Skip to content

Thiessen Method

5 messages · federico eccel, Sarah Goslee, federico.eccel +1 more

Original
Sarah Goslee
# 
Federico,

You've asked this a couple times. Did you try to find the answer yourself?

If you search at rseek.org for Thiessen or Voronoi several functions
appear that should do what you need.

You didn't tell us anything about your data, so there's no possible
way we can provide more specific advice.

Sarah

On Wed, May 25, 2011 at 6:04 PM, federico eccel
<federico.eccel at gmail.com> wrote:
federico.eccel
# 
Dear Sarah,

I have a grid in which 8 raingauges are locted, in my case the dataset is
composed by 8 hourly timeseries, one for each raingauge. I would like to
obtain from these timeseries using the Thiessen method the values of the
precipitation in all the grid. In particular I would like to create the
thiessen polygons around the raingauges that have to be limited on my grid. 

Thanks a lot

Federico

--
View this message in context: http://r.789695.n4.nabble.com/Thiessen-Method-tp3551393p3552865.html
Sent from the R help mailing list archive at Nabble.com.
Sarah Goslee
# 
Federico,

That's an improvement, but a long way from the reproducible example
requested by the posting guide. I and others who might help are more
interested in the way the data and coordinates are organized and a
detailed explanation of what you expect the results to look like, etc,
than in a verbal description of the problem. From your description, I
can come up with many ways in which your data might be specified, and
results that you might want.

The posting guide provides valuable tips on how to provide a
well-formed question.

Did you try looking at the packages and functions I suggested in my
previous reply?

Sarah

On Thu, May 26, 2011 at 11:35 AM, federico.eccel
<federico.eccel at gmail.com> wrote:

  
    

      
      
    

  


  


      

    

    

      
      

        


  

        

      Thomas Adams
    

    

      
      #
    

  


  

      
Federico,

I understand what you are after ? you want time-series estimates based on the Thiessen polygon estimates
taken from the station time-series data. My recommendation is that the process of doing this would be far
easier using something like GRASS GIS, possibly in conjunction with R (since they play together very well).

Unfortunately, lots of coding/scripting is needed ? I can not see that there are a few R commands you can
make to pull this off.

The process would look something like:

(1) import all station data for time step 1 (all data could be imported at one time, but this complicates the process)
(2) make Thiessen polygons based on (1)
(3) write-out results from (2)
(4) repeat (1)-(3) for each time step
(5) concatenate individual Thiessen polygon time-series results sequentially

This kind of thing is pretty straight-forward and keeps computers happy!

Regards,
Tom