WRF EMS Clouds too low in % (ARW core)

All other topics about postprocessing model data (GrADS and other software), about other numerical weather prediction software (including WRF-NMM and WRF-ARW discussion unrelated to UEMS/WRF EMS), and general meteorology talk go in this forum.
meteoadriatic
Posts: 1602
Joined: Wed Aug 19, 2009 10:05 am

Re: WRF EMS Cloud's to low in %

Post by meteoadriatic » Fri May 28, 2010 5:42 am

I currently do cloud cover from relative humidity field in ARW. This is only solution that works for me.

j0nes2k
Posts: 56
Joined: Wed Feb 03, 2010 5:06 pm

Re: WRF EMS Cloud's to low in %

Post by j0nes2k » Fri Jun 04, 2010 7:27 am

Hello,

thank you for your help! Would it be correct to define the total cloud cover as an average for relative humidity throughout all model levels?

Code: Select all

ave(rhprs, lev=10, lev=1013)
Or should I average on low/mid/high levels?

Best regards,
Jonas Kaufmann

meteoadriatic
Posts: 1602
Joined: Wed Aug 19, 2009 10:05 am

Re: WRF EMS Cloud's to low in %

Post by meteoadriatic » Fri Jun 04, 2010 10:29 am

No, neither one would work if you want to show total cloud cover as you see it from the ground. If you assume that 8/8 cloud cover (when whole sky is covered) = 100% then it is not matter if these clouds are 10 kilometers deep or there is one thin layer of low clouds near ground and no clouds above them. In that case, if you want to show that as total cloud cover, you would have to search for layer with highest relative humidity. Currently I don't have clear idea how to realize this.

Keep in mind however, that relations between relative humidity and cloud cover are something like this:

a) at low levels, if layer has approx. 70% RH, there should be about 1-2/8 cloud cover in that particular layer
b) at about 80% RH in low level layer, there shoud be about 5/8 cloud cover in that layer

c) at higher levels, clouds will develop easier at lower RH values, so you will need something about 50% RH for 1-2/8 cloud cover at high levels and about 70% RH for aboud 4-5/8 cloud cover.

I can give you the script I develop for composite cloud cover in ARW. Please, look here how it looks:
http://maps.meteoadriatic.net/wrf-arw-3 ... blaka.html

This is attempt to show how clouds will look when you observe them from the ground. So, darkest area will be low level clouds, grey area means cloud cover at mid levels and white areas are clouds at high levels. Blue areas are areas with no or very low quantity of clouds. This is the script, you may need to tune somewhat RH numbers to get more precise forecast for your particular area:

Code: Select all

'set rgb 99 0 0 200'
'set mpt 0 99 1 6'

# High clouds
  'set rgb 31 238 238 238'
  'set rgb 32 235 239 245'
  'set rgb 33 209 240 252'
  'set rgb 34 194 232 252'
  'set rgb 35 179 224 252'
  'set rgb 36 164 219 252'
  'set rgb 37 149 214 252'
'define hum=ave(rhprs, lev=400, lev=275)'
  'set clevs 0 40 50 60 70 80 90'
  'set ccols 37 36 35 34 33 32 31'
'define hcloudmask = hum-75'
'd hum'

# Mid clouds
  'set rgb 24 186 186 186'
  'set rgb 25 197 197 197'
  'set rgb 26 206 206 206'
  'set rgb 27 215 215 215'
  'set rgb 28 220 220 220'
  'set rgb 29 225 225 225'
  'set rgb 30 232 232 232'
  'set rgb 31 238 238 238'
'define hum=ave(rhprs, lev=750, lev=600)'
  'set clevs 70 75 80 85 90 96 99'
  'set ccols 31 30 29 28 27 26 25 24'
'define mcloudmask = hum-75'
'd maskout(hum,mcloudmask)'

# Low clouds
  'set rgb 17  99  99  99'
  'set rgb 18 110 110 110'
  'set rgb 19 120 120 120'
  'set rgb 20 133 133 133'
  'set rgb 21 145 145 145'
  'set rgb 22 160 160 160'
  'set rgb 23 175 175 175'
  'set rgb 24 186 186 186'
'define hum=ave(rhprs, lev=825, lev=800)'
  'set clevs 80 84 88 92 95 97 99'
  'set ccols 24 23 22 21 20 19 18 17'
'define lcloudmask = hum-75'
'd maskout(hum,lcloudmask)'
Hope this helps :)

delinend
Posts: 65
Joined: Wed Oct 28, 2009 7:07 pm

Re: WRF EMS Cloud's to low in %

Post by delinend » Mon Jun 07, 2010 11:06 am

Hello.

It seems that my cloud problems is solved by Robert Rozumalski.

He have made a new WRFpost to me, so that these VARS is working like those in GFS. :D

LCDClcl 0 73,214,0 ** low cloud level Low level cloud cover [%]
MCDCmcl 0 74,224,0 ** mid-cloud level Mid level cloud cover [%]
HCDChcl 0 75,234,0 ** high cloud level High level cloud cover [%]
TCDCclm 0 71,200,0 ** atmos column Total cloud cover [%]

We have testet the new WRFpost for a long time now, and it seems to work perfect.

You´r welcome to try the new modifyed WRFpost here: http://soostrc.comet.ucar.edu/wrfems/users/

Best regards

j0nes2k
Posts: 56
Joined: Wed Feb 03, 2010 5:06 pm

Re: WRF EMS Cloud's to low in %

Post by j0nes2k » Mon Jun 07, 2010 12:02 pm

Hello,

thank you for your help! I think I will try the new wrfpost version as this is the easier solution and if this won't work, I will try meteoadriatics solution.

Best regards,
Jonas

delinend
Posts: 65
Joined: Wed Oct 28, 2009 7:07 pm

Re: WRF EMS Cloud's to low in %

Post by delinend » Mon Jun 07, 2010 12:14 pm

Remember!

Don´t use TCDCclm*100 anymore.

They are alle clean % (0 -> 100) now :)

Best regards

j0nes2k
Posts: 56
Joined: Wed Feb 03, 2010 5:06 pm

Re: WRF EMS Cloud's to low in %

Post by j0nes2k » Mon Jun 14, 2010 9:25 am

Hello,

I have the new wrfpost version running for about 1 week at the moment, and it seems to me that the cloud cover effects are much better than before, but are still not as good (or realistic) as the ARWpost ones - at least for my location in northern Germany. wrfpost still seems to underestimate the cloud cover. I have not done any scientific tests on this, just my first impressions.

Best regards,
Jonas Kaufmann

delinend
Posts: 65
Joined: Wed Oct 28, 2009 7:07 pm

Re: WRF EMS Cloud's to low in %

Post by delinend » Mon Jun 14, 2010 5:47 pm

Hello Jonas.

Thats great.

By edit physichs, you can get a look-alike arwPOST.

We build WRF over scandinavia, and the new WRFpost is mutch better, with all the water and coast lines, when we compare with sat images.

Best regards.

j0nes2k
Posts: 56
Joined: Wed Feb 03, 2010 5:06 pm

Re: WRF EMS Cloud's to low in %

Post by j0nes2k » Mon Jun 14, 2010 7:16 pm

Hello delinend,
By edit physichs, you can get a look-alike arwPOST.
Can you give me a hint of what you mean with this? Should I edit physics settings in my model or is it simply a output (GrADS) thing?

As a sidenote (and as my main interest is northern Germany which is not that far from Scandinavia), which settings do you use to run your WRF and how satisfied are you with the results?

Best regards,
Jonas Kaufmann

j0nes2k
Posts: 56
Joined: Wed Feb 03, 2010 5:06 pm

Re: WRF EMS Cloud's to low in %

Post by j0nes2k » Tue Jun 15, 2010 1:44 pm

Hello,

another thing that came to my mind: which fields do you extract via wrfpost? I currently only extract about 10 variables, mostly surface related. Do I have to include other output variables to get better cloud cover data?

Best regards,
Jonas Kaufmann

Post Reply