Back

High frequency soil polarization can phenotype crop roots noninvasively

Gu, H.; Cseresnyes, I.; Butnor, J.; Li, B.; Mary, B.; Sun, H.; Zhang, X.; Lu, Y.; Liu, X.

2023-01-13 plant biology
10.1101/2023.01.12.523853 bioRxiv
Show abstract

Noninvasive and nondestructive root phenotyping techniques under field conditions are sorely needed to advance plant root science. Soil polarization measured by electrical capacitance (ECsoil) has the potential to meet this requirement, but whether it specifically detects root properties remains unexplored. We carried out manipulative experiments where wheat (Triticum aestivum L.) and maize (Zea mays L.) roots were buried in soil or immersed in hydroponic solution combined with pot trials to reveal the mechanism of root trait detection by ECsoil, while a field experiment was conducted to test its feasibility to determine root depth distribution. We found that ECsoil measured at low current frequency (< 1 kHz) was not significantly affected by the addition of roots to the system either by burying roots in soil or immersing them in solution. At frequency greater than10 kHz a shift occurred, and root polarization contributed more to ECsoil which was positively correlated with root volume. When ECsoil was measured at high frequency (30 kHz -100 kHz) it was well correlated with root volume vertical distribution in the field. The measurement error after soil moisture calibration at depths of 10 cm, 20 cm, 30 cm and 40 cm was 0.4%, 12.0%, 1% and 34%, respectively. Our results demonstrate that ECsoil is a robust method to measure in situ root distribution and we believe the newly available high frequency measurement equipment combined with novel root prediction models will enable ECsoil to be widely used for root phenotyping in the future.

Matching journals

The top 7 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.