Re: subscripts do not match number of dimensions]

This is correct.

Also, it is better to use "eofunc". It s faster.

http://www.ncl.ucar.edu/Document/Functions/Built-in/eofunc.shtml

or, if meta data is to be preserved,

http://www.ncl.ucar.edu/Document/Functions/Contributed/eofunc_Wrap.shtml

Rick Brownrigg wrote:
> Hi,
>
> I *think* this originates from this line of your script:
>
> xa3 = eof2data(eofa(0:2,:),eof_tsa(0:2,:)) ; constract array using
>
> If I understand correctly, "eofa" is a 3-D array, and if so, a
> reference such as
> "eofa(0:2,:)" is under-specified (??)
>
> Rick
>
> On Fri, 27 Nov 2009 15:45:23 +0100 (CET)
> siswanto@climate.unibe.ch wrote:
>> Hi everyone,..
>>
>> Iam very new in using NCL. I tried to apply EOF analysis using both
>> controlrun1990 and freshwater hosing experiment flux004 (CCSM3
>> output). My
>> final goal is to see the changes in PDO pattern during the hosing
>> exp.
>> Here I projecting EOF pattern of control run into SST flux, but I
>> have
>> some bugs in the script and the error massage such as :
>>
>> fatal:Number of subscripts do not match number of dimensions of
>> variable,(2) Subscripts used, (3) Subscripts expected
>> fatal:Execute: Error occurred at or near line 78 in file
>> ProjectedEOF.ncl
>>
>> Could you please give me some clues to solve this problem
>>
>> Many thanks in advance
>> Best wishes,
>>
>> Siswanto
>>
>> ----------------------------------------------------
>> My run look like this :
>>
>> University Corporation for Atmospheric Research
>> NCAR Command Language Version 5.0.0
>> The use of this software is governed by a License Agreement.
>> See http://www.ncl.ucar.edu/ for more details.
>>
>>
>> Variable: ssta
>> Type: float
>> Total Size: 259459200 bytes
>> 64864800 values
>> Number of Dimensions: 3
>> Dimensions and sizes: [time | 1001] x [lat | 180] x [lon | 360]
>> Coordinates:
>> lat: [-89.5..89.5]
>> lon: [ 0..359]
>> Number Of Attributes: 3
>> _FillValue : -999
>> missing_value : -999
>> spatial_op : Bilinear remapping: 1st order: destarea: NCL:
>> /home/siswanto/script/map_gx3v5_to_1x1d_bilin_da_040122.nc
>>
>>
>> Variable: sstb
>> Type: float
>> Total Size: 51321600 bytes
>> 12830400 values
>> Number of Dimensions: 3
>> Dimensions and sizes: [time | 198] x [lat | 180] x [lon | 360]
>> Coordinates:
>> lat: [-89.5..89.5]
>> lon: [ 0..359]
>> Number Of Attributes: 3
>> _FillValue : -999
>> missing_value : -999
>> spatial_op : Bilinear remapping: 1st order: destarea: NCL:
>> /home/siswanto/script/map_gx3v5_to_1x1d_bilin_da_040122.nc
>>
>>
>> Variable: vara
>> Type: float
>> Total Size: 21029008 bytes
>> 5257252 values
>> Number of Dimensions: 3
>> Dimensions and sizes: [lat | 52] x [lon | 101] x [time | 1001]
>> Coordinates:
>> lat: [19.5..70.5]
>> lon: [140..240]
>> Number Of Attributes: 3
>> _FillValue : 1e+35
>> missing_value : -999
>> spatial_op : Bilinear remapping: 1st order: destarea: NCL:
>> /home/siswanto/script/map_gx3v5_to_1x1d_bilin_da_040122.nc
>>
>>
>> Variable: varb
>> Type: float
>> Total Size: 4159584 bytes
>> 1039896 values
>> Number of Dimensions: 3
>> Dimensions and sizes: [lat | 52] x [lon | 101] x [time | 198]
>> Coordinates:
>> lat: [19.5..70.5]
>> lon: [140..240]
>> Number Of Attributes: 3
>> _FillValue : 1e+35
>> missing_value : -999
>> spatial_op : Bilinear remapping: 1st order: destarea: NCL:
>> /home/siswanto/script/map_gx3v5_to_1x1d_bilin_da_040122.nc
>> warning:dim_rmvmean: 1565 rightmost sections of the input array
>> contained
>> all missing values
>> warning:dim_rmvmean: 1565 rightmost sections of the input array
>> contained
>> all missing values
>>
>>
>> Variable: eofb
>> Type: float
>> Total Size: 105040 bytes
>> 26260 values
>> Number of Dimensions: 3
>> Dimensions and sizes: [evn | 5] x [lat | 52] x [lon | 101]
>> Coordinates:
>> evn: [1..5]
>> lat: [19.5..70.5]
>> lon: [140..240]
>> Number Of Attributes: 7
>> eval : ( 384.5412, 105.1702, 100.4963, 68.89677, 53.60186 )
>> pcvar : ( 38.52052, 10.53518, 10.06699, 6.901573, 5.369441 )
>> trace : 998.2763
>> pcrit : 95
>> eof_function : 2
>> _FillValue : 1e+35
>> matrix : covariance
>> fatal:Number of subscripts do not match number of dimensions of
>> variable,(2) Subscripts used, (3) Subscripts expected
>> fatal:Execute: Error occurred at or near line 78 in file
>> ProjectedEOF.ncl
>> -----------------------------------------------------------
>> My script is as following :
>> ;******************************************************************
>> ; EOF analysis for North Pacific Region (Control projection to Flux)
>> ; Siswanto
>> ;******************************************************************
>> load "$NCARG_ROOT/lib/ncarg/nclscripts/csm/gsn_code.ncl" ; High
>> Level
>> load "$NCARG_ROOT/lib/ncarg/nclscripts/csm/gsn_csm.ncl" ; plot
>> interfaces
>> load "$NCARG_ROOT/lib/ncarg/nclscripts/csm/shea_util.ncl"
>> load "$NCARG_ROOT/lib/ncarg/nclscripts/csm/contributed.ncl"
>>
>> begin
>>
>> a =
>> addfile("/data03/siswanto/dataocn/regridded/annual_means/flux004_TEMP_1x1d_y_439-1439.nc","r")
>> ; data file flux004
>> b =
>> addfile("/data03/siswanto/ctrl_1990/ocn/annual_means/regrided/ctrl_1990_TEMP_1x1d_y_933-1130.nc","r")
>> ; data file control run
>> neval = 5 ; # eigenvectors to compute
>> ssta = a->TEMP(:,:,:) ;change variable name
>> sstb = b->TEMP(:,:,:) ;variable in control run
>> dataa = ssta(:,109:160,140:240) ;and specify the area of file a;
>> datab = sstb(:,109:160,140:240) ;and specify the area of file b;
>> printVarSummary(ssta)
>> printVarSummary(sstb)
>> ;;;;;;;;;;;;;;;;;;;;;do not change for most of
>> cases;;;;;;;;;;;;;;;;;;;;;;;;;;
>> dataa@_FillValue = 1e+35 ; missing value
>> datab@_FillValue = 1e+35 ; missing value
>>
>> vara = dataa(lat|:,lon|:,time|:) ; reorder so time is last dim
>> varb = datab(lat|:,lon|:,time|:) ; reorder so time is last dim
>> vara@_FillValue = 1e+35 ; missing value
>> varb@_FillValue = 1e+35 ; missing value
>> printVarSummary(vara)
>> printVarSummary(varb)
>>
>> vara=dim_rmvmean(vara) ; remove mean flux
>> varb=dim_rmvmean(varb) ; remove mean control run
>> dims=dimsizes(vara) ; get dim sizes for variable a
>> dims=dimsizes(varb)
>> lat=vara&lat ; assign latitudes to variable
>> lon=vara&lon
>> ;time=vara&time
>> lat=varb&lat
>> lon=varb&lon
>> ;time=varb&time
>> do nl=0,dims(0)-1 ; wgt by sqrt(cos(latitude))
>> vara(nl,:,:) =
>> vara(nl,:,:)*doubletofloat(sqrt(cos(lat(nl)*0.0174532)))
>> end do
>> do nl=0,dims(0)-1 ; wgt by sqrt(cos(latitude))
>> varb(nl,:,:) =
>> varb(nl,:,:)*doubletofloat(sqrt(cos(lat(nl)*0.0174532)))
>> end do
>> ;printVarSummary(dims)
>>
>> timesa=ispan(0,1001,1)
>> yrsa=ispan(0,1000,1)
>> timesb=ispan(0,199,1)
>> yrsb=ispan(0,198,1)
>>
>> ;;;;;;;;;;;;;
>>
>> eofa=eofcov_pcmsg_Wrap(vara,neval,95.) ; calculate eof
>> eofb=eofcov_pcmsg_Wrap(varb,neval,95.) ; calculate eof
>> eofa@_FillValue=1e+35 ; missing value
>> eofb@_FillValue=1e+35 ; missing value
>> eof_tsa = eofcov_ts_Wrap(vara,eofa) ; calculate time series of eof
>> file a
>> eof_tsb = eofcov_ts_Wrap(varb,eofb) ; calculate time series of eof
>> file b
>> eof_tsa@_FillValue=1e+35 ; missing value
>> eof_tsb@_FillValue=1e+35 ; missing value
>> eof_tsc = eofcov_ts_Wrap(vara,eofb) ; calculate time series of eof
>> control
>> run projected into flux004 sst
>> eof_tsc@_FillValue=1e+35 ; missing value
>> ;;;;;;;;;;;;;;;;;
>>
>> printVarSummary(eofb)
>>
>> ;reconstruction data using eof
>>
>> do n=0,neval-1
>> eof_tsa(n,:)= eof_tsa(n,:)+eof_tsa@ts_mean(n) ; add time series
>> mean
>> of eof file a
>> end do
>>
>> xeof= eof2data(eofa, eof_tsa) ; reconstruct
>> orriginal array
>> xa3 = eof2data(eofa(0:2,:),eof_tsa(0:2,:)) ; constract array
>> using
>> first 3 EOF using EOF file b
>> xb3 = eof2data(eofb(0:2,:),eof_tsb(0:2,:))
>> xc3 = eof2data(eofb(0:2,:),eof_tsa(0:2,:))
>>
>> ; Projected control PDO mode into Flux SST
>>
>>
>> do n=0,neval-1
>> do time=0,1000
>> do lat=19.5,70.5
>> do lon=140,240
>> PC(n,:) = eofb(n,lat,lon)*vara(lat,lon,time)
>> end do
>> end do
>> end do
>> end do
>> ;printVarSummary(PC)
>>
>> wks = gsn_open_wks("ps" ,"NorthPacific_projection_sst1") ;
>> open ps
>> file
>>
>> plot = new(4, "graphic")
>> pres = True ; mod panel plot
>> pres@gsnDraw=True
>> pres@gsnFrame=True
>> pres@gsnMaximize = True ; blow up plot
>> pres@gsnPaperWidth=8.5
>> pres@gsnPaperHeight=11.0
>> pres@txString = "North Pacific 20N-70N 140E-150W EOF SSTA
>> Projected
>> with Control Run"
>> pres@txFontHeightF = .005
>> pres@gsnPaperOrientation="portrait"
>> pres@gsnPaperMargin=0.02
>> pres@gsnPanelLeft = 0.02
>> pres@gsnPanelRight = 0.98
>> pres@gsnPanelBottom = .02
>> pres@gsnPanelTop = .98
>> pres@gsnPanelYWhiteSpacePercent = 2
>> pres@gsnPanelXWhiteSpacePercent = 5
>> gsn_panel(wks,plot3,(/4,3/),pres) ; create panel plot
>>
>> resxy = True ; xy plot mods
>> desired
>> resxy@gsnDraw = False ; don't draw yet
>> resxy@gsnFrame = False ; don't advance
>> frame
>> yet
>> resxy@vpWidthF = 0.80 ; plot width
>> resxy@vpHeightF= 0.50 ; plot height
>> resxy@tiYAxisString=""
>> do ne=0,3
>> pres@txString = "North Pacific 20N-70N 140E-150W EOF SSTA 0-1000
>> Yrs
>> (Int 1000 Yrs) projected with control run"
>> resxy@gsnLeftString = "Covariance" ; titles
>> resxy@gsnCenterString = "EOF "+(ne+1)
>> resxy@gsnRightString = "%Var=" + sprintf("%4.1f",
>> eofb@pcvar(ne))
>> plot(ne) = gsn_csm_xy(wks,yrsa,proj_tsc(ne,:), resxy)
>> end do
>> gsn_panel(wks,plot3(0:3),(/4,1/),False) ; draw: 1-across,
>> 3-down
>>
>> end
>>
>>
>>
>> Siswanto
>> Climate and Environmental Physics siswanto@climate.unibe.ch
>> Physics Institute, University of Bern ph: +41 (0)31 631 48 71
>> Sidlerstrasse 5, 3012 Bern, Switzerland
>>
>>
>>
>> Siswanto
>> Climate and Environmental Physics siswanto@climate.unibe.ch
>> Physics Institute, University of Bern ph: +41 (0)31 631 48 71
>> Sidlerstrasse 5, 3012 Bern, Switzerland
>>
>>
>> _______________________________________________
>> ncl-talk mailing list
>> List instructions, subscriber options, unsubscribe:
>> http://mailman.ucar.edu/mailman/listinfo/ncl-talk
>
> _______________________________________________
> ncl-talk mailing list
> List instructions, subscriber options, unsubscribe:
> http://mailman.ucar.edu/mailman/listinfo/ncl-talk
_______________________________________________
ncl-talk mailing list
List instructions, subscriber options, unsubscribe:
http://mailman.ucar.edu/mailman/listinfo/ncl-talk
Received on Fri Nov 27 12:24:07 2009