Lee et al. (2025) Development of an automated hyperspectral system to monitor hyperspectral reflectance and sun-induced chlorophyll fluorescence with directional and hemispheric view geometries simultaneously

Identification

• Journal: Remote Sensing of Environment
• Year: 2025
• Date: 2025-10-31
• Authors: Jeongho Lee, Youngryel Ryu, Jongmin Kim, Benjamin Dechant, Sangjun Lee, Joongbin Lim
• DOI: 10.1016/j.rse.2025.115100

Research Groups

• Interdisciplinary Program in Agricultural and Forest Meteorology, Seoul National University, Seoul, Republic of Korea
• Department of Landscape Architecture and Rural Systems Engineering, Seoul National University, Seoul, Republic of Korea
• Research Institute for Agricultural and Life Sciences, Seoul National University, Seoul, Republic of Korea
• SNU Energy Initiative, Seoul National University, Seoul, Republic of Korea
• Department of Smart Farm Science, Kyung Hee University, Yongin, Republic of Korea
• German Centre for Integrative Biodiversity Research (iDiv) Halle-Jena-Leipzig, Leipzig, Germany
• Leipzig University, Leipzig, Germany
• National Forest Satellite Information & Technology Center, National Institute of Forest Science, Seoul, Republic of Korea

Short Summary

This study develops and evaluates an automated hyperspectral system (enhanced RotaPrism) for simultaneous monitoring of hyperspectral reflectance and sun-induced chlorophyll fluorescence (SIF) using both directional (conical) and hemispherical view geometries. It demonstrates the system's reliability and reveals significant, seasonally variable differences in spectral reflectance and SIF between these viewing geometries across rice paddy and mixed forest ecosystems.

Objective

• To develop an automated hyperspectral system for simultaneous measurements of surface reflectance and SIF with two geometric view configurations.
• To evaluate the instrumental reliability and uncertainty of the system, including from long-term measurements in the field.
• To examine how different viewing configurations influence spectral reflectance, SIF, and vegetation indices across different ecosystems and temporal scales.

Study Configuration

• Spatial Scale:
  • Rice paddy site (Cheorwon, South Korea):
    • Conical configuration: 1.76 meters diameter at ground level, 1.42 meters at peak canopy height (0.8 meters).
    • Hemispherical configuration: 31 meters diameter at ground level, 25 meters at peak canopy height.
  • Mixed forest site (Jeju Island, South Korea):
    • Conical configuration: 11 meters diameter at ground level, 4.4 meters at approximate canopy height (15 meters).
    • Hemispherical configuration: 137.7 meters diameter at ground level, 55.1 meters at canopy height.
• Temporal Scale:
  • Rice paddy site: Throughout the 2024 growing season (March to October).
  • Mixed forest site: April to November 2022.
  • Continuous data acquisition from 05:30 to 20:30 local time, with measurements taken every two minutes.

Methodology and Data

• Models used:
  • Singular Vector Decomposition (SVD) method for SIF retrieval.
  • 3FLD, iFLD, and SFM methods for SIF retrieval (for comparison).
  • Third-order polynomial for wavelength calibration.
  • Gaussian Convolution for spectral resolution matching between spectrometers.
• Data sources:
  • Proximal remote sensing: Enhanced RotaPrism system integrating two spectrometers.
  • Spectrometers: QEPro-CUS (730-786 nanometers, 0.15-0.17 nanometers FWHM for SIF) and HR2000+ES (200-1100 nanometers, 0.44-0.52 nanometers FWHM for Visible and Near-Infrared reflectance).
  • Reference spectrometer: ASD FieldSpec4 for cross-calibration.
  • Integrating sphere (LabSphere Inc.) with standard light source (SCL-050) for indoor radiometric calibration.
  • Phenocam images for visual phenological tracking.
  • DHT11 temperature and humidity modules for environmental monitoring.
  • Mercury Argon Wavelength Calibration Light Source (Ocean Insight, Model HG-1) for wavelength calibration.

Main Results

• The enhanced RotaPrism system demonstrated high reliability, with quality control pass rates exceeding 91% for detector saturation and 94% for temperature stability.
• Radiometric consistency between the two spectrometers showed mean relative differences of 10.4% for incoming irradiance, 8.8% for radiant exitance, and 5.1% for reflected radiance in the overlapping spectral range (735-780 nanometers).
• The system successfully captured distinct seasonal variations between bi-hemispherical reflectance (BHR) and hemispherical-conical reflectance factor (HCRF) in both ecosystems. Maximum relative differences between BHR and HCRF were wavelength-dependent and varied seasonally, with larger differences in the visible range (400-700 nanometers) and smaller differences in the near-infrared (700-900 nanometers) during full canopy development.
• Far-red SIF and vegetation indices exhibited distinct viewing-geometry-dependent patterns. In the rice paddy, SIFBH was consistently 15-20% higher than SIFHC during peak growing season, and a reversal in SIF and NDVI patterns (BH > HC to HC > BH) was observed after a data gap. In the mixed forest, SIFBH was higher (25-30%) during early/late growing season, while SIFHC was substantially higher (40-50%) during peak growing season.
• The relationship between SIFBH and SIFHC showed clear seasonal dynamics: the slope decreased from 1.35 in June to 0.87 in September in the rice paddy, while it increased from 0.69 in May to 1.20 in November in the mixed forest, reflecting the influence of canopy density and structural complexity.

Contributions

• Development of a novel automated hyperspectral system capable of simultaneous measurements of surface reflectance and sun-induced chlorophyll fluorescence (SIF) using both hemispherical and conical observation geometries.
• Provides direct, continuous, and long-term comparisons of viewing configuration effects on vegetation optical signals under identical environmental conditions, bridging a critical gap in proximal remote sensing.
• Enables unprecedented high-frequency (2-minute intervals) characterization of surface anisotropy through BHR/HCRF ratio measurements, crucial for understanding rapid vegetation structural dynamics and improving BRDF models.
• Offers essential ground-truth data for the validation and improvement of satellite-derived reflectance, albedo, and SIF products from current and upcoming hyperspectral missions (e.g., Landsat, Sentinel-2, EnMAP, DESIS, PACE, PRISMA, CHIME, FLEX).
• Advances the understanding of viewing configuration effects on ecosystem-scale SIF-Gross Primary Production (GPP) relationships and vegetation monitoring, particularly in heterogeneous ecosystems.
• Contributes to the establishment of a global network of standardized ground reference systems for hyperspectral remote sensing.

Funding

• National Institute of Forest Science (Project No. ‘FM0103-2021-01-2025’ and ‘FM103-2021-02-2025’)
• National Research Foundation of Korea (RS-2024-00348585)
• sDiv, the Synthesis Centre of the German Centre for Integrative Biodiversity Research (iDiv) Halle-Jena-Leipzig (DFG FZT 118, 202548816)

Citation

@article{Lee2025Development,
  author = {Lee, Jeongho and Ryu, Youngryel and Kim, Jongmin and Dechant, Benjamin and Lee, Sangjun and Lim, Joongbin},
  title = {Development of an automated hyperspectral system to monitor hyperspectral reflectance and sun-induced chlorophyll fluorescence with directional and hemispheric view geometries simultaneously},
  journal = {Remote Sensing of Environment},
  year = {2025},
  doi = {10.1016/j.rse.2025.115100},
  url = {https://doi.org/10.1016/j.rse.2025.115100}
}