Hi David,
How interesting this hasn't come up before. This looks like a
long-standing Ferret bug. We'll get it corrected.
Ansley
David Wang wrote:
Hi Ferreters,
I found that Ferret couldn't recognize the edges attribute if the axis
is monotonically decreasing. For example,
yes? use grid.glob.nc
*** NOTE: Edges "" do not enclose point on axis Z
*** NOTE: Substituting coordinate midpoints
$ ncdump -h grid.glob.nc
| grep Z:
Z:long_name = "vertical coordinate of cell center" ;
Z:units = "meters" ;
Z:positive = "up" ;
Z:edges = "Zp1" ;
$ ncdump -v Z,Zp1 grid.glob.nc
| tail -n 14
data:
Z = -5, -15, -25, -35, -45, -55, -65, -75, -85, -95, -105, -115, -125,
-135,
-145, -155.565, -168.36, -185.5, -208.88, -240.09, -280.325,
-330.32,
-390.325, -460.09, -538.88, -625.5, -718.36, -815.565, -915, -1015,
-1115, -1215, -1320.71, -1448.385, -1616.65, -1837.95, -2116.65,
-2448.385, -2820.71, -3215, -3615, -4015, -4415, -4815, -5215,
-5615 ;
Zp1 = 0, -10, -20, -30, -40, -50, -60, -70, -80, -90, -100, -110, -120,
-130, -140, -150, -161.13, -175.59, -195.41, -222.35, -257.83,
-302.82,
-357.82, -422.83, -497.35, -580.41, -670.59, -766.13, -865, -965,
-1065,
-1165, -1265, -1376.42, -1520.35, -1712.95, -1962.95, -2270.35,
-2626.42,
-3015, -3415, -3815, -4215, -4615, -5015, -5415, -5815 ;
I'd like to keep the values of Z and Zp1 as they are. Is there a
workaround? This is 32-bit Ferret v6.2 (Linux).
Thanks,
David
--
turn and live.
|