Artificial intelligence (AI) can help crop scientists collect and analyze volumes of data that would be impossible with traditional methods. Researchers at the University of Zurich have now used big data, machine learning and field observations in the university's experimental garden to show how plants respond to environmental changes.
Climate change makes it even more important to know how plants survive and thrive in a changing environment. Standard experiments in the laboratory show that plants accumulate pigments in response to environmental conditions. Until now, such measurements were made by sampling, which required the removal and damage of parts of the plant.
"This laborious method is not practical when thousands or millions of samples are needed. In addition, repeated sampling damages plants, which in turn makes plants react to environmental conditions. There was no suitable method for long-term monitoring of individual plants in an ecosystem," says Reiko Akiyama, the first author of the study. .
Supported by UZH University's flagship research program "Evolution in Action" (URPP), a team of researchers has developed a method that allows scientists to observe plants in the wild with high precision. PlantSerration is a method that uses robust image processing hardware and deep learning-based software to analyze field images and work in all weather conditions. The study was published in the journal Nature Communications .
With PlantServation, the researchers collected images (top view) of Arabidopsis plants during three field seasons (covering five months from autumn to spring) at UZH Irchel University and then analyzed more than four million images using machine learning.
The data indicate species-specific concentrations of plant pigment anthocyanins as a function of seasonal and annual temperature, light intensity and rainfall variation.
PlantSerration allowed scientists to experimentally reproduce what happens after the isolation of naturally hybridized polyploid species. These species arise by duplicating the entire genome of their ancestors, a common variation in plant species. Many wild and cultivated plants such as wheat and coffee originated in this way.
In the present study, the anthocyanin content of the hybrid polyploid species A. kamchatica was similar to the two ancestors; Winter in spring was similar to other species of the cold region.
"The results of the study therefore confirm that these hybrid polyploids share the environmental response of their ancestors, which supports the hypothesis of long-term polyploid evolution," said Ri Shimizu-Inatsugi, one of the two authors of the study.
PlantSerration was developed in the experimental garden of the University of Irchel, UZH.
"It was important for us to use Irchel University's garden to develop the plant monitoring hardware and software, but the application goes further; combined with solar energy, the devices can be used even in remote locations," says Kentaro Shimisu, corresponding author and URPP Evolution in Action.
"With its cost-effective, reliable hardware and open source software, PlantServation paves the way for many biodiversity studies that use artificial intelligence to study plants other than Arabidopsis, from wheat to wild plants that play a key role in the environment."
More information . DOI: 10.1038/s41467-023-41260-3
Citations: Experimental horticulture uses artificial intelligence to show how plants respond to environmental changes (September 22, 2023). Retrieved September 22, 2023, from https://phys.org/news/2023-09-experimental-garden-ai-environmental. .html:
This document is protected by copyright. No part may be reproduced without written permission, except for the good purposes of personal study or research. The content is provided for informational purposes only.
