Media Contact: Carmen Ramos Chandler, carmen.chandler@csun.edu, (818) 677-2130
California State University, Northridge environmental biologist Jeremy Yoder has received more than $800,000 from the federal government to continue work on a software package he calls TARDIS. Similar to Doctor Who’s fantastical blue police box, the project will help researchers “travel through space and time” to understand the health of plant species.
Yoder developed TARDIS, which stands for “temporal analysis of reproduction distributed in space,” as a method to assist in his studies of Joshua tree populations across California, Nevada, Arizona and Utah. The analysis uses machine-learning models trained on data collected by the members of the public who capture photos of a particular plant and upload them onto the smartphone app iNaturalist to paint a picture of what is happening with a species across time and space.
“The grant allows us to continue to develop this new method for understanding how plant populations respond to variations in their environments, whether they are in warmer, wetter, cooler, drier settings,” said Yoder, who teaches in CSUN’s College of Science and Mathematics. “It’ll help us understand how population health changes in those different conditions. There are a lot of different features that go into the health of a plant population, but it’s basically down to our plants growing, reproducing and making the seeds that will grow the next generation of plants.”
Yoder said the $814,012 from the National Science Foundation includes money for undergraduate and graduate students to expand use of the TARDIS — which until this point has only been used to study Joshua trees and California’s iconic toyon plants — to work with data from any plant species.
“We have found that we have really good records of the flowering and fruit production from iNaturalist,” he said. “We can use those to train machine learning models that relate weather variation to reproduction activity. That gives us a look at how variation across the different places you find a plant species impacts its ability to reproduce. That tells us something new about the species’ biology. Hopefully, it tells us something about how the species might cope with changing conditions, too.”
The grant includes an undergraduate research experience that will engage CSUN students in reproducible data science practices and methods to find interesting patterns and test hypotheses. The students will use the skills they learn to the develop their own projects with the TARDIS methodology, focused on species of their own choosing.
“The goal is that at the end of the project, after we’ve had at least three iterations of the course, is to take all of the student projects and put them together into a big study of how plants respond to habitat variation,” Yoder said. “We’ll have a much more complete picture of how plants with different life histories respond to varying environments: long-lived trees versus short-lived weeds, wind-dispersed seeds versus seeds that just sort of drop from the plant — even species with different pollinator needs.”
The work will help hone the TARDIS’ ability to paint a more complete picture of a plant species’ history and the impact environmental changes could have on its future health, information that could prove invaluable to anyone who works with plants, from farmers to conservationists.
“We hope to make TARDIS user friendly and more useful for folks beyond my lab,” Yoder said. “The goal is to produce something that is relatively easy for someone else to pick up without having me do all of the coding and, hopefully, people will find uses for it in contexts that I haven’t even thought of yet.”
You can learn more about Yoder’s work with Joshua trees in Episode 23 of CSUN’s “Diving into the Discourse” podcast.
