The research presented herein describes the development of a virtual rice plant simulation model that combines genotype specific primary phenotypic traits and 3D-based light and nutrient mediated hormonal control, which define a cultivar’s underlying physiological processes and resultant seasonal pattern of growth, development, maturation and yield.

Our rice model simulates the growth and develop characteristics of a genotype through a cultivar’s set of ca. 30 primary phenotypic traits and their interaction with the abiotic and biotic environment. Primary phenotypic traits, such as the per unit area of leaf photosynthesis rate (=f(˚C, N concentration, CO₂ concentration, photon flux density), maximum rate of node production and maximum grain size, are near invariant across environments, as contrasted with secondary phenotypic traits, such as average node production rate, average grain size, and timing of crop maturation, which are not.

Our virtual model is a work in progress with further linkage of the physiological and architectural components ongoing. The model simulates the temporal and spatial dynamics of rice plants and their organs as affected by the biotic and abiotic environment, incorporating the concept of repeating functional units or phytomers, represented by main tillers and sub-tillers, with each tiller comprised of nodes and each node represented by a culm, a leaf sheath, a leaf blade, a subtiller bud or a subtiller, a root bud or root, and for the ultimate node a panicle. Although only partially tested, the emergence of a new tiller from its bud is controlled by the photon flux density absorbed by a bud, and the carbohydrate, nitrogen, and phosphate supply/demand balance of the subtending tiller.

An important aspect of our research focuses on quantifying the impact of each primary phenotypic trait on crop growth, development, and yield. To date, primary phenotypic traits have been estimated for 17 rice genotypes through detailed field experiments and through review of existing literature. Preliminary model-based analysis address help to identify which traits explain the largest majority of observed yield variability across genotypes.