[R-sig-Geo] Input data for kriging with gstat

[Pankaj Agarwal]

Hi,

I am working on preparing input data set for kriging using the gstat package in order to interpolate missing data values.  Following along the example in the manual, I have been able to go through the steps with the meuse and meuse.grid data package.   I am not sure how I would need to prepare my own data set for use in the function:

x <- krige(log(zinc)~1, meuse, meuse.grid, model = m, beta = 5.9)

Since the meuse and meuse.grid data set is used as is in the function, do I have to prepare my data set similar to these?

My data has level of measured air pollutant at the county centroid and the corresponding Long/Lat location.  There are data for a few counties but missing for several, so I am trying to interpolate the missing values.  I have already projected the Long/Lat using the rgdal package.  Here is part of the data set that I have.


[1] 2.67190 2.67190 2.67190 3.12994 2.67190 2.87500 2.67190      NA      NA     NA......


   Longitude Latitude
1  -76.94736 35.51745
2  -79.26456 35.73348
3  -78.71155 34.28487
4  -80.25679 36.10971
5  -77.10624 35.83633
6  -80.81196 35.21543
7  -78.97348 36.37938
8  -79.41320 36.07369
9  -81.19022 35.89239
10 -81.10411 36.49739

....

How would I convert or use my data set to an equivalent of meuse and meuse.grid so that I could use the "krige" and other functions.

Thank you,

- Pankaj

----------------------------------------------------------------
Pankaj Agarwal, M.S
Bioinformatician
Database Analyst II
Surgical Sciences Applied Therapeutics Section
Department of Surgery
Duke University
919-244-6389
p.agarwal at duke.edu<mailto:p.agarwal at duke.edu>


	[[alternative HTML version deleted]]