11:00 AM - 1:00 PM
[AAS11-P04] Analysis of enhancement ratios for greenhouse gas sources based on GOSAT-2 data
Keywords:GOSAT-2, carbon dioxide, methane, carbon monoxide
Column-averaged dry-air mole fractions of carbon dioxide, methane, and carbon monoxide (XCO2, XCH4, and XCO) have been retrieved from short-wavelength infrared spectra observed with TANSO-FTS-2 (Thermal and Near-infrared Sensor for Carbon Observation - Fourier Transform Spectrometer -2) on board GOSAT-2 (Greenhouse gases Observing SATellite -2). In this study, we have globally calculated enhancement ratio (ER) by a new unified method using GOSAT-2 XCO2, XCH4 and XCO data obtained in the same fields of views. ER is defined by a ratio of enhanced concentrations of two gases which are obtained by subtracting background concentrations from observed concentrations [Andreae and Merlet, 2001]. ER represents emission information for the gases and is also required to calculate emission factors (EF) for estimating gas emissions from fires.
First, we calculated background concentrations of each of the three gases for each month by using GOSAT-2 column-averaged concentrations data. The global distributions of the calculated background concentrations reflected characteristics of sources and sinks of each gas. Then, we compared our background concentrations based on column-averaged concentrations with the ground-based background concentrations from the Japan Meteorological Agency (JMA), and found that the seasonal variations of our background concentrations lagged by one or two months behind those of the JMA’s background data. We also found differences in concentrations between the two background data, which may be attributable to their differences in vertical profiles and seasonal variations.
Next, we calculated enhanced concentrations for each gas (ΔXCO2, ΔXCH4, and ΔXCO) from GOSAT-2 XCO2, XCH4 and XCO data. The global distributions of the calculated enhanced concentrations showed seasonal and regional characteristics reflecting their sources. We finally defined ER for each 5-degrees grid as a gradient of the linear regression fit between the enhanced concentrations of two gases. We found significant positive correlations between ΔXCH4 and ΔXCO2 (ΔXCH4/ΔXCO2) and between ΔXCH4 and ΔXCO (ΔXCH4/ΔXCO) and therefore could define their ER values globally. In contrast, there was less significant correlation between ΔXCO and ΔXCO2 and the ER values for ΔXCO/ΔXCO2 could not be defined globally.
We also calculated ΔXCH4/ΔXCO2 and ΔXCH4/ΔXCO for several characteristic regions. In the regions where biomass burning occurs frequently, the correlation between ΔXCH4 and ΔXCO2 got stronger during the active burning season than the other seasons. The calculated ER values of ΔXCH4/ΔXCO2 in active wildfire regions were 4.93, 3.20, 2.27, 2.16 and 1.69 ppb/ppm for savanna around Port Harcourt, temperate forest around Sydney, tropical forest around Amazon, boreal forest around Alaska, and Siberia, respectively. Our calculated ER value was almost consistent with the result by Parker et al. [2016] for savanna burning, although they were not consistent for tropical forest burning. In Mumbai and Tehran, ΔXCH4/ΔXCO showed a sharp increase at the end of their rainy season.
Our results suggested that calculated ER values would depend on the definition of background concentrations for each gas and the setting of the analysis area. We have also evaluated ER values based on several different sets of background concentrations.
First, we calculated background concentrations of each of the three gases for each month by using GOSAT-2 column-averaged concentrations data. The global distributions of the calculated background concentrations reflected characteristics of sources and sinks of each gas. Then, we compared our background concentrations based on column-averaged concentrations with the ground-based background concentrations from the Japan Meteorological Agency (JMA), and found that the seasonal variations of our background concentrations lagged by one or two months behind those of the JMA’s background data. We also found differences in concentrations between the two background data, which may be attributable to their differences in vertical profiles and seasonal variations.
Next, we calculated enhanced concentrations for each gas (ΔXCO2, ΔXCH4, and ΔXCO) from GOSAT-2 XCO2, XCH4 and XCO data. The global distributions of the calculated enhanced concentrations showed seasonal and regional characteristics reflecting their sources. We finally defined ER for each 5-degrees grid as a gradient of the linear regression fit between the enhanced concentrations of two gases. We found significant positive correlations between ΔXCH4 and ΔXCO2 (ΔXCH4/ΔXCO2) and between ΔXCH4 and ΔXCO (ΔXCH4/ΔXCO) and therefore could define their ER values globally. In contrast, there was less significant correlation between ΔXCO and ΔXCO2 and the ER values for ΔXCO/ΔXCO2 could not be defined globally.
We also calculated ΔXCH4/ΔXCO2 and ΔXCH4/ΔXCO for several characteristic regions. In the regions where biomass burning occurs frequently, the correlation between ΔXCH4 and ΔXCO2 got stronger during the active burning season than the other seasons. The calculated ER values of ΔXCH4/ΔXCO2 in active wildfire regions were 4.93, 3.20, 2.27, 2.16 and 1.69 ppb/ppm for savanna around Port Harcourt, temperate forest around Sydney, tropical forest around Amazon, boreal forest around Alaska, and Siberia, respectively. Our calculated ER value was almost consistent with the result by Parker et al. [2016] for savanna burning, although they were not consistent for tropical forest burning. In Mumbai and Tehran, ΔXCH4/ΔXCO showed a sharp increase at the end of their rainy season.
Our results suggested that calculated ER values would depend on the definition of background concentrations for each gas and the setting of the analysis area. We have also evaluated ER values based on several different sets of background concentrations.