Re: time units in ferret & netcdf

[verschel@xxxxxxxxxxxxxxxxxxxxx (Mark A. Verschell)]

As usual Steve supplies the most elegant method!

In case of model months (what steve calls abstract, e.g. when the year is 360
days with 12 thirty day months), I use a very easy method for generating the
time axis. Here's an example:

Let's say I have 20 years of model runs with monthly output, representing the
years 1975 to 1994, then I define the time axis:

define axis/t="15-jan-1975:12:00":"15-dec-1994:12:00"/npoints=240 mytime

where 240 is 12 months*20 years. This places the time steps pretty closely to
mid month for the whole 20 years.

If you really want to represent model years as 1,2,...20 (for example), you can
try:

def ax/t="15-jan-0001:12:00":"15-dec-0002:12:00"/np=240/t0="01-jan-0000" mytime

This labels everything but the first year correctly (the years 0 and 1 are
special in ferret and are considered climatological years). There is probably
some way elegant way around this - I usually just change the axis label
definitions when plotting.

Mark

-- 
 Mark Verschell                     NASA/Goddard Space Flight Center
 verschell@neptune.gsfc.nasa.gov    Code 970/Lab. for Hydrospheric Proc.
 301-286-2027  Fax: 301-286-1761    Greenbelt, MD 20771