Big Data’s Big Role in Agriculture


September 03, 2014

The world’s population is increasing. We all know that. We also know that feeding that population is going to require implementing a number of strategies. Some of those strategies begin in the lab, are implemented in the field, affect harvested crops during transportation and sale, and change consumers’ eating habits. Some of those strategies involve creating heartier plants and animals, increasing crop yields, and ensuring safe food supply chains. In each of those strategies, big data analytics is going to play a significant role. An article published by Troy Media, states, “In the rapidly-expanding world of DNA sequencing, throw out everything you think you know about ‘big data’ out the window. Scientists are hoping to increase the world’s food supply by mining the ultimate data source: the genome.” [“How analytics can feed the world,” Troy Media, 6 March 2013] The article admits, “It’s a big job. … Calling it ‘Big Data’ doesn’t begin to do justice to the process.” Despite the fact that sequencing genomes is a big job, the article goes on to note that the time and money required to sequence specific genomes has decreased dramatically. Dr. Glen Plastow, a professor at the University of Alberta, who was interviewed for the article, stated that “the sequencing, amazingly, was becoming the easy part.” Plastow stated:

“Making sense of the data takes a lot of analytical and computing power. Converting this data into information, and understanding what it really means, we will be able to make faster genetic improvements to improve food production [and] reduce the time to market.”

The article continues, “Getting more from less is what it’s all about.” John Schlageck writes, “Some believe ‘big data’ may be the next renaissance in agriculture. Others call it the greatest advance in agriculture since the Green Revolution during the 1940s, ’50s and ’60s when one of the biggest waves of research and technology spurred the growth of agricultural production around the world. Some compare big data with the biotech revolution.” [“The ‘Big data’ deal,” Leader & Times, 22 July 3014] Schlageck admits there are concerns associated with gathering and using big data; but, its use seems inevitable. He notes that data will be gathered in a number of new ways:

“Drones flying above farm land recording high resolution images, and field sensors providing immediate information concerning crop conditions including moisture, nutrients, pests, etc., may become commonplace during the big ag-data era. No matter what beatitudes are bestowed on big data, most believe and hope it will improve farmers’ yields and productivity. Some say it will help feed the growing population expected to hit 9 billion in 2050. Agri-business companies are banking on its future. … Prescriptive planting or relating soil, climate and seed data with a farmer’s productions records seems to be some of the potential of big data in agriculture. The potential for an increase in grain yields is another potential.”

A report from Iowa television station WHO states, “Data science is a new driving force in agriculture.” [“Big Data Can Help With Farming Decisions,” 4 August 2014] The report goes on to note, “According to [the] USDA, U.S. farmers spent more than $367 billion on agricultural production in 2013. Matt Darr with Iowa State University says about two-thirds of every dollar spent on agriculture is focused on decisions of seed selection, fertility, and land access. Mass amounts of data could have an influence on those decisions.” The goal is to let big data analytics help farmers make better decisions to increase yields as well as make production more efficient. Getting food safely from the farm to family is another area in which big data is playing an increasingly significant role. Mary Shacklett (@MaryShacklett), Chief Executive Officer of Transworld Data, discusses the importance of a safe food supply chain. “In response to growing public safety concerns about food safety,” she writes, “the FDA’s Food Safety Modernization Act (FSMA) was signed into law on January 4, 2011. In its first phases, the FSMA focused on regulating food sources and the food supply chain; in 2014, its focus is broadening to the transport of food from producer to consumer.” [“Big data’s vital role in solving urgent food safety problems,” TechRepublic, 23 June 2014] She continues:

“What makes the FSMA interesting as a big data initiative is its acute sense of urgency. There are deadlines that food producers and transportation companies must meet, and a US public that sees one in six Americans (or 48 million people) get sick each year from eating the food that they buy. Of these victims, 128,000 are hospitalized, and 3,000 die annually, according to 2011 data from the Centers for Disease Control and Prevention. To improve this situation and to ensure that they don’t fall under the microscope of government auditors, food producers and transporters are turning to sensor-based technologies and analytics that will inject big data into their supply chains and provide them with visibility that can be as granular as pegging a contaminated shipment to a particular farmer’s field. Equally important is ‘track and trace’ monitoring and visibility of food shipments during transport. This track and trace monitoring will use sensor- and RFID-based technologies that can follow food shipments from their points of origin through the logistics network and into warehouses, distribution centers, and retail outlets. If the foods are ‘cold chain’ goods that require refrigeration or other types of strict environmental controls, such as maintenance of the goods in specific humidity ranges during transport, sensors for temperature and humidity will be expected in transport vehicles and in the containers that they carry. The sensor-based technology that controls the quality and safety of food in transport containers works in two ways: It can provide GPS data that delivers real-time information about the location of the transporter carrying the goods, and it can monitor the temperature and humidity parameters within food containers, immediately sending out automated alerts if a particular container’s environmentals begin to fail.”

Anyone who has ever suffered from food poisoning will appreciate efforts being made to ensure that the food we eat is safe. Shacklett concludes:

“With continuing food safety problems in the US, public and governmental pressures will stay focused on companies to move to sensor-based technologies and big data monitoring and analytics that can collect machine-generated data for shipment track and tracing, and also for continuously in-transit monitoring of produce and other perishable foods as they travel from producers to consumers.”

Perhaps the biggest worry shared by those concerned with global food security is how climate change is going to affect the global food supply in the decades ahead. Big data is poised to help find solutions to this conundrum as well. Mark Kinver (@Mark_Kinver) reports, “Researchers are developing mathematical models to identify genetic material that could help improve food crops’ resilience to climate change. Impacts – such as drought, pest and disease – could hit harvests and undermine global food security. Scientists hope the models will speed up the process of identifying traits, such as drought resistance, allowing breeders to grow climate-proof crops.” [“Mathematics helps find food crops’ climate-proof genes,” BBC News, 15 August 2014] He continues:

“Globally, there are 1,700 major agricultural genebanks that house in excess of seven million samples – a vast resource that researchers say makes the task of locating the sought-after traits a bit like finding a needle in a haystack. [Dr. Abdallah Bari, a senior scientist at Syria-based International Center for Agricultural Research in the Dry Area (Icarda)] said that developing mathematical models would help focus the search by ‘targeting the [samples] with a high probability of finding those traits and reducing the time it takes’. He explained that the Icarda team were developing a technique that used a ‘learning algorithm’ to harvest the necessary data that would allow plant breeders to ‘zone in on the desired traits, such as tolerance to pests, diseases, drought and heat’. Without a model, plant breeders would have to rely on the traditional and time-consuming ‘trial-and-error’ approach, which requires plants to be cross-bred and the progeny being exposed to the conditions they would be expected to encounter in fields during extreme weather events. Those that display an improved ability to cope with harsh conditions are kept as seed stock, while those without the ability to cope with the conditions often perish or are not used as a seed stock and the plant breeders have to start the process again.”

Although 35 years may sound like a long time to be able to figure out how to feed the world, it isn’t. Scientists understand the urgency of the situation and are looking for all methods that will help them save time and resources in the search for better food security. Big data has a significant role to play in those efforts.