The Weather Forecasting Forum

Wind Direction U and V
Page 1 of 1

Author:  Ian [ Thu Apr 12, 2007 8:18 pm ]
Post subject:  Wind Direction U and V

Hi Jeff,

I am using DEGRIB to convert grib2 files into a XML format, but i need to calulate wind direction and wind speed.

I believe this is stored in a grib message as

both in M/S

I used the formla atan2(y,x)

Where y is the V and x is U

is this correct? and is the output in DEG.

I am getting a little lost here, and wondered if you could help me out.


Author:  jkrob [ Fri Apr 13, 2007 2:16 am ]
Post subject: 


Well, you are half right...but not quite. Yes, UGRD & VGRD are the way wind is 'packed' into GRIB messages. However, the numbers are not straight forward. Here are the formulas to get wind speed (M/S) & direction (grid relative) from the U & V


Clear as mud...right? In essence, the UGRD is the North/South wind component (North is + & South is -) and VGRD is the East/West wind component (East is + West is - [I think...]). asked for it ;-)


Author:  Ian [ Fri Apr 13, 2007 7:23 am ]
Post subject: 

Hi Jeff,

Thanks for that. However (DIRECTION=57.29578*(arctangent(UGRD,VGRD))+180)

Does this formula work in what ever quadrant?

Or does the +180 change to suit V and U values?


180 if V>=0
0 if U<0 and V<0>0 and V<0


Author:  Ian [ Fri Apr 13, 2007 8:13 am ]
Post subject: 

Hi again,

Tried the DIRECTION=57.29578*(arctangent(UGRD,VGRD))+180 Could not get it to work.

Entered DIRECTION=57.29578*(ATAN2(UGRD,VGRD))+180

Works OK - where U = 6 and V = 5, result dir= 220 deg

Speed = 8 m/s

Could you check it please.


Author:  jkrob [ Fri Apr 13, 2007 7:01 pm ]
Post subject: 


Yes, those numbers look good. Sorry for the confusion; arctangent & ATAN2 were the same thing. I did not know if you knew Fortran functions so I wrote the formula out longhand.


Author:  Ian [ Fri Apr 13, 2007 9:09 pm ]
Post subject: 


No probs mate for the confusion, just glad you could help me out.

While Iam picking your brain for the maths, perhaps you would be able to help me out again.

I have two more calcs to do

1. PRATE, [kg/(m^2 s) to mm/h
2. Weather strings.

Now i was going to use the DAVIS weather software which came with my weather station to give a description of the weather based on the output of the GRIB model, but this is only based on pressure trends.
Is there a WMO standard for simple weather strings. I have looked on google, but have not found anything as of yet which made much sence to me.


Wingridds is getting better with every relaese - keep up the good work.

Author:  jkrob [ Sat Apr 14, 2007 2:15 am ]
Post subject: 


Sorry...I don't really follow what it is you are trying to accomplish. Can you be more 'verbose' in your explaination? ;-)


Page 1 of 1 All times are UTC
Powered by phpBB® Forum Software © phpBB Group