[R-sig-Geo] Singular Matrix Error

[Jesse Berman]

Hi All,

I want to perform ordinary kriging on a series of air pollution values, but
am getting the error

/> IL.Jul7.ok<-krige(Arithmetic.Mean~1, IL.Jul7, model=ILvgm.jul7,
newdata=IL2.map)
[using ordinary kriging]

"chfactor.c", line 131: singular matrix in function LDLfactor()
Error in predict.gstat(g, newdata = newdata, block = block, nsim = nsim,  : 
  LDLfactor/

I know that this error is typically caused by either duplicate points or a
variogram without a nugget effect.  However, I have neither (see code
below).  I also have set the same projection for points and prediction grid,
plus ran an IDW successfully /[CRS("+proj=longlat +ellps=GRS80 +datum=NAD83
+no_defs")]
/
/> ILvgm.jul7
  model     psill   range
1   Nug  1.537923   0.000
2   Sph 53.045879 266.323

> 	zerodist(IL.Jul7)
     [,1] [,2]

> summary(IL.Jul7 at data$Arithmetic.Mean)
   Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
  13.50   17.60   21.95   22.82   27.83   32.10 /

Any thoughts as to what might be causing this problem?  Sorry for not
providing a reproducible example, as I can't reproduce it and have run
similar code without problems on other data.  Thanks for the suggestions.

Jesse



-----
--------------------------------------------------------
Jesse D Berman, PhD
Johns Hopkins Bloomberg School of Public Health
Department of Environmental Health Sciences
Post-Doc 
--
View this message in context: http://r-sig-geo.2731867.n2.nabble.com/Singular-Matrix-Error-tp7584203.html
Sent from the R-sig-geo mailing list archive at Nabble.com.