Comparison of GOSAT SWIR and Aircraft Measurements of XCH 4 over West Siberia

Transcription

1 160 SOLA, 2015, Vol. 11, , doi: /sola Comparison of GOSAT SWIR and Aircraft Measurements of XCH 4 over West Siberia Akiko Ono 1, Sachiko Hayashida 1, Takafumi Sugita 2, Toshinobu Machida 2, Motoki Sasakawa 2, and Mikhail Arshinov 3 1 Faculty of Science, Nara Women s University, Nara, Japan 2 National Institute for Environmental Studies, Tsukuba, Japan 3 V. E. Zuev Institute of Atmospheric Optics, Russian Academy of Sciences, Siberian Branch, Russia Abstract We investigated the validity of column-averaged dry air mole fractions of methane (XCH 4 : V02.21) retrieved from shortwave infrared (SWIR) spectra obtained by Greenhouse gases Observing SATellite (GOSAT) over Siberia, which is known as a major source area of methane (CH 4 ). We compared the GOSAT XCH 4 set with aircraft measurements of XCH 4 that have been collected in Novosibirsk and Surgut, West Siberia since the 1990s. The average difference between the GOSAT XCH 4 and aircraft-based XCH 4 was 1.0 ± 22.0 ppb in Novosibirsk and 0.4 ± 14.8 ppb in Surgut when we selected appropriate pairs. These results indicate that GOSAT XCH 4 obtained over West Siberia is consistent with aircraft measurements and assure the reliability of the GOSAT XCH 4 product for scientific analysis. (Citation: Ono, A., S. Hayashida, T. Sugita, T. Machida, M. Sasakawa, and M. Arshinov, 2015: Comparison of GOSAT SWIR and aircraft measurements of XCH 4 over West Siberia. SOLA, 11, , doi: /sola ) 1. Introduction The Greenhouse gases Observing SATellite (GOSAT) was launched on 23 January, 2009 (Yokota et al. 2009; Kuze et al. 2009). Since then, the Thermal And Near-infrared Sensor for carbon Observations Fourier Transform Spectrometer (TANSO- FTS) on board GOSAT has collected global on carbon dioxide (CO 2 ) and methane (CH 4 ) systematically and continuously. Some validation studies of column-averaged dry air mole fractions of methane (XCH 4 ) obtained from GOSAT have been conducted (Morino et al. 2011; Yoshida et al. 2013) by comparing the GOSAT to those of a worldwide network of groundbased FTS called the Total Carbon Column Observing Network (TCCON) (Wunch et al. 2011). The average difference between GOSAT and TCCON XCH 4 has been reported to be less than 1%. All of the TCCON sites, however, are located in background regions of atmospheric CH 4 ; validation studies in CH 4 source regions have been limited. In this study, we compared GOSAT XCH 4 (V02.21) with aircraft measurements that were taken over Novosibirsk and Surgut, West Siberia. Siberia is one of the most significant CH 4 source regions, with CH 4 from wetlands and oil/gas fields. Furthermore, this study can confirm the validity of GOSAT XCH 4 at high latitudes; to date, few studies have reported on GOSAT quality at high latitudes (Gavrilov et al. 2014; Inoue et al. 2014). We describe the GOSAT products and the aircraft measurements used in this study in Section 2. The estimation method of XCH 4 from aircraft measurements is presented in Section 3, and the results of the comparison of aircraft-based XCH 4 with GOSAT products are shown in Section 4. We conclude the study in Section 5. Corresponding author: Akiko Ono, Nara Women s University, Kita-Uoya Nishimachi, Nara , Japan. ono@cc.nara-wu.ac.jp. 2015, the Meteorological Society of Japan. 2. GOSAT and aircraft 2.1 GOSAT SWIR XCH 4 GOSAT has a sun-synchronous orbit with an equator crossing time of 13:00 and a 3-day revisit time. Nominal observation by the TANSO-FTS sensor was carried out by the 5-point cross-track scan mode until July In August 2010, the grid observation pattern changed from the 5-point cross-track scan mode to the 3-point cross-track scan mode with 260 km spacing. As shown in Fig. 2c of Kuze et al. (2012), GOSAT measures the spectra three times at approximately the same location. When multiple were available at almost the same location, we selected the XCH 4 with the lowest spectral fitting residuals (χ 2 ) as a representative for that location (see Section 4). In this study, we used GOSAT SWIR XCH 4 V02.21 retrieved by the National Institute for Environmental Studies (NIES). The retrieved SWIR XCH 4 are screened using post-processing procedures. After the post-screening, the selected are accessible to registered users who applied to the GOSAT Research Announcement (RA); we will refer to this as RA. Some of the are further screened using stricter criteria, and the selected are accessible to general users (); we will refer to this as. 2.2 Aircraft measurements Since the 1990s, NIES has used chartered airplanes (AN-30 and AN-24) to perform monthly flask air sampling of greenhouse gasses in West Siberia (Umezawa et al. 2012). We used vertical CH 4 profiles obtained from these aircraft measurements in the vicinity of Novosibirsk (54.3 N, 82.2 E) and Surgut (61.5 N, 73.0 E). The air samples are collected in 550 ml glass flasks at altitudes ranging from 0.5 to 7 km with km intervals. These samples are sent to NIES and analyzed for CH 4 concentration using a gas chromatograph equipped with a flame ionization detector (FID). The precision of gas chromatograph analysis for CH 4 measurement is estimated to be 1.7 ppb. The standard gases used in this analysis are calibrated against the NIES 94 CH 4 scale. According to the results of the WMO Round-Robin intercomparison (Zhou et al. 2009; NOAA/ESRL 2015), the NIES 94 scale was higher than the NOAA/GMD scale by ppb in a range of 1,750 1,840 ppb. 3. Analysis methods XCH 4 estimation from aircraft To compare the CH 4 concentrations obtained by aircraft sampling with GOSAT XCH 4, we must infer the CH 4 profiles for the altitudes at which the aircraft cannot make measurements. In this study, we followed a method similar to Saitoh et al. (2012). For the altitudes between 0.5 and 7.0 km (about 400 hpa), aircraft measurements are available. For the altitudes below the lowest altitude of aircraft sampling ( km, about 950 hpa) and the altitudes between the highest altitude of aircraft sampling and the tropopause, we assumed CH 4 concentrations to be equal to the values obtained at the nearest altitude. The tropopause height was

2 Ono et al., Comparison of GOSAT SWIR and Aircraft Measurements of XCH 4 over West Siberia 161 taken from the NCEP/NCAR reanalysis (Kalnay et al. 1996). Between the tropopause and 65 km (0.1 hpa), we used the a priori profile calculated by the NIES transport model (Maksyutov et al. 2008; Saeki et al. 2013). The stratospheric part of the model (up to ~10 hpa) is constrained by the climatology of the Halogen Occultation Experiment (HALOE) (Russell et al. 1993; Grooß and Russell 2005). We applied averaging kernels of the GOSAT retrievals to the CH 4 profile construction mentioned above. Subsequently, the aircraft-based XCH 4 values were estimated by accumulating CH 4 number densities using the CH 4 profile and by dividing the accumulated CH 4 densities by the total number density of dry air molecules. The dry air molecules were calculated using the temperature and water vapor from the Japan Meteorological Agency Grid Point Value (JMA-GPV) (e.g., Nakakita et al. 1996). Temperature from the Committee on Space Research (COSPAR) International Reference Atmosphere (CIRA-86) (Fleming et al. 1990) and water vapor from the Air Force Geophysics Laboratory (AFGL) (Anderson et al. 1986) were also used for altitudes above 10 hpa for which no JMA-GPV were available. The measured and assumed CH 4 concentrations used to calculate XCH 4 are shown by the solid curve of Fig. S1. 4. Comparison between GOSAT and aircraft-based XCH 4 Figure 1 shows the time series of the aircraft-based XCH 4 and of the GOSAT XCH 4 that were observed around Novosibirsk and Surgut from 2009 through In the figure, red solid circles and red open circles represent all of the aircraft-based XCH 4, and grey circles represent daily values of GOSAT XCH 4 that were observed ± 5 from the flight locations in Novosibirsk and Surgut. The black open circles represent the monthly average value of GOSAT XCH 4, and the black error bars show the standard deviations of GOSAT XCH 4 for the corresponding month, which is about 14 ppb, on average. During winter, the GOSAT XCH 4 over Novosibirsk and Surgut were not retrieved at high latitudes due to a large solar zenith angle (> 70 ) (Yoshida et al. 2011). The aircraft-based XCH 4 values were remarkably high in winter, but there were no comparable GOSAT ; as such, those were beyond the scope of this study. Except for the from 6 August, 2012 and 30 July, 2013 in Novosibirsk and the from 10 September, 2012, 26 May, 2013, and 28 July, 2013 in Surgut (red solid circles), aircraft-based XCH 4 values were close to the average monthly GOSAT XCH 4 and fell within the range of the corresponding standard deviations. In both series, an increasing trend is clearly observable. To compare aircraft with GOSAT quantitatively, we selected matching pairs using the following procedure. First, we defined the distance between aircraft and GOSAT observation locations as L and the difference between aircraft and GOSAT observation dates as DD. We assumed a matching condition of L to be within 300 km (L 300 km) and a matching condition of DD to be within 1 day (DD = 1, 0, 1). As mentioned in Section 2, in August 2010 the grid observation pattern changed to the 3-point cross-track scan mode, and three were available at approximately the same location. Therefore, multiple GOSAT near a given location were often selected in correspondence to an aircraft measurement. Including these redundant, the total number of matched pairs was 126. We refer to this set as case-1, in which the were selected using the criteria of L 300 km and 1 DD 1. To refine the selection of corresponding pairs, we added the criteria described below. As described in Section 2.1, GOSAT obtained three measurements at approximately the same location. When those three GOSAT points corresponded to one aircraft s measurement, the GOSAT with the lowest χ 2 was selected from the cluster of GOSAT at almost the same point. After selecting the with the lowest χ 2, we obtained the set case-2. The number of pairs in case-2 was 49. Furthermore, when multiple GOSAT at different locations within L 300 km or on different days corresponded to a given aircraft, the GOSAT with the shortest L and lowest DD was selected. We refer to the set based on this selection as case-3 ; the number of pairs in case-3 was 23. Though we examined all of the above cases, we focused on the results of case-3 in this paper. The detailed results of case-1 and case-2 are shown in Table S1 and S2 for reference. To evaluate the accuracy and precision of GOSAT XCH 4, we calculated the difference between GOSAT and aircraft-based XCH 4. We defined and the ratio of as = GOSAT XCH 4 Aircraft-based XCH 4, (1) XCH4 Diff Ratio of = _. (2) Aircraft-based XCH 4 A scatter plot of corresponding aircraft-based and GOSAT XCH 4 is shown in Fig. 2, and pairs of case-3 are summarized in Table 1. As shown in the figure, variability of GOSAT XCH 4 is greater than that of aircraft-based XCH 4. The coefficient Fig. 1. Time series of the GOSAT SWIR XCH 4 and aircraft-based XCH 4 that were observed in (a) Novosibirsk and (b) Surgut from 2009 to The grey circles indicate daily values of GOSAT XCH 4 that were observed ±5 from the flight locations. The black open circles represent the monthly average value of GOSAT XCH 4, and the black error bars show the standard deviations of GOSAT XCH 4 for the corresponding month. The red solid circles indicate aircraft-based XCH 4 used for comparison in this study, and the red open circles indicate aircraft-based XCH 4 that was not matched with GOSAT. Fig. 2. Scatter plot of aircraft-based XCH 4 and GOSAT SWIR XCH 4 in (a) Novosibirsk and (b) Surgut. The black open circles indicate pairs of case-1, the grey circles indicate pairs of case-2, and the red circles indicate pairs of case-3. The black lines show one-to-one correspondence.

3 162 SOLA, 2015, Vol. 11, , doi: /sola Table 1. Pairs of aircraft-based XCH 4 and GOSAT XCH 4 in Novosibirsk and Surgut for case-3 Flight area Observation date (YYYYMMDD) Aircraft GOSAT Type of GOSAT CH km Aircraft measurement value Aircraft-based A XCH 4 GOSAT G Ratio of ()/A Novosibirsk RA Surgut Table 2. Average values and standard deviations of matching case-3 pairs for each category in Novosibirsk and Surgut Average values Novosibirsk Surgut Flightarea Number of Ratio of ()/A Number of Ratio of ()/A All ± ± ± ± 0.82 Type of GOAST RA ± ± ± ± 0.82 Observation same day (DD = 0) ± ± ± ± 0.84 Total average of both areas 0.7 ± 18.9 ppb ( 0.04 ± 1.05%) of determination (R 2 ) in case-3 is not necessarily higher than that in other cases because of less selected, although the selection criterion is the strictest. In Table 2, we show the average of () based on the type of GOSAT ( or RA) and on the observation date. For all pairs, obtained was 1.0 ± 22.0 ppb ( 0.05 ± 1.23%) at Novosibirsk, and 0.4 ± 14.8 ppb ( 0.02 ± 0.82%) at Surgut. Yoshida et al. (2013) performed a validation of GOSAT XCH 4 (V02.xx * ) using the 723 measurements provided by TCCON and showed that was 5.9 ± 12.6 ppb. Their study represented global coverage, but all the TCCON sites used in the study are located in background regions of CH 4 ; their validation study did not cover source regions of CH 4. There are only a few validation studies that cover high latitude regions with large CH 4 emissions. Gavrilov et al. (2014) compared GOSAT XCH 4 (V02. xx * ) with 256 ground-based FTS measurements obtained near St. Petersburg, Russia and reported that was 1.9 ± 14.5 * The mixed version including various minor versions such as V02.00, V02.10, and V ppb. Inoue et al. (2014) compared GOSAT XCH 4 (V02.00) with 3 aircraft measurements from Yakutsk, Siberia and reported that the was 9.2 ± 15.2 ppb (3.7 ± 16.7 ppb) within ± 2 (± 5 ). In this study, we found between GOSAT XCH 4 (V02.21) and aircraft-based XCH 4 in West Siberia was 0.7 ± 18.9 ppb ( 0.04 ± 1.05%). The shown in our study (within ~1%) is almost in the same range as those in the aforementioned studies. As shown in Table 1, the absolute value of ( XCH 4 _ Diff ) on 14 June, 2010 (44.2 ppb) in Novosibirsk was notably large despite the fact that the paired were observed on the same day (DD = 0). One possible explanation is that the GOSAT XCH 4 was categorized as RA, which was screened with less strict criteria. If this RA is excluded, the increases from 1.0 to 2.9 ppb, while the value of the standard deviation decreases from 22.0 to 18.5 ppb. Figure 3 shows the dependency of on L. In Novosibirsk (Fig. 3a), the value increases with respect to L, as expected. On the other hand, in Surgut (Fig. 3b), most of the values are positive regardless of the value of L. On the whole, it is difficult to find reasonable dependency of

4 Ono et al., Comparison of GOSAT SWIR and Aircraft Measurements of XCH 4 over West Siberia 163 Fig. 3. Relationship between the distance from aircraft site to GOSAT observation location (L) and difference between GOSAT and aircraft-based XCH 4 () in (a) Novosibirsk and (b) Surgut. Average of XCH 4 _ Diff and the standard deviation, root mean square deviation (RMSD), and mean absolute deviation (MAD) are shown in figures. The black open circles indicate pairs of case-1, the grey circles indicate pairs of case-2, and the red circles indicate pairs of case-3. on parameter L in the range examined in this study because of limited. This is also true for parameter DD. Estimated aircraft-based XCH 4 values are subject to uncertainties attributable to the assumed CH 4 profile (e.g., Sepúlveda et al. 2014). The uncertainty of XCH 4 arises from the accuracy of climatology of CH 4 in the stratosphere, tropopause height, temperature profiles, and CH 4 in the surface layer. We investigated the impact of those factors on aircraft-based XCH 4 estimations. First, we applied the stratospheric CH 4 variations within 1σ standard deviation provided in the HALOE climatological set (ranging from ±3.2% to ±10.0% at ~215 hpa). The results of the sensitivity analysis are shown in Fig. 4. The aircraft-based XCH 4 values changed by ±17.7 ppb and ±20.6 ppb, on average, in Novosibirsk and Surgut, respectively. Second, we shifted the tropopause height downward or upward corresponding to ±70 hpa. The aircraftbased XCH 4 values changed ±22.4 ppb and ±24.8 ppb, on average, in Novosibirsk and Surgut, respectively, as shown in Fig. 4. Third, when temperature (GPV) were replaced with NCEP/NCAR reanalysis, the aircraft-based XCH 4 values in Novosibirsk and Surgut changed, on average, by 0.3 ppb and 1.2 ppb, respectively. Therefore, the different meteorological temperature sets examined here had only a minor effect on the aircraft-based XCH 4 estimation. Fourth, when the CH 4 concentration at the ground level was fixed at a specific value (1,950 ppb), the aircraft-based XCH 4 values increased to 1.7 ppb in Novosibirsk and 0.4 ppb in Surgut. As the air column below 500 m can contribute only about 5% to the total air column, the variation of CH 4 at this altitude range (below 500 m) cannot have a critical effect on. The CH 4 content in the lower troposphere (LT), below ~3 km, often increases in the summer because of strong methane emissions from the surface. For example, the CH 4 profiles on 26 July, 2010 at Novosibirsk and on 28 July, 2013 at Surgut show clear LT CH 4 enhancement (Fig. S2). On those days, the GOSAT XCH 4 values indicated good consistency with aircraft-based XCH 4 (see Table 1 and Fig. 4). These cases demonstrate the potential of GOSAT for detecting LT CH 4 enhancement in the summer over West Siberia. However, some other showing LT CH 4 enhancement (e.g., on 30 July, 2013 at Novosibirsk) are beyond the uncertainty range. This difference between aircraft and GOSAT XCH 4 may be attributable to the variability in CH 4 content in the lower atmosphere, the effect of aerosols/cirrus (Ohyama et al. 2015), or the large interval between air sampling levels, which is not sufficient to detect the thin-layered structure of CH 4 profiles. The target accuracy for application of the XCH 4 to inverse analysis has been estimated to be ~1% (e.g., Meirink et al. 2006). Because the shown in our study was within 1%, The monthly HALOE climatological products can be downloaded from the web ( suplement.tar). Fig. 4. Uncertainty of aircraft-based XCH 4 for each case-3 pair at (a) Novosibirsk and (b) Surgut. The red circles indicate aircraft-based XCH 4, the blue circles indicate GOSAT XCH 4, the black error bars show the variation of 1σ standard deviation of the HALOE climatological set, and the green error bars show the variation caused by shifting the tropopause height downward or upward by ±70 hpa. this result suggests the GOSAT CH 4 have enough quality for used in inverse analysis for detecting CH 4 sources on the global scale. 5. Summary and conclusions In order to confirm the validity of the GOSAT XCH 4 (V02.21) product obtained from the SWIR band and retrieved by NIES, we compared the GOSAT with aircraft measurements over Novosibirsk and Surgut in West Siberia. For each aircraft measurement, we selected GOSAT that satisfied the matching conditions of L 300 km and 1 DD 1. We selected 12 pairs for Novosibirsk and 11 pairs for Surgut. For all pairs, calculated as 1.0 ± 22.0 ppb ( 0.05 ± 1.23%) at Novosibirsk, and 0.4 ± 14.8 ppb ( 0.02 ± 0.82%) at Surgut. The values of both sites were within 1%. As most of the CH 4 measurement sites, such as the TCCON sites, are located in background regions of atmospheric CH 4, only a few validation studies have been conducted over CH 4 source regions prior to our study. The result of this study demonstrates that GOSAT SWIR XCH 4 is also reliable over CH 4 source regions and thus may be helpful in scientific applications such as inverse analysis of global CH 4 emissions. Acknowledgements This study was supported by the Environmental Research and Technology Development Fund (A-1202) of the MOE and the Green Network of Excellence (GRENE-ei) of the MEXT. This study was performed within the framework of the GOSAT RA. The authors thank the staff of the Zuev IAO, Russia and the WIM, Russia for supporting the air sampling over Siberia, as well as the RME for partial financial support for research flights under State Contract No (RFMEFI61314X0013). Edited by: S. Morimoto

5 164 SOLA, 2015, Vol. 11, , doi: /sola Supplement Figure S1. An example of the aircraft profile observed on 30 July, 2013 (black circles) with the HALOE climatology profile assumed for the altitudes above the aircraft measurements (solid curve). Aircraft-based XCH 4 was calculated using the CH 4 profile shown in the figure. See text for details. Figure S2. Methane profiles used in this study were obtained by aircraft at Novosibirsk and Surgut. Dates of observation are indicated in the legend as yyyymmdd and correspond to the list in Table 1. Table S1. Matching case-1 pairs of GOSAT XCH 4 and aircraft-based XCH 4 in Novosibirsk and Surgut. Table S2. Matching case-2 pairs of GOSAT XCH 4 and aircraft-based XCH 4 in Novosibirsk and Surgut. References Anderson, G. P., S. A. Clough, F. X. Kneizys, J. H. Chetwynd, and E. P. Shettle, 1986: AFGL Atmospheric Constituent Profiles (0 120 km). Tech. Rep. AFGL-TR , Air Force Geophysics Laboratory, Hansom AFB, MA, USA. Fleming, E. L., S. Chandra, J. J. Barnett, and M. Corney, 1990: Zonal mean temperature, pressure, zonal wind, and geopotential height as functions of latitude, COSPAR International Reference Atmosphere: 1986, Part II: Middle Atmosphere Models. Adv. Space Res., 10, 12, 11 59, doi: / (90)90386-e. Gavrilov, N. M., M. V. Makarova, A. V. Poberovskii, and Y. M. Timofeyev, 2014: Comparisons of CH 4 ground-based FTIR measurements near Saint Petersburg with GOSAT observations. Atmos. Meas. Tech., 7, , doi: /amt Grooß, J.-U., and J. M. Russell, 2005: Technical note: A stratospheric climatology for O 3, H 2 O, CH 4, NO x, HCl and HF derived from HALOE measurements. Atmos. Chem. Phys., 5, , doi: /acp Inoue, M., I. Morino, O. Uchino, Y. Miyamoto, T. Saeki, Y. Yoshida, T. Yokota, C. Sweeney, P. P. Tans, S. C. Biraud, T. Machida, J. V. Pittman, E. A. Kort, T. Tanaka, S. Kawakami, Y. Sawa, K. Tsuboi, and H. Matsueda, 2014: Validation of XCH 4 derived from SWIR spectra of GOSAT TANSO-FTS with aircraft measurement. Atmos. Meas. Tech., 7, , doi: / amt Kalnay, E., M. Kanamitsu, R. Kistler, W. Collins, D. Deaven, L. Gandin, M. Iredell, S. Saha, G. White, J. Woollen, Y. Zhu, A. Leetmaa, B. Reynolds, M. Chelliah, W. Ebisuzaki, W. Higgins, J. Janowiak, K. C. Ropelewski, J. Wang, R. Jenne, and D. Joseph, 1996: The NCEP/NCAR 40-Year Reanalysis Project. Bull. Amer. Meteor. Soc., 77, Kuze, A., H. Suto, M. Nakajima, and T. Hamazaki, 2009: Thermal and near infrared sensor for carbon observation Fourier-transform spectrometer on the Greenhouse Gases Observation Satellite for greenhouse gases monitoring. Appl. Opt., 48, 35, , doi: /ao Kuze, A., H. Suto, K. Shiomi, T. Urabe, M. Nakajima, J. Yoshida, T Kawashima, Y. Yamamoto, F. Kataoka, and H. Buijs, 2012: Level 1 algorithms for TANSO on GOSAT: Processing and on-orbit calibrations. Atmos. Meas. Tech., 5, , doi: /amt Maksyutov, S., P. K. Patra, R. Onishi, T. Saeki, and T. Nakazawa, 2008: NIES/FRCCGC global atmospheric tracer transport model: Description, validation, and surface sources and sinks inversion. J. Earth Sim., 9, Meirink, J. F., H. J. Eskes, and A. P. H. Goede, 2006: Sensitivity analysis of methane emissions derived from SCIAMACHY observations through inverse modeling. Atmos. Chem. Phys., 6, , doi: /acp Morino, I., O. Uchino, M. Inoue, Y. Yoshida, T. Yokota, P. O. Wennberg, G. C. Toon, D. Wunch, C. M. Roehl, J. Notholt, T. Warneke, J. Messerschmidt, D. W. T. Griffith, N. M. Deutscher, V. Sherlock, B. Connor, J. Robinson, R. Sussmann, and M. Rettinger, 2011: Preliminary validation of column-average volume mixing ratios of carbon dioxide and methane retrieved from GOSAT short-wavelength infrared spectra. Atmos. Meas. Tech., 4, , doi: /amt Nakakita, E., S. Ikebuki, T. Nakamura, M. Kanmuri, M. Okuda, A. Yamaji, and T. Takasao, 1996: Short-term rainfall prediction method using a volume scanning radar and grid point value from numerical weather prediction. J. Geophys. Res., 101, D21, NOAA/ESRL, 2015: The 6th WMO/IAEA Round Robin Comparison Experiment. results.php?rr=rr6&param=ch4 (accessed ). Ohyama, H., S. Kawakami, T. Tanaka, I. Morino, O. Uchino, M. Inoue, T. Sakai, T. Nagai, A. Yamazaki, A. Uchiyama, T. Fukamachi, M. Sakashita, T. Kawasaki, T. Akaho, K. Arai, and H. Okumura, 2015: Observations of XCO 2 and XCH 4 with groundbased high-resolution FTS at Saga, Japan and comparisons with GOSAT products. Atmos. Meas. Tech. Discuss., 8, , doi: /amtd Russell, J. M. III, L. L. Gordley, J. H. Park, S. R. Drayson, D. H. Hesketh, R. J. Cicerone, A. F. Tuck, J. E. Frederick, J. E. Harries, and P. J. Crutzen, 1993: The Halogen Occultation Experiment. J. Geophys. Res., 98, D6, Saeki, T., R. Saito, D. Belikov, and S. Maksyutov, 2013: Global high-resolution simulations of CO 2 and CH 4 using a NIES transport model to produce a priori concentrations for use in satellite retrievals. Geosci. Model Dev., 6, , doi: / gmd Saitoh, N., M. Touno, S. Hayashida, R. Imasu, K. Shiomi, T. Yokota, Y. Yoshida, T. Machida, H. Matsueda, and Y. Sawa, 2012: Comparisons between XCH 4 from GOSAT Shortwave and Thermal Infrared Spectra and Aircraft CH 4 Measurements over Guam. SOLA, 8, , doi: /sola Sepúlveda, E., M. Schneider, F. Hase, S. Barthlott, D. Dubravica, O. E. García, A. Gomez-Pelaez, Y. González, J. C. Guerra, M. Gisi, R. Kohlhepp, S. Dohe, T. Blumenstock, K. Strong, D. Weaver, M. Palm, A. Sadeghi, N. M. Deutscher, T. Warneke, J. Notholt, N. Jones, D. W. Griffith, D. Smale, G. W. Brailsford, J. Robinson, F. Meinhardt, M. Steinbacher, T. Aalto, and D. Worthy, 2014: Tropospheric CH 4 signals as observed by NDACC FTIR at globally distributed sites and comparison to GAW surface in situ measurements. Atmos. Meas. Tech., 7, , doi: / amt Umezawa, T., T. Machida, S. Aoki, and T. Nakazawa, 2012: Contributions of natural and anthropogenic sources to atmospheric methane variations over western Siberia estimated from its carbon and hydrogen isotopes. Global Biogeochem. Cycles, 26, GB4009, doi: /2011gb Wunch, D., G. C. Toon, J. -F. L. Blavier, R. A. Washenfelder, J. Notholt, B. J. Connor, D. W. T. Griffith, V. Sherlock, and P. O. Wennberg, 2011: The Total Carbon Column Observing Network. Philos. Trans. R. Soc. A, 369, , doi: /rsta Yokota, T., Y. Yoshida, N. Eguchi, Y. Ota, T. Tanaka, H. Watanabe, and S. Maksyutov, 2009: Global concentrations of CO 2 and CH 4 retrieved from GOSAT: first preliminary results. SOLA, 5, , doi: /sola Yoshida, Y., Y. Ota, N. Eguchi, N. Kikuchi, K. Nobuta, H. Tran, I. Morino, and T. Yokota, 2011: Retrieval algorithm for CO 2 and CH 4 column abundances from short-wavelength infrared spectral observations by the Greenhouse gases observing satellite. Atmos. Meas. Tech., 4, , doi: /amt Yoshida, Y., N. Kikuchi, I. Morino, O. Uchino, S. Oshchepkov, A. Bril, T. Saeiki, N. Schutgens, G. C. Toon, D. Wunch, C. M. Roehl, P. O. Wennberg, D. W. T. Griffith, N. M. Deutscher, T. Warneke, J. Notholt, J. Robinson, V. Sherlock, B. Connor, M. Rettinger, R. Sussmann, P. Ahonen, P. Heikkinen, E. Kyro, J. Mendonca, K. Strong, F. Hase, S. Dohe, and T. Yokota, 2013: Improvement of the retrieval algorithm for GOSAT SWIR XCO 2 and XCH 4 and their validation using TCCON. Atmos. Meas. Tech., 6, , doi: /amt Zhou, L. X., D. Kitis and P. P. Tans, 2009: Report of the Fourth WMO Round-Robin Reference Gas Intercomparison, In Report of the 14th WMO/IAEA Meeting of Experts on Carbon Dioxide, Other Greenhouse Gases and Related Tracers Measurement Techniques, Helsinki, September 2007, Tuomas Laurila, Ed., WMO/GAW Report No. 186, Manuscript received 19 August 2015, accepted 28 October 2015 SOLA: jstage. jst. go. jp/browse/sola/