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 ...
170 Choice of foliation and spatial coordinates Taking the flat-divergence of this relation and using relation (9.80) (with the commutation property of ∂/∂t and D) yields ∂ ˜ Γ i ∂t = = −2NDj Ãij − 2A ij DjN + β k Dk ˜ Γ i − ˜ Γ k Dkβ i + 2 3 ˜ Γ i Dkβ k ˜γ jk DjDkβ i + 1 3 ˜γij DjDkβ k . (9.83) Now, we may use the momentum constraint (6.110) to express Dj Ãij : with ˜Dj Ãij = −6 Ãij ˜ Dj lnΨ + 2 3 ˜ D i K + 8πΨ 4 p i , (9.84) ˜Dj Ãij = DjÃij + ˜Γ i jk − ¯ Γ i kj jk à + ˜Γ j jk − ¯ Γ j jk à =0 ik , (9.85) where the “= 0” results from the fact that 2˜ Γ j ˜γ := det ˜γij = detfij =: f [Eq. (6.19)]. Thus Eq. (9.83) becomes ∂ ˜ Γ i jk = ∂ ln ˜γ/∂xk and 2 ¯ Γ j jk = ∂ ln f/∂xk , with ∂t = ˜γjk DjDkβ i + 1 3 ˜γij DjDkβ k + 2 3 ˜ Γ i Dkβ k − ˜ Γ k Dkβ i + β k Dk ˜ Γ i −2N 8πΨ 4 p i − Ãjk ˜ Γ i jk − ¯ Γ i jk − 6 Ãij Dj ln Ψ + 2 3 ˜γij DjK We conclude that the Gamma freezing condition (9.81) is equivalent to ˜γ jk DjDkβ i + 1 3 ˜γij DjDkβ k + 2 3 ˜ Γ i Dkβ k − ˜ Γ k Dkβ i + β k Dk ˜ Γ i = 2N 8πΨ 4 p i − Ãjk ˜ Γ i jk − ¯ Γ i jk − 6 Ãij Dj ln Ψ + 2 3 ˜γij DjK − 2 Ãij DjN. + 2 Ãij DjN. (9.86) (9.87) This is an elliptic equation for the shift vector, which bears some resemblance with Shibata’s approximate minimal distortion, Eq. (9.75). 9.3.5 Gamma drivers As seen above the Gamma freezing condition (9.81) yields to the elliptic equation (9.87) for the shift vector. Alcubierre and Brügmann [7] have proposed to turn it into a parabolic equation by considering, instead of Eq. (9.81), the relation ∂βi ∂t = k∂˜ Γi ∂t , (9.88) where k is a positive function. The resulting coordinate choice is called a parabolic Gamma driver. Indeed, if we inject Eq. (9.88) into Eq. (9.86), we clearly get a parabolic equation for the shift vector, of the type ∂β i /∂t = k ˜γ jk DjDkβ i + 1 3 ˜γij DjDkβ k + · · · .
9.3 Evolution of spatial coordinates 171 An alternative has been introduced in 2003 by Alcubierre, Brügmann, Diener, Koppitz, Pollney, Seidel and Takahashi [9] (see also Refs. [181] and [59]); it requires ∂2βi ∂t2 = k∂˜ Γi ∂t − η − ∂ ∂βi ln k ∂t ∂t , (9.89) where k and η are two positive functions. The prescription (9.89) is called a hyperbolic Gamma driver [9, 181, 59]. Indeed, thanks to Eq. (9.86), it is equivalent to ∂2βi + η − ∂t2 ∂ ∂βi ln k = k ˜γ ∂t ∂t jk DjDkβ i + 1 3 ˜γij DjDkβ k + 2 3 ˜ Γ i Dkβ k − ˜ Γ k Dkβ i + β k Dk ˜ Γ i −2N 8πΨ 4 p i − Ãjk Γ˜ i jk − ¯ Γ i jk − 6ÃijDj ln Ψ + 2 3 ˜γij DjK − 2Ãij DjN , (9.90) which is a hyperbolic equation for the shift vector, of the type of the telegrapher’s equation. The term with the coefficient η is a dissipation term. It has been found by Alcubierre et al. [9] crucial to add it to avoid strong oscillations in the shift. The hyperbolic Gamma driver condition (9.89) is equivalent to the following first order system ⎧ ⎪⎨ ⎪⎩ ∂βi ∂t = kBi ∂B i ∂t = ∂˜ Γ i ∂t − ηBi . (9.91) Remark : In the case where k does not depend on t, the Gamma driver condition (9.89) reduces to a previous hyperbolic condition proposed by Alcubierre, Brügmann, Pollney, Seidel and Takahashi [10], namely ∂2βi ∂t2 = k∂˜ Γi ∂t − η∂βi . (9.92) ∂t Hyperbolic Gamma driver conditions have been employed in many recent numerical computations: • 3D gravitational collapse calculations by Baiotti et al. (2005, 2006) [28, 30], with k = 3/4 and η = 3/M, where M is the ADM mass; • the first evolution of a binary black hole system lasting for about one orbit by Brügmann, Tichy and Jansen (2004) [70], with k = 3/4NΨ −2 and η = 2/M; • binary black hole mergers by – Campanelli, Lousto, Marronetti and Zlochower (2006) [73, 74, 75, 76], with k = 3/4; – Baker et al. (2006) [32, 33], with k = 3N/4 and a slightly modified version of Eq. (9.91), namely ∂˜Γ i /∂t replaced by ∂˜Γ i /∂t − β j ∂˜Γ i /∂x j in the second equation; – Sperhake [248], with k = 1 and η = 1/M.
- Page 120 and 121: 120 Asymptotic flatness and global
- Page 122 and 123: 122 Asymptotic flatness and global
- Page 124 and 125: 124 Asymptotic flatness and global
- Page 126 and 127: 126 The initial data problem Notice
- Page 128 and 129: 128 The initial data problem where
- Page 130 and 131: 130 The initial data problem 8.2.3
- Page 132 and 133: 132 The initial data problem where
- Page 134 and 135: 134 The initial data problem Figure
- Page 136 and 137: 136 The initial data problem Figure
- Page 138 and 139: 138 The initial data problem In par
- Page 140 and 141: 140 The initial data problem Accord
- Page 142 and 143: 142 The initial data problem Remark
- Page 144 and 145: 144 The initial data problem Remark
- Page 146 and 147: 146 The initial data problem 8.4.1
- Page 148 and 149: 148 The initial data problem Since
- Page 150 and 151: 150 The initial data problem • fo
- Page 152 and 153: 152 Choice of foliation and spatial
- Page 154 and 155: 154 Choice of foliation and spatial
- Page 156 and 157: 156 Choice of foliation and spatial
- Page 158 and 159: 158 Choice of foliation and spatial
- Page 160 and 161: 160 Choice of foliation and spatial
- Page 162 and 163: 162 Choice of foliation and spatial
- Page 164 and 165: 164 Choice of foliation and spatial
- Page 166 and 167: 166 Choice of foliation and spatial
- Page 168 and 169: 168 Choice of foliation and spatial
- Page 172 and 173: 172 Choice of foliation and spatial
- Page 174 and 175: 174 Choice of foliation and spatial
- Page 176 and 177: 176 Evolution schemes been used by
- Page 178 and 179: 178 Evolution schemes Comparing wit
- Page 180 and 181: 180 Evolution schemes Now the ∇-d
- Page 182 and 183: 182 Evolution schemes coordinates (
- Page 184 and 185: 184 Evolution schemes = 1 2 −∆
- Page 186 and 187: 186 Evolution schemes corresponds t
- Page 188 and 189: 188 Evolution schemes
- Page 190 and 191: 190 Lie derivative Figure A.1: Geom
- Page 192 and 193: 192 Lie derivative
- Page 194 and 195: 194 Conformal Killing operator and
- Page 196 and 197: 196 Conformal Killing operator and
- Page 198 and 199: 198 Conformal Killing operator and
- Page 200 and 201: 200 BIBLIOGRAPHY [13] A. Anderson a
- Page 202 and 203: 202 BIBLIOGRAPHY [43] T.W. Baumgart
- Page 204 and 205: 204 BIBLIOGRAPHY [75] M. Campanelli
- Page 206 and 207: 206 BIBLIOGRAPHY [105] G. Darmois :
- Page 208 and 209: 208 BIBLIOGRAPHY [135] H. Friedrich
- Page 210 and 211: 210 BIBLIOGRAPHY [163] J. Isenberg
- Page 212 and 213: 212 BIBLIOGRAPHY [195] S. Nissanke
- Page 214 and 215: 214 BIBLIOGRAPHY [226] M. Shibata :
- Page 216 and 217: 216 BIBLIOGRAPHY [255] K. Taniguchi
- Page 218 and 219: Index 1+log slicing, 161 3+1 formal
9.3 Evolution <strong>of</strong> spatial coordinates 171<br />
An alternative has been introduced in 2003 by Alcubierre, Brügmann, Diener, Koppitz,<br />
Pollney, Seidel <strong>and</strong> Takahashi [9] (see also Refs. [181] <strong>and</strong> [59]); it requires<br />
∂2βi ∂t2 = k∂˜ Γi ∂t −<br />
<br />
η − ∂<br />
<br />
∂βi ln k<br />
∂t ∂t<br />
, (9.89)<br />
where k <strong>and</strong> η are two positive functions. The prescription (9.89) is called a hyperbolic<br />
Gamma driver [9, 181, 59]. Indeed, thanks to Eq. (9.86), it is equivalent to<br />
∂2βi <br />
+ η −<br />
∂t2 ∂<br />
<br />
∂βi ln k = k ˜γ<br />
∂t ∂t jk DjDkβ i + 1<br />
3 ˜γij DjDkβ k + 2<br />
3 ˜ Γ i Dkβ k − ˜ Γ k Dkβ i + β k Dk ˜ Γ i<br />
<br />
−2N 8πΨ 4 p i − Ãjk Γ˜ i<br />
jk − ¯ Γ i <br />
jk − 6ÃijDj ln Ψ + 2<br />
3 ˜γij <br />
DjK − 2Ãij <br />
DjN , (9.90)<br />
which is a hyperbolic equation for the shift vector, <strong>of</strong> the type <strong>of</strong> the telegrapher’s equation.<br />
The term with the coefficient η is a dissipation term. It has been found by Alcubierre et al. [9]<br />
crucial to add it to avoid strong oscillations in the shift.<br />
The hyperbolic Gamma driver condition (9.89) is equivalent to the following first order<br />
system<br />
⎧<br />
⎪⎨<br />
⎪⎩<br />
∂βi ∂t<br />
= kBi<br />
∂B i<br />
∂t = ∂˜ Γ i<br />
∂t − ηBi .<br />
(9.91)<br />
Remark : In the case where k does not depend on t, the Gamma driver condition (9.89) reduces<br />
to a previous hyperbolic condition proposed by Alcubierre, Brügmann, Pollney, Seidel <strong>and</strong><br />
Takahashi [10], namely<br />
∂2βi ∂t2 = k∂˜ Γi ∂t<br />
− η∂βi . (9.92)<br />
∂t<br />
Hyperbolic Gamma driver conditions have been employed in many recent <strong>numerical</strong> computations:<br />
• 3D gravitational collapse calculations by Baiotti et al. (2005, 2006) [28, 30], with k = 3/4<br />
<strong>and</strong> η = 3/M, where M is the ADM mass;<br />
• the first evolution <strong>of</strong> a binary black hole system lasting for about one orbit by Brügmann,<br />
Tichy <strong>and</strong> Jansen (2004) [70], with k = 3/4NΨ −2 <strong>and</strong> η = 2/M;<br />
• binary black hole mergers by<br />
– Campanelli, Lousto, Marronetti <strong>and</strong> Zlochower (2006) [73, 74, 75, 76], with k = 3/4;<br />
– Baker et al. (2006) [32, 33], with k = 3N/4 <strong>and</strong> a slightly modified version <strong>of</strong><br />
Eq. (9.91), namely ∂˜Γ i /∂t replaced by ∂˜Γ i /∂t − β j ∂˜Γ i /∂x j in the second equation;<br />
– Sperhake [248], with k = 1 <strong>and</strong> η = 1/M.