Page 4 of 5

Re: Temperature 2m ​​above the ground too cold

Posted: Fri Jan 03, 2014 5:36 pm
by Gippox
meteoadriatic wrote:Gippox, I can see that you have data in weasdsfc variable. This should be amount of snowfall like apcpsfc is for total precipitation. I think. But my weasdsfc is always empty. Do you know why? I use UPP 2.1 compiled myself, like you do?

BTW colors are very easy to control in GrADS, which part you don't understand? :)
I also use UPP compiled myself. In wrf_cntrl.parm i see this, maybe help you

Code: Select all

(ACM SNOWFALL        ) SCAL=( 5.0)
 L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000)
 (BUCKET SNOW  PRECIP ) SCAL=( 5.0)
 L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000)
 (BUCKET TOTAL PRECIP ) SCAL=( 5.0)
 L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000)
 (BUCKET CONV PRECIP  ) SCAL=( 5.0)
 L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000)
 (BUCKET GRDSCALE PRCP) SCAL=( 5.0)
 L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000)
 (ACM SNOW TOTAL/MELT ) SCAL=( 4.0)
 L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000)
I also discovered that, unlike arwpost, apcpsfc is the cumulative total precipitation. :(

I need a wide palette of colors and associated codes, because i do not know how to get the tones different from the standard Grads colors...

Re: Temperature 2m ​​above the ground too cold

Posted: Fri Jan 03, 2014 5:55 pm
by meteoadriatic
You define new color with
'set rgb CC xxx yyy zzz':
http://www.iges.org/grads/gadoc/gradcomdsetrgb.html

Then use that color code later:
'set ccol CC' or whatever you do with color.
("CC" is any number from 16 to 200 and some more, don't know exactly maximum).

You can also use color.gs, it is nice script for fields that behave linearly like temperature. For exponential fields like MLCAPE it is not very good. In fact maybe it is but I haven't found way to make exponential scale with it.
http://kodama.fubuki.info/wiki/wiki.cgi ... gs?lang=en

To find out what RGB mix is some color you find for example on the other maps, you can use this little utility:
http://gcolor2.sourceforge.net/

Re: Temperature 2m ​​above the ground too cold

Posted: Fri Jan 03, 2014 5:56 pm
by dominic
gippox tell me what scale do you like that you realize the I

Re: Temperature 2m ​​above the ground too cold

Posted: Fri Jan 03, 2014 5:57 pm
by Gippox
SNOW and SNOWH are not present as output parameters in wrf do you think is correct to make this change

Code: Select all

  
ncap2 -O -s 'where (SNOW < 10) SNOW=0' /home/andrea/wrfems/runs/arw/wrfinput_d01 /home/andrea/wrfems/runs/arw/wrfinput_d01
ncap2 -O -s 'where (SNOWH < 0.1) SNOWH=0' /home/andrea/wrfems/runs/arw/wrfinput_d01 /home/andrea/wrfems/runs/arw/wrfinput_d01

ncap2 -O -s 'where (SNOW < 10) SNOW=0' /home/andrea/wrfems/runs/arw/wrfinput_d02 /home/andrea/wrfems/runs/arw/wrfinput_d02
ncap2 -O -s 'where (SNOWH < 0.1) SNOWH=0' /home/andrea/wrfems/runs/arw/wrfinput_d02 /home/andrea/wrfems/runs/arw/wrfinput_d02

to the file wrfinput_d01 and wrfinput_d02 that are generated by real.exe ?

Re: Temperature 2m ​​above the ground too cold

Posted: Fri Jan 03, 2014 5:59 pm
by meteoadriatic
Gippox wrote:SNOW and SNOWH are not present as output parameters
Where? In gribs or in wrfout fields? in wrfout fields they should be if I recall correctly. In gribs snodsfc is SNOWH field.

Re: Temperature 2m ​​above the ground too cold

Posted: Fri Jan 03, 2014 6:00 pm
by dominic
Gippox wrote:
meteoadriatic wrote:Gippox, I can see that you have data in weasdsfc variable. This should be amount of snowfall like apcpsfc is for total precipitation. I think. But my weasdsfc is always empty. Do you know why? I use UPP 2.1 compiled myself, like you do?

BTW colors are very easy to control in GrADS, which part you don't understand? :)
I also use UPP compiled myself. In wrf_cntrl.parm i see this, maybe help you

Code: Select all

(ACM SNOWFALL        ) SCAL=( 5.0)
 L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000)
 (BUCKET SNOW  PRECIP ) SCAL=( 5.0)
 L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000)
 (BUCKET TOTAL PRECIP ) SCAL=( 5.0)
 L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000)
 (BUCKET CONV PRECIP  ) SCAL=( 5.0)
 L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000)
 (BUCKET GRDSCALE PRCP) SCAL=( 5.0)
 L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000)
 (ACM SNOW TOTAL/MELT ) SCAL=( 4.0)
 L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000)
I also discovered that, unlike arwpost, apcpsfc is the cumulative total precipitation. :(

I need a wide palette of colors and associated codes, because i do not know how to get the tones different from the standard Grads colors...

Quello che mi piacerebbe ottenere è l'accumulo in centimetri. Ho notato però un particolare, capita alcune volte che il modello vede la neve sul mare e questa cosa non va bene. Avevo pensato di sviluppare un sistema in linguaggio web che tenesse conto solo delle precipitazioni. A me principalmente interessa la previsione tipo triorarie e nello sviluppo delle mappe grafiche più che plottare con grads.

Voi che dite come si può derivare il dato degli accumuli nevosi? , attivando queste variabili in wrf cosa si può ottenere?
What I'd like to get is the accumulation in centimeters. I noticed, however, a particular, it happens sometimes that the model can see the snow on the sea and this thing is not good. I had planned to develop a system in web language that takes into account only of precipitation. I mainly interested in predicting type every three hours and in the development of graphical maps plotting with more than grads.

What do you say how can arise because of the snow accumulation? Activating these variables in wrf what can you get?

Re: Temperature 2m ​​above the ground too cold

Posted: Fri Jan 03, 2014 6:03 pm
by dominic
Boys lately (I do not remember the link) I had read that it would be useful to activate the variable in wrf QSNOW, which pretty much gives you the falling snow at a certain height. This variable is called Snow Mixing ratio in kg

Re: Temperature 2m ​​above the ground too cold

Posted: Fri Jan 03, 2014 6:12 pm
by meteoadriatic
dominic wrote:Boys lately (I do not remember the link) I had read that it would be useful to activate the variable in wrf QSNOW, which pretty much gives you the falling snow at a certain height. This variable is called Snow Mixing ratio in kg
I have that enabled (it is by default, from WRF registry - don't know about EMS version though, Robert changed a lot in v3.4 for precipitation fields), I have snow mixing ratio on all model levels and use that, for example for this:
http://meteoadriatic.net/prognoza/karte ... -kise.html

Re: Temperature 2m ​​above the ground too cold

Posted: Fri Jan 03, 2014 6:16 pm
by dominic
meteoadriatic wrote:
dominic wrote:Boys lately (I do not remember the link) I had read that it would be useful to activate the variable in wrf QSNOW, which pretty much gives you the falling snow at a certain height. This variable is called Snow Mixing ratio in kg
I have that enabled (it is by default, from WRF registry - don't know about EMS version though, Robert changed a lot in v3.4 for precipitation fields), I have snow mixing ratio on all model levels and use that, for example for this:
http://meteoadriatic.net/prognoza/karte ... -kise.html

Superb Ivan O_O how did you do this plot. I would not know how to get started with grads: D

would be nice to understand how it is made ​​up of your script

Re: Temperature 2m ​​above the ground too cold

Posted: Fri Jan 03, 2014 6:27 pm
by meteoadriatic
dominic wrote:Superb Ivan O_O how did you do this plot. I would not know how to get started with grads: D
would be nice to understand how it is made ​​up of your script
OK here it is.

First, enable snow and rain mixing ratios on lowest model level:

Code: Select all

 (SNOW ON MDL SFCS    ) SCAL=( 6.0)
 L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000)
 (RAIN ON MDL SFCS    ) SCAL=( 6.0)
 L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000)
Add those lines in your *cntrl.parm in static directory. If they are there make them look like that (with "1" on first digit). Then after next postprocess you will have variables RWMRhlev1 for rain mixing ratio and SNMRhlev1 for snow mixing ratio on lowest model level near ground, that is where you probably want to get data. If you want on higher levels, change also higher "zeros" to "ones".

Now with those two variables you will know where there is much more snow than rain (it is snow), where it is much more rain than snow (it is rain), or where it is approx. equal amount (it is mixed precipitation - sleet).

In grads script:

Code: Select all

'set gxout shade2b'
'set clevs 0 0.2 10'
'set ccols 2 2 3 11'
'define snowrainratio=(SNMRhlev1+0.0000001)/(RWMRhlev1+0.0000001)+0.01'
'define precthreshold=((SNMRhlev1+RWMRhlev1)*1000000)-1'
'd maskout(snowrainratio,precthreshold)'
Don't ask me why those strange numbers... it is not easy to explain in details (it has to do with very small and very large (infinity) numbers when you divide with zero or like that... to be computable at all just do like I did :) )

That's all, everything else in your script do as always you do.