Agriculture Reference
In-Depth Information
point kriging
block kriging
sampled point
Fig. 2.6
Principles of block-kriging and point-kriging (From Whelan et al.
2002
, altered)
map for later use is made from the sampled or sensed data, the averaging can occur
during mapping.
The sampling or sensing of the data usually occurs on a
punctual basis
. Hence for
mapping of a continuous surface, some interpolation is needed. This
interpolation
of the punctual signals is called
kriging
in honour of D.G. Krige, a South African
geologist, who was a pioneer in this processing of data for mapping. Today, many
kriging methods are available. These methods are based on the assumption that geo-
logical properties in close proximity to each other are more likely to be similar than
those separated by longer distances. So the basic premise - the concept of spatial
dependence - is the same as for semivariances (Sect.
2.2
).
Kriging can be subdivided into either block- or point methods, depending on the
size of the respective map area, for which the property is estimated by interpolation.
In block kriging, the average value of the property for a block within the map is
calculated, whereas in point kriging the target is just a point within the area. This
means that point kriging can be considered as the limiting case of block kriging
when the block size approaches the sampling- or sensing size (Fig.
2.6
). For details
to interpolation methods and to computer programs for this see Hengl (
2007
),
Webster and Oliver (
2007
) and Whelan et al. (
2002
).
An important feature of block-kriging is that its estimate - the block mean - may
gain in reliability as the block area increases. This follows from basic features of
means. This averaging advantage of
block-kriging
is effective as long as the size of
the blocks does not exceed the cell-size that is effective in the machinery operations.
But when the sizes of the kriged blocks get larger than the cells of the machines, this
averaging gets detrimental. This is because then another averaging feature - its
leveling effect - has consequences. This leveling effect deteriorates the objectives of
