Advanced Search
Article Contents

Ice Nucleation of Cirrus Clouds Related to the Transported Dust Layer Observed by Ground-Based Lidars over Wuhan, China


doi: 10.1007/s00376-021-1192-x

  • Cirrus clouds related to transported dust layers were identified on 22 occasions with ground-based polarization lidar from December 2012 to February 2018 over Wuhan (30.5°N, 114.4°E), China. All the events occurred in spring and winter. Cirrus clouds were mostly located above 7.6 km on top of the aloft dust layers. In-cloud relative humidity with respect to ice (RHi) derived from water vapor Raman lidar as well as from ERA5 reanalysis data were used as criteria to determine the possible ice nucleation regimes. Corresponding to the two typical cases shown, the observed events can be classified into two categories: (1) category A (3 cases), in-cloud peak RHi ≥ 150%, indicating competition between heterogeneous nucleation and homogeneous nucleation; and (2) category B (19 cases), in-cloud peak RHi < 150%, revealing that only heterogeneous nucleation was involved. Heterogeneous nucleation generally took place during instances of cirrus cloud formation in the upper troposphere when advected dust particles were present. Although accompanying cloud-top temperatures ranged from –51.9°C to –30.4°C, dust-related heterogeneous nucleation contributed to primary ice nucleation in cirrus clouds by providing ice nucleating particle concentrations on the order of 10−3 L−1 to 102 L−1. Heterogeneous nucleation and subsequent crystal growth reduced the ambient RHi to be less than 150% by consuming water vapor and thus completely inhibited homogeneous nucleation.
    摘要: 利用地基偏振激光雷达在2012年12月至2018年2月期间在中国武汉观测到了(30.5°N, 114.4°E)22例受远距离输送而来沙尘影响的卷云事件。所有事件都出现在春季和冬季。观测到的卷云主要位于沙尘层的顶部,高度大于7.6 km。为了鉴别卷云的冰核化机制,我们使用了分别来自水汽拉曼激光雷达观测和ERA5再分析数据的云内冰面相对湿度(RHi)的峰值作为判据。基于此,对2个典型事件进行了个例分析。根据相同判据,将观测到受沙尘影响的卷云分为2类:(1)第一类(3例):云内峰值RHi超过150%,表明了异质核化和均质核化存在竞争关系;(2)第二类(19例):云内峰值RHi小于150%,表明仅发生异质核化。当存在平流输送而来的沙尘时,异质核化在高层卷云的形成中起主导作用。研究表明,在较低的云顶温度下(–51.9°C– –30.4°C),沙尘可提供 10−3 L−1至102 L−1量级的冰核数浓度,沙尘导致的异质核化过程对卷云的初始冰晶生成具有重要影响。除此之外,异质核化过程以及冰晶的生长会消耗水汽,使周围环境的RHi降低至150%以下,进而抑制均质核化的发生。
  • 加载中
  • Figure 1.  Time-height contour plots (1-min/30-m resolution) of (b) range-corrected signal and (d) volume depolarization ratio measured by a 532 nm polarization lidar on 12 January 2018 during 2030–2400 LST. Height profiles of (a) temperature and relative humidity (RH with respect to water and ice respectively), and (c) wind speed of the U (westerly), V (southerly), and W (vertical velocity) components obtained from ECMWF ERA5 reanalysis data during 2200–2300 LST. The horizontal dashed lines indicate cloud top level with a temperature of –47.8°C. The vertical dotted line indicates 100% relative humidity. Several cirrus clouds (8.6–9.3 km) appeared on the top of the dust layer.

    Figure 2.  Profiles of (a) relative humidity (RH with respect to water and ice) derived by lidar and ERA5 reanalysis data and (b) water vapor mixing ratio and saturated water vapor mixing ratio derived by lidar during 2100–2200 LST, and (c) temperature measured by radiosonde launched at 2000 LST on 12 January 2018.

    Figure 3.  Profiles of (a) dust and total (dust + non-dust) extinction coefficient, (b) dust and total (dust + non-dust) backscatter coefficient, (c) volume depolarization ratio $ {\delta }_{\mathrm{v}} $, and (d) particle depolarization ratio $ {\delta }_{\mathrm{p}} $ derived by the 532 nm polarization lidar during 2000–2030 LST on 12 January 2018. Horizontal error bars denote the relative errors of each parameter.

    Figure 4.  Profiles of (a) dust mass concentration $ {M}_{\mathrm{d}} $, (b) dust surface area concentration $ {S}_{\mathrm{d}} $, and (c) ice-nucleating particle concentration derived by the POLIPHON method using the U17-D parameterization scheme during 2000–2030 LST on 12 January 2018. The horizontal dashed lines denote the temperature height levels of –35°C and –47.8°C (cloud top temperature). $ {S}_{\mathrm{i}} $ represents ice saturation ratio.

    Figure 5.  Time-height contour plots (1 min by 30 m resolution) of (b) range-corrected signal and (d) volume depolarization ratio measured by a 532 nm polarization lidar for 0000–0200 LST 7 February 2018. Height profiles of (a) temperature and relative humidity (RH with respect to water and ice, respectively), and (c) wind speed of the U (westerly), V (southerly), and W (vertical velocity) components obtained from ECMWF ERA5 reanalysis data during 0000–0100 LST. The horizontal dashed lines indicate cloud top level with a temperature of –51.9°C. The vertical dotted line indicates 100% relative humidity. Several cirrus clouds (9.0–10.1 km) appeared embedded in the upper part of the dust layer.

    Figure 6.  Profiles of (a) relative humidity (RH with respect to water and ice) derived by lidar and ERA5 reanalysis data and (b) water vapor mixing ratio and saturated water vapor mixing ratio derived by lidar during 0000–0100 LST on 7 February 2018, and (c) temperature profile measured by radiosonde launched at 2000 LST on 6 February 2018.

    Figure 7.  Profiles of (a) dust and total (dust + non-dust) extinction coefficient, (b) dust and total (dust + non-dust) backscatter coefficient, (c) volume depolarization ratio $ {\delta }_{\mathrm{v}} $, and (d) particle depolarization ratio $ {\delta }_{\mathrm{p}} $ derived by the 532 nm polarization lidar during 2000–2030 LST on 7 February 2018. Horizontal error bars denote the relative errors of each parameter.

    Figure 8.  Profiles of (a) dust mass concentration $ {M}_{\mathrm{d}} $, (b) dust surface area concentration $ {S}_{\mathrm{d}} $, and (c) ice-nucleating particle concentration derived by the POLIPHON method during 0140–0150 LST on 7 February 2018. The horizontal dashed lines denote the temperature height levels of –35°C and –51.9°C (cloud top temperature). The INPC profiles obtained by the U17-D parameterization schemes are given. $ {S}_{\mathrm{i}} $ represents ice saturation ratio.

    Figure 9.  Overview of cloud top temperature and in-cloud peak RHi (obtained by ERA5 reanalysis data) of cirrus clouds for each case. The red circles (CTT < –36°C and also RHi ≥ 150%) indicate the cases showing competition between heterogeneous and homogeneous nucleation. The dark blue circles (100% < RHi < 150%) indicate that only heterogeneous nucleation takes place.

    Table 1.  Estimated uncertainties in the dust-related optical parameters and cloud-relevant properties (Kafle and Coulter, 2013; Ansmann et al., 2019b).

    ParameterUncertainty
    Dust backscatter coefficient $ {\beta }_{\mathrm{d}} $10%
    Dust extinction coefficient $ {\alpha }_{\mathrm{d}} $20%
    Volume depolarization ratio $ {\delta }_{\mathrm{v}} $5%
    Particle depolarization ratio $ {\delta }_{\mathrm{p}} $5%–10%
    Dust mass concentration ${M }_{\mathrm{d} }$20%–30%
    Dust surface area concentration $ {S}_{\mathrm{d}} $30%–40%
    INP number concentration $ {n}_{\mathrm{I}\mathrm{N}\mathrm{P}} $50%–500%
    DownLoad: CSV

    Table 2.  Applied values of the conversion parameters and constants required in the POLIPHON retrieval.

    ParametervalueUnitReference
    $ {c}_{\mathrm{v},\mathrm{d}} $0.77×10−12Mm(Ansmann et al., 2019b)
    $ {c}_{\mathrm{s},\mathrm{d}} $3.1×10−12Mm m2 cm−3(Ansmann et al., 2019b)
    Dust lidar ratio LR45sr(He et al., 2021a)
    Dust density $ {\mathrm{\rho }}_{\mathrm{d}} $2.6g cm−3(Wagner et al., 2009)
    DownLoad: CSV

    Table 3.  Information on 22 cirrus cloud layer cases with dust-related ice nucleation during December 2012–February 2018 at Wuhan (30.5°N, 114.4°E) obtained from polarization lidar observations and ECMWF ERA5 reanalysis data. Cloud top height (CTH) and the dust-layer-averaged values of dust extinction coefficient (Dust Ext.), dust mass concentration (Dust mass), volume depolarization ratio (VDR), and particle depolarization ratio (PDR) are calculated based on the polarization lidar data. Cloud top temperature (CTT), and respective in-cloud peak RHw and RHi are given by ERA5 reanalysis data. $ {S}_{\mathrm{i}} $ represents the ice saturation ratio. The maximum INPCs within the dust layer for three assumed constant $ {S}_{\mathrm{i}} $ (1.15, 1.25, and 1.35) are calculated by the POLIPHON method using U17-D parameterization schemes. Peak INPC is calculated with the CTT, cloud top $ {S}_{\mathrm{d}} $, and in-cloud peak RHi, indicating the upper limit INPC in the actual atmosphere.

    DateTime (LST)CTH
    (km)
    CTT
    (°C)
    Peak RHw (%)Peak RHi (%)VDRPDRDust Ext. (Mm−1)Dust mass (μg m−3)U17D
    (L−1) (Si=1.15)
    U17D
    (L−1) (Si=1.25)
    U17D
    (L−1) (Si=1.35)
    Peak INPC (L−1)
    201212310730−11009.9−46.3100.3153.90.040.374.318.630.0240.5715.9034139.180
    201301140527−07208.1−37.3101.9138.80.030.293.396.790.0010.0130.0850.051
    201312240510−07379.3−45.396.2140.20.050.375.3610.720.0080.1321.044NAN
    201312250445−08009.2−43.597.0140.10.180.4242.8085.690.0040.0500.32419.333
    201412201627−17187.7−38.576.1107.50.050.2310.2420.510.0160.3523.405NAN
    201503122102−22078.8−42.888.1120.20.070.2415.1530.320.0010.0100.0590.045
    201503220405−06508.6−36.198.8141.30.020.212.805.610.0010.0110.0750.278
    201603110100−02108.5−39.480.7111.20.100.3521.3642.760.0030.0470.3730.105
    201603282050−23159.1−36.4102.1154.00.070.3111.2622.55NAN0.0030.019NAN
    201611282055−24009.4−41.591.4140.80.030.283.517.030.54417.639224.9575.215
    201704130810−10008.9−34.091.6129.30.030.254.789.570.0010.0130.0790.016
    201704140220−04459.0−38.182.8123.30.030.234.519.020.0080.1501.3190.038
    201704230545−07577.8−30.492.7122.80.05NAN2.915.83NANNANNANNAN
    201704272335−23579.7−39.781.2109.40.030.303.577.160.0320.7958.413NAN
    201704280000−06209.0−33.495.8129.20.030.324.529.06NANNANNANNAN
    1746−18009.4−38.578.3116.10.030.282.114.230.0892.57430.2280.042
    201705172220−230810.0−42.788.4130.60.030.303.136.260.1685.07461.5672.521
    201712092340−24008.3−30.493.0127.90.050.247.2714.56NANNANNANNAN
    201712100300−04558.1−34.493.1126.00.090.3416.1032.24NANNANNANNAN
    201801090557−07147.6−36.183.6111.40.030.184.869.730.0010.0070.041NAN
    201801122055−23289.7−47.8100.0157.40.070.3210.4020.820.0812.31526.96618739.800
    201802070013−013210.2−51.983.3133.90.050.336.1312.272.647102.3141484.9306576.980
    DownLoad: CSV
  • Ansmann, A., and Coauthors, 2019a: Ice-nucleating particle versus ice crystal number concentrationin altocumulus and cirrus layers embedded in Saharan dust: A closure study. Atmospheric Chemistry and Physics, 19, 1 5087−1 5115, https://doi.org/10.5194/acp-19-15087-2019.
    Ansmann, A., R.-E. Mamouri, J. Hofer, H. Baars, D. Althausen, and S. F. Abdullaev, 2019b: Dust mass, cloud condensation nuclei, and ice-nucleating particle profiling with polarization lidar: Updated POLIPHON conversion factors from global AERONET analysis. Atmospheric Measurement Techniques, 12, 4849−4865, https://doi.org/10.5194/amt-12-4849-2019.
    Barahona, D., and A. Nenes, 2009: Parameterizing the competition between homogeneous and heterogeneous freezing in cirrus cloud formation – monodisperse ice nuclei. Atmospheric Chemistry and Physics, 9, 369−381, https://doi.org/10.5194/acp-9-369-2009.
    Bolton, D., 1980: The computation of equivalent potential temperature. Mon. Wea. Rev., 108(7), 1046−1053, https://doi.org/10.1175/1520-0493(1980)108<1046:TCOEPT>2.0.CO;2.
    Bühl, J., P. Seifert, A. Myagkov, and A. Ansmann, 2016: Measuring ice- and liquid-water properties in mixed-phase cloud layers at the Leipzig Cloudnet station. Atmospheric Chemistry and Physics, 16, 1 0609−1 0620, https://doi.org/10.5194/acp-16-10609-2016.
    Bühl, J., P. Seifert, R. Engelmann, and A. Ansmann, 2019: Impact of vertical air motions on ice formation rate in mixed-phase cloud layers. npj Climate and Atmospheric Science, 2, 36, https://doi.org/10.1038/s41612-019-0092-6.
    Campbell, J., and H. K. Christenson, 2018: Nucleation- and emergence-limited growth of ice from pores. Physical Review Letters, 120, 165701, https://doi.org/10.1103/PhysRevLett.120.165701.
    Campbell, J. R., S. Lolli, J. R. Lewis, Y. Gu, and E. J. Welton, 2016: Daytime cirrus cloud top-of-the-atmosphere radiative forcing properties at a midlatitude site and their global consequences. J. Appl. Meteorol. Climatol., 55, 1667−1679, https://doi.org/10.1175/JAMC-D-15-0217.1.
    Cziczo, D. J., and Coauthors, 2013: Clarifying the dominant sources and mechanisms of cirrus cloud formation. Science, 340, 1320−1324, https://doi.org/10.1126/science.1234145.
    David, R. O., and Coauthors, 2019: Pore condensation and freezing is responsible for ice formation below water saturation for porous particles. Proceedings of the National Academy of Sciences of the United States of America, 116, 8184−8189, https://doi.org/10.1073/pnas.1813647116.
    DeMott, P. J., M. P. Meyers, and W. R. Cotton, 1994: Parameterization and impact of ice initiation processes relevant to numerical model simulations of cirrus clouds. J. Atmos. Sci., 51, 77−90, https://doi.org/10.1175/1520-0469(1994)051<0077:PAIOII>2.0.CO;2.
    DeMott, P. J., D. J. Cziczo, A. J. Prenni, D. M. Murphy, S. M. Kreidenweis, D. S. Thomson, R. Borys, and D. C. Rogers, 2003: Measurements of the concentration and composition of nuclei for cirrus formation. Proceedings of the National Academy of Sciences of the United States of America, 100(25), 1 4655−1 4660, https://doi.org/10.1073/pnas.2532677100.
    Dierens, K., 2003: On the transition between heterogeneous and homogeneous freezing. Atmospheric Chemistry and Physics, 3, 437−446, https://doi.org/10.5194/acp-3-437-2003.
    Fernald, F. G., 1984: Analysis of atmospheric lidar observations: Some comments. Appl. Opt., 23, 652−653, https://doi.org/10.1364/AO.23.000652.
    Freudenthaler, V., and Coauthors, 2009: Depolarization ratio profiling at several wavelengths in pure Saharan dust during SAMUM 2006. Tellus B, 61, 165−179, https://doi.org/10.1111/j.1600-0889.2008.00396.x.
    Froyd, K. D., and Coauthors, 2013: Cirrus cloud formation and the role of heterogeneous ice nuclei. AIP Conference Proceedings, 1527, 976−978, https://doi.org/10.1063/1.4803436.
    Froyd, K. D., D. M. Murphy, P. Lawson, D. Baumgardner, and R. L. Herman, 2010: Aerosols that form subvisible cirrus at the tropical tropopause. Atmospheric Chemistry and Physics, 10, 209−218, https://doi.org/10.5194/acp-10-209-2010.
    Gamage, S. M., R. J. Sica, G. Martucci, and A. Haefele, 2020: A 1D var retrieval of relative humidity using the ERA5 dataset for the assimilation of Raman lidar measurements. J. Atmos. Oceanic Technol., 37(11), 2051−2064, https://doi.org/10.1175/JTECH-D-19-0170.1.
    Haag, W., B. Kärcher, J. Ström, A. Minikin, U. Lohmann, J. Ovarlez, and A. Stohl, 2003: Freezing thresholds and cirrus cloud formation mechanisms inferred from in situ measurements of relative humidity. Atmospheric Chemistry and Physics, 3, 1791−1806, https://doi.org/10.5194/acp-3-1791-2003.
    He, Y., and F. Yi, 2015: Dust aerosols detected using a ground-based polarization lidar and CALIPSO over Wuhan (30.5°N, 114.4°E), China. Advances in Meteorology, 2015, 536762, https://doi.org/10.1155/2015/536762.
    He, Y., F. Yi, Y. Yi, F. C. Liu, and Y. P. Zhang, 2021a: Heterogeneous nucleation of midlevel cloud layer influenced by transported Asian dust over Wuhan (30.5°N, 114.4°E), China. J. Geophys. Res. Atmos., 126(2), e2020JD033394, https://doi.org/10.1029/2020JD033394.
    He, Y., Y. F. Zhang, F. C. Liu, Z. P. Yin, Y. Yi, Y. F. Zhan, and F. Yi, 2021b: Retrievals of dust-related particle mass and ice-nucleating particle concentration profiles with ground-based polarization lidar and sun photometer over a megacity in central China. Atmospheric Measurement Techniques, 14, 5939−5954, https://doi.org/10.5194/amt-14-5939-2021.
    He, Y., F. C. Liu, Z. P. Yin, Y. P. Zhang, Y. F. Zhan, and F. Yi, 2021c: Horizontally Oriented ice crystals observed by the synergy of zenith- and slant-pointed polarization lidar over Wuhan (30.5°N, 114.4°E), China. Journal of Quantitative Spectroscopy and Radiative Transfer, 268, 107626, https://doi.org/10.1016/j.jqsrt.2021.107626.
    Hersbach, H., and Coauthors, 2020: The ERA5 global reanalysis. Quart. J. Roy. Meteor. Soc., 146, 1999−2049, https://doi.org/10.1002/qj.3803.
    Heymsfield, A. J., and Coauthors, 2017: Cirrus clouds. Meteor. Monogr., 58, 2.1−2.26, https://doi.org/10.1175/AMSMONOGRAPHS-D-16-0010.1.
    Hoffmann, N., A. Kiselev, D. Rzesanke, D. Duft, and T. Leisner, 2013: Experimental quantification of contact freezing in an electrodynamic balance. Atmospheric Measurement Techniques, 6, 2373−2382, https://doi.org/10.5194/amt-6-2373-2013.
    Hoose, C., and O. Möhler, 2012: Heterogeneous ice nucleation on atmospheric aerosols: A review of results from laboratory experiments. Atmospheric Chemistry and Physics, 12, 9817−9854, https://doi.org/10.5194/acp-12-9817-2012.
    Hu, Q. Y., H. F. Wang, P. Goloub, Z. Q. Li, I. Veselovskii, T. Podvin, K. T. Li, and M. Korenskiy, 2020: The characterization of Taklamakan dust properties using a multiwavelength Raman polarization lidar in Kashi, China. Atmospheric Chemistry and Physics, 20, 1 3817−1 3834, https://doi.org/10.5194/acp-20-13817-2020.
    Huang, J. P., P. Minnis, B. Lin, T. H. Wang, Y. H. Yi, Y. X. Hu, S. Sun-Mack, and K. Ayers, 2006: Possible influences of Asian dust aerosols on cloud properties and radiative forcing observed from MODIS and CERES. Geophys. Res. Lett., 33, L06824, https://doi.org/10.1029/2005GL024724.
    Immler, F., R. Treffeisen, D. Engelbart, K. Krüger, and O. Schrems, 2008: Cirrus, contrails, and ice supersaturated regions in high pressure systems at northern mid latitudes. Atmospheric Chemistry and Physics, 8, 1689−1699, https://doi.org/10.5194/acp-8-1689-2008.
    IPCC, 2013: Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press.
    Kafle, D. N., and R. L. Coulter, 2013: Micropulse lidar-derived aerosol optical depth climatology at ARM sites worldwide. J. Geophys. Res. Atmos., 118, 7293−7308, https://doi.org/10.1002/jgrd.50536.
    Kanji, Z. A., L. A. Ladino, H. K. Wex, Y. Boose, M. Burkert‐Kohn, D. J. Cziczo, and M. Krämer, 2017: Overview of ice nucleating particles. Meteor. Monogr., 58, 1.1−1.33, https://doi.org/10.1175/AMSMONOGRAPHS-D-16-0006.1.
    Kärcher, B., and U. Lohmann, 2003: A parameterization of cirrus cloud formation: Heterogeneous freezing. J. Geophys. Res., 108(D14), 4402, https://doi.org/10.1029/2002JD003220.
    Kärcher, B., A. Dörnbrack, and I. Sölch, 2014: Supersaturation variability and cirrus ice crystal size distributions. J. Atmos. Sci., 71, 2905−2926, https://doi.org/10.1175/JAS-D-13-0404.1.
    Kojima, T., P. R. Buseck, Y. Iwasaka, A. Matsuki, and D. Trochkine, 2006: Sulfate-coated dust particles in the free troposphere over Japan. Atmospheric Research, 82(3-4), 698−708, https://doi.org/10.1016/j.atmosres.2006.02.024.
    Kong, W., and F. Yi, 2015: Convective border layer evolution from lidar backscatter and its relationship with surface aerosol concentration at a location of a central China megacity. Geophysics should be Geophysical or uniformly use the abbreviation: J. Geophys. Res. Atmos., 120, 7928−7940, https://doi.org/10.1002/2015JD023248.
    Koop, T., B. P. Luo, A. Tsias, and T. Peter, 2000: Water activity as the determinant for homogeneous ice nucleation in aqueous solutions. Nature, 406, 611−614, https://doi.org/10.1038/35020537.
    Krämer, M., and Coauthors, 2009: Ice supersaturations and cirrus cloud crystal numbers. Atmospheric Chemistry and Physics, 9, 3505−3522, https://doi.org/10.5194/acp-9-3505-2009.
    Krämer, M., and Coauthors, 2016: A microphysics guide to cirrus clouds – Part 1: Cirrus types. Atmospheric Chemistry and Physics, 16, 3463−3483, https://doi.org/10.5194/acp-16-3463-2016.
    Krämer, M., and Coauthors, 2020: A microphysics guide to cirrus – Part 2: Climatologies of clouds and humidity from observations. Atmospheric Chemistry and Physics, 20, 1 2569−1 2608, https://doi.org/10.5194/acp-20-12569-2020.
    Kuebbeler, M., U. Lohmann, J. Hendricks, and B. Kärcher, 2014: Dust ice nuclei effects on cirrus clouds. Atmospheric Chemistry and Physics, 14, 3027−3046, https://doi.org/10.5194/acp-14-3027-2014.
    Leblanc, T., I. S. McDermid, and T. D. Walsh, 2012: Ground-based water vapor Raman lidar measurements up to the upper troposphere and lower stratosphere for long-term monitoring. Atmospheric Measurement Techniques, 5, 17−36, https://doi.org/10.5194/amt-5-17-2012.
    Liu, D., Z. E. Wang, Z. Y. Liu, D. Winker, and C. Trepte, 2008: A height resolved global view of dust aerosols from the first year CALIPSO lidar measurements. J. Geophys. Res., 113, D16214, https://doi.org/10.1029/2007JD009776.
    Liu, F. C., and F. Yi, 2013: Spectrally resolved Raman lidar measurements of gaseous and liquid water in the atmosphere. Appl. Opt., 52(28), 6884−6895, https://doi.org/10.1364/AO.52.006884.
    Liu, X., X. Shi, K. Zhang, E. J. Jensen, A. Gettelman, D. Barahona, A. Nenes, and P. Lawson, 2012: Sensitivity studies of dust ice nuclei effect on cirrus clouds with the Community Atmosphere Model CAM5. Atmospheric Chemistry and Physics, 12, 1 2061−1 2079, https://doi.org/10.5194/acp-12-12061-2012.
    Lohmann, U., P. Spichtinger, S. Jess, T. Peter, and H. Smit, 2008: Cirrus cloud formation and ice supersaturated regions in a global climate model. Environmental Research Letters, 3, 045022, https://doi.org/10.1088/1748-9326/3/2/025002.
    Mamouri, R. E., and A. Ansmann, 2014: Fine and Coarse dust separation with polarization lidar. Atmospheric Measurement Techniques, 7, 3717−3735, https://doi.org/10.5194/amt-7-3717-2014.
    Mamouri, R. E., and A. Ansmann, 2015: Estimated desert-dust ice nuclei profiles from polarization lidar: Methodology and case studies. Atmospheric Chemistry and Physics, 15, 3463−3477, https://doi.org/10.5194/acp-15-3463-2015.
    Mann, H. B., and D. R. Whitney, 1947: On a test of whether one of two random variables is stochastically larger than the other. The Annals of Mathematical Statistics, 18(1), 50−60, https://doi.org/10.1214/aoms/1177730491.
    Marcolli, C., 2014: Deposition nucleation viewed as homogeneous or immersion freezing in pores and cavities. Atmospheric Chemistry and Physics, 14, 2071−2104, https://doi.org/10.5194/acp-14-2071-2014.
    Marcolli, C., 2017: Pre-activation of aerosol particles by ice preserved in pores. Atmospheric Chemistry and Physics, 17, 1595−1622, https://doi.org/10.5194/acp-17-1595-2017.
    Marcolli, C., 2020: Technical note: Fundamental aspects of ice nucleation via pore condensation and freezing including Laplace pressure and growth into macroscopic ice. Atmospheric Chemistry and Physics, 20, 3209−3230, https://doi.org/10.5194/acp-20-3209-2020.
    Marinou, E., and Coauthors, 2019: Retrieval of ice-nucleating particle concentrations from lidar observations and comparison with UAV in situ measurements. Atmospheric Chemistry and Physics, 19, 1 1315−1 1342, https://doi.org/10.5194/acp-19-11315-2019.
    Murphy, D. M., and T. Koop, 2005: Review of the vapour pressures of ice and supercooled water for atmospheric applications. Quart. J. Roy. Meteor. Soc., 131, 1539−1565, https://doi.org/10.1256/qj.04.94.
    Murray, B. J., D. O’Sullivan, J. D. Atkinson, and M. E. Webb, 2012: Ice nucleation by particles immersed in supercooled cloud droplets. Chemical Society Reviews, 41, 6519−6554, https://doi.org/10.1039/C2CS35200A.
    Murray, F. W., 1967: On the computation of saturation vapor pressure. J. Appl. Meteorol. Climatol., 6, 203−204, https://doi.org/10.1175/1520-0450(1967)006<0203:OTCOSV>2.0.CO;2.
    Pan, H. L., and Coauthors, 2019: Seasonal and vertical distributions of aerosol type extinction coefficients with an emphasis on the impact of dust aerosol on the microphysical properties of cirrus over the Taklimakan Desert in Northwest China. Atmos. Environ., 203, 216−227, https://doi.org/10.1016/j.atmosenv.2019.02.004.
    Peng, L., F. Yi, F. C. Liu, Z. P. Yin, and Y. He, 2021: Optical properties of aerosol and cloud particles measured by a single-line-extracted pure rotational Raman lidar. Optics Express, 29(14), 2 1947−2 1964, https://doi.org/10.1364/OE.427864.
    Sakai, T., T. Nagai, Y. Zaizen, and Y. Mano, 2010: Backscattering linear depolarization ratio measurements of mineral, sea-salt, and ammonium sulfate particles simulated in a laboratory chamber. Appl. Opt., 49, 4441−4449, https://doi.org/10.1364/AO.49.004441.
    Sassen, K., Z. E. Wang, and D. Liu, 2008: Global distribution of cirrus clouds from CloudSat/Cloud-Aerosol Lidar and Infrared Pathfinder Satellite Observations (CALIPSO) measurements. J. Geophys. Res., 113, D00A12, https://doi.org/10.1029/2008JD009972.
    Seifert, P., and Coauthors, 2010: Saharan dust and heterogeneous ice formation: Eleven years of cloud observations at a central European EARLINET site. J. Geophys. Res., 115, D20201, https://doi.org/10.1029/2009JD013222.
    Spice, A., D. W. Johnson, P. R. A. Brown, A. G. Darlison, and C. P. R. Saunders, 1999: Primary ice nucleation in orographic cirrus clouds: A numerical simulation of the microphysics. Quart. J. Roy. Meteor. Soc., 125, 1637−1667, https://doi.org/10.1002/qj.49712555708.
    Tesche, M., A. Ansmann, D. Müller, D. Althausen, R. Engelmann, V. Freudenthaler, and S. Groß, 2009: Vertically resolved separation of dust and smoke over Cape Verde using multiwavelength Raman and polarization lidars during Saharan Mineral Dust Experiment 2008. J. Geophys. Res., 114, D13202, https://doi.org/10.1029/2009JD011862.
    Ullrich, R., and Coauthors, 2017: A new ice nucleation active site parameterization for desert dust and soot. J. Atmos. Sci., 74, 699−717, https://doi.org/10.1175/JAS-D-16-0074.1.
    Wagner, F., and Coauthors, 2009: Properties of dust aerosol particles transported to Portugal from the Sahara Desert. Tellus B, 61, 297−306, https://doi.org/10.1111/j.1600-0889.2008.00393.x.
    Wang, W., F. Yi, F. C. Liu, Y. P. Zhang, C. M. Yu, and Z. P. Yin, 2020: Characteristics and seasonal variations of cirrus clouds from polarization lidar observations at a 30°N plain site. Remote Sensing, 12, 3998, https://doi.org/10.3390/rs12233998.
    Wang, W. C., L. F. Sheng, H. C. Jin, and Y. Q. Han, 2015: Dust aerosol effects on cirrus and altocumulus clouds in Northwest China. Journal of Meteorological Research, 29(5), 793−805, https://doi.org/10.1007/s13351-015-4116-9.
    Weger, M., and Coauthors, 2018: The impact of mineral dust on cloud formation during the Saharan dust event in April 2014 over Europe. Atmospheric Chemistry and Physics, 18, 1 7545−1 7572, https://doi.org/10.5194/acp-18-17545-2018.
    Whiteman, D. N., S. H. Melfi, and R. A. Ferrare., 1992: Raman lidar system for the measurement of water vapor and aerosols in the Earth’s atmosphere. Appl. Opt., 31(16), 3068−3082, https://doi.org/10.1364/AO.31.003068.
    Wiegner, M., and Coauthors, 2009: Numerical simulations of optical properties of Saharan dust aerosols with emphasis on lidar applications. Tellus B, 61, 180−194, https://doi.org/10.1111/j.1600-0889.2008.00381.x.
    Wu, C., and F. Yi, 2017: Local ice formation via liquid water growth in slowly ascending humid aerosol/liquid water layers observed with ground-based lidars and radiosondes. J. Geophys. Res. Atmos., 122, 4479−4493, https://doi.org/10.1002/2016JD025765.
    Yin, Z. P., F. Yi, Y. He, F. C. Liu, C. M. Yu, Y. P. Zhang, and W. Wang, 2021: Asian dust impacts on heterogeneous ice formation at Wuhan based on polarization lidar measurements. Atmos. Environ., 246, 118166, https://doi.org/10.1016/j.atmosenv.2020.118166.
  • [1] Y. BHAVANI KUMAR, C. NAGESWARA RAJU, M. KRISHNAIAH, 2006: Indo-Japanese Lidar Observations of the Tropical Middle Atmosphere During 1998 and 1999, ADVANCES IN ATMOSPHERIC SCIENCES, 23, 711-725.  doi: 10.1007/s00376-006-0711-0
    [2] Huang Shihong, Qian Changguo, Wang Weimin, Li Ruxiang, 1994: On Mechanisms of Nucleation of Ice Crystals by Aerodynamic Cooling, ADVANCES IN ATMOSPHERIC SCIENCES, 11, 57-64.  doi: 10.1007/BF02656994
    [3] Jianjun LIU, Zhanqing LI, ZHENG Youfei, Maureen CRIBB, 2015: Cloud-Base Distribution and Cirrus Properties Based on Micropulse Lidar Measurements at a Site in Southeastern China, ADVANCES IN ATMOSPHERIC SCIENCES, 32, 991-1004.  doi: 10.1007/s00376-014-4176-2
    [4] Sandeep D. WAGH, Baban NAGARE, Sanjay D. MORE, P. Pradeep KUMAR, 2017: Multiyear Observations of Deposition-Mode Ice Nucleating Particles at Two High-Altitude Stations in India, ADVANCES IN ATMOSPHERIC SCIENCES, 34, 1437-1446.  doi: 10.1007/s00376-017-7048-8
    [5] Yao Keya, Liu Chunlei, 1996: ICE Particle Size and Shape Effect on Solar Energy Scattering Angular Distribution, ADVANCES IN ATMOSPHERIC SCIENCES, 13, 505-510.  doi: 10.1007/BF03342040
    [6] Liu Chunlei, Yao Keya, 1996: Calculation of Ice Crystal Diffraction, ADVANCES IN ATMOSPHERIC SCIENCES, 13, 340-348.  doi: 10.1007/BF02656851
    [7] Sungkyun SHIN, Young Min NOH, Kwonho LEE, Hanlim LEE, Detlef M?LLER, Y. J. KIM, Kwanchul KIM, Dongho SHIN, 2014: Retrieval of the Single Scattering Albedo of Asian Dust Mixed with Pollutants Using Lidar Observations, ADVANCES IN ATMOSPHERIC SCIENCES, 31, 1417-1426.  doi: 10.1007/s00376-014-3244-y
    [8] MIN Min, WANG Pucai, James R. CAMPBELL, ZONG Xuemei, XIA Junrong, 2011: Cirrus Cloud Macrophysical and Optical Properties over North China from CALIOP Measurements, ADVANCES IN ATMOSPHERIC SCIENCES, 28, 653-664.  doi: 10.1007/s00376-010-0049-5
    [9] ZHONG Lingzhi, LIU Liping, DENG Min, ZHOU Xiuji, 2012: Retrieving Microphysical Properties and Air Motion of Cirrus Clouds Based on the Doppler Moments Method Using Cloud Radar, ADVANCES IN ATMOSPHERIC SCIENCES, 29, 611-622.  doi: 10.1007/s00376-011-0112-x
    [10] Huang Runheng, Kuo_Nan Liou, 1984: REMOTE SOUNDING OF THE CIRRUS OPTICAL DEPTH AND TEMPERATURE FROM 3.7 AND 11 MICROMETER WINDOWS, ADVANCES IN ATMOSPHERIC SCIENCES, 1, 150-178.  doi: 10.1007/BF02678128
    [11] GAO Yongqi, SUN Jianqi, LI Fei, HE Shengping, Stein SANDVEN, YAN Qing, ZHANG Zhongshi, Katja LOHMANN, Noel KEENLYSIDE, Tore FUREVIK, SUO Lingling, 2015: Arctic Sea Ice and Eurasian Climate: A Review, ADVANCES IN ATMOSPHERIC SCIENCES, 32, 92-114.  doi: 10.1007/s00376-014-0009-6
    [12] Huang Runheng, Kuo-Nan Liou, 1985: EFFECTS OF HORIZONTAL ORIENTATION ON THE RADIATIVE PROPERTIES OF ICE CLOUDS, ADVANCES IN ATMOSPHERIC SCIENCES, 2, 20-27.  doi: 10.1007/BF03179733
    [13] Wang Angsheng, N. Fukuta, 1985: A QUANTITATIVE STUDY ON THE GROWTH LAW OF ICE CRYSTALS, ADVANCES IN ATMOSPHERIC SCIENCES, 2, 45-53.  doi: 10.1007/BF03179736
    [14] Jin Xiangze, Huang Ruixin, Yang Jiayan, 1999: Centennial Oscillations in an Ocean-ice Coupled Model, ADVANCES IN ATMOSPHERIC SCIENCES, 16, 323-342.  doi: 10.1007/s00376-999-0012-5
    [15] Cai Qiming, Kuo-Nan Liou, 1985: CALCULATION ON THE LIGHT SCATTERING FUNCTION OF HEXAGONAL ICE CRYSTALS, ADVANCES IN ATMOSPHERIC SCIENCES, 2, 446-454.  doi: 10.1007/BF02678743
    [16] K. Hussain, A. Shaukat, 1989: A Study of Supersaturation Spectra of Deposition Ice Nuclei, ADVANCES IN ATMOSPHERIC SCIENCES, 6, 247-253.  doi: 10.1007/BF02658020
    [17] Dongang LIU, Qinghua YANG, Andrei TSARAU, Yongtao HUANG, Xuewei LI, 2023: A Parameterization Scheme for Wind Wave Modules that Includes the Sea Ice Thickness in the Marginal Ice Zone, ADVANCES IN ATMOSPHERIC SCIENCES, 40, 2279-2287.  doi: 10.1007/s00376-023-2188-5
    [18] Wu Beiying, Lu Daren, 1988: SIMULATION OF THE INTENSITY AND POLARIZATION OF SKYLIGHT DURING TWILIGHT PERIOD, ADVANCES IN ATMOSPHERIC SCIENCES, 5, 15-26.  doi: 10.1007/BF02657342
    [19] Qinghua YANG, Jiping LIU, Matti LEPPÄRANTA, Qizhen SUN, Rongbin LI, Lin ZHANG, Thomas JUNG, Ruibo LEI, Zhanhai ZHANG, Ming LI, Jiechen ZHAO, Jingjing CHENG, 2016: Albedo of Coastal Landfast Sea Ice in Prydz Bay, Antarctica: Observations and Parameterization, ADVANCES IN ATMOSPHERIC SCIENCES, 33, 535-543.  doi: 10.1007/s00376-015-5114-7
    [20] ZHAO Zhen, LEI Hengchi, 2014: Aircraft Observations of Liquid and Ice in Midlatitude Mixed-Phase Clouds, ADVANCES IN ATMOSPHERIC SCIENCES, 31, 604-610.  doi: 10.1007/s00376-013-3083-2

Get Citation+

Export:  

Share Article

Manuscript History

Manuscript received: 05 June 2021
Manuscript revised: 13 September 2021
Manuscript accepted: 14 October 2021
通讯作者: 陈斌, bchen63@163.com
  • 1. 

    沈阳化工大学材料科学与工程学院 沈阳 110142

  1. 本站搜索
  2. 百度学术搜索
  3. 万方数据库搜索
  4. CNKI搜索

Ice Nucleation of Cirrus Clouds Related to the Transported Dust Layer Observed by Ground-Based Lidars over Wuhan, China

    Corresponding author: Yun HE, heyun@whu.edu.cn
  • 1. School of Electronic Information, Wuhan University, Wuhan 430072, China
  • 2. Key Laboratory of Geospace Environment and Geodesy, Ministry of Education, Wuhan 430072, China
  • 3. State Observatory for Atmospheric Remote Sensing, Wuhan 430072, China

Abstract: Cirrus clouds related to transported dust layers were identified on 22 occasions with ground-based polarization lidar from December 2012 to February 2018 over Wuhan (30.5°N, 114.4°E), China. All the events occurred in spring and winter. Cirrus clouds were mostly located above 7.6 km on top of the aloft dust layers. In-cloud relative humidity with respect to ice (RHi) derived from water vapor Raman lidar as well as from ERA5 reanalysis data were used as criteria to determine the possible ice nucleation regimes. Corresponding to the two typical cases shown, the observed events can be classified into two categories: (1) category A (3 cases), in-cloud peak RHi ≥ 150%, indicating competition between heterogeneous nucleation and homogeneous nucleation; and (2) category B (19 cases), in-cloud peak RHi < 150%, revealing that only heterogeneous nucleation was involved. Heterogeneous nucleation generally took place during instances of cirrus cloud formation in the upper troposphere when advected dust particles were present. Although accompanying cloud-top temperatures ranged from –51.9°C to –30.4°C, dust-related heterogeneous nucleation contributed to primary ice nucleation in cirrus clouds by providing ice nucleating particle concentrations on the order of 10−3 L−1 to 102 L−1. Heterogeneous nucleation and subsequent crystal growth reduced the ambient RHi to be less than 150% by consuming water vapor and thus completely inhibited homogeneous nucleation.

摘要: 利用地基偏振激光雷达在2012年12月至2018年2月期间在中国武汉观测到了(30.5°N, 114.4°E)22例受远距离输送而来沙尘影响的卷云事件。所有事件都出现在春季和冬季。观测到的卷云主要位于沙尘层的顶部,高度大于7.6 km。为了鉴别卷云的冰核化机制,我们使用了分别来自水汽拉曼激光雷达观测和ERA5再分析数据的云内冰面相对湿度(RHi)的峰值作为判据。基于此,对2个典型事件进行了个例分析。根据相同判据,将观测到受沙尘影响的卷云分为2类:(1)第一类(3例):云内峰值RHi超过150%,表明了异质核化和均质核化存在竞争关系;(2)第二类(19例):云内峰值RHi小于150%,表明仅发生异质核化。当存在平流输送而来的沙尘时,异质核化在高层卷云的形成中起主导作用。研究表明,在较低的云顶温度下(–51.9°C– –30.4°C),沙尘可提供 10−3 L−1至102 L−1量级的冰核数浓度,沙尘导致的异质核化过程对卷云的初始冰晶生成具有重要影响。除此之外,异质核化过程以及冰晶的生长会消耗水汽,使周围环境的RHi降低至150%以下,进而抑制均质核化的发生。

    • Cirrus clouds, composed of ice crystals, form primarily in the upper troposphere (generally above 7 km) with temperatures below –30°C. They are common in the atmosphere, especially in tropical and midlatitude regions (Sassen et al., 2008), and play a major role in modulating the radiation budget of the Earth–Atmosphere system (IPCC, 2013; Campbell et al., 2016). Moreover, cirrus clouds are one of the most uncertain components of general circulation models, which leads to challenges in predicting the rate and geographical pattern of climate change (Heymsfield et al., 2017).

      Heymsfield et al. (2017) reviewed past studies on cirrus clouds to summarize the current knowledge and to identify the areas where more research is needed. They proposed seven overall challenges to the understanding of the cirrus physical process (i.e., cirrus formation, evolution, and microphysical properties) and its more reliable representation in climate models, which are mainly limited by available measurement techniques. One challenge is determining how to measure/document the relative importance of homogeneous versus heterogeneous nucleation of ice crystals in cirrus clouds. Dust aerosols are regarded as the most effective and abundant ice-nucleating particles (INPs) in the atmosphere at temperatures below –15°C (Hoose and Möhler, 2012; Murray et al., 2012; Kanji et al., 2017). There are three major dust sources on Earth (Saharan desert, Middle Eastern desert mainly located in the Arabian Peninsula, and Asian deserts including the Taklimakan and Gobi) that form a dust belt in the tropical and midlatitude regions of the Northern Hemisphere, leading to a mass of dust particles emitted into the atmosphere (Liu et al., 2008). These dust particles may in turn significantly influence ice nucleation within cirrus clouds in the upper troposphere, even at temperatures below –36°C (Cziczo et al., 2013; Kuebbeler et al., 2014; Weger et al., 2018), thus providing a promising opportunity to study the possible role of heterogeneous nucleation for cirrus clouds.

      In-situ measurement is the primary method used in studying ice nucleation in cirrus clouds. This is accomplished by employing an aircraft to directly sample, or detect and thus infer, the dominant ice nucleation mechanism by analyzing: (i) the chemical composition of sampled ice residuals (IR) after sublimation; (ii) relative humidity with respect to ice (RHi); and (iii) ice crystal number concentrations (Cziczo et al., 2013). Specifically, mineral dust and metallic particles are known as the dominant IRs for heterogeneous nucleation. Heterogeneous nucleation can be triggered with RHi below 150% versus the 150%–170% required by homogeneous nucleation (Koop et al., 2000; Krämer et al., 2009). Heterogeneous nucleation usually results in a much lower ice crystal number concentration in comparison to homogeneous nucleation (Froyd et al., 2010; Kuebbeler et al., 2014; Krämer et al., 2016, 2020).

      Considering the large cost of airborne measurement, remote sensing is an affordable alternative for conducting long-term observation of ice nucleation in cirrus clouds and can even realize process-level observations for a single event. Ground-based lidar is a superior remote sensing approach that provides vertical information related to cirrus clouds, ice-nucleating particles, and atmospheric parameters. In particular, the combined observations of water vapor from Raman lidar and temperature and pressure data from radiosonde measurements can accurately derive the RHi, which is a powerful way to analyze the relative importance of two ice nucleation regimes. Moreover, the POlarization LIdar PHOtometer Networking (POLIPHON) method, proposed by Mamouri and Ansmann (2014, 2015), can quantitatively determine the height profiles of ice-nucleating particle concentration (INPC), as verified by unmanned aerial vehicle measurements (Marinou et al., 2019). Based on the POLIPHON method, the closure study performed by Ansmann et al. (2019a) in Cyprus confirmed good coincidence within one order of magnitude between ice crystal concentration and ice-nucleating particle concentration for two cirrus cloud cases. However, compared with studies using model simulations or in situ measurements (Liu et al., 2012), field experiments using ground-based lidar measurements are still insufficient to promote our understanding of dust-related ice nucleation within cirrus clouds. In particular, very little attention has been given to the East Asian region, which is seriously impacted by dust aerosols transported from the Taklimakan, Gobi, and Middle Eastern deserts.

      Wuhan (30.5°N, 114.4°E), a megacity in central China, is frequently intruded by transported dust plumes in spring and winter (He and Yi, 2015). Yin et al. (2021) reported that Asian dust is responsible for an increase of 20% in ice-containing cloud fractions over Wuhan based on the statistics of 456 thin, midlevel cloud layers within two years. He et al. (2021a) found that, compared with original pure dust, the dust ice-nucleating efficiency (i.e., onset temperature) within midlevel cloud layers does not experience a distinct change over Wuhan even after the dust was transported thousands of kilometers. Wang et al. (2020) studied the statistical characteristics of cirrus clouds based on one year of polarization lidar data over Wuhan and found a positive correlation between the cirrus occurrence frequency and dust column mass density in seasons other than summer, suggesting that heterogeneous ice formation may take place therein. Consequently, the observation of dust-related ice nucleation within cirrus clouds over Wuhan is of particular interest for understanding the relative importance of homogeneous versus heterogeneous nucleation. Furthermore, the integrated observations by multiple ground-based lidars are of great importance and can provide first-hand information on the geometrical, optical, and microphysical properties of the cirrus clouds and dust layers, as well as atmospheric parameters including the water vapor mixing ratio, RHi, and temperature, which are all helpful for understanding the heterogeneous nucleating process.

      From December 2012 to February 2018, cirrus clouds relevant to transported dust layers were observed based on multiple ground-based lidar observations over Wuhan, China. In this paper, the relevant instrumentation and methodology are first briefly introduced in section 2. Then in section 3, two case studies and a statistical study are presented. In the last section, a summary and conclusions are given.

    2.   Instrumentation and methodology
    • Since 2010, routine observations of aerosols, clouds, and precipitation have been carried out by using several ground-based lidar systems together with a sun photometer at our site on the campus of Wuhan University (Wang et al., 2020; He et al., 2021a, b, c; Yin et al., 2021). In this study, a 532 nm polarization lidar was the main instrument used to observe cirrus clouds and the optical properties of their related dust layers. In addition, a water vapor Raman lidar with a 355-nm laser was employed to derive the water vapor mixing ratio (WVMR) within the cirrus cloud (Wu and Yi, 2017). Combining the lidar-derived WVMR with temperature and pressure data from radiosondes, the relative humidity with respect to water (RHw) and ice (RHi) can be derived.

    • The polarization lidar was the main instrument used to observe the dust aerosols and cirrus clouds. This lidar system began routine operation in 2010 and was already introduced in detail by Kong and Yi (2015). The outgoing laser is 532-nm linearly polarized light with a polarization purity better than 10000:1, pulse energy of ~120 mJ, a pulse width of 6 ns, and a repetition rate of 20 Hz. The receiving unit employs a 300-mm aperture Cassegrain telescope with a field-of-view (FOV) of ~1 mrad to collect the atmospheric backscattering light. Then the received light is separated into parallel and perpendicular polarized components by using double cascaded cubic polarizing beamsplitters enabling sufficient suppression of crosstalk between two receiving channels. Hamamatsu 5783P Photomultiplier tubes (PMTs) were used as the detector with a two-channel Licel TR40-160 as the signal digitizer. The gain ratio between two orthogonally polarized channels is calibrated using the Δ90° method (Freudenthaler et al., 2009). The lidar-derived volume depolarization lidar $ {\delta }_{\mathrm{v}} $ indicates the nonspherical shape of backscattering objects to effectively distinguish dust particles and ice crystals (Wang et al., 2020) from the liquid droplets and other types of spherical aerosols.

      The aerosol extinction coefficient $ \alpha $ and backscatter coefficient $ \beta $ were calculated with the method of Fernald (1984). The lidar ratio was set to 45 sr for the background/dust aerosols (He et al., 2021a). Considering $ {\delta }_{\mathrm{v}} $ value is contributed by both atmospheric molecules and aerosols and thus depends on the ratio of molecular to particulate backscatter, it is necessary to extract the individual contribution of aerosol (i.e., particle depolarization ratio $ {\delta }_{\mathrm{p}} $) to depict its microphysical properties independent of the aerosol loadings. The particle depolarization ratio can be calculated using the following equation (Freudenthaler et al., 2009):

      where the molecular depolarization ratio $ {\delta }_{\mathrm{m}} $, which depends on the specification of the narrowband filter used in the receiving channels, is 0.004 for our lidar system and $ R $ is aerosol backscattering ratio. Note that the assumption of constant lidar ratio in the Fernald method would lead to errors in retrieved $ \beta $, which would be subsequently propagated to $ {\delta }_{\mathrm{p}} $. Particle depolarization ratio is strongly sensitive to the nonsphericity of particles, while lidar ratio is highly sensitive to the absorption characteristics of particles (Wiegner et al., 2009). Lidar ratio and particle depolarization ratio are controlled by different primary factors; some secondary factors (i.e., particle size) may influence them both and result in errors in retrieved $ {\delta }_{\mathrm{p}} $ when using a single-wavelength polarization lidar.

      The backscatter coefficient for the dust component $ {\beta }_{\mathrm{d}} $ can be extracted with the following equation (Tesche et al., 2009; Mamouri and Ansmann, 2014):

      where the dust particle depolarization ratio $ {\delta }_{\mathrm{d}}=0.31 $ and the non-dust particle depolarization ratio $ {\delta }_{\mathrm{n}\mathrm{d}}=0.05 $ (Sakai et al., 2010; Mamouri and Ansmann, 2014). The dust backscatter coefficient $ {\beta }_{\mathrm{d}} $ was then multiplied by the dust lidar ratio 45 sr to obtain the dust extinction coefficient $ {\alpha }_{\mathrm{d}} $, which was then input to the POLIPHON method as described in section 2.4. The typical lidar ratio at 532 nm for Asian dust was independently measured to be 45±7 sr (Hu et al., 2020) and 47±4 sr (Peng et al., 2021) by pure rotational Raman lidar (Ångström-relationship assumption can be ignored theoretically), suggesting the assumed dust lidar ratio (45 sr) is reasonable in this study and leads to acceptable relative uncertainties (as listed in Table 1).

      ParameterUncertainty
      Dust backscatter coefficient $ {\beta }_{\mathrm{d}} $10%
      Dust extinction coefficient $ {\alpha }_{\mathrm{d}} $20%
      Volume depolarization ratio $ {\delta }_{\mathrm{v}} $5%
      Particle depolarization ratio $ {\delta }_{\mathrm{p}} $5%–10%
      Dust mass concentration ${M }_{\mathrm{d} }$20%–30%
      Dust surface area concentration $ {S}_{\mathrm{d}} $30%–40%
      INP number concentration $ {n}_{\mathrm{I}\mathrm{N}\mathrm{P}} $50%–500%

      Table 1.  Estimated uncertainties in the dust-related optical parameters and cloud-relevant properties (Kafle and Coulter, 2013; Ansmann et al., 2019b).

    • A Raman lidar system emits a 355 nm laser beam with a power of ~6 W and detects the Raman backscatter from water vapor at 407 nm and nitrogen molecules at 387 nm by using a 200-mm aperture Cassegrain telescope. The WVMR can be derived using the method from Whiteman et al. (1992), with the calibration constant determined by using the WVMR data from radiosondes launched at 0000 UTC or 1200 UTC from the Wuhan Weather Station (~24 km from our site) (Wu and Yi, 2017). To ensure a sufficient signal-to-noise ratio (SNR) at cirrus cloud altitudes, hourly integrated data were applied to derive the WVMR in this study. If the telescope is upgraded with an aperture of 600 mm and the laser is upgraded with a power of 9 W, integrated time can be substantially promoted from one hour to five minutes which is more favorable for realizing the process-level observation of cirrus cloud evolution. A similar scheme was used by Leblanc et al. (2012) for the long-term monitoring of water vapor at the upper troposphere and lower stratosphere.

      Combining the lidar-derived WVMR profile with the atmospheric pressure and temperature profiles from radiosonde measurements, RHw and RHi profiles can be calculated (Murphy and Koop, 2005). The related calculations are given below. The relationship between the vapor pressure $ E $ and WVMR can be given by (Bolton, 1980):

      where r denotes the WVMR measured by Raman lidar and $ P $ is the atmospheric pressure provided by radiosonde measurement. The saturated water vapor pressure with respect to liquid water $ {E}_{\mathrm{w}} $ and ice water $ {E}_{\mathrm{i}} $ can be calculated, respectively, using the empirical formulas as below (Murray et al., 1967; Bolton, 1980):

      where T is the temperature provided by radiosonde measurement. Therefore, RHw and RHi can be obtained, respectively, by the following.

    • ERA5 is the fifth-generation atmospheric reanalysis product produced by the European Centre for Medium-Range Weather Forecasts (ECMWF) and was released publicly in 2016 (Hersbach et al., 2020). It can provide hourly datasets of atmospheric parameters (e.g., relative humidity, temperature, three components of wind speed, horizontal divergence, fraction of cloud cover, ozone mass mixing ratio, and cloud water content) on 37 pressure levels (from 1000 hPa to 1 hPa) with an enhanced horizontal resolution of 31 km (0.25°) from 1979 onwards. In this study, the grid cell containing Wuhan (30.5°N, 114.4°E) was selected to provide the necessary meteorological parameters, including RHw, temperature (T), the eastward (U) and northward (V) components of wind speed, and vertical velocity (W). Benefiting from the one-hour temporal resolution, ERA5 data can cover the time intervals between the launches of the radiosondes (12 h). ERA5 data are employed to provide the meteorological parameters in this study. Due to the insufficient vertical resolution of ERA5, linear interpolation was conducted on the temperature profile to obtain the required cloud top temperature.

      A single cirrus cloud usually exists in lidar FOVs for only tens of minutes, which is less than the one-hour temporal resolution of ERA5. Therefore, it is assumed in this work that the one-hour meteorological parameters from ERA5 that overlap in time with the presence of a cirrus cloud are representative of the meteorological conditions during the entire lifetime of the cirrus cloud. For the water vapor condition (RH), the influences of waves or turbulence on the variation of RH values (Kärcher et al., 2014) over the short time scale are not considered. Since most of the cases lasted for more than one hour, attributing to the presence of a series of adjacent cirrus clouds, it is substantially reasonable to employ one-hour RH as the vapor condition during the cloud lifetime. In addition, in-cloud peak RH is defined as the maximum RH value within the vertical extension of the cirrus cloud.

    • The dust mass concentration $ {M}_{\mathrm{d}} $, dust surface area concentration $ {S}_{\mathrm{d}} $, and dust-related INP concentration $ {n}_{\mathrm{I}\mathrm{N}\mathrm{P}} $ can be retrieved by the POLIPHON method (Mamouri and Ansmann, 2014, 2015; Ansmann et al., 2019a; Marinou et al., 2019). Dust particles pertinent to cirrus clouds appear in the upper troposphere and hence are less influenced by anthropogenic aerosols near the surface. Therefore, we used the pure-dust conversion factors (${c}_{\mathrm{v},\mathrm{ }\mathrm{d}}$ and ${c}_{\mathrm{s},\mathrm{ }\mathrm{d}}$) obtained at the Lanzhou SACOL site by Ansmann et al. (2019b). All the uncertainties for the derived parameters are listed in Table 1. The applied conversion factors and constants are given in Table 2.

      ParametervalueUnitReference
      $ {c}_{\mathrm{v},\mathrm{d}} $0.77×10−12Mm(Ansmann et al., 2019b)
      $ {c}_{\mathrm{s},\mathrm{d}} $3.1×10−12Mm m2 cm−3(Ansmann et al., 2019b)
      Dust lidar ratio LR45sr(He et al., 2021a)
      Dust density $ {\mathrm{\rho }}_{\mathrm{d}} $2.6g cm−3(Wagner et al., 2009)

      Table 2.  Applied values of the conversion parameters and constants required in the POLIPHON retrieval.

      The specific calculation processes of dust mass concentration $ {M}_{\mathrm{d}} $ and dust-related INP concentration $ {n}_{\mathrm{I}\mathrm{N}\mathrm{P}} $ are given below. The dust mass concentration $ {M}_{\mathrm{d}} $ is retrieved using the following equation:

      where $ {\rho }_{\mathrm{d}} $ is the dust particle density, $ {\alpha }_{\mathrm{d}} $ is the dust extinction coefficient (see section 2.1), and $ {c}_{\mathrm{v},\mathrm{d}} $ is the extinction-to-volume conversion factor. At the cirrus cloud altitudes, we only consider INP concentration for deposition nucleation mode and pore condensation and freezing using the parameterization scheme U17-D (Ullrich et al., 2017). The ice nucleation active surface site (INAS) density $ {n}_{\mathrm{s}} $ can be retrieved with the following equation:

      where $ {S}_{\mathrm{i}} $ is ice saturation ratio and $ T $ is the temperature, both obtained from the measurement of the most recently launched radiosonde. The surface area concentration of dust $ {S}_{\mathrm{d}} $ can be obtained by:

      where $ {c}_{\mathrm{s},\mathrm{d}} $ is the extinction-to-surface area conversion factor (Ansmann et al., 2019b). Finally, we can calculate the INPC for deposition freezing mode $ {n}_{\mathrm{I}\mathrm{N}\mathrm{P}} $ with the following equation (Marinou et al., 2019):

    3.   Observational results
    • From December 2012 to February 2018, 22 cases of cirrus clouds related to transported dust layers were identified based on polarization lidar observations over Wuhan, China. Water vapor Raman lidar began to operate in November 2017, thus lidar-derived WVMR and RH data were available for only a few cases. The criteria for cirrus clouds were described in detail by Wang et al., (2020). The criteria for an aloft dust layer are as follows: (i) $ {\delta }_{\mathrm{p}} $ value is larger than 0.1 throughout the layer; (ii) layer thickness exceeds 0.3 km; (iii) layer base is above the planetary boundary layer (He et al., 2021a). As long as any edge (top or base) of a cirrus cloud layer was in contact with an edge of a dust layer for >10 min, we determine that the formation or evolution of this cirrus cloud layer was affected by the dust layer (i.e., the cirrus cloud is related to the dust layer). Note that a single bulk cloud or a series of very adjacent clouds near to the same dust layer would be considered a single case. Finally, all the selected cases were checked individually by eye. In the following, we first provide two typical case studies (both with lidar-derived WVMR, temperature, and RH data available) and then some brief statistics.

    • Dust aerosols were observed to intrude into Wuhan on 12 January 2018. Figures 1b and 1d show the time-height contour plots (1-min/30-m resolution) of the range-corrected signal and volume depolarization ratio $ {\delta }_{\mathrm{v}} $ during 2030–2400 LST (Local Standard Time = UTC + 8 h) measured by the polarization lidar. The corresponding height profiles of the RHw and RHi, temperature, horizontal wind speed, and vertical velocity obtained from ERA5 reanalysis data during 2200–2300 LST are shown in Figs. 1a and 1c. Two dust aerosol layers were identified by the volume depolarization ratio (Fig. 1d). The lower layer located from the surface to approximately 3.6 km showed relatively small $ {\delta }_{\mathrm{v}} $ values of < 0.1, while the upper layer that appeared at altitudes of 7–9 km had larger $ {\delta }_{\mathrm{v}} $ values up to 0.16.

      Figure 1.  Time-height contour plots (1-min/30-m resolution) of (b) range-corrected signal and (d) volume depolarization ratio measured by a 532 nm polarization lidar on 12 January 2018 during 2030–2400 LST. Height profiles of (a) temperature and relative humidity (RH with respect to water and ice respectively), and (c) wind speed of the U (westerly), V (southerly), and W (vertical velocity) components obtained from ECMWF ERA5 reanalysis data during 2200–2300 LST. The horizontal dashed lines indicate cloud top level with a temperature of –47.8°C. The vertical dotted line indicates 100% relative humidity. Several cirrus clouds (8.6–9.3 km) appeared on the top of the dust layer.

      As seen from plots of the volume depolarization ratio, three optically transparent cirrus clouds, with $ {\delta }_{\mathrm{v}} $ values larger than 0.2, began to appear on top of the dust layer (above ~8.7 km) at ~2100 LST. Each cirrus cloud existed for less than half an hour, which is very different from stratiform midlevel cloud layers that typically have longer durations (He et al., 2021a). The cloud top temperature was –47.8°C as denoted by the horizontal dashed lines. The base of the first cirrus cloud (2100–2130 LST) was embedded in the dust layer below, hinting that the ice crystals within the cloud were probably related to heterogeneous nucleation triggered by dust particles. In general, only heterogeneous nucleation can take place at temperatures ranging from –36°C to 0°C. DeMott et al. (2003) found that homogeneous freezing requires a somewhat higher ice supersaturation as temperature decreases. Therefore, even at temperatures below –36°C, homogeneous nucleation cannot be initiated unless the RHi is strongly supersaturated up to 150% (Koop et al., 2000; Ansmann et al., 2019a). To confirm the specific nucleating regime in this case, the RHi condition within the cirrus cloud needs to be checked further.

      Figure 2 shows the profiles of RHw, RHi, WVMR, saturated WVMR, and temperature derived by the combination of water vapor Raman lidar and radiosonde measurements during 2200–2300 LST. As a comparison, the simultaneous RHw and RHi profiles obtained from the observation of water vapor Raman lidar are also given here. As seen in Fig. 2a, the RHw and RHi values from lidar measurements and ERA5 data agree well with each other at altitudes below 8.0 km. Above 8.0 km, both lidar measurements and ERA5 data show an enhancement of RH (both RHw and RHi), which is attributed to the occurrence of cirrus clouds. The RH values from lidar measurements are consistently smaller than those from ERA5 data. Nevertheless, the peak values of RHi provided by the two methods both exceed 150% (i.e., 183.2% from lidar and 157.4% from ERA5) at the altitudes where clouds are located (8.6–9.8 km), indicating that homogeneous nucleation may also participate in ice formation. Consequently, we can conclude that the ice-nucleating pattern here is the competition between heterogeneous and homogeneous nucleation. This situation was found only three times during the whole observation period, as seen in subsection 3.3. In addition, it is found that the lidar-derived peak RHi during 2000–2100 LST, before cirrus clouds appeared in the lidar FOV, is as high as 133.5% (not shown here). This provides a great RHi condition for the possible initiation of subsequent homogeneous nucleation. As a result of competition, the homogeneous nucleation process is generally suppressed because heterogeneous nucleation advances the onset of ice nucleation and consumes the available water vapor (Kärcher and Lohmann, 2003; Barahona and Nenes, 2009). Based on an investigation using the Community Atmospheric Model version 5 (CAM5), Liu et al. (2012) reported that dust-related heterogeneous nucleation reduces the occurrence of homogeneous nucleation and thus the ice crystal number concentration for upper atmospheric cirrus clouds in the Northern Hemisphere.

      Figure 2.  Profiles of (a) relative humidity (RH with respect to water and ice) derived by lidar and ERA5 reanalysis data and (b) water vapor mixing ratio and saturated water vapor mixing ratio derived by lidar during 2100–2200 LST, and (c) temperature measured by radiosonde launched at 2000 LST on 12 January 2018.

      The dust-related INPC, which determines the significance of heterogeneous nucleation (Kärcher and Lohmann, 2003), can be obtained quantitatively by the POLIPHON method (Mamouri and Ansmann, 2014, 2015). Figure 3 shows profiles of the total (dust + non-dust) and dust extinction coefficient, total (dust + non-dust) and dust backscatter coefficient, volume depolarization ratio $ {\delta }_{\mathrm{v}} $, and particle depolarization ratio $ {\delta }_{\mathrm{p}} $ for a cloud-free period during 2000–2030 LST measured by polarization lidar. Considering that the vertical distribution of the dust layer at 7–9 km was uniform over time, we assume that the dust optical properties during these 30 minutes are representative of the entire event. As seen in Fig. 4a, dust extinction predominantly contributes to the total aerosol extinction at altitudes of 7.0–9.8 km with a mean of 10.4 Mm−1, indicating the dominance of pure dust particles within this layer. Moreover, the presence of a pure dust layer is also verified by $ {\delta }_{\mathrm{p}} $ values larger than 0.3 throughout the layer, as shown in Fig. 3d (Sakai et al., 2010).

      Figure 3.  Profiles of (a) dust and total (dust + non-dust) extinction coefficient, (b) dust and total (dust + non-dust) backscatter coefficient, (c) volume depolarization ratio $ {\delta }_{\mathrm{v}} $, and (d) particle depolarization ratio $ {\delta }_{\mathrm{p}} $ derived by the 532 nm polarization lidar during 2000–2030 LST on 12 January 2018. Horizontal error bars denote the relative errors of each parameter.

      Figure 4.  Profiles of (a) dust mass concentration $ {M}_{\mathrm{d}} $, (b) dust surface area concentration $ {S}_{\mathrm{d}} $, and (c) ice-nucleating particle concentration derived by the POLIPHON method using the U17-D parameterization scheme during 2000–2030 LST on 12 January 2018. The horizontal dashed lines denote the temperature height levels of –35°C and –47.8°C (cloud top temperature). $ {S}_{\mathrm{i}} $ represents ice saturation ratio.

      Figures 4a and 4b show the concentration profiles of the dust mass and surface area during the same period as Fig. 3. The layer-averaged dust mass concentration is 20.8 μg m−3 with a peak of 49.5 μg m−3. The layer-averaged surface area concentration is 3.2 × 10−11 m2 cm−3. Considering that the dust layer mostly appeared above the –35°C isotherm and liquid water was not observed before cirrus formation, deposition nucleation and pore condensation and freezing mechanism (PCF) should be exclusively responsible for the ice crystal formation, rather than immersion/condensation and contact nucleation (Marcolli, 2014; Campbell and Christenson, 2018; David et al., 2019). In Fig. 4c, the mean (maximum) INPCs within the dust layer are estimated to be 0.02 L−1 (0.1 L−1), 0.47 L−1 (2.3 L−1), and 4.7 L−1 (27.0 L−1) for three assumed constant ice saturation ratio ($ {S}_{\mathrm{i}} $) values of 1.15, 1.25, and 1.35, respectively, based on the U17-D parameterization scheme (Ullrich et al., 2017). Note that Ullrich et al. (2017) also mentioned PCF is the responsible mechanism for part of their parametrizations, especially in the temperature range between –43°C and –23°C. The maximum INPC values in this case are nearly on the same order of magnitude as those of 0.23 L−1, 0.94 L−1, and 9.3 L−1 (respectively corresponding to $ {S}_{i} $ values of 1.15, 1.25, and 1.35) observed in a Saharan-dust-related thin cirrus case over Limassol (34.7°N, 33.0°E), Cyprus, which was generally consistent with the ice crystal concentrations in the adjacent regions (Ansmann et al., 2019a). The INPC profile calculated with the $ {S}_{\mathrm{i}} $ profile (light blue in Fig. 2a) measured by water vapor Raman lidar is shown by the gray line in Fig. 4c. Attributing to the significant enhancement of RHi at altitudes of 9–10 km, the actual INPC around the cloud top reached an order of magnitude of 103–104 L−1.

    • Another dust-intrusion event was captured by ground-based lidars over Wuhan on 7 February 2018. Figures 5b and 5d show the time-height contour plots (1-min/30-m resolution) of the range-corrected signal and volume depolarization ratio $ {\delta }_{\mathrm{v}} $ during 0000–0200 LST measured by polarization lidar. The corresponding height profiles of the RHw, RHi, temperature, horizontal wind speed, and vertical velocity obtained from ERA5 reanalysis data during 0000–0100 LST are shown in Figs. 5a and 5c. Similar to the first case, double-layer dust plumes appeared at altitudes of 0–4.8 km and 7–10 km, respectively, as seen from the volume depolarization ratio of approximately 0.1 (Fig. 5d).

      Figure 5.  Time-height contour plots (1 min by 30 m resolution) of (b) range-corrected signal and (d) volume depolarization ratio measured by a 532 nm polarization lidar for 0000–0200 LST 7 February 2018. Height profiles of (a) temperature and relative humidity (RH with respect to water and ice, respectively), and (c) wind speed of the U (westerly), V (southerly), and W (vertical velocity) components obtained from ECMWF ERA5 reanalysis data during 0000–0100 LST. The horizontal dashed lines indicate cloud top level with a temperature of –51.9°C. The vertical dotted line indicates 100% relative humidity. Several cirrus clouds (9.0–10.1 km) appeared embedded in the upper part of the dust layer.

      Above 9 km, several short-lived cirrus clouds were observed to embed in the upper dust layer (7–10 km) since 0012 LST (Fig. 5d). The cloud top temperature was –51.9°C, as denoted by the horizontal dashed lines. Within approximately 80 minutes, each cloud top overlapped with the top edges of the dust layer. Therefore, the ice formation within these cirrus clouds is considered to be relevant to the dust-related heterogeneous nucleation. To verify whether homogeneous nucleation was involved, the RHi profiles during 0000–0100 LST obtained from both lidar detection and ERA5 reanalysis data are given in Fig. 6a. The corresponding profiles of the water vapor mixing ratio from water vapor Raman lidar and temperature from radiosonde are shown in Figs. 6b and 6c. The simultaneous RHi values from lidar measurements and ERA5 data are in accordance with each other well below 7.5 km. An RHi enhancement appeared above 8 km, reflecting the presence of cirrus clouds. Different from the first case, the peak ice saturation ratio (i.e., 124.2% from lidar and 133.9% from ERA5) at cloud altitudes (9–10 km) does not exceed 150%, which is insufficient to initiate homogeneous nucleation, indicating that only heterogeneous nucleation is likely to be responsible for in-cloud ice formation. Before the cirrus clouds appeared in the lidar FOV, the lidar-derived peak RHi during 2300–2400 LST on 6 February was only 74.2% (not shown here), which is much less than that of 133.5%, as observed in the first case, indicating that the RHi condition for the subsequent occurrence of homogeneous nucleation is insufficient. In addition, the invariably low vertical velocity (approximately 0 m s−1, see Fig. 5c) throughout the vertical extension of the cloud region also supports the probable absence of homogeneous nucleation, since a strong vertical velocity, such as >1 m s−1, is essential to the onset of homogeneous freezing (Spice et al., 1999). However, there is still a possibility that homogeneous nucleation had previously occurred and was subsequently inhibited by concomitant heterogeneous nucleation with the consumption of water vapor and thus the decreased RHi to be <150%. This situation was frequently observed during the whole observation period, as discussed in subsection 3.3.

      Figure 6.  Profiles of (a) relative humidity (RH with respect to water and ice) derived by lidar and ERA5 reanalysis data and (b) water vapor mixing ratio and saturated water vapor mixing ratio derived by lidar during 0000–0100 LST on 7 February 2018, and (c) temperature profile measured by radiosonde launched at 2000 LST on 6 February 2018.

      Figure 7 shows profiles of the total (dust + non-dust) and dust extinction coefficient, total (dust + non-dust) and dust backscatter coefficient, volume depolarization ratio, and particle depolarization ratio for a cloud-free period during 0140–0150 LST measured by polarization lidar. It is assumed that this period depicts the representative optical properties of dust aerosols for this case. As seen in Fig. 7a, dust extinction contributes most of the total aerosol extinction at altitudes of 7–10 km, with a mean of 6.1 Mm−1. The $ {\delta }_{\mathrm{p}} $ values invariably exceed 0.3 above 7.7 km, suggesting that the upper part of the dust layer is composed of pure dust particles, while $ {\delta }_{\mathrm{p}} $ values decrease to 0.2–0.3 at 7–7.8 km (see Fig. 7d), revealing that mixed dust (i.e., dust particles mixed with other urban aerosols that usually show a $ {\delta }_{\mathrm{p}} $ value of <0.3) is dominant in the lower part of the dust layer (Mamouri and Ansmann, 2014, 2015; He et al., 2021b; Yin et al., 2021).

      Figure 7.  Profiles of (a) dust and total (dust + non-dust) extinction coefficient, (b) dust and total (dust + non-dust) backscatter coefficient, (c) volume depolarization ratio $ {\delta }_{\mathrm{v}} $, and (d) particle depolarization ratio $ {\delta }_{\mathrm{p}} $ derived by the 532 nm polarization lidar during 2000–2030 LST on 7 February 2018. Horizontal error bars denote the relative errors of each parameter.

      Figure 8 shows the concentration profiles of the dust mass, surface area, and ice nucleating particles during the same period as Fig. 9. The layer-averaged dust mass concentration is 12.3 μg m−3, with a peak of 17.5 μg m−3 appearing at ~8.6 km. The layer-averaged surface area concentrations are 1.6 × 10−11 m2 cm−3. For the study of heterogeneous nucleation within cirrus clouds, the possibility of immersion/condensation freezing and contact freezing can be completely excluded because liquid water was not observed previously, as seen from the volume depolarization ratio in Fig. 5b (Hoffmann et al., 2013). At cirrus cloud altitudes (9–10 km), dust particles probably initiate ice nucleation through deposition freezing (Ansmann et al., 2019a). Note that preactivation can be another process responsible for ice crystal formation here, because pore ice and water can survive even below ice saturation and may subsequently remerge once the ice saturation ratio exceeds 100% again (Marcolli, 2017, 2020). Deposition nucleation was considered in the INPC calculation using the parameterization scheme U17-D (see Fig. 8c). The mean (maximum) INPCs above 7.5 km are estimated to be 0.8 L−1 (2.6 L−1), 27.2 L−1 (102.3 L−1), and 374.9 L−1 (1484.9 L−1) for three assumed constant ice saturation ratio ($ {S}_{i} $) values of 1.15, 1.25, and 1.35, respectively, based on the U17-D parameterization scheme (Ullrich et al., 2017). The INPC profile calculated with the $ {S}_{\mathrm{i}} $ profile (light blue in Fig. 6a) measured by water vapor Raman lidar is shown by the gray line in Fig. 8c. The actual INPCs generally varied between the INPC values calculated from the fixed $ {S}_{\mathrm{i}} $ of 1.15 and 1.35.

      Figure 8.  Profiles of (a) dust mass concentration $ {M}_{\mathrm{d}} $, (b) dust surface area concentration $ {S}_{\mathrm{d}} $, and (c) ice-nucleating particle concentration derived by the POLIPHON method during 0140–0150 LST on 7 February 2018. The horizontal dashed lines denote the temperature height levels of –35°C and –51.9°C (cloud top temperature). The INPC profiles obtained by the U17-D parameterization schemes are given. $ {S}_{\mathrm{i}} $ represents ice saturation ratio.

      Figure 9.  Overview of cloud top temperature and in-cloud peak RHi (obtained by ERA5 reanalysis data) of cirrus clouds for each case. The red circles (CTT < –36°C and also RHi ≥ 150%) indicate the cases showing competition between heterogeneous and homogeneous nucleation. The dark blue circles (100% < RHi < 150%) indicate that only heterogeneous nucleation takes place.

    • During the period from December 2012 to February 2018, 22 cases of cirrus clouds (a single bulk cloud or a series of very adjacent clouds) relevant to transported dust layers were identified over Wuhan, China. All the events occurred in spring and winter when Wuhan was frequently intruded by dust aerosols. In addition to the two case studies shown above, the remaining 20 occasions were also examined. The detailed statistics, including the local time, cloud top height and temperature, dust extinction coefficient, dust mass concentration, volume depolarization ratio, particle depolarization ratio, in-cloud peak RHw and RHi, maximum INPCs for three assumed constant $ {S}_{\mathrm{i}} $, and actual upper limited INPC (calculated with the CTT, cloud top $ {S}_{\mathrm{d}} $, and in-cloud peak RHi) are listed in Table 3.

      DateTime (LST)CTH
      (km)
      CTT
      (°C)
      Peak RHw (%)Peak RHi (%)VDRPDRDust Ext. (Mm−1)Dust mass (μg m−3)U17D
      (L−1) (Si=1.15)
      U17D
      (L−1) (Si=1.25)
      U17D
      (L−1) (Si=1.35)
      Peak INPC (L−1)
      201212310730−11009.9−46.3100.3153.90.040.374.318.630.0240.5715.9034139.180
      201301140527−07208.1−37.3101.9138.80.030.293.396.790.0010.0130.0850.051
      201312240510−07379.3−45.396.2140.20.050.375.3610.720.0080.1321.044NAN
      201312250445−08009.2−43.597.0140.10.180.4242.8085.690.0040.0500.32419.333
      201412201627−17187.7−38.576.1107.50.050.2310.2420.510.0160.3523.405NAN
      201503122102−22078.8−42.888.1120.20.070.2415.1530.320.0010.0100.0590.045
      201503220405−06508.6−36.198.8141.30.020.212.805.610.0010.0110.0750.278
      201603110100−02108.5−39.480.7111.20.100.3521.3642.760.0030.0470.3730.105
      201603282050−23159.1−36.4102.1154.00.070.3111.2622.55NAN0.0030.019NAN
      201611282055−24009.4−41.591.4140.80.030.283.517.030.54417.639224.9575.215
      201704130810−10008.9−34.091.6129.30.030.254.789.570.0010.0130.0790.016
      201704140220−04459.0−38.182.8123.30.030.234.519.020.0080.1501.3190.038
      201704230545−07577.8−30.492.7122.80.05NAN2.915.83NANNANNANNAN
      201704272335−23579.7−39.781.2109.40.030.303.577.160.0320.7958.413NAN
      201704280000−06209.0−33.495.8129.20.030.324.529.06NANNANNANNAN
      1746−18009.4−38.578.3116.10.030.282.114.230.0892.57430.2280.042
      201705172220−230810.0−42.788.4130.60.030.303.136.260.1685.07461.5672.521
      201712092340−24008.3−30.493.0127.90.050.247.2714.56NANNANNANNAN
      201712100300−04558.1−34.493.1126.00.090.3416.1032.24NANNANNANNAN
      201801090557−07147.6−36.183.6111.40.030.184.869.730.0010.0070.041NAN
      201801122055−23289.7−47.8100.0157.40.070.3210.4020.820.0812.31526.96618739.800
      201802070013−013210.2−51.983.3133.90.050.336.1312.272.647102.3141484.9306576.980

      Table 3.  Information on 22 cirrus cloud layer cases with dust-related ice nucleation during December 2012–February 2018 at Wuhan (30.5°N, 114.4°E) obtained from polarization lidar observations and ECMWF ERA5 reanalysis data. Cloud top height (CTH) and the dust-layer-averaged values of dust extinction coefficient (Dust Ext.), dust mass concentration (Dust mass), volume depolarization ratio (VDR), and particle depolarization ratio (PDR) are calculated based on the polarization lidar data. Cloud top temperature (CTT), and respective in-cloud peak RHw and RHi are given by ERA5 reanalysis data. $ {S}_{\mathrm{i}} $ represents the ice saturation ratio. The maximum INPCs within the dust layer for three assumed constant $ {S}_{\mathrm{i}} $ (1.15, 1.25, and 1.35) are calculated by the POLIPHON method using U17-D parameterization schemes. Peak INPC is calculated with the CTT, cloud top $ {S}_{\mathrm{d}} $, and in-cloud peak RHi, indicating the upper limit INPC in the actual atmosphere.

      The studied cirrus clouds were mostly located on top of the aloft dust layers. The dust-layer-averaged values of the dust extinction coefficient, dust mass concentration, volume depolarization ratio, and particle depolarization ratio were 2.1–42.8 Mm−1, 4.2–85.7 µg m−3, 0.03–0.18, and 0.18–0.42, respectively. In total, 11 cases showed a $ {\delta }_{\mathrm{p}} $ value of ≥ 0.3, indicating the presence of pure dust particles (Sakai et al., 2010). Kojima et al. (2006) confirmed that these uncoated dust particles can still exist in the free troposphere after thousands of kilometers of transport, which can be of particular importance for cirrus cloud formation in the atmosphere. The INPC values mostly varied between the order of magnitude of 10−3 L−1 and 102 L−1. DeMott et al. (2003) also measured a similar cirrus-active INPC of < 30 L−1 in western America. These INPC values are consistent with the typical ice crystal number concentration of 1–100 L−1 caused by heterogeneous freezing (Cziczo et al., 2013).

      Figure 9 shows the overview of cloud top temperature (CTT) and in-cloud peak RHi obtained by ERA5 reanalysis data for each cirrus cloud case. It should be mentioned that the lidar-derived RHi data are available only for the last five cases listed in Table 3. We compared the RHi profiles from lidar measurements and ERA5 data and found that they generally agree with each other well. Hence, the RHi values derived from ERA5 data were used in the statistical study (Immler et al., 2008; Gamage et al., 2020). The RH peak, indicating the occurrence of cloud, from ERA5 and Raman lidar appeared at almost the same altitudes with similar values. Benefiting from the much higher vertical resolution, Raman lidar observation showed more details of RH vertical variation compared with ERA5. For cases with CTT < –36°C and RHi ≥ 150%, we infer that there is competition between heterogeneous and homogeneous nucleation within the cirrus cloud (Cziczo et al., 2013). The respective proportion (or importance) of these two nucleating regimes depends on the velocity of updraft within the cloud (Kärcher and Lohmann, 2003). Considering the large amount of dust INP supplied here, a stronger updraft on the order of 1 m s−1 is necessary to favor the transition from heterogeneous to homogeneous freezing (Dierens, 2003). However, the necessary high-temporal-resolution vertical velocity data (e.g., from Doppler lidar) are absent and should be a focus of future research. Similarly, using the RHi measured during the INCA campaign, Haag et al. (2003) also reported that in-situ cirrus clouds are probably formed from the combination of heterogeneous and homogeneous nucleation in midlatitude regions of the Northern Hemisphere. Once RHi reduces to the range of 100%–150%, we infer that only heterogeneous nucleation takes place in cirrus cloud formation. As a result, the observed events can be classified into two categories corresponding to the two case studies: (1) category A (3 cases): competition between heterogeneous and homogeneous nucleation (red circles in Fig. 9); (2) category B (19 cases): only heterogeneous nucleation takes place (dark blue circles in Fig. 9).

      To evaluate the possible differences in lidar-derived dust optical properties between the two case types, a Mann-Whitney U test was performed for dust extinction coefficient and particle depolarization ratio (Mann and Whitney, 1947). For dust extinction coefficient, both U1 for category A (=37) and U2 for category B (=20) are larger than the critical value U0.05 (=7). For particle depolarization ratio, both U1 for category A (=41) and U2 for category B (=13) are also larger than the critical value U0.05 (=7). Therefore, there are no significant differences in the dust extinction coefficient and particle depolarization ratio between the two categories. This suggests that the optical properties of the dust aerosols in the vicinity of the investigated cirrus clouds are possibly not responsible for the distinction between the two case types. Hence, we conjecture that ambient ice saturation ratio and temperature should be the major factors leading to the two case types, and sufficient INPCs were provided in all the cases for catalyzing heterogeneous nucleation during the dust intrusion period.

      Overall, the significance of heterogeneous nucleation within the cirrus cloud triggered by transported dust particles was verified in the upper troposphere over Wuhan. Although accompanying cloud top temperatures were low (–51.9°C to –30.4°C), heterogeneous nucleation can still predominantly contribute to the total ice nucleation and even completely inhibit the homogeneous nucleation process. This is in accordance with the earlier finding that if the upper-tropospheric INPCs reach up to the same magnitude as the lower troposphere and the vertical velocity is as weak as < 0.3 m s−1, homogeneous nucleation is almost negligible (DeMott et al., 1994; Spice et al., 1999). Based on the analysis of the cirrus residue composition, relative humidity, and cirrus particle concentration measurements, Froyd et al. (2013) inferred that heterogeneous nucleation is a dominant cirrus formation mechanism for midlatitude, subtropical, and tropical regions. This effect can strongly influence the microphysical properties of cirrus clouds because heterogeneous nucleation generally leads to a low number concentration of ice crystals compared with homogeneous nucleation (Froyd et al., 2010) and thus induces a net cloud forcing of –0.4 W m−2 (Liu et al., 2012) or –2.0 W m−2 (Lohmann et al., 2008) as estimated by the atmosphere model. In addition, these dust-related cirrus clouds were also reported to show modified optical and microphysical properties, including decreases in cloud optical depth, cloud ice water path, and cloud effective particle radius (Huang et al., 2006; Wang et al., 2015; Pan et al., 2019).

    4.   Summary and conclusions
    • During the period from December 2012 to February 2018, cirrus clouds relevant to transported dust layers were observed over Wuhan, China. Based on polarization lidar observations assisted by water vapor Raman lidar measurements, 22 events were captured during spring and winter. Cirrus clouds were mostly located on top of the aloft dust layers. The optical and microphysical properties of the dust layer and dust-INP-related parameters were derived based on the POLIPHON method. ECMWF ERA5 reanalysis data were employed to provide meteorological parameters, including temperature, relative humidity, horizontal wind speed, and wind direction. The mean values of the dust extinction coefficient, dust mass concentration, volume depolarization ratio, and particle depolarization ratio were 2.1–42.8 Mm−1, 4.2–85.7 µg m−3, 0.03–0.18, and 0.18–0.42, respectively. The INPC values mostly varied between the order of magnitude of 10−3 L−1 and 102 L−1.

      Two typical events were examined in detail to reveal the ice nucleation in cirrus clouds when effective INP dust particles existed. In the first case (on 12 January 2018), three optically transparent cirrus clouds appeared on top of an aloft dust layer with a cloud top temperature of –47.8°C and a cloud top height of 9.7 km. The in-cloud peak RHi values from lidar measurement (183.2%) and ERA5 data (157.4%) both exceeded 150%, revealing the competition between heterogeneous and homogeneous nucleation within the clouds. In the second case (on 7 February 2018), several short-living cirrus clouds were embedded in a dust layer at altitudes of 7–10 km with a cloud top temperature of –51.9°C and a cloud top height of 10.2 km. Their top edges invariably overlapped with each other. Considering that the in-cloud peak RHi values from lidar measurement (124.2%) and ERA5 data (133.9%) were both less than 150%, we inferred that only heterogeneous nucleation should be responsible for cirrus cloud formation. During the one-hour period before the cirrus clouds appeared in the lidar FOV, the lidar-derived peak RHi at cirrus altitudes for the first case (133.5%) is much larger than that for the second case (74.2%), which provides a great water vapor condition for the subsequent occurrence of homogeneous nucleation on 12 January 2018.

      Three cases showed a similar feature as the first case and were classified into category A, while the remaining nineteen cases showed the typical feature of the second case and were classified into category B. Therefore, based on the remote sensing approach with multiple devices and data, we verified that heterogeneous nucleation generally took place in cirrus cloud formation in the upper troposphere over Wuhan, a midlatitude site in the Northern Hemisphere, in the presence of advected dust particles. Although accompanying cloud top temperatures were as low as –51.9°C to –30.4°C, dust-related heterogeneous nucleation could contribute significantly to primary ice nucleation in cirrus clouds by providing sufficient dust INPs and even inhibit homogeneous nucleation (19 out of 22 cases) by consuming water vapor and thus reducing required RHi (≥ 150%).

      In future work, we plan to record more cases when water vapor Raman lidar measurements are available so that the RHi values within or around the cirrus clouds can be more accurate, providing more robust results. The simultaneous observation of millimeter-wave radar will also be involved so that the ice crystal number concentrations (ICNCs) within the cirrus clouds can be obtained (Bühl et al., 2016). Because the ICNCs formed from homogeneous nucleation are much larger than those from heterogeneous nucleation, it is possible to distinguish the respective importance of the two nucleating mechanisms (Cziczo et al., 2013). Furthermore, Doppler lidar observations are also necessary for future studies so that more plausible vertical velocity data are available as another quantitative criterion to distinguish heterogeneous and homogeneous nucleation (Seifert et al., 2010; Bühl et al., 2019). Lastly, an updated spectrally resolved water Raman lidar, which is able to simultaneously detect gaseous, liquid, and ice water, is also anticipated to be involved in future dust-related cirrus cloud observations (Liu and Yi, 2013).

      Acknowledgements. This research is funded by the National Natural Science Foundation of China (Grant Nos. 42005101 and 41927804), Hubei Provincial Natural Science Foundation of China (Grant No. 2020CFB229), and the Fundamental Research Funds for the Central Universities Grant (Grant Nos. 2042020kf0018 and 2042021kf1066). The Meridian Space Weather Monitoring Project (China) also provides financial support for lidar maintenance. The authors thank the European Centre for Medium-Range Weather Forecasts (ECMWF) for the ERA5 reanalysis data (https://climate.copernicus.eu/climate-reanalysis) and the University of Wyoming for providing Wuhan radiosonde data (http://weather.uwyo.edu/upperair/sounding.html). We also thank all the colleagues who participated in the operation of the lidar systems at our site. Data used to generate the results of this paper are available from the authors upon request (e-mail: yf@whu.edu.cn).

Reference

Catalog

    /

    DownLoad:  Full-Size Img  PowerPoint
    Return
    Return