Which sensors to use, how many, how to install them and how to interpret the data to program the risks are some of the most frequently asked questions of growers interested in implementing this technology.
In recent years, sensors have become an effective tool for managing irrigation in agriculture, since they make it possible to know the water state of the soil in real time.
However, to make proper use of these devices and for the data they provide to be reliable, several technical criteria and concepts must be taken into account. For example, there are sensors that determine the availability of water in the soil (matric potential sensors), others allow to determine the amount of water in the soil (humidity sensors).
Edgar Hincapié, researcher in the Soil and Water area of Cenicaña, explains some aspects of the use of sensors to monitor the matric potential in the soil and irrigation management in the cultivation of sugar cane. In the next edition of the Information Letter, criteria to be taken into account for the use of humidity sensors will be presented.
What are soil matric potential sensors?
Soil matric potential sensors (PMS) are those that directly or indirectly measure the force with which the soil holds water. This potential reflects the hydric state of the soil: the lower (more negative) is the value, the greater the force with which the soil retains water and vice versa.
There are several types of PMS sensors: tensiometers, granular matrix sensors, gypsum blocks, dielectric and thermal sensors. The most widely used sensors today are sphygmomanometers and granular matrix.
What is the difference between sphygmomanometers and granular matrix sensors?
Its way of functioning. While tensiometers measure suction or vacuum, granular matrix ones are based on the principle of variable electrical resistance.
When a tensiometer is installed in the ground it changes according to the inflows and outflows of water, creating differences in suction. For example, if the soil is dry, water comes out of the tensiometer and creates a negative pressure; When it rains or irrigation occurs the opposite happens: the water enters the tensiometer and the suction is reduced. Suction is measured on the sphygmomanometer with a pre-calibrated either a clock or electronic vacuum gauge.
On the other hand, when a granular matrix sensor is inserted into the soil, it is balanced with the moisture content; once installed, an electrical current is passed through the sensor. When the soil and the sensor are wet the current is easily conducted; on the contrary, if the soil is dry and therefore the sensor, the passage of the current is limited. By measuring the resistance of the sensor to the passage of current, the water state of the soil can be determined. These types of sensors require calibration for each floor where they are installed.
How many and where to install the PMS sensors?
The number of sensors that are installed depends on the way of making the measurements: manually or automated. It is important to clarify that the measurements are made by irrigation units and since an irrigation unit can contain different soils, the sensors must be installed in the most representative soil, that is, the one that occupies the largest area.
Manual measurements
- Five to six measurement points per irrigation unit.
- Two sensors are installed at each point: a superficial one (between 20 and 30 cm deep) in the area where the greatest volume of roots is concentrated and a deep one (between 40 and 45 cm).
- They are installed 20 cm from the center of the cane stock.
Automatic measurements
- One monitoring station per irrigation unit and located in the most representative soil.
- The station must include 6 sensors that are installed in 3 different grooves: 3 surface sensors (between 20 and 30 cm deep) and 3 deep (between 40 and 45 cm), and a temperature sensor installed at 30 cm depth.
The data from the surface sensors are used to program the irrigation and the data from the deep sensors, to verify the depth to which the water reached.
How is the timing of irrigation determined with PMS sensors?
The monitoring of PMS sensors allows determining the moment to start irrigation in order to optimize water use and be more efficient in the work. Typically, the need for irrigation can be predicted two to three days in advance.
In the case of sugarcane in the Cauca river valley, Cenicaña has determined that between -20 to -80 kPa is the optimal range of matric potential for the varieties currently planted in the region, that is, at this time plants can extract soil water without much effort.
Therefore, when plants begin to have difficulty extracting water from the soil, it is time to water. For crops in soils of medium to fine texture, irrigation should be started when the threshold of matric potential of the soil is -75±5kPa. For sandy textured soils the threshold for the start of irrigation is -60±5kPa.
What do the measurements made by the PMS sensors indicate?
PMS sensors installed in medium to fine and very fine textured soils indicate:
- 0 to -10 kPa: soil close to saturation point. This condition is normal for a period of approximately 24 hours after irrigation or heavy rainfall (> 25 mm).
- -10 to -20 kPa: the soil retains water with little energy and the plant can take it with minimal effort.
- -20 to -75 kPa: optimum range for sugarcane cultivation. Plants can extract water from the soil without much effort.
- Less than -75 kPa: threshold for the start of irrigation. Plants begin to have difficulty in extracting water from the soil.
Step by step for the installation of matric potential sensors (tensiometers or granular matrix sensors):
1. Saturate the sensors before installation. For this they are immersed in distilled water for 24 hours.
2. Select the points for the installation of the sensors, clean the area and dig a hole to the required depth and with a diameter slightly greater than the diameter of the sensor, using a drill hole.
3. Once the holes have been dug, take a little of the soil that was extracted from the hole, add water and prepare a thick and homogeneous mixture, then pour part of this mixture into the hole and immediately introduce the sensor to the determined depth. As the sensor is inserted, the soil / water mixture should flow upward and evacuate the contained air. This operation is performed to obtain a good contact between the sensor and the ground and therefore a correct measurement.
4. Once the sensors have been inserted, fill the holes again with the soil and lightly compact it with a bar or piece of wood in order to avoid the formation of a preferential water flow channel that would affect the sensor measurement.
The installation of the sensors must be carried out after irrigation or precipitation greater than 25 mm.
Readings can be done 24 hours after installation, once sensor / ground has been balanced.
How are PMS sensors read?
PMS sensors (sphygmomanometers or granular matrix sensors) can be read manually or automatically. In Cenicaña low cost manual and automatic readers have been designed and built.
Manual reading: To read the tensiometers and granular matrix sensors, electronic meters are used that are inserted or attached to the equipment. In both cases the readings are made sensor by sensor (one by one) and the data is recorded in a format previously designed for this purpose.
Automated reading: A datalogger is used that allows the data to be recorded continuously and to program the recording of information through the time of six PMS sensors and a soil temperature sensor. Data can be manually downloaded to a computer or transmitted to a workstation using a mobile phone modem so that users can view current soil conditions from any computer or mobile phone, making irrigation management easier.
How are these sensors calibrated?
The granular matrix sensors used to measure PMS require calibration, since they measure the current flow that passes through two electrodes. Therefore, it is necessary to generate a calibration curve that relates this variable to the matric potential of the soil. Curves must be generated for each floor where the sensor is installed.
For this, soil samples with a built-in sensor are subjected to a different suction voltage (matric potential) and in parallel the output current of the sensor is measured in kOhms. Later these two variables are related and the calibration curve is constructed.
Table 1 presents an example of manually measured matric potential records with sphygmomanometers at two depths (30 cm and 45 cm) during an irrigation cycle.
Day No. |
Matric potential (kPa) |
|
30 cm |
45 cm |
|
0 |
3 |
10 |
2 |
12 |
16 |
5 |
21 |
17 |
7 |
30 |
21 |
10 |
35 |
24 |
14 |
42 |
28 |
16 |
49 |
32 |
18 |
54 |
36 |
20 |
61 |
42 |
21 |
65 |
48 |
22 |
69 |
51 |
23 |
74 |
53 |
24 |
77 |
55 |
The first reading (zero day) corresponds to the measurement made 24 hours after watering and the reading made 24 days later corresponds to the threshold for re-watering.