I was wondering if anyone has some hints or solutions to following behavior I have had with the " positive:'up' " z axis attribute.
I have several 3D variables defined with lat, long and depth. In the netcdf files I defined a depth axis with a "positive up" convention and depths are negative values and decreasing monotonically when examining the 'ncdump' of the file.
"depth = 0, -1, -5, " …. Etc ! I wonder if that is a correct convention (decreasing monotonically) for 'positive up'?
I use ferret to do some involved calculations with the variables (creating a new variable) and then save the results to a new netcdf file. Ferret then behaves according to the manual:
"If the coordinates of an axis are monotonically decreasing, instead of increasing, Ferret will transparently reverse both the axis coordinates and the dependent variables that are defined upon that axis. Note that if Ferret writes a reverse-ordered variable to a new netCDF file (with the SAVE command), the coordinates and data in the output file will be in monotonically increasing coordinate order—reversed from the input file."
The convention in the output file also changed to "positive down"
When I then take the new netcdf output file and the original file to do some 3D visualization work (Xvision), the files cannot be compared because depth axes are now mismatched.
I would like for ferret to either preserve the "positive up" attribute when saving (/ORDER switch or something) OR is there some other technique for restoring the positive up convention and reversing the depths in the output netcdf file (netcdf operators etc.) …
Thanks for any hints,
Dr. A. Rost Parsons
Oceanographer, Naval Research Laboratory
Building 1009, Room A119, Code 7333
Stennis Space Center, MS 39529-5004