Welcome:
  Home Page
  Our Project
  Grant Page
  GWEN
  Jobs
People:
  Directory
  Mail Lists
  Advisory Board
Documents:
  Publications
  Presentations
  Newsletter
  Movies
Links:
  Other Networks
  Software Links
  CVS
Meetings:
  Past
  Upcoming
Projects:
Network:
  Initial Data
  Black Holes
  Neutron Stars
  PN Theory
Related:
  AEI Group
  Cactus
  A.S.C.
Internal:
  Resources
Contact:
  General
  WebMaster
Legal Notice
   Please read
next up previous contents
Next: References Up: Andrea Nerozzi (Portsmouth) Previous: Mini-CV

Relativistic irrotational fluids: 3D simulations in flat space

The following paragraphs will describe the project I am currently working on, whose final aim is to study the behaviour of relativistic irrotational fluids in a curved spacetime, such as Schwarzschild or Kerr. In order to do this, I am developing a code within the Cactus environment, to study numerically the problem. I present the recent results obtained in flat space, which show that the code is valid and ready to be extended to general metrics.

General equations

In order to define an irrotational fluid, we have to introduce the vorticity tensor:

displaymath583

where tex2html_wrap_inline1357 is the projection tensor, h is the enthalpy of the fluid and tex2html_wrap_inline1361 is the four velocity. The vorticity tensor is identically null for an irrotational fluid. From the Euler's equations for a perfect fluid:

displaymath596

we can derive a simple expression for the vorticity tensor which we write below :

  equation602

Equation gif tells us that, if tex2html_wrap_inline1363 , then we can write the quantity tex2html_wrap_inline1365 as the gradient of a scalar field, that is tex2html_wrap_inline1367 . We then obtain the following expression for the conservation of particle density, in terms of tex2html_wrap_inline1369 :

  equation615

This equation must be matched with the one coming from the normalization condition for the four velocity tex2html_wrap_inline1361 :

  equation623

Equations gif and gif have been studied analytically in literature to study fluids in curved metrics such as Schwarzschild and Kerr. A first attempt to solve these equations can be found in [2] where equation gif is linearized by setting n=h. This further approximation corresponds to the speed of sound being equal to the speed of light. A perturbative approach can instead be found in [1]. Our aim is to solve numerically equation gif, starting from the simple case of a flat space, and then moving to curved space-times, thus dropping the further approximation tex2html_wrap_inline1375 .

Thermodynamic considerations

We define the pressure to be given by the following polytropic expression:

  equation635

Some straightforward thermodynamic calculations lead to the following equations which relate the particle density n to the enthalpy h.

  eqnarray639

The speed of sound can also be calculated and it turns out to be:

  equation649

The conservation law system

In order to study numerically equation gif, we have to reduce it in first order conservation form (for further details, see [3] or [4]). We can do it for a flat space metric as follows: we write equation gif in explicit form:

equation660

We introduce the variables:

eqnarray667

We thus obtain the following conservation law system:

  eqnarray671

Equation gif can be rewritten in term of the new variables in the following manner:

  equation680

The conservation law system gif is evolved numerically using a second order numerical method (Lax-Wendroff, Mac-Cormack). At the end of each timestep, the new value of h is computed using a Raphson-Newton iteration on equation gif. The value of n is then computed by using equation gif.

Numerical Results

In figures gif and gif some 1D results are shown. The evolution of an enthalpy perturbation with some initial velocity tex2html_wrap_inline1383 has been studied. The code reproduces quite well our expectations for 1D evolution. We in fact expect the initial data to evolve following the characteristics given by the following expression:

  equation692

Following equation gif, we see that in figure gif, the initial data are subsonic, so that the initial perturbation is divided into two waves propagating in both directions. Figure gif represents supersonic initial data, so the two characteristics are both in the same direction.

     figure701
Figure: Fluid simulation with initial data: tex2html_wrap_inline1389 , tex2html_wrap_inline1391
Figure: Fluid simulation with initial data: tex2html_wrap_inline1385 , tex2html_wrap_inline1387


next up previous contents
Next: References Up: Andrea Nerozzi (Portsmouth) Previous: Mini-CV


This work has been supported by the EU Programme 'Improving the Human Research Potential and the Socio-Economic Knowledge Base' (Research Training Network Contract HPRN-CT-2000-00137).