Hands-On Geospatial Analysis with R and QGIS

Kriging

Kriging is an interpolation process in which interpolated values are due to a Gaussian process. In kriging, a variogram is used to interpolate geostatistical data. Kriging gives us the best, linear, unbiased prediction.

With the use of observed data and a variogram, kriging computes estimates and uncertainties at unobserved points. There are different types of kriging; we'll discuss some of them briefly now. Ordinary kriging assumes that the mean is constant. Simple kriging assumes that the generalized least squares estimate of the trend coefficients is known. Universal kriging has a local trend component.

Now we'll use some of these kriging methods. We'll use the meuse.grid dataset, which contains coordinates of points on a regular ...