Combining XCO2 Measurements Derived from SCIAMACHY and GOSAT for Potentially Generating Global CO2 Maps with High Spatiotemporal Resolution

Tianxing Wang; Jiancheng Shi; Yingying Jing; Tianjie Zhao; Dabin Ji; Chuan Xiong

doi:10.1371/journal.pone.0105050

Abstract

Global warming induced by atmospheric CO₂ has attracted increasing attention of researchers all over the world. Although space-based technology provides the ability to map atmospheric CO₂ globally, the number of valid CO₂ measurements is generally limited for certain instruments owing to the presence of clouds, which in turn constrain the studies of global CO₂ sources and sinks. Thus, it is a potentially promising work to combine the currently available CO₂ measurements. In this study, a strategy for fusing SCIAMACHY and GOSAT CO₂ measurements is proposed by fully considering the CO₂ global bias, averaging kernel, and spatiotemporal variations as well as the CO₂ retrieval errors. Based on this method, a global CO₂ map with certain UTC time can also be generated by employing the pattern of the CO₂ daily cycle reflected by Carbon Tracker (CT) data. The results reveal that relative to GOSAT, the global spatial coverage of the combined CO₂ map increased by 41.3% and 47.7% on a daily and monthly scale, respectively, and even higher when compared with that relative to SCIAMACHY. The findings in this paper prove the effectiveness of the combination method in supporting the generation of global full-coverage XCO₂ maps with higher temporal and spatial sampling by jointly using these two space-based XCO₂ datasets.

Citation: Wang T, Shi J, Jing Y, Zhao T, Ji D, Xiong C (2014) Combining XCO₂ Measurements Derived from SCIAMACHY and GOSAT for Potentially Generating Global CO₂ Maps with High Spatiotemporal Resolution. PLoS ONE 9(8): e105050. https://doi.org/10.1371/journal.pone.0105050

Editor: Juan A. Añel, University of Oxford, United Kingdom

Received: January 3, 2014; Accepted: July 20, 2014; Published: August 13, 2014

Copyright: © 2014 Wang et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Funding: The work described in this paper has been jointly supported by project of “Climate Change: Carbon Budget and Related Issues” (Grant nr. XDA05040402) from Chinese Academy of Sciences (CAS), the CAS/SAFEA International Partnership Program for Creative Research Teams (Grant nr. KZZD-EW-TZ-09) and National Natural Science foundation of China (Grant nr. 41301177). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Competing interests: The authors have declared that no competing interests exist.

Introduction

In recent years, global warming caused by emission of CO₂ has attracted considerable attention from the public. During the past decade, although tremendous efforts have been made toward improving the understandings of the mechanism between CO₂ increase in the atmosphere and global warming, some uncertainties still exist in the spatiotemporal characteristics of CO₂ sinks/sources on regional and global scales due to the lack of high-density measurements of such variables with good accuracy [1], [2]. To date, the estimates of CO₂ flux from inverse methods rely mainly on ground-based measurements [3], [4]. Although providing highly accurate atmospheric CO₂ records, the traditional ground-based networks intrinsically suffer from sparse spatial coverage [2], [5]. Satellite-based measurements with various spatial and temporal resolutions provide a unique opportunity to accurately map atmospheric CO₂ in both daytime and nighttime over large areas, thus having the potential to bridge this gap. As a result, various satellite-based platforms have been equipped in recent years for deriving the CO₂ concentrations.

Generally, methods for retrieving CO₂ from space can be grouped into two categories: (1) inferring CO₂ concentrations by measuring shortwave infrared (SWIR) reflected solar radiation around 1.6 and 2.0 µm with sufficient spectral resolution. This includes the Greenhouse gases Observing SATellite (GOSAT), operating since 2009 [6], the Scanning Imaging Absorption spectrometer for Atmospheric CartograpHY (SCIAMACHY), in orbit since 2002 [7], and the second Orbiting Carbon Observatory (OCO-2), which, as a rebuild of OCO [8], [9], is planned to be launched in July 2014. In addition, CarbonSat will also be scheduled to be launched in 2018 (http://www.iup.uni-bremen.de/carbonsat/). These measurements have a nearly uniform sensitivity to CO₂ from the surface up through the middle troposphere, and thus are frequently used to derive the column-average dry air mole fraction of atmosphere CO₂ (XCO₂) during the daytime; (2) retrieving CO₂ concentrations by interpreting the recorded spectra of the Earth-atmosphere system in thermal infrared (TIR) bands (around 15 µm). Instruments that work in such a way include AIRS [10], [11], IASI [12], [13], and FTS (Band 4) of GOSAT [6]. These measurements bring the advantage that they can detect CO₂ during both day and night time, while the lack of sensitivity in the lower troposphere makes them inappropriate to estimate CO₂ near the surface where the largest signals of CO₂ sources and sinks occur [1]. The complementarities of these platforms allow us to combine the SWIR and TIR measurements for obtaining enhanced understanding of CO₂ spatiotemporal variations globally. Since XCO₂ is much less affected by vertical transport of CO₂, it is particularly useful for investigation of CO₂ sources and sinks using inversion modeling [14], [15]. On the other hand, the spatial and temporal variations in XCO₂ are even smaller than that in the surface CO₂; therefore, unprecedented measurement precision and accuracy are highly required for such column measurements [16]–[19]. SCIAMACHY (operation stopped in April 2012) and GOSAT are two typical instruments that can be used to derive XCO₂ from space, and a variety of retrieval algorithms have been developed for SCIAMACHY [1], [20]–[27] and GOSAT [2], [4], [5], [28]–[30] with eyes on improving XCO₂ retrieval accuracy to a great extent. At present, a number of XCO₂ products have been released. These will definitely enhance our understanding of the global carbon cycle.

Unfortunately, almost all typical instruments currently used to derive atmospheric CO₂ concentration are working in the infrared spectral range (less than 16 µm). Thus, except for the instrument’s observation mode (for example, GOSAT observes in lattice points), the spatial coverage of the derived CO₂ is severely restricted by the presence of clouds. In addition, the lower signal-to-noise level over ice/snow covered surfaces and ocean for SWIR instruments (e.g., SCIAMACHY) also contributes to the CO₂ sparse coverage. For instance, it has been pointed out that only about 10% of GOSAT data can be used for retrieval of XCO₂ due to the cloud contaminations [4]. The amount of CO₂ measurements will be even smaller if additional screening criteria such as quality of spectral fit, aerosol loadings, etc. are further applied. Although the amount of remaining CO₂ measurements from certain space-based instruments may largely surpass that of ground-based sites, it is still not sufficient enough for accurately quantifying the spatiotemporal distribution of CO₂ over the global scale. As a result, it is greatly desired to jointly use these available CO₂ measurements derived from various space-based data. Recently, a novel method has been proposed for combining CO₂ values from seven different algorithms, and a new Level-2 CO₂ database (EMMA) from one algorithm is composed according to the median of monthly average of seven CO₂ products in each 10°×10° latitude/longitude grid box [31]. In fact, this method cannot increase the number of CO₂ observations but chooses a product with moderate oscillation among the available products. Despite the usefulness of the XCO₂ measurements (Level 2) in their own right, further spatiotemporal analysis for interpreting their scientific merit is essentially necessary due to the retrieval uncertainties and sparse coverage of such Level-2 observations [32]. For this point, many works have attempted to generate global full-coverage (i.e., Level 3) maps from XCO₂ values derived from single satellite observations using a geospatial statistics approach [32]–[34]. However, as reflected in these studies (for instance, Fig. 1 in the work of [33]), a compromise has to be made between the interpolated accuracy and the spatiotemporal resolution of Level-3 product because of the limited amount of Level-2 XCO₂ observations being used. For this point, instead of using Level-2 XCO₂ from a single dataset (e.g., GOSAT or OCO-2) as performed in the existing literature, we attempt to explore the potential of combining two CO₂ datasets (GOSAT and SCIAMACHY) in assisting in global Level-3 generation, aiming to: (1) propose a general strategy for combining (fusing) various CO₂ datasets with different instruments, algorithms, averaging kernels, etc.; and 2) increase the number of daily CO₂ points (utilized in Level-3 map interpolations) through the combination of two datasets, so that potentially improved Level-3 maps with higher accuracy and shorter time scale can be generated. The better the interpretation of the satellite-based CO₂ observations one can make, the higher the resolution (both temporal and spatial) of the generated global CO₂ maps.

Download:

Figure 1. Validation of the BESD products against in situ TCCON CO₂ measurements over globe for 2009–2010.

https://doi.org/10.1371/journal.pone.0105050.g001

Datasets

For GOSAT, the Fourier transform spectrometer (FTS) on GOSAT is the fundamental unit to retrieve atmospheric CO₂ and CH₄. It observes sunlight reflected from the earth’s surface, and light emitted from the atmosphere and the surface. It is composed of three narrow bands in the SWIR region (0.76, 1.6, and 2.0 µm) and a wide TIR band (5.5–14.3 µm) at a spectral and spatial resolution of 0.2 cm⁻¹ and 10.5 km, respectively [35]. Specifically, four CO₂ products from GOSAT have currently been released to the public: University of Leicester product [9], [36], the RemoTeC product [28], NIES GOSAT product [35] and the product generated by NASA’s Atmospheric CO₂ Observations from Space (ACOS) team (hereafter called ACOS product) [2], [30]. The difference between some of the above mentioned products with various versions have been investigated in a recent study [37]. In the present paper, the ACOS product of 2009–2010 with version v2.9 has been employed.

SCIAMACHY was successfully launched on board Environmental Satellite (ENVISAT) in 2002 (unfortunately ceased in April 2012), which is a detector elements satellite spectrometer covering the spectral range 0.24–2.38 µm with a moderate spectral resolution of about 0.2–1.6 nm, and spatial resolution at nadir of 60×30 km [7]. It has eight spectral channels, with 1024 individual detector diodes for each band, observing the spectral regions 0.24–1.75 µm (band 1–6), 1.94–2.04 µm (band 7), and 2.26–2.38 µm (band 8) simultaneously in nadir and limb and solar and lunar occultation viewing geometries [22]. As mentioned in Section 1, till today, a number of CO₂-retrieval algorithms have been developed for SCIAMACHY. The IUP/IFE of University of Bremen has released two XCO₂ products, i.e., WFM-DOAS product [21], [22] and the Bremen Optimal Estimation DOAS (BESD) product [1], [26]. In this study, the BESD product with the versions of v02.00.08 for 2009–2010 is used.

In addition, CO₂ profiles of CT [38] are also collected here to allow the data mentioned above to be properly fused. CT is a NOAA data assimilation system, which provides the 3D profiles of CO₂ mole fractions in the atmosphere over the globe. For this study, CT data with version CT2011 is collected. This dataset provides global CO₂ profiles with 3°×2° latitude/longitude grid and 3 hours temporal resolution (a total 8 times from 01 to 22 in UTC) spanning the time period from January 2000 to December 2010. The CT dataset is used here mainly to assist in adjusting and time-shifting of the two CO₂ products being combined.

Methodologies

For combining the different space-based CO₂ measurements, three steps are adapted in this study. First, taking the global ground measurements of CO₂ as reference, remove the bias of the individual CO₂ retrievals for ensuring the accuracy of the fused CO₂ product; then make some adjustment for both the ACOS and BESD products, so that they can be physically comparable and thus combined; finally fuse the ACOS and BESD CO₂ products considering their retrieval uncertainties, spatial scales, differences in averaging kernels and overpass times, etc.

3.1 Global bias corrections

Removal of any global bias of the retrieved CO₂ when compared with the ground in situ measurements is essential before performing joint use. Many researches [4], [25] frequently pointed out that CO₂ retrievals from GOSAT are low biased with different levels due to the uncertainties in pressure, radiometric calibration, line shape model, cloud and aerosol scattering, etc. Fortunately, a recent study has proposed a method for evaluating systematic errors in CO₂ and showed that the new version of ACOS product (v2.9) has a low global bias (<0.5 ppm) [39]. Thus, there is no global bias correction for the ACOS product being conducted here, but only the ACOS retrievals that pass the filter of table B1 in the work of [38] and marked as “good” in the quality flag are used. For the BESD product, we select Total Carbon Column Observing Network (TCCON) [15] measurements for 2009–2010 as the ground truth to determine its global bias. Specifically, BESD retrievals within ±2.5° and ±2.5° latitude/longitude box centered at each TCCON site and the mean FTS value (within ±1 h time window of satellite overpass time) are extracted and compared (totally ten TCCON sites are utilized). The coincidence criteria mentioned above ultimately yield a total of 338 pairs of CO₂ measurements. The comparison result is shown in Fig. 1.

3.2 Retrieval adjustments

As pointed out by most researchers, it is not reasonable to directly compare or use two XCO₂ measurements. A suitable way to do that is to take the a priori profiles and variations in averaging kernel into account during the comparison [26], [40]. To tackle the a priori issue, after correcting their global biases, both BESD and ACOS products are adjusted for a common a priori profile, which we assume to be the CT profile interpolated at the middle of the two overpass times (Equation (1)). Specifically, the a priori CO₂ profile of both the ACOS and BESD are first interpolated or extrapolated to the level of the CT CO₂ profile according to their pressure layers. After interpolation, the a priori profiles for both ACOS and BESD have the same dimension as the CT profile. Here the reason we take the CT profile at the middle of the two overpass times is that the time difference for GOSAT (1∶00 pm) and SCIAMACHY (10∶00 am) is relative large (3 hours), if we take one satellite time as reference, the induced error would be large for the other satellite measurements considering the CO₂ natural diurnal variation. So a middle time between these two satellite overpass times is selected for minimizing the CO₂ uncertainties during the adjustment.(1)Here, is the adjusted XCO₂ for ACOS or BESD; corresponds to retrieved XCO₂ of ACOS or BESD; a is the column-averaging kernel (row vector) of ACOS or BESD; h is pressure-weighting function (column vector); I is an identity matrix; and (column vectors) are the common CT CO₂ profile and the corresponding a priori CO₂ profile for ACOS or BESD, respectively.

While it is not trivial to accurately consider the smoothing error without an estimate of the true atmospheric variability which is generally not readily available for most cases [39]. Fortunately, some works revealed that the smoothing error is generally small [26], [39]. Consequently, for the remainder of this paper, only the adjustment in Equation (1) is applied for both the ACOS and BESD CO₂ products (after bias corrections).

3.3 Combination and time shifting

Based on the processes described above, the world is divided into a number of 0.5°×0.5°latitude/longitude grid box (totally 720×360). For each grid cell, Equation (2) is used to combine the corresponding CO₂ measurements within that grid.(2)where is the combined XCO₂; m is the total number of space-based CO₂ retrievals (ACOS and/or BESD) within a certain grid; is the ith XCO₂ retrieval in a grid for which the global bias and Equation (1) are supposed to be applied; is the ratio of uncertainty of the ith XCO₂ retrieval to its XCO₂ value.

Please note that since different CO₂ retrievals have distinct overpass times, it is necessary to unify them to avoid uncertainties induced from the time discrepancy before fusion. To this end, a method for considering the CO₂ shifting along time has been developed (Equation (3)). First, designate a specific time or select one overpass time as reference, then transfer CO₂ measurements at various overpass times to that of the reference time by interpolating the CT CO₂ at temporal scale. Here, it should be pointed out that despite the CO₂ absolute values of CT not being accurate enough, the daily cycle pattern of atmospheric CO₂ it reflects is assumed to be correct.(3)Here, is the transformed XCO₂ (ACOS or BESD) at the reference time; is the retrieved XCO₂ from ACOS or BESD at overpass time t; and are CO₂ profiles of CT at times of reference and t, respectively; is the pressure-weighting vector (column vector).

Based on the time-shifting strategy proposed here, a global CO₂ map at any specific time can be theoretically produced by employing the pattern of the CO₂ daily cycle reflected by CT data. For instance, we can unify all XCO₂ retrievals being combined with various overpass times to that of UTC = 1.

Results

Evaluation analysis showed that the global bias for the BESD product is generally small. In this study, the bias of the BESD product is corrected by subtracting 0.6 ppm from all XCO₂ values according to the results in Fig. 1. Although the systematic bias of the XCO₂ retrievals is removed, it is supposed that the error characteristics (random error) within the data are still unchanged. The bias-corrected XCO₂ retrievals of both ACOS and BEDS are used as fundamental data for the combination algorithm.

By applying the series of processes shown in Section 3, daily, weekly, as well as monthly maps of combined XCO₂ for 2009 and 2010 are generated. Here, as an example, only four maps (from May to August) of monthly mean XCO₂ of 2010 are shown here (Fig. 2–Fig. 5). In addition, the total XCO₂ uncertainties of the combined product which mainly depend on the uncertainties of the original ACOS or BESD XCO₂ retrievals are also illustrated.

Download:

Figure 2. XCO₂ monthly mean maps in May of 2010.

((a) ACOS XCO₂, (b) BESD XCO₂, (c) combined product, and (d) XCO₂ uncertainties of the combined product).

https://doi.org/10.1371/journal.pone.0105050.g002

Download:

Figure 3. XCO₂ monthly mean maps in June of 2010.

((a) ACOS XCO₂, (b) BESD XCO₂, (c) combined product, and (d) XCO₂ uncertainties of the combined product).

https://doi.org/10.1371/journal.pone.0105050.g003

Download:

Figure 4. XCO₂ monthly mean maps in July of 2010.

((a) ACOS XCO₂, (b) BESD XCO₂, (c) combined product, and (d) XCO₂ uncertainties of the combined product).

https://doi.org/10.1371/journal.pone.0105050.g004

Download:

Figure 5. XCO₂ monthly mean maps in August of 2010.

((a) ACOS XCO₂, (b) BESD XCO₂, (c) combined product, and (d) XCO₂ uncertainties of the combined product).

https://doi.org/10.1371/journal.pone.0105050.g005

From Fig. 2–Fig. 5, it is not difficult to observe that the combined data realize the physical complementary of the two products in terms of spatial coverage. The number of valid CO₂ measurements in the fused product is the union of the CO₂ data from both the ACOS and BESD at the same geographical location. In addition, the combined XCO₂ demonstrates similar spatiotemporal characteristics with that of ACOS and BESD over the globe, which implies that all processes associated with the combination do not distort the essential information of the original XCO₂ products (ACOS or BESD). Similar findings can also be observed in the daily mean and weekly mean XCO₂ maps. To quantitatively investigate the improvement of fused XCO₂ in spatial coverage, the fractional coverage of all three variables (ACOS, BESD, and combined XCO₂) on both daily and monthly scales is calculated (Fig. 6). From Fig. 6, it can be seen that the average global coverage of ACOS and BESD is around 0.46% and 0.21%, respectively, on a daily scale. The monthly mean coverage of such products accounts for about 5.70% and 3.75%, respectively. While spatial coverage of combined XCO₂ can reach up to 0.65% and 8.42% on daily and monthly scales, respectively, it accounts for increments of 41.3% and 47.7% on the daily and monthly scales with respect to that of GOSAT and it is even higher relative to the coverage of SCIAMACHY. Likewise, the cumulative fraction of coverage of the combined XCO₂ has risen to 25% when compared with 20% and 13% for ACOS and BESD, respectively. The increase in the XCO₂ spatial coverage indicates the potential advantage of the combined XCO₂ observations in generating global Level-3 XCO₂ maps when compared with any single dataset by providing more satellite-based XCO₂ retrievals used for optimal interpolating.

Download:

Figure 6. XCO₂ fraction of coverage of ACOS, BESD, and combined products.

(a) Daily coverage. (b) Monthly coverage. (c) Cumulative coverage.

https://doi.org/10.1371/journal.pone.0105050.g006

For evaluating the performance of our combination strategy, the combined XCO₂ values are compared with that retrieved from ACOS and BESD as well as XCO₂ in the EMMA database at two TCCON sites (Fig. 7). The results reveal that the XCO₂ values from the combination method show generally consistent variation in time with TCCON measurements except for a small overall bias (especially for the Lamont site). On the whole, the new combined XCO₂ product shows good consistency with the EMMA data, and they are comparable in terms of CO₂ magnitude, while the combined XCO₂ are shown with a longer time period, which is in line with the satellite observations, and possess more data points even over the same period.

Download:

Figure 7. Comparison of XCO₂ measurements from TCCON, ACOS, BESD, EMMA, and our new combination method over Wollongong and Lamont sites (distance<0.25 degree, temporal difference<1 hour).

https://doi.org/10.1371/journal.pone.0105050.g007

Discussions and Conclusions

Despite the fact that space-based measurements can provide a unique opportunity to map atmospheric CO₂ over large areas, the number of valid CO₂ measurements from a single space-based instrument is generally limited for a certain day over a specific region due to the presence of clouds. In addition, although these Level-2 XCO₂ retrievals themselves are very important for inversion modeling of surface carbon sources/sinks, further comprehensive analysis by investigating the spatiotemporal full-coverage XCO₂ (Level 3) distribution is needed for interpreting their significant scientific merit [32]. While the limited satellite observations restrict the generation of Level-3 XCO₂ maps with high spatial and temporal resolutions when only a single satellite-based XCO₂ dataset is considered. This is our main motivation in this paper.

In this study, a strategy for combining SCIAMACHY and GOSAT CO₂ measurements has been proposed by fully accounting for the CO₂ global bias, differences in averaging kernels and overpass times, and the Level-2 retrieval errors of the CO₂ measurements being used. The results indicated that the average global coverage of both ACOS and BESD is less than 0.5% on a daily scale, and less than 6% on a monthly scale. While spatial coverage of combined XCO₂ can reach up to 0.65% and 8.42% on daily and monthly scales, respectively, the comparison analysis reveals that the combined XCO₂ product is consistent with TCCON and EMMA in both temporal variation and magnitude except for a small bias when compared with the TCCON measurements. All these findings herein prove the effectiveness of the combination method in supporting generation global full-coverage XCO₂ maps with higher temporal and spatial sampling by jointly using two space-based XCO₂ datasets. Similar to the existing studies (e.g. [32]–[34]), although these combined XCO₂ are not intended to be used in inverse modeling studies, they deliver a key complement for such research, and can be deemed as an independent dataset for comparison with model predictions. Similar to the existing study [31], an improved fusion approach (based on multiple XCO₂ datasets) to create Level-2 XCO₂ measurements that can be directly used for inverse modeling is also attempted and will be presented in another paper.

A last point that needs to be addressed is that although we employed CO₂ data of GOSAT and SCIAMACHY in this study, the proposed strategies are not restricted to such data. As a general strategy, it can be refined and adapted to further combine other XCO₂ products, such as OCO-2, CarbonSat, etc. in the future, and even to be applied to the fusion of other trace gases, such as O₃, CH₄.

Acknowledgments

The authors would like to thank the SCIAMACHY team at University of Bremen IUP/IFE as well as the ACOS scientific teams for providing us the CO₂ products. The authors also thank the anonymous reviewers for their helpful and valuable comments to improve this work.

Author Contributions

Conceived and designed the experiments: TW JS YJ TZ DJ CX. Performed the experiments: TW JS YJ TZ DJ CX. Analyzed the data: TW JS YJ TZ DJ CX. Contributed reagents/materials/analysis tools: TW JS YJ TZ DJ CX. Wrote the paper: TW JS YJ TZ DJ CX.

References

1. Reuter M, Bovensmann H, Buchwitz M, Burrows JP (2010) A method for improved SCIAMACHY CO₂ retrieval in the presence of optically thin clouds. Atmos. Meas. Tech 3: 209–232.
- View Article
- Google Scholar
2. O’Dell CW, Connor B, Bösch H, O’Brien D, Frankenberg C, et al. (2012) The ACOS CO₂ retrieval algorithm – Part 1: Description and validation against synthetic observations. Atmos Meas Tech 5(1): 99–121.
- View Article
- Google Scholar
3. Baker DF, Law RM, Gurney KR, Rayner P, Peylin P, et al. (2006) TransCom 3 inversion intercomparison: Impact of transport model errors on the interannual variability of regional CO₂ fluxes, 1988–2003. Global Biogeochem Cy 20: GB1002
- View Article
- Google Scholar
4. Morino I, Uchino O, Inoue M, Yoshida Y, Yokota T, et al. (2011) Preliminary validation of column-averaged volume mixing ratios of carbon dioxide and methane retrieved from GOSAT short-wavelength infrared spectra. Atmos Meas Tech 4: 1061–1076.
- View Article
- Google Scholar
5. Butz A, Hasekamp OP, Frankenberg C, Aben I (2009) Retrievals of atmospheric CO₂ from simulated space-borne measurements of backscattered near-infrared sunlight: accounting for aerosol effects. Appl Opt 18: 3322–3336.
- View Article
- Google Scholar
6. Kuze A, Suto H, Nakajima M, Hamazaki T (2009) Thermal and near infrared sensor for carbon observation Fourier-transform spectrometer on the Greenhouse Gases Observing Satellite for greenhouse gases monitoring, Appl Opt. 35: 6716–6733.
- View Article
- Google Scholar
7. Bovensmann H, Burrows JP, Buchwitz M, Frerick J, Noel S, et al. (1999) SCIAMACHY –Mission objectives and measurement modes. J Atmos Sci 56: 127–150.
- View Article
- Google Scholar
8. Crisp D, Atlas RM, Breon FM, Brown LR, Burrows JP, et al. (2004) The Orbiting Carbon Observatory (OCO) mission. Adv Space Res 34: 700–709.
- View Article
- Google Scholar
9. Boesch H, Baker D, Connor B, Crisp D, Miller C (2011) Global Characterization of CO₂ Column Retrievals from Shortwave-Infrared Satellite Observations of the Orbiting Carbon Observatory-2 Mission. Remote Sens 3: 270–304.
- View Article
- Google Scholar
10. Aumann HH, Chahine MT, Gautier C, Goldberg MD, Kalnay E, et al. (2003) AIRS/AMSU/HSB on the Aqua Mission: Design, science objectives, data products, and processing systems. IEEE Trans Geosci Remote Sens 41: 253–264.
- View Article
- Google Scholar
11. Chahine M, Barnet C, Olsen ET, Chen L, Maddy E (2005) On the determination of atmospheric minor gases by the method of vanishing partial derivatives with application to CO₂. Geophys Res Lett 32: L22803
- View Article
- Google Scholar
12. Phulpin T, Cayla F, Chalon G, Diebel D, Schlussel P (2002) IASI on board Metop: Project status and scientific preparation. Proceedings of the 12th International TOVS Study Conference. Lorne, Vi ctoria, Australia.
13. Turquety S, Hadji-Lazaro J, Clerbaux C, Hauglustaine DA, Clough SA, et al. (2004) Operational trace gas retrieval algorithm for the Infrared Atmospheric Sounding Interferometer. J Geophys Res 109: D21301
- View Article
- Google Scholar
14. Yang Z, Washenfelder RA, Keppel-Aleks G, Krakauer NY, Randerson JT, et al. (2007) New constraints on Northern Hemisphere growing season net flux. Geophys Res Lett 34: L12807
- View Article
- Google Scholar
15. Wunch D, Toon GC, Blavier JFL, Washenfelder R, Notholt J, et al. (2011) The Total Carbon Column Observing Network. Philos T Roy Soc A 369: 2087–2112.
- View Article
- Google Scholar
16. Rayner PJ, O’Brien DM (2011) The utility of remotely sensed CO₂ concentration data in surface source inversions. Geophys Res Lett 28: 175–178.
- View Article
- Google Scholar
17. Houweling S, Breon FM, Aben I, Rodenbeck C, Gloor M, et al. (2004) Inverse modeling of CO₂ sources and sinks using satellite data: a synthetic inter-comparison of measurement techniques and their performance as a function of space and time. Atmos Chem Phys 4: 523–538.
- View Article
- Google Scholar
18. Olsen SC, Randerson JT (2004) Differences between surface and column atmospheric CO₂ and implications for carbon cycle research. J Geophys Res 109; D02301. doi:10.1029/2003JD003968.
19. Miller CE, Crisp D, DeCola PL, Olsen SC, Randerson JT, et al. (2007) Precision requirements for space-based XCO₂ data. J Geophys Res 112: D10314
- View Article
- Google Scholar
20. Buchwitz M, Beek R, Burrows JP, Bovensmann H, Warneke T, et al. (2005) Atmospheric methane and carbon dioxide from SCIAMACHY satellite data: initial comparison with chemistry and transport models, Atmos Chem Phys. 5: 941–962.
- View Article
- Google Scholar
21. Buchwitz M, Beek R, Nol S, Burrows JP, Bovensmann H, et al. (2005) Carbon monoxide, methane and carbon dioxide columns retrieved from SCIAMACHY by WFM-DOAS: year 2003 initial data set. Atmos Chem Phys 5: 3313–3329.
- View Article
- Google Scholar
22. Buchwitz M, Rozanov VV, Burrows JP (2000) A near-infrared optimized DOAS method for the fast global retrieval of atmospheric CH₄, CO, CO₂, H₂O, and N₂O total column amounts from SCIAMACHY Envisat-1 nadir radiances. J Geophys Res 105: 15231–15245
- View Article
- Google Scholar
23. Houweling S, Hartmann W, Aben I, Schrijver H, Skidmore J, et al. (2005) Evidence of systematic errors in SCIAMACHY-observed CO₂ due to aerosols. Atmos Chem Phys 5: 3003–3013.
- View Article
- Google Scholar
24. Barkley MP, Frieβ U, Monks PS (2006) Measuring atmospheric CO₂ from space using Full Spectral Initiation (FSI) WFM-DOAS, Atmos Chem Phys. 6: 3517–3534.
- View Article
- Google Scholar
25. Schneising O, Buchwitz M, Burrows JP, Bovensmann H, Reuter M, et al. (2008) Three years of greenhouse gas column-averaged dry air mole fractions retrieved from satellite – Part 1: Carbon dioxide. Atmos Chem Phys 8: 3827–3853.
- View Article
- Google Scholar
26. Reuter M, Bovensmann H, Buchwitz M, Burrows JP, Connor BJ, et al. (2011) Retrieval of atmospheric CO₂ with enhanced accuracy and precision from SCIAMACHY: Validation with FTS measurements and com-parison with model results. J Geophys Res 116: D04301
- View Article
- Google Scholar
27. Bösch H, Toon GC, Sen B, Washenfelder RA, Wennberg PO, et al. (2006) Space-based near-infrared CO₂ measurements: Testing the Orbiting Carbon Observatory retrieval algorithm and validation concept using SCIAMACHY observations over Park Falls, Wisconsin. J Geophys Res 111: D23302
- View Article
- Google Scholar
28. Butz A, Guerlet S, Hasekamp O, Schepers D, Galli A, et al. (2011) Toward accurate CO₂ and CH₄ observations from GOSAT. Geophys Res Lett 14: L14812 DOI:https://doi.org/10.1029/2011GL047888.
- View Article
- Google Scholar
29. Yoshida Y, Ota Y, Eguchi N, Kikuchi N, Nobuta K, et al. (2011) Retrieval algorithm for CO₂ and CH₄ column abundances from short-wavelength infrared spectral observations by the Greenhouse Gases Observing Satellite. Atmos Meas Tech 4: 717–734.
- View Article
- Google Scholar
30. Crisp D, Fisher B, O’Dell C, Frankenberg C, Basilio R, et al. (2012) The ACOS CO₂ retrieval algorithm - Part II: Global XCO₂ data characterization. Atmos Meas Tech 5: 687–707.
- View Article
- Google Scholar
31. Reuter M, Bösch H, Bovensmann H, Bril A, Buchwitz M, et al. (2013) A joint effort to deliver satellite retrieved atmospheric CO₂ concentrations for surface flux inversions: the ensemble median algorithm EMMA. Atmos Chem Phys 13: 1771–1780.
- View Article
- Google Scholar
32. Hammerling DM, Michalak AM, O’Dell C, Kawa SR (2012) Global CO₂ distributions over land from the Greenhouse Gases Observing Satellite(GOSAT). Geophys Res Lett 39: L08804
- View Article
- Google Scholar
33. Hammerling DM, Michalak AM, Kawa SR (2012) Mapping of CO₂ at high spatiotemporal resolution using satellite observations: Global distributions from OCO-2. J Geophys Res 117: D06306
- View Article
- Google Scholar
34. Zeng Z, Lei L, Hou S, Ru F, Guan X, et al. (2014) A Regional Gap-Filling Method Based on Spatiotemporal Variogram Model of CO₂ Columns. IEEE Trans Geosci Remote Sens 6: 3594–3603.
- View Article
- Google Scholar
35. Yokota T, Yoshida Y, Eguchi N, Ota Y, Tanaka T, et al. (2009) Global Concentrations of CO₂ and CH₄ Retrieved from GOSAT: First Preliminary Results. SOLA 5: 160–163.
- View Article
- Google Scholar
36. Parker R, Boesch H, Cogan A, Fraser A, Feng L, et al. (2011) Methane observations from the Greenhouse Gases Observing SATellite: Comparison to ground-based TCCON data and model calculations. Geophys Res Lett 15: L15807
- View Article
- Google Scholar
37. Wang TX, Shi JC, Jing YY, Xie YH (2013) Investigation of the consistency of atmospheric CO₂ retrievals from different space-based sensors: Intercomparison and spatio-temporal analysis, Chin Sci Bull. 33: 4161–4170.
- View Article
- Google Scholar
38. Peters W, Jacobson AR, Sweeney C, Andrews AE, Conway TJ, et al. (2007) An atmospheric perspective on North American carbon dioxide exchange: Carbon Tracker. Proc Natl Acad Sci U.S.A. 48: 18925–18930.
- View Article
- Google Scholar
39. Wunch D, Wennberg PO, Toon GC, Connor BJ, Fisher B, et al. (2011) A method for evaluating bias in global measurements of CO₂ total columns from space. Atmos Chem Phys 11: 12317–12337.
- View Article
- Google Scholar
40. Rodgers CD (2000) Inverse Methods for Atmospheric Sounding: Theory and Practice. World Scientific Publishing Co. Ltd.193 p.

[ref1] 1. Reuter M, Bovensmann H, Buchwitz M, Burrows JP (2010) A method for improved SCIAMACHY CO₂ retrieval in the presence of optically thin clouds. Atmos. Meas. Tech 3: 209–232.
View Article
Google Scholar

[2] View Article

[3] Google Scholar

[ref2] 2. O’Dell CW, Connor B, Bösch H, O’Brien D, Frankenberg C, et al. (2012) The ACOS CO₂ retrieval algorithm – Part 1: Description and validation against synthetic observations. Atmos Meas Tech 5(1): 99–121.
View Article
Google Scholar

[5] View Article

[6] Google Scholar

[ref3] 3. Baker DF, Law RM, Gurney KR, Rayner P, Peylin P, et al. (2006) TransCom 3 inversion intercomparison: Impact of transport model errors on the interannual variability of regional CO₂ fluxes, 1988–2003. Global Biogeochem Cy 20: GB1002
View Article
Google Scholar

[8] View Article

[9] Google Scholar

[ref4] 4. Morino I, Uchino O, Inoue M, Yoshida Y, Yokota T, et al. (2011) Preliminary validation of column-averaged volume mixing ratios of carbon dioxide and methane retrieved from GOSAT short-wavelength infrared spectra. Atmos Meas Tech 4: 1061–1076.
View Article
Google Scholar

[11] View Article

[12] Google Scholar

[ref5] 5. Butz A, Hasekamp OP, Frankenberg C, Aben I (2009) Retrievals of atmospheric CO₂ from simulated space-borne measurements of backscattered near-infrared sunlight: accounting for aerosol effects. Appl Opt 18: 3322–3336.
View Article
Google Scholar

[14] View Article

[15] Google Scholar

[ref6] 6. Kuze A, Suto H, Nakajima M, Hamazaki T (2009) Thermal and near infrared sensor for carbon observation Fourier-transform spectrometer on the Greenhouse Gases Observing Satellite for greenhouse gases monitoring, Appl Opt. 35: 6716–6733.
View Article
Google Scholar

[17] View Article

[18] Google Scholar

[ref7] 7. Bovensmann H, Burrows JP, Buchwitz M, Frerick J, Noel S, et al. (1999) SCIAMACHY –Mission objectives and measurement modes. J Atmos Sci 56: 127–150.
View Article
Google Scholar

[20] View Article

[21] Google Scholar

[ref8] 8. Crisp D, Atlas RM, Breon FM, Brown LR, Burrows JP, et al. (2004) The Orbiting Carbon Observatory (OCO) mission. Adv Space Res 34: 700–709.
View Article
Google Scholar

[23] View Article

[24] Google Scholar

[ref9] 9. Boesch H, Baker D, Connor B, Crisp D, Miller C (2011) Global Characterization of CO₂ Column Retrievals from Shortwave-Infrared Satellite Observations of the Orbiting Carbon Observatory-2 Mission. Remote Sens 3: 270–304.
View Article
Google Scholar

[26] View Article

[27] Google Scholar

[ref10] 10. Aumann HH, Chahine MT, Gautier C, Goldberg MD, Kalnay E, et al. (2003) AIRS/AMSU/HSB on the Aqua Mission: Design, science objectives, data products, and processing systems. IEEE Trans Geosci Remote Sens 41: 253–264.
View Article
Google Scholar

[29] View Article

[30] Google Scholar

[ref11] 11. Chahine M, Barnet C, Olsen ET, Chen L, Maddy E (2005) On the determination of atmospheric minor gases by the method of vanishing partial derivatives with application to CO₂. Geophys Res Lett 32: L22803
View Article
Google Scholar

[32] View Article

[33] Google Scholar

[ref12] 12. Phulpin T, Cayla F, Chalon G, Diebel D, Schlussel P (2002) IASI on board Metop: Project status and scientific preparation. Proceedings of the 12th International TOVS Study Conference. Lorne, Vi ctoria, Australia.

[ref13] 13. Turquety S, Hadji-Lazaro J, Clerbaux C, Hauglustaine DA, Clough SA, et al. (2004) Operational trace gas retrieval algorithm for the Infrared Atmospheric Sounding Interferometer. J Geophys Res 109: D21301
View Article
Google Scholar

[36] View Article

[37] Google Scholar

[ref14] 14. Yang Z, Washenfelder RA, Keppel-Aleks G, Krakauer NY, Randerson JT, et al. (2007) New constraints on Northern Hemisphere growing season net flux. Geophys Res Lett 34: L12807
View Article
Google Scholar

[39] View Article

[40] Google Scholar

[ref15] 15. Wunch D, Toon GC, Blavier JFL, Washenfelder R, Notholt J, et al. (2011) The Total Carbon Column Observing Network. Philos T Roy Soc A 369: 2087–2112.
View Article
Google Scholar

[42] View Article

[43] Google Scholar

[ref16] 16. Rayner PJ, O’Brien DM (2011) The utility of remotely sensed CO₂ concentration data in surface source inversions. Geophys Res Lett 28: 175–178.
View Article
Google Scholar

[45] View Article

[46] Google Scholar

[ref17] 17. Houweling S, Breon FM, Aben I, Rodenbeck C, Gloor M, et al. (2004) Inverse modeling of CO₂ sources and sinks using satellite data: a synthetic inter-comparison of measurement techniques and their performance as a function of space and time. Atmos Chem Phys 4: 523–538.
View Article
Google Scholar

[48] View Article

[49] Google Scholar

[ref18] 18. Olsen SC, Randerson JT (2004) Differences between surface and column atmospheric CO₂ and implications for carbon cycle research. J Geophys Res 109; D02301. doi:10.1029/2003JD003968.

[ref19] 19. Miller CE, Crisp D, DeCola PL, Olsen SC, Randerson JT, et al. (2007) Precision requirements for space-based XCO₂ data. J Geophys Res 112: D10314
View Article
Google Scholar

[52] View Article

[53] Google Scholar

[ref20] 20. Buchwitz M, Beek R, Burrows JP, Bovensmann H, Warneke T, et al. (2005) Atmospheric methane and carbon dioxide from SCIAMACHY satellite data: initial comparison with chemistry and transport models, Atmos Chem Phys. 5: 941–962.
View Article
Google Scholar

[55] View Article

[56] Google Scholar

[ref21] 21. Buchwitz M, Beek R, Nol S, Burrows JP, Bovensmann H, et al. (2005) Carbon monoxide, methane and carbon dioxide columns retrieved from SCIAMACHY by WFM-DOAS: year 2003 initial data set. Atmos Chem Phys 5: 3313–3329.
View Article
Google Scholar

[58] View Article

[59] Google Scholar

[ref22] 22. Buchwitz M, Rozanov VV, Burrows JP (2000) A near-infrared optimized DOAS method for the fast global retrieval of atmospheric CH₄, CO, CO₂, H₂O, and N₂O total column amounts from SCIAMACHY Envisat-1 nadir radiances. J Geophys Res 105: 15231–15245
View Article
Google Scholar

[61] View Article

[62] Google Scholar

[ref23] 23. Houweling S, Hartmann W, Aben I, Schrijver H, Skidmore J, et al. (2005) Evidence of systematic errors in SCIAMACHY-observed CO₂ due to aerosols. Atmos Chem Phys 5: 3003–3013.
View Article
Google Scholar

[64] View Article

[65] Google Scholar

[ref24] 24. Barkley MP, Frieβ U, Monks PS (2006) Measuring atmospheric CO₂ from space using Full Spectral Initiation (FSI) WFM-DOAS, Atmos Chem Phys. 6: 3517–3534.
View Article
Google Scholar

[67] View Article

[68] Google Scholar

[ref25] 25. Schneising O, Buchwitz M, Burrows JP, Bovensmann H, Reuter M, et al. (2008) Three years of greenhouse gas column-averaged dry air mole fractions retrieved from satellite – Part 1: Carbon dioxide. Atmos Chem Phys 8: 3827–3853.
View Article
Google Scholar

[70] View Article

[71] Google Scholar

[ref26] 26. Reuter M, Bovensmann H, Buchwitz M, Burrows JP, Connor BJ, et al. (2011) Retrieval of atmospheric CO₂ with enhanced accuracy and precision from SCIAMACHY: Validation with FTS measurements and com-parison with model results. J Geophys Res 116: D04301
View Article
Google Scholar

[73] View Article

[74] Google Scholar

[ref27] 27. Bösch H, Toon GC, Sen B, Washenfelder RA, Wennberg PO, et al. (2006) Space-based near-infrared CO₂ measurements: Testing the Orbiting Carbon Observatory retrieval algorithm and validation concept using SCIAMACHY observations over Park Falls, Wisconsin. J Geophys Res 111: D23302
View Article
Google Scholar

[76] View Article

[77] Google Scholar

[ref28] 28. Butz A, Guerlet S, Hasekamp O, Schepers D, Galli A, et al. (2011) Toward accurate CO₂ and CH₄ observations from GOSAT. Geophys Res Lett 14: L14812 DOI:https://doi.org/10.1029/2011GL047888.
View Article
Google Scholar

[79] View Article

[80] Google Scholar

[ref29] 29. Yoshida Y, Ota Y, Eguchi N, Kikuchi N, Nobuta K, et al. (2011) Retrieval algorithm for CO₂ and CH₄ column abundances from short-wavelength infrared spectral observations by the Greenhouse Gases Observing Satellite. Atmos Meas Tech 4: 717–734.
View Article
Google Scholar

[82] View Article

[83] Google Scholar

[ref30] 30. Crisp D, Fisher B, O’Dell C, Frankenberg C, Basilio R, et al. (2012) The ACOS CO₂ retrieval algorithm - Part II: Global XCO₂ data characterization. Atmos Meas Tech 5: 687–707.
View Article
Google Scholar

[85] View Article

[86] Google Scholar

[ref31] 31. Reuter M, Bösch H, Bovensmann H, Bril A, Buchwitz M, et al. (2013) A joint effort to deliver satellite retrieved atmospheric CO₂ concentrations for surface flux inversions: the ensemble median algorithm EMMA. Atmos Chem Phys 13: 1771–1780.
View Article
Google Scholar

[88] View Article

[89] Google Scholar

[ref32] 32. Hammerling DM, Michalak AM, O’Dell C, Kawa SR (2012) Global CO₂ distributions over land from the Greenhouse Gases Observing Satellite(GOSAT). Geophys Res Lett 39: L08804
View Article
Google Scholar

[91] View Article

[92] Google Scholar

[ref33] 33. Hammerling DM, Michalak AM, Kawa SR (2012) Mapping of CO₂ at high spatiotemporal resolution using satellite observations: Global distributions from OCO-2. J Geophys Res 117: D06306
View Article
Google Scholar

[94] View Article

[95] Google Scholar

[ref34] 34. Zeng Z, Lei L, Hou S, Ru F, Guan X, et al. (2014) A Regional Gap-Filling Method Based on Spatiotemporal Variogram Model of CO₂ Columns. IEEE Trans Geosci Remote Sens 6: 3594–3603.
View Article
Google Scholar

[97] View Article

[98] Google Scholar

[ref35] 35. Yokota T, Yoshida Y, Eguchi N, Ota Y, Tanaka T, et al. (2009) Global Concentrations of CO₂ and CH₄ Retrieved from GOSAT: First Preliminary Results. SOLA 5: 160–163.
View Article
Google Scholar

[100] View Article

[101] Google Scholar

[ref36] 36. Parker R, Boesch H, Cogan A, Fraser A, Feng L, et al. (2011) Methane observations from the Greenhouse Gases Observing SATellite: Comparison to ground-based TCCON data and model calculations. Geophys Res Lett 15: L15807
View Article
Google Scholar

[103] View Article

[104] Google Scholar

[ref37] 37. Wang TX, Shi JC, Jing YY, Xie YH (2013) Investigation of the consistency of atmospheric CO₂ retrievals from different space-based sensors: Intercomparison and spatio-temporal analysis, Chin Sci Bull. 33: 4161–4170.
View Article
Google Scholar

[106] View Article

[107] Google Scholar

[ref38] 38. Peters W, Jacobson AR, Sweeney C, Andrews AE, Conway TJ, et al. (2007) An atmospheric perspective on North American carbon dioxide exchange: Carbon Tracker. Proc Natl Acad Sci U.S.A. 48: 18925–18930.
View Article
Google Scholar

[109] View Article

[110] Google Scholar

[ref39] 39. Wunch D, Wennberg PO, Toon GC, Connor BJ, Fisher B, et al. (2011) A method for evaluating bias in global measurements of CO₂ total columns from space. Atmos Chem Phys 11: 12317–12337.
View Article
Google Scholar

[112] View Article

[113] Google Scholar

[ref40] 40. Rodgers CD (2000) Inverse Methods for Atmospheric Sounding: Theory and Practice. World Scientific Publishing Co. Ltd.193 p.

Combining XCO₂ Measurements Derived from SCIAMACHY and GOSAT for Potentially Generating Global CO₂ Maps with High Spatiotemporal Resolution

Combining XCO₂ Measurements Derived from SCIAMACHY and GOSAT for Potentially Generating Global CO₂ Maps with High Spatiotemporal Resolution

Correction

Figures

Abstract

Introduction

Datasets

Methodologies

3.1 Global bias corrections

3.2 Retrieval adjustments

3.3 Combination and time shifting

Results

Discussions and Conclusions

Acknowledgments

Author Contributions

References