Why is soil more conductive than water? Compare the following two sources:

High quality deionized water has a conductivity of about $5.5~\mathrm{\mu S/m},$
  typical drinking water in the range of $5$-$50~\mathrm{ m S/m,}$ while sea water
  about $5 ~\mathrm{S/m}$[2] (i.e., sea water's conductivity is one million times
  higher than that of deionized water).


Source
It seems highly unintuitive to me that sand can be more conductive than drinking water. I think I am missing something here.
 A: The comparison certainly depends on the soil, and the dissolved salts in the water, and the interaction between the two.  Recall from early chemistry that pure water has very few charge carriers.  Add any amount of dissolved salts and it quickly becomes conductive.  In the natural world, 1mS/m conductivity in water is pretty rare.  It's more often 10mS/m to 100mS/m  - and that's still called "fresh(ish)".
"Soil" has a wide range of conductivity, but it is strongly influenced by the water content and the clay content.  Dry, pure sand is very, very resistive.  The resistivity of natural sand is defined by the water content and conductivity  (look up Archie's law).  Clay soil resistivity also depends on water content and salinity, but dry clay still has some conductivity because of the free ions of the clay minerals.
A very readable technical note on the subject is Technical Note 5 Electrical Conductivity of Soils and Rocks by Duncan McNeill, and available on-line from the website of Geonics, in their library (www.geonics.com).
A: Pure water is actually a really good electrical insulator. 
The conductive properties of water depends on the ions dispersed in it. For pure water, the only ions present are $\text{OH}^-$ and $\text{H}^+$ ions produced in the water dissociation equilibrium. This is why sea water, which contains a large quantity of $\text{Na}^+$, $\text{Cl}^-$ and other ions has such a large conductivity.
Conductivity of water also depends on the quantity of gas contained in it: the more gas it contains, the less conductive it is, because gases are terrible electrical conductors. Quoting from Wikipedia:

The electrical conductivity of ultra-pure water is $5.5 × 10^{−8}$ S/cm (18 MΩ·cm in the reciprocal terms of electrical resistivity) and is due only to H$^+$ and OH$^−$ ions produced in the water dissociation equilibrium. This low conductivity is only achieved, however, in the presence of dissolved monatomic gases. Completely de-gassed ultrapure water has a conductivity of $1.2 × 10^{−4}$ S/m, whereas on equilibration to the atmosphere it is $7.5 × 10^{−5}$ S/m due to dissolved CO$_2$ in it.

(Notice that $5.5 × 10^{−8}$ S/cm =  $5.5 × 10^{−6}$ S/m.)
We therefore have, in order:


*

*Ultra-pure water + monatomic gases: $5.5 × 10^{−6}$ S/m

*Ultra-pure water + CO$_2$: $7.5 × 10^{−5}$ S/m

*De-gassed ultra-pure water: $1.2 × 10^{−4}$ S/m

*Drinking water: $5 × 10^{−3} - 5 × 10^{−2}$ S/m

*Seawater: $5$ S/m


You can therefore see why soil is generally more conductive than water: it is because it contains such a large number of ions. Of course, you also have to take into consideration the quantity of gas contained in the soil, because as we have seen a large quantity of gas will usually decrease the conductivity. For example it is clear that dry sand will have a relatively low conductivity because of the large quantity of air trapped between the grains.
