New video gives a glimpse of CIAT's work on big data

It follows the story of Blanca, a 75-year-old Colombian rice grower in the country’s Tolima department. She’s seen the climate change over the years, and how it’s affected her crop.

CIAT has been working with the country’s rice growers’ association, Fedearroz, to provide advice to farmers like Blanca, and help them respond to both climate change and seasonal variations in the weather.

Using award-winning data mining techniques to analyse huge quantities of crop and climate data, they have generated new insights that can be quickly translated into action on the ground.

These include:

• characterisation of the most successful, site-specific management practices to allow farmers to achieve the best yields under prevailing conditions.
• characterisation of the different weather patterns that occur in their area, their frequency, and sets of management practices so that farmers can continue to produce crops and make money – even in adverse conditions. These are combined with seasonal forecasts to anticipate upcoming conditions and release appropriate recommendations.

These new approaches follow successful work in 2014 with 170 rice farmers in the country’s Caribbean department of Córdoba.

By analyzing crop and climate data from Colombia’s National Institute of Hydrology, Meteorology and Environmental Studies (IDEAM, by it’s Spanish acronym), and a range of other sources, CIAT scientists predicted the likelihood of failed rains that could jeopardise a forthcoming planting season. They shared the insights with Fedearroz, who advised the farmers not to plant.

The analysis was correct; the rains didn’t come. Had farmers sown at the usual time, they would have faced potentially catastrophic financial losses on seeds, fertiliser, labour and other inputs. Instead, they skipped a season, producing a good harvest a few months later.

“Central to the success of this work was the free, open access to large amounts of weather data from IDEAM, and the willingness of Fedearroz to engage with data mining to extract value from years of crop data that had not been used much until then,” said Sylvain Delerce, a CIAT data expert.

“We were able to spot the main limiting factors of crop productivity at a site-specific scale, combine it with weather predictions, and get the information to Fedearroz in good time.”

It’s one of the strongest examples yet of the work of CIAT and its partners to use big data to bring quick, measurable impacts to farmers.

“Ultimately, there is limited value in raw data,” continued Delerce. “It needs to be processed, analysed and combined with other sources of information in order to unlock its true power. When organisations understand this, data starts to flow.

“It is about changing the mindset.”

Similar insights are anticipated for a wide range of crops and regions, following the launch of the CGIAR Platform for Big Data in Agriculture, which counts GODAN as a partner.

Jointly led by CIAT and the International Food Policy Research Institute (IFPRI), and involving all 15 CGIAR centers, the new Platform represents the largest concerted effort to bring together data experts, scientific organisations and governments, to share and process information to help strengthen agriculture.

“This kind of work is set to explode – not just in Colombia and not just for rice, but worldwide and for numerous vital crops,” said CIAT’s Andy Jarvis, a self-described “big data evangelist”, who is closely involved in setting up the six-year, initiative.

“We’re talking literally thousands of analysts – ‘geeks’ as I prefer to call them – from across the pubic and private sectors, crunching numbers, searching for trends and providing new insights into ways to make smallholder agriculture more sustainable, more productive and more responsive to challenges like climate change.

“It’s a huge opportunity to bring the data revolution to millions of smallholder farmers around the world.”

*