Work

Adaptive background suppression-based plume identification, emission estimation, and ground validation of coal mine methane in Changzhi, Shanxi, China

Year: 2026

Type: article

Abstract: This study developed a coal mine methane remote-sensing monitoring framework centered on adaptive background suppression using high-spatial-resolution multispectral data from Sentinel-2. The framework integrates methane column enhancement retrieval, automated plume detection, and emission flux estimation. Specifically, bilateral filtering and the multi-band single-pass (MBSP) method were used to retrieve methane column enhancements; a methane plume detection algorithm combining NDVI-based stratification, adaptive segmentation, morphological processing, and wind-field constraints was developed; and coal mine methane emission fluxes were quantified using the integrated mass enhancement (IME) method. Application to a pilot coal enterprise in Changzhi during 2024 demonstrated that the proposed framework effectively reduced weak vegetation-background effects and heterogeneous high-intensity interference while preserving plume topological consistency. Sentinel-2 also provided an observation frequency more than eight times higher than that of EMIT. The mean satellite-derived emission rate for the pilot enterprise was 1023.57 ± 266.84 kg ⋅ h − 1 , with a mean relative deviation of 14.47% from ground-based measurements. The relatively large deviations and daily-scale fluctuations observed during some periods were mainly attributed to uncertainties associated with the spectral resolution of Sentinel-2, surface heterogeneity, algorithm parameter settings, and ERA5 wind-field data. A comparison with the EDGAR inventory revealed broad consistency in regional spatial emission patterns. However, the higher-resolution Sentinel-2 data detected potential coal-related emission signals in several grid cells absent from EDGAR, accounting for 22.2% of the potential emission grids identified in Changzhi. On a regional scale, EDGAR’s annual emission estimate for Changzhi was 32.11% higher than the Sentinel-2 annualized estimate derived in this study.

Source: International Journal of Applied Earth Observation and Geoinformation

DOI: 10.1016/j.jag.2026.105587

Authors: Xuan Wu, Shenshen Li, Qiang Ge, Xuefei Hu, Kun Cai

Institutions: Henan University, Chinese Academy of Sciences, Aerospace Information Research Institute, Sun Yat-sen University

Language: English

Topic: Atmospheric and Environmental Gas Dynamics

Subfield: Global and Planetary Change

Field: Environmental Science

Domain: Physical Sciences

SDG: Climate action

Open Access status: gold

Estimated APC paid: $3,550