Green spin

What is the ideal time for sowing and harvesting? How much fertiliser is needed? Is the use of pesticides justified? Mofato, the web application developed by the German startup Green spin, optimises the work of farmers by providing recommendations at the field level. Green spin’s algorithms, which analyse satellite images and meteorological data from the past 15 years, are not just useful to farmers: by limiting soil pollution, they are also good for the planet.

Over the past 100 years, agricultural productivity has been in a state of constant development. This is due to advances in plant breeding and genetics, the introduction of fertilisers and other chemical agents, and the use of larger and more efficient agricultural machinery. But intensive agriculture has shown its limits, and many farmers believe that the next revolution will be driven by new technologies, particularly big data and machine learning techniques.

Today, farmers use automatic GPS-controlled machines to fertilise the soil with a level of accuracy that would previously have been impossible, and have digital support tools to help manage their crops. "At Agritechnica, the world's largest trade fair for agricultural machinery and equipment, held in Hanover in 2015, participants have never seen so many applications and intelligent machines, all of them making serious promises. But for the managers of small farms, the benefits of these solutions are not always obvious", explains Sebastian Fritsch, Operational Director of Green spin. "In many cases, the publisher or manufacturer is the only winner: applications or machines collect data that allows them to improve their tools and ensures that they are among the market leaders of tomorrow. But that’s not all. Farmers may also become dependent on a few publishers, who could eventually abuse this unbalanced relationship. This is not good for either farmers or consumers.”

For these reasons, Green spin has devised an alternative to the established solutions. A solution that uses freely accessible data, without holding farmers captive.

In 2012, Sebastian Fritsch and Gunther Schorcht, co-founders of Green Spin, who had previously worked at the University of Würzburg (Bavaria), decided to put theory into practice by launching their startup. They were joined by Clemens Delatrée, who had previously worked as an environmental consultant in the private sector. With the help of a grant from the Federal Ministry of Economic Affairs and Energy (Bundesministerium für Wirtschaft und Energie, BMWi), they officially founded their company in 2013, while maintaining close relations with the university's remote sensing department, where they had carried out a large amount of research that subsequently became the basis for Green spin technologies.

"Very early on, we understood the potential benefits to agriculture of exploiting geolocated data such as satellite images, archives and weather forecasts. Most of this data is free and accessible to all, but rarely used because of its complexity. Hence, our idea was to create a tool that processes this data in the background, and provides the farmer with recommendations that are easy to understand and implement, without the need to understand all the details and calculations involved."

The Mofato application downloads the latest high-definition satellite images made available by NASA and, in particular, ESA (European Space Agency), which provide valuable information on the state of the crops, and links them to the available meteorological data, to provide the user with a dashboard indicating, on a plot-by-plot basis, the ideal periods for planting or harvesting, as well as the most favorable moments for spreading fertiliser (and how much), or whether pesticides are needed. "In the medium term, the farmer who uses Mofato as part of his planning tools will reduce his fertiliser and pesticide consumption, and can hope to improve his yield. It is good for the planet, and for the bank accounts of those who tend it."

"A fundamental requirement of our application is the scalability of the underlying IT infrastructure," says Fritsch. For each agricultural field, we have to analyse satellite images and weather data from the last 15 years. This mass of data to be processed requires high computational power. We have joined the OVH cloud Startup program and migrated our infrastructure to OVHcloud dedicated servers, which we complement with Public Cloud instances. These allow us to run calculations in parallel, to show our recommendations to our users faster, and to upscale more easily. With OVHcloud as our partner, we are perfectly prepared for all the challenges ahead."

Since it was unveiled to the public at the Agritechnica exhibition in November 2015, the application’s number of users has been steadily increasing. This is partly because the application is free for small operators, in return for which they allow the startup to continuously improve and enrich its solution to better market it to large enterprises, who also benefit from personalised support and custom functionality.

"We have over 130 users in Germany. We are now exploring Eastern Europe, notably Romania, and are aiming for a wider deployment by 2017. We help these farms optimise their processes. We are also planning to go further, promoting networking amongst independent farmers, so they can share their experiences and help each other. We believe that encouraging independent agriculture, amongst both software publishers and multinational agri-food companies, is good for consumers."