Towards a Stochastic Model to Simulate Grapevine Architecture: A Case Study on Digitized Riesling Vines Considering...

Area of Science:

• Plant science
• Computational biology
• Agricultural modeling

Background:

• Functional-structural plant models (FSPMs) are crucial for in silico studies of environmental impacts on plant growth.
• Predictive capabilities of FSPMs can reduce extensive on-field experimental efforts.
• Incorporating natural variability through stochasticity enhances model robustness and prediction accuracy.

Purpose of the Study:

• To develop stochastic model components for the Virtual Riesling FSPM using Bayesian calibration.
• To enhance the model's predictive capabilities for grapevine phenology and growth under future environmental conditions.
• To investigate the impact of elevated CO2 on grapevine growth and variability.

Main Methods:

• Bayesian model calibration and multi-objective optimization were employed.
• Development of stochastic components for phenology (budburst variability) and growth (phytomer development, internode elongation).
• Utilized grapevine data from two seasons under free-air carbon dioxide enrichment (FACE) conditions.

Main Results:

• Bayesian calibration with mixed models effectively captured natural shoot growth variability.
• Natural variability complicates the detection of treatment effects, such as elevated CO2.
• Cardinal temperatures were estimated for phenology and growth modeling using a development days approach.

Conclusions:

• The developed stochastic components enhance the realism of virtual plant simulations.
• Future extensions of stochastic models will improve in silico studies on canopy microclimate, grape health, and quality.
• This approach provides a robust framework for studying climate change impacts on viticulture.