First provoke the plant and then listen to what it says

Drought is not good for a tomato plant. Leaves droop and turn brown. Flowers wither, fruits stop growing. Once the effects of drought are visible, it is almost impossible for a plant to fully recover. But a plant already gives many signals that it is suffering from drought before it can be seen in its leaves and flowers. You just have to have eyes and ears for that.

Researchers at Plantenna, a cooperation program of TU Delft, Eindhoven University of Technology, Twente University and Wageningen University have developed these in recent years. This summer they measured whether their new sensors to measure drought stress are doing what they should. Today is the last day of measurement, next week their greenhouse in the Delphy Improvement Center, a horticultural research complex in Bleiswijk, will be emptied.

Disinfecting hands twice, also a cloth over the recording device please, clumsily walking over a moving shoe brush, wearing a protective suit, hairnet, gloves and plastic over the shoes. It takes ten minutes for our party – three researchers, a press officer and a reporter – to enter the greenhouse. Everything to protect the tomato plants against accidental diseases. Tomatoes are sensitive plants.

With the roots in glass wool

“It was sometimes sweating this summer. We know from our measurements that the temperature at the top of the greenhouse was regularly around forty degrees,” says Marie-Claire ten Veldhuis. She walks considerably more handily over the shoe brush. Ten Veldhuis is an associate professor of water management at TU Delft and one of Plantenna’s researchers. “The plants were already mature when we started here in May, other research has been done before us.”

There are six rows of tomato plants in the greenhouse. At hip height, the plants are placed in a small black container in which the irrigation sticks have also been inserted. Underneath that is a pack of glass wool – the same stuff used to insulate roofs – in which the roots grow. Thick bundles of bare stems run horizontally along the rows. Each stem, fifteen meters long, belongs to a plant that grows towards the ceiling. Colored labels help distinguish the plants in the tangle.

“They grow twenty to thirty centimeters a week,” says Ten Veldhuis. Yellow flowers are visible at the top, small green fruits below and ripe fruits hang at breast height – handy picking height. Just like in a real greenhouse. “The plants grow up along a thread. At the end of the week, Delphy’s managers lower the plant a bit, and move it up a spot if necessary. That’s why the stems go all the way down below.” The excess leaves at the bottom of the plant are pulled off and collected on the ground.

How far can you go?

“Drought is already a problem for some crops. Not with crops in a greenhouse, everything can be controlled there, but in the field,” says Ten Veldhuis. “In the future you can imagine that there will also be a scarcity of nutrients, such as phosphate. If everything is no longer optimally available, you will shift the cultivation conditions. Ultimately, we want to know how far can you go before a plant is damaged and can you detect that point with sensors?”

Let’s get to grips with the drought first. New, innovative sensors have been developed within the Plantenna project. Such as a ‘stomatoscope’, a microscope that hangs high in the greenhouse, just above a leaf, where it shows the movement of the stomata in the leaf. Hidden in a protective package of gray foam around a stem hangs a sap flow meter that emits heat pulses and looks a little further on what is left of that heat. From this, the speed of the sap flow can be measured. A plant radar hangs from another stem, which emits electromagnetic waves close to the stem. Water absorbs part of the signal, from the part that has been reflected it can be deduced how much water is in the stem.

There are also existing sensors in the greenhouse, such as a sensitive spring that measures the shrinkage and expansion of the stems. They are there to validate the measurements of the new sensors.

The trials are not taking place in the field, where drought was also a problem this summer, but in a greenhouse because the conditions can be controlled well here. The plants get their nutrients from the irrigation water, in which everything is dissolved in exactly the right amount. Every few minutes there is a hiss, then CO₂ sprayed into the greenhouse. And in the corner is a cardboard box. “It contained bumblebees, every eight weeks a new load of bumblebees is released to pollinate the plants,” says Mariska Koning, postdoc researcher at TU Delft and supervisor of Plantenna’s fieldwork.

Each department involved has its ‘own’ sensors. “This is really a collaboration between microelectronics on the one hand and water management and botany on the other,” says Ten Veldhuis.

Koning, with a background in physics and meteorology, is particularly interested in the environmental factors of the plants. A net of fiber optic cables hanging between the plants is intended to provide a three-dimensional picture of the temperature in the greenhouse. The temperature is measured at every thirty centimeters with flashes of light. “It can be ten degrees warmer at the top than at the bottom. The position of the sun makes a big difference, of course, but even a leaf that provides shade on a leaf below already creates a temperature difference,” says Koning.

“Temperature variation creates airflow, a crop needs it to be able to evaporate,” says Koning. „Everything a plant absorbs except CO₂ enters through the juice. If there is too little air flow, the moisture will stick to the leaf, then evaporation will stop and juice will no longer be absorbed at the bottom.” The temperature network also provides additional information for the other sensors: if heat stress occurs, at what temperature does it occur?

A lot of variety

Of course there is already a thermometer in greenhouses, solar radiation and humidity are usually already monitored. “But that happens with one or two such boxes, which hang in the middle of the greenhouse.” Koning points to a measuring box that is also hanging in this research greenhouse. “So it measures in one place, far from most plants. We see that there is a lot of variation in the greenhouse. And our sensors are very close to the plant.”

The researchers have conducted four stress tests in recent months. “Then we took out the drippers with irrigation, usually for a day or two,” says Koning. Always with a different row of plants, so that the initial values ​​were almost the same for each stress test. “You don’t know whether the plants are damaged and whether they react differently to new drought stress as a result.” When the glass wool slabs were completely dry, the irrigation was allowed back in, and the plants were given time to recover.

“During the last stress test, we left the irrigation out a little longer,” says Koning. “The stress signals were much clearer then. You expect that, of course, but it was nice to see that reflected in the sensor data. There was still nothing to see on the outside of the plant.”

Tomato plants are fine plants for this type of research. They react quickly, so you can quickly see whether the sensors are picking up useful signals. “The recovery is also interesting. As soon as those drippers go back you will see a change in one go. Stomata open again, the sap flow immediately starts,” says Berend de Klerk. He has just graduated in high tech engineering from TU Delft. His graduation research was about a plant microphone, he has now started a startup around the sensor.

A gray package

“It’s hanging up there, on a stem,” De Klerk points to a gray package high in a tomato plant. The microphone is also encased in foam, as the hiss of the CO₂– caused spray noise. The sensor itself is only small. “We listen to the sounds the plant makes. We knew that a plant makes noise, but we wanted to learn to understand what it is saying during this measurement campaign.”

What does a plant sound like, and what does it say? De Klerk snaps his fingers quickly. “Something like that, but higher and very soft of course. We cannot hear it. When it is dry, it makes more noise.”

It is striking that although the sensors have a different approach, they all have to answer the same question: does this plant suffer from drought? “That’s right,” says Ten Veldhuis. “But whether they are all equally sensitive, and especially whether they pick up the fast dynamics well, we all have yet to find out. In the coming six months we will analyze all the data and write down our findings. Some sensors are truly scientific instruments, such as the stomatoscope. A beautiful device, but far too laborious for the field. We hope that some of the cheap sensors will turn out to be easy to use, so that you can buy a lot of them and put them in the field.”

We barely lasted upstairs in that hot greenhouse, they continue to produce fruit cheerfully

Marie-Claire ten Veldhuis Tu Delft

The temperature net has already been successfully deployed in a fruit orchard. Koning: “When it is freezing cold, flowers can freeze to pieces, and then you will not get any fruit. Fans are therefore sometimes used in the orchard, which actively mix warm and cold air so that flowers remain just above freezing. The question was, when do you turn on the fans and how much does it actually do. We were able to follow the temperature changes very well with the net.”

Even if it turns out that the other sensors also do what they promise, the way to ‘connected cultivation still far. After the science come the practical bumps. Most sensors are now supplied with power via a plug, data is read by attaching a computer. “Power should actually come from small solar panels. And you want to read data online, or collect it in a data logger,” says Koning. “We’ve had quite a bit of trouble finding enough power strips and long cords. That is not possible in a large open field.”

The Delft people, technicians without much plant knowledge before they started, have started to look at plants differently during these months of measuring. “These plants are very active, they are alive. And they are so resilient,” says Ten Veldhuis. “We had huge heat waves this summer. We barely survived in that hot greenhouse, they continue to produce fruit happily.” Koning: “We tried one of the sensors on a houseplant beforehand. So nothing happens there at all. Of course they are also grown on that, they have to last a long time, but what a world of difference.”