In an era of increasing weather variability, smart systems for agriculture enable solutions
How ´agricultural intelligence´ benefits farmers, investors and everyone else along the agricultural value chain
Your company is about delivering agricultural intelligence. What does that mean?
Farmers all over the world are experiencing a kind of weather variability that they have never experienced before. This is causing enormous consternation because farmers need the seasons to be normal. If it is too hot, too dry, too wet or too cold, it will negatively impact their yield. Agricultural intelligence is the only way to manage and adapt to changing environmental conditions.
Agricultural intelligence provides aggregated information within a certain context that is of tremendous value across all segments of the agricultural value chain. It takes a signal that starts with the weather, but we are not a weather company. We are an “ag intelligence” company and we understand how a particular weather condition influences a specific crop during a certain growth stage. The idea is to provide information to the many segments of the agricultural value chain in a way that analysts can consume to make the best decisions. We have libraries of business decisions that relate to specific crop growth stages and very specific on-farm actions that the weather would drive.
Who are your customers, and how can smallholder farmers and other people in agriculture benefit from this information?
We have customers across the whole agricultural value chain. We have clients that work directly with the farmer providing a host of services. We help them with, for example, SMS services to inform and help adapt to different environmental issues. On the output side, we have clients who have to know the quality of what they are buying. For example, a client in Eastern Europe receives the output of foliar disease models that inform their harvest sampling strategy as they must avoid mixing healthy grain with mycotoxin infected grain (we predict conditions conducive for the development of the mycotoxin forming foliar disease). On the input side, there are customers who want to know which farmers need crop protection, which could benefit from additional fertilizers or simply how to manage where they store – and move – their crop protection inventory.
If, on a very high level, all these actors along the value chain were working with what we call ‘symmetrical information,’ then they are all seeing the same environmental signal. From our perspective, that is part of ag intelligence, as this information would contribute to optimising – or making more efficient decisions within – the agricultural value chain.
What do you mean by ‘symmetrical information’?
That term comes out of the insurance industry, I believe. Insurance companies have more information than the client, but they can’t leverage that unethically – everybody has to agree. This means, if you purchase a health insurance, it is because you believe that the price that you are charged is a fair balance against the cost and the risk.
Using this term in agriculture means that the researchers and input providers need to prioritise more effectively if they know what the farmers are experiencing. As a consequence, the whole value chain would optimise. The signal of agricultural intelligence is quite imperative to make agriculture more efficient.
How do you include technology and digital solutions in your business model?
On the side of the farmer, the obvious piece is data science and ICT with a phone. Smallholder farmers are typically very cash-flow restricted. If the weather conditions when they plant are particularly good, and if they match with good soil moisture and maybe even a little bit of rain in the forecast, the obvious advice would be to fertilise. You may also tell them to fertilise if there is a significant drought forming in a neighbouring market area, as the yields of those farmers in the drought would suffer, and prices would go up.
We are doing some things with the World Bank right now, in terms of implementation and capacity-building for local agricultural observatories. Ten days ago, we did some enthusiastically-received training with about 40 people at KALRO, the Kenyan Agricultural and Livestock Research Organization. This project empowers analysts in the Ministry of Ag, and others, to leverage information for their purposes, whether on the plant breeding side or the extension side. Our goal is to deliver information to people who know how to interpret the information for the segment that they are focused on along the ag value chain.
I can find your data on apps.awhere.com?
Our information is available in two different ways. You can subscribe to an API, which allows you to pull individual locations for which you can use your own tools, with or without our capacity-development support. You can also subscribe to a geospatial product , which provides everyday updates for the entire planet. Within this option, you can buy whatever geography you are interested in and we provide the geospatial content at a 9-kilometre resolution, which allows you to identify anomalies in ag weather across the whole of a production region.
What you get from us is georeferenced and interpreted ag intelligence in formats that can be utilized by all sorts of analytical platforms. Our data is interoperable with everything from Python and ‘R’ to Excel and statistical programs like SAS and SPSS. Everything we do spatially goes into any of the GIS mapping tools.
How far can you look into the future?
Well, nobody can look into the future. That is one of the big mysteries. The problem with weather variability is that there is no mechanism – no history – by which you can predict further out into the future (even 30 days is not yet available, though some provide 15 days of daily forecast). There are some really good scientists who are working on seasonal forecasts, but the world is still waiting for someone to crack the nut of forecasts that are actionable by a farmer.
Which skills do farmers and people along the value chain need in order to make use of new digital technologies for agriculture?
I raise the surrender flag on what farmers should do. In my opinion, they need to be provided with trustworthy information that is accurate and helps them to make decisions. If you are working in the agricultural value chain, you need to appreciate more quantitative metrics to guide your decisions. Doing things the way you have always done them is a terrible plan in today’s world. You can’t just stick to what your grandfather and mother did, because the atmosphere is warmer and the ag-meteorology system is more variable and different. People need to realise that they have to be more flexible today.
In addition, we need trust to work together. Here in the United States, there are climate change deniers. In the middle of what we call our Corn Belt, people are not seeing that weather variability is impacting them. They think that it is not a big deal, but as soon as you get to the geographic edges of whatever you are producing, the impact of being too wet or too dry will be dramatic. Ag meteorology – digital agricultural advisories – is doing a good job with producing tailored recommendations, but people also have to be willing to listen.
Which technologies will be most relevant and disruptive within the next 2-5 years?
In my opinion that will be the ICT-enabled solutions that get information to the smallholder farmers to close the last mile. There will also be these Uber-like technologies that allow a farmer to rent a tractor or a harvester for one or two days to do things more efficiently. Within the next five years, we will also see private companies and investors who are increasingly looking at analytics. They already do, but innovative analytics are emerging from the agricultural sciences that will help them guide their investment decisions. 3D printing, in my opinion, does for production what solar power is already doing for energy – it is decentralising the source for a lot of goods. I can see 3D manufacturing becoming really good for all sorts of farm and household implements that you would no longer need to import from afar.
There is also a lot of talk about the digital divide. Are we going to able to overcome it?
If you define the digital divide by access to the internet, yes we are going to overcome it and it has huge ramifications for humanity. Cellular networks are getting better and better across Africa, and in some cases are even better than in rural America. My guess is that at some point, building out these networks will become something that is paid for because it is the ‘tide’ that lifts all boats – paid for by government through economic incentives to private firms.
Access to digital technology saves you money at some point. If people have access, for example, their health can be better and that saves huge costs, especially from a government perspective.
In order to bridge the digital divide, agricultural intelligence needs to be made obvious so that people can consume it. Consuming it allows them to be more educated and to ask better questions. One of the best stories from our training is when we were showing people how to use our API (to get a localized hourly forecast) via a freely available and open source platform ‘R’ . They were just copying a code, which is quite straightforward. One woman entered the latitude and longitude of her shamba – or farm – and got the forecast for the first time in her life. She cheerfully exclaimed: ‘This is great, I need to email this to my husband!’ I didn’t tell her what to do, but she already knew exactly what to do when she got it. When we enable people to have access to something, the first thing they do is innovate its use.
Thank you for these interesting insights!