In the vast fields of soybean cultivation, breeders are constantly on the hunt for the next big yield booster. Traditional methods of identifying high-yield varieties are often labor-intensive and time-consuming, relying heavily on manual measurements. However, a groundbreaking study led by Chen Yang from the Research Institute of Quantitative Remote Sensing and Smart Agriculture at Henan Polytechnic University is revolutionizing this process. By harnessing the power of unmanned aerial vehicles (UAVs) and advanced clustering algorithms, Yang and his team have identified key phenotypic traits that could significantly enhance the efficiency and accuracy of soybean variety screening.
The study, published in the journal ‘Remote Sensing’ (translated from Chinese), utilized UAV-based and field phenotyping to collect data from a staggering 1923 and 1015 soybean breeding plots across two years. The goal was to pinpoint the critical traits that correlate with high yields, thereby streamlining the selection process for breeders. “The rapid development of high-throughput phenotyping provides new possibilities for soybean phenotypic research and excellent variety screening,” Yang explains. This technology allows for the collection and processing of phenotypic data from a large number of crops in a short time, making it a game-changer for large-scale variety screening.
The research employed a two-step clustering method using a self-organizing map (SOM) and K-means clustering to group the soybean varieties based on their phenotypic traits. This approach revealed that the duration of canopy coverage remaining above 90% (Tcc90) was a critical trait for selecting high-yield varieties. Additionally, high-yield soybean varieties typically exhibited rapid development of canopy coverage, prolonged duration of high canopy coverage, a delayed decline in canopy coverage, and moderate-to-high plant height and hundred-grain weight.
The implications of this research are vast, particularly for the energy sector, which relies heavily on soybean oil as a renewable energy source. By identifying these key traits, breeders can more efficiently select high-yield varieties, potentially increasing soybean production and, consequently, the availability of soybean oil for biofuel production. “This study provides useful insights into the key traits that are valuable for identifying high-yield soybean varieties from numerous testing materials,” Yang notes. The method successfully selected 87% and 72% of high-yield varieties in 2022 and 2023, respectively, demonstrating its potential to revolutionize soybean breeding programs.
The integration of UAV-based high-throughput phenotyping into soybean variety screening not only accelerates the breeding process but also enhances its precision. This technology allows breeders to monitor crops dynamically and non-destructively, providing a wealth of data that can be analyzed to identify the most promising varieties. As Yang’s research shows, the future of soybean breeding lies in the fusion of advanced technology and traditional agricultural practices. This approach could pave the way for more sustainable and efficient agricultural practices, benefiting not only soybean farmers but also the broader energy sector.