10:45 〜 12:15
[AGE27-P08] Visualizing plant root distribution and its interaction with soil constituents using X-ray computed tomography
キーワード：plant root、X-ray computed tomography、image analysis
Plant roots are usually present in surface soils and have hydraulic and mechanical effects on the soil. The influence of these roots is expected to vary greatly depending on the root size and especially the distribution of roots in the soil. However, the growth of plant roots in the soil and their interactions with soil particles, soil water, and pore air still needs to be better understood. Therefore, this study aimed to obtain and analyze microscopic images of soil structure, including plant roots using X-ray CT, which has been remarkably developed recently.
Acrylic columns with an inner diameter of 20 mm and a height of 50 mm were filled with Toyoura sand, and Guinea grass, Italian grass, and soybean were grown. Fourteen days after germination, the soil in the acrylic column was photographed by a Metrotom 1500 (Carl Zeiss). Roots were segmented using the image analysis software ImageJ (NIH). Areas on the CT image that could be visually identified as plant roots were colored with a brightness value of zero. Soil particles, pore water, and air were segmented using ExFact VR (Visual Science, Japan), a cross-sectional image 3D software. Kriging was used for segmentation. After segmentation was completed, the distribution trends of each soil constituent were analyzed. Two types of images were selected for analysis: one near the surface, where pore air is abundant, and the other at the center of the column, where soil particles and pore water are plentiful. The size of the analyzed images was 50 voxels in the x-, y-, and z-directions, respectively.
Three-dimensional images of segmented soil particles, pore water, air, and plant roots were successfully produced. The 3D images revealed the distribution of plant roots in the soil. Thick roots grow horizontally near the surface, and thin roots grow vertically downward in the central area. This is likely because the roots near the surface are more stable when subjected to wind and other forces, allowing the plants to become self-supporting. In the central area, the stabilizing role of the roots is weakened, and vertical growth becomes dominant for the uptake of water and nutrients. Near the surface, the roots were surrounded by air, suggesting that the stabilizing role was more significant than the uptake of water and nutrients. On the other hand, the roots were covered with water at the deeper depths.
The volume fraction analysis of soil components showed that the air fraction was high in the surface layer but decreased in deeper soil while the soil particle fraction, i.e., dry bulk density, increased. Many of the plant roots in the surface layer were thick, and there was a significant difference in the volume fraction depending on whether the thick roots were included in the analysis area. In the center of the column, the proportion of soil particles was higher due to the higher dry density. The proportion of plant roots was lower in the center of the column due to the sparse growth of thin roots.
The analysis based on the X-ray CT images showed that the shape of plant roots varied with depth and that the volume fraction of soil constituents varied depending on the plant roots' growth state and condition. In the future, we would like to conduct similar evaluations using a greater variety of plants and soils and assess the effects of the interaction between plant root shape and soil components on soil properties.
Acrylic columns with an inner diameter of 20 mm and a height of 50 mm were filled with Toyoura sand, and Guinea grass, Italian grass, and soybean were grown. Fourteen days after germination, the soil in the acrylic column was photographed by a Metrotom 1500 (Carl Zeiss). Roots were segmented using the image analysis software ImageJ (NIH). Areas on the CT image that could be visually identified as plant roots were colored with a brightness value of zero. Soil particles, pore water, and air were segmented using ExFact VR (Visual Science, Japan), a cross-sectional image 3D software. Kriging was used for segmentation. After segmentation was completed, the distribution trends of each soil constituent were analyzed. Two types of images were selected for analysis: one near the surface, where pore air is abundant, and the other at the center of the column, where soil particles and pore water are plentiful. The size of the analyzed images was 50 voxels in the x-, y-, and z-directions, respectively.
Three-dimensional images of segmented soil particles, pore water, air, and plant roots were successfully produced. The 3D images revealed the distribution of plant roots in the soil. Thick roots grow horizontally near the surface, and thin roots grow vertically downward in the central area. This is likely because the roots near the surface are more stable when subjected to wind and other forces, allowing the plants to become self-supporting. In the central area, the stabilizing role of the roots is weakened, and vertical growth becomes dominant for the uptake of water and nutrients. Near the surface, the roots were surrounded by air, suggesting that the stabilizing role was more significant than the uptake of water and nutrients. On the other hand, the roots were covered with water at the deeper depths.
The volume fraction analysis of soil components showed that the air fraction was high in the surface layer but decreased in deeper soil while the soil particle fraction, i.e., dry bulk density, increased. Many of the plant roots in the surface layer were thick, and there was a significant difference in the volume fraction depending on whether the thick roots were included in the analysis area. In the center of the column, the proportion of soil particles was higher due to the higher dry density. The proportion of plant roots was lower in the center of the column due to the sparse growth of thin roots.
The analysis based on the X-ray CT images showed that the shape of plant roots varied with depth and that the volume fraction of soil constituents varied depending on the plant roots' growth state and condition. In the future, we would like to conduct similar evaluations using a greater variety of plants and soils and assess the effects of the interaction between plant root shape and soil components on soil properties.