Search:
From tk@GFDL.GOV Wed Nov 12 14:21:49 1997
To: spangler@ncar.ucar.edu
Subject: Trenberth's NH SLP data
Dear Will,
I recently downloaded Trenberth's NH SLP data set (ds010.1). I think
there may be a discontinuity at one of the locations, and I wanted to
just bring this to your attention. The location in question is 140W,70N
and the time period in question is from about 1904-1909. The problem
can be seen using a grads version of the data if you do a difference between
SLP at that point and a neighboring point (e.g., 145W, 70N). Instead of
typical differences of 0 to +3mb, the differences in that period are
typically -5 to -8 mb.
Regards,
--
Tom Knutson
--------------------------------------------------------------------------
Geophysical Fluid Dynamics Lab /NOAA | phone: +1-609-452-6509
P.O. Box 308 | fax: +1-609-987-5063
Princeton University Forrestal Campus | e-mail: tk@gfdl.gov
Princeton, New Jersey 08542 U.S.A. | www: http://www.gfdl.gov/~tk
################################################################################
From minobe@geophys.hokudai.ac.jp Sat Feb 7 00:01:54 1998
To: spangler@ncar.ucar.edu
Subject: interpolation of discontinuity in ds010.1
Dear Will:
I would like to add a bit information about the discontinuity of ds.010.1
(Trenberth's SLP data). The problem was already mentioned by Dr. T. Knutson.
I see that the problem is on the grid-point at 65N 140W (not 70N as
noted by Dr. Knutson) for the version through Dec 1997. Most of the problem
may be solved by a simple zonal interpolation program shown below.
best regards,
parameter (im=72,jm=15,iystrt=1899,iyend=1997,maxmth=12)
parameter (latgvn=65,lnggvn=220)
parameter (jgvn=(latgvn-15)/5+1,igvn=lnggvn/5+1)
parameter (mthend=12)
dimension dim(im,jm)
undef=0.
print *,igvn,jgvn
open(11,file='slp.grd',form='unformatted',access='direct',
& recl=im*jm*4)
open(21,file='slp_.grd',form='unformatted',access='direct',
& recl=im*jm*4)
nrec=0
do 100 iy=iystrt,iyend
do 110 mth=1,maxmth
nrec=nrec+1
read(11,rec=nrec) dim
if (dim(igvn,jgvn) .ne. undef .and.
& dim(igvn+1,jgvn).ne. undef .and.
& dim(igvn-1,jgvn).ne. undef) then
dif1=dim(igvn+1,jgvn)-dim(igvn,jgvn)
dif2=dim(igvn,jgvn)-dim(igvn-1,jgvn)
if (abs(dif1).gt.3. .and. abs(dif2).gt.3. .and.
& dif1*dif2.lt.0.) then
print *,iy,mth
dim(igvn,jgvn)=(dim(igvn+1,jgvn)+dim(igvn-1,jgvn))*0.5
endif
endif
write(21,rec=nrec) dim
110 continue
if (iy.eq.iyend .and. mth.eq.mthend) exit
100 continue
close(11)
stop
end
--------------------------------------------------------------------
Shoshiro Minobe minobe@geophys.hokudai.ac.jp
Department of Geophysics, Graduate School of Science,
Hokkaido University, N-10 W-8, Sapporo 060, Japan.
Tel: +81-11-706-2644, Fax: +81-11-706-4907
################################################################################
From hanawa@pol.geophys.tohoku.ac.jp Wed Apr 1 21:39:38 1998
To: "'spangler@ncar.ucar.edu'"
Subject: February 1945 data of ds010.1
Dear Dr. Will Spangler,
We have noticed the existence of erroneous data in Trenberth's
N. Hem Sea Level Press, monthly (ds010.1) data set. That is,
it is found by mapping the SLP data of February 1945 show apparently
abnormal SLP distribution. One of reasons might be mis-arrangement
of grids. If one shift the grids of 40, 50, and 60N by two grids to east, then
one can obtain the plausible SLP distribution: clear Aleutian Low and
Icelandic Low, although there are still several erroneous grids in the map.
This will be also pointed out in our recent work (Hanawa and Yasuda, 1998,
which is submitted to J. Atmos. Oceanic Technol.)
Best regards, Kimio Hanawa
Dr. Kimio Hanawa
Department of Geophysics
Graduate School of Science
Tohoku University
Aoba-ku, Sendai 980-8578
TEL. 81-22-217-6526
FAX. 81-22-217-6530
E-MAIL. hanawa@pol.geophys.tohoku.ac.jp
################################################################################
From miyamoto@pol.geophys.tohoku.ac.jp Thu Jul 23 04:47:11 1998
To: joseph@ucar.edu
Subject: February 1945 data of ds010.1
Dear Dennis Joseph,
We have already informed that the data of February 1945 seem to be
erroneous. In order to confirm this and if possible to find some
correction method, we made the comparison between COADS-derived
monthly mean SLP field and the present dataset. As a result, we got a
conclusion that the present data at latitudes of 15, 25, 35, 45, 55, 65,
75, 85N should be shifted to the west by 10 degrees.
It is reported that the data at the grid of 65N, 220E are erroneous in
several months. Dr. S. Minobe suggested the method to remove the
erroneous data. That is, if this grid value is either lower or higher
than longitudinally adjacent data by 3 hPa, then the data can be
regarded as the erroneous one. However, we found that this method is
working well but it is not perfect. Actually, even after adopting this
method, we found that there are still erroneous data.
Best regards,
Kengo MIYAMOTO and Kimio HANAWA
Physical Oceanography Laboratory, Department of Geophysics,
Graduate School of Science, Tohoku University, Sendai 980-8578, JAPAN
TEL: +81-22-217-6527, FAX: +81-22-217-6530
E-MAIL:miyamoto@pol.geophys.tohoku.ac.jp
hanawa@pol.geophys.tohoku.ac.jp
################################################################################
From minobe@neptune.sci.hokudai.ac.jp Fri Nov 20 01:27:09 1998
To: spangler@ncar.ucar.edu
Subject: anomalous value in ds010.1
Dear Will:
This is to inform an erroneous value in SLP data, ds010.1
(Dr. Trenberth's N.Hem Sea Level Press, monthly).
The value at at 95E, 50N, Dec1974, is anomalously low as 956.3 mb,
whereas the SLP values at 90E and 100E (same latitutde and month)
are 1048.5 and 1048.1, respectively. The problem is commonly seen
in the data that I've made from the molydata.pbin (time stamp
of 11/18/98) using readpc.f, and also in slp.grd and slpx.grd
in the grads-directory with time stamp of 19 and 20 Aug 1998.
This anomalous datum can be corrected by an interpolation, or
possibly by adding 100 mb.
best regards,
Shoshiro Minobe,
--------------------------------------------------------------------
Shoshiro Minobe minobe@neptune.sci.hokudai.ac.jp
Division of Earth and Planetary Sciences, Graduate School of Science,
Hokkaido University, N-10 W-8, Sapporo 060-0810, Japan.
Tel: +81-11-706-2644, Fax: +81-11-706-4907
################################################################################
|
| Updated 22 July 2005 |