Water is the only atmospheric component with three phases. In this work, ultraviolet Raman lidar is developed for synchronous measurements of water vapor, liquid water and ice water in Xi’an (34.233°N, 108.911°E), China. Different interference filters are designed to construct individual water Raman channels, and the corresponding central wavelength and bandwidth are determined by 399.0 nm (3.1 nm), 403.0 nm (5.0 nm) and 407.6 nm (0.6 nm) in ice water, liquid water and water vapor Raman channels, respectively. The mutual interference effect originating from the overlapping characteristics of water Raman spectra is further analyzed, and an accurate retrieval method based on linear simultaneous equations and mutual interference degrees is proposed for synchronous three-phase water mixing ratio profiles. Preliminary measurements are carried out in the Centre for lidar remote sensing research of Xi’an University of Technology, and representative measurement examples are obtained and validated for the performance of the Raman lidar system. Synchronous mixing ratio profiles in water vapor, liquid water and ice water are retrieved, and the corresponding extinction coefficient and relative humidity profiles are also combined to reveal the variation characteristics in three-phase waters. The possible aerosol fluorescence are analyzed as well, and it is inferred that the aerosol fluorescence might affect (possibly overestimate) the derived mixing ratio values of the liquid water and ice water. The effective detection can reach up to a height of 5 km under cloudy weather, and synchronized growth in water vapor and liquid water content is obtained in cloud layers. Continuous observations are also made under hazy weather conditions, and the temporal and spatial evolution trends of three-phase waters in clouds are successfully realized. Preliminary exploration and results validate the feasibility of ultraviolet Raman lidar for synchronous measurements of atmospheric water vapor, liquid water and ice water.
© 2019 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
Water is the only atmospheric component with three phases. Water vapor plays an important role in cloud evolution, precipitation, global water cycle and climate change. Liquid water in clouds is not only an important component in the balance of atmospheric water budget but also an important microphysical parameter for understanding cloud physical processes. Super-cooled water in the cloud (which remains liquid below 0°C) is also one of the most important parameters indicative of the potential for artificial precipitation enhancement. Therefore, the study of the three-phase water distribution with high precision and high spatial-temporal resolution is of great significance in understanding the formation of clouds and precipitation, fine prediction of precipitation, and artificial precipitation enhancement [1–3].
Lidar is a powerful tool in the remote sensing of atmospheric components and atmospheric properties. With the development of laser technology and Raman spectroscopy, Raman lidar has succeeded in measuring atmospheric temperature, water vapor and aerosols [4–6]. Ro-vibrational Raman scattering signals from water vapor molecules (H2O) and nitrogen molecules (N2) have been widely used in the detection of water vapor profiles. Many ground-based Raman lidar systems have been developed around the world, and significant achievements have been made in the measurement of atmospheric water vapor and aerosol profiles [7–11]. These achievements are focused on the water vapor Raman scattering signals with a frequency shift of 3657 cm-1.
In recent years, several institutes have carried out research on three-phase waters [12–15]. In 2000, Veselovskii et al. developed a Raman lidar to separate the Raman signals from water vapor and liquid water in the tropospheric layer, and studied their relative intensities for various weather conditions . In 2004, Wang et al. developed a Raman lidar to detect the solid-water content in cirrus clouds and proposed the inversion method of the solid-water mixing ratio . In 2012, Yi Fan et al. obtained complete Raman spectra of three-phase water by a 32-channel grating spectrometer, and achieved lidar-returned signals of atmospheric water vapor and liquid water in the range of 0-6 km . Affected by overlapping Raman spectra, synchronous measurements of water vapor, liquid water and ice have encountered great difficulties [19–22]. Further research on synchronous detection technique, especially on synchronous retrievals of accurate mixing ratio profiling, is required.
In this paper, an ultraviolet Raman lidar system is established for synchronous measurements of atmospheric three-phase waters. The system configuration is introduced in section 2. The central wavelength and bandwidth of the filters in individual water Raman channels determine the spectral interference degree. The mutual interference effect between three-phase water Raman channels is first discussed in section 3 and further applied to the synchronous accurate retrieval of three-phase water mixing ratio profiles. In sections 4 and 5, we display preliminary experiments and several representative results. Synchronous measurements of atmospheric three-phase water below 5 km can be realized, and the synchronous growth of water vapor, liquid water and ice water in clouds is achieved. The representative results validate the performance of Raman lidar for three-phase waters under different weather conditions.
2. Raman lidar system
The ultraviolet Raman lidar system used in this study was developed in the Centre for lidar remote sensing research of Xi’an University of Technology, Xi’an, China (34.233°N, 108.911°E). A schematic diagram of the system is presented in Fig. 1. The system employs a pulsed Nd:YAG laser as a light source. The laser operates at a frequency-tripled wavelength of 354.7 nm and has a 20 Hz repetition rate, an energy output of 150 mJ, and a 9 ns full width at half maximum (FWHM) pulse duration. Returned signals are collected using a 600 mm Newtonian telescope with a field of view of 0.4 mrad, and then coupled into a multimode optical fiber and guided into a spectroscopic box. A set of dichroic mirrors (DMs) and narrow-band interference filters (IFs) is used to construct a high-efficiency polychromator, which divides the returned signals into five channels. Channel 1 is used for the detection of elastic Mie-Rayleigh scattering signals. Channels 2-5 are used for the detection of different vibrational Raman scattering signals; channel 2 is used for detection of vibrational Raman lidar returns of nitrogen molecules, and channel 3, 4, and 5 are used to detect Raman scattering signals of ice, liquid water and water vapor molecules, respectively. All the signals are detected by PMTs. Analog mode with a high-gain and high-speed amplifier was used for the lidar returns from the elastic scattering and nitrogen vibrational Raman channels, and a photon-counting technique was adopted to obtain the desirable detection sensitivity in three Raman channels for water vapor, liquid water and ice. The integration time for the three-phase water measurement was 5 min, and the integration time for the aerosol measurements was 3 min. All the raw data were taken with an altitude resolution of 3.75 m and were smoothed with a sliding window length of 75-150 m. All the heights in this study were relative to the ground level.
The DMs require an incident angle of 45° and have dichroic characteristics. They transmit the spectra in a certain wavelength range with a peak transmittance of >98% and almost completely reflect other wavelengths, and the IFs are used to select the wavelengths of interest and to reduce the background radiation and block the strong elastic signals. DM1 can transmit lidar returns with spectral wavelengths greater than 380 nm, and DM2 transmits spectral wavelengths larger than 395 nm. Thus, the reflection of DM1 and DM2 is used to build channel 1 of the elastic scattering signal and channel 2 of the nitrogen vibrational Raman signals. Fine extraction of these two spectra is executed by the narrow-band interference filters IF1 and IF2, with CWLs (central wavelength) of 354.7 nm and 386.7 nm, respectively, and a bandwidth of 1.0 nm. The transmittance by DM2 is divided by BS1 and BS2 into three parts, of which one part, with 50% of the energy, is used to build the ice Raman channel 3, and IF3 is used to extract the ice Raman scattering signals with a CWL of 399.0 nm and a bandwidth of 3.1 nm. The remaining two parts are used to build the liquid water and water vapor Raman channels 4 and 5, respectively. IF4 in the liquid water Raman channel is designed with CWL of 403.0 nm and a bandwidth of 5.0 nm, and IF5 with a central wavelength of 407.6 nm and a bandwidth of 0.6 nm is used to extract water vapor Raman scattering signals in the water vapor Raman channel. The main specifications of the Raman lidar are listed in Table 1.
3. Accurate retrieval method based on mutual interference effect
3.1 Retrieval method for synchronous water mixing ratio
If the pure Raman scattering signal of individual phase water can be extracted, the corresponding water mixing ratio profile can be obtained by direct inversion according to the definition of mixing ratio, which uses the ratio of the Raman scattering signals of the three-phase water and nitrogen molecules. Due to the extremely narrow bandwidth of IF5 in our design, it is thought that the pure water vapor Raman scattering signals can be extracted from the water vapor Raman channel, and thus, the atmospheric water vapor mixing ratio WW(z) can be directly given as :
Due to the overlapping characteristics of three-phase water Raman spectrum, however, it is difficult to extract pure liquid water and ice Raman signals; the mutual interference effect is required in the synchronous inversion of three-phase water mixing ratios. In regard to liquid water, the contribution of ice water and water vapor must be considered. The liquid water mixing ratio WL(z) can be deduced as
The ice water mixing ratio WI(z) can be deduced as ,
It can be seen from the numerator of the above Eqs. (1), (3) and (4) that the purpose of subtraction is to obtain the pure Raman scattering signal intensity of three-phase waters. However, it is still not rigorous enough, because the interference between liquid water and ice water is reciprocal. For this reason, an accurate retrieval method based on linear simultaneous equations and mutual interference degrees is proposed for synchronous three-phase water mixing ratio profiles, as follows,
3.2 Interference effect from the three-phase water Raman spectrum
Due to the spectral overlapping characteristics of three-phase water, it is essential to take into account the mutual interference effects of different water phases. Figure 2 shows the Raman spectrum distribution of three-phase waters [17, 22, 28]. The Raman spectra of liquid water and ice water present obvious differences from that of water vapor. These spectra are continuous in the wavelength range of 395-408 nm with close central wavelengths and obvious overlapping regions. The peak wavelengths of liquid water and ice water are located at ∼402.9 nm and 398.7 nm, respectively. Additionally, the Raman spectrum of water vapor also overlaps partially with that of liquid water. Thus, the overlapping characteristics of Raman scattering spectra of ice, liquid water and water vapor make it difficult to detect three-phase water accurately and precisely, and the corresponding interference effects cannot be ignored among different phase waters.
To better express the interference effects, the interference degree is used and defined as the contribution proportion of x’ phase water in the x phase water Raman channel, as follows:
Figure 3 shows the distribution of signal relative intensity and the interference degree of liquid water in the water vapor Raman channel, and the star mark in the figure shows the CWL and FWHM of the filter in the experiment. The left figure Fig. 3(a) corresponds to the influence of filter parameters on signal relative intensity. The relative intensity of water vapor scattering signal can reach ∼ 0.6 with the filter central wavelength of 407.6 nm and bandwidth of less than 0.5 nm, and it gradually increases to ∼ 0.9 with the increasing bandwidth of 2.0 nm. As shown in Fig. 3(b), we can see that the interference degree has a complex relationship with the central wavelength and bandwidth of the filter. Under the same central wavelength, the interference degree of liquid water increases with increasing filter bandwidth. For example, the interference degree reaches more than 20% with a filter bandwidth of > 2 nm and with a central wavelength of 407-407.2 nm, reflecting the obvious interference effect of the liquid water signal on the water vapor signal. The interference degree is < 10% when the central wavelength of the filter is 407.4-407.6 nm and the bandwidth is less than 2 nm, and it decreases with bandwidth with the lowest value being < 5%. Thus, when the ultra-narrowband interference filter is selected in the water vapor Raman channel, the interference degree of liquid water is less than 5%; that is to say, the contribution of liquid water can be ignored in the water vapor Raman channel, and the pure atmospheric water vapor Raman scattering signal can be extracted.
In regard to liquid water, the results are more complex and conflicting because interference effects from ice water and water vapor contribute to the liquid water signals simultaneously. Figure 4 shows the influence of the filter optical parameters on the ice interference degree and water vapor interference degree in the liquid water Raman channel, and the distribution of relative intensity of liquid water Raman scattering signal is shown as well. As presented in Fig. 4(a), the relative intensity of liquid water Raman scattering signal increases with the filter bandwidth under the same central wavelength. For example, at the fixed filter central wavelength of 403.0 nm, the relative intensity increases from 0.45 to 0.9 when the filter bandwidth increases from 2 to 6 nm. Thus, the increasing filter bandwidth is conducive for signal acquisition in liquid water Raman channel. However, the interference effects from ice and water vapor reflects different distribution trends. Figure 4(b) corresponds to the interference degree results from water vapor signals. It is very low, and most values are less than 2%. With a fixed central wavelength of 403 nm in the liquid water channel, the water vapor interference degree increases from 1% to approximately < 10% when the bandwidth increases from 2 to 6 nm. Figure 4(c) corresponds to the interference degree results from the ice signals. The interference degree decreases with increasing redshift of center wavelength. When the central wavelength of the filter is in the range of 399-402 nm, the interference degree from ice water increases with increasing bandwidth, reaching more than 100% with a bandwidth of > 3 nm. With the redshift of the center wavelength increasing to 403-405 nm, the interference degree of ice is obviously reduced to <60%. Thus, the interference effect between the ice and liquid water is obvious and non-negligible. Therefore, the optical parameter selection in the liquid water Raman channel directly affects the interference degree information of ice and water vapor in the liquid water Raman channel.
Figure 5 shows influence of filter parameters on signal relative intensity and the interference degree of liquid water in the ice Raman channel. It can be seen that the maximum signal intensity exists in the range of wavelength of 399-400 nm. The interference degree of liquid water exceeds 80% when the filter central wavelength in the ice Raman channel is larger than 400 nm; it decreases to less than 40% with a central wavelength of < 399 nm and a bandwidth of < 3 nm. It should be pointed out that the interference degree of water vapor in the ice channel is ignored because the Raman spectrum of water vapor is far away from that of ice water.
This analysis not only serves as the theoretical basis for the design of Raman spectroscopic system, but also provides the corresponding values of interference degree, which play an important role in the retrieval method for synchronous water mixing ratios. According to our designed spectroscopic system in our study, the interference degrees from water vapor and ice, DLW and DLI in liquid water Raman channel, are valued ∼0.063 $\, \pm 0.003$ and 0.831 $\, \pm 0.08$, respectively. The interference degree DIL from liquid water is valued 0.493 $\, \pm 0.05$ and a negligible fraction, 0.00005, of the water vapor contribution to the signal in ice Raman channel. For the water vapor Raman channel, the interference degree from liquid water, DWL, is valued 0.052 $\, \pm 0.005$, and the interference degree from ice valued 0.025 with uncertainty of 5%, respectively.
4. Measurements examples and discussion
The Raman lidar system was built and operated at Xi’an University of Technology, Xi’an (34.233°N, 108.911°E), China. Some tests were conducted to validate the system performance under different weather conditions. Figure 6 presents a representative measurement example taken at 21:39 CST on September 29, 2018. The example is used to validate the performance of Raman lidar for three-phase water under cloudy weather. From the range-square-corrected signals (RSCS) of lidar returns, as shown in Fig. 6(a), we can see that two obvious peaks appear at heights of ∼4.5 km and ∼5.5 km from the elastic scattering signals, which correspond to the information of thin cloud layers; however, the cloud information does not appear in the attenuated nitrogen vibrational Raman scattering signals, which verifies a high rejection rate for elastic scattering signals. Synchronous growth of water vapor and liquid water in the clouds can be observed as well. Furthermore, the system signal-to-noise ratio (SNR) and the statistical errors for three-phase waters measurements are analyzed. SNR curves in individual water Raman channels are discussed and shown in Fig. 6(b). It can be seen that the channel SNRs can reach a value of 10 to a height of 5.5 km and 4.8 km for water vapor Raman channel and liquid water Raman channel, however, the SNR for ice Raman channel is relatively lower, and the SNR values greater than 10 are obtained only to a height of 3.0 km, which indicated that it is unable to reach the height of clouds.
Atmospheric extinction coefficient profile is retrieved and shown in Fig. 7(a). Atmospheric aerosols are distributed homogeneously below 4.0 km. The aerosol extinction coefficient is ∼0.2 km-1 near the surface, corresponding to a visibility of ∼30 km, which is consistent with the ground visibility meter measurements. The extinction coefficient increase greatly at 4.5 km in the cloudy layer. The simultaneous relative humidity profile is also shown as the dotted line. It varied with height within the range of 20% -60% below 4.0 km, and increased to ∼100% in the cloudy layer. According to the retrieval method, the atmospheric water vapor mixing ratio and liquid water mixing ratio profile are retrieved and shown in Figs. 7 (b) and 7(c). The atmospheric water vapor mixing ratio is approximately 3 g/kg at the bottom layer, and it decreased gradually with height. Little increase was obtained at 2-3 km and 3-3.5 km, indicating a thin water vapor layer, and a large enhancement in water vapor can be obviously achieved at the height range of 4.2-4.8 km of the cloudy layer. The local ground temperature is 20.4 °C, and the ground relative humidity is about 30%, and the obtained ground water vapor mixing ratio is ∼ 4.12 g/kg from the ground meteorological station. The averaged value of water vapor mixing ratio in the range of 1-1.5 km is about 2.2 g/kg, which is to be slightly lower than that value from the ground meteorological station, and it is thought that the water vapor data is reasonable. The 1-σ statistical error is also presented as error-bar; the errors gradually increased with height, they are less than 0.5 g/kg under a height of 4 km and are estimated to be increased with values of > 1g/kg in the cloudy layer. The simultaneous liquid water mixing ratio profile, as shown in Fig. 7 (c), shows a consistent trend with that of water vapor; however, the liquid water content is relatively lower, with an average mean of 0.2 g/kg. Corresponding error bars are also shown and indicated the relatively larger errors in liquid water in the cloudy layer. Also, it should mention that large peaks of liquid water mixing ratio at ∼ 2.5 km and ∼ 3.2 km might be due to the aerosol fluorescence at the wavelength of 403 nm [21, 22]. The right figure presents the retrieved ice water mixing ratio profile, and there is no ice over the measurement range, which might mainly related with atmospheric temperature being above zero below the height of 3 km. Thus, our Raman lidar system has successfully characterized synchronous growth trends of atmospheric water vapor and liquid water in clouds.
Another representative measurement example was taken at 20:36 CST on March 20, 2019, under cloudy conditions. As shown in Fig. 8(a), a cloud layer is observed at heights of 3-3.5 km, and synchronous increase in the RSCS of water vapor, liquid water and ice are also achieved in the cloudy layer. From the channel SNRs curves in Fig. 8(b), the effective detection range can reach to a height of 3.5 km, although with part of cloud penetration in the liquid water channel. Figure 9 presents the retrieval results of extinction coefficient and three-phase water mixing ratios. The left figure shows the extinction coefficient and relative humidity profiles. Maximum extinction coefficient was valued 1.5 km-1 in the cloudy layer, and the corresponding relative humidity reached 100%. The retrieved water mixing ratio profiles are shown in Figs. 9(b)–9(d). Atmospheric water vapor mixing ratio is obtained with lower content with averaged value of 1 g/kg under the height of 3 km, and obvious enhancement is obtained to a value of 2-5 g/kg in the clouds. The liquid water content is relatively lower, with less than 0.05 g/kg, and obviously increased to ∼ 0.2 g/kg in clouds. It can be also seen that a large peak of liquid water mixing ratio is observed below the cloud between 2.5-3.0 km, and we suspect that this may be due to Raman scattering by water-laden aerosols, as suggested by Sakai, Reichardt and Veselovskii [19,21–22]. The ice water content in clouds is also captured with maximum mixing ratio of 0.12 g/kg. The measurement error bars are also shown with increasing trends with height in individual phased water mixing ratios, and the largest errors up to > 0.05 g/kg are obtained in ice mixing ratio due to the low channel SNR.
Figure 10 presents another representative measurement example taken at 21:06 CST on March 5, 2019 under clear weather conditions. The Range-square-corrected signals in each channel are shown in the left, and signal-to-noise ratio curves in three-phase water Raman channels are at the right figure. Under SNR of 10, the effective range can reach 3.5 km, 3 km and ∼3 km for water vapor, liquid water and ice Raman channels, respectively. Individual water mixing ratio profiles are retrieved and shown in Fig. 11, and extinction coefficient and relative humidity profiles are also shown. The water vapor mixing ratio decreases gradually with height, and increases to a value of 2.2 g/kg at the height range of 2.5-3.0 km, and the corresponding humidity is 60%. From the extinction coefficient profile, the extinction coefficient declines slowly with height, and no peak exists at 2.5-3.0 km. Combined with the shapes of RSCS of Mie-Rayleigh and water vapor, the extinction coefficient and relative humidity profiles, it can be inferred that a water layer exists at the height range of 2.5-3.0 km, not cloudy layer or aerosol layer. In a view of liquid water, synchronous enhancement is clearly obtained with maximum content of 0.15 g/kg in this water layer. The ice content has a relatively stable average value of < 0.01 g/kg below the height of 2 km, and obvious growth is also seen in the height range of 2.5-3 km, indicating rich ice content. We also can see a small increase in ice water below 2.0 km, and it is most likely caused by aerosol fluorescence, and this effect might possibly overestimate the derived mixing ratio values of ice water [19, 21–22]. The above examples clearly demonstrate that the Raman lidar system has successfully realized the synchronous measurements of atmospheric three-phase water in the water layers and cloudy layers at different height ranges.
5. Continuous observations
Continuous observations were also carried out from 20:45 to 04:15 on December 25, 2018 under mild and cloudy weather. The obtained time series of mixing ratio profiles are used to investigate the time-spatial evolution of three-phase water. Figure 12 presents three sets of measurement results of RSCS at different times during the whole period: 22:43, 23:54 and 01:34 CST. It can be clearly seen from the Mie-Rayleigh scattering signals that there exists a thin cloud layer at heights of 1.8-2.1 km. The effective detection range for individual water Raman channels reaches only 2 km, which is mainly affected by mildly hazy and cloudy weather conditions. The local ground temperature is 5 °C, the visibility is 5 km, and the local AQI is 99 with PM2.5 and PM10 concentrations of 75 µg/m3 and 145 µg/m3, respectively. Almost synchronous enhancement can be obtained from the RSCS of three-phase waters in the cloud layer. Regarding the ice Raman channel, enhanced returned signals can be successfully captured in clouds for even the closest detection distance. Thus, continuous observations are also regarded as examples to validate the performance of the Raman lidar under hazy and cloudy weather conditions.
Figure 13 presents a set of representative retrieval results of three-phase water mixing ratio profiles. As shown in the left Fig. 13(a), the aerosol extinction coefficient decreases gradually with increasing height, and obvious enhancement is observed in the height range of 1.8-2.1 km of the cloudy layer. And the simultaneous relative humidity profile is also shown with ∼100% in the cloudy layer. Figures 13(b)-13(d) correspond to profiles of the water vapor mixing ratio, liquid water mixing ratio and ice mixing ratio, respectively. They show a trend of stable change at heights of 0.6-1.8 km in water vapor, and the atmospheric water vapor content is averaged with a value of 1 g/kg, which is mainly affected by mildly hazy weather. Obvious water vapor enhancement can be obtained with a maximum value of 5 g/kg in the cloudy layer, indicating abundant water vapor content in clouds. A consistent trend is also found in liquid water and ice; evident enhancement of liquid water and ice content can be obtained at the cloud layer, and the corresponding mixing ratio reaches ∼0.3 g/kg and ∼0.05 g/kg, respectively. Furthermore, the ice mixing ratio is the lowest among those of the three phases of water, and its content is approximately 0.005 g/kg below the height of 1.8 km. The peaks of extinction and the mixing ratios at the height range of 1.8-2.0 km is likely to indicate the existence of mixed-phase cloud, however, the recognition of cloud phase state is required further validation with polarization technique. In the coming work, a combined Raman-depolarization lidar system is to be improved and developed for measurements of three-phase waters and depolarization, and another reference channel is also added for monitoring aerosol fluorescence. The results reflect minimal ice presence at the bottom of the troposphere and a significant increase in high-level clouds.
To intuitively understand the spatial-temporal evolution of the three-phase water content, a THI (time-height-indication) plot is shown in Fig. 14. The first subfigure Fig. 14(a) shows the temporal evolution of extinction coefficient profiles, and the drift and motion of the cloud layer of ∼ 2 km can be clearly reflected. With the passage of time, we have also gained the evolution of the three-phase water contents during the whole detection period, as shown in Figs. 14(b)–14(d). Synchronized growth in water vapor, liquid water and ice mixing ratio is achieved in the cloud layers. It should note that three-phase waters below the cloud layer can be detected effectively, because the Raman scattering signals attenuates sharply and the corresponding SNR decreases above the cloud.
We developed an ultraviolet Raman lidar system for synchronous measurements of water vapor, liquid water and ice. Considering the overlapping characteristics of the three-phase water Raman spectrum, the interference effect in the three-phase water Raman spectrum was simulated and analyzed, and the corresponding interference degree in Raman channels of different water phases was obtained. An accurate retrieval method based on linear simultaneous equations and mutual interference degrees was proposed for synchronous three-phase water mixing ratio profiles.
Preliminary measurements were carried out at Xi’an University of Technology and demonstrated the performance of the Raman system under different weather conditions. The effective detection can reach up to a height of 5 km under cloudy weather, and synchronous measurements can also successfully achieve in clouds under hazy and cloudy conditions.
Combined with the extinction coefficient and relative humidity profiles, the synchronous water vapor, liquid water and ice profiles were retrieved to reveal the variation characteristics in three-phase waters. Synchronized growth in water vapor, liquid water and ice content was obtained in water layers and cloud layers. The results demonstrated the feasibility of Raman lidar for synchronous measurements of three-phase water.
In the future, there are still several key technologies to be further developed. The system performance of Raman lidar is to be improved for cloud measurements at high level. It is required to make comparisons with other instruments, especially, for the liquid water mixing ratio and the intergrade liquid water. Also, the error estimation for three-phase waters is also needed. In addition, polarization characteristics and fluorescence measurements are required to combine with three-phase water content for cloud phase research.
National Natural Science Foundation of China (NSFC) (U1733202, 41627807, 41575027, 61308105).
The authors wish to thank Prof. Mao Jietai of Peking University help to assistance in Raman spectroscopy theory, and also to thank postgraduate Zhang Jing help with the spectral measurement of three-phase waters.
1. H.R. Pruppacher and J.D. Klett, “Microphysics of clouds and precipitation, Second revised and enlarged edition with an introduction to cloud chemistry and cloud electricity,” Kluwer Academic publisher, Dordrecht, pp954 (1997)
2. K. Schneider E, P. Kirtman B, and S. Lindzen R, “Tropospheric Water Vapor and Climate Sensitivity,” J. Atmos. Sci. 56(11), 1649–1658 (1999). [CrossRef]
3. F. Russo, N. Whiteman D, B. Demoz, I. Veselovskii, H. Melfi S, and M. Hoff R, “Development of Raman LIDAR Techniques to Address the Indirect Aerosol Effect: Retrieving the Liquid Water Content of Clouds,” Eur. Space Agency. 1(561), 411–414 (2004).
4. I. Balin, I. Serikov, S. Bobrovnikov, V. Bimeonov, B. Calpini, Y. Arshinov, and H. Van den bergh, “Simultaneous measurement of atmospheric temperature, humidity, and aerosol extinction and backscatter coefﬁcients by a combined vibrational-pure-rotational Raman lidar,” Appl. Phys. B: Lasers Opt. 79(6), 775–782 (2004). [CrossRef]
5. A. Behrendt, T. Nakamura, M. Onishi, R. Baumgart, and T. Tsuda, “Combined Raman lidar for the measurement of atmospheric temperature, water vapor, particle extinction coefficient, and particle backscatter coefficient,” Atmos. Chem. Phys. 41(36), 7657–7666 (2002). [CrossRef]
6. E. Hammann, A. Behrend, F. Le Mounter, and V. Wulfmeyer, “Temperature profiling of the atmospheric boundary layer with rotational Raman lidar during the HD(CP)2 observational Prototype experiment,” Atoms.Chem.Phys. 15(5), 2867–2881 (2015). [CrossRef]
7. J. Nott G, J. Duck T, G. Doyle J, E. W. Coffin M, C. Perro, P. Thackray C, R. Drummond J, F. Fogal P, E. McCullough, and J. Sica R, “A Remotely Operated Lidar for Aerosol, Temperature, and Water Vapor Profiling in the High Arctic,” J. Atmos. Oceanic Technol. 29(2), 221–234 (2012). [CrossRef]
8. I. Mattis, A. Ansmann, D. Althausen, V. Jaenisch, U. Wandinger, D. Müller, Y. F. Arshinov, S. M. Bobrovnikov, and I. B. Serikov, “Relative-humidity profiling in the troposphere with a Raman lidar,” Appl. Opt. 41(30), 6451–6462 (2002). [CrossRef]
9. J. Reichardt, U. Wandinger, V. Klein, 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]
10. T. Sakai, T. Nagai, M. Nakazato, T. Matsumura, N. Orikasa, and Y. Shoji, “Comparisons of Raman Lidar Measurements of Tropospheric Water Vapor Profiles with Radiosondes, Hygrometers on the Meteorological Observation Tower, and GPS at Tsukuba Japan,” J. Atmos. Oceanic Technol. 24(8), 1407–1423 (2007). [CrossRef]
11. T. Leblanc, I. S. McDermid, and T. D. Walsh, “Ground-based water vapor raman lidar measurements up to the upper troposphere and lower stratosphere for long-term monitoring,” Atmos. Meas. Tech. 5(1), 17–36 (2012). [CrossRef]
12. P. Bobylev L, V. Zabolotskikh E, M. Mitnik L, and L. Mitnik M, “Atmospheric Water Vapor and Cloud Liquid Water Retrieval Over the Arctic Ocean Using Satellite Passive Microwave Sensing,” IEEE Trans. Geosci. Remote Sensing. 48(1), 283–294 (2010). [CrossRef]
13. L. Morrison A, E. Kay J, H. Chepfer, R. Guzman, and V. Yettella, “Isolating the Liquid Cloud Response to Recent Arctic Sea Ice Variability Using Spaceborne Lidar Observations,” J. Geophys. Res: Atmos. 123(1), 473–490 (2018). [CrossRef]
14. N. Whiteman D and H. Melfi S, “Cloud liquid water, mean droplet radius, and number density measurements using a Raman lidar,” J. Geophys. Res. 104(D24), 31411–31419 (1999). [CrossRef]
15. D. Spinhirne J, R. Boers, and D. Hart W, “Cloud Top Liquid Water from Lidar Observations of Marine Stratocumulus,” J. Appl. Meteorol. 28(2), 81–90 (1989). [CrossRef]
16. I. A. Veselovskii, H. K. Cha, D. H. Kim, S. C. Choi, and J. M. Lee, “Study of atmospheric water in gaseous and liquid state by using combined elastic-Raman depolarization lidar,” Appl. Phys. B: Lasers Opt. 73(7), 739–744 (2001). [CrossRef]
17. Z. Wang, N. Whiteman D, B. Demoz B, and I. Veselovskii, “A new way to measure cirrus cloud ice water content by using ice Raman scatter with Raman lidar,” Geophys. Res. Lett. 31(311), 121–141 (2004). [CrossRef]
18. F. Liu and F. Yi, “Spectrally resolved Raman lidar measurements of gaseous and liquid water in the atmosphere,” Appl. Opt. 52(28), 6884–6895 (2013). [CrossRef]
19. T. Sakai, N. Whiteman D, F. Russo, D. Turner D, I. Veselovskii, H. Melfi S, T. Nagai, and Y. Mano, “Liquid Water Cloud Measurements Using the Raman Lidar Technique: Current Understanding and Future Research Needs,” J. Atmos. Oceanic Technol. 30(7), 1337–1353 (2013). [CrossRef]
20. R. Vincenzo, I. Marcro, R. Giuseppe, and V. Guido, “Raman lidar observations of cloud liquid water,” Appl. Opt. 43(35), 6440–6453 (2004). [CrossRef]
21. I. A. Veselovskii, H. K. Cha, D. H. Kim, S. C. Choi, and J. M. Lee, “Raman lidar for the study of liquid water and water vapor in the troposphere,” Appl. Phys. B: Lasers Opt. 71(1), 113–117 (2000). [CrossRef]
22. J. Reichardt, “Cloud and Aerosol Spectroscopy with Raman Lidar,” J. Atmos. Oceanic Technol 31(9), 1946–1963 (2014). [CrossRef]
23. D. N. Whiteman, “Examination of the traditional Raman lidar technique. II. Evaluating the ratios for water vapor and aerosol,” Appl. Opt. 42(15), 2593–2608 (2003). [CrossRef]
24. A Ansmann and D Müller, “Lidar and Atmospheric Aerosol Particles,” Springer Series in Optical Sciences, Springer, New York, 102, pp 105–141 (2005)
25. W. Yufeng, F. Qiang, Z. Meina, D. H. Gao Fei, S. Yuehui, and H. Dengxin, “A UV multifunctional Raman lidar system for the observation and analysis of atmospheric temperature, humidity, aerosols and their conveying characteristics over Xi'an,” J. Quant. Spectrosc. Radiat. Transfer 205, 114–126 (2018). [CrossRef]
26. Di Huige, Hua Dengxin, Wang Yufeng, and Yan Qing, “Investigation on the correction of the Mie scattering lidar’s overlapping factor and echo signals over the total detection range,” Acta physica sinica 62(9), 094215 (2013). [CrossRef]
27. T. Plakhotnik and J. Reichardt, “Accurate absolute measurements of the Raman backscattering differential cross-section of water and ice and its dependence on the temperature and extinction wavelength,” J. Quant. Spectrosc. Radiat. Transfer 194, 58–64 (2017). [CrossRef]
28. D. Kim, I. Song, H.-D. Cheong, Y. Kim, S. Baik, and J. Lee, “Spectrum Characteristics of Multichannel Water Raman Lidar Signals and Principal Component Analysis,” OPT REV 17(2), 84–89 (2010). [CrossRef]