UC Merced tests AI irrigation in a Parlier almond orchard, sprinkler by sprinkler
Researchers at UC Merced and UC ANR set up two test beds at the Kearney center in Parlier. One is watered the usual way and one is run by an algorithm, so they can compare water use and production.
AI-assisted summary of reporting by University of California. How we report
Key Takeaways
- UC Merced and UC ANR researchers installed an AI-run irrigation system in an almond orchard at the Kearney research center in Parlier.
- One test bed is irrigated the traditional way and one by the AI system, so researchers can compare water use and production.
- Professor Wan Du leads the project, which was funded through the Fall 2023 Climate Action Seed Funds.
- Students working with the nonprofit From Farms to Incubators documented the project in stories, photos and videos.
"We have sprinklers under almost every tree so we can control them," Wan Du said.
That's the setup in an almond block at the Kearney Agricultural Research and Extension Center in Parlier. Researchers from UC Merced and University of California Agriculture and Natural Resources have put an artificial intelligence system in charge of when each sprinkler opens and how long it runs. They want to know if an algorithm can grow the same crop on less water.
Two test beds, one comparison
Du is a computer science and engineering professor at UC Merced, and he leads the project. There are two test beds. One gets watered the traditional way, and the other runs on the AI system.
"We will compare how much water each test bed will use," Du said in a UC account of the project, "and then compare the production and result of these two fields."
The system measures moisture in the soil, then the potential for water to move up through a tree and out the leaves. That data gets uploaded to the internet. An algorithm reads it and decides whether to open a sprinkler and how long to leave it on.
Two other professors built the parts underneath. Safeeq Khan, a civil and environmental engineering professor who works in water resource management and soil science, built the soil model. Stefano Carpin, who teaches computer science and engineering, developed the AI model.
What it's supposed to save
Du said the goal is money in growers' pockets as much as water in the ground.
"We want to save as much water as we can," Du said. "Saving water will reduce the cost to growers, and we can save the natural resource for the next generation."
The money came from the Fall 2023 round of the Climate Action Seed Funds. When UC wrote about the project in September 2025, it said the system would be operating by the following spring, once the trees started to flower. That puts the first real run at bloom.
The students documenting it
Part of the grant's goal was spreading the word. Amy Wu, founder and chief content director of the nonprofit From Farms to Incubators, built a program to teach students how to communicate about ag tech. Wu met the professors at UC Merced, then drove out to Kearney to see the system and document it.
"We trained them on interviewing and writing, to document the journey of the AI-generated crop irrigation system," Wu said.
Two of the students were UC Merced undergraduates, and the third had recently graduated from California State University, Monterey Bay. They put together a package of stories, photos and videos meant for everyone from consumers curious about new technology to growers looking for help with climate change.
Anvi Kudaraya, a UC Merced computer science and engineering undergraduate from Pleasanton, wrote Python scripts to process real sensor data from the field. She said it taught her "how to handle messy, real-world datasets and prepare them for machine learning models."
Savio Jabbo, a computer science and engineering major from San Diego, saw the payoff on a small scale.
"Usually the work that goes into these projects goes unnoticed," Jabbo said. "But when you see the impact that it has even on such a small scale, it feels like the whole world should know about it."
Wu called the pilot a success.
"I thought it went very, very well," Wu said. "We went into this not knowing what to expect, but we wanted to expose young people to the possibilities of communicating food and farming."
Central Valley AI is produced by the CVAI Agriculture Desk team and developed by Kaweah Tech, a regional firm that builds, deploys, and integrates AI solutions for businesses across California's Central Valley.