Analyses and numerical evaluation of integrated time-series monitoring datasets including CO2 concentration and fluxes at controlled CO2 release site in South Korea

• Published in: Journal of hydrology (Amsterdam) 2020-11, Vol.590, p.125213, Article 125213
• Main Authors: Seo, Do Hyun, Han, Weon Shik, Park, Eungyu, Jeong, Jina, Oh, Yun-Yeong, Kim, Hyun-Jun, Yoo, Gayoung, Jun, Seong-Chun, Yun, Seong-Taek
• Format: Article
• Language: English
• Subjects: CO2 flux; CO2 mass; CO2 release; Meteorological; Soil moisture content
• ISSN: 0022-1694; 1879-2707
• DOI: 10.1016/j.jhydrol.2020.125213

•Weather, soil moisture, CO2 concentration and fluxes were monitored for 965 days in South Korea.•Heterogeneities of soils have little effect on simulated CO2 fluxes.•Increased soil moisture content results to high CO2 concentration in soil profile. In Eumseong, Republic of Korea, a controlled CO2 release site was designed by Korea CO2 Storage Environmental Management (K-COSEM) Research Center to evaluate properly integrated CO2 monitoring tools at the soil surface. CO2 was released from the 55 m-length PVC-pipe buried at 2.5 m-depth, and then, CO2 concentration and flux were monitored from August 1, 2015 to March 21, 2018 (965 days). Influencing environmental factors such as meteorological data, soil moisture and temperature at 5 different depths were monitored concurrently. The measured soil moisture content decreased rapidly when the soil temperature dropped below 0 °C. When CO2 release began, both surface CO2 flux and soil CO2 concentration increased with a time lag. However, with occurrence of precipitation events, CO2 flux decreased to almost 0 cm3 cm−2 day−1 but soil CO2 concentration increased rapidly. The HYDRUS-SOILCO2 predicted soil moisture, temperature and natural background CO2 flux compared to measured data (R2 = 0.74, 0.89, and 0.85). The 10 scenarios were constructed with different soil characteristic curves. Unlike soil moisture content, the simulated natural CO2 flux did not depend heavily on soil characteristics. The mass of released CO2 was estimated by the difference between the measured CO2 flux data during the release test and natural CO2 prediction model. At the 2nd CO2 release test, there was 1.5 times difference between the actual released CO2 mass and calculated CO2 mass.