Skip to main content
ARS Home » Midwest Area » Wooster, Ohio » Application Technology Research » Research » Publications at this Location » Publication #413605

Research Project: Sustainable Production and Pest Management Practices for Nursery, Greenhouse, and Protected Culture Crops

Location: Application Technology Research

Title: Developing supervised machine learning algorithms to classify lettuce foliar tissue samples into interpretation zones for 11 plant essential nutrients

Author
item VEAZIE, PATRICK - North Carolina State University
item CHEN, HSUAN - North Carolina State University
item HICKS, KRISTIN - North Carolina Department Of Agriculture & Consumer Services
item HOLLEY, JAKE - Colorado State University
item EYLANDS, NATHAN - University Of Minnesota
item MATTSON, NEIL - Cornell University
item Boldt, Jennifer
item BREWER, DEVIN - Michigan State University
item LOPEZ, ROBERTO - Michigan State University
item WHIPKER, BRIAN - North Carolina State University

Submitted to: Urban Agriculture and Regional Food Systems
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 10/21/2024
Publication Date: 11/5/2024
Citation: Veazie, P., Chen, H., Hicks, K., Holley, J., Eylands, N., Mattson, N., Boldt, J.K., Brewer, D., Lopez, R., Whipker, B. 2024. Developing supervised machine learning algorithms to classify lettuce foliar tissue samples into interpretation zones for 11 plant essential nutrients. Urban Agriculture and Regional Food Systems. 9(1) Article e70002. https://doi.org/10.1002/uar2.70002.
DOI: https://doi.org/10.1002/uar2.70002

Interpretive Summary: Proper fertilizer management is an economic and environmental goal of commercial greenhouse growers. The timely and accurate detection of nutrient deficiencies or toxicities in a crop allows growers to produce high yields while minimizing fertilizer use. However, crop differences and interactions between nutrients can make accurate detection a challenge. This research utilized machine learning to provide a new method to interpret leaf nutrient status for lettuce grown in controlled environments. The machine learning models provided a more accurate interpretation of nutrient status than the traditional method. These models will help growers and technical specialists more accurately interpret lettuce leaf data. Improved interpretation of leaf nutrient status will improve crop yields, reduce overuse of fertilizers, and improve grower profitability.

Technical Abstract: Greenhouse crop nutrient management recommendations based on foliar tissue testing rely heavily on human interpretation which can result in recommendation variations and errors. Critical nutrient ranges vary for each species and the potential for error in interpretation increases due to this complexity. Machine learning can be utilized to develop algorithms to accurately classify new information using models developed on known data from a training data set. This study examines four different machine learning algorithms (J48, Random Forest, SMO, and MLP) by two different cross-validation strategies (10-fold and 66% split) to determine if machine learning can be utilized to accurately classify foliar tissue samples within corresponding nutrient ranges. Lettuce (Lactuca sativa L.) foliar tissue samples (n=1950) from a variety of controlled experiments and diagnostic samples from state and private labs were compiled and assigned to one of five nutrient ranges of deficient, low, sufficient, high, or excessive for each of eleven plant essential nutrients of interest based on Gamma or Weibull distributions. Individual machine-learning algorithms were developed for each nutrient. For all examined essential nutrients, J48 or random forest (RF) yielded the greatest percent correct classification when compared to MLP or SMO. This study establishes the novel use of machine learning for lettuce foliar nutrient analysis results interpretation with a higher accuracy rate by traditional statistical methods.