Scientists at Rothamsted have developed a novel mathematical model for seed germination. The new tool is derived from field data and marks a significant improvement in accuracy over previous models.
Germination models use water and temperature as the main drivers of seed activity. Currently these are based largely on water potential, which is the energy that drives water to move from one system to another. This is challenging to record accurately in the field. The new model uses water content which is much easier to measure, according to a press release.
These models are essential to the seed industry as it is important to develop lines where each individual seed germinates at roughly the same time and grows uniformly. This is particularly important for crops like carrots or onions where having similar sized plants makes them more profitable.
Seed suppliers must therefore test seed samples to ensure a certain germination rate is met, a process that is difficult and time consuming. So modelling is used to assess each batch based on a given sample.
However, current models do not always accurately reflect field conditions.
That is where the new model comes in. Water content is much easier to measure accurately in situ than the drivers of water potential. That makes the new model far more likely to accurately predict germination times in field situations.
“This is the first time a germination model has been able to be developed using field data,” said Dr Xiaoxian Zhang, corresponding author of the new study. “Seed germination is a crucial stage in plant development, intricately regulated by various environmental stimuli. Understanding these interactions is essential for optimizing planting and seedling management but remains challenging due to the trade-off effects of environmental factors on the germinating process.”
The research team developed a new model by conceptualizing seed germination as a dynamic process. To validate this model, they carried out field experiments by planting wheat seeds on different dates to establish a temperature gradient and in various plots to create a gradient of soil water content.
Over seven years, they compared the experimental data with the model’s calculated results. The findings revealed that the model accurately reproduces all germination patterns and subsequent seedling tillering, achieving a remarkable 95% accuracy.
“In germination there a trade-off effect of soil water on bioavailable water and oxygen. Introducing a physiological dimension enables seed germination and the subsequent tillering process to be modelled as a continuous physiological process, providing deeper insight into plant growth dynamics,” said Zhang.
“We believe this new approach offers a genuinely new approach to germination modelling and for the first time takes on tiller number. We also hope that it will be more stable than the hydrothermal time models and hence more useful for field application,” he added.
