[selig5576]

Hi all,

I have been working on a 2D hydrodynamics code. Recently I have implemented the AUSM (Advection Upwind Splitting Method) scheme for the 2D compressible Euler equations. Previously I had been using the Rusanov method as it was simple. I'm testing my code against the bechmark problem of Wendroff and Liska 2003. While my density contour plot looks +- exactly the same (I'm not expecting it to be the same as they use a different method), my pressure contour plot has some strange behavior. The paper Wendroff & Liska does not offer pressure contour plots so I am unsure if there is a numerical problem.

With that said when I plot the pressure and density with the Rusanov method I do not get the oscillations (as attached), but as we know the Rusanov method is not as accurate the AUSM scheme when it comes to the accuracy of the method. I have attached some images for illustration.

Comment: I am less inclined to believe my code is wrong as my density plot is more or less the same and does not display strange oscillations.

It would be much appreciated if someone had some thoughts on if there is something wrong with my code or if it is just the product of the method I get those weird lines.

EDIT: My code is written in Fortran 90 and I can include the code if that helps.
EDIT 2: I ran the Rusanov code and I get the same lines, but no wiggles.

[duri]

Is that appearing in first order scheme as well. If not, verify your limiter implementation.