For the issue of missing and fill values, note this section from the
Users' Guide in the description of using NetCDF files in Ferret
(Ch 3, sec 1.6.1). I've put a setence in bold type. Once
the
data is read in, Ferret recognizes only one missing value flag.
For NetCDF input data sets the missing value flag(s) is
indicated by the values of the attributes "missing_value" and
"_FillValue." If both attributes are defined to have different
values both will be recognized and used by Ferret as missing
value indicators, however the occurrences of _FillValue will
be replaced with the value of missing_value as the data are
read into Ferret's memory cache so that only a single missing
value flag is apparent inside of Ferret. The command
SET VARIABLE/BAD= can also be applied to NetCDF variables,
thereby temporarily setting a user-imposed value for _FillValue.
But you could set the variable to some particular value at the locations
you want to mark. This flag would not be the same as the missing
value,
and shade it with, for example:
shade var/lev=(0,1000,100) ! the normal range of the
data
shade/over/lev=(-9999)
! the value of your flag
Ansley Manke
Yunyao Qu wrote:
Because the data set is just a 512*512 matrix(and we have the lat-lon position description for each point value also), a simple way to change the origin of Axes is using YLIMITS=hi:lo instead of YLIMITS=lo:hi in SHADE command for this data set.
As for differentiating the Missing values and Fill Values(secondary invalid values) in one file, we just sacrifice this, and contour the lat and lon lines to mark the user interested region(it is north or south hemisphere in this scenario).It seems we could specify TWO invalid values( one by Missing, another by _Fill) in Ferret User Guide, can we?
If the answer is YES, can we draw it in different colors?Thanks,
FrankAnsley Manke wrote:
> Hi Frank,
> Did you get a satisfactory solution to your questions? If so could
> you post the answer back to the Ferret Users's list, so there is
> a resolution there which others might find?
>
> Ansley
>
> Yunyao Qu wrote:
>
> > Hi, Mick,
> >
> > Sorry for not clear declaring. The coordinate is NOT REAL lat-lon representation. It is a virtual 512x512 matrix
> > dimension.
> > I attached 2 sample files you can see its dimensiont.
> > I tested (SHADE var x_page y_page, mp_....,etc) , they work fine in lat-lon gridded data. But this kind of virtual
> > COL and ROW data ....
> >
> > Thanks
> > Frank QU
> >
> > Mick Spillane wrote:
> >
> > > Hi Frank,
> > > Judging from your figure, which I take it is basically an
> > > mp_stereographic_north projection of your data, it is just the
> > > longitude that is flipped - I can see that Africa is backward but
> > > the north pole is where it ought to be.
> > > The Ferret commands you give are not the ones that made this
> > > plot, at least not in my limited understanding of the mp_* demos.
> > > In the commands for making a polar stereographic plot in my demo
> > > below you will see the variable x_page that holds the "x" part of
> > > the transformation. The simplest way for you to fix your problem
> > > may be to replace x_page by (-1)*x_page in your shade command(s).
> > > There is also a variable "MP_CENTRAL_MERIDIAN" that you may choose
> > > to use to put a different meridian at 12 o'clock.
> > >
> > > My demo:
> > >
> > > ! flipping longitudes in a polar stereographic plot
> > >
> > > use etopo60
> > > set grid rose
> > >
> > > set win/aspect=1:ax
> > > ppl axset,0,0,0,0 ; ppl axlabp,0,0
> > >
> > > go mp_stereographic_north
> > > region/x=0:360/y=0:90
> > >
> > > ! first shade in the normal way (longitude is correct here)
> > >
> > > shade/nolab/nokey rose, x_page, y_page
> > >
> > > ! then flip/"correct" longitudes and use new MP_CENTRAL_MERIDIAN
> > >
> > > message
> > >
> > > let MP_CENTRAL_MERIDIAN=45
> > > shade/nolab/nokey rose, (-1)*x_page, y_page
> > >
> > > |--****--****-*---*---***--***--|____spillane@pmel.noaa.gov____|
> > > |-*__---*-----*--*-*--*--*-*--*-|_SCIENCE APPLICATIONS SUPPORT_|
> > > |--***--*-----*-*---*-***--***--|____EPIC/Ferret/PlotPlus______|
> > > |-----*-*-----*-*****-*----*----|__Room 2070 Bldg#3 NOAA/PMEL__|
> > > |-****---****-*-*---*-*----*----|____Phone_:_(206)526-6780_____|
> > >
> > > On Thu, 17 Jan 2002, Yunyao Qu wrote:
> > >
> > > > Hi, Mick,
> > > > My problem is EXACTLY as what you said:"....that when you plot it the "map"
> > > > is upsidedown....".
> > > > So, We would like to change it back when plot it.
> > > > The data set is simply a (row, col) matrix NetCDF file.
> > > >
> > > > The following is script
> > > > ================
> > > > USE SAA_Descriptor
> > > > SET REGION/Y=90:0/X=180:-180/T=280683
> > > > SHADE/LEVELS=1/nokey/noaxis water_vapor // get label
> > > > shade/LEVELS=1/nolab/nokey/noaxis/over/pal=black missing(water_vapor,0) //
> > > > plot missing data
> > > > SHADE/NOAXES/LEVELS=23/over/key/nolab water_vapor // plot image
> > > > shade/over/nolab/pal=grey/noaxis/lev=(0) water_vapor // plot special bad data
> > > >
> > > > ==================
> > > > The generated image also attached. We would like its upside down.
> > > > http://wwwdev.saa.noaa.gov/VisData/EDRMAPNH14/GIF/2001_12_31-20:00:00-EDRMAPNH14-water_vapor-1_512_1_512-0.5.GIF
> > > >
> > > > Thanks,
> > > > Frank
> > > >
> > > >
> > > >
> > > >
> > > >
> > > >
> > > > Mick Spillane wrote:
> > > >
> > > > > Hi Frank,
> > > > > Its not quite clear to me what problem you are dealing with.
> > > > > Sometimes an xy dataset is ordered opposite to the Ferret norm of
> > > > > south to north/west to east so that when you plot it the "map" is
> > > > > upsidedown.
> > > > > In some cases it is possible to orient the plot in the way
> > > > > desired using the ppl xaxis and ppl yaxis commands; in other cases
> > > > > one has to be more tricky.
> > > > > I suggest you first take a look at the user group mail
> > > > > archives
> > > > >
> > > > > http://ferret.wrc.noaa.gov/Ferret/Mail_Archives/ferret_mail_archives.html
> > > > >
> > > > > where for example in 2001 you will find at least two notes that may
> > > > > deal with a situation like yours:
> > > > > http://ferret.wrc.noaa.gov/Ferret/Mail_Archives/fu_2001/msg00649.html
> > > > > http://ferret.wrc.noaa.gov/Ferret/Mail_Archives/fu_2001/msg00650.html
> > > > >
> > > > > In general it may be useful to be more specific in your query -
> > > > > for example describe the format of your dataset, cut and paste lines
> > > > > from your ferret session that show how you accessed the data and plotted
> > > > > it, etc.
> > > > >
> > > > > Mick
> > > > >
> > > > > |--****--****-*---*---***--***--|____spillane@pmel.noaa.gov____|
> > > > > |-*__---*-----*--*-*--*--*-*--*-|_SCIENCE APPLICATIONS SUPPORT_|
> > > > > |--***--*-----*-*---*-***--***--|____EPIC/Ferret/PlotPlus______|
> > > > > |-----*-*-----*-*****-*----*----|__Room 2070 Bldg#3 NOAA/PMEL__|
> > > > > |-****---****-*-*---*-*----*----|____Phone_:_(206)526-6780_____|
> > > >
> > > > --
> > > > Yunyao "Frank" Qu
> > > > Satellite Active Archive (SAA)
> > > > NESDIS, NOAA
> > > >
> > > >
> >
> > --
> > Yunyao "Frank" Qu
> > Satellite Active Archive (SAA)
> > NESDIS, NOAA
> >
> > ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
> > Name: 20011231_200000.cdf
> > 20011231_200000.cdf Type: Channel File (application/x-cdf)
> > Encoding: base64
> > Download Status: Not downloaded with message
>
> --
> Ansley Manke Pacific Marine Environmental Laboratory Seattle WA (206)526-6246--
Yunyao "Frank" Qu
Satellite Active Archive (SAA)
NESDIS, NOAA
--
Ansley Manke Pacific Marine Environmental Laboratory Seattle
WA (206)526-6246