A refined and different mesh was used for forward and inverse problems to, respectively, increase the simulation accuracy and avoid the inverse crime . The obtained inverted responses were compared with the responses of the true models. As for ERT, we compared true and inverted MALM responses. First, the true response was simulated with the 8 current sources overt the true ρmed. Second, a MALM response was calculated over the inverted ρmed and inverted to obtained the inverted CSD. Third, the obtained inverted CSD was used to forward calculate the inverted MALM response over the inverted ρmed. True and inverted MALM responses were then compared.We performed hydroponic and soil experiments using maize and cotton plants. In all the plant experiments, the injection electrode was positioned in the plant stem at a height of 1 cm from the surface of the growth media. For the hydroponic experiments, the plants were first grown in columns with aerated nutrient solution . They were then moved to the rhizotron for the experiments. As in the metallic roots test, the rhizotron was filled 1 day before the experiment to reach stable and homogeneous temperature and salinity conditions. The plant was positioned at the center of the rhizotron with soft rubber supports. The plants were submerged at the same level as in the growing column to avoid discrepancies caused by the plant tissue adaptation to the submerged and aerated conditions, as discussed above with regard to the growing conditions. Consequently, the root crown was approximately 3 cm below the water surface. For the soil experiments,square plastic pot seedlings were grown directly in the rhizotron to avoid damaging the roots and altering the root-soil interface. The soil was prepared by mixing equal volumes of sandy and clay natural soils acquired from an agricultural study site run by U.C. Davis, CA .
The plants were irrigated with double strength Hoagland solution . Two soil experiments were performed. In the first experiment, four cotton plants were grown for four months. For these experiments, the plants were positioned with the root crown approximately 8 cm deep . In the second experiment, a pregerminated maize seed was planted 3 cm deep and then grown for four months .Figure 3 shows the result of a synthetic numerical test performed to evaluate the iCSD resolution, inversion stability, and influence of imposed constraints. The obtained CSD matches the true position of the simulated current source. The sum of the current sources equals 1 as expected and required by the continuity constraint. The resolution of the CSD is in line with the electrode interspace . The first-order regularization does not hinder the reconstruction of simulated point source. Figure 4 shows the results of a laboratory experiment where the iCSD method was tested with a known distribution of current sources obtained with metallic wires. The use of metallic wires offered a comprehensive solution to test the overall correct functioning of the laboratory setup and inversion routine. The iCSD correctly characterized both position and intensity of the test sources with no need for prior information to constrain the solution. The asymmetrical distribution CSD agrees with the principle of path of least resistance, i.e., the closer to the return electrode, the shorter the water path, the higher the current source density. The obtained distribution of current source agrees with true CSD and resolution expected on the basis of previous numerical tests. The iCSD resolution was sufficient for the imaging of a root-like structure, e.g., differentiate between distal, proximal, or equally distributed current pathways in the roots . In this case, a distal CSD was investigated and the iCSD correctly found no proximal current source . Figures 5, 6, and 7 show the results of the synthetic tests based on the 8-source laboratory tests.
Figure 5 compares the true and inverted ρmed to highlight the overall influence of rhizotron meshing, acquisition sequence, processing, and inversion. Figure 6a offers a more quantitative comparison between true and inverted ERT responses. Similarly, Fig. 6b presents the correlation between true iCSD response and the inverted iCSD response . ERT and iCSD inverted responses correlate with the associated true responses . Figure 7 shows the inverted CSD distribution and confirms the iCSD capability to characterize both intensity and position of the current sources with a resolution of 3 cm. Plant Experimental Results. The plant experiments investigated the root current pathways combining the iCSD with the direct observation of the roots. Hydroponic and soil experiments were performed with maize and cotton plants. Figure 8a shows the root architecture for one of the two plant hydroponic experiments with maize plants. Although the roots were visually observed to extend deeper in the rhizotron, both the inverted CSD distributions show that the current leakage in this hydroponic system primarily occurred in the proximal root part at the top of the rhizotron. In particular, the highest current density was predominantly concentrated near the root crown with the remaining current leakage already occurring along the submerged stem section. The ρmed obtained from the ERT inversion confirmed the expected homogeneous temperature and salinity conditions. In the soil experiments with cotton plants, the iCSD method successfully located the root position of the four plants along the soil surface and produced consistent signals corresponding to the root crowns . The results also show a localized current leakage near the root crowns with no appreciable current leakage along the four stem sections in the soil or in the distal portions of the root systems.
By contrast to the cotton results, the CSD distribution obtained from the maize experiment conducted within the soil presents current leakage along the stem section . The ERT-based ρmed obtained from the cotton soil experiments presents a strong heterogeneity with a highly resistive layer at the top of the rhizotron, representing a dry soil layer . The evolution of this dry layer is due to water loss through evapotranspiration. The ERT results highlight the need to account for the heterogeneity of ρmed in order to correctly calculate the responses of the VRTes during the iCSD. The ERT results also show how the ERT characterization can support the interpretation of the iCSD results by providing information on the soil characteristics and thus on possibly associated physiological responses, such as drought and suberization.Both numerical and laboratory iCSD tests successfully image the CSD without requiring prior assumptions on the actual distribution of the roots and CSD. These assumptions were present in previous works but were here replaced by application-specific constraints . In this sense, not only the explicit linear formulation allowed a direct implementation and evaluation of the these constraints,25 liter pot but also enabled full characterization of the resulting liner problem and John. Relative to the field setup of Mary et al. , the laboratory setup offered the advantages of more flexible positioning of the electrodes and consequently better ERT and MALM data coverage. The other significant advantages over previous studies were the reduced number of VRTe in a 2D distribution and the new iCSD routine resulting from the 2D distribution of the VRTe and trans-dimensional ERT inversion. The inversion times was significantly reduced by the proposed iCSD, particularly thanks to its linear formulation, use of specific constraints and code optimization. Solving the linear system described above took less than 3 s on a standard laptop, compared to a few minutes using the generic optimization process used in Mary et al. . The reduced inversion time also offers the advantage of a faster calculation of the Pareto front, which allowed a more informed inversion regularization. Finally, the coupling between ERT and iCSD through the python geophysical library eased the optimization of the entire routine . Coupling ERTand iCSD is a relevant aspect of our procedure because the ERT data processing/analysis routine has a strong influence on the successive iCSD inversion, which may require the entire procedure to be repeated to test different ERT inversion parameters .The iCSD results showed that the current leakage occurred in the very proximal regions of the root systems in both soil and hydroponic conditions. The proximal leakage was observed despite the return electrode being placed at the bottom of the rhizotron to allow deep current pathways. Nonetheless, the expected influence of the return electrode position was observed in the laboratory test with metallic roots and motivates the use of this electrode configuration in future laboratory experiments. Our results are consistent with the early studies on maize root electrical properties , and corroborate the recent works that questioned the assumptions of the BIA methods . The high resistivity of the top layer of soil is expected to induce root suberization : our results would support the physiological hypothesis of water and nutrient absorption through older and possibly suberized roots .
It is worth noticing that the absence of current leakage along the section of the cotton stems in the top dry soil supports the assumption that the electrical structure of roots controls their current conduction behavior. Suberized epidermal cells can affect the movement of ions and, consequently, the current conduction in roots . Therefore, it is feasible for current to be conducted along deeper portions of the more woody roots with minor leakage. The BIA experimental results that have observed positive correlations between electrical signals and root area are likely a result of physiological correlations between the root regions that contribute to the current flow and hair roots, which contributes most to functional root surface area . While correlations between BIA electric signals and investigated root traits appear to be indirect, the correlations observed across experimental platforms and species continue to validate its value for in-situ root phenotyping . In their study on field grapevines, Mary et al. concluded that the CSD could be used to infer the root depth distribution. Because of the significant suberization of major roots in grapevines and orange trees, the electric current could penetrate deep into the root system before significant leakage occured. The deeper current penetration allowed the iCSD method to access and phenotype the root system. On the contrary, limited current results in lower sensitivity of the iCSD to distal and younger parts of the root system that are likely dominated by less suberized, finer roots. We attribute differences in current penetration among root systems of maize, cotton, grapevine, and orange tree to the differences in physiological traits such as the extent of suberization and lignification. If on one hand the sensitivity to the root physiological traitsis a promising opportunity, on the other hand it has to be accounted for when phenotyping more herbaceous roots.The aluminum ion in acid soils is a major factor that inhibits root elongation at the root apex. Root cap cells play important roles in the protection of root apex against soil stresses such as soil compaction and pathogen attacks. Root cap cells, although they have strong binding capacity with Al, play minor roles in the amelioration of Al-caused root elongation inhibition . Acacia mangium Willd. is a highly Al-tolerant leguminous tree, in spite of the fact that only a small amount of citrate is released against Al . We have recently noticed that the root apex of this species is surrounded by a large, cap-like structure of sloughed-off tissues. To understand its role in high Al tolerance mechanisms, we examined the effects of the cap-like structure in the Al-resistant root elongation in A. mangium. We show a representative A. mangium root under hydroponic condition in Figure 1a. A cap-like structure, which was apparently consisted of the plant tissues, covered the entire root apex up to the 5 mm region behind the tip. After the structure was readily sloughed off , the root was covered again with a new one of similar size in a week after the detachment . To evaluate roles of the structure on root growth, the whole structure and resulting tissues on root surface were carefully removed with forceps, and then the roots were exposed to 0.5 mM calcium solution containing 0, 100, or 500 μM Al for 48 h. Al induced a root bending in both roots immediately after the exposure, and the proportion in the occurrence of root bending was higher in Al-treated roots without the cap-like structure .