[R] issues with krige function

Adrienne Wootten amwootte at ncsu.edu
Wed Oct 31 14:14:08 CET 2012


My apologies to all, the package is gstat that I am using not geoR.
thanks!
A

On Tue, Oct 30, 2012 at 1:35 PM, Adrienne Wootten <amwootte at ncsu.edu> wrote:
> Greetings all,
>
> Ran into a strange problem with the krige function from geoR.  The
> problem that I am having is that while the krige function seems to
> work well, the resulting predicted values are all NAs.  Given the size
> of the datasets I am working with can't attach it, but I can provide
> snippets of the datasets.
>
>> casedata
>    station year month day  obs         mpe        bias type      lat       lon
> 140  319147 2011     8  28 0.00 0.000000000  0.00000000 COOP 35.48667 -82.96833
> 141  319354 2011     8  28 0.02 0.001305799 -0.01869420 COOP 34.25722 -78.68722
> 142  319357 2011     8  28 0.00 0.045194085  0.04519409 COOP 34.40944 -78.79139
> 143  319440 2011     8  28 6.92 5.530313974 -1.38968603 COOP 35.85278 -77.03056
> 144  319461 2011     8  28 0.98 0.949224103 -0.03077590 COOP 34.14250 -77.87861
> 145  319467 2011     8  28 0.30 1.492852601  1.19285260 COOP 34.32083 -77.92056
> 146  319476 2011     8  28 2.95 3.284362345  0.33436235 COOP 35.69389 -77.94556
> 147  319555 2011     8  28 0.00 0.000000000  0.00000000 COOP 36.13083 -81.22750
> 148  319667 2011     8  28 0.00 0.014472943  0.01447294 COOP 35.85583 -80.36083
> 149  319675 2011     8  28 0.00 0.000000000  0.00000000 COOP 36.13056 -80.54806
>
>> v.fit = vgm(phi=1.2914,model="Mat",range=14.1611,nugget=0.23,kappa=83.1047)
>> krige.pred.out = krige(formula=bias~1,locations=~lon+lat,data=casedata,newdata=gridmeta,model=v.fit)
> [using ordinary kriging]
>
>> krige.pred.out
>    lon      lat var1.pred var1.var
> 1  -86.28771 33.91335       NaN      NaN
> 2  -86.27388 33.94722       NaN      NaN
> 3  -86.26003 33.98111       NaN      NaN
> 4  -86.24615 34.01500       NaN      NaN
> 5  -86.23225 34.04890       NaN      NaN
> 6  -86.21835 34.08280       NaN      NaN
> 7  -86.20441 34.11672       NaN      NaN
> 8  -86.19045 34.15064       NaN      NaN
> 9  -86.17648 34.18457       NaN      NaN
> 10 -86.16249 34.21851       NaN      NaN
>
>>  summary(krige.pred.out)
>       lon              lat                 var1.pred        var1.var
>  Min.   :-86.29   Min.   :30.51   Min.   : NA     Min.   : NA
>  1st Qu.:-82.38   1st Qu.:33.73   1st Qu.: NA     1st Qu.: NA
>  Median :-79.88   Median :35.22   Median : NA     Median : NA
>  Mean   :-79.91   Mean   :35.26   Mean   :NaN     Mean   :NaN
>  3rd Qu.:-77.48   3rd Qu.:36.76   3rd Qu.: NA     3rd Qu.: NA
>  Max.   :-73.16   Max.   :40.16   Max.   : NA     Max.   : NA
>                                                NA's   :44408   NA's   :44408
>
>
> Any ideas are greatly appreciated, I'm not getting any warnings or
> errors from krige, so this is perplexing to me.
>
> --
> Adrienne Wootten
> Graduate Research Assistant
> State Climate Office of North Carolina
> Department of Marine, Earth and Atmospheric Sciences
> North Carolina State University



-- 
Adrienne Wootten
Graduate Research Assistant
State Climate Office of North Carolina
Department of Marine, Earth and Atmospheric Sciences
North Carolina State University




More information about the R-help mailing list