I am aware of computational costs while working with high resolution data, so this question is more of hypotetical nature and is related to the relevance of friction parameter in inundation modeling. The question is also software(model)-dependent.