News Release

Enhancing intercropping efficiency: A UAV-based model for precise quantification of shading impact in maize-soybean systems

Peer-Reviewed Publication

Plant Phenomics

Fig. 2


The workflow for high-throughput estimation of canopy height differences between soybean strip and the adjacent maize strips, including UAV image acquisition and processing (1), the establishment of canopy height model (CHM) (2), and the calculation of Hms (3). UAV image acquisition and processing were composed of UAV campaigns (A) and strip segmentations on UAV-derived digital ortho-mosaic (B) and digital surface model (DSM; C). The establishment of a CHM included the calculation of the difference between DSM and the reference ground (D) and the removal of soil and vegetative pixels of the other adjacent crop (E) from the CHM. Lastly, the strips of maize (G) and soybean (H) were separated by the threshold (F), and the canopy heights were then estimated from the corresponding CHMs, followed by the calculation of Hms.

view more 

Credit: Plant Phenomics

Intercropping, the simultaneous cultivation of multiple crops, is beneficial for maximizing resource use but poses challenges due to the shading of lower crops by higher ones. This shading, quantified as Cumulative Shading Capacity (CSC), significantly affects light interception and crop efficiency but remains poorly understood. Previous research mainly has focused on light distribution and crop responses without directly quantifying CSC and its variability across rows. Despite attempts using geometrical and statistical models, accurate, direct CSC quantification and understanding inter-row heterogeneity remain elusive.

In November 2023, Plant Phenomics published a research article entitled by “Quantification of the cumulative shading capacity in a maize-soybean intercropping system using unmanned aerial vehicle”.

In this study, researchers utilized unmanned aerial vehicles (UAVs) to measure canopy height differences and solar position, leading to the development of a Shading Capacity Model (SCM) for quantifying the shading impact on soybeans in maize-soybean intercropping systems. The SCM demonstrated that the southernmost row of soybean consistently experienced the highest shading proportion, with significant variations across strip configurations and plant densities. The maximum cumulative shading capacity (CSC) observed was 123.77 MJ m-2. A quantitative relationship was established between CSC and soybean canopy height increment, indicating that as the canopy height increased, so did the CSC. The model also assessed the effects of canopy height difference, latitude, and planting direction on shading, finding that greater height differences and latitudes resulted in increased shading, while a planting direction between 90° to 120° minimized it.

The study found that the shading distance of maize on adjacent soybean rows changed over growth stages, influenced by canopy height and solar trajectory. Detailed analysis revealed that the first row of soybean typically had the largest shading proportion, with the amount of shading differing significantly among different intercropping treatments and configurations. Overall, treatments with smaller distances between maize and soybean (Dms) experienced greater shading. Furthermore, a positive correlation was identified between the canopy height increment of soybeans and CSC, implying that taller soybeans encountered increased shading. Finally, the study examined how different factors influenced the overall shading proportion of the soybean strip. It was noted that the canopy height difference, latitude, and planting direction significantly affected the shading proportion, with certain directions and configurations minimizing shading effects.

Overall, this study is valuable for optimizing intercropping patterns by adjusting factors like planting configurations and directions, indicating potential for future applications in enhancing crop growth, light competition, and microclimate management on a larger scale through integration with crop models or remote sensing data.




Min Li1, Pengcheng Hu2,3, Di He3, Bangyou Zheng4, Yan Guo1, Yushan Wu5, and Tao Duan6*


1College of Land Science and Technology, China Agricultural University, Beijing, China.

2School of Agriculture and Food Sustainability, The University of Queensland, St Lucia, QLD, Australia.

3Agriculture and Food, CSIRO, GPO Box 1700, Canberra ACT 2601, ACT, Australia.

4Agriculture and Food, CSIRO, Queensland Biosciences Precinct, St Lucia, QLD, Australia.

5College of Agronomy, Sichuan Agricultural University, Chengdu, China.

6Institute of Microelectronics of Chinese Academy of Sciences, Beijing, China.

Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.