Most plant-based proteins are derived from the soybean. Currently, the EU imports around 32 million tons of soya (mainly as processed soya meal) per year to feed its livestock. This makes the EU highly dependent on soya producing countries such as Brazil, Argentina and the US. Furthermore, the European production of legumes has been steadily declining over the last 50 years. The EU is thus increasingly dependent on other countries’ protein crops. It also has adverse effects on soil fertility, biodiversity and water quality due to over-farming with other crops and a shorter crop rotation. This use case addresses the current lack of technological innovation in the cultivation and processing of protein plants. Through smart farming technologies, such as decision support systems and better sensor data to optimise machine task operations, this use case aims to reintroduce and increase soybean cultivation in the EU. Soy makes up around 40% of all plant protein used for animal feed in the EU. Amid the high dependency on imports of plant-based protein, this use case supports the EU’s production to diminish the global environmental challenges posed by soybean plantations overseas such as the loss of primary forest.
Boosting the EU soybean sector
For Georg Spreitzer, the Use Case Coordinator, standing still is going backwards. Therefore, he and his team developed a field management system based on sensors and high-end technology for intensive agricultural systems like soybeans.
Their decision support system (DSS) helps farmers to determine the ideal time and amount for irrigation, based on crop models and sensors, does not only reduce water consumption but also lowers electricity use for the pumps. Furthermore, the collected data are used as a basis for variable rate application of seed density, which means to adapt sowing according to soil properties. Ultimately, all of this leads to a higher yield and quality. For the development of the DSS, the team of Donau Soja and Soia Italia worked together with IT specialists from Sysman P&S. The latter are also active in our fruits use case Fresh Table Grapes Chain. The benefit of collaboration with use cases in other sectors of the project, first and foremost, is to reap the hard-earned knowledge and proven infrastructure of other agricultural techniques. Collaboration allows to apply this knowledge to an agricultural environment beyond the one it originated from. However, creating such a productive ecosystem with mutual understanding requires patience, empathy and, last but not least, resources. As a result, the team leveraged some modules from the Bluleaf decision support software from Sysman - originally developed for fruits and vegetables – and adapted the crop model and the irrigation advice to the specific needs of soybean growers.
Quality (% of protein)
To develop and validate their solution, the researchers rely on 2 test fields within one farm in Austria and 8 fields within one farm in Italy in the year 2020. On the latter, a trial was conducted to analyse variable seed rates depending on a soil fertility map. It was carried out under two tillage conditions (minimum and conventional) in two replicates. Apart from this, their general set up includes several sensors distributed throughout the fields. The sensors measure air temperature, air humidity, wind speed and direction, solar radiation, precipitation along with soil moisture in three depths (10cm, 20cm, 30cm). Additionally, data on the soil structure (with electrical conductivity mapping), the harvest amount and quality indicators like protein content or humidity of the harvested grain (with Near Infrared Sensors (NIR) mounted to the combine harvester system) is gathered. All the information is supplemented by location data and can thus be displayed as a map. This provides the possibility to correlate data from various sensors to a certain location at a certain time. While an external contractor provides the sensors, Sysman P&S implements the DSS and subsequently takes care of the distribution, hosting and maintenance. Farmers can take advantage of the gathered data by estimating the ideal irrigation time and amount. Moreover, it gives them a better understanding of different field zones’ productivity or assistance with variety selection.
Challenges during implementation
The challenge of connectivity is a deciding factor regarding the success or failure of a project. The weather sensors of this use case rely on a sufficient coverage of the Sigfox network.
Sigfox is a network operator that builds wireless networks to inexpensively as well as reliably connect low-power objects such as devices or sensors, which need to be continuously on and emitting small amounts of data. Even though the Sigfox IoT network covers a total of 5.8 million km² in 72 countries as of November 2020, the network in some rural areas in Austria in Italy can still be poor. Nevertheless, the use case team managed to connect the sigfox sensors in their test fields, enabling them to visualise the gathered weather data. While connectivity was an issue that could be addressed by the researchers, the actual weather obviously could not. In 2020 the weather provided more as well as more equally distributed rain as one would expect based on experiences of previous years in Austria. On the one hand, this is beneficial for the farmers since sufficient irrigation is one of the requirements of a good soya propagation. For the researchers, on the other hand, less irrigation actions carried out by the farmers meant less real-life data for the DSS irrigation feature. Nevertheless, they succeeded in bringing the feature to the market!
Promoting sustainable agricultural production requires farmers to adopt new technologies to increase the agricultural productivity, while conserving the environment, making adoption and diffusion important issues in agriculture. At the moment, the technologies for mapping quantitative and qualitative harvest indicators (NIR, yield monitor) are not broadly diffused among Italian or Austrian farmers. As a consequence, neither the farmers themselves nor the agricultural technicians are very familiar with the usage of such novel technologies and depend on support as well as time to collect their own experiences. Furthermore, each combination of technology and the specific environment it is deployed in can cause individual challenges. Especially the accuracy of quality indicators measured during harvest with the NIR-Sensor, like protein content, moisture, oil, etc. are highly dependent on the very specific setting of sensor- and combine-type. The NIR-Sensor needs to be placed in a position where the amount of soybean passing the sensor is high enough, while the design of the combine, especially the elevator where the soybeans are transported to the tank, specifies the possible positions for mounting the sensor to the combine.
That innovation is a constant cycle of reiteration and subsequent improvement is also demonstrated by the latest feature called comprehensive soya variety selection which the use case team worked on during the home stretch of the project period. Since this feature focuses on a number of different soybean varieties, a vast amount of agronomic and climate data is needed. This ultimately enables the researchers to determine the specific relations between agronomic as well as climate parameters and the quantity and quality of the harvest. These insights, in turn, help them to further explore features based on the correlation of soil conditions, the particular soybean variety and eventually the protein outcome of each variety.
Achievements, products & services
Decision Support System for soybean growers to maximise yield and protein content
Information about production efficiency of different field zones through GIS based field management
Easy to install and movable IoT weather sensors
Mapping of protein grown on the field through NIR-sensor
Increased profit and optimised irrigation feature
Documentation of actions