Leaf chlorophyll content (LCC) and canopy chlorophyll content (CCC) are widely used indicators of photosynthetic capacity, nitrogen status, and overall plant health. Accurate monitoring of chlorophyll content in apple leaves and canopies is essential for optimizing orchard management and advancing smart agriculture. Traditional chemical assays to measure chlorophyll are destructive and impractical for large-scale application. Remote sensing techniques, especially vegetation indices (VIs), have emerged as alternatives, yet their accuracy is highly sensitive to shadows within canopies. While satellite-based methods often ignore shadow effects due to coarse resolution, UAVs provide centimeter-level imagery where shadows become prominent and problematic. Conventional approaches attempt to mask shadowed pixels, but this introduces new uncertainties. Therefore, developing shadow-resistant methods is a key step toward reliable chlorophyll monitoring in modern precision orchards.
A study (DOI: 10.1016/j.plaphe.2025.100015) published in Plant Phenomics on 6 March 2025 by Hao Yang’s team, Beijing Forestry University, establishes a shadow-resistant UAV and modeling framework that enables accurate chlorophyll monitoring of individual apple trees, advancing precision management in modern smart orchards.
In this study, researchers combined three-dimensional radiative transfer modeling (3D RTM), UAV-based multispectral imagery, and Gaussian process regression (GPR) to overcome the effects of canopy shadows on chlorophyll estimation in apple orchards. First, the accuracy of simulated canopy reflectance was validated against field measurements, showing nearly identical spectral curves (R² = 0.999, RMSE = 0.03), confirming the reliability of the simulated spectra for subsequent analyses. Simulations revealed that the proportion of canopy shadows followed a parabolic pattern over the day, with the lowest shadow fraction and highest photon escape probability at noon, making midday the optimal time for UAV data collection. Spectral analysis further demonstrated that visible band reflectance peaked when shadows were minimal, whereas near-infrared reflectance exhibited the opposite trend. Next, the team assessed the resistance of vegetation indices (VIs) to shadow interference, ranking them by robustness. Five indices—NDVI-RE, Cire, Cigreen, TVI, and GNDVI—were identified as least sensitive to shadow variations, and four of them (NDVI-RE, Cire, Cigreen, and TVI) were sufficient to achieve the highest inversion accuracy. Comparisons between simulated and UAV-measured VIs confirmed consistency, validating the use of simulated datasets to support modeling. Based on these insights, a hybrid inversion model was developed that integrated selected VIs with GPR. The model performed well, with LCC estimates reaching R² = 0.78 and RMSE of 6.86 μg/cm², and CCC estimates achieving R² = 0.78 and RMSE of 32.33 μg/cm². Importantly, the model provided comparable or superior results to traditional shadow-masking approaches while avoiding their limitations. Finally, mapping across two orchards (429 and 215 trees) revealed significant spatial heterogeneity in both LCC and CCC, with values showing normal distributions, underscoring the method’s ability to capture fine-scale variability in canopy chlorophyll content for individual apple trees.
This shadow-resistant inversion method has immediate applications in smart orchard management. By providing accurate, non-destructive chlorophyll data at the individual tree level, growers can detect nutrient deficiencies, monitor stress, and optimize fertilizer use with unprecedented precision. The approach reduces dependence on destructive sampling and overcomes the limitations of masking algorithms, offering a scalable solution for large orchards. Moreover, mapping CCC and LCC distributions enables managers to visualize variability within orchards, supporting site-specific interventions that improve productivity, sustainability, and resource efficiency.
###
References
DOI
10.1016/j.plaphe.2025.100015
Original URL
https://doi.org/10.1016/j.plaphe.2025.100015
Funding information
This research was funded by the Natural Science Foundation of China (42171303, 42371373) and the Special Fund for Construction of Scientific and Technological Innovation Ability of Beijing Academy of Agriculture and Forestry Sciences (KJCX20230434).
About Plant Phenomics
Science Partner Journal Plant Phenomics is an online-only Open Access journal published in affiliation with the State Key Laboratory of Crop Genetics & Germplasm Enhancement, Nanjing Agricultural University (NAU) and distributed by the American Association for the Advancement of Science (AAAS). Like all partners participating in the Science Partner Journal program, Plant Phenomics is editorially independent from the Science family of journals. Editorial decisions and scientific activities pursued by the journal's Editorial Board are made independently, based on scientific merit and adhering to the highest standards for accurate and ethical promotion of science. These decisions and activities are in no way influenced by the financial support of NAU, NAU administration, or any other institutions and sponsors. The Editorial Board is solely responsible for all content published in the journal. To learn more about the Science Partner Journal program, visit the SPJ program homepage.