Soil probes are expanding in capability and some systems are now able to measure salinity, nutrient levels and temperature in addition to moisture. Some, like those of Quebec, Canada-based ChrysaLabs, can provide almost real-time measurements of moisture and nutrients. Their probes are currently being tested in collaboration with Olds College in Alberta, Canada for moisture sensing in western Canadian soils, which differ significantly in soil type profiles compared to those in Eastern Canada.
Soil sampling with CrysaLabs equipment on the Olds College Smart Farm fields.
Photo: Olds College
At the Tasmanian Institute for Agriculture, a joint partnership between the Department of Primary Industries, Parks, Water and Environment and the University of Tasmania, Research Fellow Dr. Marcus Hardie and his colleagues continue to develop various sensor technologies. The ‘Smart’ Shovel’ can sense soil moisture, salinity and compaction. Hardie’s team is also working on ‘Below Ground Sensor Data Transmission’ so that sensors can be fully buried without risk of damage from machinery etc., with commercialization expected by the end of the year.
At the same time, Hardie and his team compare their sensors to existing ones, and have found that differences in sensor accuracy “can be massive.”
“The issue of accuracy needs to be carefully unpacked as there are different issues which may or may not be important,” he explains. “For example, many growers use the patterns in the data to decide when to irrigate, and thus they need regular data (hourly or every few hours), and actual numbers aren’t that important. For them, issues with temperature fluctuations and electronic drift are more important, as is the quality of data connections. But for growers who use the actual numbers and tend to take infrequent measurements, accuracy is quite important.” Actual numbers also matter if you’re using the sensors to determine when to irrigate as opposed to how much to irrigate.
Dr. Vasudha Sharma, extension irrigation specialist at the University of Minnesota in the U.S., has also found a wide range in sensor accuracy and agrees that degree of accuracy required depends on what information the farmer needs. She has been testing soil moisture sensors since 2019 and will be comparing five this year. “There is a lot of research going on into accuracy,” she says, adding that “salinity can interfere with the accuracy of electromagnetic sensors, so farmers using those sensors in that situation will need to do site-specific calibration using soil samples taken to a lab.
Dr. Vasudha Sharma with the University of Minnesota installs a watermark sensor.
Photo: University of Minnesota
To review, soil sensors, attached to probes, are either put in for the entire season (‘in situ,’ installed via an access hole usually with multiple sensors at different depths on the probe), or moved around (more
expensive) by being poked into the soil. Both have limitations.
In addition, sensors can measure moisture in two ways. “Electromagnetic conductivity sensors directly give soil volumetric water content, are more accurate and behave better in sandy soils,” Sharma explains. “They currently cost about USD$350. The other category costs about $50 each and measure moisture indirectly by measuring soil tension (how much the roots have to work to extract water) and converting that to soil volumetric water content. Some are better than others in terms of that conversion. In sandy soils, some skill is needed in ensuring good soil contact.”
Daniel Stefner, a research technician at Olds College, has been testing sensors for several years in
collaboration with colleagues at Alberta Innovates. “We focus on not on accuracy but on connectivity and usability to measure moisture, salinity and temperature,” he explains. “Data connectivity is a big issue in Canada. We’ve tested close to 40 probes at this point from five companies for moisture, and we’re now looking at how the sensors can be used to get a ‘water infiltration picture’ for fields versus pasture, different types of pasture, and so on. For soil nutrients, testing is ongoing as sensors become available. We’ve looked at dozens of sensors from four different companies.”
Daniel Stefner installs soil sensors with colleagues on the Olds College Smart Farm fields during the 2021 growing season. Photo: Olds College
Advice for farmers As with implementing any technology, farmers need to know what information they require from soil sensors that might impact a financial decision, and also do their homework on what’s available, but here is some specific guidance.
“For those with a small budget, install a few low-cost sensors in your field for two or three seasons, following recommendations of extension specialists or your agronomist for the right depth depending on your soil and your crops,” says Sharma. “Get to know your soils, how fast they dry out. Track humidity and temperature daily as well to make yourself a rudimentary model to follow. But if you have the money, invest in reputable electromagnetic sensors that can measure moisture and nutrients and temperature with a data logger. The more information you have, the better.”
Dr. Vasudha Sharma with the University of Minnesota takes soil moisture measurements. Photo: University of Minnesota
Stefner suggests farmers understand level of connectivity required for various systems, and also that they honestly assess their own IT skill level in setting up and using a particular system. He adds that simplicity is important “because you won’t use it or won’t use it as much if it’s frustrating. Some sensors work with cell networks but if you have no coverage, a private LoRaWAN (low range wide area network) that you have to set up might be affordable and worthwhile.” (In 2021 in the journal Sensors, a team from Italy published their results in creating a low-power wireless LoRaWAN network using commercial components and free or open-source software libraries. It demonstrates “the feasibility of a modular system built with cheap off-the-shelf components, including sensors.”)
Hardie agrees that “it’s not always about the sensor, it’s often more important to get the back-end sorted. That is, how are you going to access and store the data?”
He adds that if your irrigation system is limited in how much you can irrigate, or you can only apply irrigation at a single rate for a whole field, focusing on when to irrigate is the priority. Sensing should focus on the wettest and driest parts of the field. Hardie also stresses that “what’s commonly not understood is most growers use sensors to inform them when they don’t need to irrigate, rather than when they do need to irrigate.”
Sensing in future Looking forward, Stefner expects sensor technology to become simpler with more capabilities, but emphasizes that it’s up to growers to build their knowledge, both of their fields and their ability to use the data to make better economically-driven decisions.
All agree that connectivity is sure to progress in the years to come. Sharma says that “in five years, I see farmers being able to see their data online and not have to go to the field to download it. This is advancing very fast. Cellular connectivity is increasing in some parts of the world, but I think the soil sensor industry is moving towards use of satellites because there are areas where there won’t be cellular coverage for the foreseeable future.”
Hardie doesn’t think there will be much new innovation in sensors themselves, and notes that current sensor technology is 20 years old. The “bigger” issue for growers, he says, “is whether to focus on using soil sensors or plant sensors to decide when and how much to irrigate.” ●
Sensors for plants, trees and pests Exciting innovation is happening with sensors for trees, plants and pests, with this data typically being used in conjunction with soil sensor data and other data to guide maximization of yield.Israel-based SupPlant, for example, guides irrigation scheduling through integrating soil moisture data in an algorithm with current and projected weather data – and data from the tree or plant in question.Their “dendrometer” sensor measures micro-changes in the circumference of a tree trunk which corresponds to charges in weather conditions and soil moisture, but also correlates to tree growth, yield and growth stage.
SupPlant’s dendrometer on table gapevine. Photo: SupPlant
“We know from years of experience and thousands of sensors how to detect good healthy growth and how to recognize stress,” explains Nitzan Shatzkin, SupPlant’s chief commercial agronomist. “We’ve learned from experience that sometimes a few small mistakes along the way can affect the rest of the season, and we aid our farmers to avoid that.”
The company also use fruit sensors that measure the diameter of the fruit because measuring fruit growth on a daily base is crucial to ensure reaching fruit size and quality goals.
A SupPlant fruit sensor on a macadamia nut. Photo: SupPlant
Additionally, SupPlant is employing leaf temperature sensors, which provide insight into water stress. “We use them less as part of our commercial product but more for research and creating better models,” explains Shatzin, “and crop-specific features like managing stress in wine grapes to ensure both yield and quality.
”Vivent of Switzerland is offering Phytlsigns sensors that detect stress signals in the plant’s internal network long before visible symptoms of disease, pest attack or other conditions appear. Pest sensors are also on the way. Dr. Arezoo Emadi at the University of Windsor in Ontario, Canada has developed low-cost, disposable micro sensors that detect volatile chemicals in the air which signify the presence of different pests and plant stressors long before they’re visible to the human eye. It is now being trialed through a project led by Ontario Greenhouse Vegetable Growers.
More than 30 percent of honeybee colonies are disappearing each year, a rate that is not only economically devastating to farmers but represents a severe risk to global food supplies. While various diseases, pests, and pesticides lead to colony collapse, no other factor has been more devastating to the bees than climate change.
Climatech precision robotics company Beewise offers a solution. Utilizing 24/7 monitoring and smart technology, the company states it can significantly increase pollination capacity and honey production, Beewise's proprietary robotic beehive, the Beehome, detects threats to a honeybee colony such as pesticides and the presence of pests and immediately defends against them. Its automatic robotic system responds to threats in real time and requires no human intervention.
In addition to protecting and defending, Beewise helps honeybees thrive and flourish by reversing the trend of colony collapse. To help combat the detrimental effects of climate change on bees, Beehomes are thermally regulated; protect from fires, flooding, and Asian wasps; and provide enhanced feeding techniques for when forage (food supply) is not available to the bees.
The company states that Beehome reduces bee mortality by 80 percent, resulting in increased yields of at least 50 percent, while eliminating approximately 90 percent of manual labour when compared to traditional beehives. Beewise currently manages more than seven billion bees, which equates to 25,000 acres of pollinated crops.
Earlier this year, the company unveiled a new lighter-weight Beehome that increases hive mobility, enabling farmers to effortlessly care for millions of bees and ensure seasonal crop pollination.
Beewise also recently announced a USD$80 million Series C funding round led by New York-based global venture capital and private equity firm Insight Partners, with participation from Fortissimo Capital, Corner Ventures, lool ventures, Atooro Fund, Meitav Dash Investments Ltd, and Sanad Abu Dhabi. This brings the company's total funding to over $120 million. ●
Detecting crop disease invisible to the naked eye could soon be a reality for UK farmers, thanks to the development of a handheld imaging device, the Fotenix Echo.
The palm-sized device, from diagnostics company Fotenix, uses light across and beyond the visible spectrum, to enable the identification of plant stress.
According to Fotenix, trials have shown that Echo can identify the presence of disease in wheat, and will now be refined for launch into the agricultural market for use by breeders, agronomists and farmers.
According to Charles Veys, founder of Fotenix, the company has been entertaining the idea for some time. “In fact, our first prototype was in this form, however, the quality of the data was never good enough, so the system didn’t make it to the catalogue,” he noted.
“In this incarnation, the Echo had really only been trialed internally and needed development on user ergonomics and mode of operation in an end-user setting. Enabling crop diagnostics in this way will provide early detection of disease, mitigating its impact on crop health and informing protection regimes to safeguard yield.”
The trial and evaluation work was conducted by agri-tech innovation centre Crop Health and Protection (CHAP), with support from Rothamsted Research as part of the RTO Grant Support Scheme operated by Innovate UK EDGE. The scheme enables SMEs to apply for grants of up to £15,000 to access services offered by research and technology organizations and the UK’s Catapult network, and is aimed at businesses looking to grow and scale through innovation.
The CHAP team, familiar with technology evaluation, provided crucial information on ergonomics and usability of the device. Results showed that the Echo has potential for a variety of applications in agriculture, extending laboratory capabilities into the field in a cost-effective and simple package. ●
Researchers have created a wearable sensor for plant leaves that monitor plant health, including leaf water content — a key marker of metabolism and drought stress.
While metal electrodes have previously been used to monitor thirsty plants, it has proven difficult attaching these devices to hairy leaves, which reduces their accuracy.
Reporting in the journal ACS Applied Materials and Interfaces , Renato Lima and colleagues wanted to identify an electrode design that was reliable for long-term monitoring of plants’ water stress, while also staying put. The researchers created two types of electrodes: one made of nickel deposited in a narrow, squiggly pattern, and the other cut from partially burnt paper that was coated with a waxy film.
When the team affixed the electrodes to soybean leaves with adhesive tape, the nickel-based electrodes performed better, producing larger signals as the leaves dried out. And the metal electrodes adhered to the leaves more strongly than the ones made of burnt paper in windy conditions, likely because the thin squiggly design of the metallic film allowed more of the tape to connect with the leaf surface.
Next, the researchers created a plant-wearable device with the metal electrodes and attached it to a living plant. The device wirelessly shared data to a smartphone app and website, and a simple, fast machine learning technique successfully converted these data to the percent of water content lost by the plant.
The researchers say that monitoring water content on leaves can indirectly provide information on exposure to pests and toxic agents.
Because the plant-wearable device provides reliable data indoors, the team now plans to test the devices in outdoor gardens and crops to determine when plants need to be watered, potentially increasing yields. ●
Living organisms are constantly sending out electrophysical signals. Plants use these electrical signals to detect stress when attacked by pests and diseases. The Business Unit Greenhouse Horticulture and Flower Bulbs of Wageningen University & Research is investigating whether sensors can recognize electrophysical signals.
With this knowledge, diseases and pests can be recognized at an earlier stage, and control measures can therefore be taken more quickly. This is important for propagators: they want to supply clean plants to growers, who in turn want to detect possible contamination as quickly as possible.
Earlier research by WUR showed that it is indeed possible to intercept electrophysical signals. For this study, a strawberry plant was infected with thrips. With an electrophysical sensor, the contamination could be detected after only two days. This means a significant time saving compared to visual inspection: usually it is only after about five to six days that the leaves take on a silvery glow, which can only be seen by a good inspection of the plants.
The new two-year study focuses on a vegetable crop (tomato) and an ornamental plant (gerbera, petunia or Helianthus) and on three threats: mildew, thrips and a virus. During the first weeks of the research, Vivent's electrophysical sensor measures which signals the crop transmits in a normal state, and passes them on to the underlying software. This is followed by targeted infections with the damage organisms. The idea is that the software learns through artificial intelligence to recognize the specific signals of the crops on the damage organisms. ●
U.S.-based Solinftec is expanding the launch of its new agtech robotic platform, Solix Ag Robotics, into Canada in partnership with Stone Farms and the University of Saskatchewan.
The new technology aims to provide farmers and agronomists a new level of information to increase yields, improve the usage of inputs, lower environmental impact and support the global demand for food supply. The goal is to have the robot commercially available for the 2023 season for wheat crops.
Solinftec will run the new robot at multiple farms, one being Stone Farms in Davidson, Saskatchewan, aiming to adapt the technology to the specific needs of the Canadian grower and improve the algorithm for identifying weeds.
Solinftec’s Solix Ag Robotics is integrated with the company’s artificial intelligence platform, ALICE A.I., capturing information directly from the crops. Programmed with a neurological network featuring a complex detection algorithm, the new in-field robotic device has the ability not only to scan for crop health and nutrition, disease, insects, and weeds, but is built to monitor the entire field ecosystem and provide real-time insights.
“As we’ve done in regions across the globe, we are taking the robot directly to the fields to learn at a hyper-local level how they perform,” said Leonardo Carvalho, Solinftec’s operational director lead in Canada.”
Solinftec will also work in partnership with the University of Saskatchewan in Saskatoon to validate field results provided by the new robotics technology. ●
