Topics

In recent years, domestic fertilizer prices have surged due to global geopolitical instability and the depreciation of the Japanese yen. At the same time, there is growing international and domestic demand for reducing the environmental impact of agriculture to achieve a more sustainable society. In response to these societal challenges, the development of rice varieties better suited to cultivation with low-fertilizer inputs has become an urgent priority.

To develop rice varieties adapted for low-fertilizer cultivation, it is necessary to identify genotypes from underutilized genetic resources that can maintain yield under reduced fertilizer input, and to determine breeding traits associated with nutrient uptake. Among these breeding traits, root system architecture (RSA) is a key factor influencing nutrient uptake. However, when roots are excavated from paddy fields, their three-dimensional (3D) structure within the soil is disrupted, making accurate observation of RSA difficult. In addition, extracting and observing roots from flooded fields requires substantial labor and time. These limitations have long been a bottleneck in breeding rice varieties with improved RSA.

To overcome the above-mentioned challenges, NARO has developed a novel method for non-destructive observation of rice root systems using X-ray computed tomography (CT). In this method, soil blocks containing rice roots are carefully collected from the field to preserve their structure, and then they are scanned using X-ray CT to visualize the roots within the soil without disruption. Although X-ray CT images also capture non-root materials such as soil particles and straw, NARO has developed a new image processing algorithm that enables the extraction of root-specific data. This advancement allows for accurate, 3D observation of rice root systems cultivated in paddy fields without disturbing their natural shape. The image processing is fully automated, requiring only the collection of soil samples and approximately 10 minutes of CT scanning per sample. In contrast, conventional methods involving root excavation and washing typically allow a single person to process only one sample per day. For example, evaluating the root systems of 500 rice genotypes or breeding lines would require around 500 person-days using traditional methods. With the new CT-based method, the same evaluation can be completed in approximately 10 person-days, making root system phenotyping feasible within practical breeding workflows. NARO aims to apply this technology to accelerate the development of rice varieties that are adapted to low-fertilizer cultivation, thereby contributing to the realization of sustainable agriculture.