Deployment of irrigation-related sensors and probes

There has been a change in the perspective of the stewards and managers of the irrigation projects. It is no longer simply a matter of knowing that the irrigation system is working correctly or whether the equipment is performing well, rather they also have to manage the whole upstream and downstream aspect of the water supply, i.e. for the irrigation system, from the withdrawal of the resource right through to the point where the water comes in contact with the soil or is lost through drainage. 

Until recently, many questions have remained unanswered: What happens to the water that is applied through irrigation? Has the water been applied efficiently? Has the rainfall been effective? What is the actual effect of syringing (light irrigation) on the temperature and humidity of the surface of my golf course? How much water has been consumed per day, week or month? What effect does the withdrawal or abstraction of water have on the natural environment?  Quite simply, is irrigation being managed properly?

Recent developments in technology now enable us, in 2021, to answer some of these questions. The idea is to use sensors that consume little energy, connected to wireless and energy-efficient communication panels or boxes, which send their data directly to the Internet. A web-based management platform then deals with displaying and processing the data so that it can become usable.

 

Here are the main uses that are widely operational and generally deployed:

• Among the “technical sensors”, the automatic reading of the meters not only allows for the volume of water passing through the meter to be recorded but it also provides an instantaneous estimate of the flow rate. The reading of the pressure in the system, as well as the water level in the ponds, the groundwater level in the boreholes, tanks, etc. are of particular value because they not only allow the user to know the level from a functional perspective and also enable him of her to know the soil water reserve expressed in terms of volume.

This latter aspect is typical of the connected objects technological revolution: taking a reading of the groundwater level of a borehole is nothing new, many elaborate pumping installations already have this type of sensor, but up to now they have required a sophisticated form of automation and a software to read the data, with the user even having to read the recorded values manually while on site. Nowadays, a sensor can transmit this information quite simply to an Internet platform, without any pre-programming and the user can independently manage the graphic or digital display.

We should also mention the sensors that allow the user to know the condition of the remote pumping installations, whether or not there are any faults, or information about the configuration of the meter that triggers certain devices with a trip counter (treatment, pumping-out, etc.).

• The so-called « agronomic » sensors relate to those that measure the moisture content of the turf or trees, and the temperature of the soil, etc. We should also mention, that, traditionally, they form a part of the weather station (solar radiation, humidity and air pressure, wind speed and direction).

The (soil) moisture sensor is one that must be installed in order to have a clear vision of the soil water status. The amount of information that can be supplied is endless, providing that the proper equipment is used. It is not enough to simply place a sensor or probe in the soil. The ideal solution is to measure the moisture at two or three depths in order to have an understanding of the movement of the water, the infiltration rate and ensuring that the soil water reserve in the zone where irrigation is applied is actually being replenished. Ideally, there should even be a sensor quite close to the surface, which provides information that lets the user know whether the water is really penetrating the soil. Sometimes, the results can be very surprising!

Being able to ascertain the temperature of the soil is also an irreplaceable asset, if it can be measured continuously and the data analysed.

One particular aspect of the rain gauges should be mentioned: the use and automatic reading of the precision rain gauges has opened the eyes of many irrigation managers with regard to the real effectiveness of the rainfall during the hot season, and this can be achieved via a sensor and a simple communication box used on an urban area or large site, such as a park, golf course or sports complex.

In the same way, it is now possible to incorporate into the traditional irrigation controllers certain rain gauges that measure the rainfall in real time (with an accuracy of around 0.2 mm), so as to be able to stop irrigation when a precisely defined threshold is reached (4.2 mm, for example). This is not the case with the traditional rain shut-off devices used on green spaces, where the level of precision is relative…

We can add less well-known sensors to this list, such as those that measure the salinity, pH, oxygen level in the water, etc., which are becoming more and more widely used.

• The communications networks provided by these sensors vary greatly, which means that, in theory, this allows for their deployment over a large area, such as a town, of sensors of all types with the quality of the data adapted to each individual need.

Explanation: the low-energy sensors use low-power networks, such as Sigfox or the LoRaWan networks of certain telephone operators, even if the country’s outlying or rural agricultural areas have poor coverage. The sensors are operated by batteries or even by solar energy and thus they are self-reliant for the duration of the irrigation season. The installations that require greater supervision use the 2G-3G-4G networks or even the affiliated networks (LTE, for example). The energy required for the use of these networks often prohibits the use of battery-powered sensors, so an electricity supply is required. However, it allows for large amounts of data to be sent and thus very precise information becomes available, sometimes almost within a second.

• There should be a particular focus on an Internet platform that handles the data rapidly and with a good processing capacity. In fact, the sensors have relatively little on-board intelligence and the processing, reconciliation and displaying or read-out of the data is conducted entirely by the remote software.

Bearing in mind the constant development of the servers, communication networks and sensors, the platforms that manage these sensors are different from the software packages that traditionally manage the irrigation process (irrigation software or access to irrigation controllers, equipment that is now well-known to the farmers). Once the data is transmitted to the Internet server, its display or read-out is entirely managed by the user. The display of historical data, subsequent data comparison and use of the measurements is completely free, from the moment the values are recorded at some point or another.

The internet platform allows the user to compile and interpret the data in conjunction with the use of an irrigation system that will benefit from them.

It is the principle of assembling brick by brick: the user only needs to record the reading of one single sensor (the reading of the water meter or even the rainwater gauge), then complete the installation over time with the subsequent addition of more sensors.

 

Does the use of these tools represent a valuable contribution compared with the traditional form of irrigation management?

The response is clearly positive. Reading the moistures sensors directly on site and recording the rainfall and temperature allows for the irrigation process to be thoroughly analysed. It becomes possible to define the soil water-holding capacity as well as the critical thresholds; confirm whether or not the rainfall is effective during the hot periods; reduce irrigation while being fully informed of the facts when the soil water reserve has reached full capacity and the plant’s consumption is low.

Having an analysis that provides information about the effectiveness of the irrigation scheduling (irrigating once or several times, early or late at night, in large or small amounts, etc.) is clearly an advantage, and this is directly due to the connected tools that have been available recently.

This effectiveness coincides with the decline in the number of weather stations installed for use in irrigation, as they do not provide direct responses as far as the practice of irrigation is concerned.

Going beyond simple experiments, the managers who have these tools available have already substantially modified the way they irrigate since these sensors were installed. The method of irrigation used on the tramways, green spaces, sports fields and golf courses has changed significantly over the last few years as a result of these tools, insofar as there is less water used and irrigation is more efficient (less leaching of the soils, for example).

Despite these positive results, a certain number of points must be clarified. Understanding them will help to prevent the inconveniences related to a technology that is still flourishing and which leads to an array of solutions that could appear to be quite appealing but you must know how to separate the good from the bad.

First of all we have the problem of compatibility. We have the open communication solutions, passing through the operated networks (supplied by the communication operators), or even solutions that use a private captive communication, often making use of a Gateway, which acts as an intermediary between the sensors and Internet, meaning that only the original manufacturer can provide the components and support required to keep the installation running.

The compatibility of the sensors is also an issue. The modems must have connectivity with the industrial sensors, distributed by the manufacturers and easily replaceable by a similar modem produced by another manufacturer.  Generally-speaking, the industrial products are preferable to the customised (made-to-measure) products, even though they may seem quite attractive.

Furthermore, the modems and other connected objects must have been specifically designed or adapted for irrigation. How many sensors or modems imported from a role for which they were adapted clearly do not work once they are installed in the irrigation systems. We have lost count of the number of sensors that exhaust their batteries by relaying information during the winter period, when the irrigation system is shut down, or even those that cannot withstand the moisture caused by the presence of the sprinklers. Or even quite simply situations where a battery only allows for a report to be sent once a month, which is unacceptable for a whole irrigation season.

We must also take into consideration the level of on-board intelligence of each sensor. It must be remembered that the more intelligence and advanced programming there is on board a sensor, the more costly it is to install and it is more complicated to keep it operational over time.

We must remember that wireless communication relies on a radio connection and that very frequently the information is lost. Thus, nothing beats reading the water meter visually in order to record the consumption. This does not call into question the automatic recording of a water meter, which offers some useful opportunities nevertheless, rather it is obvious that we have to be clear-sighted enough, so as not to be disappointed with these tools. Furthermore, the positioning of the transmission boxes in the inspection chambers or valve boxes could significantly interfere with the transmission of the data, particularly in an urban environment. This is all the more the case for certain devices designed to be used by the gardeners (we are thinking of the battery-powered connected controllers) and the efficiency of the transmission obviously depends on the device being positioned properly.

In the same way, although the low-power networks are excellent for relaying the information collected by the sensors,  two-way communication (sending a message or a command to a control board or box) becomes far more complicated. Even though, theoretically, the Sigfox and LoRaWan networks, for example, authorise several messages known as “downloads” per day, in reality, the technical complexity of sending the messages as well as the delays means that they are far from being 100% reliable.  Controlling or managing the system is, therefore, possible but with a success rate that cannot be guaranteed and this must be taken into consideration.

Furthermore, it is important to recognise that the number of sensors to be used should be limited to the same site or the same manager. In fact, an analysis has to be conducted in order to be able to interpret the results. Then we have the task of adapting the programming to the irrigation system and sprinklers or the installation, after this analysis has been completed. Increasing the number of measurements and readings taken can often be counter-productive.

In the irrigation sector, the specialised software packages have performed particularly well for more than 20 years. Many solutions available on the connected sensors market for irrigation also purport to be able to schedule irrigation from software available on the Internet, which has a practical side, but we can see that the technological gap is immense when compared with traditional software. If the capacity of these new tools appears to be sufficient to manage limited irrigation in green spaces, its use on sports fields or complex installations can still be challenging (especially if we compare it with the capacities of decoder-type controllers, which have been available on the marker for more than 15 years now). Or perhaps we will have to accept going back to a basic form of irrigation management, which would be the last straw for the products where technological modernisation has also become a reality.

We have seen in particular the introduction of fully-connected irrigation controllers, uniquely connected, which are impossible to use without employing the internet server, remote communications and smartphone applications. Whereas all the renovations of the irrigation system over the last 10 years consists of installing controllers, which the gardeners or technicians in the field could use on site (even automatically reloading every evening the customised programmes decided at the highest level).

Lastly, the Holy Grail of full automated irrigation has not yet been achieved by the Internet-connected sensors. Analysing the soil moisture measurements in conjunction with irrigation and rainfall does not yet allow irrigation scheduling to be automatic. The most that can be achieved is to shut off the irrigation automatically when the situation is clear-cut (heavy rain or waterlogged soil). But this does not mean that the irrigation modulation can be adjusted to a tenth of a millimetre, as a serious greenkeeper would do... In fact, automatic irrigation scheduling does not actually exist.

Finally, it should be remembered that wireless sensors are battery-operated (less frequently powered by a solar panel). Over these last few years, we have seen an explosion in the number of sensors scattered around different sites but which will soon need to be maintained, if only to replace the batteries!

In conclusion, the deployment of the new connected tools for irrigation has been in decline for a number of years now. The equipment must be chosen for its suitability to the specific characteristics of irrigation and preferably from industrial manufacturers who will still have a presence on the market in a few years’ time. The sensors installed, if they are targeted and limited, allow for records and analyses to become available, which are clearly relevant to the sound management of the proper management of the irrigation system, on a daily basis, and more generally the results is an efficient use of the water resource. They represent accurate unmatched on-site sources of information, which are actually very relevant. In order to exploit this data and ensure that these technological devices remain in operation, it would be advisable to appreciate their contribution instead of indulging in the “totally connected” devices, which would only exacerbate the shortcomings inherent to this type of advanced technology. Finally, these tools are efficient when they are used in conjunction with well-functioning centrally controlled irrigation systems, without which there would be nothing to gain from having so much data available.

Author
Raphael Karm, Terideal