The data is not available !
library(gstat)
var1 <- variogram(Total_C ~ 1, ~x.axis + y.axis, cutoff = 500, width = 30, data = mc)
# create figure of empirical semivariogram
plot(var1, plot.numbers = TRUE, xlab = "Distance", ylab = "Semivariance", cex = 1, cex.axis = 2, xlim=c(0, 550), ylim=c(0.2, 0.7))
To fit theoretical semivariograms, the function ‘fit.variogram’ of the package ‘gstat’ is used. Sill, nugget and range are set to be calculated based on the empirical variogram data, which is also used to fit the model ‘(fit.method=1)’. This method fits the variogram model to the experimental variogram, using weighted least squares with weight = Nj, where Nj is the number of observations in the j -th distance class (bin) (from: http://www.gstat. org/gstat.pdf, Table 4.2). The exponential model is used here. Semivariogram models are only fitted to undirected empirical semivariograms as the number of point pairs per bin in the directed ones is very low and predictions therefore have less power (the number of points per bin (np) for the undirected semivariograms is as high as for all directed semivariograms together).
mod1 <- fit.variogram(var1, vgm(psill = NA, "Exp", range = NA, 1), fit.sills = TRUE, fit.ranges = TRUE, fit.method = 1)
## Warning in fit.variogram(var1, vgm(psill = NA, "Exp", range = NA, 1), fit.sills
## = TRUE, : No convergence after 200 iterations: try different initial values?
## Warning in fit.variogram(object, model, fit.sills = fit.sills, fit.ranges =
## fit.ranges, : singular model in variogram fit
Descriptive statistics of soil organic C (Total_C) concentration, nitrogen concentration (Total_N), and organic matter (OM). Values are based on the 1st sample set (n = 150). Soil properties represent the top 15 cm of Duke Farm’s agriculture (grazed field) soil
N | C | OM | ||||||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Soil series | n | Min | Median | Max | Mean | SD | Min | Median | Max | Mean | SD | Min | Median | Max | Mean | SD | ||
KkoC | 25 | 0.100 | 0.180 | 0.310 | 0.188 | 0.046 | 0.990 | 1.800 | 3.120 | 1.874 | 0.464 | 1.700 | 3.100 | 5.380 | 3.220 | 0.801 | ||
KkoD | 13 | 0.130 | 0.220 | 0.390 | 0.231 | 0.076 | 1.240 | 2.180 | 4.050 | 2.317 | 0.841 | 2.140 | 3.760 | 6.970 | 3.992 | 1.449 | ||
PenB | 36 | 0.060 | 0.205 | 0.310 | 0.192 | 0.060 | 0.250 | 2.150 | 3.520 | 1.952 | 0.696 | 0.420 | 3.705 | 6.070 | 3.366 | 1.202 | ||
PeoC | 26 | 0.120 | 0.210 | 0.390 | 0.218 | 0.065 | 1.260 | 2.090 | 3.800 | 2.199 | 0.677 | 2.170 | 3.600 | 6.540 | 3.790 | 1.167 | ||
where N = Total Nitrogen, C = Total Carbon and OM = organic matter
Analysis of variance for soil organic C (Total_C) concentration, nitrogen concentration (Total_N), and organic matter (OM) for Duke Farm’s agriculture (grazed field) soil n = 150.
Parameters | df | Sum Sq | Mean Sq | F value | Pr(>F) | |
---|---|---|---|---|---|---|
Total_C | S_Series | 3 | 2.620 | 0.875 | 1.996 | 0.120 |
Residuals | 96 | 42.070 | 0.438 | - | - | |
Total_N | S_Series | 3 | 0.026 | 0.009 | 2.373 | 0.075 |
Residuals | 96 | 0.354 | 0.004 | - | - | |
OM | S_Series | 3 | 7.800 | 2.599 | 1.993 | 0.120 |
Residuals | 96 | 125.200 | 1.305 | - | - | |
You can use the plotting functions spplot or bubble as illustrated below (note: the x- and y-axis are the spatial coordinates)**
data = data.frame(x,y,Total_C)
library(sp)
coordinates(data) = ~x+y
class(data)
## [1] "SpatialPointsDataFrame"
## attr(,"package")
## [1] "sp"
spplot(data, "Total_C", colorkey = TRUE, main = " Grazed feilds 30m Triangular sampling C concentrations (n = 150 Random)")
# The End