Raman Lidar for the Measurement of Temperature, Water Vapor, and Aerosol in Beijing in the Winter of 2014

  • cc icon
  • ABSTRACT

    To measure atmospheric temperature, water vapor, and aerosol simultaneously, an efficient multi-function Raman lidar using an ultraviolet-wavelength laser has been developed. A high-performance spectroscopic box that utilizes multicavity interference filters, mounted sequentially at small angles of incidence, is used to separate the lidar return signals at different wavelengths, and to extract the signals with high efficiency. The external experiments are carried out for simultaneous detection of atmospheric temperature, water vapor, and aerosol extinction coefficient in Beijing, under clear and hazy weather conditions. The vertical profiles of temperature, water vapor, and aerosol extinction coefficient are analyzed. The results show that for an integration time of 5 min and laser energy of 200 mJ, the mean deviation between measurements obtained by lidar and radiosonde is small, and the overall trend is similar. The statistical temperature error for nighttime is below 1 K up to a height of 6.2 km under clear weather conditions, and up to a height of 2.5 km under slightly hazy weather conditions, with 5 min of observation time. An effective range for simultaneous detection of temperature and water vapor of up to 10 km is achieved. The temperature-inversion layer is found in the low troposphere. Continuous observations verify the reliability of Raman lidar to achieve real-time measurement of atmospheric parameters in the troposphere.


  • KEYWORD

    Lidar , Rotational Raman , Temperature , Water vapor , Aerosol

  • I. INTRODUCTION

    Temperature, water vapor, and aerosols are three important atmospheric meteorological parameters. Continuous observation of these parameters is fundamental to improving our understanding of weather and climate change, particularly of the radiation and heat balance of the earth’s atmosphere, and of atmospheric chemistry. The accurate observation of diurnal variation of these meteorological parameters is useful for environment assessment to understand and foretell weather changes [1]. Cloud formation and the hydrological cycle are required for understanding water-vapor concentration and motion. Aerosol particles can have a great effect on the balance of Earth’s atmospheric radiation system, by absorbing and scattering radiation [2, 3]. Human activities produce many particles and secondary aerosols that lead to significantly reduced visibility and additional occurrence of haze [4-6]. Above all, small particles (PM2.5) suspended in the air will lead to air pollution problems and undermine human health [7].

    The government introduced a series of temporary policies, such as limiting cars on the streets according to their license-plate numbers, restricting dust, and forbidding emission-heavy production. Beijing and its neighboring regions, includeing Hebei, Tianjin, Shandong, and Inner Mongolia adopted these measures to guarantee air quality according to 2014 Asia-Pacific Economic Cooperation (APEC) Summit. The measurement experiments were carried out in the atmospheric observing site of the University of Chinese Academy of Sciences (UCAS) at Yanqi Lake (40.41°N, 116.68°E), Huairou District, Beijing from November to December. They can help us to comprehend the evolution of air pollution, and evaluate the influence of air-pollution control measures [8, 9].

    Lidar is a mature technique for measuring temperature, water vapor, and aerosols [1, 10, 11]. Here we introduce the latest phase of a system that is designed with the implementation of vibrational Raman channels for water-vapor measurements, purely rotational Raman channels for temperature measurements, and elastic channels for aerosol measurements. The lidar designed by the Anhui Institute of Optics and Fine Mechanics, Chinese Academy of Sciences, is carried out in the suburbs of Beijing during winter, and the observed results are presented. Many previous experiments in Beijing failed to simultaneously detect temperature, water vapor, and aerosol. On the other hand, their observation periods tended to be short [12-18]. Therefore, this experiment provides good data support for Beijing’s atmospheric-parameter measurements.

    II. MEASUREMENT PRINCIPLE

    We can use rotational Raman backscattering signals to measure atmospheric- temperature profile. The intensity distribution of the rotational Raman spectrum contains information on the atmospheric temperature. The intensities Raman signals of low and high quantum numbers have opposite temperature dependences. For calculating the atmospheric- temperature profile, we often use the ratio

    image
    image

    Q(T,z) is the ratio of these two rotational Raman signals, and Sjlow(T,z) and Sjhigh(T,z) are the two rotational Raman signals of opposite temperature dependences. A, B, and C are calibration coefficients that could be determined by fitting, and z is the height. The water-vapor mixing ratio (WVMR) at a height of z can be calculated with

    image

    where Cw is a calibration constant, PWV and PRR are respectively the background-subtracted lidar signal for water vapor, and the rotational Raman lines of N2 and O2 [19]. Tr(Z) is a differential transmission factor for the different values of atmospheric extinction Tr at the two wavelengths λRR and λH2O. The backscattering at the laser wavelength is caused by Rayleigh scattering from air molecules and Mie scattering from aerosol particles. We obtain the aerosol’s extinction coefficient at the transmitted wavelength:

    image

    where Nα (z) is the molecular number density, S(z) is background-corrected signal, αm can either be determined from the best available meteorological data for temperature and pressure or approximated from appropriate standard atmospheres, λ0 is the laser wavelength, λN is the N2 vibration-rotational Raman wavelength, and k is a constant equal to (1).

    III. INSTRUMENTAL SETUP

    The schematic overview of the Temperature Water Vapor and Aerosols Raman Lidar (TWAR) is shown in Fig. 1. The technical specifications of the system are listed in Table 1. The TWAR is mounted in a mobile container, which can be easily moved to different observation sites by a truck.

    The fundamental radiation at 1064 nm, the second-harmonic radiation at 532 nm, and the third-harmonic radiation at 354.67 nm are emitted into the atmosphere simultaneously. The emissions at 1064 nm and 532 nm are transmitted to the atmosphere through a small window, and the return signals received by a telescope. Using a Cassegrain telescope with a diameter of 200 mm and a focal length of 4 m (labeled A in Fig. 1), and in the after-optics receiver box, there are three detection channels to detect the elastic signal at 1064 nm and parallel and perpendicular polarizations at 532 nm, which are separated with a polarizer. These signals are used to analyze the distributions and optical characteristics of aerosols and clouds. The third-harmonic radiation at 354.67 nm is separated from the radiation at 532 nm and 1064 nm and transmitted into the atmosphere through the big window after a beam expander, which can decrease both beam divergence and energy density. The beam expander can narrow the divergence of the laser beam to 0.1 mrad, for eye safety, and to reduce the solar background noise. The pulse energy of the laser beam in the ultraviolet is approximately 200 mJ, with a 20-Hz pulse-repetition frequency.

    The return signals are received by a Cassegrain telescope with a primary mirror diameter of 400 mm and a focal length of 4 m (labeled B in Fig. 1). After collimation with a lens, the incoming light is divided by a beam splitter (BS1), to separate the vibrational Raman signals from the rotational Raman signal and the elastic signal. Wavelengths longer than 370 nm are transmitted efficiently, while shorter wavelengths are reflected.

    The transmission efficiency of the BS1 is 0.95 for longer wavelengths and 0.01 for shorter. The wavelengths of the vibrational-rotational Raman signals for water vapor and N2 excited by 354.67 nm are about 407 nm and 386 nm respectively. They are separated by BS2 and then pass through the interference filters (IFs) IF4 and IF5, finally to be focused on the photomultiplier tubes PMT4 and PMT5, respectively. The suppression magnitude of IF4 and IF5 is about 10-12 at 355 nm, and 10-6 at other wavelengths. The rotational Raman signal and elastic signal pass the daylight-reducing IF0 with a transmission band of 8 nm, and are separated on IF1. The elastic signal that will be used to analyze aerosol is focused on photomultiplier tube PMT1, and the rotational Raman signal is almost totally reflected. The low-quantum-number rotational Raman signal is transmitted by IF2a and IF2b and focused on photomultiplier tube PMT2. The high-quantum-number rotational Raman signal is reflected by IF2a, transmitted by IF3, and focused on photomultiplier tube PMT3. Because the transmission band of IF2a verges on the laser’s wavelength, the low-quantum-number rotational Raman channel needs two filters to achieve sufficient suppression of the elastic signal. Filters IF1-IF3 are mounted at small angles of incidence, 5.7-6.5°, so that we can select the central wavelengths of the extracted rotational Raman signals by changing these angles, to confirm the most suitable filter parameters [20, 21]. Filter parameters are also shown in Table 1.

    IV. MEASUREMENT EXAMPLES

       4.1. Temperature Measurement

    The atmospheric measurements with the new instrument were taken in November and December 2014 at Yanqi Lake (40.41°N, 116.68°E). Typical measurements are shown in Figs. 2 and 3. The lidar signals were obtained with a temporal resolution of 5 min and height resolution of 7.5 m, corresponding to 5000 laser shots. Figure 2 shows the lidar temperature profiles and those for a local radiosonde that was used to calibrate the lidar constants, plus the profiles of statistical temperature uncertainty. The statistical temperature uncertainty can be calculated as

    image

    where PRR1 is the background-corrected signal in the first Raman channel, PB1 is the background in this channel, PRR2 and PB2 are the same for the second Raman channel, and Q is the ratio of these two rotational Raman signals. As shown in the figure, the temperature profiles measured by lidar and radiosonde have the same trend, up to a height of 10 km. Atmospheric temperatures decreased, with a gradient of approximately 6 K/km. However, below 500 m the lidar profiles are not very accurate, on account of different overlap of the two rotational Raman channels in the near field.

    The lidar signals are smoothed with a gliding average of 120 m in length up to 3 km, and 600 m above 3 km. The response function Q(T) = exp(a / T2+b / T+c) with fitting coefficients a = 1.35 × 105, b = −4.7 × 103, and c = 24.71 is used. The profile of atmospheric temperature is measured for a clear atmosphere at 20:30 on 8 November 2014 in (a), and for mildly hazy weather on 13 November 2014 in (b). In the clear-weather measurement shown in Fig. 2(a), the lidar profile is consistent with that for the radiosonde, below 10 km. A statistical error of less than 1 K is obtained, up to 6.2 km in height, while for the measurement obtained in mild haze shown in Fig. 2(b), the error of less than 1 K is obtained only up to 2.5 km in height. Furthermore, a cloud appears at about 9 km in height. Intense Mie backscattering by the cloud is detected in the rotational Raman channel, leading to large statistical uncertainty in temperature above the cloud, but the statistical temperature uncertainty does not exceed 3 K up to 6 km in height. It can be seen that the lidar profile matches the radiosonde profile very closely, below the cloud.

    Figure 3 shows consecutive temperature profiles measured with TWAR lidar between 20:00 and 22:00 on 8 November 2014. The lidar data were acquired with a temporal resolution of 5 minutes. To observe the spatio-temporal transformation of temperature, consecutive profiles are shifted by about 13 K in turn. As we can see from the picture, the atmospheric temperature distribution is relatively stable over two hours. Temperature from 2 to 3 km in altitude tends to increase gradually over the observation period, probably because the near-surface hot air gradually moves upward, due to human activity and volatilization of solar radiation. Though the fluctuations seen in the lidar temperature data are a little high due to the statistical uncertainty of the data, many temperature perturbations are seen by TWAR lidar.

       4.2. Water-Vapor Mixing Ratio Measurement

    The measurement example presented in Fig. 4 was from 28 July 2014, at the site of the Anhui Institute of Optics and Fine Mechanics, Chinese Academy of Sciences, Hefei. The data were acquired between 19:30 and 20:00, with radiosonde being close in time and space to the lidar measurement. The radiosonde used to calibrate the lidar was launched at a distance of 100 meters from the lidar-observation location. Figure 4(a) compares the water-vapor profiles measured with lidar and concurrent radiosonde. Under cloud-free conditions at night, the profile’s top height is 10 km. Figures. 4(b) and 4(c) show the deviation and relative error between the lidar and radiosonde profiles. In this case the relative error mostly does not exceed 5% up to 4 km, and is well below 20% up to 7.5 km. However, below 400 m the lidar data deviation is relatively large, due to the different overlap of the nitrogen vibrational Raman channel and water vibrational Raman channel in the near field. We can use the photon counting (PC) signals in the far range, where the signal-to-noise ratio of the analog data (A/D) is low, and the A/D signals in the near range, where PC signals are saturated. To get a single profile, we need to merge the A/D and PC signals that are formed by using the A/D signals up to 2 km and the PC signals above.

    Figure 5 compares the water-vapor profiles measured with lidar and radiosonde on 9 and 10 November 2014 at the University of Chinese Academy of Sciences (UCAS) at Yanqi Lake (40.41°N, 116.68°E), Huairou District, in Beijing. For each comparison, 10 min of lidar raw data were integrated. Blue bars indicate the statistical error in the lidar measurements. The resolution of the raw data is 7.5 m, and signal profiles are smoothed with a gliding average of 150 m. These values can be seen as typical of this lidar system’s performance when it has been moved to Beijing. It can be seen from the picture that the results measured in Beijing are not as good as the data measured in Hefei, because the radiosonde’s release location is about 15 km away from the observation location. Also, we can find that winter in Beijing is very dry, and the water-vapor mixing ratio is generally below 2 g/kg.

    Figure 6 shows the water-vapor mixing ratio during a three-day measurement period between 11 November 2014 at 00:00 UTC and 13 November 2014 at 23:45 UTC, which is approximately 72 continuous hours of lidar data. The water-vapor concentration is appreciable below about 5 km, and the water-vapor mixing ratio is approximately 1.5 g/kg during the continuous observation. The effective height of the measurement is about 8 km at night, whereas in the daytime only the water-vapor content in the boundary layer can be measured, and the height of the measurement is limited to about 2.5 km, due to the strong background noise. During these observation days the weather in Beijing was very dry, and the maximum water-vapor mixing ratio was less than 2 g/kg.

       4.3. Aerosol Measurement

    This measurement example was obtained between 19:30 and 20:00 on 28 July 2014. The profiles for range-corrected signal, retrieved extinction coefficient, water-vapor mixing ratio, and temperature shown in Fig. 7 clearly indicate the atmospheric boundary layer. The retrieved water-vapor mixing ratio and temperature profiles correspond well to those for the radiosonde, and reveal the boundary layer at heights of 1.5-1.8 km where a temperature inversion of 2 K is observed. From the range-corrected signal and retrieved extinction coefficient, we can see from the 532-nm and 1064-nm Mie-Rayleigh scattering signals that an obvious small peak appears at heights of 4.5-5 km, which shows the sign of a thin cloud layer. The extinction-coefficient profiles indicate that it increases, with a maximum value of 0.08 km-1 (532 nm) in the cloud layer. However, no signal enhancement from the two rotational Raman signals was observed, indicating the high level of rejection of the rotational Raman compared to the elastic scattering. This example proves that our lidar system has the ability to identify the atmospheric boundary layer in the lower troposphere.

    The vertical distribution of the aerosol backscattering coefficient and the total aerosol relative error measurements at 355 nm, 532 nm, and 1064 nm are given in Fig. 8 (black curves). In addition, the vertical distributions of the backscattering coefficient of air molecules at these three wavelengths are also given (dashes).

    image

    where zc is the reference point’s height, X(z) is the range-corrected signal, S1 and S2 indicate the lidar ratio of aerosol and atmospheric molecules respectively, and β1(z) and β2(z) indicate the backscattering coefficients of aerosol and atmospheric molecules at a height of z, respectively. In Eq. (6) [22], there are four measurement errors that will be transmitted to the atmospheric aerosol backscattering coefficient: β1(z), β2(z), S1, and X(z). In the formula for the aerosol backscattering coefficient, it is difficult to calculate the partial derivative of the aerosol backscattering coefficient for direct measuring of variables. To solve this problem, we use a direct error formula whose reasonability and reliability are tested by comparison calculation [23]. Based on experience, the relative error in the backscattering coefficient of atmospheric molecules obtained by the atmospheric model is less than one percent. The lidar ratio at 532 nm is generally between 30 and 70 sr (355 nm: 10-30 sr, 1064 nm: 30-60 sr). The relative error in the aerosol backscattering coefficient is nearly one hundred percent, which is selected as the reference point in the clean layer near the top of the troposphere. The relative error in the lidar return signal is less than two percent [23]. According to the sizes of errors in the above variables, we can estimate the total relative error by using the Eq. (6). As can be seen from the figure, the height of the boundary layer is about 2.4 km at night. Under the boundary layer, the relative error of the aerosol backscattering coefficient at 355 nm is less than 40%, and 10% at 1064 nm. The relative error at 532 nm is nearly 20%. In the analysis of the relative error in the aerosol backscattering coefficient, we found that the main source of relative error in the measurements at 355 nm and 532 nm is the lidar ratio, while the error in the reference-point value is the main source of the relative error at 1064 nm.

    V. CONCLUSION

    We have developed a multifunction lidar for simultaneous high-resolution measurement of atmospheric temperature profile, water vapor, and aerosol. With seven lidar signals [elastic signals at 1064nm, 532 nm (parallel and perpendicular) and 355 nm, two rotational Raman signals, and the vibrational-rotational Raman signals of water vapor and N2], we can determine the temperature profile, water-vapor mixing ratio profile, aerosol extinction coefficient, color ratio at the two wavelengths, and depolarization ratio at 532 nm. These results indicate that our new instrument has the ability for continuous and automatic observation.

    The nighttime statistical temperature error is less than 1 K up to 6.2 km in height in the clear-atmosphere measurement, and up to 2.5 km in height in slightly haze weather, for an integration time of 5 min. Our lidar system has still some room for improvement, such as the performance of the temperature measurements near the ground, the ability to detect water vapor and temperature during the day, and enhancing the instrument’s signal-to-noise ratio. These results indicate that this efficient, mobile, and compact Raman lidar system can be used as a practical and continuously operating tool for measuring atmospheric parameters, which can help us understand many meteorological processes and phenomena.

  • 1. Behrendt A., Nakamura T., Onishi M., Baumgart R., Tsuda T. 2002 Combined Raman lidar for the measurement of atmospheric temperature, water vapor, particle extinction coefficient, and particle backscatter coefficient [Appl. Opt.] Vol.41 P.7657-7666 google doi
  • 2. Bardouki H., Liakakou H., Economou C., Sciare J., Smolik J., ?dimal V., Eleftheriadis K., Lazaridis M., Dye C., Mihalopoulos N. 2003 Chemical composition of size-resolved atmospheric aerosols in the eastern Mediterranean during summer and winter [Atmos. Environ.] Vol.37 P.195-208 google doi
  • 3. Ramanathan V., Feng Y. 2009 Air pollution, greenhouse gases and climate change: Global and regional perspectives [Atmos. Environ.] Vol.43 P.37-50 google doi
  • 4. Harrison R. G. 2012 Aerosol-induced correlation between visibility and atmospheric electricity [J. Aerosol Sci.] Vol.52 P.121-126 google doi
  • 5. Pui D. Y. H., Chen S., Zuo Z. 2014 PM2.5 in China: Measurements, sources, visibility and health effects, and mitigation [Particuology] Vol.13 P.1-26 google doi
  • 6. Xiao S., Wang Q. Y., Cao J. J., Huang R., Chen W. D., Han Y. M., Xu H. M., Liu S. X., Zhou Y. Q., Wang P., Zhang J. Q., Zhan C. L. 2014 Long-term trends in visibility and impacts of aerosol composition on visibility impairment in Baoji, China [Atmos. Res.] Vol.149 P.88-95 google doi
  • 7. O. G 2001 Pulmonary effects of inhaled ultrafine particles [Int. Arch. Occup. Environ. Health] Vol.74 P.1-8 google
  • 8. Chen Z., Zhang J., Zhang T., Liu W., Liu J. 2015 Haze observations by simultaneous lidar and WPS in Beijing before and during APEC, 2014 [Sci. China Chem.] Vol.58 P.1385-1392 google doi
  • 9. Zifa W., Jie L. I., Zhe W., Wenyi Y., Xiao T., Baozhu G. E., Pinzhong Y. A. N., Lili Z. H. U., Xueshun C., Huansheng C., Wei W., Jianjun L. I., Bing L. I. U., Xiaoyan W., Wei W., Yilin Z., Ning L. U., Debin S. U. 2014 Modeling study of regional severe hazes over mid-eastern China in January 2013 and its implications on pollution prevention and control [Sci. China Earth Sci.] Vol.57 P.3-13 google doi
  • 10. Radlach M., Behrendt A., Wulfmeyer V. 2008 Scanning rotational Raman lidar at 355 nm for the measurement of tropospheric temperature fields Scanning rotational Raman lidar at 355 nm for the measurement of tropospheric temperature fields [Atmos. Chem. Phys.] Vol.8 P.159-169 google doi
  • 11. Hammann E., Behrendt A., Le Mounier F., Wulfmeyer V. 2015 Temperature profiling of the atmospheric boundary layer with rotational Raman lidar during the HD (CP) 2 [Atmos. Chem. Phys.] Vol.15 P.2867-2881 google doi
  • 12. Yunlei L., Hua L., Yubao C., Yuchun G. 2014 Aerosol detection experiment with 532nm lidar based on Fernald method [Electron. Des. Engimeering] Vol.22 P.88-90 google
  • 13. Zhanshan W., Yunting L., Qian L., Lihua W., Baoxian L. 2017 Analysis on a Dust Pollution Event in Beijing in May, 2017 Based on the Observation of an Atmospheric Supersite [Environ. Monit. China] Vol.33 P.28-34 google
  • 14. Wei W., Nan Y., Yaolong S., Zhen Z., Linjun C., Wenxuan C., Baobin C., Qiang F., Jianjun L. 2017 Application of Mobile Lidar in Analying Regional Pollutants Transportation During a Haze Episode over Beijing-Tianjin-Hebei Aera [Environ. Monit. China] Vol.33 P.7-13 google
  • 15. Chuanyao D., Liping Y., Mian W., Jingjin M., Dong L., Chunbo Z., Lei M., Lu W. 2015 Comprehensive detection of fog and haze process [Meteorol. Mon.] Vol.41 P.1525-1530 google
  • 16. Chen Y., An J.-L., Lin J., Sun Y.-L., Wang X.-Q., Wang Z.-F., Duan J. 2017 Observation of nocturnal low-level wind shear and particulate matter in urban Beijing using a Doppler wind lidar [Atmos. Ocean. Sci. Lett.] Vol.10 P.411-417 google doi
  • 17. Tiemin Z., Jihong W., Linmao W., Xueming C., Jianqing W., Xu Z., Hongyan P. 2017 Observations of sodium layer over Beijing and Haikou in july 2012 [Chin. J. Sp. Sci] Vol.37 P.424-431 google
  • 18. Zhaobin S., Xiaonong L., Zhanshan W., Ziming L., Xiujuan Z., Cong H. 2016 Scavenging effect of rime and east wind on PM2.5 under air heavy pollution in Beijing [Environ. Sci.] Vol.37 P.4-10 google
  • 19. Whiteman D. N., Melfi S. H., Ferrare R. A. 1992 Raman lidar system for the measurement of water vapor and aerosols in the Earth’s atmosphere [Appl. Opt.] Vol.31 P.3068-3082 google doi
  • 20. Weitkamp C. 2006 Lidar: Range-Resolved Optical Remote Sensing of the Atmosphere Vol.102 google
  • 21. Behrendt A., Nakamura T. 2002 Calculation of the calibration constant of polarization lidar and its dependency on atmospheric temperature [Opt. Express] Vol.6 P.6587-6595 google
  • 22. Bo C. T. W. D. L., Yuan C. K. W. Z. G., Jun L. Z. 2010 A new method for determining aerosol baekscatter coefficient boundary value in the lower troposphere [Acta Opt. Sin.] Vol.6 P.3 google
  • 23. Tao Z., Wu D., Liu D., Hu S., Nie M., Shi B. 2011 Estimation of aerosol backscatter coefficient error in lidar data processing [Zhongguo Jiguang (Chinese J. Lasers)] Vol.38 P.1214001-1214005 google
  • [] 
  • [] 
  • [] 
  • [] 
  • [FIG. 1.] Schematic diagram and photograph of the developed TWAR lidar.
    Schematic diagram and photograph of the developed TWAR lidar.
  • [TABLE 1.] Technical data
    Technical data
  • [] 
  • [FIG. 2.] Lidar and radiosonde temperature profiles measured at Yanqi Lake (40.41°N, 116.68°E) on (a) 8 November 2014 and (b) 13 November 2014. Error profiles show the statistical temperature uncertainty.
    Lidar and radiosonde temperature profiles measured at Yanqi Lake (40.41°N, 116.68°E) on (a) 8 November 2014 and (b) 13 November 2014. Error profiles show the statistical temperature uncertainty.
  • [FIG. 3.] Consecutive temperature profiles measured with TWAR lidar.
    Consecutive temperature profiles measured with TWAR lidar.
  • [FIG. 4.] (a) Profiles of water-vapor mixing ratio, measured with lidar (circles) and radiosondes launched at night (black curves) on 28 July 2014. 900 s of lidar data were integrated, beginning at the radiosonde’s launch. Lidar signals were smoothed with a gliding average of 150 m. (b) Difference between lidar and local radiosonde measurements. (c) Relative error profile.
    (a) Profiles of water-vapor mixing ratio, measured with lidar (circles) and radiosondes launched at night (black curves) on 28 July 2014. 900 s of lidar data were integrated, beginning at the radiosonde’s launch. Lidar signals were smoothed with a gliding average of 150 m. (b) Difference between lidar and local radiosonde measurements. (c) Relative error profile.
  • [FIG. 5.] Profiles of water-vapor mixing ratio, measured with lidar (black curves) and radiosonde (black dashs) for 9 and 10 November 2014.
    Profiles of water-vapor mixing ratio, measured with lidar (black curves) and radiosonde (black dashs) for 9 and 10 November 2014.
  • [FIG. 6.] Continuous observation of water-vapor mixing ratio measured by lidar, between 11 November and 13 November.
    Continuous observation of water-vapor mixing ratio measured by lidar, between 11 November and 13 November.
  • [FIG. 7.] Profiles of range-corrected signal, aerosol extinction, water vapor mixing ratio and temperature taken at 19:30 and 20:00 on 28 July 2014.
    Profiles of range-corrected signal, aerosol extinction, water vapor mixing ratio and temperature taken at 19:30 and 20:00 on 28 July 2014.
  • [FIG. 8.] Profiles for aerosol backscattering coefficient and total aerosol relative error taken at 02:00 on 11 November 2014.
    Profiles for aerosol backscattering coefficient and total aerosol relative error taken at 02:00 on 11 November 2014.
  • []