On a windy day high up in the hilly country in Rwanda’s Northern Province, Paul Tuyisingize observes the rapid flight of a drone above his wheat fields with an expression of wonder and delight. “I’m very happy with this project,” he says as the drone’s multi-spectral sensor gathers information about the condition of his crop. “I think it will help me to increase my yields.”
The use of Unmanned Aerial Systems (UAS), or drones, could have a transformative impact on agriculture. They are already widely used on large landholdings to map farm boundaries, conduct crop inventories and measure biomass development. Now, we are beginning to see how they could help smallholder farmers increase their yields and incomes.
Paul is participating in a research project supported by CTA and Airinov, a pioneer in drone-based farming applications, and managed by two Rwandan organisations, Charis Unmanned Aerial Solutions (Charis UAS) and the Regional Research Centre for Integrated Development (RCID). The ultimate aim of the project is to optimise the use of nitrogen fertilisers on wheat.
“When we set up Charis in 2014, I was interested in building and selling drones, and teaching people how to use them,” says Eric Rutayisire, Charis’s chief executive officer, “but I soon realised that people didn’t want to buy drones. They wanted drone services.” Eric began discussing the potential use of drones with agronomist Jules Kazungu, the Director General of RCID. Together they submitted a concept note to Giacomo Rambaldi, senior programme manager for Information and Communication Technologies (ICTs) at CTA. “Giacomo recommended that we should make this a research project focusing on the nitrogen requirements for wheat, an important cash crop for many smallholders,” says Eric.
In 2017, Charis was one of 14 African organisations to benefit from a CTA-funded training programme in the use of drones, held at Airinov’s Paris headquarters. Each trainee returned home with a winged drone equipped with a multispectral sensor and RGB camera, manufactured by Parrot. The companies paid 40% of the equipment, with CTA covering the rest.
Research in the field
“We are very keen to improve technologies that advance precision agriculture, and I believe that the use of drones is going to do that,” explains Jules Kazungu, whose company was responsible for gathering data on the ground. The project looked at the performance of wheat at three farms at different altitudes. At each farm, 24 micro-plots received six different fertiliser treatments. There were also control plots which received no treatment.
The Parrot drone was used to take aerial images five times during the growing cycle. On each occasion, Jules and his team took samples from the micro-plots and sent them for analysis at the laboratories of the Rwanda Agricultural Board (RAB). Charis then sent the imagery gathered by the drone to Airinov, together with the data from the soil samples. Airinov’s task – still in progress at the time of going to press – is to calibrate an algorithm for wheat based on correlations between the reflectance of the crop and dry matter and nitrogen content. The algorithm will enable extension agencies to use imagery taken from drones to assess the nutritional status of wheat and its nitrogen fertiliser requirements, and advice farmers how much fertiliser to apply and where.
Although the farmers involved in the pilot project didn’t directly benefit from the data collection, they were introduced to new methods of production. “In the past I never used fertilisers with wheat,” says Paul Tuyisingize. “I planted wheat after potatoes and relied on the fertility left over after the potato harvest.” And like most farmers, he used to broadcast his seeds, rather than sow them in straight lines. The project encouraged the farmers to sow in straight lines and apply fertilisers in prescribed quantities. All were anticipating that their yields would be 50% higher than in previous years.
The Rwandan government recognises that the agricultural sector could benefit from mechanisation, especially when large numbers of farmers plant the same crop in contiguous areas. “The same applies to the use of drones,” says Giacomo Rambaldi. “It doesn’t make sense to provide services to individual smallholder farmers, but it does if they come together as a group, such as a cooperative.”
Giacomo is keen to build on the achievements of the pilot project by establishing a larger research programme which will look at the costs and benefits of using drone services to provide advice to smallholder farmers. “For example, if we can establish that advice leads to an increase in yields worth an extra US$50 per ha, and the advice costs farmers US$8, that means that the farmer will make an extra US$42 per ha, less the additional inputs he or she may have to use,” he says. The use of drones, in other words, could significantly improve farm productivity and household incomes for smallholder farmers.