Open Access
Add to BibSonomy
Abstract
Pure rotational Raman lidar is often used for atmospheric temperature profile measurements. However, high elastic scattering suppression ratios (>107) are required for temperature measurement in clouds and haze, which imposes stringent requirements on spectral separation techniques. To solve this problem, a lidar measurement technique based on vibrational and rotational Raman spectra is proposed. Using nitrogen vibrational and rotational Raman scattering to obtain temperature profiles under strong elastic scattering, combined with the dual-rotational Raman temperature measurements under weak elastic scattering, a vertical distribution of atmospheric temperature including cloud and haze layers, can be obtained. The feasibility of the method was verified by numerical simulation. The Raman lidar for temperature measurements was established in Xi'an University of Technology, and the obtained temperature results show good agreement with the radiosonde measurements. The proposed method combines the high sensitivity of the dual-rotational Raman method and the high Mie-scattering suppression of the vibrational Raman method, thus further improving the adaptability of Raman lidar to cloudy and hazy air conditions and supporting atmospheric and cloud physics research.
© 2022 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
Temperature is an important parameter for describing the physical state of the atmosphere, in which it directly affects the thermal process and water vapor condensation [1–2]. Currently, radiosondes are generally used to obtain vertical profiles of atmospheric temperature. Although this approach can provide can provide high measurement accuracy, it is susceptible to the influence of external objective conditions such as airspace, and its low temporal resolution and horizontal drift can no longer meet the urgent needs of real-time meteorological process analysis. As an active remote sensing technology with high spatial and temporal resolution, lidar can effectively enrich the methods of measuring atmospheric temperature profiles.
At present, lidar technologies that can be used for atmospheric temperature measurement mainly include hyperspectral acquisition Rayleigh spectroscopy technology and rotational Raman technology. Among them, the application and development of hyperspectral lidar has been limited to a certain extent owing to its high structural complexity and high spectral resolution spectroscopic technical requirements [3–4]. In relative terms, the theoretical foundation of using rotational Raman spectra for temperature retrieval is solid, and the corresponding Raman lidar system is less complex and easier to build. Rotational Raman (RR) relative temperature measurement is performed by extracting two portions of pure rotational Raman line signals with opposite temperature dependencies [5–16]. In addition, the fine extraction of a single Raman spectral line can provide technical support for atmospheric temperature measurement [5–9]. Notably, Wuhan University designed the single-line-extraction technique for 532 nm band to extract single Raman line signals and retrieved the temperature without a calibration process [7–9]. Because the intensity of elastic scattering is at least three to four orders of magnitude higher than that of rotational Raman scattering, to finely extract the weak rotational Raman signal, the elastic scattering suppression ratios reach at least 60 dB to avoid the contamination of elastic scattering [10]. In recent years, RR temperature measurement technology has developed rapidly [11], and its measurement performance has been verified under clear sky conditions [10–21]. Behrendt and Reichardt developed a dual-rotational Raman temperature measurement system with a laser wavelength of 532.25 nm, and the elastic scattering suppression ratio greater than 60 dB was found to be suitable for temperature measurement under the condition of a few thin clouds [15]. When the elastic scattering suppression ratio is high enough, pure rotational Raman lidar can accurately measure tropospheric temperature including optically-thin aerosols and clouds [8–9]. However, most rotational Raman lidars cannot be well for measuring temperature of cloud and haze layers owing to the fact that the low quantum number (low-J) rotational Raman line signals, which decrease with increasing temperature, are highly susceptible to elastic scattering signals.
The occurrence and dissipation of haze are closely related to the vertical distribution of temperature. The temperature distribution in cloud layers is also crucial for exploring cloud processes. However, haze (aerosol) and cloud particles lead to a strong elastic scattering process, which is unfavorable for extracting weak rotational Raman signals. For a 354.7 nm incident laser light and a central wavelength (CWL) of the high-quantum-number (high-J) rotational Raman spectrum channel (CH-H) of 352.5 nm, the spectral spacing is approximately 1.8 nm, and it is easy to achieve a higher elastic scattering suppression ratio. However, for the CWL of the low quantum number (low-J) rotational Raman spectrum channel (CH-L) of 353.9 nm, the spectral spacing is approximately 0.8 nm, and it is difficult to achieve efficient suppression of strong elastic scattering. Therefore, elastic scattering contamination in CH-L has become a major problem hindering the practical application of RR thermometry lidars. To improve the elastic scattering suppression ratio, some scholars have used a combination of three narrow-band interference filters [21]. However, this method reduces the spectral transmittance and affects the signal-to-noise ratio (SNR) of the system. In addition, other scholars have proposed a method for Mie-scattering correction for CH-L, which can eliminate the elastic scattering signal using software [22]. However, the delivery process is complicated, and the timeliness and accuracy are low.
To measure the temperature in cloud or haze layer by Raman lidar and to improve the adaptability of Raman temperature measurement lidar in practical applications, we propose a segmented spliced vibrational-rotational Raman atmospheric temperature measurement method. Because the vibrational Raman scattering spectra have a large wavelength shift relative to elastic scattering, a high elastic scattering suppression ratio can be easily achieved. Therefore, based on the nitrogen vibrational Raman spectra and temperature-dependent rotational Raman spectra, the temperature retrieval under strong elastic scattering conditions can be achieved [23–24]; this is called the vib-rotational Raman (VR) temperature measurement. Combined with the RR method, a temperature stitching inversion method suitable for sub-measurement intervals under different conditions was constructed. The system measurement performance under different weather conditions was simulated and analyzed, and the feasibility of segmental fusion atmospheric temperature measurement based on vibrational and rotational Raman scattering spectra was verified. An experiment based on a multichannel Raman lidar, which was established at Xi’an University of Technology (XAUT), was carried out, and the temperature profiles on cloudy and hazy days were obtained.
2. Temperature measurement method
The theoretical basis for a temperature Raman lidar to retrieve temperature profiles is that the signal intensity ratio of the high-J and low-J Raman lines from atmospheric molecules (N2 or O2) has a simple temperature dependence, which follows a Boltzmann distribution [9,25]. A temperature Raman lidar usually extracts two portions of Raman line signals with opposite temperature dependencies: high-J Raman line signals, whose intensity increases with increasing temperature, and low-J Raman line signals. Atmospheric temperature measurement can be achieved by the correlation between the signal intensity ratio and temperature [15,26], and several formulas are used for the inversion algorithm [19–20].
Regarding the weakly dependence between the vibrational Raman spectral signal and temperature, the high-J Raman line signals are the core variables of VR. Similarly, the temperature profile can be retrieved by the signal intensities ratio. The signal intensity corresponding to the CH-H and the nitrogen vibrational Raman channel (CH-V) can be expressed as [18–22,24,27]:
Because the nitrogen vibrational Raman scattering has a large wavelength shift relative to the incident wavelength, the transmittances of the two cannot be easily ignored. The aerosol extinction coefficient α of the atmospheric transmission path is obtained using the Raman method. The transmittance difference between the two signals is calculated, and the ratio can be corrected accordingly. The correction process can effectively improve the system's ability to identify the temperature-inversion phenomenon. A quadratic function is used to fit Gvr [25], and the functional relationship between the ratio and temperature is obtained. The system constant ratio can be simplified by the system calibration process, and the temperature measurement can be achieved after system calibration with other parallel devices.
The sensitivity S of lidar to temperature measurement is defined as the fractional change in the influence function when the temperature changes by 1 K under a given temperature condition [28], as follows:
Assuming that the solar background light during the day, the dark current of the photodetector, and the various kinds of noise caused by the load resistance are all irrelevant, the SNR of each channel measurement can be expressed as:
The statistical error depending on the sensitivity and SNR can be expressed as:
3. Comparative analysis of RR and VR
3.1 Simulation and analysis under clear sky
The spectral band of the low-J Raman lines is relatively narrow and closer to the elastic scattering spectra. And CH-L may be affected by the oxygen molecules Raman line signal intensity which increases with increasing temperature [12]. Therefore, the out-of-band suppression rate and adjacent crosstalk should be considered when constructing the RR system. The full width at half height (FWHM) corresponding to the CH-H, CH-L and CH-V were 1.0 nm, 0.5 nm and 0.5 nm, respectively.
According to the atmospheric standard model and the formulas described in Section 2, when the CWLs of CH-H and CH-L are 352.5 nm and 353.9 nm, respectively, the atmospheric echo signal under a clear sky was simulated. The primary simulation parameters of the system are listed in Table 1.
Table 1. Main simulation parameters
The thick red solid line in Fig. 1(b) is the RSCS of Mie-Rayleigh scattering, namely, the elastic scattering signal; the black dashed line and the purple long dashed line correspond to the rotational Raman echo signals of CH-L and CH-H, respectively; and the thin blue solid line represents the nitrogen vibrational Raman signal. It can be seen that there was no strong Mie-scattering process under a clear sky, and the elastic scattering signal was mainly Rayleigh scattering, which is approximately three orders of magnitude higher than the low-J Raman signal.
Fig. 1. Clear sky condition. (a) Backscatter ratio; (b) range squared correction signal (RSCS); (c) temperature measurement sensitivity.
The temperature measurement sensitivity can be calculated, as shown by the solid line in Fig. 1(c). As vibrational Raman scattering is not severely affected by temperature, the sensitivity of the VR temperature measurement is slightly lower than that of RR.
As each Raman spectrum has a different temperature sensitivity, the position of the CWL corresponding to the filter directly affects the sensitivity. In view of the small Raman spectral range corresponding to CH-L, it was set as a fixed value here, and the corresponding relationship between the spectral separation position and temperature sensitivity is analyzed by changing the CWL of CH-H. The dotted and dot-dashed lines in Fig. 1(c) are the temperature measurement sensitivities when the CWL of the spectral separation curve was blue-shifted and red-shifted. The results show that the corresponding sensitivities of the two methods increased with decreasing temperature. The sensitivity increased when the CWL was blue-shifted, and the corresponding sensitivity decreased when the central wavelength was red-shifted. Although the sensitivity in the blueshift situation was significantly increased, this does not mean that the measurement performance of the system improved. The intensity of the rotational Raman spectra corresponding to CH-H dropped sharply at this time, which would seriously affect the SNR of the system, and introduce a large system error. Therefore, the selection of the CWL corresponding to the spectral separation should consider both sensitivity and SNR.
Although the VR temperature measurement sensitivity is relatively low, the temperature measurement accuracy can be improved by increasing the SNR. Figure 2 shows the SNR and statistical errors under clear sky conditions. The black and blue data in the figure correspond to RR and VR, respectively, where the solid lines are the SNRs and the dotted lines are the statistical errors. Setting the SNR to be greater than 100 can meet the measurement requirements corresponding to the effective measurement height of the system. The maximum measurement heights corresponding to RR and VR are both above 8 km, and the statistical errors at 8 km correspond to 0.4 K and 0.7 K, respectively. In theory, CH-H should extract high-J line signals as completely as possible to improve the SNR while ensuring high sensitivity. However, owing to the Lorentzian tail characteristic of the spectral separation curve, it cannot be achieved. This low sensitivity limits the effective identification of small-scale temperature inversion processes. The results show that the detection performance of RR is better than that of VR under clear sky conditions; therefore, RR is more suitable for temperature detection in a clean atmosphere.
3.2 Simulation and analysis in the haze and cloud layers
3.2.1 Analysis of the influence of the out-of-band suppression rate
The strong elastic scattering in the cloud and haze layers imposes strict technical requirements for the RR temperature measurement lidar system. To explore the applicable conditions of RR and VR for temperature measurement, a simulation analysis was performed under two weather conditions (hazy and cloudy), and the sensitivity and measurement error were further discussed. In the simulation, the elastic scattering suppression ratios of the CH-H and CH-V were set to 70 dB. The influence of the mixing of elastic scattering signals on the temperature measurement accuracy was simulated by changing the elastic scattering suppression ratio of the CH-L.
The value of aerosol backscatter ratio corresponding to hazy days is approximately 2-10. The simulation process was based on the changing trend of the backscatter coefficients in the atmospheric model, and the maximum backscatter ratios in the range of 0-3 km were constructed to be 2, 5, and 8. The aerosol distribution conditions are shown in Fig. 3(a). The curves shown in Fig. 3(b) correspond to the distribution of different atmospheric scattering echo signals with height under the condition that the elastic scattering suppression ratios of CH-H is set to 50 dB and the maximum backscatter ratio of the bottom layer is 5. As the echo intensity is affected by aerosols, it was significantly attenuated in areas with large backscatter.
Figure 4(a) shows the temperature simulation results when the elastic scattering suppression ratio of the CH-L is 50 dB and the maximum backscatter ratio is 5. Figures 4(b)–(d) show the relative deviations when the elastic scattering suppression ratios are 50 dB, 55 dB, and 60 dB. As VR can effectively avoid elastic scattering contamination, it will not be discussed further. RR has a temperature deviation of approximately 1.5 K at the position of the maximum backscatter ratio, while the theoretical temperature deviation of VR is almost zero.
The simulation results show that the temperature measurement process of RR will introduce a measurement deviation of at least 1.5 K if the suppression ratio is 50 dB and the backscatter ratio is greater than 5; when the suppression ratio is 55 dB and the backscatter ratio is greater than 8, the temperature deviation is greater than 0.6 K. Only when the suppression ratio is higher than 60 dB, can the temperature measurement process of RR achieve high-precision measurement of atmospheric temperature with a deviation of less than 0.1 K.
The backscatter ratio corresponding to the cloud layer is usually greater than 10. Based on the measured cloud backscatter coefficient, the distributions of the intracloud backscatter ratios with maximum backscatter ratios of 10, 20, and 30 were constructed, as shown in Fig. 5(a). Figure 5(b) shows the atmospheric echo signal intensity when the elastic scattering suppression ratio of CH-L is 50 dB and the backscatter ratio is 20. It can be seen that if there is strong Mie-scattering, the intensity of the elastic scattering signal increases significantly, whereas the vibrational and rotational Raman signals are significantly attenuated.
Figure 6 shows the temperature simulation results and relative deviations if the elastic scattering suppression ratio of the CH-L is 50 dB and the backscatter ratio is 20. Additionally, due to the contamination of elastic scattering, a temperature deviation of approximately 6.5 K occurs at the position of the maximum backscatter ratio in RR, whereas the temperature deviation of VR is nearly zero.
The simulation results show that if the suppression ratio is 50 dB and the backscatter ratio is greater than 10, the temperature deviation of RR is approximately 3 K, and the deviation with the backscatter ratio (∼5) is approximately 1.7 K; when the suppression ratio is 55 dB and the backscatter ratio is greater than 20, a temperature deviation of at least 1.5 K will introduced. A deviation of less than 0.3 K can be achieved only when the suppression ratio is 60 dB. However, the temperature deviation is reversed at this time. This is because the CH-H signal attenuates sharply with an increasing backscatter ratio. The temperature results are positively biased.
Based on the above simulation analysis, RR has very strict technical requirements for the optical parameters of the system, and it is difficult to achieve effective measurement of the complete atmospheric temperature profile under cloudy and hazy conditions. To further explore the feasibility of the segmented spliced vibrational-rotational Raman temperature measurement (the vibrational-rotational splicing method), a simulation analysis of the SNR corresponding to RR and VR was carried out.
3.2.2 Measurement performance on cloudy and hazy days
Compared to RR, the spectral separation relationship of VR is beneficial for atmospheric temperature measurement in strong elastic scattering regions. By setting the elastic scattering suppression ratio of CH-L to 50 dB, the SNRs and statistical errors of hazy and cloudy days under different backscatter ratios can be obtained, as shown in Fig. 7. The black solid line, dotted line, and dot-dashed line in Figs. 7(a),(b) represent the SNRs and statistical errors corresponding to RR when the backscatter ratio is no more than 2, 5, and 8, respectively. The representation of Figs. 7(c),(d) is the same as that of Figs. 7(a),(b), but with maximum backscatter ratios of 10, 20, and 30.
The maximum allowable error was set to 3 K. When the backscatter ratio is 5 (8), the maximum measurement height of RR on hazy days is approximately 4.3 km (7.3 km), respectively, and those of VR are approximately 3.8 km (6.6 km). On cloudy days, when backscatter ratio is 20 (30), the maximum measurement height of RR from the cloud base height (∼5 km) is approximately 2.7 km (4.3 km), respectively, and those of VR are approximately 2 km (3.5 km). In addition, the SNR decreases significantly with an increase in the scattering ratio, which shows that the laser beam will not be able to penetrate the heavy haze and dense cloud layer, as a result of the strong attenuation. The SNR corresponding to RR at the same height is slightly higher than that of VR. The maximum relative difference in the haze layer between the two statistical errors is about 0.5 K; the maximum difference in the cloud layer is approximately 0.7 K. Under the same conditions, the temperature difference between RR and VR in the haze layer or cloud layer is small. Therefore, to account for the high sensitivity of RR and the high elastic scattering suppression ratio of VR, RR and VR can be used to obtain the atmospheric temperature under clean air and strong elastic scattering, respectively. A complete atmospheric temperature profile was obtained by segmented splicing.
For clouds and haze, weather conditions can be divided into four categories: clear sky, haze and no clouds, clouds and no haze (CNH), and haze and clouds (CH). Among them, RR can be directly used in the first condition, and the vibrational-rotational splicing method can be used in other cases. Figure 8 shows the backscatter ratio and temperature simulation results under haze and cloud conditions. The red solid and dashed lines in Fig. 8(b) correspond to the temperature simulation profiles of RR, and the blue solid line corresponds to that of VR. It is clear that RR has an obvious temperature deviation in the shaded region, whereas VR has good consistency with the model temperature. In fact, in the case of haze and clouds, the effective measurement height will decrease owing to large signal attenuation. Based on this result, the segmented and spliced atmospheric temperature measurement method can effectively obtain a complete profile of atmospheric temperature for cloudy and hazy days.
Fig. 8. Simulation results of the vibrational-rotational splicing method. (a) Backscatter ratio; (b) temperature.
4. Lidar system and temperature observations
Experimental observations were performed based on the XAUT lidar, and the vibrational-rotational splicing method was used to retrieve the atmospheric temperature profile. The feasibility of the segmented spliced atmospheric temperature measurement method was verified by comparing the radiosonde data. The composition of the system is illustrated in Fig. 9.
The system employs a pulsed Nd: YAG laser as a light source operating at a frequency-tripled wavelength of 354.7 nm with a 20 Hz repetition rate, an energy output of 240 mJ, and a 9 ns FWHM pulse duration. A Cassegrain telescope was used as the optical receiver, and a narrow-band interference filter was used as the core filter device to finely separate the Raman spectra. A photomultiplier tube and data acquisition card were combined to perform photoelectric conversion and data acquisition. The elastic scattering suppression ratios of the CH-H, CH-L and CH-V were approximately higher than 65 dB, 50 dB, and 65 dB, respectively.
When the backscatter ratio is relatively small, the temperature results corresponding to the two methods have a small difference; therefore, the overlapping area near the haze boundary and the cloud boundary can be selected for temperature splicing. The backscatter ratio (∼3) was set as the boundary criterion and the upper and lower boundaries of the aerosol layer were discriminated based on the derivative of the backscatter ratio. Taking the boundary value as the starting point, a range of 300 m in the direction of decreasing backscatter ratio was considered, and the height value corresponding to the minimum temperature difference between the two methods was indexed in this range. The 100 m upper and lower intervals at this height were taken as the boundaries of the temperature splicing area, and the backscatter ratio was used as the weight for average processing.
Figure 10 shows the experimental observation results for the CNH condition on May 15, 2019. The red solid line in Fig. 10(a) is the RSCS of the elastic scattering channel, and the blue solid line, black dashed line and purple long dashed line correspond to the CH-V, CH-L, and CH-H signals, respectively. According to the elastic scattering, it can be clearly seen a significant cloud layer with a thickness of approximately 2.5 km begins to appear from 4.5 km. The bottom aerosol level is relatively low, and it can be considered that the bottom layer is approximately haze-free. In the strong elastic scattering region, the inelastic scattering channels all show significant signal attenuation, which is consistent with theory. In addition, compared with the CH-H and CH-V, CH-L suffers from signal distortion under the action of Mie scattering; that is, Mie-scattering crosstalk occurs in this region.
The black solid line in Fig. 10(b) represents the radiosonde data, and the blue and red solid lines correspond to the temperature of RR and VR, respectively. The blue dotted line represents the temperature inversion result of RR inside the cloud layer. By comparison, it was found that under the existing system parameters, RR has a large temperature deviation and cannot effectively achieve temperature inversion in the cloud layer, whereas VR can accurately obtain the vertical distribution of atmospheric temperature. In the effective measurement range (∼6.7 km), the vibrational-rotational splicing temperature is highly consistent with the radiosonde data. Owing to the attenuation of the signal, it cannot completely penetrate the cloud layer to effectively measure the complete cloud layer. Nevertheless, the results still show that the vibrational-rotational splicing method can obtain a relatively complete vertical profile of atmospheric temperature.
Figure 11 shows the experimental observation results of the CH condition on December 15, 2020. Affected by objective conditions, the laser energy was set to ∼180 mJ during this experiment. The subject of the illustration is the same as that of Fig. 10. It is clear that a relatively strong elastic scattering process is caused by the underlying aerosol and low-level cloud layers. At the same time, the corresponding signal distortion occurs in the CH-L, resulting in a large temperature deviation in the corresponding interval of RR. In contrast, VR can obtain temperature profiles consistent with the radiosonde data in the haze layer and cloud layer, thus effectively compensating for the deficiency of RR.
Figure 10(c) and Fig. 11(c) correspond to the relative temperature deviations under the CNH and CH conditions, respectively. The results show that the vibrational-rotational splicing method can achieve a measurement accuracy with a maximum deviation of 2 K within the effective measurement range.
Figure 12 shows the continuous observation results of the cloud layer on the night of May 15, 2019. The left part shows the RSCS of Mie-Rayleigh scattering signal and the aerosol backscatter ratio. The black dotted line with asterisk indicates the cloud base height. It can be seen that there are continuous cloud layers above 3.8 km, and there is no haze at the bottom. Furthermore, an aerosol layer with a thickness of about 2 km appears in the interval from 1.3 km to 3.2 km. The right part shows the continuous temperature distribution and the temperature profiles. As shown, this method can effectively measure the temperature inside the cloud layer and provide data support for the in-depth study of cloud processes. These experimental results show that this method can effectively identify the temperature inside the haze layer and cloud layer, which is helpful to improve the adaptability of Raman temperature measurement lidar to external conditions.
Fig. 12. Continuous observation results. (a) RSCS of M-Rayleigh signal; (b) atmospheric temperature.
5. Discussion and conclusions
To improve the ability of Raman lidar for temperature measurement in cloud or haze layers, a segmented spliced atmospheric temperature measurement method based on the principle of rotational and vibrational Raman scattering is proposed. Based on the atmospheric standard model, the system measurement performance corresponding to the RR and VR was theoretically simulated. It was found that the penetration ability of the two methods for haze and clouds is similar, but VR easily achieves a higher elastic scattering suppression ratio, and thus can obtain the atmospheric temperature in strong elastic scattering areas. The feasibility of the vibrational-rotational segmented splicing temperature measurement method was verified through simulation analysis. Finally, experimental verification was carried out using the multichannel lidar system of Xi'an University of Technology. The results show that this method can effectively eliminate the blind spot of RR in the measurement of strong elastic scattering regions and obtain atmospheric temperature profiles on cloudy and hazy days with deviations of less than 2 K. It is worth noting that owing to the relatively weak Raman scattering intensity, the effective measurement height of the Raman method is limited by harsh atmospheric transmission conditions. In summary, the segmented spliced vibrational-rotational Raman temperature measurement technology can improve the adaptability of Raman lidar to cloudy and hazy conditions and is helpful for the analysis and research of meteorological processes. Nevertheless, certain noteworthy technical requirements for the practical application. For example, a higher SNR is required to obtain a more complete temperature profiles inside the cloud. In addition, special attention should be paid to the optical path consistency of the vibrational and rotational Raman channels in the system construction; otherwise, a complex optical path correction will be required.
The simulation analysis in this study is mainly based on the optical parameters of the existing multi-channel lidar system, such as the CH-H with 1.0 nm@352.5 nm and the CH-L with 0.5 nm@353.9 nm. Although this parameter does not affect the analysis and verification of the vibration-rotation Raman splicing temperature measurement technology, the optimal selection of system parameters is an important factor affecting the system detection performance. To explore the optimal settings of the parameters (CWL and FWHM) of CH-H and CH-L, a modified Gaussian curve was used to extract the signals from the anti-Stokes branch in the simulations. This indicates that the SNR and statistical errors, which were simulated by the optimized system parameters (CH-H with 0.5 nm@353 nm and CH-L with 0.3 nm@354.05 nm), improved accordingly. In comparison, the detection performance of RR significantly improved. Although a smaller FWHM limits the extraction of rotational Raman signals, it can improve daytime detection performance to a certain extent. Choosing between the two still requires a more in-depth analysis and discussion. This shows that the existing lidar system still has room for further optimization.
Funding
National Natural Science Foundation of China (42130612, 61875163).
Acknowledgments
The authors gratefully acknowledge the Xi'an Atmospheric Exploration Center for the data support. Writing assistance was provided by Huige Di and Dengxin Hua.
Disclosures
The authors declare no conflicts of interest.
Data availability
Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.
References
1. Committee on Developing Mesoscale Meteorological Observational Capabilities to Meet Multiple Needs, National Research Council, Observing Weather and Climate from the Ground Up: A Nationwide Network of Networks, The National Academies, (2009).
2. V. Wulfmeyer, R. M. Hardesty, D. D. Turner, A. Behrendt, M. P. Cadeddu, P. Di Girolamo, P. Schlüssel, J. Van Baelen, and F. Zus, “A review of the remote sensing of lower tropospheric thermodynamic profiles and its indispensable role for the understanding and the simulation of water and energy cycles,” Rev. Geophys. 53(3), 819–895 (2015). [CrossRef]
3. P. Piironen and E. W. Eloranta, “Demonstration of a high-spectral-resolution lidar based on an iodine absorption filter,” Opt. Lett. 19(3), 234–236 (1994). [CrossRef]
4. D. Hua, M. Uchida, and T. Kobayashi, “Ultraviolet high-spectral-resolution Rayleigh-Mie lidar with a dual-pass Fabry–Perot etalon for measuring atmospheric temperature profiles of the troposphere,” Opt. Lett. 29(10), 1063–1065 (2004). [CrossRef]
5. S. Chen, Z. Qiu, Y. Zhang, H. Chen, and Y. Wang, “A pure rotational Raman Lidar using double-grating monochromator for temperature profile detection,” J. Quant. Spectrosc. Radiat. Transfer 112(2), 304–309 (2011). [CrossRef]
6. J. Jia and F. Yi, “Atmospheric temperature measurements at altitudes of 5–30 km with a double-grating-based pure rotational Raman lidar,” Appl. Opt. 53(24), 5330–5343 (2014). [CrossRef]
7. M. Weng, F. Yi, F. Liu, Y. Zhang, and X. Pan, “Single-line-extracted pure rotational Raman lidar to measure atmospheric temperature and aerosol profiles,” Opt. Express 26(21), 27555–27571 (2018). [CrossRef]
8. F. Liu, F. Yi, Y. Zhang, and Y. Yi, “Double-Receiver-Based Pure Rotational Raman LiDAR for Measuring Atmospheric Temperature at Altitudes Between Near Ground and Up To 35 km,” IEEE Trans. Geosci. Remote Sensing 57(12), 10301–10309 (2019). [CrossRef]
9. F. Liu, R. Wang, F. Yi, W. Huang, C. Ban, W. Pan, Z. Wang, and H. Hu, “Pure rotational Raman lidar for full-day troposphere temperature measurement at Zhongshan Station (69.37°S, 76.37°E), Antarctica,” Opt. Express 29(7), 10059–10076 (2021). [CrossRef]
10. J. Reichardt, U. Wandinger, V. Klein, I. Mattis, B. Hilber, and R. Begbie, “RAMSES: German Meteorological Service autonomous Raman lidar for water vapor, temperature, aerosol, and cloud measurements,” Appl. Opt. 51(34), 8111–8131 (2012). [CrossRef]
11. J. Cooney, “Measurement of Atmospheric Temperature Profiles by Raman Backscatter,” J. Appl. Meteor. 11(1), 108–112 (1972). [CrossRef]
12. M. Radlach, A. Behrendt, and V. Wulfmeyer, “Scanning rotational Raman lidar at 355 nm for the measurement of tropospheric temperature fields,” Atmos. Chem. Phys. 8(2), 159–169 (2008). [CrossRef]
13. Y. F. Arshinov, S. M. Bobrovnikov, V. E. Zuev, and V. M. Mitev, “Atmospheric temperature measurements using a pure rotational Raman lidar,” Appl. Opt. 22(19), 2984–2990 (1983). [CrossRef]
14. G. Vaughan, D. P. Wareing, S. J. Pepler, L. Thomas, and V. Mitev, “Atmospheric temperature measurements made by rotational Raman scattering,” Appl. Opt. 32(15), 2758–2764 (1993). [CrossRef]
15. A. Behrendt and J. Reichardt, “Atmospheric temperature profiling in the presence of clouds with a pure rotational Raman lidar by use of an interference-filter-based polychromator,” Appl. Opt. 39(9), 1372–1378 (2000). [CrossRef]
16. J. Totems, P. Chazette, and A. Baron, “Mitigation of bias sources for atmospheric temperature and humidity in the mobile Raman Weather and Aerosol Lidar (WALI),” Atmos. Meas. Tech. 14(12), 7525–7544 (2021). [CrossRef]
17. P. Achtert, M. Khaplanov, F. Khosrawi, and Gumbel, “Pure rotational-Raman channels of the Esrange lidar for temperature and particle extinction measurements in the troposphere and lower stratosphere,” Atmos. Meas. Tech. 6(1), 91–98 (2013). [CrossRef]
18. R. K. Newsom, D. D. Turner, and J. E. M. Goldsmith, “Long-Term Evaluation of Temperature Profiles Measured by an Operational Raman Lidar,” J. Atmos. Ocean. Technol. 30(8), 1616–1634 (2013). [CrossRef]
19. D. Wu, Z. Wang, P. Wechsler, N. Mahon, M. Deng, B. Glover, M. Burkhart, W. Kuestner, and B. Heesen, “Airborne compact rotational Raman lidar for temperature measurement,” Opt. Express 24(18), A1210–1223 (2016). [CrossRef]
20. P. Di Girolamo, R. Marchese, D. N. Whiteman, and B. B. Demoz, “Rotational Raman Lidar measurements of atmospheric temperature in the UV,” Geophys. Res. Lett. 31(1), L01106 (2004). [CrossRef]
21. E. Hammann, A. Behrendt, F. Le Mounier, and V. Wulfmeyer, “Temperature profiling of the atmospheric boundary layer with rotational Raman lidar during the HD(CP)2 Observational Prototype Experiment,” Atmos. Chem. Phys. 15(5), 2867–2881 (2015). [CrossRef]
22. J. Su, M. P. McCormick, Y. Wu, R. B. Lee, L. Lei, Z. Liu, and K. R. Leavor, “Cloud temperature measurement using rotational Raman lidar,” Journal of Quantitative Spectroscopy and Radiative Transfer 125, 45–50 (2013). [CrossRef]
23. W. S. Heaps, J. Burris, and J. A. French, “Lidar technique for remote measurement of temperature by use of vibrational-rotational Raman spectroscopy,” Appl. Opt. 36(36), 9402–9405 (1997). [CrossRef]
24. J. Su, M. P. McCormick, and L. Lei, “New technique to retrieve tropospheric temperature using vibrational and rotational Raman backscatter,” Earth Space Sci. 7(4), e2019EA000817 (2020). [CrossRef]
25. A. Behrendt, “Temperature measurements with lidar,” in Lidar: Range-resolved Optical Remote Sensing of the Atmosphere, C. Weitkamp, ed. (Springer, 2005).
26. D. Hua, J. Liu, K. Uchida, and T. Kobayashi, “Daytime Temperature Profiling of Planetary Boundary Layer with Ultraviolet Rotational Raman Lidar,” Jpn. J. Appl. Phys. 46, 5849–5852 (2007). [CrossRef]
27. F. Liu and F. Yi, “Lidar-measured atmospheric N2 vibrational-rotational Raman spectra and consequent temperature retrieval,” Opt. Express 22(23), 27833–27844 (2014). [CrossRef]
28. D. Hua, M. Uchida, and T. Kobayashi, “Ultraviolet Rayleigh-Mie lidar with Mie-scattering correction by Fabry–Perot etalons for temperature profiling of the troposphere,” Appl. Opt. 44(7), 1305–1314 (2005). [CrossRef]
