J. Meteor. Res.  2017, Vol. 31 Issue (3): 576-585   PDF    
http://dx.doi.org/10.1007/s13351-017-6044-3
The Chinese Meteorological Society
0

Article Information

Hongke CAI, Yunfei FU, Quanliang CHEN, Xiao FENG, Xin TIE, Ranting TAO, Kepiao XU. 2017.
Optical Properties of Cirrus Transition Zones over China Detected by CALIOP. 2017.
J. Meteor. Res., 31(3): 576-585
http://dx.doi.org/10.1007/s13351-017-6044-3

Article History

Received May 10, 2016
in final form November 8, 2016
Optical Properties of Cirrus Transition Zones over China Detected by CALIOP
Hongke CAI1,2, Yunfei FU2, Quanliang CHEN1, Xiao FENG1, Xin TIE1, Ranting TAO1, Kepiao XU1     
1. School of Atmospheric Science, Chengdu University of Information Technology/Plateau Atmospheric and Environment Laboratory of Sichuan Province, Chengdu 610225;
2. School of Earth and Space Sciences, University of Science and Technology of China, Hefei 230026
ABSTRACT: A transition zone near cirrus lateral boundaries can be detected by CALIOP (cloud–aerosol lidar with orthogonal polarization). In the present study, for such transition zones over China, a number of optical properties, such as the backscatter coefficient and depolarization ratio, showed transitional characteristics between cirrus and clear sky. The stepped horizontal profile showed sharp changes in particle number and morphology between cirrus clouds and clear sky. The color ratio, however, was unable to show cirrus transition features because of the low signal-to-noise ratio. Typical ice particles presented a color ratio of 0.55–1.25 and a depolarization ratio of greater than 0.12, which were significantly higher than those of clear sky. Therefore, optical properties in transition took the form of stepwise horizontal profiles. The proportion of typical-featured particles also demonstrated a stepped horizontal profile similar to the optical characteristics, but the relationship between the proportion and the optical characteristics was not uniform in the cirrus clouds, transition zone, and clear sky. Therefore, the optical changes in the transition zone were caused by not only the change in particle concentration, but also the change in the particles themselves. The probability density distribution of the transition-zone widths showed a positive skewness distribution, and transition zones with widths of 3–5 km occurred most frequently. Overall, transition-zone width decreased with increasing temperature and increased with increasing vertical and horizontal wind speeds. This trend demonstrated independence with the direction of the vertical and horizontal winds. These observations implied that the transitional features were caused by material exchange, such as entrainment and turbulent transport, near the cirrus lateral boundaries, and by the phase transformation of particles, such as sublimation.
Key words: cirrus     transition zone     lidar     temperature     wind speed    
1 Introduction

Recent research has indicated that clouds are surrounded by a transition zone, in which optical properties change sharply. At short distances (less than 20 km) to the nearest cloud, observations of parameters such as aerosol optical thickness, shortwave albedo, color ratio, and attenuated backscatter coefficient, show transitional characteristics (Su et al., 2008; Redemann et al., 2009; Tackett and Di Girolamo, 2009; Varnai and Marshak, 2011). The formation (such as condensation) and removal (such as evaporation and sedimentation) of cloud particles, together with the hydration and dehydration of aerosol particles, cause dramatic changes in microphysical characteristics, such as number, phase, and scale, and will also change the thermal structure and optical properties of the atmosphere (Hegg et al., 1990; Lelieveld and Heintzenberg, 1992; Hoppel et al., 1994; Alkezweeny, 1995; Clarke et al., 1998; Lu et al., 2003; Hegg et al., 2004; Liu et al., 2014).

Cirrus clouds play an important role in the radiation budget and water cycle processes in the earth–atmosphere system (Liou, 1986; Wang et al., 1996). They exhibit a significant radiative forcing effect at the top of the atmosphere if their optical thickness is greater than 0.06 (Comstock et al., 2002) or if the ice water path is greater than 0.1 g m–2 (Berry and Mace, 2014), and long-term aggregation may enhance their contribution (Berry and Mace, 2014; Campbell et al., 2016). The most significant effects that cirrus clouds have on the water cycle and stratosphere–troposphere exchange are the processes of dehydration and aerosol scavenging (Lelieveld et al., 2007), which reduce the aerosol concentration density in the middle and upper troposphere simultaneously (Liu et al., 2001) and cause the occurrence frequency of high clouds and the lower stratosphere water vapor mixing ratio to have a significantly negative correlation in the tropics (Wang et al., 1996). Not only the seasonal cycle of cirrus occurrence, but also the seasonal cycle of the cirrus ice water content, are highly anti-correlated with the water vapor variation in the tropical tropopause layer (TTL), which supports the hypothesis that the total water at 100 hPa is roughly constant (Flury et al., 2012). In addition, heterogeneous chemical reactions occurring on cirrus ice particle surfaces can change the chemical composition of the tropopause region, affecting ozone budgets (Borrmann et al., 1996; Solomon et al., 1997).

Relying on the high resolution of instruments, significant gradients of transitional characteristics can be observed at the edges of cirrus clouds (Bowdle et al., 1991). For optically thin cirrus clouds (optical thickness less than 0.3; Sassen and Cho, 1992) and thinner cirrus transition zones, lidar is currently the most practical observational technique (Haladay and Stephens, 2009). High-altitude cirrus clouds reside in a relatively cold and dry environment, primarily composed of nonspherical ice crystals, and fundamentally different from the low-altitude clouds examined in the studies mentioned above. Li et al. (2014) attributed cirrus transition characteristics to the sublimation of ice crystals in an unsaturated environment and proposed that changes in the concentration density and formation of ice crystals result in changes in the radiation characteristics. A further exploration of cirrus transitional characteristics (over China) was conducted in the present study.

2 Data and method

CALIOP (cloud–aerosol lidar with orthogonal polarization) is a dual-wavelength polarization lidar aboard the CALIPSO (cloud–aerosol lidar and infrared pathfinder satellite observations) satellite, and provides 532- and 1064-nm vertical cloud and aerosol profiles (Winker et al., 2003). The Level 1B (L1B) product provides the 532- and 1064-nm attenuated backscatter coefficients ( $\beta {'_{{\rm{532}}}}$ and $\beta {'_{{\rm{1064}}}}$ ) to analyze properties of the scattering particles, while the level 2 vertical feature mask (L2-VFM) product profiles cloud and aerosol particles to identify the boundary between cloud and clear sky. From 8.3 to 20.2 km, the horizontal and vertical resolutions of L1B data are 1 km and 60 m, respectively, and the vertical resolution of the L2-VFM product is the same as that of the L1B data. The version 3 datasets of L2-VFM were used in the present study (Liu et al., 2014).

The depolarization ratio is defined as the ratio (δ) of the orthogonal component $(\beta {'_ \bot }) $ and the parallel component ( $\beta {'_{//}}$ ) of the 532-nm polarization backscattering coefficient,

$\delta = \frac{{\beta {'_ \bot }}}{{\beta {'_{//}}}}.$

Depolarization properties are used to describe the shape of particles, which can be used to distinguish aerosols from clouds and cloud-phase nonspherical particles. The depolarization ratio of ice crystal particles is closely related to temperature, and the depolarization ratio typically increases with a decrease in temperature (Noel et al., 2002; Wang et al., 2008). Further, the color ratio (χ) is defined as the backscatter intensity ratio of 1064- and 532-nm signals,

$\chi = \frac{{\beta {'_{1064}}}}{{\beta {'_{{\rm{532}}}}}}, $

and can reflect the scale characteristics of the scattering particles. These two parameters can be used as important indicators of the microphysical properties and optical characteristics of scattering particles.

NCEP climate forecast system reanalysis data (resolution: 0.5° × 0.5°; Saha et al., 2010) were used to investigate the relationship between the properties of cirrus transition zones and regional thermodynamic parameters. The model time interval was four times daily, and the nearest 1800 UTC data were used.

The selection of samples in this study followed the principles described below:

(1) The quality of the CALIOP product is limited by the signal-to-noise ratio (SNR), which is relatively low during the day because of the interference from solar radiation (Kim et al., 2008). This effect can be particularly significant for thin cirrus detection with small optical thickness. Thus, CALIOP data with high SNR at night were selected for use in this study.

(2) Only the internal parts of cirrus clouds were chosen, while the levels of cloud top and base were removed in calculation, in order to highlight the lateral characteristics of cirrus and avoid the influence of the upper and lower boundaries.

(3) The cloud and the clear sky are wide enough on the horizontal axis to ensure that the transition zone is affected by one direction of a lateral boundary only. Previous studies chose different horizontal widths. Tackett and Girolamo (2009) concluded that the backscatter 2.99 km away from clouds is not significantly different from the backscatter at greater distances, while Varnai and Marshak (2011) believed that the transition zone can extend to 15 km beyond the clouds. To fully show the characteristics of cirrus clouds and clear sky, a broader horizontal width was selected in the present study. One hundred and thirty-five profiles (45 km) in cirrus clouds and clear sky were chosen, which is approximately twice the transition zone width identified in this paper.

(4) The type and phase identification in the CALIOP products are sufficiently reliable, with a CAD (cloud–aerosol discrimination) score higher than 70 in the L2-VFM products (Liu et al., 2004, 2009).

(5) The 532-nm backscatter of cirrus clouds should be stronger than that of the clear sky to confirm that the cloud does exist and to avoid possible misjudgments in the L2-VFM products. The backscatter in the cirrus cloud interior $(\beta _c) $ should be larger than the sum of the monthly average value $ \left (\left. {{{\bar \beta }_a}} \right|_{15}^{45} \right)$ and two times the standard deviation $2\sigma \left({\left. {{\beta _a}} \right|_{15}^{45}} \right) $ in clear sky, which is 15 to 45 km outside the cloud,

${\beta _c} > \left. {{{\bar \beta }_a}} \right|_{15}^{45} + 2\sigma \left({\left. {{\beta _a}} \right|_{15}^{45}} \right).$

(6) The cirrus optical thickness should be relatively small to ensure that the cloud is transmissive to the laser and the cloud bottom height should be higher than 6 km so as to avoid possible misjudgment in the L2-VFM products.

(7) The cirrus lateral boundary should be explicit, and the uncertainty should be less than the horizontal width of the transition zone. The horizontal averaging required for detection of cirrus clouds in the L2-VFM algorithm should be no more than 1 km, which provides a coarse measure of backscatter intensity. The wider the averaging range, the more vague the actual position of the cloud boundary may be, and the lower the reliable resolution.

3 Optical properties in the cirrus transition zone

This research is based on CALIOP observations over mainland China between 20° and 40°N and between 105° and 125°E. A total of 1279 samples were selected in the summer from 2006 to 2011; they occurred between 1754 and 1912 UTC, and the cirrus optical thicknesses were all less than 1.

Because average values are easily affected by outliers, whereas the median expresses the central tendency of the samples and reflects the characteristics of the statistics more robustly, median values were used to describe the horizontal profile observed by CALIOP, as shown in Fig. 1. The median ${\beta _{{\rm{1064}}}}$ , ${\beta _{{\rm{532}}}}$ , $\chi $ , and ${\delta _{{\rm{532}}}}$ fitting results for cirrus clouds, the transition zone, and clear sky are indicated by the red dashed lines, with different fitting methods used in the three regions: quadratic fitting was used for cirrus clouds, while a linear fitting method was used in the transition zone and in the clear sky.

Figure 1 Optical properties near cirrus lateral boundaries detected by CALIOP: (a) 1064-nm backscatter coefficient, (b) 532-nm backscatter coefficient, (c) color ratio, and (d) depolarization ratio. The blue line represents the median; the black dashed lines represent the quartiles; and the red dashed lines represent the fitting result of the median in cirrus, the transition zone, and clear sky. On the horizontal axis along the CALIOP track, the negative, zero, and positive coordinates represent the cirrus, cloud boundaries, and clear sky, respectively, as identified by the L2-VFM products.

The fitting values were identical to the observed values, and the percentage fit residuals were all no more than 3.5%, except for the color ratio in the transition zone and clear sky. The linear fitting in the clear sky described the optical properties in the quiescent atmosphere condition, which was displaced from clouds, and the quadratic fitting in cirrus clouds reflected the symmetrical structure that was caused by the two corresponding lateral boundaries. Both the backscatter coefficient and depolarization ratio demonstrated gradient transition characteristics near the lateral boundaries, and the scope of the transition zone indicated by the two parameters was accordant. Based on the fitting results near the lateral boundaries, systematic deviation between the cirrus cloud and clear sky was considered to be the criterion of the transition zone, and the layer was between –11.5 and 8.2 km.

In the range of –4.8 to 1.2 km, the backscatter coefficient and depolarization ratio changed most dramatically. The median and the lower quartile of the color ratio had a similar stepped feature, but the upper quartile showed a reverse step function, meaning the color ratio in the cloudless atmosphere was even larger than that in cirrus clouds. Compared to the theoretical color ratio of atmospheric molecules, which is 0.0625 for 532/1064 nm Rayleigh scattering, there was a large deviation between them. Additionally, the color ratio in clear sky was larger than that calculated by Varnai and Marshak (2011), which was derived from the transition zone outside the cloud below 3 km over the ice-free ocean. This was attributed to the low SNR in the thin atmosphere, especially the very low SNR of the 1064-nm laser in the high-altitude clear sky at night (Tao et al., 2008; Wu et al., 2011), which led to the result that the color ratio could not show transitional features near cirrus clouds.

The horizontal profiles of optical properties demonstrate the transitional features of scattering particles: from strong backscatter to weak backscatter, from large scale to small scale, and from significantly irregular to globular. This latter feature is associated with changes in particle numbers and in the microphysical properties of scattering particles. The probability density curves (Fig. 2) of color ratio and depolarization ratio inside cirrus clouds, the transition zone, and clear sky reflect differences in scattering particle types. Compared to the intersection points of the probability density function curve, a color ratio that varies from 0.55 to 1.25 and a depolarization ratio that is greater than 0.12 (log10δ532 > –0.9) can be regarded as typical characteristics of ice crystals within cirrus. The proportion of such particles also displayed stepped transitional characteristics, as shown in Fig. 3.

Figure 2 The probability density function (PDF) of (a) color ratio and (b) depolarization ratio. The blue, black, and red lines represent the cirrus cloud (C), transition zone (0), and clear sky (A), respectively.

The fitting method in Fig. 3 is the same as in Fig. 1. The stepped transitional characteristics were similar to the horizontal profiles of optical properties, such as backscatter coefficient, color ratio, and depolarization ratio. Additionally, the position of the transition zone, determined by the proportion of typical particles, was roughly equivalent to that determined by the optical properties, which reflects the objective law of the intrinsic nature in the cirrus–sky continuum. This method may avoid abnormally large values of color ratio, and so can reflect the transitional characteristics of particle size.

Figure 3 Horizontal profiles of the particle proportion with the typical characteristics of cirrus particles: (a) the color ratio was between 0.55 and 1.25 and (b) the depolarization ratio was greater than 0.12. The blue lines represent the observational results, while the red dashed lines represent the fitting results.

The relationship between optical properties and typical particle proportion is shown in Fig. 4. By using the fitting results (solid lines in Fig. 4) of optical properties and typical particle proportion with cirrus clouds (from –45 to –12 km), the microphysical characteristics of scattering particles within the transition zone were extrapolated. In the transition zone (0.03 < δ < 0.2), the depolarization ratio decreased faster than the typical particle proportion. This means that changes in optical properties within the transition zone may not only be caused by the mixing ratio of different types of particles; the microphysical characteristics (depolarization ratio stands for the nonsphericity of particles) of scattering particles may also contribute to the changes in optical properties. Near the clear sky (δ < 0.04), the scatter points suddenly trended downward, indicating that the proportion of cloud droplets decreased rapidly. This observation indicates possible mechanisms of the transitional features: material exchange, such as entrainment and turbulent transport near the cirrus lateral boundaries and changes in particles themselves caused by phase transformation, such as sublimation, melting, and so on. Although the SNR of the color ratio is very low in the transition zone and in clear sky, the parameter still reflects the decreasing sizes of scattering particles and the decreasing proportion of the larger ones (Fig. 4a).

Figure 4 Relationship between the characteristics and proportion of typical droplets: (a) color ratio and (b) depolarization ratio. The solid lines are the linear fitting results within the cirrus interior between –45 and –12 km.
4 Horizontal extent of the cirrus transition zone

Although gradient changes of optical properties in the transition zone reflect the apparent variation directly, calculation of the gradient contains a large uncertainty due to the thin cirrus transition zone. Otherwise, it is nonsense to compare the gradient of different variables— $\displaystyle\frac{{\partial {\beta _{{\rm{532}}}}}}{{\partial x}}$ , $\displaystyle\frac{{\partial {\beta _{1064}}}}{{\partial x}}$ , $\displaystyle\frac{{\partial \chi }}{{\partial x}}$ , and $\displaystyle\frac{{\partial {\delta _{532}}}}{{\partial x}}$ —which changes by itself in the transition zone. The width of the transition zone took on a negative correlation with the gradient of the optical properties at a confidence level of α = 0.05. Additionally, the horizontal extent of the transition zone was robust. The optical properties systematically deviated from that within the cirrus interior and clear sky as described above, which could be defined as the internal transition zone criterion. Therefore, the transition zone widths determined by the backscattering coefficient were almost the same as those identified by the depolarization ratio.

Previous studies on transition zone properties have shown that increasing the relative humidity can induce the growth of aerosol particles, which leads to an increase in the aerosol particle scattering cross-section (Su et al., 2008; Tackett and Di Girolamo, 2009; Twohy et al., 2009). The width of the cirrus transition zone, showing a narrowing trend with increasing temperature and decreasing relative humidity, implies sublimation under the influence of temperature and saturation rate in the transition zone (Li et al., 2014). These studies reveal that optical properties in the transition zone are influenced by the changes in microphysical properties of the scattering particles.

The probability density distribution of the transition zone width shown at different temperatures, as shown in Fig. 5, was positively skewed under certain temperature conditions. A transition zone width of 3–5 km was most common. At temperatures colder than –30°C, the width varied with temperature. The narrowest transition band (1–3 km) ratio increased as the temperature increased, while the proportion of wider transition zones was reduced. This result is in agreement with Li et al. (2014).

Figure 5 Probability density function (PDF) of transition zone widths at different temperatures.

This phenomenon is attributable to the sublimation of cirrus ice crystals under the influence of temperature. The total ice mass of the cirrus cloud and ambient water vapor in the tropical tropopause to lower stratosphere region is nearly constant, while temperature acts as a main regulator for balancing the partition between them (Flury et al., 2012). The increase in water vapor appears to be related to the increase in temperature in the TTL (Su et al., 2011). Additionally, there is spatial anti-correlation between ice water content and water vapor in the upper troposphere, where low water vapor is associated with low temperatures in regions of tropical convection (Jiang et al., 2010). Hence, it is possible to infer that temperature may strongly control sublimation in cirrus formation, which would explain why the transition width varied with temperature. Moreover, glaciated liquid water droplets may influence the features of clouds and transition zones, which increase rapidly for temperatures warmer than –37°C and likely amount to more than cirrus ice crystals for temperatures warmer than –27°C (Campbell et al., 2015).

The relationship between the width of the transition zone and the dynamic conditions of the atmosphere is shown in Fig. 6. Because CALIPSO tracks north–south at night, the wind direction is determined by meridional wind. The wind blowing to the clouds is defined as negative, and positive represents the wind blowing out of the clouds. This is the same as the positive and negative signs of horizontal distance in Figs. 1 and 3. A red asterisk represents the transition zone width averaged under the condition of a given wind speed, and the red vertical line denotes the corresponding error limit. The blue line connecting the red asterisks represents the tendency of transition zone width with wind speed, meaning that the width widens with the increase of vertical and horizontal wind speeds. This trend is not concerned with the direction of either the vertical or the horizontal wind speed. This implies that the wind would strengthen the material exchange near the cirrus boundaries, which is conducive to the horizontal diffusion of ice crystals, leading the transition zone to widen. Vertical movements would promote the transfer of materials as well.

Figure 6 Relationships between transition zone width and (a) vertical wind velocity and (b) horizontal wind velocity.

The widths of the cirrus transition zones were less than 20 km, and the thicknesses were less than 1 km in this study, which is the equivalent of middle-γ- or small-scale weather systems. Small-scale turbulence would affect sublimation by changing the distribution of ice crystals and water vapor, as well as the temperature gradient (Kärcher, 2012). This may also explain the trend of transition zone width with changes in wind speed. The red vertical line in Fig. 6 shows that the relationship between the transition zone and wind speed contains a large uncertainty. It is difficult to obtain meteorological observational data matching with the transition zone. Because the properties of the transition zone are related to multiple factors, such as temperature, humidity, and wind speed, the statistical relationship of a single factor may be obscured by other meteorological elements.

5 Conclusion and discussion

By selecting high-quality observational samples from CALIOP, statistics on the optical properties near cirrus lateral boundaries were analyzed. The horizontal profiles of backscatter coefficient and depolarization ratio demonstrated transition zones with sharp changes. Transitional phenomena were identified in the range of 11.5 km in cirrus clouds to 8.2 km in clear sky, based on the cloud boundaries in the CALIOP L2-VFM products. The horizontal gradients of optical properties within the transition zone were significantly different from those of the cirrus interior and of the clear sky. The location of the transition zone identified by the depolarization ratio agreed with the corresponding result identified by backscatter coefficient, though the color ratio could not be used to show transitional features, due to low SNR.

Upon examination of the probability distributions of scattering particle optical properties in cirrus clouds, the transition zone, and clear sky, particles whose backscatter color ratio was 0.55–1.25, and whose polarization ratio was greater than 0.12, showed representative transitional changes. The proportion of such particles also displayed stepped-transition horizontal profiles. The changes in optical properties in the transition zone may thus come from two aspects: the change in mixing ratio with different types of scattering particles, and the change in the scattering particles by themselves.

The width of the transition zone as an indicator for change in the properties of the transition zone reveals a trend that changes with thermodynamic conditions; specifically, the horizontal extent of the transition zone increases with decreasing temperature, and the transition zone also widens with increasing horizontal and vertical wind speeds. This suggests that phase transformations, such as sublimation and melting, which result in changes in particle morphology, along with material exchange, which results from entrainment and turbulent transport near the cirrus boundaries, are two possible mechanisms for the influence of the transition zone on optical properties. It is important to note, however, that the relationship of the transition zone width, temperature, and wind speed is subject to uncertainty in reanalysis data. More precise observational data could therefore play an important role in further studies to reveal the mechanisms of transition zone effects more clearly.

Acknowledgments. All figures were created by using the NCAR Command Language (NCL) (2016). The CALIOP products were downloaded from the Atmospheric Science Data Center at NASA Langley Research Center of the United States.

References
Alkezweeny A. J., 1995: Field observations of in-cloud nucleation and the modification of atmospheric aerosol size distributions after cloud evaporation. J. Appl. Meteor., 34, 2649–2654. DOI:10.1175/1520-0450(1995)034<2649:FOOICN>2.0.CO;2
Berry E., Mace G. G., 2014: Cloud properties and radiative effects of the Asian summer monsoon derived from A-Train data. J. Geophys. Res. Atmos., 119, 9492–9508. DOI:10.1002/2014JD021458
Borrmann S., Solomon S., Dye J. E., et al.,1996: The potential of cirrus clouds for heterogeneous chlorine activation. Geophys. Res. Lett., 23, 2133–2136. DOI:10.1029/96GL01957
Bowdle D. A., Rothermel J., Vaughan J. M., et al.,1991: Aerosol backscatter measurements at 10.6 micrometers with airborne and ground-based CO2 Doppler lidars over the Colorado high plains: 2. Backscatter structure . J. Geophys. Res. Atmos., 96(D3), 5337–5344. DOI:10.1029/90JD02157
Campbell J. R., Vaughan M. A., Oo M., et al.,2015: Distinguishing cirrus cloud presence in autonomous lidar measurements. Atmos. Meas. Tech., 8, 435–449. DOI:10.5194/amt-8-435-2015
Campbell J. R., Lolli S., Lewis J. R., et al.,2016: Daytime cirrus cloud top-of-the-atmosphere radiative forcing properties at a midlatitude site and their global consequences. J. Appl. Meteor. Climatol., 55, 1667–1679. DOI:10.1175/JAMC-D-15-0217.1
Clarke A. D., Varner J. L., Eisele F., et al.,1998: Particle production in the remote marine atmosphere: Cloud outflow and subsidence during ACE 1. J. Geophys. Res. Atmos., 103(D13), 16397–16409. DOI:10.1029/97JD02987
Comstock J. M., Ackerman T. P., Mace G. G., 2002: Ground-based lidar and radar remote sensing of tropical cirrus clouds at Nauru island: Cloud statistics and radiative impacts. J. Geophys. Res. Atmos., 107(D23). DOI:10.1029/2002JD002203
Flury T., Wu D. L., Read W. G., 2012: Correlation among cirrus ice content, water vapor and temperature in the TTL as observed by CALIPSO and Aura/MLS. Atmos. Chem. Phys., 12, 683–691. DOI:10.5194/acp-12-683-2012
Haladay T., Stephens G., 2009: Characteristics of tropical thin cirrus clouds deduced from joint CloudSat and CALIPSO observations. J. Geophys. Res. Atmos., 114(D8). DOI:10.1029/2008JD010675
Hegg D. A., Radke L. F., Hobbs P. V., 1990: Particle production associated with marine clouds. J. Geophys. Res. Atmos., 95(D9), 13917–13926. DOI:10.1029/JD095iD09p13917
Hegg D. A., Covert D. S., Jonsson H., et al.,2004: Observations of the impact of cloud processing on aerosol light-scattering efficiency. Tellus B, 56, 285–293. DOI:10.1111/j.1600-0889.2004.00099.x
Hoppel W. A., Frick G. M., Fitzgerald J. W., et al.,1994: Marine boundary layer measurements of new particle formation and the effects nonprecipitating clouds have on aerosol size distribution. J. Geophys. Res. Atmos., 99(D7), 14443–14459. DOI:10.1029/94JD00797
Jiang J. H., Su H., Pawson S., et al.,2010: Five year (2004–2009) observations of upper tropospheric water vapor and cloud ice from MLS and comparisons with GEOS-5 analyses. J. Geophys. Res., 115(D15). DOI:10.1029/2009JD013256
Kärcher B., 2012: Supersaturation fluctuations in cirrus clouds driven by colored noise. J. Atmos. Sci., 69, 435–443. DOI:10.1175/JAS-D-11-0151.1
Kim S.-W., Berthier S., Ra J.-C., et al.,2008: Validation of aerosol and cloud layer structures from the space-borne lidar CALIOP using a ground-based lidar in Seoul, Korea. Atmos. Chem. Phys., 8, 3705–3720. DOI:10.5194/acp-8-3705-2008
Lelieveld J., Heintzenberg J., 1992: Sulfate cooling effect on climate through in-cloud oxidation of anthropogenic SO2. Science, 258, 117–120. DOI:10.1126/science.258.5079.117
Lelieveld J., Brühl C., Jöckel P., et al.,2007: Stratospheric dryness: Model simulations and satellite observations. Atmos. Chem. Phys., 7, 1313–1332. DOI:10.5194/acp-7-1313-2007
Li R., Cai H. K., Fu Y. F., et al.,2014: The optical properties and longwave radiative forcing in the lateral boundary of cirrus cloud. Geophys. Res. Lett., 41, 3666–3675. DOI:10.1002/2014GL059432
Liou K.-N., 1986: Influence of cirrus clouds on weather and climate processes: A global perspective. Mon. Wea. Rev., 114, 1167–1199. DOI:10.1175/1520-0493(1986)114<1167:IOCCOW>2.0.CO;2
Liu H. Y., Jacob D. J., Bey I., et al.,2001: Constraints from 210Pb and 7Be on wet deposition and transport in a global three-dimensional chemical tracer model driven by assimilated meteorological fields . J. Geophys. Res. Atmos., 106(D11), 12109–12128. DOI:10.1029/2000JD900839
Liu J. J., Chen B., Huang J. P., 2014: Discrimination and validation of clouds and dust aerosol layers over the Sahara desert with combined CALIOP and IIR measurements. J. Meteor. Res., 28, 185–198. DOI:10.1007/s13351-014-3051-5
Liu Y. Z., Jia R., Dai T., et al.,2014: A review of aerosol optical properties and radiative effects. J. Meteor. Res., 28, 1003–1028. DOI:10.1007/s13351-014-4045-z
Liu Z. Y., Vaughan M. A., Winker D. M., et al.,2004: Use of probability distribution functions for discriminating between cloud and aerosol in lidar backscatter data. J. Geophys. Res. Atmos., 109(D15). DOI:10.1029/2004JD004732
Liu Z. Y., Vaughan M., Winker D., et al.,2009: The CALIPSO lidar cloud and aerosol discrimination: Version 2 algorithm and initial assessment of performance. J. Atmos. Oceanic Technol., 26, 1198–1213. DOI:10.1175/2009JTECHA1229.1
Lu M. L., Wang J., Freedman A., et al.,2003: Analysis of humidity halos around trade wind cumulus clouds. J. Atmos. Sci., 60, 1041–1059. DOI:10.1175/1520-0469(2003)60<1041:AOHHAT>2.0.CO;2
Noel V., Chepfer H., Ledanois G., et al.,2002: Classification of particle effective shape ratios in cirrus clouds based on the lidar depolarization ratio. Appl. Opt., 41, 4245–4257. DOI:10.1364/AO.41.004245
Redemann J., Zhang Q., Russell P. B., et al.,2009: Case studies of aerosol remote sensing in the vicinity of clouds. J. Geophys. Res. Atmos., 114(D6). DOI:10.1029/2008JD010774
Saha S., Moorthi S., Pan H.-L., et al.,2010: The NCEP climate forecast system reanalysis. Bull. Amer. Meteor. Soc., 91, 1015–1057. DOI:10.1175/2010BAMS3001.1
Sassen K., Cho B. S., 1992: Subvisual-thin cirrus lidar dataset for satellite verification and climatological research. J. Appl. Meteor., 31, 1275–1285. DOI:10.1175/1520-0450(1992)031<1275:STCLDF>2.0.CO;2
Solomon S., Borrmann S., Garcia R. R., et al.,1997: Heterogeneous chlorine chemistry in the tropopause region. J. Geophys. Res. Atmos., 102(D17), 21411–21429. DOI:10.1029/97JD01525
Su H., Jiang J. H., Liu X. H., et al.,2011: Observed increase of TTL temperature and water vapor in polluted clouds over Asia. J. Climate, 24, 2728–2736. DOI:10.1175/2010JCLI3749.1
Su W. Y., Schuster G. L., Loeb N. G., et al.,2008: Aerosol and cloud interaction observed from high spectral resolution lidar data. J. Geophys. Res. Atmos., 113(D24). DOI:10.1029/2008JD010588
Tackett J. L., Di Girolamo L., 2009: Enhanced aerosol backscatter adjacent to tropical trade wind clouds revealed by satellite-based lidar. Geophys. Res. Lett., 36. DOI:10.1029/2009GL039264
Tao Z. M., McCormick M. P., Wu D., et al.,2008: Measurements of cirrus cloud backscatter color ratio with a two-wavelength lidar. Appl. Opt., 47, 1478–1485. DOI:10.1364/AO.47.001478
The NCAR Command Language (Version 6. 3. 0), 2016: Boulder, Colorado: UCAR/NCAR/CISL/VETS. [Available online at http://dx.doi.org/10.5065/D6WD3XH5].
Twohy C. H., Coakley Jr J. A., Tahnk W. R., 2009: Effect of changes in relative humidity on aerosol scattering near clouds. J. Geophys. Res. Atmos., 114(D5). DOI:10.1029/2008JD010991
Varnai T., Marshak A., 2011: Global CALIPSO observations of aerosol changes near clouds. IEEE Geosci. Remote Sens. Lett., 8, 19–23. DOI:10.1109/LGRS.2010.2049982
Wang P.-H., Minnis P., McCormick M. P., et al.,1996: A 6-year climatology of cloud occurrence frequency from Stratospheric Aerosol and Gas Experiment II observations (1985–1990). J. Geophys. Res. Atmos., 101(D23), 29407–29429. DOI:10.1029/96JD01780
Wang Z. Z., Chi R. L., Liu B., et al.,2008: Depolarization properties of cirrus clouds from polarization lidar measurements over Hefei in spring. Chinese Opt. Lett., 6, 235–237. DOI:10.3788/COL20080604.0235
Winker, D. M., J. R. Pelon, M. P. McCormick, et al., 2003: The CALIPSO mission: Spaceborne lidar for observation of aerosols and clouds. SPIE 4893, Lidar Remote Sensing for Industry and Environment Monitoring III, Hangzhou, China, 24 March, SPIE, 1–11, doi: 10.1117/12.466539.
Wu D., Wang Z., Wang B., et al.,2011: CALIPSO validation using ground-based lidar in Hefei (31.9°N, 117.2°E), China. Appl. Phys. B, 102, 185–195. DOI:10.1007/s00340-010-4243-z