Open Access
Add to BibSonomy
Abstract
This work presents the design of multichannel heterodyne signal detection. The measuring capabilities of the proposed configuration were tested based on the developed multichannel laser heterodyne spectroradiometer (MLHS). The MLHS can simultaneously detect absorption features of atmospheric CO2 and CH4 in the NIR spectral range with an ultra-high spectral resolution of λ/δλ ∼ 6 × 107. Such a high resolution allows the MLHS to measure fully resolved individual line contours at 1.605 µm and 1.655 µm for CO2 and CH4 respectively. We propose a new method for synchronous measurements of dark and mixed signals and discuss its effect on the recorded data. We demonstrate advantages of the proposed technique by detailed comparison of data measured by MLHS and a single channel laser heterodyne spectroradiometer.
© 2020 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
1. Introduction
Development of robust and affordable techniques for accurate measurements of greenhouse gases (GHG) contents in the atmosphere is a highly demanded scientific and engineering task. In addition to spaceborne instruments, which provide global coverage and high accuracy of GHG column abundance [1,2], there is a need in compact, portable and low-cost devices whose functionality may vary from validation of spacecraft observations to focused atmospheric monitoring of major GHG sources like megacities, industrial plants or natural inventories. Besides compactness, high mobility and low cost, such devices should be characterized by an extremely high spectral resolution to meet general requirements to GHG observations. In particular, a fully resolved contour of an individual rotational line of a near-IR vibrational overtone measured by direct Sun observation provides information about both vertical profiles of GHG concentration [3] and wind velocity line-of-sight projection [4]. A resolution of order λ/δλ ∼ 106 could be achieved by Bruker 125HR Fourier spectrometer, which is used as a regular instrument by the worldwide network of GHG monitoring “Total Carbon Column Observing Network” (TCCON) [5]. Datasets obtained by TCCON and similar ground-based networks [6–8] are widely used for remote sensing orbital probes data validation, as well as for data assimilation in climate models [9]. Employment of a compact Fourier spectrometer [10] leads to a considerable decrease of spectral resolution, precluding measurements of a resolved contour of an individual spectral line. In contrast with the Fourier spectrometry, the heterodyne technique provides an ultra-high spectral resolution up to λ/δλ ∼ 107-108, keeping instrumentation at low cost, with high mobility and push button functionality. This fact leads to an assumption that heterodyne spectroradiometers could become an alternative to TCCON measuring instruments [11]. There are no commercially available heterodyne spectroradiometers nowadays, hence, scientific groups all over the world present their achievements in development and application of heterodyne spectroradiometers in the NIR spectral range [12–14]. In this paper we focus on heterodyne spectroradiometer developments in the NIR, as it is this range where the most practical applications are expected, in particular due to exploiting commercial telecommunication components. Nevertheless, it does not depreciate significance of heterodyne spectroradiometer developments in the MIR for astronomy [15,16], as well as for atmospheric research [17,18]. For example, recently the development of the heterodyne spectroradiometer based on interband cascade laser emitting in the vicinity of 3.3 µm has been presented [19]. This work is noteworthy, being the first ever implication of the heterodyne technique in this spectral range, which covers absorption features of two important GHG, water vapor and methane. Additionally, the algorithm for inverse problem solution for retrieving vertical distribution of water vapor and methane concentrations is presented.
A typical configuration of NIR heterodyne spectroradiometers implies LO frequency sweeping and processing of the intermediate frequency (IF) signal within the wide frequency bandwidth of 1-2 GHz [12,14] rather than IF spectral analysis by backend electronics common for microwave and mid-IR heterodyne receivers [20]. The broadband IF detection results in significant increase of a signal to noise ratio (SNR), as IF signal power is directly proportional to square root from the bandwidth. However, increasing the IF bandwidth leads to decreasing spectral resolution, which in turn leads to the loss of valuable information about the sounded atmosphere. For example, retrieval of the wind velocity vertical profile from measured transmittance spectra requires spectral resolution of few MHz [4].
The increase of integration time during measurements within the narrow IF bandwidth related to LO frequency sweeping is an inevitable measure to achieve satisfactory SNR. Such increase in data acquisition time affects the accuracy of measurements because of systematic errors caused by instrumental parameters drift, limited information rate per observation session, and more sophisticated data treatment. In order to overcome these shortcomings, J. Kurtz and S. O’Byrne [13] proposed the concept of a multichannel heterodyne receiver based on a balanced photodiode. The authors have shown that such a technique leads to decreasing integration time while keeping high spectral resolution and satisfactory SNR.
This work is dedicated to the description of a multichannel laser heterodyne spectroradiometer (MLHS) for simultaneous measurements of atmospheric CO2 and CH4 absorption features in the NIR range based on different principle, allowing for large amounts of measurement channels and advanced treatment of the dark signal. Like in [13], the multichannel concept capabilities have been tested on dual channel configuration. We believe that implementation of this technique would substantially increase the efficiency of GHG monitoring and aerologic sounding by heterodyne NIR spectroscopic measurements reported in [11].
2. Measurement method
MLHS operates in the direct Sun observation mode from a ground-based platform by analyzing solar radiation passed through the Earth’s atmosphere. The principle of heterodyning technique used in MLHS is based on mixing received solar radiation with the local oscillator (LO) radiation on the photodiode and processing the IF beat signal to get spectral density of the solar radiation in the range of LO frequency. As opposed to historical standard adopted in microwave radiometry, heterodyne detection in MLHS implies sweeping LO frequency in the vicinity of spectral lines of interest and processing IF signal in the narrow frequency bandwidth, typically few MHz. Fundamentals of heterodyne detection in the NIR range are described in [3,21,22]. In this paper we are only focused on technical side of multichannel configuration implementation question.
Heterodyne spectroscopy is the highly sensitive method, which allows detecting weak signals at the quantum limit [23]. For example, for wavelength 1.605 µm the noise equivalent power (NEP) associated with the quantum limit is $p = \frac{{hc}}{{\lambda \sqrt {B\tau } }} \approx {10^{ - 24}}\,\textrm{W}/\textrm{Hz}$. It is essential at the quantum limit for the photocurrent shot noise to be dominant source of noise, rather than noise from the amplifier electronics and LO relative intensity noise (RIN). The estimation of an optimal range of the LO power is described in [21].
Because of the necessity to provide coincidence of the wave fronts of LO and solar signals, there is a fundamental limitation on the field of view of a heterodyne spectroradiometer. This limitation Ω×S ≤ λ2 is generally known as “antenna theorem” [24], where Ω - solid angle and S – aperture area. Thus, there are two possibilities to increase signal intensity, either by increasing bandwidth at the cost of spectral resolution, or by implementing multichannel configuration, which consists of N independent channels for solar signal receiving. In the second case SNR is expected to increase by a factor of √N.
3. Dark signal measurement
The procedure of raw LHS data processing is described in [22]. In general, this procedure consist of the following steps: firstly, measurement of the LO noise power is conducted (dark signal, with only LO radiation received by the photodiode); secondly, measurement of the mixed signal noise power is conducted, with both LO and solar radiation components being detected; thirdly, the dark signal is subtracted from the mixed signal; finally, retrieval of a baseline level, linearization and frequency calibration of the measured transmission spectrum are performed. An accurate measurement of the dark signal is very important as it affects the retrieved spectral transmission on the atmospheric sounding path. Small variations in the dark signal may lead to significant uncertainties in the quantitative retrievals of the GHG concentration.
A. Rodin et al. [3] presented a method for simultaneous registration of dark and mixed signals based on balanced detector. Detected signals were subtracted on balanced scheme giving in its output the IF beat signal, which was further processed by amplitude detector, converted by ADC and saved on the PC. Such configuration was supposed to be convenient, as it eliminates the use of a chopper for signal modulation and, consequently, decreases the integration time twice. However, our experience has shown that this scheme had many shortcomings: firstly, doubling of a dark shot noise was observed; secondly, due to asymmetry of optical system, signals in two optical channels had slightly different shapes. Due to this fact, the balanced detector does not provide full compensation of registered signals, and eliminating the use of the chopper has failed; Finally, the described configuration uses an amplitude detector, which converts signal into the value proportional to the spectral density. The parameters of the amplitude detector are a compromise between linearity and bandwidth. To ensure acceptable linearity, we were forced to limit the bandwidth to 3 MHz. That is why the IF bandwidth could not be higher than 3 MHz. In presented multichannel configuration the IF bandwidth is also 3 MHz, but there are no any principal limitations to this value.
Measurement of the dark signal level could be implemented in different ways. For example, H.R. Melroy et al. [12] and H. Deng et al. [14] used synchronized chopper and lock-in amplifier for modulation of the mixed signal and then treated each component separately. This configuration is conventional and optimal as it allows a real time measurement of both dark and mixed signals. Nevertheless, this approach has several disadvantages, including increasing integration time due to sequential measurements, and the need for additional components – the chopper with controller and lock-in amplifier.
J. Kurtz et al. [13] and S. G. Zenevich et al. [22] employed sequential measurements of mixed signal and dark signal within the integration time. Apparently this method is not time efficient, especially in the case of detecting signal in the narrow IF bandwidth, where integration time could reach 10 minutes [22], resulting in about 20 minutes of the total integration time.
In order to optimize the process of dark signal acquisition, while avoiding moving parts in the MLHS design, we implement the independent IF receivers for simultaneous measurement of dark and mixed signals. Figure 1 shows the principle optical scheme of the experimental setup. The diode laser (DL) radiation is split by a fused fiber coupler (FC) into two shoulders. One shoulder goes to a reference channel; the other shoulder is in turn split into two parts. One of them goes directly to the IF receiver 1, the DL radiation in the second shoulder is mixed with solar radiation collected by a microtelescope (MT) and is sent to IF receiver 2. This configuration is however unsuitable for field applications, due to a significant independent drift of the signal amplitude on each channel caused by environment temperature variations affecting fiber optical system transmittance. The DL operates in the quasi-continuous mode. A DL pulse consists of 250 time slots of 50 ms with constant pump current and hence, emission wavelength and power within each slot. During the first 4 and last 50 slots of each pulse the DL is off, and these slots are used for defining the optical zero level of IF receiver. During time slots 5-200 the DL is on, allowing for mixed signal detection, while pump current is linearly swept from slot to slot to perform a spectral scan.
Fig. 1. Principle scheme for simultaneous measurement of dark and mixed signals (method 1). DL – diode laser; SMF – single-mode optical fiber; FC – fiber coupler; MT – microtelescope.
Figure 2 presents temperature drift of dark signal noise power during measurements at different environmental conditions. The black curve demonstrates no drift on the pulse region where the DL is off (time slot 1) during measurements in the field conditions. This fact means that temperature variations of the IF receiver electronic block do not contribute to the signal drift. The red curve also demonstrates no drift at the same pulse region (time slot 1) during measurements at laboratory conditions. The green curve shows no drift at the pulse region where the DL is on (time slot 50) during measurements in laboratory conditions. But during field measurements there was a significant drift of the signal for the same time slots, as shown by the blue curve. The variance of listed pulse regions is equal on both channels, but the presence of an uncontrolled independent drift in the signal does not allow us to quantitatively evaluate the detected signal.
Fig. 2. Noise power of the dark signal obtained from receiver 1 at specific time slots of the DL pulse as a function of number of DL scans for different measurement conditions. Red and green curves demonstrate variations of the 1st and 50st time slots of the DL pulse respectively in laboratory conditions. Black and blue curves demonstrate variations of the 1st and 50st time slots of the DL pulse respectively in field conditions. For the references to color meaning in this figure, the reader is referred to the web version of this article.
The principle optical scheme of the alternative technique for dark and mixed signals measurements is presented in the Fig. 3. This scheme is identical to the previous one with the only difference: both IF receivers detect mixed signals of equal optical power. This scheme is truly multichannel, as each channel measures mixed signal independently. Dark signal measurement is provided by implementing a micro-electro-mechanical systems (MEMS) fiber optical switch, which is installed into fiber optical system right after MT.
Fig. 3. Principle scheme for simultaneous measurement of dark and mixed signals (method 2). DL – diode laser; SMF – single-mode optical fiber; FC – fiber coupler; MT – microtelescope; MEMS - micro-electro-mechanical systems switch.
The MEMS switch is able to cut off solar radiation at the beginning and at the end of the DL pulse in order to leave the range close to the absorption line untouched. The shape of an IF receiver output signal is schematically shown in Fig. 4. The dark signal level is calculated during data post processing by the second order polynomial approximations of pulse regions marked as “2” corresponding to the time period when DL is turned on and Sun is off. Temperature drifts described above do not affect the absorption level, as both signals on the IF receiver drift simultaneously. Since the IF receiver 1 is lacking a MEMS switch, only mixed signal is detected. Subsequent averaging in such a configuration is provided by signals alignment at the post processing stage.
Fig. 4. Schematic structure of the IF receiver output signal. 1 – pulse regions where laser is turned off; 2 – regions where only DL radiation drops on the photodiode; 3 – region where mixed signal is detected.
The technique described above requires twice as low integration time, as each channel measures mixed signal and dark signal simultaneously. It resolves the problem of dark signal temperature drift as well. The MLHS setup presented in the next section exploits this configuration of dark signal level detection.
4. Experimental setup
The setup is based on the single channel LHS described in detail in [22] and retains its main characteristics. Only features specific to MLHS and its difference from a single channel LHS are presented here. The MLHS circuit is shown in Fig. 5. The main upgrades compared to a single channel configuration are: (i) multichannel entrance optics; (ii) a compact digital FPGA-based multichannel IF analyzer, (iii) simultaneous measurements of CO2 and CH4 using two separate LO, (iv) a MEMS switch for synchronous measurement of dark and mixed signals. DFB lasers with single-mode fiber output and linewidth of ∼ 2 MHz are used as LO emitting at wavelengths of λ1 = 1.605 µm and λ2 = 1.655 µm for CO2 and CH4 respectively. Lasers emit in a quasi-continuous mode with the measurement cycle repetition frequency of 20 Hz. A whole laser pulse length is 50 ms, including 39 ms (time slots 5-200) when the laser is on and 11 ms (time slots 1-4 and 201-250) when the laser is off. Emission wavelength ramping of each laser in the range of ∼1 cm-1 is provided by a sawtooth-shaped modulation of pumping current from 70 mA to 95 mA. Single-mode optical fiber (SMF) and fused fiber couplers (FC) are used for splitting the DL signal between channels and coupling the LO with solar radiation. The system of microtelescopes (MT) installed on the telescope mount, which is used as solar tracker, is applied for solar radiation injection into the SMF.
Fig. 5. Principle scheme of multichannel laser heterodyne spectroradiometer (MLHS). DL – diode laser; SMF – single-mode optical fiber; FC – fiber coupler; MT – microtelescope; MEMS - micro-electro-mechanical systems switch; PD – photodiode; OA – operational amplifier; BPF – band pass filter.
The MLHS is equipped with 6 measuring channels: 4 analytical channels for measuring atmospheric CO2 and CH4 (two channels for each gas), and 2 reference channels (one channel for each gas). Generally, the scheme of the MLHS for CH4 detection is identical to one for CO2 detection, with the only difference of exploiting different type of gas cells in the reference channels. These cells are filled with targeted gases – carbon dioxide and methane – under the pressure of ∼ 10 mbar and are used for stabilizing LO frequency with the accuracy of ∼ 5 MHz. Such a high accuracy is achieved due to feedback between the DL temperature, controlled by TEC, and the apparent position of the absorption line maximum in the reference cell, based on proportional-integration controller. For each gas the separate reference channel is implied. Due to relatively high abundance of carbon dioxide in the atmosphere, it is necessary to take measurements of weak spectral lines, consequently, CO2 absorption measurement in the reference channel demands optical path of ∼ 30 m. We used ICOS [25] cell in the CO2 reference channel, as described in detail in [22]. In the methane reference channel, a 4 cm length single pass reference cell is used.
In the analytical channels, the mixed LO and solar radiation is detected by a fast p-i-n photodiode (PD). PD output is converted into voltage by the transimpedance amplifier with conversion resistance of 4 kΩ and the high frequency cutoff at 3 MHz, then additionally amplified with a gain of ∼ 1000 and passed through a band filter with the bandpass of 200 kHz - 3 МHz. Such a configuration of the IF receiver provides spectral resolution of λ/δλ ∼ 6 × 107. The IF beat signal of each channel is individually digitized by an 8-bit ADC with the bandwidth of 10 MHz and processed by variance calculation in the FPGA. Data integration and storage is handled by a PC.
A linearization of the DL frequency scale is provided by processing of each DL signal passed through a quartz Fabri-Perot etalon, similarly as in [22]. The frequency calibration is based on synchronization of the absorption peak in reference channel to the value of frequency from HITRAN2016 database [26]. Laser control is handled by two NI USB-6215 boards, which register signals in reference channels and control MEMS switches. The MLHS operation control is provided by LabView2018 software. Dimensions of the MLHS is of 60 × 50 × 23 cm and weights ∼ 13 kg (excluding the laptop and the mount). The MLHS is powered by 8.4V lithium ion battery array with the capacity enough for uninterrupted operation during one solar day at room temperature.
5. Measurements results
Measurements of atmospheric transmittance spectra of CO2 and CH4 were carried out at MIPT’s campus in the Moscow suburb at the end of July 2019. Figure 6(b) and 6(c) present measured atmospheric transmittance spectra (blue curve) and model synthetic spectra (red curve) for CO2 and CH4 respectively. Figure 6(a) demonstrates the CO2 transmittance spectrum measured by the single channel LHS at the same spectral range in August 2017 [22], where oscilloscope R&H RTO 1012 was used as IF signal analyzer. Both Fig. 6(b), 6(c) and 6(a) spectra were registered at clear sky conditions, and suppose the optical depth of the atmosphere column in the NIR range is about 0.1, the comparison of the measured SNR is possible.
Fig. 6. Measured atmospheric CO2 and CH4 transmittance spectra (blue curve) and theirs model fitting (red curve). a – CO2 transmittance spectrum recorded by single channel LHS at 2017 [22]; b – CO2 transmittance spectrum recorded by MLHS at the same spectral range; c – CH4 transmittance spectrum recorded by MLHS.
Synthetic transmittance spectra were calculated based on the model of the plane-parallel atmosphere. Temperature and pressure profiles were taken from open reanalysis databases ERA-interim [27] and NCEP/NCAR [28]. Spectral line parameters were taken from HITRAN2016 [26].
Detailed analysis of recorded spectra is not the primary goal of this paper. The model used in the analysis as well as the column concentration retrieval procedure are described in [22]. This work is focused on the implementation of an efficient multichannel heterodyne detection technique, development of a new method for simultaneous measurements of dark and mixed signals, and theirs influences on received spectra. As a result of implementing this method, the quality of recorded spectra became noticeably higher. Simultaneous measurements of dark and mixed signals allow for integration time twice as low in comparison with standard approach. Implementation of multichannel IF receivers (N = 2) enable increasing the SNR by the factor of sqrt(2) due to the Gaussian statistics of measured signal noise component. Finally, exploiting the developed IF receivers and FPGA-based digital IF signal analyzer, with the filtering system of power supply and isolating all sensitive components from outer electromagnetic interference allow us to become closer to the quantum limit and resulted in additional increase of the SNR. After all modifications, the overall effect is the following: SNR improved three times and integration time decreased twice in comparison with the previous version of the spectroradiometer. The spectra in Fig. 6(b) and 6(c) were measured within integration time of 4 minutes with the SNR of 300.
It also worth noticing that IF receiver bandwidth broadening affects SNR as a square root proportion, thus applying 20 MHz receivers and 4 analytical channels for each gas could result in even better SNR levels and reduction of accumulation time down to 15 s, according to our tests. Although the spectral resolution will decrease expectedly.
6. Conclusion
The design of the multichannel laser heterodyne spectroradiometer (MLHS) for simultaneous measurements of atmospheric CO2 and CH4 absorption features in the NIR spectral range has been presented. The possibility of increasing the SNR and overcoming antenna theorem limitation by implementation of a two channels configuration of the instrument’s receiver has been described. The number of channels in presented technique is only limited by optical power of the local oscillator, and may be easily extended up to 8 channels. A new method for parallel dark and mixed signals measurements, which leads to significant economy in integration time, has been demonstrated. The compact and lightweight MLHS design is easy to adapt to mobile platforms including drones and orbital probes for the GHG atmospheric remote sensing. In such cases the multichannel configuration of IF receiver is essential due to the lack of time for signal accumulation during each solar occultation in orbital pass.
Funding
Russian Foundation for Fundamental Investigations (19-29-06104 (A. Rodin, I. Gazizov, M. Spiridonov), 19-32-90276 (S. Zenevich)).
Acknowledgments
The authors also thank Data Plus Co. Ltd and S. Shapotkin for providing optoelectronic components of the MLHS.
Disclosures
The authors declare no conflicts of interest.
References
1. A. Kuze, H. Suto, M. Nakajima, and T. Hamazaki, “Thermal and near infrared sensor for carbon observation Fourier-transform spectrometer on the Greenhouse Gases Observing Satellite for greenhouse gases monitoring,” Appl. Opt. 48(35), 6716–6733 (2009). [CrossRef]
2. A. Eldering, C. W. O’Dell, P. O. Wennberg, D. Crisp, M. R. Gunson, C. Viatte, C. Avis, A. Braverman, R. Castano, A. Chang, L. Chapsky, C. Cheng, B. Connor, L. Dang, G. Doran, B. Fisher, C. Frankenberg, D. Fu, R. Granat, J. Hobbs, R. A. M. Lee, L. Mandrake, J. McDuffie, C. E. Miller, V. Myers, V. Natraj, D. O’Brien, G. B. Osterman, F. Oyafuso, V. H. Payne, H. R. Pollock, I. Polonsky, C. M. Roehl, R. Rosenberg, F. Schwandner, M. Smyth, V. Tang, T. E. Taylor, C. To, D. Wunch, and J. Yoshimizu, “The Orbiting Carbon Observatory-2: first 18 months of science data products,” Atmos. Meas. Tech. 10(2), 549–563 (2017). [CrossRef]
3. A. Rodin, A. Klimchuk, A. Nadezhdinskiy, D. Churbanov, and M. Spiridonov, “High resolution heterodyne spectroscopy of the atmospheric methane NIR absorption,” Opt. Express 22(11), 13825–13834 (2014). [CrossRef]
4. A. V. Rodin, D. V. Churbanov, S. G. Zenevich, A. Y. Klimchuk, V. M. Semenov, M. V. Spiridonov, and I. S. Gazizov, “Vertical wind profiling from the troposphere to the lower mesosphere based on high-resolution heterodyne near-infrared spectroradiometry,” Atmos. Meas. Tech. 13(5), 2299–2308 (2020). [CrossRef]
5. F. Chevallier, N. M. Deutscher, T. J. Conway, P. Ciais, L. Ciattaglia, S. Dohe, M. Fröhlich, A. J. Gomez-Pelaez, D. Griffith, F. Hase, L. Haszpra, P. Krummel, E. Kyrö, C. Labuschagne, R. Langenfelds, T. Machida, F. Maignan, H. Matsueda, I. Morino, J. Notholt, M. Ramonet, Y. Sawa, M. Schmidt, V. Sherlock, P. Steele, K. Strong, R. Sussmann, P. Wennberg, S. Wofsy, D. Worthy, D. Wunch, and M. Zimnoch, “Global CO2fluxes inferred from surface air-sample measurements and from TCCON retrievals of the CO2 total column,” Geophys. Res. Lett. 38(24), L24810 (2011). [CrossRef]
6. M. Frey, M. K. Sha, F. Hase, M. Kiel, T. Blumenstock, R. Harig, G. Surawicz, N. M. Deutscher, K. Shiomi, J. E. Franklin, H. Bösch, J. Chen, M. Grutter, H. Ohyama, Y. Sun, A. Butz, G. Mengistu Tsidu, D. Ene, D. Wunch, Z. Cao, O. Garcia, M. Ramonet, F. Vogel, and J. Orphal, “Building the COllaborative Carbon Column Observing Network (COCCON): long-term stability and ensemble performance of the EM27/SUN Fourier transform spectrometer,” Atmos. Meas. Tech. 12(3), 1513–1530 (2019). [CrossRef]
7. B. Holben, T. Eck, I. Slutsker, D. Tanre, J. Buis, A. Setzer, E. Vermote, J. Reagan, Y. Kaufman, and T. Nakajima, “AERONET—a federated instrument network and data archive for aerosol characterization,” Remote Sens. Environ. 66(1), 1–16 (1998). [CrossRef]
8. M. De Mazière, A. M. Thompson, M. J. Kurylo, J. D. Wild, G. Bernhard, T. Blumenstock, G. O. Braathen, J. W. Hannigan, J.-C. Lambert, T. Leblanc, T. J. McGee, G. Nedoluha, I. Petropavlovskikh, G. Seckmeyer, P. C. Simon, W. Steinbrecht, and S. E. Strahan, “The Network for the Detection of Atmospheric Composition Change (NDACC): history, status and perspectives,” Atmos. Chem. Phys. 18(7), 4935–4964 (2018). [CrossRef]
9. X. Jiang, D. Crisp, E. T. Olsen, S. S. Kulawik, C. E. Miller, T. S. Pagano, M. Liang, and Y. L. Yung, “CO2 annual and semiannual cycles from multiple satellite retrievals and models,” Earth Space Sci. 3(2), 78–87 (2016). [CrossRef]
10. M. Gisi, F. Hase, S. Dohe, T. Blumenstock, A. Simon, and A. Keens, “XCO2-measurements with a tabletop FTS using solar absorption spectroscopy,” Atmos. Meas. Tech. 5(11), 2969–2980 (2012). [CrossRef]
11. P. I. Palmer, E. L. Wilson, G. L. Villanueva, G. Liuzzi, L. Feng, A. J. DiGregorio, J. Mao, L. Ott, and B. Duncan, “Potential improvements in global carbon flux estimates from a network of laser heterodyne radiometer measurements of column carbon dioxide,” Atmos. Meas. Tech. 12(4), 2579–2594 (2019). [CrossRef]
12. H. R. Melroy, E. L. Wilson, G. B. Clarke, L. E. Ott, J. Mao, A. K. Ramanathan, and M. L. McLinden, “Autonomous field measurements of CO2in the atmospheric column with the miniaturized laser heterodyne radiometer (Mini-LHR),” Appl. Phys. B: Lasers Opt. 120(4), 609–615 (2015). [CrossRef]
13. J. Kurtz and S. O’Byrne, “Multiple receivers in a high-resolution near-infrared heterodyne spectrometer,” Opt. Express 24(21), 23838–23848 (2016). [CrossRef]
14. H. Deng, C. Yanga, W. Wanga, C. Shana, Z. Xua, B. Chena, L. Yaoa, M. Hua, R. Kana, and Y. He, “Near infrared heterodyne radiometer for continuous measurements of atmospheric CO2 column concentration,” Infrared Phys. Technol. 101, 39–44 (2019). [CrossRef]
15. H. Nakagawa, S. Aoki, H. Sagawa, Y. Kasaba, I. Murata, G. Sonnabend, M. Sornig, S. Okano, J. R. Kuhn, J. M. Ritter, M. Kagitani, T. Sakanoi, M. Taguchi, and M. Kagitani, “IR heterodyne spectrometer MILAHI for continuous monitoring observatory of Martian and Venusian atmospheres at Mt. Haleakalā, Hawaii,” Planet. Space Sci. 126, 34–48 (2016). [CrossRef]
16. G. Sonnabend, D. Wirtz, V. Vetterle, and R. Schieder, “High-resolution observations of Martian non-thermal CO2 emission near 10 um with a new tunable heterodyne receiver,” Astron. Astrophys. 435(3), 1181–1184 (2005). [CrossRef]
17. A. Hoffmann, N. A. Macleod, M. Huebner, and D. Weidmann, “Thermal infrared laser heterodyne spectroradiometry for solar occultation atmospheric CO2 measurements,” Atmos. Meas. Tech. 9(12), 5975–5996 (2016). [CrossRef]
18. D. Weidmann, W. J. Reburn, and K. M. Smith, “Ground-based prototype quantum cascade laser heterodyne radiometer for atmospheric studies,” Rev. Sci. Instrum. 78(7), 073107 (2007). [CrossRef]
19. J. Wang, G. Wang, T. Tan, G. Zhu, C. Sun, Z. Cao, W. Chen, and X. Gao, ““Mid-infrared laser heterodyne radiometer (LHR) based on a 3.53 µm room-temperature interband cascade laser,” Opt. Express 27(7), 9610–9619 (2019). [CrossRef]
20. M. Sornig, T. A. Livengood, G. Sonnabend, D. Stupar, and P. Kroetz, “Direct wind measurements from November 2007 in Venus upper atmosphere using ground-based heterodyne spectroscopy of CO2 at 10 mm wavelength,” Icarus 217(2), 863–874 (2012). [CrossRef]
21. A. Yu. Klimchuk, A. I. Nadezhdinskii, Ya. Ya. Ponurovskii, and Yu. P. Shapovalov, and A. V. Rodin, “On the possibility of designing a high-resolution heterodyne spectrometer for near-IR range on the basis of a tunable diode laser,” Quantum Electron. 42(3), 244–249 (2012). [CrossRef]
22. S. G. Zenevich, A. Y. Klimchuk, V. M. Semenov, M. V. Spiridonov, and A. V. Rodin, “Measurements of the fully resolved contour of the carbon dioxide absorption line in a band at λ = 1.605 um in the atmospheric column using high-resolution heterodyne spectroradiometry method,” Quantum Electron. 49(6), 604–611 (2019). [CrossRef]
23. T. Kostiuk and M. J. Mumma, “Remote sensing by IR heterodyne spectroscopy,” Appl. Opt. 22(17), 2644–2654 (1983). [CrossRef]
24. A. E. Siegman, “The antenna properties of optical heterodyne receivers,” Appl. Opt. 5(10), 1588–1594 (1966). [CrossRef]
25. E. J. Moyer, D. S. Sayres, G. S. Engel, J. M. St. Clair, F. N. Keutsch, N. T. Allen, J. H. Kroll, and J. G. Anderson, “Design considerations in high-sensitivity off-axis integrated cavity output spectroscopy,” Appl. Phys. B 92(3), 467–474 (2008). [CrossRef]
26. I. E. Gordon, L. S. Rothman, C. Hill, R. V. Kochanov, Y. Tana, P. F. Bernath, M. Birk, V. Boudon, A. Campargue, K. V. Chance, B. J. Drouin, J.-M. Flaud, R. R. Gamache, J. T. Hodges, D. Jacquemart, V. I. Perevalovm, A. Perrin, K. P. Shine, M.-A. H. Smith, J. Tennyson, G. C. Toon, H. Tran, V. G. Tyuterev, A. Barbe, A. G. Császár, V. M. Devi, T. Furtenbacher, J. J. Harrison, J.-M. Hartmann, A. Jolly, T. J. Johnson, T. Karman, I. Kleiner, A. A. Kyuberis, J. Loos, O. M. Lyulin, S. T. Massie, S. N. Mikhailenkom, N. Moazzen-Ahmadi, H. S. P. Müller, O. V. Naumenkom, A. V. Nikitinm, O. L. Polyansky, M. Rey, M. Rotger, S. W. Sharpe, K. Sung, E. Starikovam, S. A. Tashkunm, J. Vander Auwera, G. Wagner, J. Wilzewski, P. Wcisło, S. Yuh, and E. J. Zak, “The HITRAN2016 molecular spectroscopic database,” J. Quant. Spectrosc. Radiat. Transfer 203, 3–69 (2017). [CrossRef]
27. D. P. Deea, S. M. Uppala, A. J. Simmons, P. Berrisford, P. Poli, S. Kobayashi, U. Andrae, M. A. Balmaseda, G. Balsamo, P. Bauera, P. Bechtold, A. C. M. Beljaars, L. van de Bergd, J. Bidlot, N. Bormann, C. Delsol, R. Dragani, M. Fuentes, A. J. Geer, L. Haimberger, S. B. Healy, H. Hersbach, E. V. Holm, L. Isaksen, P. Kallberg, M. Kohler, M. Matricardi, A. P. McNally, B. M. Monge-Sanz, J.-J. Morcrette, B.-K. Park, C. Peubey, P. de Rosnay, C. Tavolato, J.-N. Thepaut, and F. Vitarta, “The ERA-Interim reanalysis: configuration and performance of the data assimilation system,” Q. J. R. Meteorol. Soc. 137(656), 553–597 (2011). [CrossRef]
28. E. Kalnay, M. Kanamitsu, R. Kistler, W. Collins, D. Deaven, L. Gandin, M. Iredell, S. Saha, G. White, J. Woollen, Y. Zhu, A. Leetmaa, and R. Reynolds, “The NCEP/NCAR 40-Year Reanalysis Project,” Bull. Am. Meteorol. Soc. 77(3), 437–471 (1996). [CrossRef]
