FAQ soilmoisture basics

Soil moisture is essential in horticulture and agriculture. Only with the right amount of water at the right time plants will grow and produce the desired yields.

Soil moisture is determined by precipitation, groundwater, irrigation and the consumption of water by plants. Some water is held in the soil against gravity and remains there. The soil thus has a water storage capacity that is needed by plants. The more precisely the water content can be determined, the better irrigation can be controlled and thus adapted specifically to the water requirements of the plants.

Soil moisture also plays a major role in other sectors. In civil engineering, the water content in the soil determines the load-bearing capacity or the stability of dike structures. Excessive soil moisture can even lead to landslides.

However, soil moisture sensors can also be used for other purposes, ranging from snow moisture measurement and concrete curing control to monitoring the water content in grain stores.

Electrical methods are particularly suitable for measuring soil moisture. This allows continuous monitoring of the water content in the soil and connection to irrigation control systems in a variety of ways.

Soil is a mixture of water, air and a mineral/organic soil matrix. Depending on the water content, the electrical characteristics change, whereby a distinction can be made between the electrical conductivity and the so-called dielectric characteristics.

The electrical conductivity depends not only on the water content, but largely on the dissolved salts. Measurement methods based on this principle are therefore often very unreliable and not suitable. Sensors on this basis, which are usually sold at very low prices, do show some dependence on the water content, but this changes over time. In addition, electrical conductivity measurement requires direct metallic contact with the soil and leads to corrosion of the electrodes. Therefore, reliable and permanent measurements are practically impossible.

Dielectric measurement is based on the special properties of the water molecule, which can align itself as a dipole in an electric field. This interaction can be detected and converted into a so-called volumetric water content. It can also be said that the soil between the electrodes is a capacitor whose capacitance changes with soil moisture. A great advantage is that this capacitance measurement does not require direct metallic contact of the electrodes. Therefore, a long-term stable sensor system is possible on this basis. In praxis, capacitance is measured using small AC voltages. Here the frequency selection is of decisive importance. Very low-cost capacitive sensors often use frequencies that are much too low, and there is considerable cross interference from electrical conductivity. High frequency capacitive sensors are more complex and expensive, but allow a reliable and accurate determination of the water content.

Material moisture measurement technology, as the upper field of soil moisture measurement technology, is an extensive research and development field. The standard scientific work in English on this subject is available free of charge:
Electromagnetic moisture measurement

The selection of a suitable sensor for soil moisture measurement depends on several aspects:

Required measuring accuracy

Soil / Substrate

Measuring volume

Interface

Of course, you want a sensor that is as accurate as possible. However, higher accuracy or additional special properties are usually associated with higher costs.

For scientific applications or very demanding irrigation tasks, the SMT100 is usually used, while the less expensive SMT50 is mainly found in simpler irrigation tasks. Both sensors are suitable for irrigation control, but differ in important details.

The SMT100 is designed for the full measuring range from 0 to 100 %. In addition, it enables a largely soil type-independent measurement with particularly high resolution. This is due, among other things, to its high internal measuring frequency, which effectively suppresses interference effects caused, for example, by variable electrical conductivity of the soil.

The measuring range of the SMT50 only goes up to 50% and the soil type has a stronger effect on the calibration curve, since a lower internal measuring frequency is used. The resolution is also lower than the SMT100, but is easily sufficient for many applications, especially when setting thresholds for irrigation control.

Moisture sensors for soils should provide measurement results that are influenced as little as possible by the soil type. Technically, this is a very big challenge, because the soil type influences the measured dielectric properties. One can reduce the disturbing influence by a clever choice of the measuring frequency. In general, it is better to use the highest possible measuring frequency in the range of a few hundred MHz. In the SMT100 this is realized, but it requires a high electrotechnical effort, which is reflected in the costs. The SMT50 operates at lower frequencies and is therefore less expensive. The disadvantage, however, is the greater dependence on the ground type.

Another important aspect when selecting a sensor is the measuring volume. The SMT100 and SMT50 belong to the class of point sensors, i.e. the water content is determined in a limited volume around the sensor. In some irrigation applications, a larger measurement volume is desired to better average over inhomogeneities of water distribution in the soil. The AquaFlex is particularly suitable for such applications. The measuring volume extends over the sensor length of 3 m and thus enables optimal averaging. With a sensor of this type, it is more difficult to achieve the same calibration and soil type independence as with a point sensor due to its design. For the practical application in irrigation technology, however, the large measuring volume is the decisive advantage.

The interface for control or data acquisition can be distinguished between analog and digital variants. With analog interfaces, a voltage value is usually output. The voltage range can be different depending on the sensor. For example, the SMT50 is designed for 0 - 3 V signal output. The analog SMT100 and the analog AquaFlex can be configured at the factory according to customer requirements. 0 - 10 V is the standard in automation technology, other voltage ranges such as 0 - 1 V, 0 - 3 V or 0 - 5 V are also possible on request. For the digital interfaces of the SMT100, the largest choice is either RS-485 with the software protocols TBUS, ASCII and Modbus or the SDI-12 standard, which is popular in environmental measurement technology. The AquaFlex is also available with RS-485 TBUS, ASCII and Modbus digital interface.

The advantage of the analog interface is the ease of use. The disadvantage is the somewhat lower resolution, the limited cable length and the higher hardware effort in the controller when using many sensors. Several digital sensors can be connected to a single cable (bus) and addressed by software via addresses. This reduces the hardware effort in the cabling, but the demands on the controller are higher and there is some configuration effort. Especially with RS-485, extended systems with long cable lengths can be realized with digital transmission. The transmission quality is excellent and the highest measured value resolution can be realized.

In soil science, both the term water content and the so-called suction tension are used. The volumetric water content is the proportion of water in relation to a unit volume. 30% volumetric water content therefore means that 30% of the soil volume consists of water. Suction tension, on the other hand, describes how firmly the water is held by the soil. This depends mainly on the pore sizes and their distribution. In small soil pores, the water is held tightly as in a capillary. The extent to which water from the soil is available to a root therefore depends not only on the volumetric water content, but also on the pores in which the water is stored. Suction tension is therefore an important parameter in soil science, as it describes how much water is actually available to the plant roots. In principle, suction tension would be the better parameter for irrigation control than volumetric water content. However, determining the suction tension is problematic. The classic measurement method uses water-filled clay cartridges with pressure sensors, but these require regular maintenance and are usually not suitable for underground installation and simple automation. Alternative maintenance-free methods for suction measurement often suffer from poor measurement construction and inertia, i.e., the response to changes in soil moisture takes a very long time, especially when drying out. For this reason, volumetric soil moisture sensors such as the SMT100, SMT50 or AquaFlex have become widely used in many irrigation applications.

When looking for inexpensive soil moisture sensors, one inevitably comes across mostly Far Eastern products and circuit board kits for microcontrollers and Raspberry Pi. These are used to determine either the electrical conductivity of the soil or its capacitance. Also on Youtube you can find numerous videos with these low-cost sensors, which promise a solution at an unbeatable price. However, disappointment is inevitable for several reasons.

The first thing to mention is the lack of or inadequate waterproofing. Completely unsuitable are circuit boards with bare electrodes, which corrode in the ground within a very short time. But even the capacitive board sensors that are touted as better are often defective after a short time. In the magazine Make of the Heise publishing house issue 2/2000 p.40-46 this was described in the article "Greenhouse with Arduino" by M. Sixt and shown in the picture. The SMT50 was compared with a China sensor "Capacitive Soil Moisture Sensor V1.2". The result after approx. 6 months of continuous operation was: "The SMT50 shows no signs of corrosion, whereas the cheap sensor is heavily attacked".

However, it is not only the absolutely necessary waterproofness, but also the "inner values" of the electronics play a major role. It is often simplistically assumed that, from an electrical point of view, the base behaves like a pure capacitor with alternating current. In reality, however, it is much more complicated. In addition to the capacitive properties, the electrical conductivity also has an influence on the AC behavior and, together with numerous other effects, leads to a strong frequency dependence of the so-called impedance of the base.

By selecting a suitable measuring frequency, these influences can be significantly reduced and a stable measurement can be achieved. However, in terms of circuitry, this requires the generation of high-frequency signals with significantly more expensive components than with the low-cost products. In addition, high-quality sensors such as the SMT50, SMT100 or AquaFlex always have a microcontroller built in, which contains, for example, type-specific calibration functions or individual calibration parameters. The high frequency components, the microcontroller and the calibration naturally increase the manufacturing costs, as does the waterproof potting of the electronics. The TrueFlex cable, which is suitable for underground installation and resistant to microbes, is also a special feature. You should not be satisfied with anything less for carefree long-term operation.

A high-quality soil moisture sensor is therefore also a precision product that incorporates a great deal of know-how as well as manufacturing technology and calibration effort.

Soil moisture sensors are not only used in irrigation technology, but also in many scientific disciplines. Soil moisture measurement is of course particularly important in hydrology as well as in agricultural and plant science. But water content also plays a major role in civil engineering. Scientific users have particularly high demands on the quality of the sensor technology and have therefore been using the SMT100 in particular for many years in numerous research projects. This is reported on in relevant technical journals, which, however, can often only be viewed for a fee. In recent years, however, so-called open access journals have become more widespread, allowing free access to research results. Some publications related to SMT100 are listed below and show the wide range of possible applications:

Hydrological research in karst rock

A soil moisture monitoring network to characterize karstic recharge and evapotranspiration at five representative sites across the globe.
Geoscientific Instrumentation, Methods and Data Systems, 9(1), 11–23.
https://doi.org/https://doi.org/10.5194/gi-9-11-2020

Study of infiltration in the pavement

A distributed soil moisture, temperature and infiltrometer dataset for permeable pavements and green spaces.
Earth System Science Data, 12(1), 501–517.
https://doi.org/10.5194/essd-12-501-2020

Biochar for soil improvement

Impact of Biochar Reapplication on Physical Soil Properties. IOP Conference Series: Materials Science and Engineering, 603, 022068.
https://doi.org/10.1088/1757-899X/603/2/022068

Remote sensing of snow and soil properties with radiometers

Snow Density and Ground Permittivity Retrieved from L-Band Radiometry: Melting Effects. Remote Sensing, 10(2), 354.
https://doi.org/10.3390/rs10020354

Alpine research projects

Monitoring soil moisture from middle to high elevation in Switzerland: Set-up and first results from the SOMOMOUNT network.
Hydrology and Earth System Sciences, 21(6), 3199–3220.
https://doi.org/10.5194/hess-21-3199-2017