Digital twin model for optimizing wheat photosynthetic and nitrogen use efficiency

Overview

This research program develops a digital twin model that links wheat canopy architecture with leaf-level photosynthetic metabolism to predict and optimize nitrogen use efficiency (NUE). By combining drone-based canopy imaging with genomics, transcriptomics, metabolomics, proteomics, and phenomics data, the platform characterizes how varietal differences in photosynthetic metabolism interact with canopy structure and field environment to determine NUE. The goal is to identify optimal canopy traits for improved photosynthetic performance, supporting higher wheat yields and more sustainable fertilization practices.

Technical specifications

Key features:

• Integration of multi-layer 'omics' datasets (genomics, transcriptomics, metabolomics, proteomics, phenomics) with graph-based modeling
• Drone-based multispectral and RGB imaging for 3-D canopy structure quantification from stem elongation through grain-filling stages
• Leaf-scale photosynthesis and metabolic profiling of flag leaves across heading, flowering, and grain-filling stages
• Field trial design covering 18 wheat varieties under three nitrogen rates (0, 120, and 180 kg N ha⁻¹) with four replications, totaling 216 plots
• Predictive modeling built on the GroIMP 3-D plant development platform combined with the STICS crop model for nitrogen stress simulation
• Machine learning algorithms to map relationships between metabolic intermediates, canopy traits, and NUE outcomes

Applications:

• Variety selection and breeding for improved nitrogen use efficiency
• Decision support for precision nitrogen fertilization in wheat production
• Sustainable crop management through optimized canopy architecture

Technology readiness level

The platform is currently at an early-to-mid validation stage. A first growing season of field data has been collected, and preliminary ANOVA results confirm significant differences in NUE and nitrogen content across varieties and nitrogen levels. Initial findings also show significant associations between plant height and NUE during flowering. The project spans two growing seasons to enable thorough calibration and validation of the predictive digital twin before broader deployment.