Bringing science and development together through news and analysis

View on Poverty: Big data protects farmers’ livelihoods
  • View on Poverty: Big data protects farmers’ livelihoods

Copyright: G.M.B. Akash/Panos

Speed read

  • A model based on historical data warned Colombian rice farmers not to plant

  • It saved them from wasting US$3.8 million on seeds and chemicals in a drought

  • There are moves to expand the project to other crops and countries

Shares
“Don’t plant any crops” was the expert advice one group of Colombian farmers received from a research partnership last year.

Researchers from the International Center for Tropical Agriculture, working with Colombian Rice Growers Federation, had developed a computer model that weaves together ten years’ worth of agricultural data from specific farms across Colombia. It incorporates information on how individual farmers manage their land — what they planted and how they tended it — what the weather was like during the growing period and the yields each farmer achieved. The idea is for researchers to use the massive data set to identify what kind of agricultural practices have historically worked well in specific locations during particular spells of weather.

When the scientists first trialled the model, the best advice for one group of rice farmers was simply: the climate risks, including drought risk, are too high, so don’t plant.

Copy of P1030933
Scientists Daniel Jimenez (right) and Sylvain Delerce (left) present their work to a panel of experts convened by UN Global Pulse in New York City. Credit: Liora Yuklea.
Several months later, it proved sound advice. Those farmers who missed or ignored the workshops where the advice was imparted lost most of the crops they planted, as drought did indeed hit. But the 170 farmers who heeded the warning saved a collective US$3.8 million worth of seeds and other agricultural inputs.

Last month the project won the Big Data Climate Challenge run by the UN’s Global Pulse initiative, as The Guardian reported. [1]

I asked Andy Jarvis, one of the project’s lead researchers, what the prize will mean for the future of the model.

Jarvis explains that prizes like this are not necessarily about prestige — “We didn’t even get a certificate” — which usually comes from publishing in top journals. But this kind of prize can open up opportunities for collaborations and funding.

For example, he tells me that part of the prize saw two of the project’s lead scientists, Daniel Jiménez and Sylvain Delerce, meet Rajendra Pachauri, the chair of the Intergovernmental Panel on Climate Change, and Robert Orr, a UN assistant secretary-general, during the UN’s climate summit last month. “They sat with Orr for literally half an hour to go through what they’d done and why this was useful,” says Jarvis.

Although the impacts of this meeting are “probably quite intangible”, he says the visit and the prize more generally have “planted a little seed in a lot of high-level people’s brains” about opportunities for using the emerging field of big data to help advise on good agricultural practices.

Jarvis says the prize has also prompted interest in the project from countries far beyond Latin America. For example, Nigeria’s ministry of agriculture has emailed to congratulate him on the prize, and ask how such data science concepts could help their own farming.

It’s also garnered more interest from the Colombian government. So as well as applying the project in other countries, Jarvis says he hopes to expand it to crops such as maize and beans — and even livestock farming — in the coming years.
 
Joshua Howgego is SciDev.Net’s deputy news and opinions editor. @jdhowgego

References

[1] Diana Cariboni Colombia rice growers saved from ruin after being told not to plant their crop (The Guardian, 30 September 2014)
Republish
We encourage you to republish this article online and in print, it’s free under our creative commons attribution license, but please follow some simple guidelines:
  1. You have to credit our authors.
  2. You have to credit SciDev.Net — where possible include our logo with a link back to the original article.
  3. You can simply run the first few lines of the article and then add: “Read the full article on SciDev.Net” containing a link back to the original article.
  4. If you want to also take images published in this story you will need to confirm with the original source if you're licensed to use them.
  5. The easiest way to get the article on your site is to embed the code below.
For more information view our media page and republishing guidelines.