Carbon allocation governs how plants partition assimilated carbon among leaves, roots, stems, and reproductive organs, and thereby directly shapes growth and final yield. Most existing crop growth models represent partitioning using fixed coefficients or empirically derived rules. Those prescribed approaches limit the models' capacity to predict how allocation will respond to variation in genotype, environment, or genotype × environment interactions. As an alternative, eco‑evolutionary optimality theory treats allocation strategy as an emergent property determined by marginal costs and benefits of investing carbon into different organs. This conceptual shift has precedent in applications to woody plants but had not previously been developed for herbaceous crops such as wheat prior to this work.
The authors present DAESIM2‑Plant, a mechanistic plant growth model that integrates multiple physiological processes with an optimal allocation scheme. Key coupled processes in the model include:
Within this mechanistic framework, an eco‑evolutionary optimal allocation module determines how carbon is partitioned between leaves and roots based on marginal costs and benefits. Allocation is therefore dynamic and responsive to environmental conditions and the plant's current state rather than fixed a priori.
To apply the framework to a cereal crop, DAESIM2‑Plant is coupled to a source–sink grain production module. This coupling allows the same assimilate pool that constrains vegetative growth to also constrain grain development during the critical period and grain filling. Using this integrated source–sink approach, the authors simulate both vegetative carbon allocation dynamics and grain yield outcomes within a single model framework.
The study uses a series of idealized sensitivity experiments to assess whether the model reproduces well‑documented patterns of allocation plasticity. Reported model behaviours include:
These emergent patterns align qualitatively with established empirical observations and conceptual expectations for allocation plasticity under varying light and water regimes.
When simulating a full growing season across a gradient of soil moisture levels, the model produced a threshold‑like response in canopy development, total biomass accumulation, and grain yield. The authors describe yield as being constrained by the same assimilate supply dynamics that govern vegetative growth, particularly during the crop's critical period and grain filling. The threshold behaviour suggests nonlinear responses of canopy development and yield to progressive changes in soil moisture within this modeling framework.
By allowing allocation strategies to emerge from first principles of marginal benefit and cost within a mechanistic physiological model, DAESIM2‑Plant offers a path to generalize crop model behaviour across a range of environmental conditions and potentially across genotypes. The approach contrasts with fixed partitioning schemes and may improve representation of allocation plasticity under drought and light limitation.
The authors note that evaluating model outputs against experimental and field trial data remains an important next step. Because this work is presented as a preprint, it has not been peer reviewed. Details regarding parameterization, quantitative model performance metrics, and comparison with empirical datasets were not reported in the abstract and would need to be assessed in the full manuscript and supplementary material.
The authors declared no competing interests. Reported funders include the Australian National Environmental Science Program (NESP) – Climate Systems Hub and an Australian Research Council Linkage Project. The preprint is made available under a CC‑BY 4.0 license. The authorship and contact details were provided in the source, and the work is hosted on bioRxiv as a preprint.