Betting the Farm: Big Data in Agriculture

0
47

Farming can be a gamble. There are many variables that can determine success and some of them, like climate, are outside a farmer’s control. But an innovative approach to agricultural research may help tip the odds in the favor of smallholder farmers. Strategic foresight modeling is a research method that examines how farming systems interact with climate change and other factors. The resulting data can help predict outcomes on farmsand help make agriculture less a game of chance for farmers.

Dr. Steven Prager is at the forefront of this research as a Senior Scientist for Integrated Modeling at the International Center for Tropical Agriculture (CIAT). A research center within the CGIAR network, CIAT works to address hunger, poverty, and nutrition through increased eco-efficiency in farming systems. Prager is a member of the Decision and Policy Analysis (DAPA) Research Area at CIAT as well as the co-leader of the Global Futures and Strategic Foresight initiative for CGIAR. His research is interdisciplinary and examines the interface of climate change and agriculture with economic development. “I am a geographer by training,” says Prager, “and the DAPA modeling team is a group of talented folks that includes economists, statisticians, climate modelers, crop modelers, crop physiologists, and agronomists.”

The modeling team develops strategic foresight scenarios by using a series of integrated models that consider the impact of climate, water resources, and technology on agriculture. The models apply these factors to future scenarios and make detailed predictions about their outcomes. “Generally speaking, it is about trying to understand present conditions, future conditions, and what would happen if certain assumptions were to hold true,” according to Prager. These assumptions can include rates of adoption of technology, positive results from the application of technology, and climate shock, the tangible consequences of climate change.

“The whole idea behind foresight is to understand what we refer to as plausible scenarios,” says Prager. “We use mathematical results as a way to evaluate these different plausible scenarios and improve decision-making regarding investments in the agricultural development process.” Ultimately, these results can help reduce risk for farmers and improve their likelihood of success. These results can also help inform other research sectors with social implications, like policy and investment.

Foresight modeling produces high volumes of new data that could be relevant to research outside of the agriculture sector and CGIAR. The CGIAR Platform for Big Data in Agriculture, an initiative that aims to capture and analyze large data sets to identify trends that can inform farmers and decisionmakers, could improve the impact of DAPA’s work by providing a clear channel for sharing information. “Basically, the Platform could improve the impact associated with the results of research,” says Prager.

At present, there is no central platform for data discovery and methods discovery, but the Platform may be a solution. “I think the platform does have the potential to increase coordination across the CGIAR network, especially as it relates to shared methods and shared data,” says Prager. “Often times we have great ideas and someone’s already done something similar. We could accelerate the research process and more importantly, accelerate impact if we are clever about how we share data.”

There are important considerations and obstacles to enabling the tools of Big Data for agricultural development. One is privacy and the protection of sensitive data. Prager currently chairs the Institutional Review Board, which ensures compliance with the ethics of human subjects research at CIAT. Beyond foresight, Prager sees the Big Data Platform as an integral tool for “managing and sharing potentially sensitive data in the most appropriate way, and in a way that is most consistent with our commitment to our various stakeholders”. When conducting research that involves human subjects, researchers must consider the implications of technology, confidentiality, consent, and more. The Platform is committed to responsible data sharing, and CGIAR is nearing completion of a set of guidelines based on the privacy and ethics standards of the individual partner centers.

Coordination across these centers, on privacy standards and otherwise, presents a separate set of obstacles. CGIAR is a network of 15 Research Centers, and each Center is a unique organization with complex and differing approaches to agricultural research. “Coordinating all these perspectives and encouraging others to adopt the philosophy and participate in the movement while doing something with visible impact is a challenge,” says Prager. Building a cooperative network of this size takes time and patience but the mission at stake—to democratize agricultural data and solve major food system problems—can be unifying.

One goal of the Platform for Big Data in Agriculture is to convene partners, including the private sector that produces new innovations to solve development problems. “Engagement in that sector is going to be a big milestone and a big win for CGIAR if the Platform for Big Data in Agriculture is successful,” says Prager. IBM and Amazon are two of the external partners working with the Platform and collaboration with the tech giants may help research data hit the ground as functional projects and products for consumers.

Encouraging participation and focusing on visible impact are key to the Platform’s success. “We need to understand where the intersections are and the real places where an organization like CGIAR can make the difference,” says Prager. “The next set of problems is making food systems more sustainable, making sure more people have access to healthy food, and including marginalized communities in the food system.”

The post Betting the Farm: Big Data in Agriculture appeared first on Food Tank.

Source: Agric Investment Opportunities

LEAVE A REPLY

Please enter your comment!
Please enter your name here