Characteristics of CO2 and CH4 from different emission sources using mobile measurements and stable carbon isotope analysis

Achieving effective greenhouse gases (GHGs) mitigation policy requires accurate quantification of contribution from each emission source based on in-situ measurements. In this study, we investigated the spatial distribution of CO₂ and CH₄ emitted from different emission sources by conducting mobile measurements using a GLA331-GGA analyzer (ABB–LGR Inc.) mounted on a vehicle. We conducted seven mobile measurements in spring (N = 3), summer (N = 2), and fall (N = 2) over Seoul Metropolitan Area (SMA) in 2025. By comparing the correlation between two GHGs from various emission sources, we selected representative sites including livestock facilities (cattle and swine barns), industrial complexes, urban, wastewater treatment plants, LNG power plants, rural areas. Background GHGs concentrations were defined as the daily 5th percentile for each measurement day, and correlations between enhancements (ΔCO₂ and ΔCH₄) were assessed. Along with real time measurements, stable carbon isotopes samplings were also conducted to provide complementary constraints on concentration variability and the contributions of end-member of each emission source. For stable isotope measurements, two ambient air samples were collected per site using canisters (Entech, Simi Valley, CA, USA) and analyzed with Picarro G2131-i for δ¹³C–CO₂ (δ¹³C) and Picarro G2132-i for δ¹³C–CH₄ (δ¹³CH₄). Strong co-variability between the two GHGs was observed at several emission sources and seasons, including springtime cattle barns (R = 0.75), LNG power plants (R = 0.83), industrial complexes (R = 0.74), and swine barns (R = 0.64); summertime cattle barns (R = 0.66) and LNG power plants (R = 0.67); and fall industrial complexes (R = 0.70) and cattle barns (R = 0.97). These correlations suggested that CO₂ and CH₄ were likely emitted concurrently from shared sources or similar emission activities in SMA region. The observed δ¹³C values ranged from −8.2‰ to −12.5‰, while δ¹³CH₄ ranged from −47.2‰ to −48.6‰. Seasonal mean δ¹³C values were −11.2‰ in spring, −9.2‰ in summer, and −10.1‰ in fall, consistent with a summertime influence from enhanced biospheric respiration, with the most depleted values occurring in spring. In contrast, δ¹³CH₄ exhibited relatively small seasonal variability, as indicated by the coefficient of variation (sd/mean; 0.004 in spring, 0.013 in summer, and 0.012 in fall), but still provided useful constraints on source attribution. In addition, a Bayesian isotope mixing model (the ‘simmr’ package in R) was applied to quantify relative source contributions indicating that coal combustion contributed most strongly to δ¹³C, whereas wastewater treatment and natural gas were the dominant contributors to δ¹³CH₄.