From Space, AI Maps 20 Years of an Invasive Plant
leafy spurge
An invasive plant that can form dense patches and harm livestock.
deep-learning model
An AI system that learns image patterns from many examples.
Landsat
A long-running source of satellite images of Earth’s surface.
What happened
A University of Minnesota research team reconstructed how leafy spurge changed across Minnesota over two decades. They combined satellite images, a deep-learning model, and observations from community members. The observations included records posted to iNaturalist. The imagery came from open Landsat data supplied by the U.S. Geological Survey, with help from NASA satellites. The team studied 2000 through 2021 and predicted places where the plant could spread next.University of Minnesota research brief
Why leafy spurge matters
Leafy spurge was introduced from Europe in the late 19th century. It can form dense patches across acres of land. When damaged, it releases sap that is toxic to livestock. The University of Minnesota says the plant caused about $1 billion in economic losses for farmers and ranchers from 1960 to 2000. Finding new patches matters because control becomes harder when an invasion covers a larger area.
Why the method matters
Field surveys require people to visit many places repeatedly. That takes time and money. Satellite images offer a way to look across broad areas using a shared record. The AI model connected plant observations with patterns in the images. It could estimate where leafy spurge had appeared, even when no one had recorded every patch. The system does not remove the plants. It helps managers decide where field checks and control work deserve attention first.
What the study found
The model estimated a 570% increase in leafy spurge abundance from 2000 to 2021. The largest changes were concentrated around Fargo-Moorhead and Minneapolis-Saint Paul. The plant also appeared often near roads. That pattern suggests roadside management could help slow its spread. However, the study does not prove that roads caused the increase.
Drought created an important measurement problem. Leafy spurge produced fewer flowers during intense drought. With fewer bright flowers, satellites were less likely to detect it. A low satellite signal does not always mean the plant is absent. Adding newly invaded areas to the analysis improved forecasts of future invasion.
What remains uncertain
The map is a model-based reconstruction, not a perfect count of every plant. It shows where evidence points to an invasion and where future spread may be more likely. It does not guarantee what will happen in a particular field. The researchers still need to examine how climate change, insects, and rapid plant evolution affect the pattern. They also need to test the method beyond Minnesota and during conditions that make the plant harder to see.
What to watch next
The team plans to map leafy spurge across seven states. It also hopes to apply the approach to other invasive species. The key development is not that AI can see every plant from orbit. It is that long-term satellite records and public observations can be combined into a practical early-warning map. That map could help farmers, ranchers, and agencies spend limited time where it may matter most.
How AI Watches an Invasive Plant From Space
📰 Full story: From Space, AI Maps 20 Years of an Invasive Plant
Satellite pictures helped scientists follow a yellow-flowering weed for 20 years.
leafy spurge
An invasive plant that can spread across large areas.
deep-learning model
An AI system that learns patterns from many examples.
iNaturalist
An app where people record living things they see.
💡 The gist
- AI combined satellite pictures with public plant reports.
- Leafy spurge abundance rose by 570% from 2000 to 2021.
- The maps may help teams find new patches sooner.
What the team did
The University of Minnesota is a school that studies plants and nature. Its researchers studied leafy spurge. This plant came from Europe in the late 1800s. It can cover large areas. Its sap can harm livestock when the plant is damaged.
The team used satellite images from Landsat. The U.S. Geological Survey provides these open images. NASA satellites also supported the imagery. People added plant reports through iNaturalist. That app lets the public record living things.
Researchers fed these reports and images into a deep-learning model. This is an AI system that learns patterns from many examples. The model searched for signs of leafy spurge across many years. It also helped predict places where the plant may spread next.
Why the results matter
Checking every field by hand takes time. It also costs money. Satellites can watch broad areas in one shared record. The map can point people toward places that need closer checks.
The study found a 570% rise in abundance. The biggest changes occurred near Fargo-Moorhead. They also occurred near Minneapolis-Saint Paul. Leafy spurge often appeared near roads. Roadside checks may therefore help managers act earlier.
What the map cannot promise
The map is not a perfect plant count. It is a careful estimate based on reports and pictures. Roads may be linked with spread. The study does not prove roads caused it.
Drought made the plant harder to see. It produced fewer flowers then. Fewer flowers meant weaker signs in satellite pictures. A weak signal does not always mean no plants exist.
What comes next
Adding new invasion locations improved future forecasts. The researchers plan to study seven states. They will also examine climate change, insects, and plant evolution. They want to test this method on other invasive plants.
AI will not pull weeds from the ground. It can help people choose where to look first. Earlier searches may give land managers more useful choices. The original research brief is available from the University of Minnesota.
A Computer Looked for a Weed From High Above
📰 Full story: From Space, AI Maps 20 Years of an Invasive Plant
A computer helper watched a weed from high above Earth.
leafy spurge
A weed that came from another place and spreads.
satellite
A machine high above Earth that takes pictures.
AI
A computer helper that looks for patterns in pictures.
What happened
The University of Minnesota is a place where people study nature. Its scientists watched a weed called leafy spurge. This weed came from Europe. It can spread across big pieces of land. Its sticky juice can hurt farm animals.
The scientists used pictures taken by satellites. Satellites are machines that take pictures from high above Earth. They also used notes from people who saw the weed. A computer helper, called AI, studied the pictures and notes. It looked for places where the weed might be growing.
From 2000 to 2021, the amount of leafy spurge rose 570 percent. That means it became much more common. The weed often appeared near roads. People may check roads first.
When rain was scarce, the weed made fewer flowers. Fewer flowers made it harder for satellites to see. Hard to see does not mean gone.
AI cannot pull weeds from the ground. It can help people choose where to look. The team plans to check seven states next. The research note is here.