Analysis of root growth by impedance spectroscopy (EIS)
Résumé
Electrical impedance spectroscopy (EIS) is investigated as a non-destructive method for monitoring root growth of tomato. This paper aims to (i) review the basic principles of EIS applied to the characterisation of the different parts of the soil–root–stem-electrode continuum, (ii) experiment the validity of the relationship between root weight and root capacitance taking into account the influence of the soil and plant electrodes position, (iii) describe an EIS analysis of the root growth of tomato plants. All experiments were carried out in 50 dm3 containers either in hydroponics at 930 μS for the test of root fresh or dry weight and root capacitance relationships, or in a potting mix (oxisol) for electrode placement tests and EIS estimation of root growth. Electrical measurements of the soil–root–stem-electrode continuum were done with a two-electrode measuring system using unpolarisable Ag–AgCl electrodes. A ‘root cutting’ and a ‘progressively immersed root system’ experiments were carried out in order to validate the relationship between root capacitance and root mass at 1 kHz. The effects of soil electrode and plant electrode placement were also tested, pointing out the sensitivity of the method to the insertion height of the “plant electrode” into the stem. For the root growth experiment, Impedance Spectra (IS) measurements were made just before harvesting the roots for dry weight and length determination. Measurements were made 14, 22, 26 and 39 days after planting, until flowering. The IS of the soil–root–stem-electrode continuum was modelled by a lumped electric circuit consisting of a series resistor R 0 for the soil and of four parallel resistance (R i )-capacitance (C i ) circuits for the other components of the circuit. The model had nine parameters whose values were estimated by Complex Nonlinear Least Squares curve fitting. A stepwise ascendant regression was used to identify the electrical parameters that better correlated with root dry mass or length increment: C 3 and C 4 were well correlated with root dry mass with a r 2 of 0.975, whereas root length was explained by the combination of 1/R 3, C 3, 1/R 2 and 1/R 1 with a r 2 of 0.986. This work may be considered as a new methodological contribution to the understanding of root electrical properties in the non-destructive diagnosis of root systems.