Can anyone explain what the advantages and disadvantages of using different filter types in LES are? In particular, I am interested in why one might use a Gaussian filter rather than a top-hat filter. And Vice versa.

Thank you.

Surely someone has some ideas on how to answer this question?

Could you explain why you need to use explicit filtering? In most LES implimentations, I have come across, explicit filtering is not done: quantities are taken as filtered on the grid, which makes it difficult to estimate the type of filter that would have been theoretically used. But there are cases, where an actual filter is applied...

Thanks for your response. I just want to know why people use a Gaussian filter in LES. In particular, I want to know why it is used in physical space. Or are its attributes in spectral space?

Assuming that quantities are filtered on the grid amounts to top-har filter if the cell values are averaged based on their neighbours. I want to know how you implement (and why you would want to) a gaussian filter in a LES.

Thanks.