Goren et al. (2026) Fluorescence noise analysis as a sensitive indicator of dehydration stress in plants

Identification

• Journal: Smart Agricultural Technology
• Year: 2026
• Date: 2026-01-03
• Authors: Naama Goren, Naama Maroudas-Sklare, Sivan Mazaki, Hagit Zer, Shira Yochelis, G. Jung, Nir Keren, Yossi Paltiel
• DOI: 10.1016/j.atech.2026.101787

Research Groups

• Department of Applied Physics, Hebrew University of Jerusalem, Jerusalem, Israel
• Department of Plant & Environmental Sciences, The Alexander Silberman Institute of Life Sciences, Hebrew University of Jerusalem, Jerusalem, Israel
• Department of Physics, Ben Gurion University of the Negev, Beer Sheva, Israel
• Instytut Fizyki PAN, Warszawa, Poland

Short Summary

This study introduces a novel, non-invasive method for detecting plant dehydration stress by analyzing the statistical properties of ambient chlorophyll fluorescence fluctuations ("noise"). It demonstrates that changes in noise amplitude distributions and power spectral density patterns reliably correlate with water deficiency, offering a more precise and scalable tool for early drought detection in agriculture than conventional methods.

Objective

• To introduce a novel approach utilizing noise analysis of ambient chlorophyll fluorescence from plants to assess dehydration status.
• To identify characteristic changes in fluorescence fluctuations ("noise") that correlate with water deficiency.
• To develop a reliable, cost-effective, non-invasive, and scalable method for early drought detection in agricultural settings.

Study Configuration

• Spatial Scale: Individual plant leaves (Arabidopsis thaliana, Sorghum bicolor, Solanum lycopersicum/tomato). Focus on chloroplast and thylakoid membrane structure via Transmission Electron Microscopy (TEM).
• Temporal Scale: Experiments involved continuous monitoring over 7 to 11 days for dehydration progression, with measurements taken over several weeks of plant growth. Technical replicas were performed every 20 minutes for continuous monitoring.

Methodology and Data

• Models used:
  • Multi-Gaussian and single-Gaussian models for fitting noise amplitude distributions (histograms).
  • First-order power law model for fitting Power Spectral Density (PSD) in specific frequency regimes.
  • Second-order exponential model for detrending signals prior to Fast Fourier Transform (FFT).
  • First-order polynomial model for detrending Arabidopsis signals.
• Data sources:
  • Ambient chlorophyll fluorescence measurements using a custom setup (450 nm laser, chopper, photodetector, lock-in amplifier, spectrum analyzer).
  • Pulse-Amplitude Modulation (PAM) fluorometer (Dual-PAM) for continuous fluorescence monitoring and Photosystem II (ΦPSII) quantum yield measurements.
  • Transmission Electron Microscopy (TEM) for visualizing chloroplast and thylakoid membrane structure.
  • Plant growth under controlled environmental conditions (temperature, light cycles, irrigation levels).

Main Results

• Fluorescence noise amplitude distributions (histograms) showed distinct changes with dehydration: healthy plants exhibited multi-Gaussian fits (ordered energy transfer), dry plants displayed broader distributions (disorder), and very dry plants narrowed to a single dominant Gaussian (cessation of photosynthetic activity).
• The Gaussian width of noise distributions consistently increased during early dehydration in Sorghum, demonstrating higher precision and reliability in detecting water deficit compared to standard ΦPSII measurements, which showed greater variability.
• Power Spectral Density (PSD) analysis revealed frequency-dependent changes: Sorghum showed an overall flattening of the PSD curve and a reduction in white noise level, while Tomato exhibited shifts in power law exponents (e.g., from 1/√f to 1/f at low frequencies, and to 1/f² at high frequencies), indicating changes in the dimensionality of the fluorescing system.
• Transmission Electron Microscopy (TEM) confirmed structural changes in chloroplasts of dry Sorghum plants, including a lower ratio of stacked (grana) vs. unstacked (lamellar) thylakoid membranes, and increased lipid droplets and starch deposits, which correlated with the observed noise patterns.
• Changes in noise patterns were detectable even in the initial days of water deficit, highlighting the method's potential for early dehydration detection.

Contributions

• Introduces a novel signal-processing framework for plant stress detection based on the statistical properties of chlorophyll fluorescence fluctuations, extending beyond conventional fluorescence intensity methods.
• Demonstrates that fluorescence noise analysis provides a more precise and reliable indicator of dehydration stress compared to standard photosynthetic efficiency measurements (ΦPSII).
• Establishes a mechanistic link between fluorescence noise patterns and structural changes at the molecular level (e.g., thylakoid membrane disorganization, pigment aggregation) confirmed by TEM.
• Proposes a scalable, non-invasive, and cost-effective method compatible with automated sensor networks or drone-based platforms for large-scale agricultural monitoring and real-time irrigation decision-making.
• Offers a powerful tool for probing the energetic landscape in biophysical systems through advanced signal-processing techniques.

Funding

• Grant 663/23 from the Israeli Science Foundation.

Citation

@article{Goren2026Fluorescence,
  author = {Goren, Naama and Maroudas-Sklare, Naama and Mazaki, Sivan and Zer, Hagit and Yochelis, Shira and Jung, G. and Keren, Nir and Paltiel, Yossi},
  title = {Fluorescence noise analysis as a sensitive indicator of dehydration stress in plants},
  journal = {Smart Agricultural Technology},
  year = {2026},
  doi = {10.1016/j.atech.2026.101787},
  url = {https://doi.org/10.1016/j.atech.2026.101787}
}