Hi Glenn What's happening here is that each FILL command has no way to extrapolate to the outer edges of the grid boxes. z1[L=12] doesn't know about z2[L=13], and vice versa, so has no way to fill the gap between the last coordinate point of the first variable and the first coordinate point of the second. If you did a SHADE command, I think you'd see the whole region filled in, but you want a filled contour. What you need is a single variable covering the whole 5 years, or whatever it is. I haven't tested this, but what about something along these lines let z1 = z[d=1,gt=y5@asn] Glenn Carver wrote: I have a monthly mean netcdf files from a run of a model which has run 2001 repeatedly. Each year is in one netcdf file. I want to view the files as a time sequence rather than treat them as a climatology but am not sure how best to create the new time axis. As each file is for '2001' I need to create a new time axis in order to see the sequence of years. I've done this by: |