Agriculture Reference
In-Depth Information
gravimetr. soil water cont.
66.7 % wet basis
50.0 % w. b.
27.8 % w. b.
16.7 % w. b.
0.6
0.4
0.2
0.0
-0.2
-0.4
-0.6
-0.8
potassium cation
nitrate anion
ion-selective electrodes
pH (hydrogen cation)
Fig. 9.3
Relative activity of ions and the water content of the soil samples. The ion-selective
electrodes that were used correspond to the recommendations mentioned in the previous section.
The results have been standardized by subtracting the respective average records for 27.8 % water
content from each measurement. The
bars on top
of the columns indicate standard deviations
(From Sethuramasamyraja et al.
2007
, altered)
sensing the soil as it presents itself in the field implies that the moisture varies to
some extent. This holds between different fields and times as well as within a field
when site-specific sensing takes place.
Experimental results for sensing of soil samples with varying water content did
reveal that the ions react on this differently (Fig.
9.3
). If hydrogen ions are sensed
when mapping the pH of soils, there hardly is an effect of varying soil water content.
Contrary to this, the results for sensing of nitrate- as well as of potash ions do depend
on the soil: water ratio. Especially with nitrate ions, the effect is obvious (Fig.
9.3
).
The technique as shown in Fig.
9.4
provides
pH readings
from naturally moist
soil samples in cycles of about 10 s. This means that with a travel speed of 2 m/s
(= 7.2 km/h) and a transect distance of 20 m, a reading is obtained for successive
cells of 400 m
2
. So compared to the conventional method of taking one sample per
ha for analyzing in special laboratories, the
spatial resolution
is 25 times better.
In non-uniform fields, this is a substantial improvement. Because in most cases,
it is not possible to solve the site-specific problem that is associated with the con-
ventional sampling resolution by
interpolation
between the mapped points. Any
interpolation method can only help to alleviate the problem if the adjacent points are
spatially related. Yet with a sampling cell size of 1 ha, the distance between neigh-
boring locations is about 100 m. For the pH situation of most fields, this is well
beyond the range of semivariance (Mulla and McBratney
2000
). This means that
