CRootBox is a fast and flexible functional-structural root model that is based on state-of-the-art computational science methods. Its aim is to facilitate modelling of root responses to environmental conditions as well as the impact of roots on soil. In the future, this approach will be extended to the above-ground part of the plant.

Root systems of crops play a significant role in agroecosystems. The root system is essential for water and nutrient uptake, plant stability, symbiosis with microbes, and a good soil structure. Minirhizotrons have shown to be effective to noninvasively investigate the root system. Root traits, like root length, can therefore be obtained throughout the crop growing season. Analyzing datasets from minirhizotrons using common manual annotation methods, with conventional software tools, is time-consuming and labor-intensive. Therefore, an objective method for high-throughput image analysis that provides data for field root phenotyping is necessary. In this study, we developed a pipeline combining state-of-the-art software tools, using deep neural networks and automated feature extraction. This pipeline consists of two major components and was applied to large root image datasets from minirhizotrons. First, a segmentation by a neural network model, trained with a small image sample, is performed. Training and segmentation are done using “RootPainter.” Then, an automated feature extraction from the segments is carried out by “RhizoVision Explorer.” To validate the results of our automated analysis pipeline, a comparison of root length between manually annotated and automatically processed data was realized with more than 36,500 images. Mainly the results show a high correlation (r=0.9) between manually and automatically determined root lengths. With respect to the processing time, our new pipeline outperforms manual annotation by 98.1-99.6%. Our pipeline, combining state-of-the-art software tools, significantly reduces the processing time for minirhizotron images. Thus, image analysis is no longer the bottle-neck in high-throughput phenotyping approaches.

4‐yr monitoring of soil water content dynamics for different plant types. Horizontal crosshole GPR at the field‐plot scale sampled ∼65 m3. Natural, irrigated, and sheltered treatments in gravelly and clayey–silty soils. Consistent patches of soil water content were observed along horizontal slices using GPR. Results enable improved investigation of soil–plant–atmosphere interactions. Ground penetrating radar (GPR) has shown a high potential to derive soil water content (SWC) at different scales. In this study, we combined multiple horizontal GPR measurements at different depths to investigate the spatial and temporal variability of the SWC under cropped plots. The SWC data were analyzed for four growing seasons between 2014 and 2017, two soil types (gravelly and clayey–silty), two crops (wheat [Triticum aestivum L.] and maize [Zea mays L.]), and three different water treatments. We acquired more than 150 time‐lapse GPR datasets along 6‐m‐long horizontal crossholes at six depths. The GPR SWC distributions are distinct both horizontally and vertically for both soil types. A clear change in SWC can be observed at both sites between the surface layer (>0.3 m) and subsoil. Alternating patches of higher and lower SWC, probably caused by the soil heterogeneity, were observed along the horizontal SWC profiles. To investigate the changes in SWC with time, GPR and time‐domain reflectometry (TDR) data were averaged for each depth and compared with changes in precipitation, treatment, and soil type. The high‐temporal‐resolution TDR and the large‐sampling‐volume GPR show similar trends in SWC for both sites, but because of the different sensing volumes, different responses were obtained due to the spatial heterogeneity. A difference in spatial variation of the crosshole GPR SWC data was detected between maize and wheat. The results for this 4‐yr period indicate the potential of this novel experimental setup to monitor spatial and temporal SWC changes that can be used to study soil–plant–atmosphere interactions.

Soils with high stone content represent a challenge to root development, as each stone is an obstacle to root growth. A high stone content also affects soil properties such as temperature or water content, which in turn affects root growth. We investigated the effects of all soil properties combined on root development in the field using both experiments and modeling. Field experiments were carried out in rhizotron facilities during two consecutive growing seasons (wheat [Triticum aestivum L.] and maize [Zea mays L.]) in silty loam soils with high (>50%) and low (<4%) stone contents. We extended the CPlantBox root architecture model to explicitly consider the presence of stones and simulated root growth on the plot scale over the whole vegetation period.We found that a linear increase of stone content resulted in a linear decrease of rooting depth across all stone contents and developmental stages considered, whereas rooting depth was only sensitive to cracks below a certain crack density and at earlier growth stages. Moreover, the impact of precipitation-influenced soil strength had a relatively stronger impact on simulated root arrival curves during the vegetation periods than soil temperature. Resulting differences between stony and non-stony soil of otherwise the same crop and weather conditions show similar trends as the differences observed in the rhizotron facilities. The combined belowground effects resulted in differences in characteristic root system measures of up to 48%. In future work, comparison of absolute values will require including shoot effects-in particular, different carbon availabilities.

Traits of the plant root system architecture (RSA) play a key role in crop performance. Therefore, architectural root traits are becoming increasingly important in plant phenotyping. In this study, we use a mathematical model to investigate the sensitivity of characteristic root system measures, obtained from different classical field root sampling schemes, to RSA parameters. Methods Root systems of wheat and maize were simulated and sampled virtually to mimic real field experiments using the root system architecture (RSA) model CRootBox. By means of a sensitivity analysis, we found RSA parameters that significantly influenced the virtual field sampling results. To identify correlations between sensitivities, we carried out a principal component analysis. Results We found that the parameters of zero order roots are the most sensitive, and parameters of higher order roots are less sensitive. Moreover, different characteristic root system measures showed different sensitivity to RSA parameters. RSA parameters that could be derived independently from different types of field observations were identified. Conclusions Selection of characteristic root system measures and parameters is essential to reduce the problem of parameter equifinality in inverse modeling with multi-parameter models and is an important step in the characterization of root traits from field observations.