Open Access
Add to BibSonomy
Abstract
Conventional lidar methods for deriving particle optical properties suffer from the fact that two unknowns (backscatter and extinction coefficients) need to be determined from only one lidar equation. Thus, additional assumptions (constant lidar ratio or Ångström relationship) have to be introduced to settle this problem. In contrast, a single-line-extracted pure-rotational-Raman (PRR) lidar method allows the strict retrieval of backscatter and extinction coefficients without additional assumptions. Based on the observations of our single-line-extracted PRR lidar from February 2016 to December 2017, the optical properties (backscatter coefficient, extinction coefficient and lidar ratio) of continental polluted aerosols, dust aerosols, and cirrus cloud particles over Wuhan (30.5°N, 114.4°E) are well characterized. The mean values of the measured lidar ratios are respectively 60 ± 7 sr for continental polluted aerosols, 47 ± 4 sr for dust aerosols and 22 ± 4 sr for cirrus cloud particles. The backscatter and extinction coefficients measured by the single-line-extracted PRR lidar deviate as a whole by 7-13% and 13-16%, respectively, from those retrieved by the traditional Fernald method. The optical properties measured by the single-line-extracted PRR lidar can serve as observational standards for particle optical properties (backscatter/extinction coefficient and lidar ratio) at 532 nm wavelength.
© 2021 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
1. Introduction
Atmospheric aerosols influence air quality and chemistry in the troposphere and play an important role in controlling climate through direct radiative forcing (scattering and absorbing solar and thermal infrared radiation) and indirect radiative forcing (modifying the microphysical and hence the radiative properties and lifetime of clouds) [1]. The corresponding radiative forcing varies with particle types and vertical distributions [1,2]. Cirrus clouds are composed primarily of ice crystals that form in the cold upper troposphere with a variety of forms and shapes [3]. Studies show that cirrus clouds play a significant role in the Earth-Atmosphere system radiation budget for the widely coverage (30%) of the Earth’s surface [4]. The shapes, vertical and horizontal coverage of ice crystals significantly impact cirrus radiative properties and feedbacks to climate [5]. Moreover, cirrus clouds are one of the most uncertain components of general circulation models [6]. Considering that the optical properties (especially lidar ratios) are closely linked to particle sizes and shapes, measurements to obtain the temporally-varying vertical structures of particle optical parameters will lead to a better representation in modeling particle current and future effects on climate [7,8].
Lidar serves as an effective tool for remote sensing of atmospheric particles with high temporal-spatial resolution. Modern lidar techniques about particle measurement can be divided into three categories. One is the elastic backscatter lidar. Solutions to lidar equation for obtaining aerosol optical properties were first given by Fernald (1984) via assuming a constant lidar ratio [9]. Sasano and Browell [10] used a three-wavelength elastic backscatter lidar to discriminate aerosol types with a retrieval method that reduces the uncertainty of lidar ratio. The elastic lidar theory, practice, and analysis methods were described in detail by Kovalev et al. (2004) [11]. The other one is called “high spectral resolution lidar (HSRL)”. In 1983, Shipley et al. pointed out that the optical-scattering properties of aerosols could be measured by HSRL [12]. Subsequently by using this lidar technique, the optical properties of cirrus clouds were obtained by Grund and Eloranta (1991) [13] and aerosol profiles were measured by Hair et al. (2008) [14]. Recently, the HSRL has been widely used in classification of particle types [15–17]. The third one is designated “Raman lidar”. The Raman lidar technique was applied to measure particle optical properties more than 30 years ago, which can be divided into two subcategories, i.e., vib-rotational Raman (VRR) lidar and pure rotational Raman lidar (PRR). Ansmann et al. (1990) demonstrated that particle backscatter coefficient and extinction coefficient can be acquired from the VRR lidar measurements [18]. Later, Ferrare et al. presented the retrieved extinction and backscatter coefficients of atmospheric aerosols with this technique [19,20]. In 1992, Mitev et al. examined the potential to retrieve the aerosol extinction profiles from the PRR lidar measurements [21]. Behrendt et al. implemented this retrieval by building a temperature-independent reference signal in 2002 [22]. Achtert et al. (2013) calculated the particle backscatter and extinction coefficients by using the PRR lidar measurements [23]. Recently, dual-wavelength Raman lidar came into operation and provided a solid way for comprehensive studies of various particles [24–27].
However, the above-mentioned retrieval algorithms of particle optical properties are all ill-posed. In the case of elastic backscatter lidar measurement, due to the fact that two unknowns (particle backscatter coefficient and extinction coefficient) exist in one available lidar equation, the assumption of constant lidar ratio (the ratio of extinction coefficient to backscatter coefficient) is introduced to solve this problem [9,28]. A similar situation takes place also in VRR lidar measurement. There are three unknowns (particle extinction coefficients at both the transmitted and Raman-shifted wavelengths as well as particle backscatter coefficient at the transmitted wavelength), whereas only two independent lidar equations are available. Thus, an assumed wavelength dependence about the particle extinction coefficient (Ångström relationship) is introduced to solve the lidar equations [18]. Obviously, if the lidar ratio or Ångström relationship is not well assumed, both the retrieval algorithms might lead to large errors [27]. The retrieval of particle properties from the common PRR lidar measurements also requires an impractical assumption that molecule backscatter cross section is temperature independent. Although the HSRL measurement provides a good solution for retrieving the particle optical properties, its technical complexity and very demanding requirement in transmitting laser and receiver optics limit its application and development [12]. Recently, Weng et al. (2018) found that all-day temperature profiles can be calculated exactly by using two isolated PRR line signals [29]. Making use of the retrieved temperature profile, the particle optical properties can be further strictly determined from the measurements of the single-line-extracted pure rotational Raman lidar.
In this paper, based on a self-developed single-line-extracted pure rotational Raman (PRR) lidar by Wuhan University, the methodology for accurate retrieval of particle optical properties without additional assumptions is described in more detail. We utilize the observational data from this lidar between February 2016 and December 2017 to accurately obtain the optical properties (backscatter coefficient, extinction coefficient and lidar ratio) of continental polluted aerosols, dust aerosols, and cirrus cloud particles over Wuhan (30.5°N, 114.4°E), China. Individual observational examples show how the optical parameters for different particle types vary with altitudes. The relevant statistical results provide the observational basis for atmospheric particle classification.
2. Instruments and methodology
2.1 Instruments
In this study, particle measurements were made together by a single-line-extracted PPR lidar and a polarization lidar (PL), both of which were located at our atmospheric observation site on the campus of Wuhan University (30.5°N, 114.4°E and ∼80 m above sea level). The single-line-extracted PRR lidar allows for the profile measurements of atmospheric temperature, particle backscatter coefficient, particle extinction coefficient, and the lidar ratio, while the PL performs the profile measurements of particle depolarization ratio. In addition, the radiosonde released from Wuhan Weather Station about 23.4 km northwest of our lidar site provides temperature profiles to verify the temperature results retrieved by the single-line-extracted PRR lidar.
2.1.1 Single-line-extracted PRR lidar
The single-line-extracted PRR lidar was already introduced in detail by Weng et al. (2018) [29], so here only provides a brief description. The lidar transmitter utilizes an injection-seeded frequency-doubled Nd:YAG laser to emit a 532.237-nm laser light with a pulse energy of ∼800 mJ, a repetition rate of 30 Hz, and a very narrow linewidth of ∼3 pm. In the receiver, a 200-mm Cassegrain telescope receives the backscatter signals with a field of view of ∼0.4 mrad. Three lidar detection channels are designed for recording the elastic backscatter signal and two isolated N2 molecule PRR line signals. The detection of PRR line signals for the two Raman channels is based on the application of Fabry-Perot interferometer (FPI). Each FPI has a bandwidth of ∼30 pm and a peak transmission of ∼60% centered at the wavelength of the N2 anti-Stokes PRR line with rotational quantum number of J=6 or 16. As seen from Fig. 1, the two N2 anti-Stokes PRR lines with J=6 and 16 have spectral distances lager than 140 pm from adjacent O2 PRR lines. Hence, the usage of FPI precisely extracts the wanted N2 line signal and provides enough suppression to the adjacent O2 PRR line signals. Besides, by inserting additional narrow-band (∼0.3 nm) interference filters (IF), each Raman channel also has enough suppression to the elastic signal. In conclusion, the combination of narrow-band IFs and FPIs in the two Raman channels ideally extract the two PRR line signals with J=6 and 16 without contamination from other PRR line signals or elastic signals. Last, all the recorded signals are stored with a vertical resolution of 7.5 m and a temporal resolution of 1 minute.
Fig. 1. Differential backscatter cross sections of anti-Stokes PPR lines of N2 (blue) and O2 (red) molecules calculated for an incident laser wavelength of 532.237 nm and at a temperature of 200 K. The respective relative volume abundances are 0.78 for N2 and 0.21 for O2.
2.1.2 Polarization lidar
The lidar emits a 532.2-nm laser light with pulse energy of ∼120 mJ and repetition rate of 20 Hz. A 300-mm Cassegrain telescope collects the backscatter signal with a field of view of ∼1 mrad. Two orthogonal detection channels are designed to detect the parallel (P) and perpendicular (S) signal components. The spatial and temporal resolutions are 3.75 m and 1 minute for the stored original signals. For detailed descriptions of the polarization lidar can refer to Kong and Yi (2015) [30].
2.1.3 Radiosonde
The radiosondes are launched daily at 0000 UTC (0800 Local Time, LT) and 1200 UTC (2000 LT) from Wuhan Weather Station (∼23.4 km away from our atmospheric observation site). Profiles of air pressure, temperature, relative humidity (RH), and horizontal wind speed are measured. The temperature profiles are used to verify the reliability of the PPR lidar measurements by comparing concurrent temperature results. Besides, molecular backscatter and extinction profiles are also derived from the radiosonde data.
2.2 Methodology
2.2.1 Single-line method
The single-line-extracted PRR lidar extracts exactly the N2 anti-Stokes PRR line signals with J=6 and 16. The corresponding lidar equation can be written as [31]:
Where N(z) is the recorded photon counts at altitude z after background subtraction. C stands for the system constant for a detection channel. $\beta $ is backscatter coefficient. $\alpha $ means extinction coefficient. Subscripts a and m denote particle and molecule, respectively. ${\lambda _0}$ is incident laser wavelength and ${\lambda _J}$ is Raman-shifted wavelength.
According to the Raman scattering theory [32], temperature T can be strictly retrieved from the ratio of the two extracted line signals in the following form [29]:
With
Here Q is Raman signal ratio. E is molecule rotational energy. k is Boltzmann’s constant. Constant A can be ideally determined by theory. While constant B depends on two parts: the first part, ${C_{J = 16}}/{C_{J = 6}}$ is ratio of overall optical transmitting efficiencies for the two Raman channels and can be calibrated in-lab with the help of a dye laser with tunable output wavelength [28]; the second part, X(J=16)/X(J=6), is theoretically determined to be a value of 88/31.
The lidar equation for the elastic channel is given by:
Considering the fact that the Raman-shifted signal wavelengths and the incident laser wavelength have differences of <4.0 nm (see Fig. 1), it is reasonable to make the following approximation:
Hence, the ratio of Eq. (6) to Eq. (1) can be simplified as:
From Eq. (8), particle backscatter coefficient can be derived by:
According to definition of backscatter coefficient, ${\beta _m}$ can be expressed as:
Where n(z) is the number density of atmospheric molecules; $\sigma $ is differential backscatter cross section; 0.78 is the volume abundance of N2 molecule in the atmosphere. Combining Eqs. (9)–(11), the particle backscatter coefficient ${\beta _{a,{\lambda _0}}}(z)$ can be written as:
By selecting a reference altitude ${z_0}$ where the atmosphere is clear (${\beta _{a,{\lambda _0}}}({z_0})\textrm{ = 0}$), we can get:
By inserting Eq. (13) into Eq. (12), the particle backscatter coefficient ${\beta _{a,{\lambda _0}}}(z)$ is replaced by:
According to definition of Raman differential backscatter cross section [31,33], we get:
Here, h is Planck’s constant, c is velocity of light, and ${B_{{N_2}}}$ is rotational constant for N2.
After the temperature retrieval using Eq. (2), particle backscatter coefficient ${\beta _{a,{\lambda _0}}}(z)$ and backscatter ratio R(z) can be obtained by combining Eqs. (14) and (15):
By inserting the above-derived ${\beta _{a,{\lambda _0}}}(z)$ into Eq. (6), the particle extinction coefficient can be further retrieved by:
Finally, according to the derived backscatter coefficient and extinction coefficient, the ratio of extinction coefficient to backscatter coefficient, or namely the lidar ratio S, can also be retrieved. It is concluded that the single-line-extracted PRR lidar allows strict solution of aerosol optical parameters (backscatter/extinction coefficient and lidar ratio) without introducing additional assumptions (constant lidar ratio or Ångström relationship) or weight coefficients (making the sum of PRR lines to be temperature insensitive but not absolutely independent on temperature in common PRR lidar measurements). The relevant retrieval approach is designated as “single-line method” in this work. Obviously, this method could avoid the quantitative uncertainties resulted by the assumptions or weight coefficients in particle optical property retrieval.
2.2.2 Particle depolarization ratio
From the PL, the volume linear depolarization δv is calculated by:
Where Ns and Np are the PL-recorded parallel and perpendicular signals, respectively. G is the gain ratio between the two channels which was calibrated by the Δ90° method proposed by Freudenthaler et al. (2009) [34]. Then particle linear depolarization ratio δp can be obtained by [35]:
Here δm is the molecular linear depolarization ratio with a value of 0.004 according to the optical parameters of the PL [36,37].
3. Observational results
3.1 Case studies
In this part, the concurrent measurement data between 27 February 2016 and 21 December 2017 from both lidars are used for studying the particle properties. First, a total of 53 particle layer events are screened out. For picking out a particle layer event, the layer must have signal intensities with reasonable signal-to-noise (SNR) and persist for more than 2 hours; besides, when viewing the time-height contour plots of the range-corrected-signal (RCS) obtained from the elastic channel of the PRR lidar (e.g., Fig. 2(a)) and δv from the PL (e.g., Fig. 2(b)), the layer must has clear-cut boundaries and stable δv values. Then, the selected particle layers are roughly categorized into three typical types according to their δp values, lidar ratios and appearance altitudes: the layer having δp of ∼0.03-0.07, lidar ratios of ∼50-70 sr and altitude below 3 km is designated as continental polluted aerosol [17,38]; while the layer having δp of ∼0.15-0.4, lidar ratios of ∼40-55 sr and top altitude below ∼7 km is regarded as dust [39,40]; finally the layer having δp of ∼0.2-0.4, lidar ratios of ∼5-40 sr and base altitude reaching ∼7 km is attributed as cirrus cloud [37]. Measurement result of each typical type of particle layer is presented below.
Fig. 2. Time-height plots of: a. range-corrected signal (RCS) obtained by the elastic channel of the single-line-extracted PRR lidar; b. volume linear depolarization ratio (δv) measured by the polarization lidar. The respective spatial and temporal resolutions are 7.5 m and 1 minute. Note that a continental polluted aerosol layer persisted at altitudes of ∼1.7-2.8 km on the observational date of 3 September 2016.
3.1.1 Continental polluted aerosol case
Figure 2 shows a particle layer measurement result between 0000 and 0300 LT on 3 September 2016. According to MODIS true color image and backward trajectories analysis, this layer was a locally formed aerosol layer. Figure 2(a) provides the time-height plot of RCS measured by the elastic channel of the single-line-extracted PRR lidar, while Fig. 2(b) plots ${\delta _v}$ by the PL. It is seen that at altitudes of 1.7∼2.8 km during this observational period, the layer generally exhibited intensified backscatter and relatively low ${\delta _v}$ (∼0.02-0.05), indicating that it contained mainly continental polluted aerosols.
Figure 3 gives detailed altitude-resolved information of the layer. Figure 3(a) plots lidar-derived temperature between 2000 and 2100 LT (dashed); for comparison, the concurrent radiosonde temperature is also added (solid). The lidar and radiosonde temperatures agree well with each other at the altitudes of 1.7-2.8 km with absolute deviation of less than 2 K (Fig. 3(b)), indicating that the PRR lidar retrieves reliable temperature results. Figure 3(c) presents the relative humidity (RH) obtained from radiosonde at 0800 LT on 3 September 2016. Figure 3(d) shows three successive temperature profiles during the observational period of the layer. For discrimination, the colors of green, blue and red are used to stand for the time periods of 0000-0100, 0100-0200 and 0200-0300 LT on 3 September 2016, respectively. Figures 3(e)–3(h) plots successively profiles of aerosol backscatter coefficient, extinction coefficient, lidar ratio and δp with 150-m and 1-h resolution. As seen from Figs. 3(e) and 3(f), the layer top altitudes decreased slightly from 2.9 to 2.6 km while the base heights increased slightly from 1.6 to 1.8 km. Besides, the layer has lidar ratio values spanning from 40 to 75 sr and mean δp of 0.05 ± 0.01 (Figs. 3(g) and 3(h)). This demonstrates that this measured layer was a typical continental polluted aerosol layer. Figure 3(g) also shows that the lidar ratios tended to increase with altitude while the δp decreased with altitude. The RH obtained from radiosonde was found to be increasing with altitude during 1.8-2.6 km (Fig. 3(c)). The reason that the significant altitude-dependence of lidar ratio is hence attributed to aerosol hygroscopic growth [41].
Fig. 3. Profile results of: a. lidar-derived temperature (dashed) during 2000-2100 LT and radiosonde temperature (solid) lunched at 2000 LT on 2 September 2016; b. absolute temperature deviation between PRR lidar and radiosonde; c. the relative humidity (RH) obtained from radiosonde lunched at 0800 LT on 3 September 2016; d. lidar-derived temperature; e. backscatter coefficient; f. extinction coefficient; g. lidar ratio; h. particle linear depolarization ratio. Error bars denote the 1-σ uncertainty induced by signal noise. The spatial resolution is 150 m and temporal resolution is 1 h. In panels d to h, colors of green for 0000-0100, blue for 0100-0200 and red for 0200-0300 LT on 3 September 2016, respectively.
Table 1 summarizes the vertical mean optical properties of this continental polluted aerosol layer. Generally, the aerosol optical parameters varied less than 15% in the observed three hours, indicating that the measured layer existed stably. In detail, the backscatter (extinction) coefficient derived by the single-line method is 2.8 ± 0.6 Mm-1 sr-1 (160 ± 48 Mm-1), while the lidar ratio is approximately 58 ± 8 sr. These results agree well with those of the earlier study by Wang et al. (2016) [42].
Table 1. Lidar-derived mean aerosol optical properties on 3 September 2016
3.1.2 Dust layer case
Figure 4 provides a particle layer measurement result between 2100 and 0100 LT on the date of 22-23 April 2017. As seen from Figs. 4(a) and 4(b), the descending particle layer persisted at altitudes of 3.5-5.5 km during the observational period and generally exhibited intensified backscatter and relatively high δv (∼0.24), indicating that it contained mainly dust aerosols.
Fig. 4. Same as Fig. 2 but on the date of 22-23 April 2017. Note that a descending dust layer persisted at altitudes of ∼3.5-5.5 km.
Detailed height-resolved information is presented in Fig. 5. As seen from Fig. 5(a), lidar-derived temperature between 2000 and 2100 LT (dashed) matched well with concurrent radiosonde temperature (solid) at altitudes of 3-6 km. The absolute deviation is less than 2 K (Fig. 5(b)). Figure 5(c) shows four successive temperature profiles during the observational period of the layer. For discrimination, the colors of green, blue, purple and red are used to stand for the time periods of 2100-2200, 2200-2300, 2300-2400 and 0000-0100 LT on 22-23 April 2017, respectively. Figures 5(d)–5(g) plots successively profiles of aerosol backscatter coefficient, extinction coefficient, lidar ratio, and δp with 150-m and 1-h resolution. It is indicated that the layer has lidar ratio values range from 40 to 60 sr (Fig. 5(f)) and mean δp of 0.34 ± 0.02 (Fig. 5(g)). This demonstrates that this measured layer was a typical dust layer.
Fig. 5. Same as Fig. 3 but on the date of 22-23 April 2017. Error bars denote 1-σ uncertainty. In panels c to g, colors of green for 2100-2200, blue for 2200-2300, purple for 2300-2400 LT on 22 April 2017 and red for 0000-0100 LT on the next day, respectively.
Table 2 summarizes the vertical mean optical properties of this dust layer. It is found that the mean lidar ratio varies less than 4% in these hours, indicating that the composition of this layer is stable during the observational period. In detail, the backscatter (extinction) coefficient derived by single-line method is 1.7 ± 0.6 Mm-1 sr-1 (80 ± 29 Mm-1), while the lidar ratio is 48 ± 8 sr, which shows a good agreement with the value (45 ± 7 sr) reported by Hu et al. (2020) [43]. Figure 6 provides the 48-h backward trajectories ending at 2000 LT (1200 UTC) on 22 April 2017 for air masses at 4.6 km. As seen from Fig. 6, the air masses at this altitude originated from Taklamakan desert and the transport speed of this dust layer was fast as it only took about 28 hours to arrive at Wuhan. Thus, the sedimentation of the dust layer was mild during the ∼3600 km transport and the optical properties were close to the dust aerosol of source area.
Fig. 6. The 48-h backward trajectories calculated by HYSPLIT ending at 2000 LT (1200 UTC) on 22 April 2017 for air masses at 4.6 km.
Table 2. Lidar-derived mean aerosol optical properties on 22-23 April 2017
3.1.3 Cirrus cloud case
Figure 7 gives a particle layer measurement result between 2000 and 2400 LT on 21 December 2017. Combining the relatively strong backscatter lidar signals at altitudes of ∼6.5-9.5 km (Fig. 7(a)) and the large δv (∼0.3; Fig. 7(b)), the observed layer is regarded as cirrus cloud layer.
Fig. 7. Same as Fig. 2 but on the date of 21 December 2017. Note that cirrus cloud layer persisted at altitudes of ∼6.5-9.5 km between 2000 and 2400 LT.
Figure 8 presents detailed altitude-resolved information of the layer. Figure 8(a) plots lidar-derived temperature between 2000 and 2100 LT (dashed) and concurrent radiosonde temperature (solid). The temperature results by lidar and radiosonde agree well with each other in the altitudes of 6.5-9 km with absolute deviation of less than 2 K (Fig. 8(b)). Figure 8(c) shows three successive temperature profiles during the observational period of the layer. For better discrimination, the colors of green, blue and red are used to stand for the time periods of 2030-2130, 2130-2230 and 2230-2330 LT on 21 December 2017, respectively. Figures 8(d)–8(g) plots successively profiles of aerosol backscatter coefficient, extinction coefficient, lidar ratio, and δp with 150-m and 1-h resolution. As shown in Figs. 8(d) and 8(e), the base and top heights of this layer were ∼6.6 and 9 km respectively. Besides, the layer has lidar ratio values ranged from 15 to 30 sr (Fig. 8(f)) and mean δp of 0.25 ± 0.07 (Fig. 8(g)). This demonstrates that this measured layer was a typical cirrus cloud layer. During 2030-2130 LT, the backscatter coefficient showed a prominent peak at the altitude of 9.2 km, while both the lidar ratio and δp took their minimum values at the corresponding altitude, indicating that horizontally oriented ice crystals occurred [44,45].
Fig. 8. Same as Fig. 3 but on the date of 21 December 2017. In panels c to g, colors of green for 2030-2130, blue for 2130-2230 and red for 2230-2330 LT, respectively.
Table 3 summarizes the vertical mean optical properties of this cirrus cloud layer. The mean backscatter (extinction) coefficient is 33 Mm-1 sr-1 (610 Mm-1) during 2030-2130 LT, but dropped sharply to 13 Mm-1 sr-1 (250 Mm-1) during 2230-2330 LT, indicating that the cirrus cloud was disappearing from our lidar view. In detail, the backscatter (extinction) coefficient derived by the single-line method is 25 ± 10 Mm-1 sr-1 (350 ± 160 Mm-1) for the cirrus cloud, and the lidar ratio is 19 ± 3 sr. Wang et al. (2020) reported that the lidar ratio and δp of cirrus cloud layer in winter of Wuhan were mostly ranged from 15 to 20 sr and 0.2 to 0.3, respectively [37], which agree well with the results of this case.
Table 3. Lidar-derived mean aerosol optical properties on 21 December 2017
3.1.4 Discussion on the Fernald method
For evaluating the retrieval performance of particle optical properties by the traditional Fernald method, Fig. 9 compares the results of backscatter and extinction coefficients obtained by the single-line method and Fernald method for the above-presented three particle cases. In terms of the previous studies, the assumed lidar ratios for Fernald method are 55 sr for continental polluted aerosol, 50 sr for dust aerosol, and 20 sr for cirrus cloud, respectively [37,46,47].
Fig. 9. Profiles of backscatter coefficient, extinction coefficient and lidar ratio for a. a continental polluted aerosol case during 0000-0100 LT on 3 September 2016; b. a dust case during 0000-0100 LT on 23 April 2017; c. a cirrus cloud case during 2130-2230 LT on 21 December 2017. In all panels, red lines stand for results by single-line method and green lines for results by Fernald method. Assumed lidar ratios for Fernald method are 55 sr for continental polluted aerosol, 50 sr for dust aerosol, and 20 sr for cirrus cloud particle, respectively.
As shown in Fig. 9, the retrieved results of backscatter and extinction coefficients by the Fernald method (green) exhibit discernible discrepancies from those by the single-line method (red). For the continental polluted aerosol case during 0000-0100 LT on 3 September 2016 (Fig. 9(a)), the deviation is 12-14% (mean deviation 13 ± 1%) for the backscatter coefficient and 1-39% (mean deviation 13 ± 9%) for the extinction coefficient. For the dust case during 0000-0100 LT on 23 April 2017 (Fig. 9(b)), the deviation is 2-15% (mean deviation 7 ± 3%) for the backscatter coefficient and 1-35% (mean deviation 15 ± 9%) for the extinction coefficient. While for the cirrus cloud case during 2130-2230 LT on 21 December 2017 (Fig. 9(c)), the deviation is 1-25% (mean deviation 9 ± 8%) for the backscatter coefficient and 1-41% (mean deviation 16 ± 12%) for the extinction coefficient. It is also observed that the Fernald method always yields larger backscatter coefficient values for all three particle cases.
3.2 Statistical results
For a better investigation of discriminating optical characteristics of the three-type particles, Fig. 10 provides the scatterplot of particle linear depolarization ratio δp versus lidar ratio S retrieved for all the 53 particle layer events. Note that for generating a data pair of (δp, S) plotted in Fig. 10, the profiles of δp and S are first calculated with 1-h resolution; then the vertical mean values of δp and S profiles are taken and serve as one “data point”. Totally, there are 51 data points obtained for continental polluted aerosol, 135 for dust, and 104 for cirrus cloud, respectively.
Fig. 10. Scatterplot of lidar ratio S versus particle linear depolarization ratio δp for continental polluted aerosols (square; 51 points), dust aerosols (asterisk; 135 points), and cirrus cloud particles (dot; 104 points) measured between 27 February 2016 and 21 December 2017 over Wuhan.
As seen from Fig. 10, the three-type particles exhibit obviously distinct value ranges for S and δp. For continental polluted aerosol (square), it has the largest lidar ratios ranging from 43 to 74 sr with a mean value of 60 ± 7 sr, while its δp is small with a mean of 0.06 ± 0.01. For the dust aerosol (asterisk), the lidar ratio value is moderate spanning a range of 38-56 sr with a mean of 47 ± 4 sr, but the δp is large with a mean of 0.27 ± 0.05. For cirrus cloud particles (dot), the lidar ratio is the smallest falling into the range of 12-30 sr with a mean of 22 ± 4 sr, while the δp is the largest with a mean of 0.29 ± 0.08; note there are cirrus cloud cases with δp values of <0.2, this is attributed to the occurrence of horizontally oriented ice crystals during the 1-h average as horizontally oriented ice crystals can yield near-zero depolarization ratio [44,45].
Table 4 summarizes the lidar ratio values for the three-type particles obtained by previous studies and this work. Hu et al. (2020) found that the lidar ratio for Taklamakan dust aerosol was 45 ± 7 sr [43]; the reason of slightly larger value (47 ± 4 sr) for dust aerosol by this work is possibly due to the sedimentation of large size of particles and the mixing of anthropogenic aerosols during the transport pass. Wang et al. (2016) [42] reported measurement results of continental polluted aerosols over Wuhan with mean lidar ratio value of 57 ± 7 sr; our results show a slightly larger mean value of 60 ± 7 sr and this might be caused by mixing of smoke which has larger lidar ratio [48]. Moreover, the mean lidar ratio for cirrus cloud particles is 22 ± 4 sr by this study, which is nearly identical to the value (22 ± 8 sr) provided by Wang et al. (2020) [37]. It is concluded from Table 4 that the lidar ratios measured by the single-line-extracted PRR lidar generally show good agreements with results by previous studies.
Table 4. Summaries of lidar ratios for continental polluted aerosol, dust aerosol, and cirrus cloud by previous studies and this work
4. Summary and conclusion
This work presents retrieval method and measurement results of particle optical properties (backscatter/extinction coefficient and lidar ratio) using a single-line-extracted pure rotational Raman (PRR) lidar. The lidar exactly extracts two N2 anti-Stokes PRR line signals with J=6 and 16. Temperature is first retrieved by the ratio of the two Raman line signals. Then by combining the obtained temperature and elastic signal, a single-line method is proposed for strict deriving of particle backscatter/extinction coefficient and lidar ratio.
Based on the observations of the single-line-extracted PRR lidar from February 2016 to December 2017, the optical properties (backscatter/extinction coefficient and lidar ratio) of continental polluted aerosols, dust aerosols, and cirrus cloud over Wuhan (30.5°N, 114.4°E) are studied. Typical measurement cases are provided for detailed description of the vertical optical characteristics of the three-type particles. Statistical results of 53 particle layer events reveal that the mean lidar ratio value is 60 ± 7 sr for continental polluted aerosol, 47 ± 4 sr for dust aerosol and 22 ± 4 sr for cirrus cloud. These measured lidar ratio values show good agreement with those of previous studies.
Traditional Fernald method is also verified and shows general deviations of 7-13% for backscatter coefficients and deviations of 13-16% for extinction coefficients, when compared to results by the single-line method. It is concluded that the optical properties measured by the single-line-extracted PRR lidar can serve as observational standards for particle optical properties (backscatter/extinction coefficient and lidar ratio) at 532 nm wavelength.
Funding
National Natural Science Foundation of China (41927804); Meridian Space Weather Monitoring Project; Fundamental Research Funds for the Central Universities.
Disclosures
The authors declare no conflicts of interest.
Data availability
Date 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. A. Ansmann, M. Tesche, P. Seifert, D. Althausen, R. Engelmann, J. Fruntke, U. Wandinger, I. Mattis, and D. Müller, “Evolution of the ice phase in tropical altocumulus: SAMUM lidar observations over Cape Verde,” J. Geophys. Res. 114(D17), D17208 (2009). [CrossRef]
2. H. Guan, B. Schmid, A. Bucholtz, and R. Bergstrom, “Sensitivity of shortwave radiative flux density, forcing, and heating rate to the aerosol vertical profile,” J. Geophys. Res. 115, D06209 (2010). [CrossRef]
3. K. N. Liou, “Influence of cirrus clouds on weather and climate process: A global perspective,” Mon. Weather Rev. 114(6), 1167–1199 (1986). [CrossRef]
4. H. Nazaryan, M. P. McCormick, and W. P. Menzel, “Global characterization of cirrus clouds using CALIPSO data,” J. Geophys. Res. 113(D16), D16211 (2008). [CrossRef]
5. P. Seifert, A. Ansmann, D. Müller, U. Wandinger, D. Althausen, A. J. Heymsfield, S. T. Massie, and C. Schmitt, “Cirrus optical properties observed with lidar, radiosonde, and satellite over the tropical Indian Ocean during the aerosol-polluted northeast and clean maritime southwest monsoon,” J. Geophys. Res. 112(D17), D17205 (2007). [CrossRef]
6. K. Van Tricht, Z. Ulanowski, G. McFarquhar, U. Lohmann, P. Lawson, C. Franklin, D. J. Cziczo, P. Brown, A. Luebke, M. Krämer, and A. J. Heymsfield, “Cirrus Clouds,” in Ice Formation and Evolution in Clouds and precipitation: Measurement and Modeling Challenges (AMS American Meteorological Society, 2017), Chap. 2.
7. A. J. Heymsfield and C. M. R. Platt, “A parameterization of the particle-size spectrum of ice clouds in terms of the ambient-temperature and the ice water content,” J. Atmos. Sci. 41(5), 846–855 (1984). [CrossRef]
8. Y. He, F. Yi, Y. Yi, F. Liu, and Y. Zhang, “Heterogeneous Nucleation of Midlevel Cloud Layer Influenced by Transported Asian Dust Over Wuhan (30.5°N, 114.4°E), China,” J. Geophys. Res. 126, e2020JD033394 (2021). [CrossRef]
9. F. G. Fernald, “Analysis of atmospheric lidar observations: some comments,” Appl. Opt. 23(5), 652–653 (1984). [CrossRef]
10. Y. Sasano and E. V. Browell, “Light scattering characteristics of various aerosol types derived from multiple wavelength lidar observations,” Appl. Opt. 28(9), 1670–1679 (1989). [CrossRef]
11. V. A. Kovalev and W. E. Eichinger, Elastic Lidar: Theory, Practice, and Analysis Methods (John Wiley & Sons Inc, 2004). [CrossRef]
12. S. T. Shipley, D. H. Tracy, E. W. Eloranta, J. T. Trauger, J. T. Sroga, F. L. Roesler, and J. A. Weinman, “High spectral resolution lidar to measure optical scattering properties of atmospheric aerosols 1: Theory and instrumentation,” Appl. Opt. 22(23), 3716–3724 (1983). [CrossRef]
13. C. J. Grund and E. W. Eloranta, “University of Wisconsin High Spectral Resolution Lidar,” Opt. Eng. 30(1), 6–12 (1991). [CrossRef]
14. J. W. Hair, C. A. Hostetler, A. L. Cook, D. B. Harper, R. A. Ferrare, T. L. Mack, W. Welch, L. R. Izquierdo, and F. E. Hovis, “Airborne High Spectral Resolution Lidar for profiling aerosol optical properties,” Appl. Opt. 47(36), 6734–6752 (2008). [CrossRef]
15. S. P. Burton, R. A. Ferrare, C. A. Hostetler, J. W. Hair, R. R. Rogers, M. D. Obland, C. F. Butler, A. L. Cook, D. B. Harper, and K. D. Froyd, “Aerosol classification using airborne High Spectral Resolution Lidar measurements – methodology and examples,” Atmos. Meas. Tech. 5(1), 73–98 (2012). [CrossRef]
16. S. Groß, M. Esselborn, B. Weinzierl, M. Wirth, A. Fix, and A. Petzold, “Aerosol classification by airborne high spectral resolution lidar observations,” Atmos. Chem. Phys. 13(5), 2487–2505 (2013). [CrossRef]
17. S. P. Burton, R. A. Ferrare, M. A. Vaughan, A. H. Omar, R. R. Rogers, C. A. Hostetler, and J. W. Hair, “Aerosol classification from airborne HSRL and comparisons with the CALIPSO vertical feature mask,” Atmos. Meas. Tech. 6(5), 1397–1412 (2013). [CrossRef]
18. A. Ansmann, M. Riebesell, and C. Weitkamp, “Measurement of atmospheric aerosol extinction profiles with a Raman lidar,” Opt. Lett. 15(13), 746–748 (1990). [CrossRef]
19. R. A. Ferrare, S. H. Melfi, D. N. Whiteman, and K. D. Evans, “Raman lidar measurements of Pinatubo aerosols over southeastern Kansas during November-December 1991,” Geophys. Res. Lett. 19(15), 1599–1602 (1992). [CrossRef]
20. R. A. Ferrare, S. H. Melfi, D. N. Whiteman, K. D. Evans, and R. Leifer, “Raman lidar measurements of aerosol extinction and backscattering: 1. Methods and comparisons,” J. Geophys. Res. 103(D16), 19663–19672 (1998). [CrossRef]
21. V. M. Mitev, I. V. Grigorov, and V. B. Simeonov, “Lidar measurement of atmospheric aerosol extinction profiles: a comparison between two techniques—Klett inversion and pure rotational Raman scattering methods,” Appl. Opt. 31(30), 6469–6474 (1992). [CrossRef]
22. 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,” Appl. Opt. 41(36), 7657–7666 (2002). [CrossRef]
23. P. Achtert, M. Khaplanov, F. Khosrawi, and J. 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]
24. D. Muller, I. Mattis, U. Wandinger, A. Ansmann, D. Althausen, and A. Stohl, “Raman lidar observations of aged Siberian and Canadian forest fire smoke in the free troposphere over Germany in 2003: Microphysical particle characterization,” J. Geophys. Res. 110, D17201 (2005). [CrossRef]
25. S. Gross, M. Tesche, V. Freudenthaler, C. Toledano, M. Wiegner, A. Ansmann, D. Althausen, and M. Seefeldner, “Characterization of Saharan dust, marine aerosols and mixtures of biomass-burning aerosols and dust by means of multi-wavelength depolarization and Raman lidar measurements during SAMUM 2,” Tellus B: Chemical and Physical Meteorology 63(4), 706–724 (2011). [CrossRef]
26. T. Kanitz, R. Engelmann, B. Heinold, H. Baars, A. Skupin, and A. Ansmann, “Tracking the Saharan Air Layer with shipborne lidar across the tropical Atlantic,” Geophys. Res. Lett. 41(3), 1044–1050 (2014). [CrossRef]
27. I. Veselovskii, D. N. Whiteman, M. Korenskiy, A. Suvorina, and D. Pérez-Ramírez, “Use of rotational Raman measurements in multiwavelength aerosol lidar for evaluation of particle backscattering and extinction,” Atmos. Meas. Tech. 8(10), 4111–4122 (2015). [CrossRef]
28. A. Ansmann, M. Riebesell, U. Wandinger, C. Weitkamp, E. Voss, W. Lahmann, and W. Michaelis, “Combined raman elastic-backscatter LIDAR for vertical profiling of moisture, aerosol extinction, backscatter, and LIDAR ratio,” Appl. Phys. B 55, 18–28 (1992). [CrossRef]
29. 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]
30. W. Kong and F. Yi, “Convective boundary layer evolution from lidar backscatter and its relationship with surface aerosol concentration at a location of a central China megacity,” J. Geophys. Res. Atmos. 120(15), 7928–7940 (2015). [CrossRef]
31. 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]
32. P. Di Girolamo, A. Behrendt, and V. Wulfmeyer, “Spaceborne profiling of atmospheric temperature and particle extinction with pure rotational Raman lidar and of relative humidity in combination with differential absorption lidar: performance simulations,” Appl. Opt. 45(11), 2474–2494 (2006). [CrossRef]
33. A. Behrendt, “Temperature Measurements with Lidar,” in Lidar Range-Resolved Optical Remote Sensing of the AtmosphereC. Weikamp, ed. (Springer, 2005), pp. 273–305.
34. V. Freudenthaler, M. Esselborn, M. Wiegner, B. Heese, M. Tesche, A. Ansmann, D. Müller, D. Althausen, M. Wirth, A. Fix, G. Ehret, P. Knippertz, C. Toledano, J. Gasteiger, M. Garhammer, and M. Seefeldne, “Depolarization ratio profiling at several wavelengths in pure Saharan dust during SAMUM 2006,” Tellus B: Chemical and Physical Meteorology 61(1), 165–179 (2009). [CrossRef]
35. J. Biele, G. Beyerle, and G. Baumgarten, “Polarization lidar: Corrections of instrumental effects,” Opt. Express 7(12), 427–435 (2000). [CrossRef]
36. A. Behrendt and T. Nakamura, “Calculation of the calibration constant of polarization lidar and its dependency on atmospheric temperature,” Opt. Express 10(16), 805–817 (2002). [CrossRef]
37. W. Wang, F. Yi, F. Liu, Y. Zhang, C. Yu, and Z. Yin, “Characteristics and Seasonal Variations of Cirrus Clouds from Polarization Lidar Observations at a 30°N Plain Site,” Remote Sensing 12(23), 3998 (2020). [CrossRef]
38. D. Nicolae, J. Vasilescu, C. Talianu, I. Binietoglou, V. Nicolae, S. Andrei, and B. Antonescu, “A Neural Network Aerosol Typing Algorithm Based on Lidar Data,” Atmos. Chem. Phys. 18(19), 14511–14537 (2018). [CrossRef]
39. Y. He and F. Yi, “Dust Aerosols Detected Using a Ground-Based Polarization Lidar and CALIPSO over Wuhan (30.5 degrees N, 114.4 degrees E), China,” Advances in Meteorology 2015, D536762 (2015). [CrossRef]
40. Z. Liu, N. Sugimoto, and T. Murayama, “Extinction-to-backscatter ratio of Asian dust observed with high-spectral-resolution lidar and Raman lidar,” Appl. Opt. 41(15), 2760–2767 (2002). [CrossRef]
41. G. Zhao, C. S. Zhao, Y. Kuang, J. C. Tao, W. S. Tan, Y. X. Bian, J. Li, and C. C. Li, “Impact of aerosol hygroscopic growth on retrieving aerosol extinction coefficient profiles from elastic-backscatter lidar signals,” Atmos. Chem. Phys. 17(19), 12133–12143 (2017). [CrossRef]
42. W. Wang, W. Gong, F. Mao, Z. Pan, and B. Liu, “Measurement and Study of Lidar Ratio by Using a Raman Lidar in Central China,” IJERPH 13(5), 508 (2016). [CrossRef]
43. Q. Hu, H. Wang, P. Goloub, Z. Li, I. Veselovskii, T. Podvin, K. Li, and M. Korenskiy, “The characterization of Taklamakan dust properties using a multiwavelength Raman polarization lidar in Kashi, China,” Atmos. Chem. Phys. 20(22), 13817–13834 (2020). [CrossRef]
44. A. Ansmann, I. Mattis, D. Müller, U. Wandinger, M. Radlach, D. Althausen, and R. Damoah, “Ice formation in Saharan dust over central Europe observed with temperature/humidity//aerosol Raman lidar,” J. Geophys. Res. 110, D18S12 (2005). [CrossRef]
45. V. Noel and K. Sassen, “Study of ice crystals orientation in ice clouds based on polarized observations from the fars scanning lidar,” in 22nd International Laser Radar Conference, G. Pappalardo and A. Amodeo, eds. (2004), pp. 309–312.
46. D. Müller, A. Ansmann, I. Mattis, M. Tesche, U. Wandinger, D. Althausen, and G. Pisani, “Aerosol-type-dependent lidar ratios observed with Raman lidar,” J. Geophys. Res. 112(D16), D16202 (2007). [CrossRef]
47. Y. M. Noh, Y. J. Kim, and D. Müller, “Seasonal characteristics of lidar ratios measured with a Raman lidar at Gwangju, Korea in spring and autumn,” Atmos. Environ. 42(9), 2208–2224 (2008). [CrossRef]
48. U. Wandinger, D. Müller, C. Bockmann, D. Althausen, V. Matthias, J. Bosenberg, V. Weiss, M. Fiebig, M. Wendisch, A. Stohl, and A. Ansmann, “Optical and microphysical characterization of biomass-burning and industrial-pollution aerosols from multiwavelength lidar and aircraft measurements,” J. Geophys. Res. 107, D8125 (2002). [CrossRef]
49. A. Hänel, H. Baars, D. Althausen, A. Ansmann, R. Engelmann, and J. Y. Sun, “One-year aerosol profiling with EUCAARI Raman lidar at Shangdianzi GAW station: Beijing plume and seasonal variations,” J. Geophys. Res. 117, D13201 (2012). [CrossRef]
50. Y. Jin, K. Kai, T. Shibata, K. Zhang, and H. Zhou, “Validation of the Dust Layer Structure over the Taklimakan Desert, China by the CALIOP Space-Borne Lidar Using Ground-Based Lidar,” SOLA 6, 121–124 (2010). [CrossRef]
51. J. Hofer, D. Althausen, S. F. Abdullaev, A. N. Makhmudov, B. I. Nazarov, G. Schettler, R. Engelmann, H. Baars, K. W. Fomba, K. Mueller, B. Heinold, K. Kandler, and A. Ansmann, “Long-term profiling of mineral dust and pollution aerosol with multiwavelength polarization Raman lidar at the Central Asian site of Dushanbe, Tajikistan: case studies,” Atmos. Chem. Phys. 17(23), 14559–14577 (2017). [CrossRef]
52. M. Filioglou, E. Giannakaki, J. Backman, J. Kesti, A. Hirsikko, R. Engelmann, E. O’Connor, J. T. T. Leskinen, X. Shang, H. Korhonen, H. Lihavainen, S. Romakkaniemi, and M. Komppula, “Optical and geometrical aerosol particle properties over the United Arab Emirates,” Atmos. Chem. Phys. 20(14), 8909–8922 (2020). [CrossRef]
53. E. Giannakaki, D. S. Balis, V. Amiridis, and S. Kazadzis, “Optical and geometrical characteristics of cirrus clouds over a Southern European lidar station,” Atmos. Chem. Phys. 7(21), 5519–5530 (2007). [CrossRef]
54. D. Dionisi, P. Keckhut, G. L. Liberti, E. Cardillo, and E. Congeduti, “Midlatitude cirrus classification at Rome Tor Vergata through a multichannel Raman-Mie-Rayleigh lidar,” Atmos. Chem. Phys. 13(23), 11853–11868 (2013). [CrossRef]
55. D. A. Gouveia, B. Barja, H. M. J. Barbosa, P. Seifert, H. Baars, T. Pauliquevis, and P. Artaxo, “Optical and geometrical properties of cirrus clouds in Amazonia derived from 1 year of ground-based lidar measurements,” Atmos. Chem. Phys. 17(5), 3619–3636 (2017). [CrossRef]
