SOIL MOISTURE MONITORING


Why Do It?		Assisting growers with the decision on when to irrigate, requires a sound knowledge of what is happening to water as it moves through the soil profile. The aim is to ensure water is applied when the water level in the soils reaches a designated refill point for the particular crop. Irrigating too early can create unnecessary periods of water-logging, where crop growth is lost. Irrigating too late can create periods where growth stops because the crop is too dry. Getting the balance right requires knowledge of the crop, the moisture status of the soil, how much water is applied, how deep it reaches and where the water is used. Soil moisture sensors play an important role in garnering this information
		Soils are normally made up of particles of various sizes. The ratios between the small, medium and large particles determine whether the soil is classified as a clay, a loam, a sand, or some sub-classification of each. The distribution of particle sizes in turn impacts on both the soil’s water holding capacity and how easily the water can be accessed by a plant. Water coats the soil particles and occupies the pore spaces between them. Water is easily drawn away from the large soil pores but is difficult to remove from around the particles.
		The way the water behaves can be explained by considering 3 separate forces acting on it: Soil bond: the strength of the force the soil particle is exerting on the water to hold it in place. The smaller the particle, the tighter the bond Gravity: when the soil reaches saturation, the forces holding the water to the soil are lower than the pull exerted by gravity, so the water drains through the soil profile Matric potential: this is a measure of the suction force exerted by the roots of the plant to draw moisture away from the soil.
		When soil is dry, the only water being held is that which is closely bonded to the soil particles. As water is applied, it starts to fill the soil pores. However as more and more water is applied, the force holding it to the soil weakens. Eventually the force of gravity exceeds the force holding the water to the soil and the water is lost through the soil profile as drainage. If a droplet of water is released onto a dry soil surface, some water will be drawn down by gravity. The dry soil particles around the area will however exhibit a strong attraction to the water molecules and the water will be drawn out in all directions. Eventually the forces in the soil will reach equilibrium and the water will stop moving. Each new drop of water added changes the balance of forces, driving the soil to a new equilibrium. The wetter a soil is, the faster the soil water levels will change. The aim of irrigation management is to keep the level of water in the soil below the onset of drainage (saturation) and above the onset of stress. There are exceptions to this rule: for instance in applying deliberate water stress to maximise fruit quality in red wine grape production .
Content and Tension		The level of moisture in the soil can be expressed in two ways: Soil water tension Matric potential sensors measure the force binding the water to the soil. Matric potential is expressed as a suction or negative tension and is measured in kilopascals (kPa) Soil water content Volumetric sensors measure the quantity of water in a given volume of soil (Some systems measure gravimetric soil water content, which is the ratio by weight of water in a given weight of soil). Volumetric soil moisture content is expressed as a percentage.
		For each soil type a unique relationship exists between the two measures. When graphed, it is designated as the “soil water release curve”. Samples of the soil are placed in containers of a known volume. The samples are then weighed. The samples are then placed on a pressure plate apparatus where suction is applied to draw away the moisture. The sample weights are taken at a number of tension values. The wet and dry weights and sample volume are used to determine the percentage of moisture in the soil at each point and this is then plotted against the tension values.
From DUL to CLL		As irrigation managers, we are interested in a few key points on this curve Saturation: assumed to occur at a tension of -8 kPa The onset of stress: where this point occurs is a function of plant physiology. In annual crops it may occur at -20 kPa whilst in permanent crops it may occur at -40 or -60 kPa Wilting point: the point from which the plant can not recover .
TAW RAW and PAW		Although a soil calibration can show the total amount of water which can be held in the soil (the total available water or TAW), irrigators are normally interested in what can be held between the point of saturation and the onset of stress. This is defined as the readily available water (RAW). Water obtained once the plant is in stress is defined as the deficit available water (DAW). Although soils with a higher clay content generally hold more water, much of it is too tightly bound to the soil particles to be available to the plant. In general, clay loam soils make the best growing soils as they have the highest ratio of available water. In light soils, irrigation events must be relatively short and must be applied frequently. In heavier soils, irrigation can be applied for longer (because the soils hold more water) and the interval between irrigation extended.
		The rate at which irrigation is applied is also important. Different soils have different infiltration characteristics. Water passes rapidly through light textured soils and slowly through heavy soils. If water is applied at a rate above the infiltration rate, it will pool on the surface. If the site is sloped, water will run away down the slope, leaving some areas dry and others waterlogged. The rate at which irrigation is applied is usually measured in mm per hour. A soil survey will identify the typical or actual infiltration rate for the soil sample. A rough estimate can be made by welling up an area of dry soil with the rim of a plastic bucket, then applying a couple of litres of water. An hour later, a hole can be dug and the depth to which the water infiltrated measured. When designing an irrigation system, a designer will choose emitters whose application rate is less than the infiltration rate. The faster the water moves vertically, the less the lateral spread. In very sandy soils, water can move as far as 1cm in a minute – or 60cm per hour. Whereas in heavy clay soils, the infiltration rate may be as low as 1cm per hour. When using individual emitters such as drippers, the spacing between emitters must be altered to suit the rate at which water is moving laterally. It is thus common to see emitters spaced very close together (300m) in a light sand and further apart (1m) in heavy clay soils.
Sensor Technologies		Just as the level of moisture may be expressed in two ways, sensors which measure moisture can give the results in either form: that is as a volumetric reading or as a matric potential. In general the two types of measurements (and hence sensor technologies) have distinct characteristics. Matric potential sensors tell the picture from the plant’s perspective. As the tension level at which a plant exhibits stress occurs at a fixed point, the reading holds true for all soil types. Matric potential sensors thus make it very easy to determine when to start and stop irrigation. This is very important in annual crops such as small vegetables, where errors in irrigation management can wipe out a young crop very quickly. Volumetric sensor readings must be interpreted with knowledge of the soil type: a refill point of -25 kPa may correspond to 15% in one soil type but 33% in another. An experienced irrigation agronomist will be able to assess the soil texture and have a good guess at a suitable full and refill point, however the only way to know them with any immediacy and certainty is to perform a soil calibration (thus giving the soil water release curve for the soil). A good proxy can also be obtained by using portable devices such as electro-tensiometers to take spot readings alongside those from a volumetric sensor. If this is done at two or more water levels, the generic soil water release curve can be fitted to the soil, without having to do a full gravimetric analysis.
Matric Potential		Matric potential sensors may measure soil tension directly or indirectly. Although some US literature refers to matric potential in centobars (cb), the standard is the kilopascal (kPa). Fortunately 1cb = 1 kPa
		A tensiometer is made up of a plastic or glass column, on the end of which a porous ceramic tip is fitted. The column is partly filled with water and the top sealed. The tensiometer is installed vertically into the soil, with the tip centred at the level at which moisture measurement is required. As the plant roots try to draw water from the soil, the same suction force is applied through the ceramic tip to the water in the column, drawing water into the soil. As water is drawn form the sealed tube, it creates a vacuum in the top of the column. A pressure gauge or electronic pressure transducer can be fitted to the top of the unit to directly measure the tension
		Tensiometers do however have a number of drawbacks. These centre around high maintenance and limited range. As the reservoir of water in the column is limited, over time the column will dry out. If this occurs, air enters the ceramic tip and must be purged before the unit can read again. To counter this problem, some newer tensiometers are fitted with a level sensor which can be wired to give an alarm should the unit run low on water. Another advance is the self-filling tensiometer, which, when the soil is wet, can draw water back out of the soil, keeping the internal reservoir filled. This approach however only works if the site is regularly irrigated (such as in vegetable crops). Another complication occurs at high tension levels. Above a figure of 60 to 70 kPa, the tension becomes to great to hold a solid column of water. This break in tension, causes a rapid draining of water out of the tensiometer body. Once again the tensiometer needs to be purged and refilled before it will read properly.
		Tensiometers which have a simple mechanical vacuum gauge are relatively cheap and easy to install. However the tension reading changes as the column of air heats and cools – it is common to see the sensors covered to keep them out of the sun, but in doing so, users must ensure that the cover does not impact on delivery of water to the soil. Electro-tensiometers (units fitted with an electronic pressure sensor) get over some of these limitations: they can be completely buried, avoiding sunlight exposure; the transducers can be made more accurate than the manual gauges; But, they are considerably more expensive (typically by a factor of 10) than manual units.
		A number of sensors are available which measure soil tension indirectly, using a porous media installed into the soil. The media is chosen so that it will wet and dry easily with the surrounding soil. The water content of the media can them be measured and used to infer the water level in the soil.
Gypsum Blocks		The gypsum block was the first readily available, indirect soil tension sensor. The sensor comprises a block of gypsum into which two electrodes are inserted. As the sensor wets and dries its resistance changes: low when wet and high when dry. The resistance characteristics can in turn be mapped to matric potential with a single calibration. A very crude estimate of soil water content can be made by measuring the resistance of the soil. Resistance is a measure of how readily a material conducts electricity. The reciprocal of resistance is conductivity: a material which is highly resistive is a poor conductor and vice versa. Soil exhibits a high resistance when dry and a low resistance when wet, or has low conductivity when dry and high conductivity when wet. Gypsum blocks are very cheap, easy to install and easy to interpret but have a limited life.
Porous Blocks		Porous ceramic materials, such as those used in tensiometers can also be used to cover a sensor which would otherwise measure soil moisture content. As the water release curve of the porous media is constant, it can be measured once and then applied to all sensors made using that media. The sensor will thus return matric potential. Conversion tables could then be generated for different soil types, allowing the sensor to return both soil tension and soil content. The first sensor developed around porous media was the EquiTensiometer form Delta-T Devices of the UK. The Equitensiometer is simply a Theta Probe (see section ) with the tips embedded in a ceramic media. With their background in plant science, Delta-T wanted the Equitensiometer to read over a wide tension range. Developing a porous media which could cover this range meant they ended up with a material which absorbs contaminants from the soil and whose behaviour thus changes over time. So in addition to having a high price, the units need to be periodically re-calibrated to continue operating at their rated accuracy. Decagon, the maker of the Echo probe have recently released their own porous media sensor
Volumetric Sensors		Volumetric sensors infer soil moisture levels by measuring changes in the physical properties of the soil. The most common technique is to measure changes in the soil’s dielectric constant. A material’s dielectric constant indicates its ability to store energy. A high dielectric constant means a material can store a lot of energy and a low dielectric constant, means little energy can be stored. In the field of electronics, the capacitor is the basic device for storing energy. Capacitors are made up of pairs of parallel plates, separated by a dielectric material. Generally the material is chosen for its high dielectric constant, that is its high ability to store energy. If a capacitor is connected across a battery, the capacitor will “charge up”. A favourite trick of young electronics students is to charge a large capacitor across a battery and then to sneak up on people and discharge the capacitor on a nearby piece of metal. A large bang and spark ensue as the energy in the capacitor is rapidly discharged across this short circuit. Most capacitors are put to useful work: as stores for energy in power supplies or in the tuning components of oscillators.
		Those who grew up before “digital radios” will recall winding the tuning knob on a radio and watching a dial rotate as the radio changed frequency. The dial was invariably connected (directly or via a string) to a variable capacitor. The capacitor was made up of pairs of plates separated by air. As the dial rotated, the plates were moved from minimum overlap to full overlap. This corresponded to maximum and minimum energy storage, or maximum and minimum capacitance. The capacitor in turn made up half of a “tuned circuit” (the other half coming from a coil of wire, an inductor). So changing the overlap of the plates changed the capacitance and hence the frequency to which the radio would tune. In soil, the dielectric properties are dominated by the presence of water molecules and to a lesser extent by the presence of other ions (such as salts). Water molecules are polar in nature: that is the ends of the molecules exhibit a positive and negative charge. As the water content increases, so too does its ability to store energy. For many years the dominant method for measuring soil water content has been to use sensors which mimicked the behaviour of the old radios. The sensor elements took the role of the variable capacitor, so as the moisture level changed, the frequency of the oscillator changed and this change could be used to infer the moisture level.
		Other properties of the soil change as it wets and dries. One is the speed at which energy can pass through it. Measuring the speed at which a pulse travels along a sensor embedded in the soil, or the time taken for a pulse to travel between two points can thus also be used to infer moisture content. The first commercial volumetric soil moisture sensors measured the absorption by the soil of small amounts of radiation. The neutron probe, as these devices were called, were accurate but required a licensed operator who at best could visit a property once or twice a week to take readings. Safety issues and the desire to know what was happening real time have spelt the end of the neutron probe.
Capacitance Probes		Sentek pioneered the market for multi-level, variable frequency capacitance probes. This measurement method gives the probes a number of benefits and also contributes to their key limitations. To best assess irrigation efficacy, irrigators need to know what depths irrigation is reaching and from where in the profile the sensors are drawing water. Sentek designed a probe built around a vertical column to which multiple sensors could be attached (at 10cm increments). A printed circuit board on top of the probe is responsible for turning on individual sensors and reading the values. This approach allowed installers to fit sensors at the key depths required to give a full picture of water movement in the soil. The original design was based on the dimensions of the neutron probe tubes (50mm ID) but the sensors were installed in PVC rather than aluminium tubes. To install the probes, a slightly undersized hole is augered into the soil. A PVC tube is then fitted with a cutting edge and driven into the hole. The cutting edge shaves the edge of the hole ensuring a tight fit of tube to the surrounding soil. The inside of the tube is then cleaned out and the bottom sealed with a rubber bung. A plastic head or cap assembly is then glued on to the top of the tube. A PVC column, holding the sensors and top PCB is then lowered into the tube and the cap closed up.
		The sensor elements are made up of a pair of brass rings. The rings form the capacitance component of an L-C oscillator. As the inductive (L) component is fixed, the frequency of the oscillator will change if the capacitance between the rings changes. The dielectric component for this capacitor is the soil outside the probe: as it wets, the dielectric constant changes, changing the capacitance of the rings and hence changing the frequency of the oscillator. Although simple in concept the approach has a number of limitations The dimensions of the rings, the thickness of the access tube and the width of the air gap between the sensors and PVC all change, which means each sensor will operate over a different frequency range. The probes are “normalised” to reduce this effect The penetration of the electric field around the sensors changes with soil type and the soil’s moisture content thus changing the volume of soil which is contributing to the measurement The dielectric properties of the soil are influenced by soil temperature so the performance of the sensor changes with temperature Changes in soil conductivity (through salt and nutrients) will change the reading Good installation technique is critical to reliable data: air gaps between the access tube and soil reduce the effective measurement volume and create a preferential path down which water can flow Over time the plastic deforms and the effectiveness of the bottom stoppers can reduce, allowing moisture to enter into the probe column. This can be absorbed with satchels of silica desiccant, but over time water is absorbed into the layers of the PCBs, changing their behaviour. The need to check the probe tubes, dry them out and change the desiccant bags creates an ongoing maintenance requirement.
		The net result of all these factors is a design which has low accuracy but high repeatability: the sensor may give the same reading in the same conditions at two different times, but the value it returns is difficult to relate to the true moisture reading. Absolute numbers can only be obtained if the probe is calibrated for the soil in which it is installed. Most capacitance probes have been installed in permanent crops. Here data can be watched over the first season of operation and full and refill points estimated from the trends in the data. Growers of annual crops do not have this luxury.
Irrigation Scheduling		The aim in irrigation scheduling is to match irrigation to the plant requirements. In many crops this means keeping the plant out of stress. This is vital in crops where the aim is to generate maximum yield: each hour of stress means a reduction in yield per Ha or yield per Ml of irrigation water. Although in some crops, such as red wine grapes, the plants are kept in water stress for a period of time, to maximise fruit quality – favouring small, rich berries, rather than large water filled ones. Soil moisture sensors provide a simple means of identifying where the plant is on the soil moisture release curve at any time. Irrigators can then make informed decisions on: When in the season to commence irrigation When in the season to finish irrigation When to trigger an irrigation event How much water to apply.