Re: monthly mean climatology

[1] It is surprising that you did not get errors immediately
     because 'min' and 'max' are NCL builtin functions. They
     can not be redefined. EG

ncl 0> timeUnits = "hours since 1997-01-01 00:00:00"

        The following trys to redefine 'min' .... not allowed

ncl 1> min = ut_inv_calendar( 1999, 04, 01, 00, 00, 00, timeUnits, 0 )
fatal:syntax error: line 1 before or near =
min =
-----^

fatal:error in statement

===========================================================================

[2]

ncl 2> mint = ut_inv_calendar( 1999, 04, 01, 00, 00, 00, timeUnits, 0 )
ncl 3> maxt = ut_inv_calendar( 2008, 08, 31, 00, 00, 00, timeUnits, 0 )

[3] I have no idea why you are doing the following,
     even using 'mint' and 'maxt'

    time = ut_convert( fin->time({mint:maxt}), timeUnits )

[4] You could do this directly

    x = short2flt(fin->sst({mint:maxt},{-30:30},:))
    printVarSummary(x)

    yyyymmdd = cd_convert( x&time, -3) ; information only
    print( yyyymmdd )

    xMonthAvg = calculate_monthly_values(x, "avg", 0, False)
    printVarSummary(xMonthAvg)

    xMonthClm = clmMonTLL( xMonthAvg )
    printVarSummary(xMonthClm)

On 1/23/13 4:51 PM, Xi Chang wrote:
> Thanks for all kind responses,
>
> I did a test something like this:
>
> f = addfile( sst.nc, "r" )
> timeUnits = "hours since 1997-01-01 00:00:00"
> min = ut_inv_calendar( 1999, 04, 01, 00, 00, 00, timeUnits, 0 )
> max = ut_inv_calendar( 2008, 08, 31, 00, 00, 00, timeUnits, 0 )
> time = ut_convert( fin->time({min:max}), timeUnits )
> lat = fin->lat ({-30:30})
> lon = fin->lon
>
> x = short2flt(fin->sst({min:max},{-30:30},:))
> xMonthAve = calculate_monthly_values(x, "avg", 0, False)
>
> then, it produced this error:
> *fatal:Execute: Error occurred at or near line 11768 in file
> $NCARG_ROOT/lib/ncarg/nclscripts/csm/contributed.ncl*
>
> If I do not set the range of the time, everything is just ok.
>
> Xi.
>
>
>
>
> On Thu, Jan 24, 2013 at 8:35 AM, Will Hobbs <Will.Hobbs@utas.edu.au> wrote:
>
>> Xi
>>
>> Obviously, if you want the monthly climatology you could just take the
>> mean for each month calculated using Dennis' function.
>> Or even easier, assuming you have a valid time array:
>>
>> > timetmp = cd_calendar(time, -5) ;timetmp is an integer array with
>> (/year,month,day,hour,min,sec/) foir each time step
>>
>> >clim = new(12, typeof(x))
>>
>> > do i = 1,12
>> > ii = ind(timetmp(:,1).eq.i) ;all values in month 'i'
>>> clm(i-1) = avg(ii) ;monthly climatology is
>> the average of all values in month 'i'
>>> delete(ii)
>>> end do
>>
>>
>> Will
>>
>> From: Xi Chang <xi.chang01@gmail.com>
>> Date: Thursday, 24 January 2013 10:19 AM
>> To: "ncl-talk@ucar.edu" <ncl-talk@ucar.edu>
>> Subject: Re: monthly mean climatology
>>
>> Thank you for the hints. It worked fine but it just simply returned the
>> values from daily to monthly *not to monthly climatology*. If anyone has
>> an idea how to figure it out, i'll be really appreciate it.
>>
>> Xi.
>>
>>
>> On Wed, Jan 23, 2013 at 10:22 AM, Dennis Shea <shea@ucar.edu> wrote:
>>
>>> There are several approaches
>>>
>>> [1] http://www.ncl.ucar.edu/**Support/talk_archives/2012/**1256.html<http://www.ncl.ucar.edu/Support/talk_archives/2012/1256.html>
>>>
>>> [2] there is an unsupported and undocumented function in
>>> contributed.ncl: calculate_monthly_values
>>>
>>> It requires a time coordinate variable associated
>>> with the input variable 'x'. This time coordinate
>>> variable must be udunits compatible:
>>>
>>> "days since ...", "hours since ...", "seconds since ..."
>>>
>>> See cd_calendar and cd_inv_calendar
>>>
>>> ;-----------------------------**----
>>> ; undocumented and unsupported
>>> ;-----------------------------**----
>>> undef ("calculate_monthly_values")
>>> function calculate_monthly_values (x:numeric, arith:string,
>>> nDim[1]:integer, opt[1]:logical)
>>>
>>> ; calculate monthly values [avg, sum, min, max]
>>> ; x: numeric array of 5D or less [eg: time,lev,lat,lon]
>>> ; *must* have time coordinate recognized by cd_calendar
>>> ; if 5D [case,time,lev,lat,lon]
>>> ; arith: "avg" [also, "ave"], "sum","min","max" others may be added
>>> later
>>> ; nDim : scalar integer that specifies the 'time' dimension [generally, 0]
>>> ; opt : option ... not used here
>>> ;
>>> ; Sample usage: x(time,lat,lon) where time are n-hrly or daily values.
>>> ; xMonthAvg = calculate_monthly_values(x, "avg", 0, False)
>>> ; xMonthSum = calculate_monthly_values(x, "sum", 0, False)
>>> ; xMonthMin = calculate_monthly_values(x, "min", 0, False)
>>> ; xMonthMax = calculate_monthly_values(x, "max", 0, False)
>>> ;
>>> ; It is assumed that there will be multiple elements for the dim_avg_n
>>> ; calculation.
>>> ;
>>>
>>>
>>> On 1/22/13 4:32 PM, Xi Chang wrote:
>>>
>>>> Hallo NCL users,
>>>>
>>>> does anyone has a trick how to calculate monthly mean climatology from
>>>> daily dataset?
>>>>
>>>> Thank you for the help.
>>>>
>>>>
>>>>
>>>> ______________________________**_________________
>>>> ncl-talk mailing list
>>>> List instructions, subscriber options, unsubscribe:
>>>> http://mailman.ucar.edu/**mailman/listinfo/ncl-talk<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 Wed Jan 23 18:10:34 2013