Skip to main content
Log in

Assessing the fine root growth dynamics of Norway spruce manipulated by air humidity and soil nitrogen with deep learning segmentation of smartphone images

  • Research Article
  • Published:
Plant and Soil Aims and scope Submit manuscript

Abstract

Purpose

Measuring continuous root growth with less time- and resource-intensive methods is challenging for plant ecologists. We propose an approach where photos of fine root growth were taken with smartphones and analyzed by the deep learning method-based program RootPainter.

Methods

Picea abies saplings were grown in transparent boxes in growth chambers at either moderate (mRH) or elevated air humidity (eRH) and in nitrate (NO3) or ammonium (NH4+) dominated soil. To evaluate the partitioning of roots, we analyzed both fine root projection area (PA) of total and young white roots. We aimed to measure the ability of RootPainter to pick up on variation in fine root growth caused by changes in air humidity and soil nitrogen source.

Results

Trees growing at mRH-NH4+ had the highest total PA, 9.4 ± 1.9 cm2, while the lowest was in trees growing at eRH-NO3, 3.9 ± 0.6 cm2. The young root PA was highest at the beginning of the experiment and was affected by the initial soil nitrogen source. The older (brown) root PA increased over time. At mRH, the relative growth peak of root browning followed the peak of white roots, while at eRH, the peaks coincided. We found strong agreement between the automatic segmentation and manual annotation (F-measure = 0.88); the error measures showed no treatment-specific bias.

Conclusions

We conclude that the increased air humidity reduced the fine root growth and diminished sequential developmental peaks. The combination of smartphone images and RootPainter gives reliable results and is easy to use in future plant growth manipulation experiments.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9

Similar content being viewed by others

Data availability

The photo dataset used in the current study will be available from Zenodo. RootPainter is available from this link: https://colab.research.google.com/drive/104narYAvTBt-X4QEDrBSOZm_DRaAKHtA.

References

Download references

Acknowledgements

We thank Dr. Katrin Rosenvald for the pre-submission review of the manuscript and Prof. Krista Lõhmus for her advice in statistics. This study was funded by the Estonian Research Council grants (PRG916, PUT1350), the European Regional Development Fund (Center of Excellence EcolChange), and the Academy of Finland (PEATSPEC, decision no 341963, Iuliia Burdun).

Funding

This study was funded by the Estonian Research Council grants (PRG916, PUT1350), the European Regional Development Fund (Center of Excellence EcolChange), and the Academy of Finland (PEATSPEC, decision no 341963, Iuliia Burdun).

Author information

Authors and Affiliations

Authors

Contributions

This research was designed by Ivika Ostonen and Marili Sell. The growth chamber experiment was led by Priit Kupper. Marili Sell, Gristin Rohula-Okunev and Priit Kupper took care of plants on a daily basis. Marili Sell and Ivika Ostonen performed the imaging and analysis. Abraham G. Smith and Iuliia Burdun introduced deep-learning approach. Marili Sell, Iuliia Burdun and Ivika Ostonen conducted statistical analysis. Marili Sell wrote the manuscript with the support from Ivika Ostonen. All authors contributed to the manuscript revisions and improvements until the final version.

Corresponding author

Correspondence to Marili Sell.

Ethics declarations

Competing interests

The authors have no relevant financial or non-financial interests to disclose.

Additional information

Responsible Editor: Andrea Schnepf.

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Below is the link to the electronic supplementary material.

ESM 1

(DOCX 7.31 MB)

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Sell, M., Smith, A.G., Burdun, I. et al. Assessing the fine root growth dynamics of Norway spruce manipulated by air humidity and soil nitrogen with deep learning segmentation of smartphone images. Plant Soil 480, 135–150 (2022). https://doi.org/10.1007/s11104-022-05565-4

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s11104-022-05565-4

Keywords

Navigation