Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Comparison of XCH 4 Derived from g-b FTS and GOSAT and Evaluation Using Aircraft In-Situ Observations over TCCON Site

Comparison of XCH 4 Derived from g-b FTS and GOSAT and Evaluation Using Aircraft In-Situ... It is evident that evaluating the measurement of greenhouse gases (GHGs) obtained from multi-platform instruments against accurate and precise instrument such as aircraft in-situ is very essential when using remote sensing GHGs results for source/sink estimations with inverse modeling. The results of the inverse models are very sensitive even to small biases in the data (Rayner and O’Brien 2001). In this work, we have evaluated ground-based high resolution Fourier Transform Spectrometer (g-b FTS) and the Greenhouse gases Observing SATellite (GOSAT) column-averaged dry air mole fraction of methane (XCH ) through aircraft o o in-situ observations over Anmyeondo station (36.538 N, 126.331 E, 30 m above sea level). The impact of the spatial coinci- dence criteria was assessed by comparing GOSAT data against g-b FTS. We noticed there was no any systematic difference based on the given coincidence criteria. GOSATexhibited a bias ranging from 0.10 to 3.37 ppb, with the standard deviation from 4.92 to 12.54 ppb, against g-b FTS with the spatial coincidence criteria of ±1, ±3, ±5 degrees of latitude and longitude and ± 1 h time window. Data observed during ascent and descent of the aircraft is considered as vertical profiles within an altitude range of 0.2 to a maximum of 9.0 km so that some assumptions were applied for the construction of the profiles below 0.2 and above 9.0 km. In addition, the suitability of aircraft data for evaluation of remote sensing instruments was confirmed based on the assessment of o o uncertainties. The spatial coincidence criteria is ±1 latitude and ± 2 longitude and for temporal difference is ±1 h of the satellite observation overpass time were applied, whereas g-b FTS data are the mean values measured within ±30 min of the aircraft observation time. Furthermore, the sensitivity differences of the instruments were taken into account. With respect to aircraft, the g-b FTS data were biased by −0.19 ± 0.69%, while GOSAT data were biased by −0.42 ± 0.84%. These results confirm that both g-b FTS and GOSAT are consistent aircraft observations and assure the reliability of the datasets for inverse estimate of CH . . . . Keywords Aircraft in-situ G-b FTS GOSAT XCH 1 Introduction such as burning of fossil fuel and changes in land use. CH plays a major role in the chemistry of the Earth’satmosphere Next to CO , atmospheric methane (CH ) is one of the potent through the decomposition process; its increase is considered 2 4 greenhouse gases. Its concentrations have been increasing to change the balance of related chemical species (Cicerone since pre-industrial era as a result of intense human activities and Oremland, 1998). Therefore, accurate and precise mea- surements of CH play a substantial role for better compre- hension in global carbon cycle as well as its contribution to the Responsible Editor: Soon-Il An. global warming (Jain et al. 2000). Even though CH is the dominant anthropogenic greenhouse gas, there are still high * Samuel Takele Kenea uncertainties in CH sources and sinks at a global scale samueltake@yahoo.ca 4 (Frankenberg et al. 2008). A number of instruments deployed 1 onboard at various platforms (ground-based, airborne, and Climate Research Division, National Institute of Meteorological space-borne) have been involved in measuring of atmospheric Sciences (NIMS), 33, Seohobuk-ro, Seogwipo-si, Jeju-do 63568, Republic of Korea concentrations of CH . Aircraft in-situ measurements are highly accurate and precise and, therefore, capable to validate Department of Electrical Eng. and Center for Edge Plasma Science, Hanyang University, Seoul, Republic of Korea the ground-based instruments such as g-b FTS, as well as Korean Meteorological Society 416 S. T. Kenea et al. evaluate satellites instruments such as GOSAT. However, air- 3 Data and Methods craft observations are very sparse; this gap needs to be filled by highly accurate and precise measurements from satellite. 3.1 Aircraft Measurements Several studies on validation of remote sensing products of greenhouse gases such as CO and CH were conducted based To provide in-situ measurements of atmospheric CO ,CH , 2 4 2 4 on aircraft in-situ measurements at various sites (Araki et al. CO, and H O concentrations, the aircraft was equipped with a 2010; Messerschmidt et al. 2011; Geibeletal. 2012; Tanaka Wavelength-Scanned Cavity Ring Down Spectrometer et al. 2012;Miyamotoet al. 2013; Inoue et al. 2013, 2014). (CRDS; Picarro, G2401-mc) providing mixing ratio data re- The satellite products of CH should attain a demanding pre- corded at ~0.3 Hz intervals. The position (latitude, longitude, cision of <2% (< 34 ppb), in order to improve the precision of and height) of the aircraft was monitored by GPS, and infor- inversion models. In addition, achieving high relative accura- mation on the outside temperature, static pressure, and ground cy (< 10 ppb for XCH ) is more crucial and demanding than speed was provided by the aircraft’s instruments. Figure 2 precision to derive reliable surface fluxes via inverse model- displays schematic views of the CRDS instruments. Data col- ling (Buchwitz et al. 2016). lected during ascent and descent of the aircraft is considered as In this study, we have addressed two major issues. First, vertical profiles of CH over the Anmyeondo station. Typical observations of XCH and comparison between GOSAT and durations are in the range of 0.5–3 h. The temperature and g-b FTS XCH observations over the Anmyeondo station are pressure of the gas sample have to be tightly controlled at 45 discussed. Second, we assessed the suitability of aircraft in- C and 140 Torr in the CRDS (variations of less than 20 mK situ observations for validating other datasets and then evalu- and 0.1 Torr, respectively), which leads to highly stable spec- ate the correlative remote sensing measurements (g-b FTS and troscopic features (Chen et al. 2010). Any deviations from GOSAT) against aircraft. Here, the aircraft in-situ XCH was these values cause a reduction of the instrument’sprecision. computed based on the approach suggested by Inoue et al. Data recorded beyond these range of variations in cavity pres- (2013) and Ohyama et al. (2015). In fact, our g-b FTS was sure and temperature were discarded in this analysis. Variance calibrated with respect to TCCON (Total Carbon Column of the cavity pressure and temperature during flight results in Observing Network) common scale factor which is basically noise in the CH mixing ratios. The Picarro CRDS instrument derived from the aircraft data made at the other TCCON sites, has been regularly calibrated with respect to the standard gases therefore, our study will be useful for the scale factor exami- within the error range recommend by WMO. nation for Anmyeondo station by increasing the number of sample size. This paper is organized as follow: Section 2 gives 3.2 In-Situ Observation Data short descriptions about the Anmyeondo station. A brief over- view of data and method is provided in Section 3. Section 4 As for complementary information below the lower boundary presents results and discussion and followed by conclusions in of the aircraft observation, we utilized the CH concentration Section 5. data measured by the meteorological tower in the Anmyeondo ο ο site (36.53 N, 126.32 E, and 47 m above sea level), in close vicinity to the TCCON station. Atmospheric concentrations of 2 The Anmyeondo Station CH at 86 m above sea level were continuously measured with a precision <2 ppb using CRDS and provided as hourly aver- ο ο The Anmyeondo station is located at 36.538 N, 126.331 E, aged data. Sensors for detection of wind speed and direction, and 30 m above sea level. The topographic feature of the air temperature, barometric pressure, and humidity have been Anmyeondo station is a complex terrain which consists of also installed at this height. The in-situ data closest to the hills and valley within a few hundred meters. The climatic aircraft measurement time were selected to complement CH condition of the site is categorized as: winter is the coldest profiles. These data can be obtained from the WMO World season (a minimum temperature is about 2.7 C) while sum- Data Center for Greenhouse Gases (WDCGG) (http://ds.data. mer is the warmest season (a maximum temperature is about jma.go.jp/gmd/wdcgg/catalogue.cgi). 25.6 C), (the Anmyeondo station description is also given in Oh et al. (2018) paper). In other aspects, industries are avail- 3.3 Ground-Based FTS able within 100 km of the station. This area consists of agri- culture, forests, and urban areas. Several instruments are being The g-b FTS has been operating at Anmyeondo station within operated at the Anmyeondo station which makes this site is the network of TCCON since 2014; and detailed information important for validating remote sensing products from differ- about g-b FTS at this station was recently reported in Oh et al. ent platforms such as GOSAT. Figure 1 depicts all TCCON (2018). The TCCON is a worldwide network of ground-based sites (including operational, future, and previous sites) FTSs that was founded in 2004. It has been widely used as a globally. calibration and validation resource for satellite measurements Korean Meteorological Society Comparison of XCH Derived from g-b FTS and GOSAT and Evaluation Using Aircraft In-Situ Observations... 417 Fig. 1 Global distributions of the TCCON sites. (https://tccondata.org/) (e.g. Morino et al. 2011), but it is also offered for better un- sensors: the TANSO-FTS (Thermal And Near-infrared Sensor derstanding of the carbon cycle (e.g. Yang et al. 2007). The g- for carbon Observation Fourier Transform Spectrometer) and b FTS provides spectra in the near infrared spectrum with a TANSO-CAI (TANSO Cloud and Aerosol Imager) with the −1 high spectral resolution of 0.02 cm and a temporal resolu- IFOVof 10.5 km and 0.5–1.5 km, respectively. The TANSO- tion of ~2 min. From the recorded spectra, the target species FTS onboard GOSAT makes global observations both at nadir (XCH ) was retrieved with the GFIT nonlinear least-squares and off-nadir modes; and makes use of four spectral bands for fitting algorithm, which verifies a vertical scale factor (γ)of deriving CO and CH (Kuze et al. 2009) accurately. The 2 4 an a priori vertical profile based on the best spectral fit of the TANSO-FTS records the solar radiation reflected from the solar absorption signal. The scaled profile is then vertically surface at three Short Wavelength InfraRed (SWIR) bands at integrated, and the resulting column abundance is divided by the respective wavelength of 0.76, 1.6, and 2.0 μm and the the vertical column of dry air, calculated using the retrieved Earth’s radiation from the surface and atmosphere at the wide column of oxygen (O )(Wunch et al. 2011). Thermal InfraRed (TIR) band which is in the range between −1 5.5 and 14.3 μm with a resolution of 0.2 cm . The retrieval 3.4 Greenhouse Gases Observing SATellite (GOSAT) algorithm is a non-linear maximum a posteriori method with linear mapping based on Rodgers (2000). In this work, we have used column-averaged dry air mole fractions of CH GOSAT was launched into a sun-synchronous orbit on 23 (XCH ) V022 derived from the NIES retrieval algorithm January 2009 by H-IIA launch vehicle. It was placed in a (Yoshida et al. 2011, 2013). The achieved single measurement sun-synchronous orbit at a 666-km altitude and has a 3-day precision (random error) of the GOSAT XCH is revisit orbit cycle and a 12-day operation cycle. It carries two 4 Fig. 2 Left panel is CRDS instrument and right panel depicts the CRDS laboratory including gas cylinders for calibration purposes and computer Korean Meteorological Society 418 S. T. Kenea et al. approximately 16 ppb (1.0%) while systematic errors (relative derived from radiosonde observation and ERA-Interim accuracy or relative bias) are approximately 6 ppb (0.3%) reanalysis data were utilized for the calculation of the (Buchwitz et al. 2016). We accessed the GOSAT data through total column amounts of CH . The total column website (https://data2.gosat.nies.go.jp). amounts of CH were numerically integrated from the in-situ aircraft profiles weighted by dry air density from the surface up to the altitude of 70 km using the fol- 3.5 Methods lowing equation (e.g., Tanaka et al. 2012): The vertical profiles of CH mixing ratio are obtained dry f 1−f s CH H O during ascent and descent of the aircraft in spiral path 4 2 VC ¼ ∫ dp ð1Þ g:m (see Fig. 3) over Anmyeondo station. Since the altitude range of the aircraft measurements were limited to ap- where P is surface pressure, f -molefractionof H O proximately 0.2–5.0 km from 2012 to 2016 aircraft H O, f -molefractionof CH ,m – molecular mass 2 4 CH campaign and the maximum flight altitude was 9.0 km of air, and g - gravitational acceleration. The column in 2017, the in-situ data were utilized near the surface averaged dry-air mole fractions of CH are calculated to complement the CH profiles of aircraft-based data, from the integrated column amounts using the equation while above the aircraft ceiling, the highest altitude of given below: the aircraft observation data were extended up to the tropopause level to construct the complete CH profiles column CH XCH ¼ ð2Þ in a similar way as proposed by Miyamoto et al. column air (2013); Ohyama et al. (2015). We have used the in- where XCH is the column-averaged dry-air mole frac- situ data at the surface level to complement the aircraft 4 tion of CH . We set the coincidence criteria for compar- profile. Local planetary boundary layer (PBL) heights 4 ison between satellite data (GOSAT) and aircraft data as were obtained from National Centers for follows: GOSAT data are considered within ±1 degree Environmental Prediction (NCEP) NOAA (National latitude and ± 2 degree longitude boxes centered at the Oceanic and Atmospheric Administration) reanalysis da- Anmyeondo station and the aircraft data temporally ta. The mole fractions between the uppermost aircraft nearest to the GOSAT overpass time were selected, a measurement and the tropopause are assumed to be maximum of 1 h difference. While the g-b FTS data maintained constant as the highest aircraft measurements were averaged within ±30 min of the aircraft overpass, because at these higher altitudes the air is well mixed. which reduce the random error. During validation of For this analysis, the tropopause height was determined remote sensing vertical profiles, it is reasonable to con- from European Center for Medium range Weather sider the effect of vertical resolution and sensitivity of Forecasting (ECMWF) ERA-Interim reanalysis data, a the data (Rodgers and Connor 2003). In this analysis, horizontal resolution of 0.75 × 0.75 degrees. Above the vertically highly resolved aircraft CH profiles were de- tropopause height, GFIT a priori profiles were fixed to 4 graded by applying the column averaging kernels (a)of the aircraft data as shown by black dashed line in the j low resolution of vertical profiles derived from the left panel of Fig. 4. The dry air number density profiles Fig. 3 Typical flights path of the aircraft taken on May 09, 10, and 16, 2015 from left to right panels, respectively, are depicted. The right panel depicts vertical spiral flight path over Anmyeondo Korean Meteorological Society Comparison of XCH Derived from g-b FTS and GOSAT and Evaluation Using Aircraft In-Situ Observations... 419 Fig. 4 The left and right panels display the complete CH profile and tropopause and planetary boundary heights, respectively, and green aircraft profile appended with surface measurements, respectively, broken line shows the maximum aircraft observation altitude October 05, 2014. The red and blue broken lines represent the remote sensing instruments based on the following column to perturbations of the partial columns at the equation: various atmospheric levels. The typical column- averaging kernels for the g-b FTS and GOSAT at the in−situ a Anmyeondo station are shown in Fig. 5. The differences XCH ¼ XCH þ ∑ h aðÞ t −t ð3Þ j j in−situ a 4 4 of XCH between the aircraft and the remote sensing in−situ instruments or between g-b FTS and GOSAT are where XCH is the column-averaged dry-air mole expressed in terms of absolute and relative differences fraction of CH from the aircraft in-situ measurement, and the following mathematical expressions are: XCH and t are a priori column-averaged dry-air mole fraction and profile of CH , respectively (from g-b FTS Abso:diff : ¼ X −X ð4Þ or GOSAT), h is the pressure weighting function, and ins in−situ=FTS t is the in-situ profile from aircraft measurement. in − situ X −X ins in−situ=FTS The averaging kernel for the column retrieval is a vec- Rel:diff : ¼ 100% ð5Þ tor representing the sensitivity of the retrieved total in−situ=FTS where X and X are XCH of the aircraft and the in − situ/FTS ins 4 correlative remote sensing instruments (g-b FTS and GOSAT) or g-b FTS and GOSAT (which is used for the comparison be- tween g-b FTS and GOSAT), respectively. We examined the weather conditions at the measurement station during all obser- vation periods. For this period, we showed meteorological pa- rameters such as relative humidity, temperature, wind speed, and wind direction from radiosonde data in Fig. 6 in order to specu- late the weather conditions at the Anmyeondo station. Approximately above 3 km, the amount of atmospheric moisture was very low which was depicted by the relative humidity (RH) of below 40% in Fig. 6(a). As showninpanelscanddofFig. 6, the northerly and north easterly winds was blowing with a mag- −1 nitude of lower than 8 m s below 2 km, whereas between 2 and 6 km, the westerly and north westerly winds advected at a max- −1 imum speed of 20 m s over the measurement station. Therefore, this could bring a continental air mass from the Northern Hemisphere to the Anmyeondo station. Fig. 5 Column Averaging Kernels (CAKs) of CH are shown Korean Meteorological Society 420 S. T. Kenea et al. Fig. 6 Radiosonde measurements of (a) relative humidity, (b) temperature, (c) wind speed, and (d) wind direction, taken on October 05–07, 2014 4 Results and Discussion 4.1 Observations of XCH and Comparisons between g-b FTS and GOSAT In the following subsections, we have discussed obser- vations and comparison of XCH between g-b FTS and Here, the time series of XCH comparison between g-b 4 4 GOSAT,andthenevaluated them basedontheaircraft FTS and GOSAT was performed in the period between in-situ observations. 2014 and 2016 (see Fig. 7). We assessed to what extent Fig. 7 Time series of XCH obtained from the g-b FTS and the GOSAT at the spatial coincidence criteria of ±1, ±3, ±5 degrees latitude/longitude (left top panel) and GOSAT versus g-b FTS (right panel) in the period of green, blue, and red colors, respectively, and error bars indicate the 2014 to 2016 with a one-to-one dashed line, green circle denotes hourly standard deviations of the coincident datasets. Bottom left panel shows mean values of the g-b FTS, while the asterisks represent single day the time series of XCH on monthly mean basis and bottom right panel observations of the GOSAT. Right top panel shows g-b FTS vs GOSAT depicts annual cycle 2014–2016 Korean Meteorological Society Comparison of XCH Derived from g-b FTS and GOSAT and Evaluation Using Aircraft In-Situ Observations... 421 Table 1 Statistical results for the XCH difference between GOSAT and g-b FTS (bias = GOSAT - g-b FTS) data based on coincidence criteria of ±1, ±3, and ± 5 degrees of lat/lon and ± 1 h. N = number of coincident data, R = correlation coefficient Lat./ N R RMSE (ppb) bias ± std. (ppb) Rel. diff. ± std. (%) lon.(deg) ±1 9 0.86 5.73 3.37 ± 4.92 0.18 ± 0.27 ±3 16 0.96 6.01 1.68 ± 6.00 0.09 ± 0.33 ±5 27 0.76 12.31 0.10 ± 12.54 0.01 ± 0.68 the impact of spatial coincidence criteria affects the com- and enhanced concentrations over this region are predict- parison results. To match up the GOSAT data against g-b ed by model studies (Houweling et al. 2000). FTS, we chose geometric coincidence criteria of ±1, ±3, Dlugokencky et al. (1993) reported that significantly el- and ± 5 degrees of latitude/longitude centered at the evated CH mixing ratio was observed at Tae-ahn ο ο Anmyeondo station within the temporal window of Peninsula, Korea (36 44’ N, 126 08′ E) during summer ±1 h. Following the match-ups, all g-b FTS data coin- correlated with northwesterly airflow from northeast ciding with one satellite observation which were within China and east Siberia. In fact, taking a coarse colloca- 1 h time window are averaged, minimizing the g-b FTS tion criteria can induce collocation errors, but consider- random error. As a result of those coincidence criteria, ing a very strict criterion leads to a small sample set we obtained different sample size that might also affect (which affects statistics) due to the sparseness of the statistics to have robust conclusion. As can be seen GOSAT soundings. Buchwitz et al. (2017) described the in the top left panel of Fig. 7, the overall results sug- spatial variability in the bias of XCH termed as “relative gested that both instruments agreed in capturing the sea- bias”. It can arise from different surface reflectivity, aero- sonal variability of XCH over the Anmyeondo station. sol interference, and sloping terrain. They estimated Relatively large discrepancies were detected in peak 10 ppb relative bias for solar backscatter satellite obser- methane season, summer which might reflect the high vations. Correlations, root mean square error (RMSE), spatial heterogeneity of methane source and sink strength bias, and relative differences for these comparisons are (details are beyond the scope of this paper). The seasonal detailed in Tables 1 and 2 for varying collocation and annual cycles of XCH derived from the GOSAT criteria. As RMSE values increases from 5.73 to were compared with in g-b FTS observations over the 12.31 ppb with increasing the spatial window ±1 to ±5 Anmyeondo station, which are provided in the bottom degrees, which cover land and ocean parts since the panels of Fig. 7. Because of sample size, we used Anmyeondo station is located at the coastal area. In gen- GOSAT data that extracted within ±5 degrees latitude/ eral, there was no systematicdifferencenoticedby longitude coincidence criteria. As can be seen in Fig. 7 changing the collocation space. It was found low bias bottom panels, the overall patterns of seasonal and annu- (±σ) in GOSAT XCH against g-b FTS, which is about al cycle of the g-b FTS XCH are reproduced by GOSAT 3.37 ± 4.92 ppb, with a corresponding relative difference XCH . The maximum and minimum amounts of methane of 0.18 ± 0.27%, when applying the strict collocation were observed during summer and winter seasons, re- criteria within a time window of ±1 h and ± 1 degree of spectively. However, the seasonal cycle of CH in the latitude/longitude (see Table 1). Since we set the same Northern Hemisphere is more complex (Dlugokencky time window but changing the spatial window by ±3 and et al. 1994). While the destruction of CH due to reac- ± 5 degrees that resulted in the mean bias of 1.68 ± 6.0 tion with OH is expected to be stronger in summer, and 0.10 ± 12.54 ppb, respectively. The standard devia- source strengths also strongly vary with the seasons. tions of the differences are progressively increasing as East Asia is one of the largest source regions of methane, increasing the collocation space, but those values are Table 2 The same as Table 1,but Lat./lon.(deg) N R RMSE (ppb) bias ± std. (ppb) Rel. diff. ± std. (%) for daily mean basis. (bias = GOSAT minus g-b FTS) ±1 12 0.80 6.10 0.77 ± 6.30 0.04 ± 0.34 ±3 20 0.90 8.30 1.68 ± 8.34 0.09 ± 0.45 ±5 35 0.79 10.90 0.36 ± 10.98 0.04 ± 0.60 Korean Meteorological Society 422 S. T. Kenea et al. compatible with the combined measurement errors of the instruments. Similarly, we investigated the impact of the coincidence criteria by setting the time window on daily mean basis of g-b FTS measurements of XCH with varying the spatial coincidence ±1, ±3, and ± 5 degrees of latitude/longitude, and the bias was estimated to be less than 1.68 ppb. For a case of ±1 degree latitude/lon- gitude, GOSAT was biased by 0.78 ± 6.3 ppb with re- spect to g-b FTS. Those values are within the range of validation results reported in previous findings (e.g. Yoshida et al. 2013; Gavrilov et al. 2014;Ohyama et al. 2015) but with slightly smaller biases. Yoshida et al. (2013) performed a validation of GOSAT XCH (V02.xx) using the 723 measurements provided by Fig. 8 The comparisons of XCH between the aircraft observation versus TCCON and showed that bias was −5.90 ± 12.6 ppb. g-b FTS data (represented by blue square) and GOSAT (denoted by red Gavrilov et al. (2014) compared GOSAT XCH square) over Anmyeondo station are shown. The dashed line shows one- (V02.xx) with 256 ground-based FTS measurements ob- to-one line Table 3 Summary for the Instruments N Date (KST) Bias ± std. (ppb) Rel. diff. ± std. (%) statistics of XCH difference between remote sensing and aircraft in-situ data is given. The Aircraft vs. g-b FTS 2014-10-05 statistical estimators are 10:29:03–10:45:01 −14.60 −0.78 expressed in terms of bias and 10:46:10–11:00:01 −15.50 −0.83 relative differences with respect to 11:45:19–12:01:05 −13.00 −0.70 aircraft 13:23:10–13:38:21 −16.40 −0.88 13:39:30–13:53:33 −13.70 −0.74 14:23:46–14:48:23 −13.70 −0.74 14:49:40–15:04:30 −10.30 −0.56 2014-10-07 13:39:32–13:54:00 −8.00 −0.43 13:54:15–14:09:07 −4.50 −0.24 14:39:38–14:54:00 −8.70 −0.47 15:55:32–15:10:00 −0.30 −0.56 15:39:47–15:54:21 −6.50 −0.35 15:39:40–15:54:00 −11.60 −0.63 15:55:32–16:09:00 −11.00 −0.59 2017-10-29 09:59:16–10:31:08 0.90 0.05 10:31:09–11:03:24 4.90 0.26 12:58:58–13:37:07 15.10 0.82 13:37:07–14:19:40 −2.70 −0.14 2017-11-12 11:12:20–11:38:01 14.40 0.78 11:38:02–12:13:00 15.10 0.82 14:14:46–14:45:55 16.50 0.89 14:45:56–15:23:47 13.00 0.70 4 −3.66 ± 11.50 −0.19 ± 0.61 Aircraft vs. GOSAT (NIES V0221) Date 2012-10-17 4.53 0.25 2012-10-18 2.40 0.13 2014-10-05 −20.90 −1.13 3 −4.65 ± 14.11 −0.42 ± 0.84 Korean Meteorological Society Comparison of XCH Derived from g-b FTS and GOSAT and Evaluation Using Aircraft In-Situ Observations... 423 Fig. 9 The XCH total errors are shown for GOSAT (left panel) and for g-b FTS (right panel). (It was displayed only for the coincident days) tained near St. Petersburg, Russia and reported that the Aleks et al. 2011). The overall results indicated that g-b mean difference was −1.9 ± 14.5 ppb. Ohyama et al. FTS estimated slightly lower than aircraft. The statistical (2015) reported that the average differences XCH be- results for XCH comparisons between aircraft and g-b 4 4 tweenTANSO-FTSand g-bFTS data (TANSO-FTS mi- FTS are shown in Fig. 8 and summarized in Table 3,(cor- nus g-b FTS) is −7.6 ± 13.7 ppb. Right panel of Fig. 7 relation coefficient was not computed because of small demonstrates the results of XCH comparisons between sample numbers). The mean absolute difference of XCH 4 4 the GOSAT and g-b FTS at the spatial coincidence between aircraft and g-b FTS is −3.66 ± 11.50 ppb, with a criteria of ±1, ±3, ±5 degrees latitude/longitude green, corresponding mean relative difference of −0.19 ± 0.69%. blue, and red colors, respectively, and error bars indicate Previous findings have revealed that the unsampled part the standard deviations of the coincident datasets. This of the atmosphere above the aircraft ceiling contributes to depicts that the GOSAT data is well consistent with the the largest uncertainty in the total column calculated from g-b FTS that lie on the best line. Therefore, we can infer the aircraft profiles (Wunch et al. 2010). We estimated the that the impact the coincidence criteria (at least, up to error contributions on aircraft-based XCH . The error com- ±5) for performing the comparison of XCH over the ponents below the aircraft ceiling were derived by adding Anmyeondo station is insignificant. twice the precision of the aircraft data to the profile, and then re-integrated the profile (Wunch et al. 2010;Ohyama et al. 2015), the amount of error resulted in 3.03 ppb. The 4.2 Aircraft XCH Comparison with g-b FTS tropopause height variation induced an error of 0.40 ppb on averaged in estimating aircraft XCH and the stratospheric Several aircraft observation campaigns over Anmyeondo 4 error contribution were estimated by vertical shifting of the site were carried out in the period between 2012 and a priori by 1.0 km. That resulted in 6.30 ppb change on the 2017. However, a few numbers of aircraft data matched aircraft XCH values. The total errors were estimated to be with the remote sensing instruments were available during 4 7.0 ppb (Table 4). this observation period. The g-b FTS XCH was compared with the aircraft measurements. Here, g-b FTS data were averaged over a time window of ±30 min with respect to the aircraft measurement time. In addition, the averaging Table 4 Error budget for the estimated aircraft measurement of XCH . kernel of the g-b FTS was applied to the aircraft data to The total error is the sum, in quadrature, of the three errors. The aircraft equalize the sensitivities of CH mole fraction at each alti- error was estimated by adding the precision of the aircraft measurements to the profile and re-integrated the profile. Tropopause error was tude for the total column. Wunch et al. (2010)reported that determined by varying 1.0 km, while stratospheric error was estimated the airmass-dependent artifacts in XCH due to spectro- by shifting the stratospheric apriori profile of CH by 1.0 km. The fourth scopic inadequacies (e.g. line widths, inconsistencies in row on the given table shows the average value of the total retrieval errors the relative strengths of weak and strong lines) in from individual sounding of XCH from g-b FTS and GOSAT TCCON instruments were not seen. Here, a total number Aircraft error Tropopause error Stratospheric error Total of the aircraft measurements that matched with g-b FTS were only four during the observation period of 2014 to 3.03 ppb 0.40 ppb 6.30 ppb 7.00 ppb 2017. The diurnal range of g-b FTS data reflects not only g-b FTS total retrieval error GOSAT total retrieval error variability of airmass transport but source also sink process- 3.60 ppb 7.70 ppb es and the effect of measurement errors as well (Keppel- Korean Meteorological Society 424 S. T. Kenea et al. 4.3 Aircraft XCH Comparison with GOSAT such as aircraft in-situ is very essential when utilizing re- mote sensing GHGs results for source/sink estimations with In this section, the comparison of the GOSAT retrieval product inverse modeling. The results of the inverse models are very (V02.xx) of XCH against the aircraft observations over the sensitive even to small biases in the data (Rayner and Anmyeondo station was analyzed. Based on the coincidence O’Brien 2001). In this work, we carried out the comparison criteria, we obtained only three coincident days of observa- of XCH between g-b FTS and GOSAT over the tions. We took the mean values when obtaining more than one Anmyeondo station. Based on the comparison results be- GOSAT measurements. As noted the aircraft data was tween g-b FTS and GOSAT, both instruments are generally smoothed by GOSAT column averaging kernels. The compar- well captured the seasonal variability of XCH , the maxi- ison results of XCH between aircraft and GOSAT revealed a 4 mum and minimum amount of methane was observed dur- better agreement. The mean absolute difference of XCH was ing summer and winter seasons, respectively. The overall about −4.65 ± 14.11 ppb, also shown in Table 3. The absolute results indicate that a relatively high variability was exhibit- value of XCH difference on 5 October, 2014 was 20.9 ppb, ed during a peak methane season. In addition, the impact the which is larger than the other two coincident dates despite the coincidence criteria was assessed and there was no system- fact that the matching data were observed on the closest time atic difference was observed. The bias was estimated to be window. The discrepancy occurred at this particular date sig- from 0.1 to 3.37 ppb, and standard deviation was from 4.92 nificantly affected the mean of difference and standard devia- to 12.54 ppb. Column-averaged dry air mole fraction of CH tions. The difference could be attributable to the CH variabil- from 2012 to 2017 over the Anmyeondo station were de- ity in the lower atmosphere, the effect of aerosols/cirrus rived by using CH profiles measured by aircraft. Aircraft (Ohyama et al. 2015), or the large interval between air sam- measurements have good accuracy, but are limited in alti- pling levels, which is not sufficient to be captured the thin- tude floor and ceiling, and so we have to use additional layered structure of CH profiles by GOSAT. We tried to look information for surface and the stratosphere. As to my at atmospheric condition and aerosols during 5th October, knowledge, this is the first report on evaluation of remote 2015 using the information obtained from COMIS.4 (http:// sensing observations based on aircraft in-situ measurements uis.comis4.kma.go.kr/comis4/uis/common/index.do#), of XCH over this station. The uncertainty analysis of the Cloud-Aerosol Lidar and Infrared Pathfinder Satellite aircraft measurements of XCH confirmed the suitability of Observations (CALIPSO), and Aerosol Robotic Network data for evaluating the remote sensing products. These in- (AERONET) (see Appendix Figs. 10, 11, and 12). However, situ observations of the target species were compared against we noticed tropospheric aerosols that might be suggesting that g-b FTS, and GOSAT over there. It is noted that the aver- it tends to cause an underestimation of XCH retrievals. Still aging kernels of the remote sensing instruments were ap- further investigation is required in the future. However, the plied into the aircraft measurements. The retrieved XCH overall mean bias shown here is consistent with the previous values from the g-b FTS and GOSAT measurements showed results reported by Inoue et al. (2014, 2016). Inoue et al. a better agreement with the aircraft in-situ observations. (2014) made a comparison of GOSAT XCH (V02.00) with Both instruments revealed a negative bias against aircraft. 3 aircraft measurements from Yakutsk, Siberia and reported The relative differences of XCH were found to be −0.19 that the bias was 9.2 ± 15.2 ppb (3.7 ± 16.7 ppb) within ±2° ± 0.69% and − 0.42 ± 0.84% with respect to g-b FTS and (±5°). When considering 2 days average excluding a date of GOSAT, respectively. The small number of coincidences October 05, 2014, we obtained a bias 3.47 ± 1.51 ppb, which considered here hinders more robust conclusions, so we rec- is agreed with Inoue et al. (2016) who showed a 4.5 ± 15. ommend to carry out further works on validation by taking 20 ppb over land. Fig. 4 shows the total retrieval errors (the more coincident data using in-situ and remote sensing instru- room sum of the squares of smoothing error, retrieval noise, ments, as well as combining the model simulations in the and interference error components) of XCH obtained from future. This will allow us to improve clear identification of GOSAT and g-b FTS. The standard deviation of the differ- all the potential sources of uncertainties, as well as to under- ences that we obtained here is slightly larger than the total stand the role of local source/sink and dynamics. estimated error of GOSAT (see Fig. 9 and Table 4) and the aircraft XCH . Further work is still required to have more Acknowledgements This research was supported by the Research and robust conclusion by increasing the sample numbers. Development for KMA Weather, Climate, and Earth system Services Support to use of Meteorological Information and value Creation (KMA- 2018-00122). We acknowledge for those who provide the access for in-situ data from World Data Centre for Greenhouse Gases (WDCGG) (https://ds. 5 Conclusions data.jma.go.jp/gmd/wdcgg/cgi-bin/wdcgg/catalogue.cgi). We also greatly acknowledge the GOSAT science teams for the satellite data used in this Evaluating measurement of GHGs derived from multi- work. Many thanks goes to the science teams for the provision of ECMWF ERA-interim and NOAA NCAR reanalysis data utilized in this study. platform instruments against accurate and precise instrument Korean Meteorological Society Comparison of XCH Derived from g-b FTS and GOSAT and Evaluation Using Aircraft In-Situ Observations... 425 Appendix Fig. 10 Atmospheric sky conditions for 17th, 18th Oct. 2012, and 5th Oct. 2014, left to right panels, respectively, from COMIS.4. Red star shows Anmyeondo site. (http://uis.comis4.kma.go.kr/comis4/uis/common/index.do#) Fig. 11 Cloud and aerosol information from CALIOP data Korean Meteorological Society 426 S. T. Kenea et al. Fig. 12 Aerosol Optical Depth (AOD) from AERONET Level 2.0 data during 5th October, 2014 over Anmyeondo Open Access This article is distributed under the terms of the Creative Buchwitz, M., Schneising, O., Reuter, M., Heymann, J., Krautwurst, S., Bovensmann, H., Burrows, J.P., Boesch, H., Parker, R.J., Somkuti, Commons Attribution 4.0 International License (http:// P., Detmers, R.G., Hasekamp, O.P., Aben, I., Butz, A., Frankenberg, creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appro- C., Turner, A.J.: Satellite-derived methane hotspot emission esti- mates using a fast data-driven method. Atmos. Chem. Phys. 17, priate credit to the original author(s) and the source, provide a link to the 5751–5774 (2017). https://doi.org/10.5194/acp-17-5751-2017 Creative Commons license, and indicate if changes were made. Chen, H., Winderlich, J., Gerbig, C., Hoefer, A., Rella, C.W., Crosson, E.R., Van Pelt, A.D., Steinbach, J., Kolle, O., Beck, V., Daube, B.C., Publisher’sNote Springer Nature remains neutral with regard to juris- Gottlieb, E.W., Chow, V.Y., Santoni, G.W., Wofsy, S.C.: High accu- dictional claims in published maps and institutional affiliations. racy continuous airborne measurements of greenhouse gases (CO and CH ) using the cavity ring-down spectroscopy (CRDS) tech- nique. Atmos. Meas. Tech. 3,375–386 (2010) Cicerone, R.J., Oremland, R.S.: Biogeochemical aspects of atmospheric References methane. Glob. Biogeochem. Cycles. 2,299–327 (1988) Dlugokencky, E.J., Haris, J.M., Chung, Y.S., Tans, P.P., Fung, I.: The Araki, M., Morino, I., Machida, T., Sawa, Y., Matsueda, H., Ohyama, H., relationship between the methane seasonal cycle and regional Yokota, T., Uchino, O.: CO column-averaged volume mixing ratio 2 sources and sinks at Tae-ahn peninsula, Korea. Atmos. Environ. derived over Tsukuba from measurements by commercial airlines. Part A. 27,2015–2120 (1993) Atmos. Chem. Phys. 10, 7659–7667 (2010). https://doi.org/10. Dlugokencky, E.J., Steele, L.O., Lang, P.M., Masarie, K.A.: The growth 5194/acp-10-7659-2010. rate and distribution of atmospheric methane. J. Geophys. Res. 99, Buchwitz,M.,B.Dils,H.Boesch,C.Crevoisier, D.Detmers,C. 1702117044 (1994). https://doi.org/10.1029/94JD01245. Frankenberg, O. Hasekamp, W. Hewson, A. Laeng, S. Nol, J. Frankenberg, C., Bergamaschi, P., Butz, A., Houweling, S., Meirink, J.F., Notholt, R. Parker, M. Reuter, and O. Schneising, 2016: ESA Notholt, J., Petersen, A.K., Schrijver, H., Warneke, T., Aben, I.: Climate Change Initiative (CCI) Product Validation and Tropical methane emissions: a revised view from SCIAMACHY Intercomparison Report (PVIR) for the Essential Climate Variable onboard ENVISAT. Geophys. Res. Lett. 35, L15811 (2008). (ECV) Greenhouse Gases (GHG) for data set Climate Research https://doi.org/10.1029/2008GL034300 Data Package No. 3 (CRDP3), Version 4.0, 24. Feb. 2016. Korean Meteorological Society Comparison of XCH Derived from g-b FTS and GOSAT and Evaluation Using Aircraft In-Situ Observations... 427 Gavrilov, N.M., Makarova, M.V., Poberovskii, A.V., Timofeyev, Y.M.: Connor, B., Robinson, J., Sussmann, R., Rettinger, M.: Preliminary validation of column-averaged volume mixing ratios of carbon di- Comparisons of CH4 ground-based FTIR measurements near Saint Petersburg with GOSAT observations. Atmos. Meas. Tech. 7, 1003– oxide and methane retrieved from GOSAT short-wavelength infra- 1010 (2014). https://doi.org/10.5194/amt-7-1003-2014 red spectra. Atmos. Meas. Tech. 4,1061–1076 (2011). https://doi. Geibel, M.C., Messerschmidt, J., Gerbig, C., Blumenstock, T., Chen, H., org/10.5194/amt-4-1061-2011. Hase, F., Kolle, O., Lavriˇc, J.V., Notholt, J., Palm, M., Rettinger, Oh, Y.-S., Kenea, S.T., Goo, T.-Y., Chung, K.-S., Rhee, J.-S., Ou, M.-L., M., Schmidt, M., Sussmann, R., Warneke, T., Feist, D.G.: Byun, Y.-H., Paul Wennberg, O.P.P., Kiel, M., DiGangi, J.P., Diskin, Calibration of column-averaged CH over European TCCON FTS G.S., Velazco, V.A., Griffith, D.W.T.: Characteristics of greenhouse gas sites with airborne in-situ measurements. Atmos. Chem. Phys. 12, concentrations derived from ground-based FTS spectra at Anmyeondo, 8763–8775 (2012) South Korea. Atmos. Meas. Tech. 11(1–14), 2018 (2018) Houweling, S., Dentener, F., Lelieveld, J.: Simulation of preindustrial Ohyama, H., Kawakami, S., Tanaka, T., Morino, I., Uchino, O., Inoue, atmospheric methane to constrain the global source strength of nat- M., Sakai, T., Nagai, T., Yamazaki, A., Uchiyama, A., Fukamachi, ural wetlands. J. Geophys. Res. 105, 17243–17255 (2000) T., Sakashita, M., Kawasaki, T., Akaho, T., Arai, K., Okumura, H.: Inoue, M., Morino, I., Uchino, O., Miyamoto, Y., Yoshida, Y., Yokota, T., Observations of XCO and XCH with ground-based high-resolu- 2 4 Machida, T., Sawa, Y., Matsueda, H., Sweeney, C., Tans, P.P., tion FTS at Saga, Japan, and comparisons with GOSAT products. Andrews, A.E.,C,S.,Biraud,T.,Tanaka,S.K.,Patra,P.K.: Atmos. Meas. Tech. 8,5263–5276 (2015) Validation of XCO derived from SWIR spectra of GOSAT Rayner, P.J., O’Brien, D.M.: The utility of remotely sensed CO concen- TANSO-FTS with aircraft measurement data. Atmos. Chem. Phys. tration data in surface source inversions. Geophys. Res. Lett. 28, 13, 97719788–97719788 (2013). https://doi.org/10.5194/acp-13- 175178 (2001) 9771-2013 Rodgers, C. D. 2000: Inverse methods for atmospheric sounding: Theory Inoue, M., Morino, I., Uchino, O., Miyamoto, Y., Saeki, T., Yoshida, Y., and Practic, World Scientific Publishing Co. Pte. Ltd, Singapore, Yokota, T., Sweeney, C., Tans, P.P., Biraud, S.C., Machida, T., ISBN-10: 981–02-2740-X. Pittman, J.V., Kort, E.A., Tanaka, T., Kawakami, S., Sawa, Y., Rodgers and Connor: Intercomparison of remote sounding instruments. J. Tsuboi, K., Matsueda, H.: Validation of XCH derived from SWIR Geophys. Res. 108(405), 4116 (2003). https://doi.org/10.1029/ spectra of GOSAT TANSO-FTS with aircraft measurement data. 2002JD002299 Atmos. Meas. Tech. 7,2987–3005 (2014). https://doi.org/10.5194/ Tanaka, T., Miyamoto, Y., Morino, I., Machida, T., Nagahama, T., Sawa, amt-7-2987-2014. Y., Matsueda, H., Wunch, D., Kawakami, S., Uchino, O.: Aircraft Inoue, M., I. Morino, O. Uchino, T. Nakatsuru, Y. Yoshida, T. Yokota, D. measurements of carbon dioxide and methane for the calibration of Wunch, P. O. Wennberg, C. M. Roehl, D. W. T. Griffith, V. A. ground-based high-resolution Fourier transform spectrometers and a Velazco, N. M. Deutscher, T. Warneke, J. Notholt, J. John comparison to GOSAT data measured over Tsukuba and Moshiri. Robinson, V. Sherlock, F. Hase, T. Blumenstock, M. Rettinger, R. Atmos. Meas. Tech. 5, 20032012–20032012 (2012). https://doi.org/ Sussmann, E. Kyr, R. Kivi, K. Shiomi, S. Kawakami, M. De Mazire, 10.5194/amt-5-2003-2012 S. G. Arnold, D. G. Feist, E. A. Barrow, J. James Barney, M. Dubey, Wunch, D., Toon, G.C., Wennberg, P.O., Wofsy, S.C., Stephens, B.B., M.,Schneider, L.T.Iraci,J.R.Podolske,P. W. Hillyard,T. Fischer, M.L., Uchino, O., Abshire, J.B., Bernath, P., Biraud, S.C., Machida, Y. Sawa, K. Tsuboi, H. Matsueda, C. Sweeney, P. P. Blavier, J.-F.L., Boone, C., Bowman, K.P., Browell, E.V., Campos, Tans, A. E. Andrews, S. C. Biraud, Y. Fukuyama, J. V. Pittman, E. T., Connor, B.J., Daube, B.C., Deutscher, N.M., Diao, M., Elkins, A. Kort, and T. Tanaka, 2016: Bias corrections of GOSAT SWIR J.W., Gerbig, C., Gottlieb, E., Griffith, D.W.T., Hurst, D.F., Jimnez, XCO and XCH with TCCON data and their evaluation using 2 4 R., Keppel-Aleks, G., Kort, E.A., Macatangay, R., Machida, T., aircraft measurement data. Atmos. Meas. Tech., 9, 34913512. Matsueda, H., Moore, F., Morino, I., Park, S., Robinson, J., Roehl, Jain, A.K., Briegleb, B.P., Minschwaner, K., Wuebbles, D.J.: Radiative C.M., Sawa, Y., Sherlock, V., Sweeney, C., Tanaka, T., Zondlo, forcing and global warming potentials of 39 greenhouse gases. J. M.A.: Calibration of the Total carbon column observing network Geophys. Res. 105(D16, 380 August 27), 20773–20790 (2000) using aircraft profile data. Atmos. Meas. Tech. 3, 13511362– Keppel-Aleks, G., Wennberg, P.O., Schneider, T.: Sources of variations in 13511362 (2010). https://doi.org/10.5194/amt-3-1351-2010 total column carbon dioxide. Atmos. Chem. Phys. 11,3581–3593 Wunch, D., Toon, G.C., Blavier, J.-F.L., Washenfelder, R.A., Notholt, J., (2011). https://doi.org/10.5194/acp-11-3581-2011. Connor, B.J., Griffith, D.W.T., Sherlock, V., Wennberg, P.O.: The Kuze, A., Suto, H., Nakajima, M., Hamazaki, T.: 2009: thermal and near Total carbon column observing network. Philos. T. R. Soc. A. 369, infrared sensor for carbon observation Fourier-transform spectrom- 20872112–20872112 (2011). https://doi.org/10.1098/rsta.2010.0240 eter on the greenhouse gases observing satellite for greenhouse gas- Yang, Z., Washenfelder, R.R., Keppel-Aleks, G., Krakauer, N., es monitoring. Appl. Opt. 48,6716–6733 (2009) Randerson, J., Tans, P., Sweeney, C., Wennberg, P.: New constraints Messerschmidt, J., Geibel, M.C., Blumenstock, T., Chen, H., Deutscher, on northern hemisphere growing season net flux. Geophys. Res. N.M.,Engel, A.,Feist,D.G.,Gerbig,C.,Gisi,M.,Hase,F., Lett. 34, L12807 (2007). https://doi.org/10.1029/2007GL029742 Katrynski, K., Kolle, O., Lavric, J.V., Notholt, J., Palm, M., Yoshida, Y., Ota, Y., Eguchi, N., Kikuchi, N., Nobuta, K., Tran, H., Ramonet, M., Rettinger, M., Schmidt, M., Sussmann, R., Toon, Morino, I., Yokota, T.: Retrieval algorithm for CO and CH column G.C., Truong, F., Warneke, T., Wennberg, P.O., Wunch, D., 2 4 abundances from short-wavelength infrared spectral observations by Xueref-Remy, I.: Calibration of TCCON column-averaged CO : the greenhouse gases observing satellite. Atmos. Meas. Tech. 4, the first aircraft campaign over European TCCON sites. Atmos. 717–734 (2011). https://doi.org/10.5194/amt-4-717-2011 Chem. Phys. 11, 1076510777–1076510777 (2011). https://doi.org/ 10.5194/acp-11-10765-2011 Yoshida, Y., Kikuchi, N., Morino, I., Uchino, O., Oshchepkov, S., Bril, Miyamoto, Y., Inoue, M., Morino, I., Uchino, O., Yokota, T., Machida, T., A., Saeki, T., Schutgens, N., Toon, G.C., Wunch, D., Roehl, C.M., Sawa, Y., Matsueda, H., Sweeney, C., Tans, P.P., Andrews, A.E., Wennberg, P.O., Griffith, D.W.T., Deutscher, N.M., Warneke, T., Patra, P.K.: Atmospheric column-averaged mole fractions of carbon Notholt, J., Robinson, J., Sherlock, V., Connor, B., Rettinger, M., dioxide at 53 aircraft measurement sites. Atmos. Chem. Phys. 13, Sussmann, R., Ahonen, P., Heikkinen, P., Kyr, E., Mendonca, J., 52655275 (2013) Strong, K., Hase, F., Dohe, S., Yokota, T.: Improvement of the Morino, I., Uchino, O., Inoue, M., Yoshida, Y., Yokota, T., Wennberg, retrieval algorithm for GOSAT SWIR XCO and XCH and their 2 4 P.O., Toon, G.C., Wunch, D., Roehl, C.M., Notholt, J., Warneke, T., validation using TCCON data. Atmos. Meas. Tech. 6,1533–1547 Messerschmidt, J., Griffith, D.W.T., Deutscher, N.M., Sherlock, V., (2013). https://doi.org/10.5194/amt-6-1533-2013. Korean Meteorological Society http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png "Asia-Pacific Journal of Atmospheric Sciences" Springer Journals

Comparison of XCH 4 Derived from g-b FTS and GOSAT and Evaluation Using Aircraft In-Situ Observations over TCCON Site

Download PDF
13 pages

Loading next page...
 
/lp/springer-journals/comparison-of-xch-4-derived-from-g-b-fts-and-gosat-and-evaluation-EKHMz9LCn6

References (35)

Publisher
Springer Journals
Copyright
2019 The Author(s)
ISSN
1976-7633
eISSN
1976-7951
DOI
10.1007/s13143-019-00105-0
Publisher site
See Article on Publisher Site

Abstract

It is evident that evaluating the measurement of greenhouse gases (GHGs) obtained from multi-platform instruments against accurate and precise instrument such as aircraft in-situ is very essential when using remote sensing GHGs results for source/sink estimations with inverse modeling. The results of the inverse models are very sensitive even to small biases in the data (Rayner and O’Brien 2001). In this work, we have evaluated ground-based high resolution Fourier Transform Spectrometer (g-b FTS) and the Greenhouse gases Observing SATellite (GOSAT) column-averaged dry air mole fraction of methane (XCH ) through aircraft o o in-situ observations over Anmyeondo station (36.538 N, 126.331 E, 30 m above sea level). The impact of the spatial coinci- dence criteria was assessed by comparing GOSAT data against g-b FTS. We noticed there was no any systematic difference based on the given coincidence criteria. GOSATexhibited a bias ranging from 0.10 to 3.37 ppb, with the standard deviation from 4.92 to 12.54 ppb, against g-b FTS with the spatial coincidence criteria of ±1, ±3, ±5 degrees of latitude and longitude and ± 1 h time window. Data observed during ascent and descent of the aircraft is considered as vertical profiles within an altitude range of 0.2 to a maximum of 9.0 km so that some assumptions were applied for the construction of the profiles below 0.2 and above 9.0 km. In addition, the suitability of aircraft data for evaluation of remote sensing instruments was confirmed based on the assessment of o o uncertainties. The spatial coincidence criteria is ±1 latitude and ± 2 longitude and for temporal difference is ±1 h of the satellite observation overpass time were applied, whereas g-b FTS data are the mean values measured within ±30 min of the aircraft observation time. Furthermore, the sensitivity differences of the instruments were taken into account. With respect to aircraft, the g-b FTS data were biased by −0.19 ± 0.69%, while GOSAT data were biased by −0.42 ± 0.84%. These results confirm that both g-b FTS and GOSAT are consistent aircraft observations and assure the reliability of the datasets for inverse estimate of CH . . . . Keywords Aircraft in-situ G-b FTS GOSAT XCH 1 Introduction such as burning of fossil fuel and changes in land use. CH plays a major role in the chemistry of the Earth’satmosphere Next to CO , atmospheric methane (CH ) is one of the potent through the decomposition process; its increase is considered 2 4 greenhouse gases. Its concentrations have been increasing to change the balance of related chemical species (Cicerone since pre-industrial era as a result of intense human activities and Oremland, 1998). Therefore, accurate and precise mea- surements of CH play a substantial role for better compre- hension in global carbon cycle as well as its contribution to the Responsible Editor: Soon-Il An. global warming (Jain et al. 2000). Even though CH is the dominant anthropogenic greenhouse gas, there are still high * Samuel Takele Kenea uncertainties in CH sources and sinks at a global scale samueltake@yahoo.ca 4 (Frankenberg et al. 2008). A number of instruments deployed 1 onboard at various platforms (ground-based, airborne, and Climate Research Division, National Institute of Meteorological space-borne) have been involved in measuring of atmospheric Sciences (NIMS), 33, Seohobuk-ro, Seogwipo-si, Jeju-do 63568, Republic of Korea concentrations of CH . Aircraft in-situ measurements are highly accurate and precise and, therefore, capable to validate Department of Electrical Eng. and Center for Edge Plasma Science, Hanyang University, Seoul, Republic of Korea the ground-based instruments such as g-b FTS, as well as Korean Meteorological Society 416 S. T. Kenea et al. evaluate satellites instruments such as GOSAT. However, air- 3 Data and Methods craft observations are very sparse; this gap needs to be filled by highly accurate and precise measurements from satellite. 3.1 Aircraft Measurements Several studies on validation of remote sensing products of greenhouse gases such as CO and CH were conducted based To provide in-situ measurements of atmospheric CO ,CH , 2 4 2 4 on aircraft in-situ measurements at various sites (Araki et al. CO, and H O concentrations, the aircraft was equipped with a 2010; Messerschmidt et al. 2011; Geibeletal. 2012; Tanaka Wavelength-Scanned Cavity Ring Down Spectrometer et al. 2012;Miyamotoet al. 2013; Inoue et al. 2013, 2014). (CRDS; Picarro, G2401-mc) providing mixing ratio data re- The satellite products of CH should attain a demanding pre- corded at ~0.3 Hz intervals. The position (latitude, longitude, cision of <2% (< 34 ppb), in order to improve the precision of and height) of the aircraft was monitored by GPS, and infor- inversion models. In addition, achieving high relative accura- mation on the outside temperature, static pressure, and ground cy (< 10 ppb for XCH ) is more crucial and demanding than speed was provided by the aircraft’s instruments. Figure 2 precision to derive reliable surface fluxes via inverse model- displays schematic views of the CRDS instruments. Data col- ling (Buchwitz et al. 2016). lected during ascent and descent of the aircraft is considered as In this study, we have addressed two major issues. First, vertical profiles of CH over the Anmyeondo station. Typical observations of XCH and comparison between GOSAT and durations are in the range of 0.5–3 h. The temperature and g-b FTS XCH observations over the Anmyeondo station are pressure of the gas sample have to be tightly controlled at 45 discussed. Second, we assessed the suitability of aircraft in- C and 140 Torr in the CRDS (variations of less than 20 mK situ observations for validating other datasets and then evalu- and 0.1 Torr, respectively), which leads to highly stable spec- ate the correlative remote sensing measurements (g-b FTS and troscopic features (Chen et al. 2010). Any deviations from GOSAT) against aircraft. Here, the aircraft in-situ XCH was these values cause a reduction of the instrument’sprecision. computed based on the approach suggested by Inoue et al. Data recorded beyond these range of variations in cavity pres- (2013) and Ohyama et al. (2015). In fact, our g-b FTS was sure and temperature were discarded in this analysis. Variance calibrated with respect to TCCON (Total Carbon Column of the cavity pressure and temperature during flight results in Observing Network) common scale factor which is basically noise in the CH mixing ratios. The Picarro CRDS instrument derived from the aircraft data made at the other TCCON sites, has been regularly calibrated with respect to the standard gases therefore, our study will be useful for the scale factor exami- within the error range recommend by WMO. nation for Anmyeondo station by increasing the number of sample size. This paper is organized as follow: Section 2 gives 3.2 In-Situ Observation Data short descriptions about the Anmyeondo station. A brief over- view of data and method is provided in Section 3. Section 4 As for complementary information below the lower boundary presents results and discussion and followed by conclusions in of the aircraft observation, we utilized the CH concentration Section 5. data measured by the meteorological tower in the Anmyeondo ο ο site (36.53 N, 126.32 E, and 47 m above sea level), in close vicinity to the TCCON station. Atmospheric concentrations of 2 The Anmyeondo Station CH at 86 m above sea level were continuously measured with a precision <2 ppb using CRDS and provided as hourly aver- ο ο The Anmyeondo station is located at 36.538 N, 126.331 E, aged data. Sensors for detection of wind speed and direction, and 30 m above sea level. The topographic feature of the air temperature, barometric pressure, and humidity have been Anmyeondo station is a complex terrain which consists of also installed at this height. The in-situ data closest to the hills and valley within a few hundred meters. The climatic aircraft measurement time were selected to complement CH condition of the site is categorized as: winter is the coldest profiles. These data can be obtained from the WMO World season (a minimum temperature is about 2.7 C) while sum- Data Center for Greenhouse Gases (WDCGG) (http://ds.data. mer is the warmest season (a maximum temperature is about jma.go.jp/gmd/wdcgg/catalogue.cgi). 25.6 C), (the Anmyeondo station description is also given in Oh et al. (2018) paper). In other aspects, industries are avail- 3.3 Ground-Based FTS able within 100 km of the station. This area consists of agri- culture, forests, and urban areas. Several instruments are being The g-b FTS has been operating at Anmyeondo station within operated at the Anmyeondo station which makes this site is the network of TCCON since 2014; and detailed information important for validating remote sensing products from differ- about g-b FTS at this station was recently reported in Oh et al. ent platforms such as GOSAT. Figure 1 depicts all TCCON (2018). The TCCON is a worldwide network of ground-based sites (including operational, future, and previous sites) FTSs that was founded in 2004. It has been widely used as a globally. calibration and validation resource for satellite measurements Korean Meteorological Society Comparison of XCH Derived from g-b FTS and GOSAT and Evaluation Using Aircraft In-Situ Observations... 417 Fig. 1 Global distributions of the TCCON sites. (https://tccondata.org/) (e.g. Morino et al. 2011), but it is also offered for better un- sensors: the TANSO-FTS (Thermal And Near-infrared Sensor derstanding of the carbon cycle (e.g. Yang et al. 2007). The g- for carbon Observation Fourier Transform Spectrometer) and b FTS provides spectra in the near infrared spectrum with a TANSO-CAI (TANSO Cloud and Aerosol Imager) with the −1 high spectral resolution of 0.02 cm and a temporal resolu- IFOVof 10.5 km and 0.5–1.5 km, respectively. The TANSO- tion of ~2 min. From the recorded spectra, the target species FTS onboard GOSAT makes global observations both at nadir (XCH ) was retrieved with the GFIT nonlinear least-squares and off-nadir modes; and makes use of four spectral bands for fitting algorithm, which verifies a vertical scale factor (γ)of deriving CO and CH (Kuze et al. 2009) accurately. The 2 4 an a priori vertical profile based on the best spectral fit of the TANSO-FTS records the solar radiation reflected from the solar absorption signal. The scaled profile is then vertically surface at three Short Wavelength InfraRed (SWIR) bands at integrated, and the resulting column abundance is divided by the respective wavelength of 0.76, 1.6, and 2.0 μm and the the vertical column of dry air, calculated using the retrieved Earth’s radiation from the surface and atmosphere at the wide column of oxygen (O )(Wunch et al. 2011). Thermal InfraRed (TIR) band which is in the range between −1 5.5 and 14.3 μm with a resolution of 0.2 cm . The retrieval 3.4 Greenhouse Gases Observing SATellite (GOSAT) algorithm is a non-linear maximum a posteriori method with linear mapping based on Rodgers (2000). In this work, we have used column-averaged dry air mole fractions of CH GOSAT was launched into a sun-synchronous orbit on 23 (XCH ) V022 derived from the NIES retrieval algorithm January 2009 by H-IIA launch vehicle. It was placed in a (Yoshida et al. 2011, 2013). The achieved single measurement sun-synchronous orbit at a 666-km altitude and has a 3-day precision (random error) of the GOSAT XCH is revisit orbit cycle and a 12-day operation cycle. It carries two 4 Fig. 2 Left panel is CRDS instrument and right panel depicts the CRDS laboratory including gas cylinders for calibration purposes and computer Korean Meteorological Society 418 S. T. Kenea et al. approximately 16 ppb (1.0%) while systematic errors (relative derived from radiosonde observation and ERA-Interim accuracy or relative bias) are approximately 6 ppb (0.3%) reanalysis data were utilized for the calculation of the (Buchwitz et al. 2016). We accessed the GOSAT data through total column amounts of CH . The total column website (https://data2.gosat.nies.go.jp). amounts of CH were numerically integrated from the in-situ aircraft profiles weighted by dry air density from the surface up to the altitude of 70 km using the fol- 3.5 Methods lowing equation (e.g., Tanaka et al. 2012): The vertical profiles of CH mixing ratio are obtained dry f 1−f s CH H O during ascent and descent of the aircraft in spiral path 4 2 VC ¼ ∫ dp ð1Þ g:m (see Fig. 3) over Anmyeondo station. Since the altitude range of the aircraft measurements were limited to ap- where P is surface pressure, f -molefractionof H O proximately 0.2–5.0 km from 2012 to 2016 aircraft H O, f -molefractionof CH ,m – molecular mass 2 4 CH campaign and the maximum flight altitude was 9.0 km of air, and g - gravitational acceleration. The column in 2017, the in-situ data were utilized near the surface averaged dry-air mole fractions of CH are calculated to complement the CH profiles of aircraft-based data, from the integrated column amounts using the equation while above the aircraft ceiling, the highest altitude of given below: the aircraft observation data were extended up to the tropopause level to construct the complete CH profiles column CH XCH ¼ ð2Þ in a similar way as proposed by Miyamoto et al. column air (2013); Ohyama et al. (2015). We have used the in- where XCH is the column-averaged dry-air mole frac- situ data at the surface level to complement the aircraft 4 tion of CH . We set the coincidence criteria for compar- profile. Local planetary boundary layer (PBL) heights 4 ison between satellite data (GOSAT) and aircraft data as were obtained from National Centers for follows: GOSAT data are considered within ±1 degree Environmental Prediction (NCEP) NOAA (National latitude and ± 2 degree longitude boxes centered at the Oceanic and Atmospheric Administration) reanalysis da- Anmyeondo station and the aircraft data temporally ta. The mole fractions between the uppermost aircraft nearest to the GOSAT overpass time were selected, a measurement and the tropopause are assumed to be maximum of 1 h difference. While the g-b FTS data maintained constant as the highest aircraft measurements were averaged within ±30 min of the aircraft overpass, because at these higher altitudes the air is well mixed. which reduce the random error. During validation of For this analysis, the tropopause height was determined remote sensing vertical profiles, it is reasonable to con- from European Center for Medium range Weather sider the effect of vertical resolution and sensitivity of Forecasting (ECMWF) ERA-Interim reanalysis data, a the data (Rodgers and Connor 2003). In this analysis, horizontal resolution of 0.75 × 0.75 degrees. Above the vertically highly resolved aircraft CH profiles were de- tropopause height, GFIT a priori profiles were fixed to 4 graded by applying the column averaging kernels (a)of the aircraft data as shown by black dashed line in the j low resolution of vertical profiles derived from the left panel of Fig. 4. The dry air number density profiles Fig. 3 Typical flights path of the aircraft taken on May 09, 10, and 16, 2015 from left to right panels, respectively, are depicted. The right panel depicts vertical spiral flight path over Anmyeondo Korean Meteorological Society Comparison of XCH Derived from g-b FTS and GOSAT and Evaluation Using Aircraft In-Situ Observations... 419 Fig. 4 The left and right panels display the complete CH profile and tropopause and planetary boundary heights, respectively, and green aircraft profile appended with surface measurements, respectively, broken line shows the maximum aircraft observation altitude October 05, 2014. The red and blue broken lines represent the remote sensing instruments based on the following column to perturbations of the partial columns at the equation: various atmospheric levels. The typical column- averaging kernels for the g-b FTS and GOSAT at the in−situ a Anmyeondo station are shown in Fig. 5. The differences XCH ¼ XCH þ ∑ h aðÞ t −t ð3Þ j j in−situ a 4 4 of XCH between the aircraft and the remote sensing in−situ instruments or between g-b FTS and GOSAT are where XCH is the column-averaged dry-air mole expressed in terms of absolute and relative differences fraction of CH from the aircraft in-situ measurement, and the following mathematical expressions are: XCH and t are a priori column-averaged dry-air mole fraction and profile of CH , respectively (from g-b FTS Abso:diff : ¼ X −X ð4Þ or GOSAT), h is the pressure weighting function, and ins in−situ=FTS t is the in-situ profile from aircraft measurement. in − situ X −X ins in−situ=FTS The averaging kernel for the column retrieval is a vec- Rel:diff : ¼ 100% ð5Þ tor representing the sensitivity of the retrieved total in−situ=FTS where X and X are XCH of the aircraft and the in − situ/FTS ins 4 correlative remote sensing instruments (g-b FTS and GOSAT) or g-b FTS and GOSAT (which is used for the comparison be- tween g-b FTS and GOSAT), respectively. We examined the weather conditions at the measurement station during all obser- vation periods. For this period, we showed meteorological pa- rameters such as relative humidity, temperature, wind speed, and wind direction from radiosonde data in Fig. 6 in order to specu- late the weather conditions at the Anmyeondo station. Approximately above 3 km, the amount of atmospheric moisture was very low which was depicted by the relative humidity (RH) of below 40% in Fig. 6(a). As showninpanelscanddofFig. 6, the northerly and north easterly winds was blowing with a mag- −1 nitude of lower than 8 m s below 2 km, whereas between 2 and 6 km, the westerly and north westerly winds advected at a max- −1 imum speed of 20 m s over the measurement station. Therefore, this could bring a continental air mass from the Northern Hemisphere to the Anmyeondo station. Fig. 5 Column Averaging Kernels (CAKs) of CH are shown Korean Meteorological Society 420 S. T. Kenea et al. Fig. 6 Radiosonde measurements of (a) relative humidity, (b) temperature, (c) wind speed, and (d) wind direction, taken on October 05–07, 2014 4 Results and Discussion 4.1 Observations of XCH and Comparisons between g-b FTS and GOSAT In the following subsections, we have discussed obser- vations and comparison of XCH between g-b FTS and Here, the time series of XCH comparison between g-b 4 4 GOSAT,andthenevaluated them basedontheaircraft FTS and GOSAT was performed in the period between in-situ observations. 2014 and 2016 (see Fig. 7). We assessed to what extent Fig. 7 Time series of XCH obtained from the g-b FTS and the GOSAT at the spatial coincidence criteria of ±1, ±3, ±5 degrees latitude/longitude (left top panel) and GOSAT versus g-b FTS (right panel) in the period of green, blue, and red colors, respectively, and error bars indicate the 2014 to 2016 with a one-to-one dashed line, green circle denotes hourly standard deviations of the coincident datasets. Bottom left panel shows mean values of the g-b FTS, while the asterisks represent single day the time series of XCH on monthly mean basis and bottom right panel observations of the GOSAT. Right top panel shows g-b FTS vs GOSAT depicts annual cycle 2014–2016 Korean Meteorological Society Comparison of XCH Derived from g-b FTS and GOSAT and Evaluation Using Aircraft In-Situ Observations... 421 Table 1 Statistical results for the XCH difference between GOSAT and g-b FTS (bias = GOSAT - g-b FTS) data based on coincidence criteria of ±1, ±3, and ± 5 degrees of lat/lon and ± 1 h. N = number of coincident data, R = correlation coefficient Lat./ N R RMSE (ppb) bias ± std. (ppb) Rel. diff. ± std. (%) lon.(deg) ±1 9 0.86 5.73 3.37 ± 4.92 0.18 ± 0.27 ±3 16 0.96 6.01 1.68 ± 6.00 0.09 ± 0.33 ±5 27 0.76 12.31 0.10 ± 12.54 0.01 ± 0.68 the impact of spatial coincidence criteria affects the com- and enhanced concentrations over this region are predict- parison results. To match up the GOSAT data against g-b ed by model studies (Houweling et al. 2000). FTS, we chose geometric coincidence criteria of ±1, ±3, Dlugokencky et al. (1993) reported that significantly el- and ± 5 degrees of latitude/longitude centered at the evated CH mixing ratio was observed at Tae-ahn ο ο Anmyeondo station within the temporal window of Peninsula, Korea (36 44’ N, 126 08′ E) during summer ±1 h. Following the match-ups, all g-b FTS data coin- correlated with northwesterly airflow from northeast ciding with one satellite observation which were within China and east Siberia. In fact, taking a coarse colloca- 1 h time window are averaged, minimizing the g-b FTS tion criteria can induce collocation errors, but consider- random error. As a result of those coincidence criteria, ing a very strict criterion leads to a small sample set we obtained different sample size that might also affect (which affects statistics) due to the sparseness of the statistics to have robust conclusion. As can be seen GOSAT soundings. Buchwitz et al. (2017) described the in the top left panel of Fig. 7, the overall results sug- spatial variability in the bias of XCH termed as “relative gested that both instruments agreed in capturing the sea- bias”. It can arise from different surface reflectivity, aero- sonal variability of XCH over the Anmyeondo station. sol interference, and sloping terrain. They estimated Relatively large discrepancies were detected in peak 10 ppb relative bias for solar backscatter satellite obser- methane season, summer which might reflect the high vations. Correlations, root mean square error (RMSE), spatial heterogeneity of methane source and sink strength bias, and relative differences for these comparisons are (details are beyond the scope of this paper). The seasonal detailed in Tables 1 and 2 for varying collocation and annual cycles of XCH derived from the GOSAT criteria. As RMSE values increases from 5.73 to were compared with in g-b FTS observations over the 12.31 ppb with increasing the spatial window ±1 to ±5 Anmyeondo station, which are provided in the bottom degrees, which cover land and ocean parts since the panels of Fig. 7. Because of sample size, we used Anmyeondo station is located at the coastal area. In gen- GOSAT data that extracted within ±5 degrees latitude/ eral, there was no systematicdifferencenoticedby longitude coincidence criteria. As can be seen in Fig. 7 changing the collocation space. It was found low bias bottom panels, the overall patterns of seasonal and annu- (±σ) in GOSAT XCH against g-b FTS, which is about al cycle of the g-b FTS XCH are reproduced by GOSAT 3.37 ± 4.92 ppb, with a corresponding relative difference XCH . The maximum and minimum amounts of methane of 0.18 ± 0.27%, when applying the strict collocation were observed during summer and winter seasons, re- criteria within a time window of ±1 h and ± 1 degree of spectively. However, the seasonal cycle of CH in the latitude/longitude (see Table 1). Since we set the same Northern Hemisphere is more complex (Dlugokencky time window but changing the spatial window by ±3 and et al. 1994). While the destruction of CH due to reac- ± 5 degrees that resulted in the mean bias of 1.68 ± 6.0 tion with OH is expected to be stronger in summer, and 0.10 ± 12.54 ppb, respectively. The standard devia- source strengths also strongly vary with the seasons. tions of the differences are progressively increasing as East Asia is one of the largest source regions of methane, increasing the collocation space, but those values are Table 2 The same as Table 1,but Lat./lon.(deg) N R RMSE (ppb) bias ± std. (ppb) Rel. diff. ± std. (%) for daily mean basis. (bias = GOSAT minus g-b FTS) ±1 12 0.80 6.10 0.77 ± 6.30 0.04 ± 0.34 ±3 20 0.90 8.30 1.68 ± 8.34 0.09 ± 0.45 ±5 35 0.79 10.90 0.36 ± 10.98 0.04 ± 0.60 Korean Meteorological Society 422 S. T. Kenea et al. compatible with the combined measurement errors of the instruments. Similarly, we investigated the impact of the coincidence criteria by setting the time window on daily mean basis of g-b FTS measurements of XCH with varying the spatial coincidence ±1, ±3, and ± 5 degrees of latitude/longitude, and the bias was estimated to be less than 1.68 ppb. For a case of ±1 degree latitude/lon- gitude, GOSAT was biased by 0.78 ± 6.3 ppb with re- spect to g-b FTS. Those values are within the range of validation results reported in previous findings (e.g. Yoshida et al. 2013; Gavrilov et al. 2014;Ohyama et al. 2015) but with slightly smaller biases. Yoshida et al. (2013) performed a validation of GOSAT XCH (V02.xx) using the 723 measurements provided by Fig. 8 The comparisons of XCH between the aircraft observation versus TCCON and showed that bias was −5.90 ± 12.6 ppb. g-b FTS data (represented by blue square) and GOSAT (denoted by red Gavrilov et al. (2014) compared GOSAT XCH square) over Anmyeondo station are shown. The dashed line shows one- (V02.xx) with 256 ground-based FTS measurements ob- to-one line Table 3 Summary for the Instruments N Date (KST) Bias ± std. (ppb) Rel. diff. ± std. (%) statistics of XCH difference between remote sensing and aircraft in-situ data is given. The Aircraft vs. g-b FTS 2014-10-05 statistical estimators are 10:29:03–10:45:01 −14.60 −0.78 expressed in terms of bias and 10:46:10–11:00:01 −15.50 −0.83 relative differences with respect to 11:45:19–12:01:05 −13.00 −0.70 aircraft 13:23:10–13:38:21 −16.40 −0.88 13:39:30–13:53:33 −13.70 −0.74 14:23:46–14:48:23 −13.70 −0.74 14:49:40–15:04:30 −10.30 −0.56 2014-10-07 13:39:32–13:54:00 −8.00 −0.43 13:54:15–14:09:07 −4.50 −0.24 14:39:38–14:54:00 −8.70 −0.47 15:55:32–15:10:00 −0.30 −0.56 15:39:47–15:54:21 −6.50 −0.35 15:39:40–15:54:00 −11.60 −0.63 15:55:32–16:09:00 −11.00 −0.59 2017-10-29 09:59:16–10:31:08 0.90 0.05 10:31:09–11:03:24 4.90 0.26 12:58:58–13:37:07 15.10 0.82 13:37:07–14:19:40 −2.70 −0.14 2017-11-12 11:12:20–11:38:01 14.40 0.78 11:38:02–12:13:00 15.10 0.82 14:14:46–14:45:55 16.50 0.89 14:45:56–15:23:47 13.00 0.70 4 −3.66 ± 11.50 −0.19 ± 0.61 Aircraft vs. GOSAT (NIES V0221) Date 2012-10-17 4.53 0.25 2012-10-18 2.40 0.13 2014-10-05 −20.90 −1.13 3 −4.65 ± 14.11 −0.42 ± 0.84 Korean Meteorological Society Comparison of XCH Derived from g-b FTS and GOSAT and Evaluation Using Aircraft In-Situ Observations... 423 Fig. 9 The XCH total errors are shown for GOSAT (left panel) and for g-b FTS (right panel). (It was displayed only for the coincident days) tained near St. Petersburg, Russia and reported that the Aleks et al. 2011). The overall results indicated that g-b mean difference was −1.9 ± 14.5 ppb. Ohyama et al. FTS estimated slightly lower than aircraft. The statistical (2015) reported that the average differences XCH be- results for XCH comparisons between aircraft and g-b 4 4 tweenTANSO-FTSand g-bFTS data (TANSO-FTS mi- FTS are shown in Fig. 8 and summarized in Table 3,(cor- nus g-b FTS) is −7.6 ± 13.7 ppb. Right panel of Fig. 7 relation coefficient was not computed because of small demonstrates the results of XCH comparisons between sample numbers). The mean absolute difference of XCH 4 4 the GOSAT and g-b FTS at the spatial coincidence between aircraft and g-b FTS is −3.66 ± 11.50 ppb, with a criteria of ±1, ±3, ±5 degrees latitude/longitude green, corresponding mean relative difference of −0.19 ± 0.69%. blue, and red colors, respectively, and error bars indicate Previous findings have revealed that the unsampled part the standard deviations of the coincident datasets. This of the atmosphere above the aircraft ceiling contributes to depicts that the GOSAT data is well consistent with the the largest uncertainty in the total column calculated from g-b FTS that lie on the best line. Therefore, we can infer the aircraft profiles (Wunch et al. 2010). We estimated the that the impact the coincidence criteria (at least, up to error contributions on aircraft-based XCH . The error com- ±5) for performing the comparison of XCH over the ponents below the aircraft ceiling were derived by adding Anmyeondo station is insignificant. twice the precision of the aircraft data to the profile, and then re-integrated the profile (Wunch et al. 2010;Ohyama et al. 2015), the amount of error resulted in 3.03 ppb. The 4.2 Aircraft XCH Comparison with g-b FTS tropopause height variation induced an error of 0.40 ppb on averaged in estimating aircraft XCH and the stratospheric Several aircraft observation campaigns over Anmyeondo 4 error contribution were estimated by vertical shifting of the site were carried out in the period between 2012 and a priori by 1.0 km. That resulted in 6.30 ppb change on the 2017. However, a few numbers of aircraft data matched aircraft XCH values. The total errors were estimated to be with the remote sensing instruments were available during 4 7.0 ppb (Table 4). this observation period. The g-b FTS XCH was compared with the aircraft measurements. Here, g-b FTS data were averaged over a time window of ±30 min with respect to the aircraft measurement time. In addition, the averaging Table 4 Error budget for the estimated aircraft measurement of XCH . kernel of the g-b FTS was applied to the aircraft data to The total error is the sum, in quadrature, of the three errors. The aircraft equalize the sensitivities of CH mole fraction at each alti- error was estimated by adding the precision of the aircraft measurements to the profile and re-integrated the profile. Tropopause error was tude for the total column. Wunch et al. (2010)reported that determined by varying 1.0 km, while stratospheric error was estimated the airmass-dependent artifacts in XCH due to spectro- by shifting the stratospheric apriori profile of CH by 1.0 km. The fourth scopic inadequacies (e.g. line widths, inconsistencies in row on the given table shows the average value of the total retrieval errors the relative strengths of weak and strong lines) in from individual sounding of XCH from g-b FTS and GOSAT TCCON instruments were not seen. Here, a total number Aircraft error Tropopause error Stratospheric error Total of the aircraft measurements that matched with g-b FTS were only four during the observation period of 2014 to 3.03 ppb 0.40 ppb 6.30 ppb 7.00 ppb 2017. The diurnal range of g-b FTS data reflects not only g-b FTS total retrieval error GOSAT total retrieval error variability of airmass transport but source also sink process- 3.60 ppb 7.70 ppb es and the effect of measurement errors as well (Keppel- Korean Meteorological Society 424 S. T. Kenea et al. 4.3 Aircraft XCH Comparison with GOSAT such as aircraft in-situ is very essential when utilizing re- mote sensing GHGs results for source/sink estimations with In this section, the comparison of the GOSAT retrieval product inverse modeling. The results of the inverse models are very (V02.xx) of XCH against the aircraft observations over the sensitive even to small biases in the data (Rayner and Anmyeondo station was analyzed. Based on the coincidence O’Brien 2001). In this work, we carried out the comparison criteria, we obtained only three coincident days of observa- of XCH between g-b FTS and GOSAT over the tions. We took the mean values when obtaining more than one Anmyeondo station. Based on the comparison results be- GOSAT measurements. As noted the aircraft data was tween g-b FTS and GOSAT, both instruments are generally smoothed by GOSAT column averaging kernels. The compar- well captured the seasonal variability of XCH , the maxi- ison results of XCH between aircraft and GOSAT revealed a 4 mum and minimum amount of methane was observed dur- better agreement. The mean absolute difference of XCH was ing summer and winter seasons, respectively. The overall about −4.65 ± 14.11 ppb, also shown in Table 3. The absolute results indicate that a relatively high variability was exhibit- value of XCH difference on 5 October, 2014 was 20.9 ppb, ed during a peak methane season. In addition, the impact the which is larger than the other two coincident dates despite the coincidence criteria was assessed and there was no system- fact that the matching data were observed on the closest time atic difference was observed. The bias was estimated to be window. The discrepancy occurred at this particular date sig- from 0.1 to 3.37 ppb, and standard deviation was from 4.92 nificantly affected the mean of difference and standard devia- to 12.54 ppb. Column-averaged dry air mole fraction of CH tions. The difference could be attributable to the CH variabil- from 2012 to 2017 over the Anmyeondo station were de- ity in the lower atmosphere, the effect of aerosols/cirrus rived by using CH profiles measured by aircraft. Aircraft (Ohyama et al. 2015), or the large interval between air sam- measurements have good accuracy, but are limited in alti- pling levels, which is not sufficient to be captured the thin- tude floor and ceiling, and so we have to use additional layered structure of CH profiles by GOSAT. We tried to look information for surface and the stratosphere. As to my at atmospheric condition and aerosols during 5th October, knowledge, this is the first report on evaluation of remote 2015 using the information obtained from COMIS.4 (http:// sensing observations based on aircraft in-situ measurements uis.comis4.kma.go.kr/comis4/uis/common/index.do#), of XCH over this station. The uncertainty analysis of the Cloud-Aerosol Lidar and Infrared Pathfinder Satellite aircraft measurements of XCH confirmed the suitability of Observations (CALIPSO), and Aerosol Robotic Network data for evaluating the remote sensing products. These in- (AERONET) (see Appendix Figs. 10, 11, and 12). However, situ observations of the target species were compared against we noticed tropospheric aerosols that might be suggesting that g-b FTS, and GOSAT over there. It is noted that the aver- it tends to cause an underestimation of XCH retrievals. Still aging kernels of the remote sensing instruments were ap- further investigation is required in the future. However, the plied into the aircraft measurements. The retrieved XCH overall mean bias shown here is consistent with the previous values from the g-b FTS and GOSAT measurements showed results reported by Inoue et al. (2014, 2016). Inoue et al. a better agreement with the aircraft in-situ observations. (2014) made a comparison of GOSAT XCH (V02.00) with Both instruments revealed a negative bias against aircraft. 3 aircraft measurements from Yakutsk, Siberia and reported The relative differences of XCH were found to be −0.19 that the bias was 9.2 ± 15.2 ppb (3.7 ± 16.7 ppb) within ±2° ± 0.69% and − 0.42 ± 0.84% with respect to g-b FTS and (±5°). When considering 2 days average excluding a date of GOSAT, respectively. The small number of coincidences October 05, 2014, we obtained a bias 3.47 ± 1.51 ppb, which considered here hinders more robust conclusions, so we rec- is agreed with Inoue et al. (2016) who showed a 4.5 ± 15. ommend to carry out further works on validation by taking 20 ppb over land. Fig. 4 shows the total retrieval errors (the more coincident data using in-situ and remote sensing instru- room sum of the squares of smoothing error, retrieval noise, ments, as well as combining the model simulations in the and interference error components) of XCH obtained from future. This will allow us to improve clear identification of GOSAT and g-b FTS. The standard deviation of the differ- all the potential sources of uncertainties, as well as to under- ences that we obtained here is slightly larger than the total stand the role of local source/sink and dynamics. estimated error of GOSAT (see Fig. 9 and Table 4) and the aircraft XCH . Further work is still required to have more Acknowledgements This research was supported by the Research and robust conclusion by increasing the sample numbers. Development for KMA Weather, Climate, and Earth system Services Support to use of Meteorological Information and value Creation (KMA- 2018-00122). We acknowledge for those who provide the access for in-situ data from World Data Centre for Greenhouse Gases (WDCGG) (https://ds. 5 Conclusions data.jma.go.jp/gmd/wdcgg/cgi-bin/wdcgg/catalogue.cgi). We also greatly acknowledge the GOSAT science teams for the satellite data used in this Evaluating measurement of GHGs derived from multi- work. Many thanks goes to the science teams for the provision of ECMWF ERA-interim and NOAA NCAR reanalysis data utilized in this study. platform instruments against accurate and precise instrument Korean Meteorological Society Comparison of XCH Derived from g-b FTS and GOSAT and Evaluation Using Aircraft In-Situ Observations... 425 Appendix Fig. 10 Atmospheric sky conditions for 17th, 18th Oct. 2012, and 5th Oct. 2014, left to right panels, respectively, from COMIS.4. Red star shows Anmyeondo site. (http://uis.comis4.kma.go.kr/comis4/uis/common/index.do#) Fig. 11 Cloud and aerosol information from CALIOP data Korean Meteorological Society 426 S. T. Kenea et al. Fig. 12 Aerosol Optical Depth (AOD) from AERONET Level 2.0 data during 5th October, 2014 over Anmyeondo Open Access This article is distributed under the terms of the Creative Buchwitz, M., Schneising, O., Reuter, M., Heymann, J., Krautwurst, S., Bovensmann, H., Burrows, J.P., Boesch, H., Parker, R.J., Somkuti, Commons Attribution 4.0 International License (http:// P., Detmers, R.G., Hasekamp, O.P., Aben, I., Butz, A., Frankenberg, creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appro- C., Turner, A.J.: Satellite-derived methane hotspot emission esti- mates using a fast data-driven method. Atmos. Chem. Phys. 17, priate credit to the original author(s) and the source, provide a link to the 5751–5774 (2017). https://doi.org/10.5194/acp-17-5751-2017 Creative Commons license, and indicate if changes were made. Chen, H., Winderlich, J., Gerbig, C., Hoefer, A., Rella, C.W., Crosson, E.R., Van Pelt, A.D., Steinbach, J., Kolle, O., Beck, V., Daube, B.C., Publisher’sNote Springer Nature remains neutral with regard to juris- Gottlieb, E.W., Chow, V.Y., Santoni, G.W., Wofsy, S.C.: High accu- dictional claims in published maps and institutional affiliations. racy continuous airborne measurements of greenhouse gases (CO and CH ) using the cavity ring-down spectroscopy (CRDS) tech- nique. Atmos. Meas. Tech. 3,375–386 (2010) Cicerone, R.J., Oremland, R.S.: Biogeochemical aspects of atmospheric References methane. Glob. Biogeochem. Cycles. 2,299–327 (1988) Dlugokencky, E.J., Haris, J.M., Chung, Y.S., Tans, P.P., Fung, I.: The Araki, M., Morino, I., Machida, T., Sawa, Y., Matsueda, H., Ohyama, H., relationship between the methane seasonal cycle and regional Yokota, T., Uchino, O.: CO column-averaged volume mixing ratio 2 sources and sinks at Tae-ahn peninsula, Korea. Atmos. Environ. derived over Tsukuba from measurements by commercial airlines. Part A. 27,2015–2120 (1993) Atmos. Chem. Phys. 10, 7659–7667 (2010). https://doi.org/10. Dlugokencky, E.J., Steele, L.O., Lang, P.M., Masarie, K.A.: The growth 5194/acp-10-7659-2010. rate and distribution of atmospheric methane. J. Geophys. Res. 99, Buchwitz,M.,B.Dils,H.Boesch,C.Crevoisier, D.Detmers,C. 1702117044 (1994). https://doi.org/10.1029/94JD01245. Frankenberg, O. Hasekamp, W. Hewson, A. Laeng, S. Nol, J. Frankenberg, C., Bergamaschi, P., Butz, A., Houweling, S., Meirink, J.F., Notholt, R. Parker, M. Reuter, and O. Schneising, 2016: ESA Notholt, J., Petersen, A.K., Schrijver, H., Warneke, T., Aben, I.: Climate Change Initiative (CCI) Product Validation and Tropical methane emissions: a revised view from SCIAMACHY Intercomparison Report (PVIR) for the Essential Climate Variable onboard ENVISAT. Geophys. Res. Lett. 35, L15811 (2008). (ECV) Greenhouse Gases (GHG) for data set Climate Research https://doi.org/10.1029/2008GL034300 Data Package No. 3 (CRDP3), Version 4.0, 24. Feb. 2016. Korean Meteorological Society Comparison of XCH Derived from g-b FTS and GOSAT and Evaluation Using Aircraft In-Situ Observations... 427 Gavrilov, N.M., Makarova, M.V., Poberovskii, A.V., Timofeyev, Y.M.: Connor, B., Robinson, J., Sussmann, R., Rettinger, M.: Preliminary validation of column-averaged volume mixing ratios of carbon di- Comparisons of CH4 ground-based FTIR measurements near Saint Petersburg with GOSAT observations. Atmos. Meas. Tech. 7, 1003– oxide and methane retrieved from GOSAT short-wavelength infra- 1010 (2014). https://doi.org/10.5194/amt-7-1003-2014 red spectra. Atmos. Meas. Tech. 4,1061–1076 (2011). https://doi. Geibel, M.C., Messerschmidt, J., Gerbig, C., Blumenstock, T., Chen, H., org/10.5194/amt-4-1061-2011. Hase, F., Kolle, O., Lavriˇc, J.V., Notholt, J., Palm, M., Rettinger, Oh, Y.-S., Kenea, S.T., Goo, T.-Y., Chung, K.-S., Rhee, J.-S., Ou, M.-L., M., Schmidt, M., Sussmann, R., Warneke, T., Feist, D.G.: Byun, Y.-H., Paul Wennberg, O.P.P., Kiel, M., DiGangi, J.P., Diskin, Calibration of column-averaged CH over European TCCON FTS G.S., Velazco, V.A., Griffith, D.W.T.: Characteristics of greenhouse gas sites with airborne in-situ measurements. Atmos. Chem. Phys. 12, concentrations derived from ground-based FTS spectra at Anmyeondo, 8763–8775 (2012) South Korea. Atmos. Meas. Tech. 11(1–14), 2018 (2018) Houweling, S., Dentener, F., Lelieveld, J.: Simulation of preindustrial Ohyama, H., Kawakami, S., Tanaka, T., Morino, I., Uchino, O., Inoue, atmospheric methane to constrain the global source strength of nat- M., Sakai, T., Nagai, T., Yamazaki, A., Uchiyama, A., Fukamachi, ural wetlands. J. Geophys. Res. 105, 17243–17255 (2000) T., Sakashita, M., Kawasaki, T., Akaho, T., Arai, K., Okumura, H.: Inoue, M., Morino, I., Uchino, O., Miyamoto, Y., Yoshida, Y., Yokota, T., Observations of XCO and XCH with ground-based high-resolu- 2 4 Machida, T., Sawa, Y., Matsueda, H., Sweeney, C., Tans, P.P., tion FTS at Saga, Japan, and comparisons with GOSAT products. Andrews, A.E.,C,S.,Biraud,T.,Tanaka,S.K.,Patra,P.K.: Atmos. Meas. Tech. 8,5263–5276 (2015) Validation of XCO derived from SWIR spectra of GOSAT Rayner, P.J., O’Brien, D.M.: The utility of remotely sensed CO concen- TANSO-FTS with aircraft measurement data. Atmos. Chem. Phys. tration data in surface source inversions. Geophys. Res. Lett. 28, 13, 97719788–97719788 (2013). https://doi.org/10.5194/acp-13- 175178 (2001) 9771-2013 Rodgers, C. D. 2000: Inverse methods for atmospheric sounding: Theory Inoue, M., Morino, I., Uchino, O., Miyamoto, Y., Saeki, T., Yoshida, Y., and Practic, World Scientific Publishing Co. Pte. Ltd, Singapore, Yokota, T., Sweeney, C., Tans, P.P., Biraud, S.C., Machida, T., ISBN-10: 981–02-2740-X. Pittman, J.V., Kort, E.A., Tanaka, T., Kawakami, S., Sawa, Y., Rodgers and Connor: Intercomparison of remote sounding instruments. J. Tsuboi, K., Matsueda, H.: Validation of XCH derived from SWIR Geophys. Res. 108(405), 4116 (2003). https://doi.org/10.1029/ spectra of GOSAT TANSO-FTS with aircraft measurement data. 2002JD002299 Atmos. Meas. Tech. 7,2987–3005 (2014). https://doi.org/10.5194/ Tanaka, T., Miyamoto, Y., Morino, I., Machida, T., Nagahama, T., Sawa, amt-7-2987-2014. Y., Matsueda, H., Wunch, D., Kawakami, S., Uchino, O.: Aircraft Inoue, M., I. Morino, O. Uchino, T. Nakatsuru, Y. Yoshida, T. Yokota, D. measurements of carbon dioxide and methane for the calibration of Wunch, P. O. Wennberg, C. M. Roehl, D. W. T. Griffith, V. A. ground-based high-resolution Fourier transform spectrometers and a Velazco, N. M. Deutscher, T. Warneke, J. Notholt, J. John comparison to GOSAT data measured over Tsukuba and Moshiri. Robinson, V. Sherlock, F. Hase, T. Blumenstock, M. Rettinger, R. Atmos. Meas. Tech. 5, 20032012–20032012 (2012). https://doi.org/ Sussmann, E. Kyr, R. Kivi, K. Shiomi, S. Kawakami, M. De Mazire, 10.5194/amt-5-2003-2012 S. G. Arnold, D. G. Feist, E. A. Barrow, J. James Barney, M. Dubey, Wunch, D., Toon, G.C., Wennberg, P.O., Wofsy, S.C., Stephens, B.B., M.,Schneider, L.T.Iraci,J.R.Podolske,P. W. Hillyard,T. Fischer, M.L., Uchino, O., Abshire, J.B., Bernath, P., Biraud, S.C., Machida, Y. Sawa, K. Tsuboi, H. Matsueda, C. Sweeney, P. P. Blavier, J.-F.L., Boone, C., Bowman, K.P., Browell, E.V., Campos, Tans, A. E. Andrews, S. C. Biraud, Y. Fukuyama, J. V. Pittman, E. T., Connor, B.J., Daube, B.C., Deutscher, N.M., Diao, M., Elkins, A. Kort, and T. Tanaka, 2016: Bias corrections of GOSAT SWIR J.W., Gerbig, C., Gottlieb, E., Griffith, D.W.T., Hurst, D.F., Jimnez, XCO and XCH with TCCON data and their evaluation using 2 4 R., Keppel-Aleks, G., Kort, E.A., Macatangay, R., Machida, T., aircraft measurement data. Atmos. Meas. Tech., 9, 34913512. Matsueda, H., Moore, F., Morino, I., Park, S., Robinson, J., Roehl, Jain, A.K., Briegleb, B.P., Minschwaner, K., Wuebbles, D.J.: Radiative C.M., Sawa, Y., Sherlock, V., Sweeney, C., Tanaka, T., Zondlo, forcing and global warming potentials of 39 greenhouse gases. J. M.A.: Calibration of the Total carbon column observing network Geophys. Res. 105(D16, 380 August 27), 20773–20790 (2000) using aircraft profile data. Atmos. Meas. Tech. 3, 13511362– Keppel-Aleks, G., Wennberg, P.O., Schneider, T.: Sources of variations in 13511362 (2010). https://doi.org/10.5194/amt-3-1351-2010 total column carbon dioxide. Atmos. Chem. Phys. 11,3581–3593 Wunch, D., Toon, G.C., Blavier, J.-F.L., Washenfelder, R.A., Notholt, J., (2011). https://doi.org/10.5194/acp-11-3581-2011. Connor, B.J., Griffith, D.W.T., Sherlock, V., Wennberg, P.O.: The Kuze, A., Suto, H., Nakajima, M., Hamazaki, T.: 2009: thermal and near Total carbon column observing network. Philos. T. R. Soc. A. 369, infrared sensor for carbon observation Fourier-transform spectrom- 20872112–20872112 (2011). https://doi.org/10.1098/rsta.2010.0240 eter on the greenhouse gases observing satellite for greenhouse gas- Yang, Z., Washenfelder, R.R., Keppel-Aleks, G., Krakauer, N., es monitoring. Appl. Opt. 48,6716–6733 (2009) Randerson, J., Tans, P., Sweeney, C., Wennberg, P.: New constraints Messerschmidt, J., Geibel, M.C., Blumenstock, T., Chen, H., Deutscher, on northern hemisphere growing season net flux. Geophys. Res. N.M.,Engel, A.,Feist,D.G.,Gerbig,C.,Gisi,M.,Hase,F., Lett. 34, L12807 (2007). https://doi.org/10.1029/2007GL029742 Katrynski, K., Kolle, O., Lavric, J.V., Notholt, J., Palm, M., Yoshida, Y., Ota, Y., Eguchi, N., Kikuchi, N., Nobuta, K., Tran, H., Ramonet, M., Rettinger, M., Schmidt, M., Sussmann, R., Toon, Morino, I., Yokota, T.: Retrieval algorithm for CO and CH column G.C., Truong, F., Warneke, T., Wennberg, P.O., Wunch, D., 2 4 abundances from short-wavelength infrared spectral observations by Xueref-Remy, I.: Calibration of TCCON column-averaged CO : the greenhouse gases observing satellite. Atmos. Meas. Tech. 4, the first aircraft campaign over European TCCON sites. Atmos. 717–734 (2011). https://doi.org/10.5194/amt-4-717-2011 Chem. Phys. 11, 1076510777–1076510777 (2011). https://doi.org/ 10.5194/acp-11-10765-2011 Yoshida, Y., Kikuchi, N., Morino, I., Uchino, O., Oshchepkov, S., Bril, Miyamoto, Y., Inoue, M., Morino, I., Uchino, O., Yokota, T., Machida, T., A., Saeki, T., Schutgens, N., Toon, G.C., Wunch, D., Roehl, C.M., Sawa, Y., Matsueda, H., Sweeney, C., Tans, P.P., Andrews, A.E., Wennberg, P.O., Griffith, D.W.T., Deutscher, N.M., Warneke, T., Patra, P.K.: Atmospheric column-averaged mole fractions of carbon Notholt, J., Robinson, J., Sherlock, V., Connor, B., Rettinger, M., dioxide at 53 aircraft measurement sites. Atmos. Chem. Phys. 13, Sussmann, R., Ahonen, P., Heikkinen, P., Kyr, E., Mendonca, J., 52655275 (2013) Strong, K., Hase, F., Dohe, S., Yokota, T.: Improvement of the Morino, I., Uchino, O., Inoue, M., Yoshida, Y., Yokota, T., Wennberg, retrieval algorithm for GOSAT SWIR XCO and XCH and their 2 4 P.O., Toon, G.C., Wunch, D., Roehl, C.M., Notholt, J., Warneke, T., validation using TCCON data. Atmos. Meas. Tech. 6,1533–1547 Messerschmidt, J., Griffith, D.W.T., Deutscher, N.M., Sherlock, V., (2013). https://doi.org/10.5194/amt-6-1533-2013. Korean Meteorological Society

Journal

"Asia-Pacific Journal of Atmospheric Sciences" – Springer Journals

Published: Aug 1, 2019

There are no references for this article.