Naorem et al. (2026) Estimation of soil organic carbon under rice-fallow system using Sentinel-2 derivatives

Identification

• Journal: Plant Science Today
• Year: 2026
• Date: 2026-09-28
• Authors: Janaki Singh Naorem, D Priya, Deepak Kumar Dwivedi, J S Rajkumar, Tamang Basant
• DOI: 10.14719/pst.11182

Research Groups

• School of Natural Resource Management, College of Post Graduate Studies in Agricultural Sciences, Central Agricultural University (Imphal), Umiam, Ri Bhoi, Shillong 793 103, Meghalaya, India
• School of Social Sciences, College of Post Graduate Studies in Agricultural Sciences, Central Agricultural University (Imphal), Umiam, Ri Bhoi, Shillong 793 103, Meghalaya, India

Short Summary

This study aimed to develop a suitable regression model for estimating soil organic carbon (SOC) in traditional rice-fallow systems across various hill slopes in northeastern India using Sentinel-2 derived remote sensing indices. It identified green-NIR-based indices as the most suitable for SOC prediction in this fragile topographical region.

Objective

• To develop a suitable regression model for estimating soil organic carbon (SOC) in century-old traditional rice-fallow systems across different hill slopes in northeastern India in the Eastern Himalaya using remote sensing indices derived from Sentinel-2.

Study Configuration

• Spatial Scale: Bhoirymbong, Meghalaya, NE India in the Eastern Himalaya. Soil samples were collected from a depth of 0–0.15 m, with composite samples prepared from 10 randomly collected soil samples within a 10 m × 10 m area.
• Temporal Scale: Soil sampling was conducted during November 2020, with nearly synchronized, cloud-free (< 10 % cloud cover) Sentinel-2 data accessed for the same period.

Methodology and Data

• Models used: Regression models were developed to predict SOC using remote sensing indices.
• Data sources:
  • 100 composite soil samples collected from rice-fallow areas on various slopes (nearly level: 0–3 %, gentle: 3–8 %, moderate: 8–15 %, steep: 15–30 %, and very steep: > 30 %).
  • Sentinel-2 satellite imagery accessed from the US Geological Survey (USGS) official website, used to derive 15 remote sensing indices.

Main Results

• The brightness index (BI), BI2, and the soil bareness index (SBI) were identified as the most influential indices for predicting SOC.
• SOC values ranged from 1.17 % (mass fraction) in rolling topography (10–15 % slope) to 2.86 % (mass fraction) in gently undulating topography (2–5 % slope).
• After rice crop harvesting, red-NIR-based indices showed no significant difference, while green-NIR-based indices were significantly different (p < 0.05).
• Green-NIR-based indices were concluded to be the most suitable for SOC prediction in the study area.

Contributions

This study provides a viable remote sensing-based solution for evaluating soil organic carbon in challenging, fragile topographies of North-east India, where conventional methods face difficulties. It specifically identifies effective Sentinel-2 derived indices (green-NIR-based) for SOC prediction in traditional rice-fallow systems, offering a practical tool for monitoring and managing soil health in the Eastern Himalaya.

Funding

Not explicitly mentioned in the provided text.

Citation

@article{Naorem2026Estimation,
  author = {Naorem, Janaki Singh and Priya, D and Dwivedi, Deepak Kumar and Rajkumar, J S and Basant, Tamang},
  title = {Estimation of soil organic carbon under rice-fallow system using Sentinel-2 derivatives},
  journal = {Plant Science Today},
  year = {2026},
  doi = {10.14719/pst.11182},
  url = {https://doi.org/10.14719/pst.11182}
}