<html>
<head>
<meta content="text/html; charset=ISO-8859-1"
http-equiv="Content-Type">
</head>
<body bgcolor="#FFFFFF" text="#000000">
<font face="Times New Roman">Dear Martin and all,<br>
<br>
For 3-d ocean fields like thetao, I think the missing value (not
0.0) should invariably be assigned to all land grid cells and for
grid cells below the bottom of the ocean. I can't find where in
the document pointed to below it says otherwise. I would like to
correct the document, so please tell me where it needs revision.<br>
<br>
thanks,<br>
Karl<br>
</font><br>
On 12/4/11 8:07 PM, <a class="moz-txt-link-abbreviated" href="mailto:martin.juckes@stfc.ac.uk">martin.juckes@stfc.ac.uk</a> wrote:
<blockquote
cite="mid:E21FBC3F00D7304687CB46529F9676D71C837B88@EXCHMBX01.fed.cclrc.ac.uk"
type="cite">
<pre wrap="">I owe MRI an apology -- the use of zero instead of a missing value is in fact recommended in (<a class="moz-txt-link-freetext" href="http://cmip-pcmdi.llnl.gov/cmip5/docs/standard_output.xls">http://cmip-pcmdi.llnl.gov/cmip5/docs/standard_output.xls</a>),
regards,
Martin
________________________________________
From: Juckes, Martin (STFC,RAL,RALSP)
Sent: 04 December 2011 08:45
To: <a class="moz-txt-link-abbreviated" href="mailto:go-essp-tech@ucar.edu">go-essp-tech@ucar.edu</a>; Parton, Graham (STFC,RAL,SSTD); <a class="moz-txt-link-abbreviated" href="mailto:cmip5-users@jiscmail.ac.uk">cmip5-users@jiscmail.ac.uk</a>
Cc: Pascoe, Stephen (STFC,RAL,RALSP)
Subject: Incorrect use of missing value in some MRI ocean data.
Hello,
It has come to light that some MRI ocean data has land grid points set to zero rather than to the declared missing value. This is likely to confuse some applications reading and plotting the data -- I've notified Seiji Yukimoto at MRI,
sincerely,
Martin
________________________________________
From: Dennis Shea [<a class="moz-txt-link-abbreviated" href="mailto:shea@ucar.edu">shea@ucar.edu</a>]
Sent: 03 December 2011 19:45
To: Zhang Tao
Cc: Juckes, Martin (STFC,RAL,RALSP); <a class="moz-txt-link-abbreviated" href="mailto:haley@ucar.edu">haley@ucar.edu</a>; Gary Strand; Karl Taylor
Subject: Re: [ncl-talk] How to regrid CMIP5 ocean temp. data to regular 1X1 grid
CMIP5 users beware! :-(
---
The first rule of data processing is to 'look at the data.'
The problem is *not* with the cdo. Rather the source data file
from CMIP5 has incorrect information. I read the surface level
a =
addfile("thetao_Omon_MRI-CGCM3_piControl_r1i1p1_230101-230102.lite.nc","r")
ta = a->thetao(0,0,:,:)
printVarSummary(ta)
printMinMax(ta,0)
The variable has the following attributes
missing_value : 1e+20
_FillValue : 1e+20
I counted the number of number of _FillValue (nmsg1) and the
min/max temperatures.
(0) Sea Water Potential Temperature: min=0 max=303.924
(0) nmsg1=0
Obviously, a temperature of 0 is physically not correct. I printed
the array. No missing values (1e20) were encountered. What about land?
The actual _FillValue (missing_value) is 0.0. In an NCL
script I reset to 0.0
ta@_FillValue = 0.0
I counted the _FillValue and the min/max
(0) nmsg2=42671
(0) Sea Water Potential Temperature: min=271.293 max=303.924
The temperature range is much more reasonable.
So the cdo treated the original 0.0 as valid values rather than
_FillValue. I used the NCO operator, ncatted, to create a copy of the
original source file but altered the _FillValue to 0.0
%> ncatted -O -h -a _FillValue,thetao,o,f,0
thetao_Omon_MRI-CGCM3_piControl_r1i1p1_230101-230102.lite.nc
thetao_Omon_MRI_FillValue.nc
Then ran the cdo to generate the weights
%> cdo genbil,r360x180 thetao_Omon_MRI_FillValue.nc
thetao_Omon_MRI_bilwgt.nc
Then did the remap
%> cdo remap,r360x180,thetao_Omon_MRI_bilwgt.nc
thetao_Omon_MRI_FillValue.nc thetao_Omon.r360x180.nc
I works!!
See: <a class="moz-txt-link-freetext" href="http://www.cgd.ucar.edu/~shea/TaoZhang.png">http://www.cgd.ucar.edu/~shea/TaoZhang.png</a>
D
On 12/2/11 4:12 PM, Zhang Tao wrote:
</pre>
<blockquote type="cite">
<pre wrap="">Dennis,
I have attached 3 plots for you to take a look. MPI.pdf is the results
from MPI model, I use "cdo remapbil,ll1deg_grid.nc <input file> <output
file> " to get it.
I plot out the contour and shaded plots of MRI-CGCM3 which is obtained
from your command. clearly, the following command does not work
well with this model. I use "cdo remapbil,ll1deg_grid.nc <input file>
<output file> " for this model, similar results are obtained. There are
lots of unrealistically points in the land.
Tao
%> cdo genbil,r360x180
thetao_Omon_MRI-CGCM3_piControl_r1i1p1_230101-230102.lite.nc
thetao_Omon_MRI_bilwgt.nc
&> cdo remap,r360x180,thetao_Omon_MRI_bilwgt.nc
thetao_Omon_MRI-CGCM3_piControl_r1i1p1_230101-230102.lite.nc
thetao_Omon.r360x180.nc
On 12/2/11 3:49 PM, Dennis Shea wrote:
</pre>
<blockquote type="cite">
<pre wrap="">%> cdo genbil,r360x180
thetao_Omon_MRI-CGCM3_piControl_r1i1p1_230101-230102.lite.nc
thetao_Omon_MRI_bilwgt.nc
&> cdo remap,r360x180,thetao_Omon_MRI_bilwgt.nc
thetao_Omon_MRI-CGCM3_piControl_r1i1p1_230101-230102.lite.nc
thetao_Omon.r360x180.nc
ftp <a class="moz-txt-link-abbreviated" href="ftp://ftp.cgd.ucar.edu">ftp.cgd.ucar.edu</a>
anonymous
email
cd pub/shea/ZhangTao
prompt
mget *
quit
=================
10997404 Dec 2 15:46 thetao_Omon_MRI_bilwgt.nc
1050660 Dec 2 15:47 thetao_Omon.r360x180.nc
Hope it worked.
I'm outta her!
D
On 12/02/2011 03:00 PM, Zhang Tao wrote:
</pre>
<blockquote type="cite">
<pre wrap="">Dennis,
Could you write down the detailed CDO command to remap Japanese model?
I can make a test.
Thanks,
Tao
</pre>
</blockquote>
</blockquote>
</blockquote>
</blockquote>
</body>
</html>