Data Tech Aids Quest to Save Topsoil

Scientists turn to cloud computing, artificial intelligence and real-time data technologies to better understand and manage soil erosion because the future of farming depends on it.

By Jason Lopez

By Jason Lopez March 10, 2022

Humans and climate change are depleting the earth’s topsoil at an alarming rate. Scientists say this is not speculation – it’s a measurable fact. If the world continues with current agricultural techniques, farms of the world will lose their topsoil due to erosion in the next 50 to 100 years. With humanity on track to grow to 9 billion people by 2050, the agricultural output needs to nearly double to feed everyone. 

On top of this, soils are being depleted of the nutrition that plants need. Conventional models of fertilizing are a temporary fix for unhealthy soils and are unsustainable. In addition, when soil is eroded it loses the structure that makes it resilient to extreme weather events such as drought and floods. 

But there is a revolution underway in the scientific community, armed with advances in information technologies such as sensors, artificial intelligence and cloud computing. These data technologies have become necessary tools to understand what makes soil healthy and how to sustainably keep it that way.


Studying the Microbiome of Topsoil to Sustain Farming’s Future

Just as medical researchers have discovered the profound influence on our health of the bacteria in our gut, some calling it our second brain, soil microbiome researchers are gaining new profound insights into the makeup of healthy soil, where bacteria – billions in a teaspoon of soil – control vital processes for plant health.

“We're very interested in the soil microbiome,” said Ryan McClure, a microbiome computational scientist at the Pacific Northwest National Laboratory in Richland, WA.

“That's the collection of all these microbial species that reside in soil because it has a lot of really critical ecosystem functions like cycling of carbon, nitrogen and the promotion of plant growth.”

Humans have planted crops in for thousands of years is something far more than just dirt. Some ancient groups such as the Maya, Zuni or Maori through centuries of trial and error had an inkling there was more to soil than meets the eye and practiced forms of sustainable farming.

But most of the human agricultural activity has relied on primitive techniques still in use today such as clearing and plowing, as well as today’s chemical-based monoculture. Recently science has led the way to precision agriculture with tools like IoT, robotics, AI, cloud and remote sensing technologies to begin a shift toward global sustainable food production.


Tackling Climate Change Challenges with Cloud Computing

Soil microbiome science takes such precision to unprecedented levels with research that couldn’t be more relevant to growing food. It wasn’t long ago science discovered that the role of bacteria and fungus in plant health was far more critical than previously thought. They are organisms that help plants absorb nutrients and water from the soil. McClure finds the investigation is exciting. He said that soil is more than a mere medium to which we add inputs like fertilizer which plants eat. 

“Direct analysis of the native and natural soil in its site is difficult because it's so complex,” said McClure in a Tech Barometer podcast interview (below). “It's hard to gather data. It's hard to interpret that data.” 

His team took soil into the lab and analyzed it in a liquid medium. The purpose was not necessarily to discover which microorganisms were present but how they interacted with each other. This was accomplished by subjecting the soil samples to 66 different cultivation conditions. Using a variety of nutrients, stress conditions, oxygen levels, antibiotics, and many other inputs, McClure witnessed different functions of various microbial species as they emerged.  

“Now we're in a position to look much more deeply at who's in them, what are they doing and how do they relate to each other so that we can better understand how do each of these 66 unique individual parts combine to give you the soil microbiome as a whole.”

Performing this kind of research requires the capture and analysis of large amounts of data. 

“We collected metatranscriptomic data, which tells you what genes are being expressed. And then we use network inference tools that are based on context, the likelihood of relatedness – so, essentially how related are two things in the context of the community as a whole.” 

“Understanding soil microbiology leads directly to better growth of agricultural products,” he said. These are the first steps into a world as mysterious as the brain or black holes. And although McClure indicates much more research is needed – he said every question generates two more – one practical result of this inquiry could be more effective use of bacteria applied to soil. 


Can Technology Rescue Our Forests Before It’s Too Late?

Bacterial products are already on the market for enhancing the health of soil much the way pro-biotic yogurt is available at the supermarket which populates our intestines with beneficial bacteria. Also, the application of off-the-shelf bacteria to soil isn’t an exact science. McClure’s current research could lead to highly specific prescriptions that repair soil with known communities of microorganisms. 

“I think it's something that could be improved and refined with some of the work that the Pacific Northwest National Lab is doing,” he said. 

As McClure delves into the minute world of the soil microbiome, Bradley Miller, a scientist at Iowa State University who leads the Geospatial Laboratory for Soil Informatics, uses geospatial technologies to literally examine the topsoil of large area of land from 20,000 feet off the ground. 

“Technology is opening up doors for us,” said Miller in a Tech Barometer podcast interview (below). “We’re able to get more data, see more, analyze more.”


Tackling Topsoil Erosion with Data Tech

His specialty is geomorphology, the study of how landscapes evolve. 

“By understanding geomorphic processes we're able to develop models where we can start asking questions about how the soil-landscape is going to change in the future,” said Miller. 

And it is changing fast. The world’s topsoil is eroding away annually at a rate of 13.5 tons per hectare. Picture it as six full-size pickup trucks hauling off the topsoil each year from a piece of farmland about the size of an international rugby field. 

Iowa possesses some of the world’s most productive soil. It, too, is degrading at the global average rate, with some spots losing as much as 55 tons per hectare. But Iowa is under the watchful eye of soil scientists eager to unlock sustainable ways of slowing erosion.  

There are multiple ways to measure soil loss: physically comparing sample cores between higher and lower points on a field, modeling erosion using a methodology developed by the Water Erosion Prediction Project, measuring sediment load in rivers or satellite tracking. 

But Miller’s team is crafting a method that could provide an even more detailed and comprehensive picture of topsoil movement. Using LIDAR (Light Detection And Ranging), scientists hope to detect soil movement of, ostensibly, any point in Iowa. It becomes a benchmark to compare to current models, significantly improving accuracy. If the process he’s developing is effective, it could be a major tool in our ability to understand the impacts of management choices. 

LIDAR is the same technology used in autonomous cars to detect objects. When used in airplanes to map topography what’s revealed is stunningly detailed elevation data. Essentially, Miller can measure changes in elevation with unprecedented detail and test soil movement models for any point in the state. 


Drone with Lidar Scans Archaeological Site High in Peruvian Andes

The state was previously mapped with LIDAR a decade ago, making an assessment of change possible. 

“We’ll be able to check the hillslope erosion models, then try to improve them so that we can actually have confidence in predicting a real future,” he said. 

“If we understand the system properly, we should be able to use appropriate models to really predict how the landscape's going to change anywhere in the world.” 

Miller said there’s a staggering amount of data to process, not just from LIDAR but from models making predictions about different landscape changes over time. His Geospatial Lab uses a variety of technologies such as GIS to analyze spatial data and AI for pattern recognition, operating on an on-prem data center to run the models with cloud backup.


Meet Rick Vanover, the Batman of Backup

The world of agricultural information technologies is brimming with innovations in plant genetics, indoor cultivation, robotic field maintenance like weed control, IoT leaf monitoring, precision inputs of fertilizer and pesticides, soil microbiome discoveries and erosion research, which are some examples from a long list. Researchers, farmers, technologists, ag companies and venture capitalists are acutely aware of what’s at stake to find solutions to the existential problem of soil degradation. 

In their respective labs, McClure and Miller are making vital incremental contributions to the solutions needed to make agriculture more sustainable and effectively feed the world. 

“In science, we frequently have this idea that you hold a candle up in a dark room and that's the world you see,” said Miller. “But as you increase the brightness of that candle the world gets bigger. And it's kind of a similar story for a lot of things that we're exploring, whether it be black holes or the human brain or even soil. As we're getting new tools, we have new ways of looking at it and digging – pun intended – deeper into it.” 

Jason Lopez is executive producer of Tech Barometer, the podcast outlet for The Forecast. He’s the founder of Connected Social Media. Previously, he was executive producer at PodTech and a reporter at NPR.

© 2022 Nutanix, Inc. All rights reserved. For additional legal information, please go here.