[ncl-talk] doubt if data is being plottted
Mary Haley
haley at ucar.edu
Thu Jan 12 09:06:16 MST 2017
As we always tell people, use "print" to look at your data.
A simple do loop with some print statements should help you understand
what's happening in this range:
nlon = dimsizes(longitude(0,:))
nlat = dimsizes(latitude(:,0))
do nln=0,nlon-1
do nlt=0,nlat-1
if(latitude(nlt,nln) .ge.13.1.and.latitude(nlt,nln) .le.13.2.and.\
longitude(nlt,nln).ge.75.2.and.longitude(nlt,nln).le.75.4) then
print("==================================================")
print(" lat/lon = " + latitude(nlt,nln) + "/" +
longitude(nlt,nln))
print(" maxZ_Conv = " + maxZ_Conv(nlt,nln))
print(" nlt = " + nlt + " nln = " + nln)
end if
end do
end do
On Thu, Jan 12, 2017 at 4:32 AM, Geeta Geeta <geetag54 at yahoo.com> wrote:
> Thanks Mary for your explanation. Now I also feel that the data is plotted
> correctly.
>
> My question was " I can clearly see 4 or more points together in the
> figure in the lat range 13.1-13.2 and long range of 75.2-75.4E and they
> have reflectivity greater than 52dBz. These points are depicted in deep
> blue color. I am unable to relate these points in the output file that i
> sent you."
>
>
> than you once again.
>
> Geeta.
>
>
>
>
> On Thursday, 12 January 2017 1:39 AM, Mary Haley <haley at ucar.edu> wrote:
>
>
> Guido,
>
> Sorry for the delay in response. I'm in the middle of an NCL workshop
> right now.
>
> I'm not sure what you are asking for. You can see where you have valid
> data with the black dots, and the red dots represent locations where you
> don't have data. You will not get any filled squares over the red dots,
> because there simply isn't any data there.
>
> Since you are drawing raster contours. you will get a little square at
> every black dot, and it will be filled in a single color. Any squares that
> are dark blue represent data that is greater than or equal to 52.
>
> So, I'm still not sure what you are trying to indicate is wrong with the
> plot. Are you saying that at locations where you have red dots, that there
> should be data values there, and hence the dots should be black and there
> should be a filled square on top of it?
>
> I'm pretty sure NCL is plotting this data correctly. If you think there
> should be a filled square somewhere, please indicate exactly which square
> you think is wrong, and we can then look at the data at that point and see
> what's going on.
>
>
> --Mary
>
>
>
>
> On Sat, Jan 7, 2017 at 10:46 PM, Geeta Geeta <geetag54 at yahoo.com> wrote:
>
> Hi Mary.
> I could not send you the data because of some problem at my end. Apologies
> for that.
> I have now uploaded the data file of 2011-04-14.
>
> I am concentrating (enlarging) on the pixels between 13-14N/75-76E which I
> have plotted. (Fig attached)
> I can see 3-4 pixels having values greater than 52 dbZ within this lat-lon
> box which are adjacent. but I cannot find 4 pixels which are together and
> have values greater than 52dBZ in the data file which is also attached.
>
> I don't know what i am missing
>
>
>
> Geeta.
>
>
>
>
> On Tuesday, 27 December 2016 8:21 PM, Mary Haley <haley at ucar.edu> wrote:
>
>
> Geeta,
>
> I tried to run your script, but the data file appears to be bad.
>
> The issue is likely that your non-missing data may not be surrounded by
> enough non-missing points. If you have a single valid point or even two
> points surrounded on all sides by missing points, then no contour can be
> drawn at those points. If you have three dots, then you end up with a
> triangular contour that connects the three dots.
>
> If I could open your data file, then I was going to add a call to
> gsn_coordinates so I could draw a dot at all your missing and non-missing
> locations.
>
> I think you just need to add this to the end of your script:
>
> mkres = True
> mkres at gsMarkerSizeF = 2 ; try values from 2 to 30 to
> increase/decrease marker size
> mkres at gsMarkerIndex = 16 ; filled dot
>
> mkres at gsnCoordsNonMissingColor = "black"
> mkres at gsnCoordsMissingColor = "red"
> gsn_coordinates(wks,plot,maxZ, mkres)
>
> This should draw the missing data as red dots and the valid data as black
> dots. You will then be able to see if you have valid data that is not
> surrounded by enough other valid data.
>
> If your grid is really high-res, then you will just end up with a blob.
> You then may need to zoom in on your map area of interest a little to
> better see the area with the problem. You already have these settings:
>
> res at mpMinLatF = 13.00
> res at mpMaxLatF = 17.0
> res at mpMinLonF = 71.0
> res at mpMaxLonF = 77.0
>
> so to further zoom in on your map, you can change these to the desired
> lat/lon area.
>
> --Mary
>
>
> On Fri, Dec 16, 2016 at 7:36 AM, Geeta Geeta <geetag54 at yahoo.com> wrote:
>
> Thanks Mary for your valuable support. I really acknowledge that.
> The plot looks different and I could see the Deep Blue Dot too. Pls check
> from line no 5784 and around this (52.05)
>
> My question is that around this point there were non-zero values, so why
> gsn_countour_map is missing some of the data points. and what method of
> filling the contours is used by "cnFillOn = True" . I have attached my
> output with this.
> I am also sending the data at your ftp address.
>
> My data looks complete now because I can see each and every pixel.
>
> warm regards
> Geeta.
>
>
> On Thursday, 15 December 2016 10:56 PM, Mary Haley <haley at ucar.edu> wrote:
>
>
> Geeta,
>
> We cannot debug this by simply looking at data values and plots. I do see
> the issue that you are pointing out, but without seeing your code, I can't
> be sure of the problem.
>
> If your value that is 52.05 is surrounded by missing values, then I think
> it's possible it won't show up on the plot. You could try doing a raster
> plot:
>
> res at cnFillMode = "RasterFill"
>
> to see if this changes anything.
>
> If it doesn't, then if you could provide the data and a clean script, this
> would really help.
>
> thanks,
>
> --Mary
>
>
> On Thu, Dec 15, 2016 at 1:48 AM, Geeta Geeta <geetag54 at yahoo.com> wrote:
>
> dear ncl-talk
> I am using a variable called Z_fnl which is filtered for one condition and
> the it is called maxZ. Then maxZ is filtered for 2 conditions and is called
> maxZ_Conv.
>
> Below output is the LIne number ofthe file, Raintype, Lat, Lon, maxZ,
> maxZ_Conv (Output is attatched)
>
> 5843 (118,0) -9999 14.7402 75.7023 0 -9999
> 5844 (118,1) 152 14.7826 75.6777 23.09 -9999
> 5845 (118,2) 210 14.8249 75.6531 34.03 34.03
> 5846 (118,3) 200 14.8669 75.6286 52.05 52.05
> 5847 (118,4) 200 14.9083 75.6044 40.67 40.67
> 5848 (118,5) 210 14.9496 75.5802 22.34 22.34
> 5849 (118,6) 120 14.9906 75.5562 23.75 -9999
> 5850 (118,7) 210 15.0313 75.5322 24.61 24.61
> 5851 (118,8) 210 15.0716 75.5085 36.33 36.33
> 5852 (118,9) 210 15.112 75.4847 42.3 42.3
> Now I am plotting the Raintype and maxZ_Conv for a small area and for a
> large area.
>
> Now my problem is that it looks like that the plots and unable to
> correspond directly to the data. For example, in the line number at 5846
> which has highest value of 52.05 at the point 14.8669/75.6286 . Pls also
> check the lines from 5843-5852.
>
> (118,3) 200 14.8669 75.6286 52.05 52.05 seems to be missing in the
> plot. At this point the DEEP Blue color is missing.
>
> I m sorry for asking this basic question.
>
>
>
> ______________________________ _________________
> ncl-talk mailing list
> ncl-talk at ucar.edu
> List instructions, subscriber options, unsubscribe:
> http://mailman.ucar.edu/ mailman/listinfo/ncl-talk
> <http://mailman.ucar.edu/mailman/listinfo/ncl-talk>
>
>
>
>
>
>
>
>
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.ucar.edu/pipermail/ncl-talk/attachments/20170112/4d54a3e7/attachment.html
More information about the ncl-talk
mailing list