The "Summary" section does not say what mesh adaptation / AMR actually does, i.e., vary resolution to reduce approximation errors "where they matter" and reduce computational costs elsewhere. This is mentioned in the "Statement of Need" section but - given the summary is meant to be for a non-specialist audience - it would be good to have a sentence explaining this. Perhaps the introductory sentence could just be moved from the "Statement of Need" section to the "Summary".
The "Summary" section does not say what mesh adaptation / AMR actually does, i.e., vary resolution to reduce approximation errors "where they matter" and reduce computational costs elsewhere. This is mentioned in the "Statement of Need" section but - given the summary is meant to be for a non-specialist audience - it would be good to have a sentence explaining this. Perhaps the introductory sentence could just be moved from the "Statement of Need" section to the "Summary".
(Part of JOSS review https://github.com/openjournals/joss-reviews/issues/6887.)