Goniometer in the air: Enabling BRDF measurement of crop canopies using a cable-suspended plant phenotyping platform

• Main Authors: Bai, G. (Author), Gamon, J.A (Author), Ge, Y. (Author), Leavitt, B. (Author), Scoby, D. (Author)
• Format: Article
• Language: English
• Published: Academic Press 2023
• Subjects: Backscattering; Band index; Bidirectional reflectance distribution functions; Cables; Canopy bidirectional reflectance distribution function; Canopy BRDF; Crops; Distribution functions; Field goniometer; Field goniometers; Goniometers; Infrared devices; Mazie; Plant phenotyping; Plant phenotyping platform; Radiative transfer; Reflection; Soybean; Vegetation; Vegetation index; VNIR spectrometer
• Online Access: View Fulltext in Publisher

 The Bidirectional Reflectance Distribution Function (BRDF) quantifies the distribution of the spectral reflectance of a target surface at various viewing and illumination angles. In-field measurement of the BRDF of vegetation canopies improves the characterization of off-nadir measurements and informs radiative transfer models of canopy reflectance, where the Lambertian assumption does not hold. However, current field goniometers are unable to measure BRDF efficiently, especially for tall vegetation across the growing season because of the limitations of clearance, field accessibility, and flexibility of the sensor field of view. In this study, we explored the potential of using a large-scale cable-suspended field phenotyping system to quantify BRDF of the canopy reflectance at selected bands and Vegetation Indices (VIs) of maize and soybean canopies. The system performance benefited from the following conditions: no crop damage, full-season field accessibility of tall canopies, automatic measurement, and accurate positioning. Correlation analysis shows that a strong correlation exists among the reflectance, VIs, and the Green Pixel Fraction. The hemispheric distributions of the spectral reflectance at selected bands and VIs were quantified at multiple dates. Hot spots were observed in the backscatter direction at visible and near-infrared (NIR) bands at the largest sensor zenith angle around the Solar Principal Plane (SPP). In contrast, cold spots of the Normalized Difference Vegetation Index (NDVI) and its related VIs were observed in the backscatter direction around solar zenith angles. The row effect was found for the maize canopy at the NIR band and the Near-infrared reflectance of vegetation (NIRv). NDVI had the lowest anisotropy index values among investigated bands and VIs. This system could be further leveraged to generate rapid and detailed BRDF data at a high spatiotemporal resolution using multiple sensors. © 2023 The Authors