We’ve looked at how water is held in the soil and why it’s important to know how much water is there. Now we consider how to measure the quantity of water in the soil.

Firstly we need to address how to quantify the soil moisture content; there are two ways to do this, by mass or by volume.

If a sample of soil has a total mass equal to mt, this has two component parts the mass of the soil (ms) and the mass of the water contined in it (mw) - clearly mt=ms+mw. The water content can be expressed as a fraction of the soil dry weight, i.e., mw/ms. This is a dimensionless quantity known as the gravitational water content, and normally represented by Θm (Greek capital letter theta with subscript m for mass).

The volumetric water content is a similar ratio but is expressed as vw/vt, where vw is the volume of water, and vt is the total volume of the soil sample. In this case vt=vs+vw+vg, where vs is the volume of soil and vg is the volume of gas (air) in the soil. The gas has neglible mass and could be ingnored when expressing the water content by mass, but the volume of gas, i.e. all of the pore space that exists between the soil grains, is significant and must be acknowledged. The volumetric water content is usually represented by Θ.

These various masses and volumes can be used to give other soil properties such as bulk density (ρb=ms/vt - that's lower case Greek letter rho with subscript b) and porosity (f={vw+vg}/vt).

### Measurement techniques

The gravitational water content, Θm, is found by weighing a sample of the soil to obtain mt, drying the soil to remove the moisture and reweighing to determine ms; mw is mt-ms. The drying should be at 105 deg C and be for long enough to remove all of the water, i.e. further drying doesn’t give a different value for ms. Note that this is not hot enough to remove the tightly bound (hygroscopic) water which gets included within ms, but this is small compared to the other components. Heating at higher temperatures will burn off organic compounds in the soil, which should be included as the mass of the soil, ms, and not added to the mass of the water, mw.

The volumetric water content is obtained by the same process except that the sample taken is of a specific, known, volume. Because it’s a known volume, the bulk density, ρb, of the sample can be calculated and the volumetric water content, Θ, is Θmρbw, where ρw is the density of water which is a known constant (1 kg/l or 1 g/cubic cm).

The requirement to dry the soil sample means that a sample must be obtained from the field and returned to a laboratory for analysis, a procedure that is destructive (soil is physically removed) and time consuming for every measurement of soil moisture required. For these reasons several alternative methods have been developed that are non-destructive, require less effort and may be automated. These are all indirect methods in which some property of soil is measured and then converted to a moisture content. As such they require calibration to ensure an accurate conversion between the instrument’s output and soil moisture content.

The first of these methods, the tensiometer, measures the pressure in the soil caused by the surface tension (suction) forces described earlier. Traditionally this pressure was read using a manometer but there are electronic sensors that can be used to automate measurement. The pressure is related to water content by a soil retention curve, which varies with soil type and usually exhibits hysteresis, i.e. different curves describe the wetting-up and drying-out of soils. These are inexpensive instruments and are routinely used for informing farmers of the need for irrigation.

Gypsum, or electrical resistance, blocks work on the principle that water in the soil lowers its electrical resistivity. The material from which the block is made is such that it is well connected with the moisture regime in the surrounding soil so that water content and pore pressure is the same in the block as in the soil. Probes embedded in the block allow the resistivity to be measured, again without the need for an operator. As with tensiometers the method requires calibration for the particular soil, and will be subject to hysteresis.

Two other techniques, time domain reflectometry (TDR) and time domain transmissometry (TDT), are also based on the changes in (dielectric) soil properties that can be measured using electromagnetic pulses. These methods are accurate, fast, non-destructive, capable of automated operation, and can be used without site-specific calibration.

Another measurement technique is based on neutron scattering. A source of fast neutrons is lowered into an access tube in the soil. The neutrons collide with hydrogen nucleii in the soil and are converted to slow (thermalized) neutrons that are reflected and monitored by a counter. The source and counter are normally mounted in a single probe that can be operated at different depths to monitor variations in soil moisture within the soil profile. The volumetric water content is derived from the count ratio that compares the number of counts with a background reading from a reference medium. Neutron scattering methods have become less widely used as other methods were developed mainly because of issues associated with operating radiation-emitting devices.

All of the technologies discussed above provide a measure of soil moisture in a very small volume of soil, they can be very labour intensive and require many such measures to describe the moisture content across a field. What has been lacking is a measurement technique that works at the field or landscape scale. Two techniques are able to provide landscape-scale assessments of soil moisture. One of these is the cosmic-ray based soil sensor (CRS), the other is remote measurement from satellites. Satellites have the potential to provide repeat observations at a global scale but still require calibration from ground-based sensors. Satellite methods are not described further here.

### The cosmic ray soil moisture sensor (CRS)

The CRS sensor works in a similar way to the neutron probe with the big difference that it uses cosmic rays as the source of fast neutrons, thus avoiding the problems of handling a radioactive source.

A large benefit of this method is that a CRS integrates soil moisture over an area up to 400 m in diameter, and to a depth of up to 80 cm; in fact these figures for area and depth are likely maxima, and decrease with water content so that in a wet soil the effective depth may be as low as 15 cm. A further benefit of the CRS is that it sits above the ground and can operate remotely with little maintenance.

Obtaining soil moisture information from the CRS requires many adjustments to the monitored counts of slow neutrons to correct for variations in in-coming cosmic-rays, atmospheric pressure, humidity and altitude. In practice calibration against volumetric soil moisture derived from multiple samples within the measurement area is also needed.

In summary, there are many measurement techniques which mainly use indirect methods. Their varied characteristics mean that different techniques are suited to different applications.

The COSMOS-UK network is based on the CRS to give the landscape-scale picture but is backed up by point measurements based on dielectric soil properties.

Understanding how these different methods can best be used to provide UK-wide real-time information is the subject of on-going scientific research (see publications under Science and Resources).