Sharma Receives USDA Grant to Conduct Cold-Stress Study of Looseleaf Lettuce

EAST GREENSBORO, N.C. (April 1, 2026) — A researcher at North Carolina Agricultural and Technical State University has been awarded a two-year federal grant of $260,533 to use hyperspectral imaging and artificial intelligence to identify varieties of looseleaf lettuce that can withstand cold and freezing temperatures that damage or kill most crops.

Harmandeep Sharma, Ph.D., a research assistant professor in the College of Agriculture and Environmental Sciences (CAES) and the principal investigator on this project, was awarded an Agriculture and Food Research Initiative grant from the U.S. Department of Agriculture’s National Institute of Food and Agriculture. Arnab Bhowmik, Ph.D., an associate professor of soil science and soil microbiology at N.C. A&T, is the project’s co-principal investigator.

Starting with 100 varieties of lettuce obtained from the U.S. National Plant Germplasm System, Sharma will grow looseleaf lettuce in her Sensors for Sustainable and Efficient Plant Production (SENSE) Lab on campus. Her goal is to identify 10 varieties most resistant to the cold.

“Lettuce is a very common crop that has been studied widely for how it handles heat and water stress, but there has been little research on its ability to survive in cold weather after germination,” said Sharma. “This study will be among the first to pinpoint cold temperature thresholds in selected looseleaf lettuce accessions. With an increase in adverse weather conditions that cause fluctuating temperatures, it’s important to understand how cold stress affects lettuce yields.”

During the project, Sharma will grow lettuce in a controlled-climate chamber and subject the young plants to progressively colder temperatures ranging from 4°C (39°F) to -6°C (21°F). Hyperspectral imaging will be used on the surviving plants to capture spectral signatures that can indicate stress and collect a range of morphological, physiological and biochemical data. This data will be used to develop deep-learning models to predict health, yields, survivability and potential of cold-resistant lettuce varieties.

Sharma said hyperspectral imaging is a more precise, efficient and non-destructive method of conducting high-throughput assessments of plant traits. In addition to identifying looseleaf lettuce cultivars that could be bred with other varieties to improve lettuce productivity in colder weather, this study could create a research protocol for studying other green leafy plants.