عمومی | Nature News & Comment

Big-data project aims to transform farming in world's poorest countries

A coalition including the Food and Agriculture Organization of the United Nations (FAO), the Bill and Melinda Gates Foundation and national governments has launched a US$500-million effort to help developing countries gather data on small-scale farmers to help fight hunger and and promote rural development.

The programme, which will run through 2030, aims to fill a yawning information gap about what an estimated 500 million farmers living in poverty are doing across 50 countries in Africa, Asia and Latin America. It will do so by expanding the reach of surveys developed separately by the FAO and the World bank gather information on factors such as livestock holdings and crop yields. Those surveys have been piloted in eight countries over the past decade.

Coalition members said on 24 September that they plan to spend $36 million over three years, starting in 2019. They aim to spend an additional $200 million through 2030, with the remainder of their $500-million goal coming from participating governments and in-kind contributions.

Gathering more accurate information about seed varieties, income and the farmers' technological capacity could help coalition members track whether their ongoing agriculture investments are making a difference. They also hope the data will enable governments to tailor policies to help their farmers.

Small farms, big push

The programme could help nations meet the UN Sustainable Development Goals, which are a set of targets adopted in 2015 to reduce poverty and improve lives among the world’s poorest countries. One of those goals is to double the agricultural productivity and incomes of small-scale farmers as part of a broader effort to eliminate hunger by 2030.

The question facing governments as well as philanthropists, development banks and funding agencies such as the US Agency for International Development (USAID) is how to monitor progress.

“We realized we really do not have a way to track whether we are achieving that goal,” says Emily Hogue, a sociologist at FAO who is working on the project while on a temporary assignment from USAID. “And if we can’t track progress, it’s going to be very hard to achieve it.”

Much of the data on rural agriculture around the world come from the FAO, which relies on reporting from individual nations that is often incomplete. Even in countries that collect detailed agricultural information during censuses, the data are often years or decades out of date. And many governments don’t have the money to repeat those larger surveys regularly.

The surveys developed by the FAO and the World Bank are more focused, cheaper and faster, allowing countries to gather data before and after harvests each year. As part of the latest initiative, dubbed 50 x 2030, governments will receive technical support on implementing the surveys and using the data that workers gather. In return, governments must commit to gradually increasing their own funding for the programme, with the aim of taking complete ownership within a decade.

“What’s exciting is that governments and donors are making big commitments for the long-term,” says Claire Melamed, who heads the Global Partnership for Sustainable Development Data in London, a network that includes government agencies, academic institutions and companies. “Data becomes more valuable the longer you invest in it.”

Better data

One of the main test sites is Uganda, which began using the World Bank survey in 2009. Those survey results showed that small farmers were having trouble accessing veterinary services for their livestock. Now, the Ugandan government is redesigning its programme for supporting rural farmers so that it's easier for people to get medical care for their animals.

Uganda will also begin using the FAO’s survey ― the second country to do so ― this year. Several hundred workers are currently being trained to conduct the surveys, and they will head into the field in October to talk to farmers and gather data about the current planting season. They will then follow up with more surveys after the first harvest; the same process will take place for the next planting season, which begins in April.

The end result, Hogue says, will be a statistically representative database with precise geographic information about rural farms that can be used to track trends and adjust government outreach efforts accordingly. The survey will target 7,000 farms and cost around $850,000 annually. By comparison, Uganda’s next full agricultural census will survey 48,000 farmers, cost $11 million and is planned for 2020 — 12 years after the last one.

“This is really changing the face of how we use data for agriculture,” Hogue says.