Space for food security

How did a space agency get involved in food security?

NSO was founded in 2009. But the Netherlands has been represented at the European Space Agency for about 30 years. At a certain point we thought it’s great that we’ve launched a number of satellites into space, but how are we going to start using them optimally? So that led to one of NSO’s missions, namely to promote the use of satellite data in society. Another mission is to maximize the use of the entire infrastructure that has been developed. And at the same time look after the government’s economic interests, especially in terms of innovation. In other words, we also want the Dutch business sector and knowledge institutes to benefit from satellite data. You could describes NSO’s role as a driver of innovation, economic growth and the use of knowledge.

NSO got involved in the discussion with the Ministry of Foreign Affairs about food security and supporting smallholder food producers in developing countries. The G4AW programme emerged from this discussion. It’s a special programme because it connects two worlds that had never directly interacted with each other before. On the one hand you have satellites – the high-tech, science side of it – and then you have the development cooperation side – smallholder farmers in developing countries, who are extremely vulnerable, because if a crop fails they’re immediately in trouble. You could say the two have nothing to do with each other, and yet we’ve discovered that they actually have a great deal to offer each other.

So how did G4AW come about?

The discussions at the Ministry of Foreign Affairs explored the question of what aerospace’s added value could be. Obviously it has added value for large organisations, for authorities, for knowledge institutes, as satellites collect basic data. And from this data you can create knowledge and act on it, often at the government level. But then someone raised the question, what exactly are the authorities doing with all of this knowledge? How, for example, does a farmer benefit from it? Do they receive advice that will help them make better decisions? We concluded that this wasn’t happening at the operational level. There were a number of pilot projects, research projects and small demonstration projects, but they tended to end without much follow-up. That’s when the ministry decided to take the link between satellite data and food security to the next stage.

So we examined that question for a year or so. We spoke to many players in the field, had a literature study carried out by students, and at a certain point it became clear that if we want to achieve our goal of improving food security, a number of things are crucial: first, quite simply, we need to connect to the smallholder farmer. If you can’t do that, then you have no added value. You need to understand their needs, otherwise you won’t be able to provide them with information that can be used to take action. So knowledge transfer and capacity building are part of this process, which is typically the domain of NGOs and extension officers from ministries of agriculture. Or companies. So that’s the first step that you have to take.

The second important condition was that while you can develop a service that’s suitable for farmers, it has to be economically viable as well, otherwise it’s pointless. There has to be a party or an organisation behind it that develops a financial model for the provision of these services, and of course the service has to be affordable. Affordable could mean it’s free, but it could come from other sources. It could be a system where customers – farmers or farmer cooperatives – pay, but then in such a way that they can benefit.

The most important criterion was that the initiative has to generate a service that has added value. We noticed that this market didn’t really exist yet. Why? Because it entails a number of risks, and the services have to be developed. The customers won’t be able to pay for it yet or may not want to, so essentially there’s a huge disincentive for organisations to invest in it. That’s why we as the government have been given permission to extend grants for projects such as MUIIS. That’s basically what’s behind G4AW.

How do you actually calculate at what point a smallholder will benefit from the service?

Assume that the operational costs of service delivery are a million euros a year. And assume a smallholder farmer earns 1,000 euros a year. Let’s say he’s prepared to invest 1% of that amount in the service, with the potential added value that he will stand to earn at least 10% more income. That means he will invest 10 dollars in the service and earn 100 dollars more a year as a result. So we said, that’s the minimum the farmer will have to spend on the service. The initial response from the sector was: ‘You guys are nuts, we cannot reach out to 100,000 farmers,’ but ultimately the sector came up with clever, inclusive business models, including aggregators and governments that make the number of farmers that need to be reached realistic.

How does the raw satellite data get processed?

ESA’s Sentinel mission is an important source. It’s a European satellite constellation basically consisting of three types of satellite, two of which deliver a new message about once a week, while another one transmits on a daily basis. These satellites supply different kinds of data. The companies and knowledge institutes have algorithms that can derive information from this data. Data that comes down from the satellite, before it even reaches the tech company, first undergoes several quality checks by the operator of the satellite, who has to make sure that the data have a certain reliability and quality. In aerospace terminology, they call it working from level 0, which is the raw data, to level 4. In other words, the information and quality improves as it goes from one level to the next.

The operator also carries out a number of corrections, for example atmospheric or geographical corrections. Then ESA will put the data into a database, and that’s where the tech companies source their data from. They have models that can derive information from this data. You could even say that the data is a ‘half’ product, because for a farmer it’s not advice or an actionable product yet. It’s the tech companies’ job to take care of the primary processing of the satellite data and package it into a product. In the G4AW project MUIIS, which is coordinated by CTA, eLEAF, for example, focuses on water and vegetation, while aWhere focuses mainly on meteorological satellite data and weather forecasts.

What happens next?

Then at a certain point all of that converges in a model, or a large algorithm. Think of it as a decision support system that contains the knowledge from which advice is distilled. Often you’ll need to supplement the data with other information to achieve your aim. You may need to combine it with information on local temperatures and moisture levels. How that’s all organised is up to the tech companies. Eventually the central system will produce specific advice for farmers covering a specific region. The advice is delivered via telecom or extension officers who go into the field with their mobile phone, while some farmers receive it through SMS or radio. Those are essentially the different channels used to reach farmers. But NSO’s role in all this is to ensure that the entire field of players come together and that the satellite data eventually gets turned into a service or financial product that reaches farmers.

How is this advice benefitting smallholder farmers?

Traditionally, a farmer in a developing country would respond to his environment or at most to a weather report that he received from the national meteorological service, but that is general information and not location-specific. So you have to realise that the reliability of weather data for farmers was poor. And there was no other way for them to source information. Farmers still depended on the traditional signals that they picked up on from nature. But unfortunately we’re having to deal with climate change. That means weather seasons are starting earlier or later, droughts are getting worse or rain is intensifying. Essentially weather has become much more unpredictable.

Satellites generate much more precise, location-based, and therefore reliable, data. Providing farmers with better and timely information in the form of weather forecasting and crop management, for example, puts them in a position to make well-founded decisions – decisions that will increase yield and income and result in the more effective use of agricultural inputs and water. And because satellite data cover large areas and are consistent in time and space, the basic information is continuously the same. As a result, you can offer a high-quality service over time. Continuous, high-quality data is important. Take a local outdated weather station, for example. They break down sometimes or work at half strength. There are people at these stations filling in numbers all day long, so sometimes they get it wrong. So it’s important to automate these systems as much as possible to eliminate human error and other factors you have no control over. In that sense, satellite data is extremely consistent and reliable.

G4AW has initiated 17 projects to date. What lessons have you learned so far?

Essentially the lessons we’ve learned is that cultural and organisational problems are often much more challenging than technological ones. Clearly, the organisation and the partnership have to be handled properly, otherwise you’ll suffer delays and risk not achieving your goals. We deliberately put a limit of three years on our projects. The idea being that in the first year you set up the partnership, and develop and offer your first service.

In the second year you roll out the service in the field and get feedback. Then you make plans to scale it up and roll out your training plans. And in the third year you have an improved service and take care of the full roll-out of the service. That’s the ideal situation. But many of these projects are being extended to a fourth year. Why? Because the reality is that the partners need more time in the first year to take all the necessary steps to work well together, to reach the farmers properly.

Another lesson is that the development of a business model can be very tricky. We have several projects in which the partnership has had to adapt the business plan, the business model. But we build space for that into our projects. We’re not dogmatic about that. We don’t tell our partners that they have to stick to every detail of their plan and business model. That’s simply not how it works. These models are based on our best insights, and if these insights change along the way we simply have to adapt the plan and activities.

But it’s important to have companies on board to pull the cart. They can provide continuity because they have the business drive. Profit is considered a dirty word because it’s sees as coming at the expense of the farmers. But these companies are often social enterprises involved in the environment, who are not in it to earn millions but want to continue to provide a service. Without that business drive, a researcher will wander from project to project, and without continuity the impact will be negligible. You certainly won’t achieve what’s needed to ensure food security.

In the end, the involvement of the private sector stems from the Ministry of Foreign Affairs’ realisation in 2011 that support for development cooperation was waning, and the way to turn that around was to introduce the economic component.