Stencil and Boundary Treatment for Jameson, Schmidt, Turkel Scheme
 

Hi all,


I have a few simple questions about the Jameson, Schmidt, Turkel (JST) scheme for artificial dissipation. I am writing a 2d solver similar to the one in the original paper and I just wan to have a few things clarified. The original paper can be found here: https://www.researchgate.net/publica...epping_Schemes

I am only confused of the artificial dissipation terms. I understand the central differencing of the flux vectors.
Consider a 2d structured domain similar to the one in the paper, using central differencing:


The a paper makes it seem like I need a 5 point stencil in both the i and j directions (see equations 11 and 12). Is this correct?


Do I need 2 layers of ghost cells at the boundaries, or should boundaries be treated differently than interior cells?


Can anybody recommend any artificial schemes that us a more compact stencil?


Does anybody have any simple codes that successfully use the JST scheme?





Thanks for any help you can provide!