R help - Apr 2011 - Windrose Percent Interval Frequencies Are Non Linear! Help!

Hi All -
I have been using the windrose function found in the circular library to create
great looking windroses.  But the dashed, circular percent interval lines that
indicate the frequency at which each "pedal" occurs are not evenly
spaced (the corresponding parameter is "cir.ind").  That is, the 5%
interval closest to the heart of the windrose is wider than the 5-10 chuck,
which inturn is wider than the 10-15% chunk.

Does anyone with specific windrose experience know how to adjust the graphic
such that the data and the percent intervals are evenly spaced? 

Hopefully I am making sense here....

Thanks.

britt



      
	[[alternative HTML version deleted]]

On 2011-04-07 14:40, Brittain Briber wrote:
> Hi All -
> I have been using the windrose function found in the circular library to
create great looking windroses.  But the dashed, circular percent interval lines
that indicate the frequency at which each "pedal" occurs are not
evenly spaced (the corresponding parameter is "cir.ind").  That is,
the 5% interval closest to the heart of the windrose is wider than the 5-10
chuck, which inturn is wider than the 10-15% chunk.
>
> Does anyone with specific windrose experience know how to adjust the
graphic such that the data and the percent intervals are evenly spaced?
>
> Hopefully I am making sense here....
How about giving us a reproducible example?
Code is better than mere description;
code + description is best.

Peter Ehlers
>
> Thanks.
>
> britt
>
>
>
>
> 	[[alternative HTML version deleted]]
>

>> Does anyone with specific windrose experience know how to adjust the
>> graphic such that the data and the percent intervals are evenly spaced?
>>
>> Hopefully I am making sense here....
>
> How about giving us a reproducible example?
> Code is better than mere description;
> code + description is best.
The "problem" is probably that A = pi r ^2, and the percent intervals
are spaced evenly on the square root scale to keep areas from being
distorted.

Hadley

-- 
Assistant Professor / Dobelman Family Junior Chair
Department of Statistics / Rice University
http://had.co.nz/

Hadley -
Thanks for your answers.  I didnt think of it that way and what you say makes
complete sense.  Truthfully, however, I dont care to maintain an equal area for
each increment.  When discussing frequency distributions on the rose, having a
diagram with equally spaced intervals would be more intuitive since it is the
distance away from the origin, not the area, that designates frequency.  Do you
two or anyone else out there happen to know if it is possible to specify the
width of the intervals in windrose?  Is there a parameter that I can pass along
that would do this?  Below is a reproducible script in which you can see the
uneven interval spacing.  And thanks again for your answers.

Best
Britt


dir <- circular(runif(100, 0, 360), units="degrees")
mag <-  rgamma(100, 15)
sample <- data.frame(dir=dir, mag=mag)
par(mfrow=c(2,2))
res <- windrose(sample)


________________________________
From: Hadley Wickham <hadley@rice.edu>
To: Peter Ehlers <ehlers@ucalgary.ca>

@r-project.org>
Sent: Thursday, April 7, 2011 9:47 PM
[[elided Yahoo spam]]
>> Does anyone with specific windrose experience know how to adjust the
>> graphic such that the data and the percent intervals are evenly spaced?
>>
>> Hopefully I am making sense here....
>
> How about giving us a reproducible example?
> Code is better than mere description;
> code + description is best.
The "problem" is probably that A = pi r ^2, and the percent intervals
are spaced evenly on the square root scale to keep areas from being
distorted.

Hadley

-- 
Assistant Professor / Dobelman Family Junior Chair
Department of Statistics / Rice University
http://had.co.nz/


      
	[[alternative HTML version deleted]]