Search

The CarbonTracker (CT) model has been used in previous studies for understanding and predicting the sources, sinks, and dynamics that govern the distribution of atmospheric CO2 at varying ranges of spatial and temporal scales. However, there are still challenges for reproducing accurate model-simulated CO2 concentrations close to the surface, typically associated with high spatial heterogeneity and land cover. In the present study, we evaluated the performance of nested-grid CT model simulations of CO2 based on the CT2016 version through comparison with in-situ observations over East Asia covering the period 2009–13. We selected sites located in coastal, remote, inland, and mountain areas. The results are presented at diurnal and seasonal time periods. At target stations, model agreement with in-situ observations was varied in capturing the diurnal cycle. Overall, biases were less than 6.3 ppm on an all-hourly mean basis, and this was further reduced to a maximum of 4.6 ppm when considering only the daytime. For instance, at Anmyeondo, a small bias was obtained in winter, on the order of 0.2 ppm. The model revealed a diurnal amplitude of CO2 that was nearly flat in winter at Gosan and Anmyeondo stations, while slightly overestimated in the summertime. The model’s performance in reproducing the diurnal cycle remains a challenge and requires improvement. The model showed better agreement with the observations in capturing the seasonal variations of CO2 during daytime at most sites, with a correlation coefficient ranging from 0.70 to 0.99. Also, model biases were within −0.3 and 1.3 ppm, except for inland stations (7.7 ppm).

Since the late 1990s, the meteorological observatory established in Anmyeondo (36.5382˚ N, 126.3311˚ E, and 30 m above mean sea level), has been monitoring several greenhouse gases such as CO2, CH4, N2O, CFCs, and SF6, as part of the Global Atmosphere Watch (GAW) Program. A high resolution ground-based (g-b) Fourier Transform Spectrometer (FTS, IFS-125HR model) was installed at such observation site in 2013, and has been fully operated within the frame work of the Total Carbon Column Observing Network (TCCON) since August, 2014. The solar spectra recorded by the g-b FTS are covered in the range between 3,800 and 16,000 cm−1 at the spectral resolution of 0.02 cm−1 during the measurement period between 2013 and 2016. In this work, the GGG2014 version of the TCCON standard retrieval algorithm was used to retrieve XCO2 concentrations from the FTS spectra. Two spectral bands (at 6220.0 and 6339.5 cm−1 center wavenumbers) were used to derive the XCO2 concentration within the spectral residual of +0.01 %. All sources of errors were thoroughly analyzed. In this paper, we introduced a new home made OASIS (Operational Automatic System for Intensity of Sunray) system to our g-b FTS instrument and that allows reducing the solar intensity variations (SIV) below 2 %. A comparison of the XCO2 concentration in g-b FTS and OCO-2 (Orbiting Carbon Observatory) satellite observations were presented only for the measurement period between 2014 and 2015. Nine coincident observations were selected on a daily mean basis. It was obtained that OCO-2 exhibited low bias with respect to the g-b FTS, which is about −0.065 ppm with the standard deviation of 1.66 ppm, and revealed a strong correlation (R = 0.85). Based on seasonal cycle comparisons, both instruments were generally agreed in capturing seasonal variations of the target species with its maximum and minimum levels in spring and late summer respectively. In the future, it is planned to exert further works in utilizing the FTS measurements for the evaluation of satellite observations such as Greenhouse Gases Observing Satellite (GOSAT) at observation sites. This is the first report of the g-b FTS observations of XCO2 species over the Anmyeondo station.