The Weather Forecasting Forum
http://www.weathergraphics.com/forum/

Analysis preferencs help
http://www.weathergraphics.com/forum/viewtopic.php?f=5&t=1760
Page 1 of 2

Author:  Greg Higgins [ Sat Sep 08, 2012 8:39 pm ]
Post subject:  Analysis preferencs help

Recently my plotted lines don't come anywhere close to what they should be and probably due to my playing with the "preferences". Please go to:
File
Preferences
Analysis

.....and when "Barnes" is checked, what are your settings for:
Smoothing coefficent
Smooting passes
Surface radius
Upper radius
Gamma

I appreciate the help! Thanks!!

Author:  gdp416 [ Sat Sep 08, 2012 11:55 pm ]
Post subject:  Re: Analysis preferencs help

.5
1
100
400
.3

which were the defaults upon install.

OT: I had to register as a new user just so I could answer this. It thinks my geophi user has attempted to login too many times (huh?), so makes me enter a "captcha" code. Entering it correctly, it tells me it was entered incorrectly. Tried this numerous times with different captcha codes, it still thinks I am entering them wrong. I had this problem before, and somehow got around it but can't remember what I did.

Author:  geophi [ Sun Sep 09, 2012 12:02 am ]
Post subject:  Re: Analysis preferencs help

Never mind, I just had it send me a new password to the old account and that worked. Silly.

Author:  Peter [ Sun Sep 09, 2012 4:36 am ]
Post subject:  Re: Analysis preferences help

Whatever Analysis type is selected different settings will be available or greyed out.

The setting I use are as follows for all Analysis types:

Smoothing Coefficient: 0.51
Smoothing Passes: 2
Gamma: 0.3
Reduction:5

Automatic Smoothing is checked.

The Surface Radius and Upper Radius are always greyed out and therefore cannot be changed.

I increased the Smoothing Passes from 1 (default) to 2 as I found it helped.

Regards Peter

Author:  Greg Higgins [ Sun Sep 09, 2012 12:55 pm ]
Post subject:  Re: Analysis preferencs help

Peter,
You are obviously using Cressman for your analysis since that is the only method that allows you to "tune" the reduction setting. That is why I was asking for "Barnes" settings. Thanks for the information.

Author:  Wally [ Sun Sep 09, 2012 8:08 pm ]
Post subject:  Re: Analysis preferencs help

Peter has given good numbers for Barnes. I have used Barnes and those numbers for some time.

Author:  Greg Higgins [ Sun Sep 09, 2012 8:55 pm ]
Post subject:  Re: Analysis preferencs help

The Barnes method does not allow you to change the "Reduction", that is ONLY for Cressman and is greyed out the rest of the time. From the DA Manual (page 89): "Reduction - Determines a reduction value used in the Cressman analysis method for upper air data."

Author:  Greg Higgins [ Sun Sep 09, 2012 9:12 pm ]
Post subject:  Re: Analysis preferencs help

Going back and checking the Analysis tab under prefences shows the following:

Analysis type:
Nearest Neighbor - Only the Smoothing Coefficient and Smoothing passes can be edited; all others greyed out, INCLUDING Automatic Smoothing.
Weighted - Gamma and Reduction are greyed out; all others can be edited. IF "Automatic smoothing" is checked, THEN Surface radius, Upper radius, Gamma and Reduction are greyed out and can not be edited.
Barnes - Only Reduction is greyed out; all others can be edited. IF "Automatic smoothing" is checked, THEN Surface radius, Upper radius and Reduction are greyed out and can not be edited.
Cressman - Only Gamma is greyed out; all others can be edited. IF "Automatic smoothing" is checked, THEN Surface radius, Upper radius and Gamma are greyed out and can not be edited.

If ya'll (Yes I'm a Texan!) are getting anything different, please let me know. Might be something Tim needs to know about for future reference.

Author:  Peter [ Sun Sep 09, 2012 11:51 pm ]
Post subject:  Re: Analysis preferencs help

Sorry Gents maybe I caused some confusion.

I gave all the setting for all the combinations of Analysis types but did clarify that by saying different combinations have some settings available and some greyed out.

I use Barnes most of the time but it your preference that counts.

Regards Peter

Author:  texsky [ Tue Sep 11, 2012 10:44 pm ]
Post subject:  Re: Analysis preferencs help

Greg,

I use the following for surface data and get very good results.

Barnes method: Checked
Domain Size: 100X100
Automatic Smoothing: Unchecked
Smoothing Coefficient: .6
Smoothing Passes: 1
Surface Radius: 100 nm
Gamma: .3

However, there is a problem when trying to use this grouping for upper air isoplething. It looks terrible until 'Automatic Smoothing' is rechecked, then it looks excellent for that too. Not sure what the issue is, perhaps the upper radius slider doesn't go high enough? Anyway, this causes problems when running automated scripts like I do. The two don't want to cooperate with the same settings. I'm hoping Tim will add a toggle for AS'ing that can be inserted into scripts.

hth,
Bob

Page 1 of 2 All times are UTC
Powered by phpBB® Forum Software © phpBB Group
https://www.phpbb.com/