RandomFields (version 3.0.5)

soil: Soil data of North Bavaria, Germany

Description

Soil physical and chemical data collected on a field in the Weissenstaedter Becken, Germany

Usage

data(soil)

source

The data were collected by Wolfgang Falk, Soil Physics Group, http://www.geo.uni-bayreuth.de/bodenphysik/Welcome.html, University of Bayreuth, Germany.

Details

For technical reasons some of the data were obtained as differences of two measurements (which are not available anymore). Therefore, some of the data have negative values.

References

Falk, W. (2000) Kleinskalige raeumliche Variabilitaet von Lachgas und bodenchemischen Parameters [Small Scale Spatial Variability of Nitrous Oxide and Pedo-Chemical Parameters], Master thesis, University of Bayreuth, Germany.

Examples

set.seed(0)################################################################
## ##
## a geostatistical analysis that demonstrates ##
## features of the package `RandomFields' ##
## ##
################################################################
\dontrun{

data(soil)
str(soil)
soil <- RFspatialPointsDataFrame(
 coords = soil[, c("x.coord", "y.coord")],
 data = soil[, c("moisture", "NO3.N", "Total.N",
 "NH4.N", "DOC", "N20N")],
 RFparams=list(vdim=6, n=1)
 )
## plot the data first
colour <- rainbow(100)
plot(soil["moisture"], col=colour)

x <- seq(min(soil@coords[, 1]), max(soil@coords[, 1]), l=121)

## fit by eye
RFgui.model <-  RFgui(soil["moisture"]) 
\dontshow{if (!interactive()) RFgui.model <- RMexp()}

## fit by ML
model <- ~1 + RMplus(RMwhittle(scale=NA, var=NA, nu=NA), RMnugget(var=NA))
fit <- RFfit(model, data=soil["moisture"])
plot(fit, fit.method=c("ml", "plain", "sqrt.nr", "sd.inv"),
     model = RFgui.model, col=1:8)


## Kriging ...
k <- RFinterpolate(fit["ml"], x=x, y=x, grid=TRUE, data=soil["moisture"])
plot(x=k, col=colour)
plot(x=k, y=soil["moisture"], col=colour)


## what is the probability that at no point of the
## grid given by x and y the moisture is greater than 24 percent?cs <- RFsimulate(model=fit["ml"], x=x, y=x, data=soil["moisture"], n=50)
plot(cs, col=colour)
plot(cs, y=soil["moisture"], col=colour)
Print(mean(apply(as.matrix(cs@data) <= 24, 2, all))) ## about 40 percent
##...

}