Tuzzi et al. (2026) Estimation of Leaf Area Index and Vegetation Fractional Cover in SBG-TIR Configuration Using SCOPE Simulated Data and Sentinel-2 Images

⚠️ Warning: This summary was generated from the abstract only, as the full text was not available.

Identification

• Journal: Remote Sensing
• Year: 2026
• Date: 2026-06-11
• Authors: Luca Tuzzi, Sara Venafra, Roberto Colombo
• DOI: 10.3390/rs18121931

Research Groups

• NASA (National Aeronautics and Space Administration)
• ASI (Italian Space Agency)

Short Summary

This study evaluates machine learning approaches to retrieve Vegetation Fractional Cover (FC) and Leaf Area Index (LAI) using the limited VNIR bands of the upcoming SBG-TIR mission. The Gaussian Process Regression (GPR) model proved most effective, demonstrating high accuracy and strong agreement with Sentinel-2 biophysical products.

Objective

• To identify the optimal machine learning configuration and input variables for the accurate retrieval of FC and LAI from the VNIR and panchromatic observations of the SBG-TIR mission.

Study Configuration

• Spatial Scale: Global (mission design context); validation performed using Sentinel-2 data.
• Temporal Scale: Not specified.

Methodology and Data

• Models used: SCOPE (Soil Canopy Observation, Photochemistry and Energy Fluxes) radiative transfer model for synthetic data generation; Gaussian Process Regression (GPR) and other machine learning algorithms for retrieval.
• Data sources: Synthetic datasets (SCOPE), Sentinel-2 satellite observations, and GBOV (Ground-Based Observations for Validation) reference measurements.

Main Results

• The optimal configuration using Gaussian Process Regression (GPR) with a seven-channel input set achieved:
  • FC: $\text{RMSE} = 0.046$ and $R^2 > 0.9$.
  • LAI: $\text{RMSE} = 0.053\text{ m}^2/\text{m}^2$ and $R^2 > 0.9$.
• The proposed framework showed strong statistical agreement with the Biophysical Processor implemented in the ESA Sentinel Application Platform (SNAP) toolbox.

Contributions

• Provides a validated methodology for estimating biophysical parameters (FC and LAI) using a limited number of spectral bands, ensuring the synergistic use of VNIR and TIR data for the SBG-TIR mission.

Funding

• Not specified in the provided text.

Citation

@article{Tuzzi2026Estimation,
  author = {Tuzzi, Luca and Venafra, Sara and Colombo, Roberto},
  title = {Estimation of Leaf Area Index and Vegetation Fractional Cover in SBG-TIR Configuration Using SCOPE Simulated Data and Sentinel-2 Images},
  journal = {Remote Sensing},
  year = {2026},
  doi = {10.3390/rs18121931},
  url = {https://doi.org/10.3390/rs18121931}
}