shift vector<br />

β.

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!