Remote Sensing of Atmospheric Methane with IR OPO Lidar System

: A differential absorption lidar (DIAL) system designed on the basis of optical parametric oscillators (OPO) with nonlinear KTiOAsO 4 (KTA) and KTiOPO 4 (KTP) crystals is described. The crystals allow laser radiation tuning in the infrared region (IR) wavelength region. The measurements in the 3.30–3.50 µ m spectral range, which includes a strong absorption band of methane, are carried out. Lidar backscattered signals in the spectral band 3.30–3.50 µ m has been measured and analyzed along the horizontal path in the atmosphere. Based on the experimental results, CH 4 concentrations ~2.085 ppm along a 800 m surface path are retrieved in the spectral range under study with a spatial resolution of 100 m. The technical characteristics of the main units of the IR OPO lidar system are given. It allows the reception and processing of backscattered signals in the IR spectral region along the surface tropospheric paths. The experimental results show applicability of the IR OPO lidar system to study the distribution of methane concentration (~2.085 ppm) at 800 m atmospheric path with a spatial resolution of 100 m in the mid-IR. The error level of methane


Introduction
Methane (CH4) is one of the most important organic substances in the atmosphere. The background value of methane concentration in the atmospheric surface layer is about 2 ppm and an increase in this value leads to an increase in the greenhouse effect, since methane intensively absorbs the thermal radiation of the earth in the infrared region (IR).
With increasing methane content, chemical processes in the atmosphere change, which can lead to a deterioration of the ecological situation on earth. Therefore, monitoring the content of CH4 to control and prevent an increase in its total content in the atmosphere is an important task. The use of lidar monitoring systems for methane in the atmosphere using OPO lasers is promising.
Advances in research and manufacture of nonlinear optical crystals make it possible to create new gas analysis systems [1][2][3] and thereby expand the possibilities of remote monitoring of the environment.
For example, a differential absorption lidar (DIAL) system based on two BBO (beta barium borate, β-barium borate or β-BaB 2 O 4 crystals) optical power amplifiers pumped by an Nd:YAG laser (pulse length is~3 ns, pulse frequency is up to 100 Hz, and tuning range is 400-2500 nm) is described in [4].
As experimental results, the authors presented the vertical distribution of water vapor along a 1500-m path at the wavelengths λ on = 1187.869 nm and λ off = 1187.716 nm measured with a spatial resolution of 15 m. The use of a LiNbO 3 crystal pumped with an Nd:YAG laser allowed the authors of [5] to design a gas analysis system with the laser radiation tuned within the 1410-1850 and 2900-4100 nm wavelength ranges, an output energy of 1-45 mJ, and a radiation line width of 3-3.5 cm −1 . The system specification includes a possibility of detecting CH 4 using topographic targets at distances of 2-5 km with an sensitivity of 1 ppm along a path.
In work [6], daily fluctuations of CO 2 at a wavelength of 1570 nm were measured for the first time with the use of a DIAL KTP-OPO system with Nd:YAG pumping and output pulse energy of 70-100 mJ.

Measurement Technique
The DIAL lidar operation is based on the fact that a part of backscattered laser radiation is converted by molecules and aerosol particles during propagation through the atmosphere. The laser beam is attenuated (by the Lambert-Beer law) because of molecular absorption and molecular and aerosol scattering. Thus, a signal detected is a function of this attenuation, the fraction of photons backscattered, the path length, and the laser radiation power. The signal P R from a scattering layer of ∆z in thickness can be represented as where P 0 (λ) is the laser radiation power; A D is the area of the receiving telescope; O(z) is the area of overlapping of a laser beam and field-of-view of the receiver; β(z,λ) is the mass coefficient of backscattered radiation; η(λ) is the efficiency of the transmitting-receiving system; ∆z is the spatial resolution along the sounding path; τ(z,λ) is the mass coefficient of attenuation. The DIAL technique estimates the concentration of an atmospheric gas from the difference between lidar signals at the wavelengths λ on and λ off with different molecular absorption. The concentration of a gas is defined as: where ∆σ abs is the absorption cross section. Equation (2) is true only if the interfering absorption by other gases is weak and scattering properties of aerosol particles do not change in the ∆λ and ∆z ranges.
The minimum detectable concentrations (MDC) of gases in experiments with the use of DIAL technique in the lidar sounding is defined as [25]: where P(ν ON ,z) and P(ν OFF ,z) are the signals recorded by the photodetector from the sounding path of z in length at the frequencies ν ON and ν OFF , which are in a narrow spectral range of the absorption line of a gas under study and in the neighboring transparency window, and is the differential absorption coefficient. Here, K(ν ON ) and K(ν OFF ) are the absorption coefficients inside and outside the absorption line. For the majority of real recording and digital processing instruments used in the experiments, the difference in logarithms in Equation (3) should satisfy the condition [25]: Considering Equation (5), minimal intensities of the absorption lines can be determined from simulation and comparison between atmospheric transmission spectra of a gas under study and of interfering gases with the concentrations corresponding to threshold limit value (TLV) concentration for a 10-km path [26]. The DIAL technique is described in more detail in [26,27].

IR OPO Lidar System
The universal scheme of the OPO lidar system developed ( Figure 1) is used in field atmospheric measurements. During the measurements in the atmosphere, a part of laser radiation reflected by CaF 2 plate is used to measure the transmitted pulse energy in the experimental setup. A visible laser at 532 nm is used to align the invisible IR radiation to maintain the trans-receiver axes. Further, the wavelength of the laser radiation is measured using spectrometer S100 as shown in Figure 1. Photodetector records a backscattered signal from the atmosphere collected by the receiving telescope Cassegrain.

Laser Source
The laser system [28] (SOLAR LS, Minsk, Russia) consists of the LQ529B pumping laser and KTA/KTP radiation converter which operates in the IR region; it provides tunable generation of nanosecond radiation pulses in the spectral ranges 1.4-1.6, 1.8-2.6, and 3-4 μm. In this work spectral radiation of the laser source from 3.30 to 3.50 μm is used. The main parameters of the laser source are given in Table 1. The appearance of the master oscillator and the radiation conversion unit are shown in Figure 2. The laser system is described in more detail in [24,28,29].

Laser Source
The laser system [28] (SOLAR LS, Minsk, Russia) consists of the LQ529B pumping laser and KTA/KTP radiation converter which operates in the IR region; it provides tunable generation of nanosecond radiation pulses in the spectral ranges 1.4-1.6, 1.8-2.6, and 3-4 µm. In this work spectral radiation of the laser source from 3.30 to 3.50 µm is used. The main parameters of the laser source are given in Table 1. The appearance of the master oscillator and the radiation conversion unit are shown in Figure 2. The laser system is described in more detail in [24,28,29].

Laser Source
The laser system [28] (SOLAR LS, Minsk, Russia) consists of the LQ529B pumping laser and KTA/KTP radiation converter which operates in the IR region; it provides tunable generation of nanosecond radiation pulses in the spectral ranges 1.4-1.6, 1.8-2.6, and 3-4 μm. In this work spectral radiation of the laser source from 3.30 to 3.50 μm is used. The main parameters of the laser source are given in Table 1. The appearance of the master oscillator and the radiation conversion unit are shown in Figure 2. The laser system is described in more detail in [24,28,29].  Laser system [28]. Figure 2. Laser system [28].

Collimator Mirrors
An initial laser beam is broadened from 4 to 40 mm by mirror collimator, which allows a decrease in the radiation divergence angle to 100 µrad. The collimator consist of two mirrors MPD00M9-P01 and MPD269-P01 (Thorlabs, Newton, NJ, USA) [30]. Parameters of collimator mirrors are given in Table 2.

Receiving Telescope
A Cassegrain telescope developed at the IAO SB RAS was used as a receiver optics to collect the backscattered signal from the atmosphere. The coating material of the telescope mirrors allows operation in the entire lasing range of the laser. Table 3 presents the telescope specification. When the telescope is included in the scheme of experimental setup, a direct connection to the spectral selection equipment is assumed. Figure 3 shows a receiving telescope and a mirror collimator.

Collimator Mirrors
An initial laser beam is broadened from 4 to 40 mm by mirror collimator, which allows a decrease in the radiation divergence angle to 100 μrad. The collimator consist of two mirrors MPD00M9-P01 and MPD269-P01 (Thorlabs, Newton, NJ, USA) [30]. Parameters of collimator mirrors are given in Table 2.

Receiving Telescope
A Cassegrain telescope developed at the IAO SB RAS was used as a receiver optics to collect the backscattered signal from the atmosphere. The coating material of the telescope mirrors allows operation in the entire lasing range of the laser. Table 3 presents the telescope specification. When the telescope is included in the scheme of experimental setup, a direct connection to the spectral selection equipment is assumed. Figure 3 shows a receiving telescope and a mirror collimator.

Photodetectors
Lidar signals are detected using thermally cooled IBSG photo diode (IBSG, St.Petersburg, Russia) [31], PD36-03-TEC-PRW with AMT-07M amplifier. Characteristics of photo detector are presented in Table 4. The power of the laser source is measured with the use of Ophir Nova II, Vega, and Thorlabs PM100USB power and energy meters with PE10-C, PE25-C, PE25BB, and ES120C measuring heads, respectively.

Data Acquisition Device
The data acquisition is provided by Rigol 1104Z-S and Acute TS2202E oscilloscopes. A Rigol 1104Z-S oscilloscope is a classic device which supports the NI-VISA software interface [32] and a capability of connecting the device to a notebook via USB and Ethernet interfaces. Acute TS2202E is a compact USB device with a possibility of expanding the number of input channels via the connection to six identical devices (12 analog inputs). Characteristics of the devices are given in Table 5.
The acquisition of experimental data is automated by software which combines individual functional units of the setup into a single system.

Calibration of the Lidar System
To calibrate the lidar system and measure the absorption coefficients at the selected informative sounding wavelengths, the scheme in Figure 1 was used. A calibration gas cell to the scheme was added. The optical scheme of the experiments is shown in Figure 4. A series of laboratory experiments on measurements of the laser radiation absorption by methane in a molecular mixture was carried out. The 3.30-3.50 μm spectral range (pairs of on and off sounding wavelengths [33]) was selected for the experiments. It contains a quite strong methane absorption band. A CH4:N2 (2:98) molecular mixture in a cell at a pressure of 1 atm was used.
The laser system radiation is guided to KG-01 gas cell (100 mm long; the diameter of the CaF2window is 40 mm) by plane-parallel CaF2 plates and a rotating mirror. Before propagating through the gas cell, a part of the radiation is directed onto the thermally cooled photodiode No. 2 because of the reflection from the CaF2 plate (division ratio is 50/50) and a reference signal is thus recorded. Thermally cooled photodiode No. 3 measures an informative signal at the gas cell exit. The gas mixture transmission spectra are calculated from the ratio of the reference to informative signals. The coefficient of OPO laser radiation absorption by the molecular mixture in the gas cell is calculated by the Lambert-Beer law.
Mirror collimator and receiving telescope within the scheme allow recording backscattered radiation from the atmosphere and, at the same time, using the cell as a calibration component during full-scale measurements. The complete methodology for atmospheric measurements in Section 3.1 is described. Table 6 shows the OPO laser radiation absorption coefficients in the calibrated CH4:N2 (2:98) gas mixture at a pressure of 1 atm in the wavelength range 3.30-3.50 μm [33], calculated based on HITRAN database [34] and experimentally measured. In this spectral range, the tuning was made to more than 60 wavelengths of the OPO laser, including both informative and uninformative wavelengths for methane sounding. The information about the latter is useful for the calibration of the OPO lidar system.  A series of laboratory experiments on measurements of the laser radiation absorption by methane in a molecular mixture was carried out. The 3.30-3.50 µm spectral range (pairs of on and off sounding wavelengths [33]) was selected for the experiments. It contains a quite strong methane absorption band. A CH 4 :N 2 (2:98) molecular mixture in a cell at a pressure of 1 atm was used.
The laser system radiation is guided to KG-01 gas cell (100 mm long; the diameter of the CaF 2 window is 40 mm) by plane-parallel CaF 2 plates and a rotating mirror. Before propagating through the gas cell, a part of the radiation is directed onto the thermally cooled photodiode No. 2 because of the reflection from the CaF 2 plate (division ratio is 50/50) and a reference signal is thus recorded. Thermally cooled photodiode No. 3 measures an informative signal at the gas cell exit. The gas mixture transmission spectra are calculated from the ratio of the reference to informative signals. The coefficient of OPO laser radiation absorption by the molecular mixture in the gas cell is calculated by the Lambert-Beer law.
Mirror collimator and receiving telescope within the scheme allow recording backscattered radiation from the atmosphere and, at the same time, using the cell as a calibration component during full-scale measurements. The complete methodology for atmospheric measurements in Section 3.1 is described. Table 6 shows the OPO laser radiation absorption coefficients in the calibrated CH 4 :N 2 (2:98) gas mixture at a pressure of 1 atm in the wavelength range 3.30-3.50 µm [33], calculated based on HITRAN database [34] and experimentally measured. In this spectral range, the tuning was made to more than 60 wavelengths of the OPO laser, including both informative and uninformative wavelengths for methane sounding. The information about the latter is useful for the calibration of the OPO lidar system. Experiments to identify the absorption lines of methane in the spectral region from 3.3460 µm to 3.4790 µm were carried out. The absorption coefficients using these experimental data were calculated. Further, these lines have been validated by analyzing the HITRAN database. Based on our analysis, laser probing wavelengths at 3.4157 µm and 3.4177 µm as online and offline wavelength for methane measurements in the atmosphere were chosen.
The discrepancy of the absorption coefficients in the spectral regions that are not informative for gas analysis can be due to the influence of interfering absorption by external gases. The absorption coefficients measured (by the laser pulse energy after the absorbing cell) are in a good agreement with the absorption coefficients calculated using the HITRAN database at the sounding-informative wavelengths.

Measurement Results and Discussion
To test the capabilities the lidar system developed for methane sounding, the tuning range was reduced to informative for the gas under study (the measurements have been carried out in the 3.30-3.50 µm spectral range, which includes a quite strong absorption band of methane). Figures 5 and 6 shows the experimental design, the main components of the OPO lidar system and experimental geometry during full-scale measurements in the atmosphere. Experiments to identify the absorption lines of methane in the spectral region from 3.3460 μm to 3.4790 μm were carried out. The absorption coefficients using these experimental data were calculated. Further, these lines have been validated by analyzing the HITRAN database. Based on our analysis, laser probing wavelengths at 3.4157 μm and 3.4177 μm as online and offline wavelength for methane measurements in the atmosphere were chosen.
The discrepancy of the absorption coefficients in the spectral regions that are not informative for gas analysis can be due to the influence of interfering absorption by external gases. The absorption coefficients measured (by the laser pulse energy after the absorbing cell) are in a good agreement with the absorption coefficients calculated using the HITRAN database at the sounding-informative wavelengths.

Measurement Results and Discussion
To test the capabilities the lidar system developed for methane sounding, the tuning range was reduced to informative for the gas under study (the measurements have been carried out in the 3.30-3.50 μm spectral range, which includes a quite strong absorption band of methane). Figures 5 and 6 shows the experimental design, the main components of the OPO lidar system and experimental geometry during full-scale measurements in the atmosphere.     Experiments to identify the absorption lines of methane in the spectral region from 3.3460 μm to 3.4790 μm were carried out. The absorption coefficients using these experimental data were calculated. Further, these lines have been validated by analyzing the HITRAN database. Based on our analysis, laser probing wavelengths at 3.4157 μm and 3.4177 μm as online and offline wavelength for methane measurements in the atmosphere were chosen.
The discrepancy of the absorption coefficients in the spectral regions that are not informative for gas analysis can be due to the influence of interfering absorption by external gases. The absorption coefficients measured (by the laser pulse energy after the absorbing cell) are in a good agreement with the absorption coefficients calculated using the HITRAN database at the sounding-informative wavelengths.

Measurement Results and Discussion
To test the capabilities the lidar system developed for methane sounding, the tuning range was reduced to informative for the gas under study (the measurements have been carried out in the 3.30-3.50 μm spectral range, which includes a quite strong absorption band of methane). Figures 5 and 6 shows the experimental design, the main components of the OPO lidar system and experimental geometry during full-scale measurements in the atmosphere.   Figure 7 shows the tuning curve of the atmospheric response, the spectral range of which includes informative methane absorption lines, for example, 3.39 μm (U is the voltage of the backscattered signal at the photodiode detector, V; R is the atmospheric path, m).  Figure 7 shows the tuning curve of the atmospheric response, the spectral range of which includes informative methane absorption lines, for example, 3.39 µm (U is the voltage of the backscattered signal at the photodiode detector, V; R is the atmospheric path, m).
Using the IR OPO lidar system (see Figure 5), the distribution of CH 4 concentrations were experimentally studied from sounding at the pair of wavelengths 3415.711 (on-line) and 3417.484 nm Atmosphere 2020, 11, 70 9 of 13 (off-line) [33] with a spatial resolution of 100 m in urban environment. The measurements were carried out in Tomsk on 1 February 2019.
Atmosphere 2020, 11, x FOR PEER REVIEW 9 of 13 Using the IR OPO lidar system (see Figure 5), the distribution of CH4 concentrations were experimentally studied from sounding at the pair of wavelengths 3415.711 (on-line) and 3417.484 nm (off-line) [33] with a spatial resolution of 100 m in urban environment. The measurements were carried out in Tomsk on 1 February 2019.   Figure 9 shows the lidar signal measured under the wavelength tuning in the spectral range 3.30-3.43 μm (alternate accumulation of signals at informative wavelengths at discrete sets of lidar radiation lines was performed; the switching time between the wavelengths was 35-50 ms and the averaging time was 6.4 s).  Atmosphere 2020, 11, x FOR PEER REVIEW 9 of 13 Using the IR OPO lidar system (see Figure 5), the distribution of CH4 concentrations were experimentally studied from sounding at the pair of wavelengths 3415.711 (on-line) and 3417.484 nm (off-line) [33] with a spatial resolution of 100 m in urban environment. The measurements were carried out in Tomsk on 1 February 2019.   Figure 9 shows the lidar signal measured under the wavelength tuning in the spectral range 3.30-3.43 μm (alternate accumulation of signals at informative wavelengths at discrete sets of lidar radiation lines was performed; the switching time between the wavelengths was 35-50 ms and the averaging time was 6.4 s).  Figure 9 shows the lidar signal measured under the wavelength tuning in the spectral range 3.30-3.43 µm (alternate accumulation of signals at informative wavelengths at discrete sets of lidar radiation lines was performed; the switching time between the wavelengths was 35-50 ms and the averaging time was 6.4 s).
Atmosphere 2020, 11, x FOR PEER REVIEW 10 of 13 The results of experimental measurements of the CH4 concentration distribution along a 800 m sounding path with the OPO lidar system are shown in Figure 10. The background concentration of methane was calculated using the equation shown in Equation (2). The path averaged methane concentration of about 2.085 ppm over the path length of 800 m was retrieved. The measurement results were compared (Figure 10b) with the data from the TOR (tropospheric ozone research) station of the IAO SB RAS [35,36], where the methane concentration is around-the-clock measured. The OPO lidar and the TOR station are located in the same building. The error level of methane concentration recovery correspond to 8-13% relative to the averaged value.

Conclusions
The IR OPO lidar system has been designed. The technical characteristics of the main units of the IR OPO lidar system are given. It allows the reception and processing of backscattered signals in the IR spectral region along the surface tropospheric paths. The experimental results show applicability of the IR OPO lidar system to study the distribution of methane concentration (~2.085 ppm) at 800 m atmospheric path with a spatial resolution of 100 m in the mid-IR. The error level of The results of experimental measurements of the CH 4 concentration distribution along a 800 m sounding path with the OPO lidar system are shown in Figure 10. The results of experimental measurements of the CH4 concentration distribution along a 800 m sounding path with the OPO lidar system are shown in Figure 10. The background concentration of methane was calculated using the equation shown in Equation (2). The path averaged methane concentration of about 2.085 ppm over the path length of 800 m was retrieved. The measurement results were compared (Figure 10b) with the data from the TOR (tropospheric ozone research) station of the IAO SB RAS [35,36], where the methane concentration is around-the-clock measured. The OPO lidar and the TOR station are located in the same building. The error level of methane concentration recovery correspond to 8-13% relative to the averaged value.

Conclusions
The IR OPO lidar system has been designed. The technical characteristics of the main units of the IR OPO lidar system are given. It allows the reception and processing of backscattered signals in the IR spectral region along the surface tropospheric paths. The experimental results show applicability of the IR OPO lidar system to study the distribution of methane concentration (~2.085 ppm) at 800 m atmospheric path with a spatial resolution of 100 m in the mid-IR. The error level of The background concentration of methane was calculated using the equation shown in Equation (2). The path averaged methane concentration of about 2.085 ppm over the path length of 800 m was retrieved. The measurement results were compared (Figure 10b) with the data from the TOR (tropospheric ozone research) station of the IAO SB RAS [35,36], where the methane concentration is around-the-clock measured. The OPO lidar and the TOR station are located in the same building. The error level of methane concentration recovery correspond to 8-13% relative to the averaged value.

Conclusions
The IR OPO lidar system has been designed. The technical characteristics of the main units of the IR OPO lidar system are given. It allows the reception and processing of backscattered signals in the IR spectral region along the surface tropospheric paths. The experimental results show applicability of the IR OPO lidar system to study the distribution of methane concentration (~2.085 ppm) at 800 m atmospheric path with a spatial resolution of 100 m in the mid-IR. The error level of methane concentration recovery correspond to 8-13% relative to the averaged value. The lidar methane measurements are in a good agreement with the data of the TOR station.
The results of this work confirm the applicability of the OPO lidar system designed to the study and control the distribution of CH 4 concentration in the surface air layer, and to retrieve CH 4 concentration close to background or near-background values.