[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
20010815: 20010814: GEMPAK Barnes upper-air analysis of CAPE?
- Subject: 20010815: 20010814: GEMPAK Barnes upper-air analysis of CAPE?
- Date: Wed, 15 Aug 2001 16:13:33 -0600
Christian,
Let me clarify your question:
>Another related question, how can I have less-smooth contours with
>Barnes analysis (is it the gamma? If so, what can I use to still get a good
>analysis and get some more details?)
It is the display programs that draw "contours". Barnes analysis creates
a gridded data set from irregularly spaced observations.
There are several factors which affect drawing smooth contours.
1) In gridding your data, it is GAMMA that affects the convergence
of dataand NPASS which create successive corrections of the data.
SEARCH is the weighting factor which determines at what weight
stations will not be included. Indirectly related to SEARCH is the
size of the grid spacing (DELTAN). Having a smaller grid point spacing
allows more of the high frequency data signal to be resolved. You
must have at least 3 points within the radius of influence of
a point in order to produce a grid point value. OAGRID is a
helpful program for automatically determining DELTAN, but it
does have linmitations, such as always creating a CED grid.
When you know more about your data and want to directly specify
the grid point spacing, you can use gdcfil.
Sucessive passes with NPASS interpolate the gridpoint back to the station
locations and decrease the mean square difference, so additional passes
restore some of the data noise that the original analysis smoothed out.
2) in contouring programs, the parameters CONTUR and CURVE affect the
smoothness of the contours drawn though the gridded data. For smoother
contours, increase the number of subboxes and smoothing passes in CONTUR.
You can draw smooth contours through noisy data- which is different from
your intemation about less-smooth analysis.
3) You can apply smoothers in your GFUNC such as SM5S and SM9S
to smooth noisy data as well. These are useful with fields like OMEG
which have lots of local noise in the data, as compared to HGHT fields
which are generally smooth if the data is hydrostatic.
The tutorial covers the barnes scheme:
http://www.unidata.ucar.edu/packages/gempak/tutorial/barnes.html
Effects of GAMMA and SEARCH are shown in the excercises:
http://www.unidata.ucar.edu/packages/gempak/tutorial/barnes_exercises.html
Steve Chiswell
Unidata User Support
>From: Christian Page <address@hidden>
>Organization: UCAR/Unidata
>Keywords: 200108152103.f7FL3A118217
>
>Hi,
>
>Thanks, it works well now. The problem was that I didn't specify SNPARM when
>needed. Another related question, how can I have less-smooth contours with
>Barnes analysis (is it the gamma? If so, what can I use to still get a good
>analysis and get some more details?)
>
>Christian Page
>UQAM
>
>On Tue, 14 Aug 2001, Unidata Support wrote:
>
>>
>> Christian,
>>
>> If you are going to use OAGRID to create the grid file, you
>> need to specify a SNPARM (eg tmpc; dwpc; hght; uwnd or vwnd)
>> so that the program can determine the mean station spacing of reporting
>> locations (set the DELTA_ values to blank prior to running. Otherwise,
>> if you already know the DELTA values you want, you can use
>> GDCFIL to create the grid file. I think your FL -13 warning is OK- it
>> likely is emmited be checking for the alias first
>>
>> Then, in OABSND, you need at least one level parameter (SNPARM) in addition
>> to your STNDEX values you will grid.
>>
>> For example:
>> GDFILE = test.grd
>> DELTAN =
>> DELTAX =
>> DELTAY =
>> GRDAREA = us
>> EXTEND = 2;2;2;2
>> DTAAREA =
>> SOURCE = sn
>> SNFILE = uair
>> SFFILE =
>> SNPARM = tmpc;dwpc
>> SFPARM =
>> DATTIM = 1200
>> LEVELS = 500
>> MAXGRD = 1000
>> GEMPAK-OAGRID>r
>>
>> OBJECTIVE ANALYSIS PARAMETERS
>>
>> GRIDAREA 19.00 -119.00 46.90 -57.62 (us 47.00 -56.
> 00)
>> DELTAN 3.730
>> DELTAX 1.860
>> DELTAY 1.860
>>
>> GRID POINTS X : 34 ( 2, 2 ) Y : 16 ( 2, 2 )
>> EXTENDAREA 15.28 -122.72 50.62 -53.90
>> DATAAREA 15.28 -122.72 50.62 -53.90
>>
>>
>> CALCULATIONS:
>> FILE $GEMDATA/upperair/20010814_upa.gem
>> SOURCE SN 19.00 -119.00 47.00 -56.00 NSTN
> 79
>> PARM TMPC LEVEL 0.0 DATTIM 010814/1200
>>
>> COMPUTED DATA SPACING 2.548
>> UNIFORM DATA SPACING 4.904
>> RECOMMENDED DELTAN = 3.726 DELTAX = 2.221 DELTAY = 1.863
>>
>> Enter <cr> to accept parameters or type EXIT:
>> Parameters requested: GDFILE,DELTAN,DELTAX,DELTAY,GRDAREA,EXTEND,DTAAREA,
>> SOURCE,SNFILE,SFFILE,SNPARM,SFPARM,DATTIM,LEVELS,MAXGRD.
>> GEMPAK-OAGRID>
>>
>>
>> SNFILE = uair
>> GDFILE = test.grd
>> SNPARM = tmpc;dwpc
>> STNDEX = cape
>> LEVELS = 500
>> VCOORD = pres
>> DATTIM = 1200
>> DTAAREA =
>> GUESS =
>> GAMMA = 0.3
>> SEARCH = 50/EX
>> NPASS = 2
>> QCNTL =
>> GEMPAK-OABSND>r
>>
>> OABSND PARAMETERS:
>>
>> Output grid file: test.grd
>> Number of x, y points: 34 16
>> Grid bnds: 19.00 -119.00 46.90 -57.62 Grid xtnd: 2 2 2 2
>> Station spacing: 3.73
>> No first guess file is requested.
>> Sounding file: $GEMDATA/upperair/20010814_upa.gem
>> Data area: 15.28;-122.72;50.62;-53.90
>> Level parameters: TMPC DWPC
>> QC threshold: 0.0 0.0
>> Levels: 500
>> Vertical coordinate: pres
>> Station parameters: CAPE
>> Time: 010814/1200
>> GAMMA: 0.30 SEARCH: 50.00 NPASS: 2
>> Data extrapolation may occur.
>> Enter <cr> to accept parameters or type EXIT:
>>
>> Processing data at time 010814/1200
>>
>> RMS values for pass 1
>> PARAMETER: TMPC DWPC CAPE
>> LEVEL: 500 500 0
>> RMS: 1.16 8.21 397.47
>> #STN: 79 79 79
>>
>> RMS values for pass 2
>> PARAMETER: TMPC DWPC CAPE
>> LEVEL: 500 500 0
>> RMS: 0.64 4.99 229.18
>> #STN: 79 79 79
>> Parameters requested: SNFILE,GDFILE,SNPARM,STNDEX,LEVELS,VCOORD,DATTIM,
>> DTAAREA,GUESS,GAMMA,SEARCH,NPASS,QCNTL.
>> GEMPAK-OABSND>
>>
>>
>> You should now be able to plot your cape grid using
>> GLEVEL = 0, GVCORD = NONE, GFUNC = CAPE.
>>
>> Steve Chiswell
>> Unidata User Support
>>
>>
>>
>>
>> >From: Christian Page <address@hidden>
>> >Organization: UCAR/Unidata
>> >Keywords: 200108141425.f7EEPK125035
>>
>> >
>> >Hi,
>> >
>> >I am trying to create a Barnes upper-air analysis of CAPE, but I am still
>> >struggling with OAGRID. I tried to do the same commands that I use for heig
> ht
>> >analysis, but it doesn't work.
>> >
>> >Here is what I get from OAGRID:
>> > GDFILE Grid file
>> >/io/httpd/gempak/grids/europe_upa_cape.grid
>> > DELTAN Station spacing 3.50
>> > DELTAX X spacing 1.75
>> > DELTAY Y spacing 1.75
>> > GRDAREA Area covered by grid 26;-20;62;40
>> > EXTEND Points to extend grid 2;2;2;2
>> > DTAAREA Data area for OA 26;-20;62;40
>> > SOURCE Data source (SN or SF) SN
>> > SNFILE Sounding data file $GEMUPA/20010813_upa.gem
>> > SFFILE Surface data file
>> > SNPARM Sounding parameter list
>> > SFPARM Surface parameter list
>> > DATTIM Date/time 010813/00
>> > LEVELS Vertical levels 500
>> > MAXGRD Maximum number of grids 5000
>> >
>> >$GEMUPA/20010813_upa.gem is defined correctly and is a valid file.
>> >
>> >I then get: [FL -13] No valid file for given type and cycle time.
>> >/io/httpd/gempak/grids/europe_upa_cape.grid is non-existent prior to runnin
> g
>> >oagrid, but it is create anyway even with the error.
>> >
>> >Then, oabsnd:
>> > SNFILE Sounding data file $GEMUPA/20010814_upa.gem
>> > GDFILE Grid file
>> >/io/httpd/gempak/grids/europe_upa_cape.grid
>> > SNPARM Sounding parameter list
>> > STNDEX Stability indices cape
>> > LEVELS Vertical levels 500
>> > VCOORD Vertical coordinate type PRES
>> > DATTIM Date/time 010814/00
>> > DTAAREA Data area for OA 26;-20;62;40
>> > GUESS Guess file*time
>> > GAMMA Convergence parameter 0.3
>> > SEARCH Search radius/Extrapolation 20/EX
>> > NPASS Number of passes 2
>> > QCNTL Quality control threshold
>> >
>> >
>> >then gdcntr (failure):
>> > GDATTIM Grid date/time 010813/00
>> > GLEVEL Grid level ALL
>> > GVCORD Grid vertical coordinate PRES
>> > GFUNC Scalar grid cape
>> > GDFILE Grid file
>> >/io/httpd/gempak/grids/europe_upa_cape.grid
>> > CINT Contour interval/min/max 0
>> > LINE Color/type/width/label/smth/fltr 15 / 3 / 2
>> > MAP Map color/dash/width/filter flag 1
>> > TITLE Title color/line/title
>> > DEVICE Device|name|x size;y size|color PS | eur_cape.ps | 11;8.5 | C
>> > SATFIL Satellite image filename(s)
>> > RADFIL Radar image filename(s)
>> > PROJ Map projection/angles/margins|dr nps
>> > GAREA Graphics area #45;14;13;23
>> > CLEAR Clear screen flag yes
>> > PANEL Panel loc/color/dash/width/regn 0
>> > TEXT Size/fnt/wdth/brdr/N-rot/just/hw 1
>> > SCALE Scalar scale / vector scale 0
>> > LATLON Line color/dash/width/label/inc 0
>> > HILO Color/symbol/rng/rad/cnt/intp 2;4/H#;L#///yes
>> > HLSYM HILO txt size/posn/font/wdth/hw 2;1/2/2/HW
>> > CLRBAR Color/ornt/anch/x;y/ln;wd/freq
>> > CONTUR Subbox/smooth 0
>> > SKIP Skip_cntr/skip_plt_x;skip_plt_y 0
>> > FINT Fill interval/min/max 0
>> > FLINE Fill colors 10-20
>> > CTYPE Contour type: C/F C
>> > LUTFIL Enhancement lookup table filenam
>> > STNPLT Txtc/txt attr|marker attr|stnfil
>> >
>> >[TG -4] No times in input file.
>> > [TG -4] No times in input file.
>> > [GR -20] No times selected.
>> >
>> >Any idea what's wrong? I tried to do the same as my height objective analys
> is,
>> >but it doesn't work...
>> >
>> >Christian Page
>> >UQAM
>> >
>>
>> ****************************************************************************
>> Unidata User Support UCAR Unidata Program
>> (303)497-8644 P.O. Box 3000
>> address@hidden Boulder, CO 80307
>> ----------------------------------------------------------------------------
>> Unidata WWW Service http://www.unidata.ucar.edu/
>> ****************************************************************************
>>
>