Ocean Modelling: state-of-the-artI am presently digesting a book by Pat Roache "Fundamentals of Verification and Validation". Previously I knew of his basic idea and had applied it in my own work --- although not with sufficient rigour! (I have to be honest.)
I had thought to do a proper job on a suite of ocean models of my own design, including addressing fundamental issues about variable grids and nesting. I figured the best way to get a coherent analysis "out there" would be by writing my own electronic book. To try to publish in oceanographic journals would cause too much fragmentation --- and, if previous experience is a guide, one encounters too much unreasonable resistance.
So I did a google search on "ocean models".
Top of the google-list was a bunch of beach babes...
After that I got to: http://www.ocean-modeling.org/
Clicking on "Test Problems", the following, rather remarkable statement appeared:"An efficient way of contrasting model behavior is to devise an efficient set of process-oriented test problems in which alternative numerical formulations can be assessed against standards of merit (cost, accuracy, robustness, most smooth solution, etc)."So it seems that "smoothness" is a "standard of merit" --- along with "etc", fine company, indeed.
Disgusted, I clicked back to the beach babes...