[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

20030624: Gempak pcconv query

John,

The pcconv.tbl functions are used for the point source (surface and
upperair) programs through SFPARM and SNPARM- so screaming PCCONV at a
grid program has no effect. ;-)

In the case of gridded data, the function $GEMPAK/source/gemlib/dg/dgdwpt.f
is used (and looks to calculate dewpoint from relative humidity or mixing
ratio....but not dewpoint depression). I made the necessary changes
to the source code here (to be distributed with 5.6.K).

I can send you a source code update if you built from source, or
we'll upgrade you to 5.6.K. Your choice. If you want the source
for the routine, let me know what version you are currently running
since I'll have to back path that code as the grid routines have all
been updated the past several releases to accomodate larger grids.

Steve
****************************************************************************
Unidata User Support                                    UCAR Unidata Program
303 497 8643                                                  P.O. Box 3000
address@hidden                                   Boulder, CO 80307
----------------------------------------------------------------------------
Unidata WWW Service              http://my.unidata.ucar.edu/content/support
****************************************************************************

On Thu, 19 Jun 2003, John Merrill wrote:

> Hello,
>    In using gridded model fields, presently nogaps analyses/6 hr
> forecasts archived by JOSS for ACE-Asia, the humidity
> variables seem to be "limited" by the use of DPDK as the available field
> from the archive. I see in pcconv.tbl that this
> should lead to dwpk being available, but I can't get to this or other
> more involved moisture parameters. I see formulations
> for the mixing ratio and other parameters which I want to plot, but no
> matter how loudly I yell PCCONV while my scripts
> are running, I get back "Input grid dwpk (or whatever) ... cannot be
> found. The hint "input grid," which I know it ain't, isn't
> sufficiently suggestive to help me, unfortunately. The variables
> available at each PRES level are tmpk, dpdk, hght, urel, vrel,
> omeg. Is there a way to force use of pcconv, or some other subtlety in
> cases like this?
>
>
>         John
>
> PS - You'd answered a simple question arising from my use of
> monthly-averaged Reanalysis grids. I was flummoxed by
> a simple syntax error, as you pointed out. I've had intermittent
> difficulties with girds from this and other GRIB archives,
> sometimes related to the variable names being unusual. Anyway, we've got
> that working now (although we are having trouble
> with a monthly average precipitation rate field found on the Reanalysis
> CDs - a question is being formulated now about this
> and will be sent in soon.) When I'm out in Boulder next week I'll ask
> you whether there's something I can offer you in exchange
> for excising these dumb questions from the email archives...
>