A new site: ground-based FTIR XCO2, XCH4 and XCO measurements at Xianghe, China

The column-averaged dry-air mole fractions of CO2 (XCO2), CH4 (XCH4) and CO (XCO) have been measured with a Bruker IFS 125HR Fourier transform infrared spectrometer (FTIR) at Xianghe (39.75 °N, 116.96 °E, North China) since June 2018. The site and the FTIR system are described in this study. The instrumental setup follows the guidelines of the Total Carbon Column Observing Network (TCCON), and the near-infrared spectra are recorded by an InGaAs detector together with a CaF2 beam splitter. The HCl cell measurements that are recorded regularly to derive the instrument line shape (ILS) 5 show that the instrument is correctly aligned. The Xianghe site lies in a polluted area in North China where there are currently no TCCON sites. It can fill the TCCON gap in this region and expand the global coverage of the TCCON measurements. The TCCON standard retrieval code (GGG2014) is applied to retrieve XCO2, XCH4 and XCO. The time series, seasonal cycles and day-to-day variations of XCO2, XCH4 and XCO measurements at Xianghe between June 2018 and July 2019 are shown and discussed. In addition, the FTIR measurements have been used to validate Orbiting Carbon Observatory-2 (OCO-2) and 10 Tropospheric Monitoring Instrument (TROPOMI) satellite observations, as also shown in this paper. The Xianghe FTIR CO2, CH4 and CO data can be accessed at https://doi.org/10.18758/71021049 (Yang et al., 2019).

Atmospheric CO is an indirect greenhouse gas and is mainly emitted from fossil fuel combustion and biomass burning (Yin et al., 2015). There are mainly two methods to measure these three gases, in situ and remote sensing measurements including ground-based Fourier Transform Infrared (FTIR) measurements and satellite measurements that are discussed hereafter.
The Total Carbon Column Observing Network (TCCON) uses ground-based FTIR spectrometers to measure the direct solar 15 radiation in the near infrared spectral region, from which the total column-averaged dry-air mole fractions of CO 2 , CH 4 , N 2 O, CO, HF, H 2 O and HDO are retrieved (Wunch et al., 2011b). Because of their relatively high precision and accuracy, TCCON data are widely used in satellite validations and model comparisons (Zhou et al., 2016;Ostler et al., 2016;Crisp et al., 2017;Borsdorff et al., 2018;Velazco et al., 2019). Today, there are 25 active TCCON sites (https://tccon-wiki.caltech.edu/) covering the latitude band from 80°N to 45°S. Most TCCON sites are in North American, Europe, East Asia (South Korea and 20 Japan) and Oceania. The Hefei station, located in Eastern China, is the first Chinese site that will potentially join the TCCON network. In 2016, a FTIR Bruker IFS 125HR instrument was installed at Xianghe (39.75°N, 116.96°E, 30m a.s.l.) and started observations following the TCCON settings in June 2018. As there are no TCCON sites in North China, the FTIR at Xianghe aims to fill the gap in the network in this region.
The Orbiting Carbon Observatory-2 (OCO-2) was launched on 2 July 2014 by NASA and is devoted to enhancing our 25 understanding of regional scale CO 2 exchanges between the surface and the atmosphere (Crisp et al., 2004;Eldering et al., 2017;Crisp et al., 2017). The Tropospheric Monitoring Instrument (TROPOMI) was launched by ESA on 13 October 2017 as the single payload of the Sentinel-5 Precursor (S5P) satellite. It aims at providing accurate and timely observations of abundances of atmospheric species, such as CH 4 and CO, for air quality and climate change research and services (Borsdorff et al., 2018). However, previous validation work (Wunch et al., 2017;Lambert et al., 2019) based on FTIR measurements 30 has no study in North China due to the absence of TCCON sites in this area, so that it is important to add Xianghe site for evaluation of satellite products in this area.
In this paper, we describe the ground-based FTIR system at Xianghe, with a focus on the measurements of atmospheric CO 2 , CH 4 and CO (Yang et al., 2019). The column-averaged dry-air mole fractions of these gases are retrieved by the GGG2014  code in the period between 14 June 2018 and 19 July 2019. The paper is structured as follows. Section 2 introduces the Xianghe site and the FTIR system. In Section 3, the retrieval and filtering methods are described. To increase the precision of the retrievals, the spectra are cloud filtered based on the separate direct solar irradiation measurements. The time series of XCO 2 , XCH 4 and XCO are shown and discussed. In the next Section, the OCO-2 (XCO 2 ) and TROPOMI (XCH 4 and XCO) satellite observations are validated with the FTIR measurements at Xianghe. Finally, conclusions are drawn in Section 5. Xianghe county acts as an integrated transportation and transfer center in Beijing-Tianjin-Hebei region, which is one of the 10 most populous and economically dynamic areas in China (Ran et al., 2016). Xianghe has a middle latitude monsoon climate, with a prevailing south-east wind in summer and a north-west wind in winter (Song et al., 2011). The maximum temperature at Xianghe site is around 38°C in summer, and the minimum temperature is around -10°C in winter. The raining days occur mainly in summer including some days with extreme precipitations larger than 100 mm/day. The Bruker IFS 125HR instrument was installed in the upper level of a four-story building in June 2016. About 2 years 15 later, in June 2018, a solar tracker was installed on the roof (50 m a.s.l.), to guide the direct solar radiation into the FTIR instrument. The distance between the solar tracker and the entrance window of the FTIR instrument is about 3 meters. The solar tracker uses a camera inside the IFS 125HR spectrometer to ensure that the center of the solar disk always focuses on the entrance aperture of the spectrometer, with an active feedback loop. This system is set up following the developments from Neefs et al. (2007) and Gisi et al. (2011). The FTIR operates only under clear-sky daytime conditions. A rain sensor and a 20 solar irradiation (both total and direct) sensor, are installed next to the solar tracker, to monitor the weather conditions and to control the opening and closing of the solar tracker hatch. To protect the mirrors (Aluminum, coating with MgF 2 ) of the solar tracker, the hatch of the tracker automatically closes under rainy conditions and during nighttime. A heating system is operated in the tracker system to keep the temperature of the rotatory stages and the mirrors close to 15°C in winter. Inside the lab, the air-conditioning keeps the room temperature stabilized around 25°C. 25 The near infrared (NIR) spectra are recorded by a indium gallium arsenide (InGaAs) detector and the middle infrared (MIR) spectra are recorded by a liquid nitrogen cooled indium antimonide (InSb) detector. The entrance window and the beamsplitter are made of CaF 2 . The spectral ranges of the NIR and MIR spectra are 3800-11000 cm −1 and 2000-5000 cm −1 , respectively.
The InSb detector at Xianghe records spectra in the AC mode. The InGaAs detector at Xianghe was operated in the AC mode before 31 May 2019, but since then in the AC + DC mode to be compliant with TCCON standards. The spectrometer 30 settings automatically alternate between NIR and MIR measurements during each clear-sky day. The entrance aperture of the spectrometer in the NIR spectrum is set to 0.5 mm and changed to 0.8 mm after 19 June 2019. There are approximately 70 NIR InGaAs for each clear day. The InGaAs spectra are recorded with a maximum optical path difference (MOPD) of 45cm, corresponding to a spectral resolution of 0.02 cm −1 . Each measurement contains 2 scans (one forward and one backward), taking about 145 s. About 50m south-west of the laboratory building, a weather station is operated on a 110m-height tower, at 62 meters above the ground, measuring pressure, temperature, humidity, wind direction and wind speed. The pressure sensor is located inside the LI-7550 Analyzer, which has an accuracy of 1 hPa. On 30 May 2019, a new weather station was installed at the same height 5 as the solar tracker. The distance between the weather station and the solar tracker is about 2 meters. The pressure sensor is PTB210A Digital Barometer, with an accuracy of 0.07 hPa.

Instrument line shape
The instrument line shape (ILS) reflects the performance and alignment of the instrument (Schneider et al., 2008), which might be distorted by the shear or angular misalignment of the instrument or the field of view (FOV) (Hase et al., 1999;Wunch 10 et al., 2015). A perfectly-aligned interferometer will perfectly center the Haidinger fringes on the field stop at all optical path differences (OPD). The offset between the moving cube-corner retro-reflector (CCRR) and the fixed CCRR will cause the Haidinger fringes moving away from the center when the mirror moves away from ZOPD and this is called shear misalignment.
The angular misalignment is caused when the IR beam is not parallel to the rails. At Xianghe 2 HCl cell spectra are recorded every day after sunset. The modulation efficiency amplitude (ME) and phase error (PE) along the OPD are retrieved using the 15 LINEFIT14.5 code (Hase et al., 1999). The ME is derived from the ratio of the misaligned fringe amplitude to the theoretical fringe amplitude. LINEFIT14.5 normalizes ME to be 1.0 at zero optical path difference (ZOPD). According to the TCCON requirement the ME changes must be within 5% and the PE must be less than 0.02 rad at MOPD (Wunch et al., 2011b). Figure   2 shows the time series of the ME and the PE at the MOPD (45 cm) at Xianghe. The last date ends on 31 March 2019 when the Tungsten lamp used to measure the HCl cell spectra broke down. The mean of the ME is 0.978 ± 0.004, and the mean of the 20 PE is -0.008 ± 0.002 rad. Figure 2 shows that the alignment of the instrument slightly declines over time, but the ME and PE remain compliant with the TCCON requirements during the whole time period. A sensitivity study performed by Hase et al. (2013) showed that the uncertainty in XCO 2 is about 0.035% (0.14 ppm) for a ME change of 4%, which is within the 0.8 ppm (SZA less than 80°) retrieval accuracy of TCCON XCO 2 . Since the ME of the FTIR instrument at Xianghe is about 2-3%, the uncertainty from the ILS on the greenhouse gas retrievals can be ignored. 25

Signal-to-noise ratio
The time series of the signal-to-noise ratio (SNR) of the InGaAs spectra at Xianghe is shown in Figure 3. There are no measurements between 7 July and 22 August 2018 due to a power cut. The SNR decreases quickly with time because Xianghe is located in a polluted area (Robert and Richard, 2015;Li et al., 2007) causing rapid degradation of the mirrors of the solar tracker (Feist et al., 2016). In order to obtain a high SNR, the mirror of the solar tracker was cleaned on 14 November 2018 30 (first yellow line in Figure 3), with the SNR increased from 300 to 500. However, the SNR decreased back to the level of 300 about 3 weeks later, probably because of the increased level of air pollution and relatively lower solar irradiation in winter.  Figure 3), making the SNR rise again above the level of 300.

Retrieval Methodology
The non-linear least-squares fitting code GGG2014  is used to retrieve XCO 2 , XCH 4 , XCO and some other gases from the NIR solar absorption spectra at Xianghe.
where T C a and T C r are the a priori and retrieved total columns, A is the column averaging kernel, x t and x a are true and 10 a priori partial column profiles, ε is the uncertainty. In the forward model, there are 70 equidistant 1km thick layers from the mean sea level up to 70km altitude. The a priori profiles of gases are generated by an empirical model based on surface in situ, ACE-FTS and MkIV measurements. The inter-annual trends and seasonal variations of the species are taken into account, and the a priori profiles are adjusted based on the local tropopause pressure at local noon . The temperature, pressure and water vapour profiles are taken from the National Centres for Environmental Prediction (NCEP) reanalysis data 15 (Kalnay et al., 1996). The surface pressure, temperature and water vapour are from the local weather station.
The column averaging kernel represents the vertical sensitivity of the retrieved total column to the true partial column profile. The typical averaging kernels of XCO 2 , XCH 4 and XCO are shown in Figure 4 of Wunch et al. (2011a). In general, the retrieved CO 2 and CH 4 total columns have good sensitivity in the troposphere and the stratosphere. The retrieved CO column underestimates a deviation from the a priori partial column but overestimates a deviation from the a priori partial column in the 20 stratosphere.
The retrieval windows of CO 2 , CH 4 , CO and O 2 are listed in Table 1 in Wunch et al. (2010). As an example, Figure 4 shows the residuals of the spectral fitting for CO 2 and O 2 for one NIR spectrum at a SZA of 22.9°at Xianghe. The root mean square of the residuals for CO 2 in the spectral window 6180-6260cm −1 , CO 2 in the window 6297-6382 cm −1 and O 2 in the window 7765-8005 cm −1 are 0.24%, 0.25% and 0.37%, respectively, which compare well to the results in Toon et al. (2009). 25 The spectroscopy is the ATM line list (Toon, 2017). The total column-averaged dry-air mole fraction of gas (X gas ) is then derived from the ratio between the retrieved total column of the target species and the retrieved total column of O 2 and from the dry-air mole fraction of O 2 (0.2095) Using the ratio between the target species and O 2 reduces the uncertainties common to both gases, e.g., the surface pressure, dependence of the retrieval results which is known to be an artifact caused by spectroscopic uncertainties, is reduced by applying an empirical airmass-dependent correction, and (2) a constant scaling factor for each gas is applied to calibrate the TCCON measurements to the WMO scale (Wunch et al., , 2010.

Data quality control
As the recording time for one InGaAs spectrum takes about 145 s, the stability of the incoming solar intensity during this 5 period is important for the quality of the spectrum (Beer, 1992). If there are clouds or heavy aerosols in the light path between the FTS and the sun during the spectrum recording, the fractional line depth in FTIR spectra will be distorted (Ridder et al., 2011;Keppel-Aleks et al., 2011). To select the good quality spectra, we have discarded all the spectra with SNR less than 200.
The DC correction can be used to remove the solar irradiation variation in each spectrum (Keppel-Aleks et al., 2007). To filter out spectra infected by the occurrence of clouds or high aerosol load before May 31, 2019 (in AC recording mode), we use the of XCO 2 , XCH 4 and XCO are calculated to compare with the expected random retrieval uncertainties. Note that only days with more than 20 measurements per day are used to calculate the daily STD. The results are shown in Table 1. The STDs of 20 XCO 2 , XCH 4 and XCO decrease with increased β or decreased γ thresholds. The β and γ values are selected mainly based on the XCO 2 data. According to Wunch et al. (2015), the retrieved XCO 2 random uncertainty of TCCON data is about 0.2 % (0.8 ppm) with SZA less than 80 degrees. To reach this precision, the STD of our XCO 2 measurements should be less than 0.8 ppm, but at the same time, we try to keep the number of the measurements as large as possible. Therefore, we have chosen to set β = 90%and γ = 0%. This filtering, together with the SNR filter, makes us reach the required precision for TCCON XCO 2 25 also in the period before 31 May 2019. and the red dots denote the SNR of spectra after both filtering. All of these SNR values are also zoomed in the middle panel of Figure 5. It is clear that the spectra polluted by the tower shadow can be filtered out by the combination of SNR filtering and SI filtering with β = 90 % and γ = 0 % (see cyan and blue dots in middle panel, appearing in the same period with the shadow).
The other blue dots in the upper panel appearing outside the shadow shows that SI filtering can also filter out spectra which are polluted by clouds.
5 Figure 6 shows the time series of XCO 2 with and without filtering between 14 June 2018  following analysis, we only use the data after both SNR and SI filtering for the whole period.
Poor instrument alignment, spectral ghost, error in the time assigned to the spectrum or faulty pressure sensor may cause a dramatic jump in X air (Washenfelder et al., 2006;Wunch et al., 2011a). The retrieved X air (after both SNR and SI filtering) are shown in Figure 7 to confirm the good quality of the retrievals. X air is defined as where m dry,air and m H2O are the molecular mass of dry air and water vapour and g is the column-averaged gravitational acceleration, P s is the surface pressure and T C dry,air , T C O2 and T C H2O are total columns of dry air, O 2 and H 2 O, respectively.
The X air is around 0.98 due to a 2.0% bias in the O 2 spectroscopy (Kivi and Heikkinen, 2016). Figure 7 shows that the X air at Xianghe all pass TCCON standard quality check (between 0.96 and 1.04) and is stable over time with a mean value of 0.982 and a STD of 0.003.

Retrieval results and discussions
The time series of XCO 2 , XCH 4 and XCO after both SNR and SI filtering from June 2018 to July 2019 are shown in Figure   8. The monthly mean of XCO 2 , XCH 4 and XCO at Xianghe, Pasadena (34.1°N ) (Wennberg et al., 2014), Lamont (36.6°N) (Wennberg et al., 2016) and Karlsruhe (49.1°N) (Hase et al., 2014) from June 2018 are also displayed in Figure 9. Based on these measurements, the seasonal variations and day-to-day variations of XCO 2 , XCH 4 and XCO are assessed. Meteorological data are from NCEP Global Forecast System, with a horizontal resolution of 0.5°×0.5°and 55 hybrid sigmapressure levels. Figure 11 shows that the air arriving at Xianghe are mainly coming from the west and north during this period.
The emission sources are mainly influencing the air arriving at 500 m altitude above Xianghe: on 8 and 13 January 2019 these are mainly coming from the north-west of Xianghe with a relative fast wind speed, while on 12 January 2019 they are mainly coming from the west of Xianghe with a much slower wind speed. As shown in Figure 1, Beijing is situated west of Xianghe,  In this section, the FTIR XCO 2 , XCH 4 and XCO measurements at Xianghe are used to validate the OCO-2 XCO 2 and TROPOMI XCH 4 and XCO satellite observations. The co-located FTIR-satellite data pairs are selected based on spatialtemporal collocation criteria. The detailed selection criteria for each target (OCO-2 XCO 2 , TROPOMI XCH 4 and TROPOMI XCO) are described in the subsections 4.2 and 4.3: they account for the scan width of the satellite instrument and the characteristics of the target species. 30 According to Rodgers and Connor (2003), the differences in a priori profiles should be taken into account when comparing ground-based FTIR and satellite observations. TCCON CO 2 , CH 4 and CO a priori profiles (70 layers (Krol et al., 2005). In this study, the satellite a priori profile (OCO-2 or TROPOMI) is taken to be the common a priori profile in the comparison. To substitute the satellite a priori profile in the FTIR retrieval we follow Rodgers and Connor (2003): where X F T IR is the FTIR retrieved total column using the satellite a priori profile, X F T IR is the original FTIR retrieval, A is the FTIR TCCON column averaging kernel, x a,F T IR and x a,SAT are the a priori partial column profiles of FTIR and satellite retrievals, respectively. As the vertical layering of the FTIR retrieval is different from that of the satellite retrieval (OCO-2 or TROPOMI), the satellite a priori profile is re-gridded to the FTIR layer. After re-gridding, the total a priori column remains 10 unchanged (Langerock et al., 2015).
To compare the FTIR and satellite column measurements, the satellite measurements are corrected for a possible difference between the altitudes of its ground pixel and that of the FTIR site at Xianghe. If the surface altitude of the satellite footprint is higher than the altitude of the FTIR instrument, the FTIR a priori profile (x a,F T IR ) is used to fill the gap between the satellite lowest level (P s,SAT ) and the FTIR height (P s,F T IR ), otherwise the satellite a priori profile is considered to be the 15 profile between the satellite lowest level and the FTIR height. Then the partial column of dry air (P C dry,air ) or target species (P C gas ) between the satellite footprint surface altitude and the FTIR surface altitude is calculated as P C dry,air = P s,F T IR P s,SAT dP g (P ) (m dry,air + m h2o ν h2o ) , 20 where g (P ) is gravitational acceleration at height P , x (P ) is the a priori VMR profile of each target gas, ν h2o is the VMR of water vapor in the dry air, calculated as where ν h2o is the VMR of water vapor in the wet air. Then each satellite pixel measurement is scaled with one scaling factor (α) related to satellite pixel level, which is computed as where T C SAT dry,air and T C SAT gas are the total column of dry air and target species in the satellite measurement column. The random error of FTIR measurements together with the systematic and random errors of satellite measurements are considered here for the comparison.

OCO-2
OCO-2 incorporates three imaging grating spectrometers to measure near-infrared spectra. The spectral resolution of OCO-2 is approximately 20 times lower than that of the TCCON FTIR (0.02 cm −1 ) instruments (Frankenberg et al., 2015). OCO-2 collects 8 soundings over its 0.8°swath width every 0.333s with a 16-day repeat cycle (https://ocov2.jpl.nasa.gov/observatory/ instrument/). The OCO-2 XCO 2 measurements are retrieved by the ACOS retrieval algorithm (O'Dell et al., 2012), based on the 5 optimal estimation method. Three bands (0.756 µm, 1.61µm and 2.06 µm) are used in the XCO 2 retrieval. The a priori surface pressure, profiles of temperature and water vapor are from 3-hourly ECMWF model forecast fields and linearly interpolated in space and time to the satellite footprint. Note that there are three versions (v7, v8 and v9) available on the NASA website for the OCO-2 data. Each version comes in two variants: full and lite. The full variant contains all the retrieved parameters, but without any post-correction applied to the data. The lite variant only includes some important parameters, but the data 10 are corrected in terms of a footprint-dependent bias, a parameter-dependent bias and a scaling bias according to the WMO trace-gas standard scale. Compared to v7, many parameters have been improved in v8, such as latitude-dependent problems, surface model, spectroscopy, potential instrumental problems, atmospheric scattering by clouds and aerosols, a spatial-temporal sampling error of a priori surface pressure and the systematic pointing offsets (O'Dell et al., 2012). Based on v8, v9 has a better estimation of the surface pressure, and it shows a better performance in regions with rough topography such as over Lauder 15 (New Zealand) (Kiel et al., 2019). In this study, the latest v9 lite data are selected (https://ocov2.jpl.nasa.gov).
The satellite measurements are selected within 5°latitude × 10°longtitude around Xianghe, these are the same criteria as adopted by Wunch et al. (2017). For each FTIR measurement, the nearby satellite measurement in the spatial collocation box, with less than 2-hours' measurement time difference, is chosen to form one FTIR-satellite data pair. Note that there are nadir and glint observational modes of OCO-2 measurements over Xianghe, and these two types of measurements are combined 20 together to get a statistically robust result because of the limited number (28 days) of data pairs. The time series of the collocated OCO-2 and ground-based FTIR data from 27 June 2018 to 31 May 2019 (last date of satellite data availability) is shown in Figure 12. To avoid the influence from the cloud, we select the co-located data pair, which has at least 20 OCO-2 measurements within the box. The upper panel in Figure 12 (left) shows the daily mean bias of measured XCO 2 from OCO-2 and FTIR. The mean of OCO-2 measurements is 0.62 ppm lower than that of the FTIR 25 measurements, with a STD of 1.20 ppm. The absolute differences between OCO-2 v9 lite data and Xianghe FTIR data are comparable with the results found for the v7 lite products in Wunch et al. (2017) for other TCCON stations with biases ranging from −0.7 ± 1.32 ppm (Wollongong) to 0.9 ± 1.49 ppm (Karlsruhe) in land glint mode and ranging from −0.1 ± 1.04 ppm (Wollongong) to 1.6 ± 2.05 ppm (Garmisch) in nadir mode. The scatter plot of OCO-2 and FTIR at Xianghe is shown in the right panel in Figure 12: the derived correlation coefficient (R) is 0.959. We can conclude that OCO-2 data are in good

TROPOMI
In this section, the TROPOMI XCH 4 and XCO are compared with the FTIR measurements at Xianghe. TROPOMI is a grating spectrometer measuring solar radiation reflected by the Earth and observes in the ultraviolet and visible, near-infrared and shortwave infrared spectral regions. It has a wide swath of around 2600 km across the track and a daily global coverage of the Earth. The spatial resolution of TROPOMI is about 7km × 7km before 6 August 2019 and then it changes to 7.2km × 5.6km.

5
The TROPOMI CO data are provided in three different data streams: the near-real-time (NRTI) stream, the Offline stream (OFFL) and the Reprocessing (RPRO) stream (Landgraf et al., a). CH 4 data are provided in bias-corrected and not-corrected versions (Landgraf et al., b). In our validation, we considered the off-line and reprocessed CO data, from processor versions 01.02 and higher. For CH 4 , we also look at bias-corrected data with processor versions of 01.02 and higher.
TROPOMI uses the RemoTeC algorithm to retrieve CH 4 column using the 0.757-0.774 µm O 2 absorption band and 2.305-10 2.385 µm CH 4 absorption band (Hasekamp et al., 2019). The requirements for the accuracy and precision for TROPOMI XCH 4 are 1% and 1.5%, respectively (Hasekamp et al., 2019). We select TROPOMI XCH 4 measurements that occur within 1 hour of FTIR measurements and within a distance of 100 km from the Xianghe station based on the collocation criteria adopted at other TCCON sites (Lambert et al., 2019). In agreement with Landgraf et al. (b), the TROPOMI pixels are selected with a quality assurance value above 0.5, which removes pixels with processing errors, anomalously high signals and increasing 15 specular reflection of sunlight by the sea surface (Hasekamp et al., 2019). Similar to OCO-2, to reduce the influence from the clouds, we only select the days when there are at least 5 co-located TROPOMI CH 4 pixels.
The left panel in Figure 13 shows the time series of co-located TROPOMI and FTIR XCH 4 daily means and their relative biases (%,(satellite-FTIR)/FTIR) from 27 June 2018 to 19 July 2019 (86 days). The mean bias is -0.60 %, which is within the S5P validation requirement of a bias of 1%. In addition, the STD of the relative biases is 0.55 %, which also meets the S5P 20 mission requirement of 1.5% (Lambert et al., 2019). The R between TROPOMI and FTIR XCH 4 daily means is 0.834 ( Figure   13, right panel). According to the TROPOMI validation report (Lambert et al., 2019), the bias at Xianghe is comparable to the ones at Tsukuba, Lamont and Rikubetsu (similar latitude band).
The TROPOMI XCO measurements are retrieved from the SICOR algorithm (Hasekamp et al., 2019) in the 2.3 µm spectral range. The retrieved TROPOMI CO data is in the unit of total column density (molecules/cm 2 ), so we converted them to to 25 XCO (ppb) values for comparison with FTIR XCO measurements. Because CO is relatively reactive compared to CH 4 , we must reduce the measurement time and location differences in the colocation criteria. Therefore, the TROPOMI observations are selected within 30 minutes of each FTIR measurement and within a maximum distance of 50 km away from the FTIR site and along the light path of the ground-based FTIR measurements. Similar to CH 4 , we only select the days when there are at least 5 co-located TROPOMI CO pixels. In addition, to reduce the impact from long light paths through the atmosphere 30 (Landgraf et al., a), the TROPOMI measurements with a SZA larger than 80°or a satellite zenith angle larger than 65°are filtered out. And we only select the TROPOMI CO products in clear sky cases with cloud height below 500 m and cloud optical depth < 0.5. The left panel of Figure 14 shows the time series and relative biases of co-located TROPOMI and FTIR XCO daily means at Xianghe from 27 June 2018 to 31 May 2019 (70 days). The mean bias and STD between TROPOMI and FTIR are 2.05% and 7.82%, respectively, which are within the S5P mission requirement (bias < 15% and STD < 10%). Compared to other TCCON sites (Lambert et al., 2019), the mean relative bias is quite lower, which is because high pollution events are frequently observed at Xianghe while the satellite CO partial column a priori profile is under-estimated in the lower troposphere over 5 Xianghe. The good agreement between TROPOMI and FTIR XCO with a R of 0.961 (Figure 14, right panel) highlights the good performance of TROPOMI over Xianghe.

Conclusions
A new atmospheric ground-based FTIR site equipped with a Bruker IFS 125HR has been in operation since 14 June 2018 at Xianghe in North China. The NIR spectra are recorded following the TCCON operation procedures. As it is a rather polluted 10 location, it can provide useful information for the study of the carbon cycle in North China and the validation of related satellite observations. Regular HCl cell measurements show that the ME loss is within 2% and the PE remains within 0.02 rad, confirming the ILS of the FTIR is well aligned and meets the TCCON requirement. The XCO 2 , XCH 4 and XCO FTIR measurements at Xianghe provide useful information for related satellite validation in 25 North China. The mean bias between FTIR and OCO-2 XCO 2 measurements is -0.62 ppm with a STD of 1.20 ppm. The mean and STD of the relative differences between FTIR and TROPOMI XCH 4 measurements are -0.60% and 0.55%, respectively.
The mean and STD of the relative differences between FTIR and TROPOMI XCO measurements are 2.05% and 7.82%, respectively. Our measurements show that these satellite observations have good performance in this region.
In summary, this study shows that the Xianghe data comply with the TCCON specifications and we aim to become a part of