Kinetically Consistent Data Assimilation for Plant PET Sparse Time Activity Curve Signals

Area of Science:

• Nuclear science
• Plant physiology
• Fluid dynamics

Background:

• Plant positron emission tomography (PET) uses time activity curve (TAC) signal processing to study fluid dynamics (FD) in plants, crucial for crop management and sustainability.
• Current TAC signal processing methods have sparse space-time sampling, limiting the extraction of FD variables to averaged values.
• A data-driven approach for FD modeling in plants has not been previously established.

Purpose of the Study:

• To develop a novel data-driven digital signal processing method for plant PET.
• To enable direct computation of dynamic noise correlations and account for numerical diffusion.
• To accurately estimate spatial velocity profiles, diffusion coefficients, and compartmental exchange rates from sparse TAC signals.

Main Methods:

• Proposed a novel sparse data assimilation digital signal processing method.
• Incorporated direct computation of dynamic evolution of noise correlations.
• Explicitly accounted for numerical diffusion caused by sparse sampling.
• Employed a sequential time-stepping procedure to estimate FD parameters.

Main Results:

• Successfully estimated spatial velocity profiles, diffusion coefficients, and compartmental exchange rates.
• Demonstrated the method's capability in analyzing transport mechanisms in zucchini sprouts.
• Overcame limitations of sparse space-time sampling in TAC signal processing.

Conclusions:

• The novel method enhances the quantitative analysis of plant fluid dynamics using PET.
• This approach provides a foundation for data-driven FD modeling in plants.
• Improved understanding of crop transport mechanisms can contribute to sustainable agriculture.