Wenxuan Guo uses drones as part of an effort to establish a high-throughput phenotyping framework that supports sustainable water use strategies for crops grown under limited water conditions. | Photo Norman Martin
As water supplies become strained across the state’s agricultural regions, a team of Texas Tech and Texas A&M AgriLife Research scientists are developing artificial intelligence-driven tools to help farmers make more precise irrigation decisions.
The research team aims to create a standardized framework capable of collecting and analyzing crop water stress data from multiple sources, including unmanned aircraft systems and satellite imagery.
The three-year effort, supported by a $300,000 grant from the U.S. Department of Agriculture's National Institute of Food and Agriculture, seeks to improve how crop water stress is measured and managed through an advanced phenotyping framework that combines remote sensing technologies, machine learning and crop modeling.
The project, titled "Developing a Multi-Scale Phenotyping Framework Integrating Machine Learning for Crop Water Stress Assessment," is led by Wenxuan Guo, associate professor of crop ecophysiology and precision agriculture at Texas Tech and Texas A&M AgriLife Research.
In semi-arid regions such as the South Plains, where agricultural productivity depends heavily on efficient water management, growers face mounting pressure to maximize yields while using less water. Yet accurately assessing crop water stress remains a challenge because plant responses are influenced by constantly changing environmental conditions.
Guo said advances in high-throughput plant phenotyping – rapid and efficient measurement of plant characteristics – are creating new opportunities to address those challenges. "By facilitating precise and scalable assessments of water stress across different scales, these tools hold great potential to advance water conservation practices," Guo said.
High-Throughput Phenotyping
The research team aims to create a standardized framework capable of collecting and analyzing crop water stress data from multiple sources, including unmanned aircraft systems, ground-based sensors and satellite imagery. Artificial intelligence will then be used to transform that information into practical recommendations for irrigation management.
According to Guo, the project’s long-term goal is to establish a high-throughput phenotyping framework that supports sustainable water use strategies for crops grown under limited water conditions. The framework will incorporate standardized data collection protocols, artificial intelligence and crop growth models to quantify and predict crop performance under varying levels of water stress.
The project’s objectives include developing a centralized database for multi-scale crop water stress data, creating machine learning models that can predict crop growth and development under different irrigation conditions, and building a decision-support tool capable of delivering real-time recommendations to growers.
By integrating these technologies into a single platform, the researchers hope to provide farmers and agricultural stakeholders with reliable, data-driven tools that support more informed and timely irrigation decisions.
Anticipated benefits include improved water-use efficiency, greater crop resilience and a scalable model for precision water management in water-limited agricultural regions.
Scientific Foundation
Guo's previous research has helped establish the scientific foundation for the project. His work has focused on assessing plant water stress using unmanned aerial systems, proximal sensing technologies and satellite imagery at both plot and field scales. He also has maintained close collaborations with regional producers to advance precision irrigation practices, providing practical insight into the challenges farmers face in managing limited water supplies.
Joining Guo on the project are Jingjing Yao, assistant professor in Texas Tech's Department of Computer Science, and Hope Njuki Nakabuye, assistant professor of irrigation engineering at the Texas A&M AgriLife Research and Extension Center in Lubbock.
Nakabuye brings extensive experience in irrigation scheduling and water management. Her research has focused on evaluating strategies to improve crop productivity while enhancing water- and nutrient-use efficiency. She also has worked closely with producers through on-farm management competitions aimed at improving profitability and resource conservation.
Yao specializes in unmanned aircraft systems networks, Internet of Things technologies, artificial intelligence and edge computing. Her work centers on improving real-time decision-making, energy efficiency and data processing within drone-based networks operating in complex agricultural environments.
Real-Wolrd Conditions
Field trials for the project will be conducted at three locations: a Texas Tech research field, a Texas A&M AgriLife research field and a commercial farm near Slaton, Texas.
Researchers say the sites will allow them to develop, test and refine the multi-scale phenotyping framework under real-world conditions. Each location is equipped with subsurface irrigation systems that enable precise control of water applications, creating a range of irrigation scenarios that simulate varying degrees of crop water stress.
The combination of controlled research environments and commercial production fields is expected to provide critical data for validating the technology and ensuring its practical value for farmers facing increasing pressure to produce more with less water.
Ultimately, researchers believe the project could help usher in a new generation of irrigation management tools that safeguard agricultural productivity while conserving scarce water resources in challenged farming regions.
CONTACT: Wenxuan Guo, Associate Professor of Crop Ecophysiology & Precision Agriculture | Joint Appointment, Texas A&M AgriLife Research; Department of Plant & Soil Science, Texas Tech University Department of Plant & Soil Science, Texas Tech University at (806) 834-2266 or wenxuan.guo@ttu.edu
0706NM26