• Subscribe

The future of farming

What are we going to eat? It’s a question we can all relate to but, with a global population expected to reach 8.5 billion by 2030 and temperatures rising worldwide due to climate change, one that is increasingly difficult to answer.

<strong>Carbon exchange.</strong> David Potere of Indigo Ag talks about the Terraton Initiative to remove carbon dioxide from the atmosphere and use it to enrich agricultural soils. At the Center for Digital Agriculture Conference at NCSA. Courtesy Darrell Hoemann.Though agriculture has been slower than other sectors to adopt digital technology, it may yet be the solution we need. Science Node recently attended the Center for Digital Agriculture Conference at the National Center for Supercomputing Applications (NCSA) to find out more about some of the most promising technologies that will keep fresh, nutritious food flowing to our collective table.

Model plants

One way to increase food production is to design enhanced crops. Various genetic traits can increase yield, improve water efficiency, or help a plant withstand heat. Growing more crops with those traits could raise food outputs around the world.

<strong>Strong roots.</strong> Computer simulation of root systems of the traditional ‘three sisters’: maize (green), bean (blue), and squash (red), demonstrating nitrate capture. Courtesy Zhu, et al.But identifying and improving specific plant traits involves complex interactions between genetics, environment, and the local ecosystem. To do this, biologists construct mathematical models of the parts of the plant they are interested in, such as root growth, molecular transport, or response to temperature or CO2.

Because biologists construct their models to answer specific questions, the models aren’t easily adapted for reuse. Crops in silico hopes to change that by working to connect thousands of existing models in order to build virtual crops, says Amy Marshall-Colon, a University of Illinois (U of I) biologist. 

This is a global scientific collaboration that gathers data from domain experts and incorporates it into Yggdrasil, a computational framework that can connect models written in different languages. By modeling whole crops researchers can capture more complex biological processes than is possible with isolated models.

Field test robot

Once biologists model and eventually create new varieties of crop plants they have to test them in the field to find out which ones have the best real-world performance. To do this, growers measure and compare various plant characteristics.

The Terrasentia robot rolls beneath the leaf canopy to collect data on crops via autonomous navigation and machine-based vision analytics. Courtesy EarthSense, Inc.

This creates a bottleneck in the production of new varieties, since most evaluation is done manually by humans. Imagine walking through row after row of corn and stretching a tape measure from the base of each plant to its tip—that’s pretty much how it’s done.

Drones can help with some assessments, but they can’t see under the plant canopy. Which is why Girish Chowdhary, of the Distributed Autonomous Systems Lab (DASLAB) at U of I, has invented the Terrasentia robot.

The ultra-compact Terrasentia is only a foot wide and can easily navigate between crop rows. Equipped with a variety of sensors, including cameras and LiDAR, the robot’s detect-and-track algorithm can assess objects as it rolls past. For example, it can estimate the width of plant stems by correlating pixel distance with lidar distance.

Machine learning analytics then convert collected data into actionable information about plant traits. And, unlike a drone’s flight which only lasts 15 minutes, Terrasentia can work for 6 hours.

Boosting trust to increase yields

Farms in North America and Europe can encompass thousands of acres cultivated on an industrial scale, but in many parts of the world, farms are operated by smallholders with very small field sizes.

<strong>If very small fields</strong> in Nepal were aggregated, crop yields could increase by up to as much as one-third. A pilot program is testing if Blockchain is the answer to fair and transparent sharing. Courtesy ActionAid.In Nepal, generations of land-splitting have resulted in a fragmented agricultural landscape. In addition to small fields, farmers are often of advanced age, and there is little access to equipment or technology. In such areas, the crop yield is one-third less than what could be derived from a field of the same size in the US, says Adam Davis, U of I crop sciences professor.

One solution could be to aggregate fields and manage them with modern farming practices. But farmers may not have sufficient faith in their neighbors to trust that their own crop share will be fairly allotted back to them at harvest time.

That’s why Davis is piloting a program in Nepal and Mozambique that will test blockchain as a way of managing field aggregations and transparently tracking output shares. Many questions remain, but if a digital ledger can reassure farmers that aggregation is safe, it could dramatically increase food production in areas most in need.

Not just more food, but safer food

Food passes through many hands—from grower to trader to processor to packager to retailer—before reaching the consumer’s plate. At any point along that chain, mistakes can be made and food can become contaminated. 

<strong>Food passes through many hands</strong> (such as this fish-processing plant) before reaching the consumer—which means many opportunities for contamination. Safefood.ai scans multiple sources to detect and predict outbreaks.  Courtesy Jabbi. <a href='https://creativecommons.org/licenses/by-sa/3.0/deed.en'>(CC BY-SA 3.0)</a>In today’s global supply chain, tainted food can quickly sicken many people in widespread areas. To prevent that, process engineering company Bühler has developed a digital platform that detects and even predicts contaminated food outbreaks before they happen. 

Safefood.ai scans regulatory websites, news reports, and even social networks for information about unsafe food. Custom reports are then delivered to a user’s dashboard warning about threats to specific products or ingredients.

But food safety is a major challenge and so, Bühler, in partnership with Alibaba Cloud, recently launched an AI challenge to use training data from safefood.ai to create algorithms that can discover incidents with high relevancy and great precision. Because no matter how much food we have, it won’t be much use unless it’s safe.

Read more:

Join the conversation

Do you have story ideas or something to contribute? Let us know!

Copyright © 2019 Science Node ™  |  Privacy Notice  |  Sitemap

Disclaimer: While Science Node ™ does its best to provide complete and up-to-date information, it does not warrant that the information is error-free and disclaims all liability with respect to results from the use of the information.

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 ScienceNode.org — 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 ScienceNode.org” containing a link back to the original article.
  4. The easiest way to get the article on your site is to embed the code below.