PBPavel Baykalov

Pavel Baykalov

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Pavel Baykalov has a total of 22 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Soil Warming Shifts Seasonal and Soil Depth‐Related Root Growth Dynamics in Subarctic Grasslands

    by Biplabi Bhattarai, Pavel Baykalov, Páll Sigurdsson, Coline Le Noir de Carlan, Bjarni D. Sigurdsson, Ivan Janssens, Boris Rewald, Ivika Ostonen

    Climate warming in northern ecosystems is accelerating, yet the responses of belowground growth dynamics under warming remain poorly understood. We assessed how soil warming affects seasonal root growth dynamics in subarctic grasslands, aiming to predict belowground responses under future climate scenarios. We investigated root growth across soil depths and seasons in geothermally warmed subarctic grasslands, comparing two warming durations: medium‐term (MTW; 11–14 years) and long‐term (LTW; > 60 years) and three warming magnitudes (ambient: +0°C, < +2°C, and > +2°C). Root length and projection area were quantified from minirhizotron images using the automated segmentation model RS_SAG. The automated segmentation model trained and used in this study performed robustly, with accuracy comparable to manual labels. Warming duration and magnitude effects on root length and projection area were season‐ and soil‐depth‐dependent. Root growth responses were more pronounced under LTW than MTW, particularly in the subsoil under warming > +2°C. Significant differences from ambient conditions occurred during summer, where root length and projection area were larger than in ambient conditions. Warming shifted root growth phenology by accelerating peak‐season growth and intensifying winter decline. Enhanced subsoil root growth was related to a higher topsoil rhizome biomass, which likely provided additional energy to support deeper root growth and a higher abundance of Kobresia myosuroides —a deep‐rooted rhizomatous sedge species with extensive belowground biomass. Higher subsoil root length and projection area under LTW indicate a shift towards deeper rooting, potentially associated with changes in species composition or nutrient availability. Enhanced root growth, proliferation, temporal shifts and seasonal amplitude may increase litter inputs, influencing microbial activity and organic matter dynamics in subsoils. These deep‐root responses may partially offset warming‐induced organic matter decomposition but could also affect the stability of soil carbon pool, emphasising the role of deep‐soil roots in carbon cycling in subarctic soils.

  • Open Access

    Spectral clustering analysis: discrimination of grass-herb roots and live-dead roots in VISNIR and SWIR regions

    by Pavel Baykalov, Gernot Bodner, Ivika Ostonen, Boris Rewald

    Background and aims: Hyperspectral imaging is becoming a key, high-throughput technique in plant research. However, its application to roots has not yet received sufficient attention. The aims of this study are to identify spectral features that distinguish fine roots from soil, non-woody roots of different species, and dead from living roots, and to identify appropriate analytical techniques. Methods: Roots of Alopecurus pratensis (meadow foxtail) and Urtica dioica (nettle) and the rhizosphere were imaged in rhizoboxes in the wavelength range 400–1700 nm, covering both visible near- (VISNIR) and shortwave infrared (SWIR) regions. Principal Component Analysis, K-means clustering, and Generalised Linear Model, Partial Least Squares Discriminant Analysis, and Distributed Random Forest models were used to classify groups. Wavebands critical for classification were identified. Results: Our results demonstrate the intricate nature of spectra clustering, highlighting the challenges in the VISNIR range and the promise of SWIR data for enhanced separability. While species differentiation is challenging, the determination of the living conditions of the roots is possible within the SWIR range. The analysis reveals the significance of specific spectral regions, notably those associated with water content and senescence, in distinguishing between living and dead roots. Water content regions (mainly 1245 nm and 1450 nm) were most important in discriminating between roots and soil. Conclusions: This study highlights the potential of spectral analysis, particularly in the SWIR region, for distinguishing roots by species and vitality. Further efforts are needed to develop robust methods for mixed data sets containing roots of different species and degrees of vitality. Graphical


  • Open Access

    Automatic Root Length Estimation from Images Acquired In Situ without Segmentation

    by Faina Khoroshevsky, Kaining Zhou, Sharon Chemweno, Yael Edan, Aharon Bar-Hillel, Ofer Hadar, Boris Rewald, Pavel Baykalov, Jhonathan E. Ephrath, Naftali Lazarovitch

    Image-based root phenotyping technologies, including the minirhizotron (MR), have expanded our understanding of the in situ root responses to changing environmental conditions. The conventional manual methods used to analyze MR images are time-consuming, limiting their implementation. This study presents an adaptation of our previously developed convolutional neural network-based models to estimate the total (cumulative) root length (TRL) per MR image without requiring segmentation. Training data were derived from manual annotations in Rootfly, commonly used software for MR image analysis. We compared TRL estimation with 2 models, a regression-based model and a detection-based model that detects the annotated points along the roots. Notably, the detection-based model can assist in examining human annotations by providing a visual inspection of roots in MR images. The models were trained and tested with 4,015 images acquired using 2 MR system types (manual and automated) and from 4 crop species (corn, pepper, melon, and tomato) grown under various abiotic stresses. These datasets are made publicly available as part of this publication. The coefficients of determination (R2), between the measurements made using Rootfly and the suggested TRL estimation models were 0.929 to 0.986 for the main datasets, demonstrating that this tool is accurate and robust. Additional analyses were conducted to examine the effects of (a) the data acquisition system and thus the image quality on the models’ performance, (b) automated differentiation between images with and without roots, and (c) the use of the transfer learning technique. These approaches can support precision agriculture by providing real-time root growth information.


  • Open Access

    Semantic segmentation of plant roots from RGB (mini-) rhizotron images—generalisation potential and false positives of established methods and advanced deep-learning models

    by Pavel Baykalov, Bart Bussmann, Richard Nair, Abraham George Smith, Gernot Bodner, Ofer Hadar, Naftali Lazarovitch, Boris Rewald

    Background: Manual analysis of (mini-)rhizotron (MR) images is tedious. Several methods have been proposed for semantic root segmentation based on homogeneous, single-source MR datasets. Recent advances in deep learning (DL) have enabled automated feature extraction, but comparisons of segmentation accuracy, false positives and transferability are virtually lacking. Here we compare six state-of-the-art methods and propose two improved DL models for semantic root segmentation using a large MR dataset with and without augmented data. We determine the performance of the methods on a homogeneous maize dataset, and a mixed dataset of > 8 species (mixtures), 6 soil types and 4 imaging systems. The generalisation potential of the derived DL models is determined on a distinct, unseen dataset. Results: The best performance was achieved by the U-Net models; the more complex the encoder the better the accuracy and generalisation of the model. The heterogeneous mixed MR dataset was a particularly challenging for the non-U-Net techniques. Data augmentation enhanced model performance. We demonstrated the improved performance of deep meta-architectures and feature extractors, and a reduction in the number of false positives. Conclusions: Although correction factors are still required to match human labelled root lengths, neural network architectures greatly reduce the time required to compute the root length. The more complex architectures illustrate how future improvements in root segmentation within MR images can be achieved, particularly reaching higher segmentation accuracies and model generalisation when analysing real-world datasets with artefacts—limiting the need for model retraining.