This study aims to evaluate the long-range transport of CO2 in East Asian region, using concentration data in a surface measurement site (Gosan Station), column averaged concentration data of satellite-borne instrument (GOSAT), and GEOS-Chem modeling results for the period of June 2009 to May 2011. We perform a validation of the data from GOSAT and GEOS-Chem with total column observations (TCCON). The analysis of the long-range transport and high concentration (HC) events using surface/satellite observations and modeling results is conducted. During the HC events, the concentrations in CO2 and other air pollutants such as SO2 and CO are higher than that of all episodes. It means that CO2, known as a globally well-mixed gas, may also act as a fingerprint of human activity with unique regional characteristics like other air pollutants. This comprehensive analysis, in particular with GOSAT CO2 observation data, shows that CO2 plume with high concentration can be long-range transported with 1-2 days’ duration with regional scale. We can find out with GEOS-Chem tagging simulation that more than 45% of the elevated CO2 concentration over central/eastern China, Korea, and Japan on high concentration days can be explained by emission sources of East Asia mainland.
Since the industrial revolution, the current level of CO2 emitted into the atmosphere has increased by nearly 41% because of fossil fuel combustion. This contributes to a higher global warming potential (approximately 82%) as compared to other greenhouse gases, such as CH4 and N2O [
The atmospheric CO2 concentration is determined by the average concentration and its short-term variation mainly due to the long-range transport and meteorological conditions [
In order to investigate the impact of global/regional carbon cycle on the regional change of CO2 concentration and air quality [
Although CO2 is a globally well-mixed gas in a climatological sense, it has also the unique regional characteristics over the source and downwind area because of the same anthropogenic emission sources driven by human socioeconomic activity like other air pollutants (e.g., CO, SO2, and aerosols), and few studies for long-range transport of CO2 and its high concentration events have been conducted at East Asia where it is one of the most hot spots of CO2 emission in the world.
In this study, we investigate CO2 long-range transport phenomena and high concentration events on the downwind regions of the Asian continent and provide observational evidences showing regionally the same behavior of CO2 as other air pollutants over East Asia, using all the available data, CO2 and other air pollutants concentration data observed at surface monitoring sites, satellite data, and results of a chemical transport model.
This study focused on the CO2 concentration data available for the period of June 2009 to May 2011, which was observed by the Greenhouse gases Observing SATellite (GOSAT) and surface monitoring stations. Surface observation data were collected from the Gosan site in Korea (
Hourly averaged CO2 concentration and other air pollutants (SO2, NO2, CO, O3, PM10, and PM2.5) concentration data have been measured at Gosan Station in Korea. Gosan Station is located at the west end of Jeju Island, Korea (33°17′N, 126°10′E) (Figure
The location of Gosan site, Jeju Island in Korea.
The CO2 sampling from a 10 m tower is automatically analyzed every 30 seconds, through a Nondispersive Infrared Analyzer (NDIR). All CO2 data are reported using the WMO CO2 mole fraction scale. We perform a statistical analysis after QA/QC (Quality Assurance/Quality Control) processes to determine whether it is a clean or a polluted atmosphere [
Also, in order to make direct comparison of GOSAT data and GEOS-Chem model results with surface measurement data, we used TCCON’s column-averaged concentration. TCCON is a network of ground-based Fourier transform spectrometers that measure column-averaged concentrations of CO2, CO, CH4, and H2O in the atmosphere. For this study, the TCCON data were selected at each site, within about ±1 hour of the GOSAT overpass time. Also, the GOSAT and GEOS-Chem data were selected within about ±1 degree of the area centered at each TCCON site.
GOSAT is a sun-synchronous satellite, which provides a global coverage in three days and crosses the equator at about 13:00 local time. Its swath width is 790 km and its spatial resolution at nadir is 10.5 km. GOSAT’s wavelength ranges between 0.76 and 14.3
We have used version 9-02-01 of GEOS-Chem, a global 3D atmospheric chemistry transport model developed by the National Aeronautics and Space Administration (NASA) and Harvard University. It is driven by assimilated meteorology from the NASA Global Modeling Assimilation Office (GMAO) and contains approximately 300 photochemical reaction mechanisms of the O3-
In this study, we used a globally uniform CO2 field of 375 ppmv as an initial condition, which was measured on January 1, 2004. This simulation is performed for the target period of June 2009 to May 2011 using the GEOS-5 meteorological field with a horizontal resolution of
The CO2 concentration (
In addition, GEOS-Chem can carry out tagged simulations for CO, O
It is difficult to show the long-range transport phenomena of CO2 in the regional scale because its lifetime is very long enough to be well-mixed even globally. Moreover timely continuous CO2 measurements with spatially wide coverage are very restricted so far. So, to distinguish each high concentration event from measurements of long-term period and to analyze the composite field of them are an easy way to identify the long-range transport phenomena.
As mentioned above, Gosan surface measurement site observed both CO2 and other air pollutants simultaneously and its geographical location is almost a middle zone of Korean Peninsula, China, and Japan. So, this site is one of the best stations for analyzing the characteristics of a long-range transport that occurred in the East Asia regions. The observed CO2, SO2, and CO concentrations of Gosan site for the period of June 2009 to May 2011 were used. SO2 and CO are well-known tracers of long-range transport, which are mostly emitted by anthropogenic activities and their lifetime (several weeks to months) is long enough to trace air pollutant plumes at the intercontinental scale [
In addition, high concentration (HC) events were defined as CO2 concentration at Gosan Station was higher than the mean + 1
Compared to TCCON’s column-averaged concentration, GOSAT concentrations underestimated nearly by
Comparison of the column-averaged CO2 concentration (a) between TCCON and GOSAT, (b) between TCCON and GEOS-Chem, and (c) between GOSAT and GEOS-Chem.
TCCON versus GOSAT
TCCON versus GEOS-Chem
GOSAT versus GEOS-Chem
Figure
Seasonal variation and global distribution of the column-averaged CO2 concentration (a) measured by GOSAT SWIR and (b) simulated by GEOS-Chem. MAM denotes the period from March to May, JJA from June to August, SON from September to November, and DJF from December to February.
GOSAT
GEOS-Chem
Nevertheless, the GOSAT and GEOS-Chem data showed very similar patterns of the seasonal variation and spatial distributions. The seasonal variation for both data is obvious that CO2 decreases in summer because of photosynthesis by vegetation while it increases in winter because of the increased usage of fossil fuel and decreased vegetation. The amplitude of the CO2 seasonal cycle is higher in the northern hemisphere than in the southern hemisphere. Also, the spatial distributions represent that CO2 concentration is higher in the northern hemisphere than in the southern hemisphere because of the higher population and greater number of industrial activities. In addition, the spatial distribution of CO2 concentration showed the possibility of a long-range transport of CO2 because relatively high concentration was detected over an ocean region far from the land source. For spring and winter period of northern hemisphere (MAM and DJF), high concentration plume of CO2 emitted from the Asian continent was expanded down to North Pacific, and for spring period of southern hemisphere (SON), the CO2 emitted from South America and Africa has been expanded to the Atlantic and the Indian Ocean.
Through correlation analysis and comparison of spatial distribution among the comprehensive dataset, TCCON, GOSAT, and GEOS-Chem, we find that, at least in East Asian region, GOSAT and GEOS-Chem data are comparable to each other in temporal/spatial distribution with well-known differences with TCCON data, so that it might be reasonable to use those data for this study.
Annual averaged CO2 concentration observed at Gosan Station in 2010 was 398.3 ppmv, which is 13.1 ppmv higher than the global average CO2 concentration of 389.0 ppmv [
Table
Daily averaged surface concentrations of CO2 and air pollutants on HC, HC − 1, and HC + 1 case and their occurrences.
CO2 |
SO2 |
CO |
O3 |
NO2 |
PM10 |
PM2.5 |
Number of days | |
---|---|---|---|---|---|---|---|---|
All | 395.9 | 2.9 | 0.7 | 43.5 | 3.9 | 49.1 | 27.0 | 606 |
|
||||||||
HC − 1 | 398.3 | 2.8 | 1.0 | 44.9 | 3.8 | 48.4 | 22.8 | 25 |
HC | 403.1 | 5.0 | 1.2 | 47.7 | 5.5 | 68.1 | 40.3 | 67 |
HC + 1 | 398.0 | 2.8 | 1.0 | 46.9 | 3.9 | 55.2 | 25.5 | 25 |
Backward trajectory and forward trajectory analyses (Figure
(a) Backward trajectories and (b) forward trajectories at 1 km altitude for the cases of high concentration (HC) case.
Backward trajectory
Forward trajectory
Figure
Monthly distribution in number of HC days classified from CO2 concentration data observed at Gosan Station.
In this section, further analysis on the long-range transport of high concentration CO2 was conducted using the GOSAT data and the GEOS-Chem modeling results, in order to expand its spatial coverage.
Figure
Difference in CO2 concentration between HC and all periods of the East Asian region using GOSAT SWIR. Region I (30–60°N, 110–120°E) covers the eastern region of China, and Region II (30–60°N, 120–130°E) covers the Yellow Sea and the Korean Peninsula, while Region III (30–60°N, 130–140°E) represents Japan.
We focused on 3 subregions (Region I, Region II, and Region III) for providing some observational evidence of long-range transport of high concentration CO2 over downwind area from Asian continent; Region I (30–60°N, 110–120°E) covers the eastern region of China, and Region II (30–60°N, 120–130°E) covers the Yellow Sea and the Korean Peninsula, while Region III (30–60°N, 130–140°E) represents Japan.
Figure
Regional mean of the column-averaged CO2 concentration and its standard deviation on HC − 1, HC, and HC + 1 and for all periods (All) using GOSAT SWIR. (a) Region I (30–60°N, 110–120°E), (b) Region II (30–60°N, 120–130°E), and (c) Region III (30–60°N, 130–140°E).
Region I
Region II
Region III
Figure
Distribution of the column-averaged CO2 concentrations for all periods (All) and high concentration cases (HC − 1, HC, and HC + 1) simulated by GEOS-Chem.
In order to clarify the source apportionment on high CO2 concentration days, we performed the so-called tagging simulation [
(a) Differences of the column-averaged CO2 concentrations between high concentration cases (HC − 1, HC, and HC + 1) and all periods (All) simulated by GEOS-Chem. (b) and (c) represent differences of column-averaged CO2 concentration contributed by emission from East Asia mainland and South and Central Europe by tagged simulation, respectively.
The contribution of sources and sinks to atmospheric CO2 concentration of the East Asian regions simulated by GEOS-Chem tagged simulation.
It is found that there are two major contributions to East Asia region for high concentration events from East Asia mainland and South/Central Europe. Over central and eastern China, Korea, and Japan, almost 45% of the contribution to China, Korea, and Japan inland area can be explained by emission sources of East Asia mainland (i.e., central/eastern China) and almost 20% to northern China by European emission sources out of the analysis domain (Figures
In this study, the ground-based data, the satellite-borne instrument (GOSAT), and the chemical transport model (GEOS-Chem) are used to investigate the long-range transport phenomena for high CO2 concentration plume in Asian continent and the downwind regions for the period of June 2009 to May 2011. We also evaluate GOSAT data and GEOS-Chem results with TCCON data to show the reliability of the derived data. Those have very similar patterns of the seasonal variation and spatial distribution and have well-known differences with TCCON data, so that it could be reasonable to use GOSAT and GEOS-Chem data for this study.
The analysis of the long-range transport and high concentration event using observation data in a surface site, the Gosan Station in Korea, shows that the ratio of high concentration days to all periods is 11.1% with CO2 concentration of 403.1 ppmv and 7.2 ppmv increases compared to the average CO2 concentration (395.9 ppmv) in all periods. We can find out from backward/forward trajectory analysis that the air originating from/out of Asian continent may be transported through the polluted area in China and flowed into Korea and Japan.
The column-averaged concentration data by GOSAT and GEOS-Chem were analyzed to expand the spatial distribution of the long-range transport of high concentration CO2 plume. With GOSAT data, concentration (389.4 ppmv) on HC days is 3.6 ppmv higher than that of all periods. It is obvious that high concentrations are also dominating in central and eastern China and Japan as a surface site, the Gosan Station in Korea, experiences high concentration events and CO2 plume with high concentration is prevailing over those areas through long-range transport and mixing processes like other air pollutants. Also, with subregional analysis, it is found that CO2 plume with high concentration can be long-range transported with 1-2 days’ duration with regional scale.
With tagging simulation using GEOS-Chem modeling, we can explain that more than 45% of the elevated CO2 concentration over central and eastern China, Korea, and Japan on HC days may be attributed to emission sources of East Asia mainland, while almost 20% over northern China may be attributed to European emission sources.
In this study, we can understand that CO2 which is known as a globally well-mixed gas may also act as a fingerprint of human activity with regional characteristics like other air pollutants. This comprehensive analysis with all of available data, although observational evidences are still quite few, would help better understand the long-range transport of CO2 and its impact on climate change and the carbon cycle in the Asian regions.
The authors declare that there is no conflict of interests regarding the publication of this paper.
The authors thank JAXA, NIES, and MOE for the GOSAT data and their continuous support as part of the Joint Research Agreement. TCCON data was obtained from the TCCON Data Archive, operated by the California Institute of Technology from the Website at