• Austin, R. T., , A. J. Heymsfield, , and G. L. Stephens, 2009: Retrieval of ice cloud microphysical parameters using the CloudSat millimeter-wave radar and temperature. J. Geophys. Res., 114 , D00A23. doi:10.1029/2008JD010049.

    • Search Google Scholar
    • Export Citation
  • de Reus, M. S., and Coauthors, 2009: Evidence for ice particles in the tropical stratosphere from in-situ measurements. Atmos. Chem. Phys., 9 , 67756792.

    • Search Google Scholar
    • Export Citation
  • Dubuisson, P., , V. Giraud, , J. Pelon, , B. Cadet, , and P. Yang, 2008: Sensitivity of thermal infrared radiation at the top of the atmosphere and the surface to ice cloud microphysics. J. Appl. Meteor. Climatol., 47 , 25452560.

    • Search Google Scholar
    • Export Citation
  • Heymsfield, A. J., , A. Bansemer, , G. Heymsfield, , and A. O. Fierro, 2009: Microphysics of maritime tropical convective updrafts at temperatures from −20° to −60 °C. J. Atmos. Sci., 66 , 35303562.

    • Search Google Scholar
    • Export Citation
  • Hunt, W. H., , D. M. Winker, , M. A. Vaughan, , K. A. Powell, , P. L. Lucker, , and C. Weimer, 2009: CALIPSO lidar description and performance assessment. J. Atmos. Oceanic Technol., 26 , 12141228.

    • Search Google Scholar
    • Export Citation
  • Knollenberg, R. G., , K. Kelly, , and J. C. Wilson, 1993: Measurements of high number densities of ice crystals in the tops of tropical cumulonimbus. J. Geophys. Res., 98 , 86398664.

    • Search Google Scholar
    • Export Citation
  • Libbrecht, K. G., 2005: The physics of snow crystals. Rep. Prog. Phys., 68 , 855895.

  • Macke, A., , and M. I. Mischenko, 1996: Applicability of regular particle shapes in light scattering calculations for atmospheric ice particles. Appl. Opt., 33 , 42914296.

    • Search Google Scholar
    • Export Citation
  • Noel, V., , D. M. Winker, , M. McGill, , and P. Lawson, 2004: Classification of particle shapes from lidar depolarization ratio in convective ice clouds compared to in situ observations during CRYSTAL-FACE. J. Geophys. Res., 109 , D24213. doi:10.1029/2004JD004883.

    • Search Google Scholar
    • Export Citation
  • Ou, S. C., , K. N. Liou, , Y. Takano, , and R. L. Slonaker, 2005: Remote sensing of cirrus cloud particle size and optical depth using polarimetric sensor measurements. J. Atmos. Sci., 62 , 43714383.

    • Search Google Scholar
    • Export Citation
  • Pitts, M. C., , L. W. Thomason, , Y. Hu, , and D. M. Winker, 2007: An assessment of the on-orbit performance of the CALIPSO wide field camera. Remote Sensing of Clouds and the Atmosphere XII, A. Comerón et al., Eds., International Society for Optical Engineering (SPIE Proceedings, Vol. 6745), 67450K, doi:10.1117/12.737377.

    • Search Google Scholar
    • Export Citation
  • Platt, C. M. R., 1973: Lidar and radiometric observations of cirrus clouds. J. Atmos. Sci., 30 , 11911204.

  • Platt, C. M. R., 1981: Remote sounding of high clouds. III: Monte Carlo calculations of multiple scattered lidar returns. J. Atmos. Sci., 38 , 156167.

    • Search Google Scholar
    • Export Citation
  • Platt, C. M. R., , D. M. Winker, , M. A. Vaughan, , and S. D. Miller, 1999: Backscatter-to-extinction ratios in the top layers of tropical mesoscale convective systems and in isolated cirrus from LITE observations. J. Appl. Meteor., 38 , 13301345.

    • Search Google Scholar
    • Export Citation
  • Platt, C. M. R., , S. A. Young, , R. T. Austin, , G. R. Patterson, , D. L. Mitchell, , and S. D. Miller, 2002: LIRAD observations of tropical cirrus clouds in MCTEX. Part I: Optical properties and detection of small particles in cold cirrus. J. Atmos. Sci., 59 , 31453162.

    • Search Google Scholar
    • Export Citation
  • Rienecker, M. M., and Coauthors, 2008: The GEOS-5 data assimilation system: Documentation of versions 5.0.1, 5.1.0, and 5.2.0. NASA Technical Report Series on Global Modeling and Data Assimilation, Vol. 27, NASA-TM-2008-104606, 101 pp.

    • Search Google Scholar
    • Export Citation
  • Stephens, G. L., and Coauthors, 2008: CloudSat mission: Performance and early science after the first year of operation. J. Geophys. Res., 113 , D00A18. doi:10.1029/2008JD009982.

    • Search Google Scholar
    • Export Citation
  • Takano, Y., , and K-N. Liou, 1989: Solar radiative transfer in cirrus clouds. Part I. Single-scattering and optical properties of hexagonal ice crystals. J. Atmos. Sci., 46 , 319.

    • Search Google Scholar
    • Export Citation
  • Takano, Y., , and K-N. Liou, 1995: Radiative transfer in cirrus clouds. Part III: Light scattering by irregular ice crystals. J. Atmos. Sci., 52 , 818837.

    • Search Google Scholar
    • Export Citation
  • Winker, D. M., 2003: Accounting for multiple scattering in retrievals from space lidar. 12th International Workshop on Lidar Multiple Scattering Experiments, C. Werner et al., Eds., International Society for Optical Engineering (SPIE Proceedings, Vol. 5059), 128–139.

    • Search Google Scholar
    • Export Citation
  • Winker, D. M., , R. Couch, , and M. P. McCormick, 1996: An overview of LITE: NASA’s Lidar In-Space Technology Experiment. Proc. IEEE, 84 , 164180.

    • Search Google Scholar
    • Export Citation
  • Winker, D. M., , W. H. Hunt, , and M. J. McGill, 2007: Initial performance assessment of CALIOP. Geophys. Res. Lett., 34 , L19803. doi:10.1029/2007GL030135.

    • Search Google Scholar
    • Export Citation
  • Winker, D. M., , M. A. Vaughan, , A. H. Omar, , Y. Hu, , K. A. Powell, , Z. Liu, , W. H. Hunt, , and S. A. Young, 2009: Overview of the CALIPSO mission and CALIOP data processing algorithms. J. Atmos. Oceanic Technol., 26 , 23102323.

    • Search Google Scholar
    • Export Citation
  • Yang, P., , B. A. Baum, , A. J. Heymsfield, , Y. X. Hu, , H-L. Huang, , S-C. Tsay, , and S. Ackerman, 2003: Single-scattering properties of droxtals. J. Quant. Spectrosc. Radiat. Transfer, 79 , 11591169.

    • Search Google Scholar
    • Export Citation
  • Young, S. A., , and M. A. Vaughan, 2009: The retrieval of profiles of particulate extinction from Cloud Aerosol Lidar Infrared Pathfinder Satellite Observations (CALIPSO) data: Algorithm description. J. Atmos. Oceanic Technol., 26 , 11051119.

    • Search Google Scholar
    • Export Citation
  • View in gallery

    Strong mesoscale convective system observed over North Western Australia on 15 Mar 2007. (a) Latitude–height plot of CALIOP 532-nm attenuated backscatter coefficient; layer boundaries shown represent the limits of integration used in determining γ′(π) and Δ. (b) Plot of integrated attenuated backscatter (sr−1 × 10) (solid line) and integrated depolarization ratio (dashed line) with latitude. (c) Integrated attenuated backscatter plotted against midcloud temperature.

  • View in gallery

    The same system as in Fig. 1, but showing corresponding images from the CloudSat millimeter radar. (top) A repeat of the CALIOP lidar image; (top middle) radar reflectivity (dBZ), (bottom middle) cloud particle effective radius (μm), and (bottom) ice water content (mg m−3).

  • View in gallery

    As in Fig. 1, but for a strong mesoscale system observed in the Pacific warm pool on 31 Mar 2007.

  • View in gallery

    As in Fig. 2, but for the system on 31 Mar 2007.

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 26 26 4
PDF Downloads 6 6 0

Characteristics of CALIPSO and CloudSat Backscatter at the Top Center Layers of Mesoscale Convective Systems and Relation to Cloud Microphysics

View More View Less
  • 1 CSIRO Marine and Atmospheric Research, Aspendale, Victoria, Australia
  • | 2 NASA Langley Research Center, Hampton, Virginia
  • | 3 Colorado State University, Fort Collins, Colorado
© Get Permissions
Full access

Abstract

Following the discovery of anomalously high values of lidar integrated attenuated backscatter near the top center layers of mesoscale convective systems (MCSs) observed by the NASA Lidar In-Space Technology Experiment (LITE), a search of Cloud Aerosol Lidar with Orthogonal Polarization (CALIOP) data on board the Cloud Aerosol Lidar and Infrared Pathfinder Satellite Observation (CALIPSO) platform revealed the same phenomena in a sample of eight MCSs investigated. The backscatter depolarization ratio also showed changes concurrent with the high integrated backscatter and either increased or decreased concurrently with the anomalous backscatter. Simultaneous CloudSat data in the A-Train formation showed a cloud-top altitude similar to that measured by CALIOP, indicating fairly large ice crystals were reaching cloud top. Based on previous work, the CALIOP and CloudSat returns were likely due to a mix of small ice droxtals or frozen drops extending in a continuous spectrum to large crystals composed of well-formed hexagonal columns, thick hexagonal plates, spheroids, and irregular particles. The CALIOP lidar would detect the whole spectrum whereas CloudSat would detect ice crystals greater than ∼30 μm in effective radius; there were apparently enough of such crystals to allow CloudSat to detect a cloud-top height similar to that found by CALIOP. Using such a model, it was estimated that the measured backscatter phase function in the most active part of the cloud could be reconciled approximately with theoretical values of the various crystal habits. However, it was harder to reconcile the changes in depolarization ratio given the absence of values of this parameter for small droxtal crystals.

Corresponding author address: C. M. R. Platt, 47 Koetong Parade, Mt. Eliza, VIC, Australia. Email: mplatt@net2000.com.au

Abstract

Following the discovery of anomalously high values of lidar integrated attenuated backscatter near the top center layers of mesoscale convective systems (MCSs) observed by the NASA Lidar In-Space Technology Experiment (LITE), a search of Cloud Aerosol Lidar with Orthogonal Polarization (CALIOP) data on board the Cloud Aerosol Lidar and Infrared Pathfinder Satellite Observation (CALIPSO) platform revealed the same phenomena in a sample of eight MCSs investigated. The backscatter depolarization ratio also showed changes concurrent with the high integrated backscatter and either increased or decreased concurrently with the anomalous backscatter. Simultaneous CloudSat data in the A-Train formation showed a cloud-top altitude similar to that measured by CALIOP, indicating fairly large ice crystals were reaching cloud top. Based on previous work, the CALIOP and CloudSat returns were likely due to a mix of small ice droxtals or frozen drops extending in a continuous spectrum to large crystals composed of well-formed hexagonal columns, thick hexagonal plates, spheroids, and irregular particles. The CALIOP lidar would detect the whole spectrum whereas CloudSat would detect ice crystals greater than ∼30 μm in effective radius; there were apparently enough of such crystals to allow CloudSat to detect a cloud-top height similar to that found by CALIOP. Using such a model, it was estimated that the measured backscatter phase function in the most active part of the cloud could be reconciled approximately with theoretical values of the various crystal habits. However, it was harder to reconcile the changes in depolarization ratio given the absence of values of this parameter for small droxtal crystals.

Corresponding author address: C. M. R. Platt, 47 Koetong Parade, Mt. Eliza, VIC, Australia. Email: mplatt@net2000.com.au

1. Introduction and theory

The Lidar In-Space Technology Experiment (LITE; Winker et al. 1996) carried a high-power three-wavelength neodymium–yttrium–aluminum–garnet (Nd:YAG) lidar. The lidar flew aboard the space shuttle Discovery for 10 days in September of 1994, during which time 53 h were devoted to lidar soundings. During this time orbits over the Pacific Ocean warm pool region crossed several mesoscale convective systems (MCS), including Typhoon Melissa. Unusual increases in the integrated attenuated lidar backscatter were discovered at 532 nm with maximum values near the storm centers. The lidar pulse was penetrating to 1–2 km into the cloud top before complete attenuation occurred. In other decaying systems the enhancements were absent (Platt et al. 1999). The LITE lidar sounded at 5° to the nadir to avoid specular reflections from the sea surface and plate crystals in clouds.

The opportunity to investigate these backscatter anomalies further and to find out if they occurred in other MCSs has been provided by the successful launch of the Cloud Aerosol Lidar and Infrared Pathfinder Satellite Observation (CALIPSO) mission (Winker et al. 2009). The primary instrument aboard CALIPSO is the Cloud Aerosol Lidar with Orthogonal Polarization (CALIOP; Hunt et al. 2009). CALIOP is similar to the LITE lidar in being an elastic lidar transmitting at 532 and 1064 nm; however, CALIOP replaces LITE’s 355-nm channel with a depolarizing channel at 532 nm. The depolarization properties of the crystals give further information on ice crystal size and habit (e.g., Platt et al. 2002; Noel et al. 2004; Ou et al. 2005). Photomultiplier tubes are used for detection at 532 nm, and they provide large linear dynamic range and high sensitivity. Dual 14-bit digitizers provide the 22-bit dynamic range required to encompass the full range of backscatter up to dense clouds (Winker et al. 2007). The calculated value of backscatter coefficient below which signal saturation will not occur is 2.0 km−1 sr−1 (W. H. Hunt 2010, personal communication), which is a factor of about 100 above values of the backscatter coefficient from the MCSs. Also included aboard CALIPSO are two passive sensors: a three-channel imaging infrared radiometer (IIR; Dubuisson et al. 2008) and a single-wavelength wide field camera (Pitts et al. 2007). The CALIPSO satellite was launched in April 2006 along with CloudSat [carrying the 94-GHz Cloud Profiling Radar (CPR; Stephens et al. 2008)] flying in formation, thus observing closely the same patch of the atmosphere in both space and time. CALIPSO follows 12.5 s behind CloudSat on average, with the radar and lidar footprints overlapping 90% of the time. The detectable radar reflectivity extends right through to the surface except in intense rain systems such as MCSs, where there is severe attenuation toward the surface. The integrated attenuated isotropic backscatter γ′(π) can be written as (Platt 1973)
i1558-8432-50-2-368-e1
where k is the isotropic backscatter-to-extinction ratio, η is an effective multiple-scattering factor (Platt 1981), and is the effective two-way transmittance, also modified by multiple scattering. When the cloud is fully attenuating to the lidar pulse, then and, from Eq. (1),
i1558-8432-50-2-368-e2
If η is known, then the isotropic backscatter-to-extinction ratio k can be derived directly from the calculated quantity γ′(π). The values of k at peak values of integrated backscatter are calculated from Eq. (2) by assuming η = 0.6. Winker (2003) calculated this value for hexagonal crystals, which is relevant to the findings in this paper, as will be shown later. The same value is used to calculate baseline values away from the peak. Winker calculated slightly higher values of η for fully attenuating cirrus layers, but the value of 0.6 is found to be best for extinction retrieval of CALIOP results from both low- and high-optical-depth cirrus (Young and Vaughan 2009). If γ′(π) is calculated from the calibrated isotropic backscatter coefficient, then the backscatter-to-extinction ratio k is equal numerically to the ice-crystal normalized backscatter phase function P(π), which itself is a function of the ice-crystal habit. Values of γ′(π) for the strongly attenuating MCSs in CALIOP are discussed in what follows. Note that the isotropic backscatter values are used because the quantity k is related directly to the backscatter phase function P(π). The CALIOP data are plotted here as the backscatter coefficient per steradian and need to be multiplied by 4π to give the isotropic backscatter coefficient.
The backscatter depolarization ratio Δ is defined as
i1558-8432-50-2-368-e3
where β and β are the perpendicular and parallel depolarized returns, respectively. Values of Δ were obtained from Eq. (3).
The equivalent radar reflectivity factor Ze is obtained from the CloudSat Cloud Geometrical Profile (“2B-GEOPROF”) standard data product. For ice particles that are sufficiently small relative to the radar wavelength (3.2 mm for the CPR), the effective reflectivity factor may be written as the product of the sixth moment of the particle diameter times a dielectric factor that accounts for the standard definition of reflectivity with respect to liquid water:
i1558-8432-50-2-368-e4
where
i1558-8432-50-2-368-e5
and m is the index of refraction for ice or liquid water. As the particle size becomes comparable to the radar wavelength, Mie-scattering effects become more important. The CloudSat Cloud Water Content (Radar Only) data product (“2B-CWC-RO”) uses an optimal estimation algorithm to retrieve the effective radius and ice water content (IWC) of ice clouds from the measured radar reflectivity factor Ze and modeled cloud temperature (Austin et al. 2009). The minimum detectable radar reflectivity is approximately −30 dBZ, with some variations with season, location, and background. The minimum detectable particle size is 30–40 μm in terms of the effective radius of a distribution of equivalent-mass spheres.

All of the temperatures referred to in this study are obtained from the National Aeronautics and Space Administration (NASA) Goddard Earth Observing System (GEOS-5) dataset. These are derived from models and assimilated radiosonde and other data. A comparison with the radiosonde data indicates an uncertainty of 1°–2°C in the regions studied (Rienecker et al. 2008). The effects of these uncertainties will be discussed later.

2. Systems investigated

The cases chosen were those MCS systems that showed either a pronounced peak in cloud-top altitude or alternatively a slower increase to maximum altitude and subsequent decrease. These features also showed maximum lidar backscatter and cloud attenuation at the centers of the peaks, similar to those detected in LITE. The maximum altitude was designated as cloud center although it may not have been at the cloud geographic center. Also, of course, the measured center where the CALIOP footprint crossed may not have been exactly over the most active region, but it was assumed for this study to be the center.

It was clear that this CALIOP footprint would not traverse these most active regions on many occasions, and this was borne out by finding only 17 such cases when browsing globally from September 2006 until April 2007, which was mainly the Southern Hemisphere summer season. Of the 17 found, 8 were chosen for further analysis. Five of these were in the Australia–Pacific warm pool regions, one in central equatorial Africa, one in South America, and one in Texas.

The IIR data showed that all of the systems exhibited low IR brightness temperature. The IIR was of limited value, however, having a scan width of 69 km, which was less than the width of five of the systems. However, three of the smaller systems indicated a circular shape in brightness temperature, although the orbit did not traverse exactly in the center of the circles.

When the lidar integrated attenuated backscatter was investigated, it was found that in all of the storms studied the same enhancement, to different extents, occurred as was found in LITE, in line with the assumption that the CALIOP footprint was close to the centers of the MCSs. This discovery was made after the selection of the MCSs. This paper presents some preliminary results from these eight systems.

The systems attained maximum altitudes of around 17 km and had various spatial extents varying from 1.5° to 9° in tropical regions. Lidar pulse penetration of the cloud was reduced to about 1.5 km at cloud centers.

For each system the integrated attenuated backscatter, the backscatter depolarization ratio of total backscatter (i.e., molecular plus particulate combined), and the midcloud temperature were calculated across the length of the system along the CALIPSO orbit. The midcloud temperature was used because it represents the cloud depth in which backscatter occurs before complete attenuation. All values were obtained from the CALIPSO version-2.01 data products available from the NASA Langley Atmospheric Science Data Center. For the CloudSat data, the radar reflectivity, equivalent radius, and IWC were calculated along equivalent paths.

These derived quantities are illustrated for two of the systems in Figs. 1 –4. Quantities are given in Table 1 for all eight systems and are discussed later.

Derived quantities for the system on 15 March 2007 are shown in Figs. 1 and 2. Figure 1a shows the 532-nm attenuated backscatter coefficient versus latitude across the mesoscale system. In Fig. 1b is shown the integrated attenuated backscatter and depolarization ratio along the orbit. Figure 1c shows the integrated attenuated backscatter plotted against the midcloud temperature. Results from the CloudSat analysis are shown in Fig. 2 depicting the lidar attenuated backscatter coefficient and the radar reflectivity, equivalent radius, and ice water concentration for the same system. The mesoscale system investigated was a large convective cloud mass in the Kimberley coastal region of North West Australia.

The backscatter coefficient in Fig. 1a shows a typical increase in value toward the center of the system with an accompanying increase in pulse attenuation. The backscatter coefficient decreases again beyond the system center. There is a similar increase in cloud-top altitude of 2–3 km toward the system center and a similar decrease beyond the center. All of the eight convective systems studied had the same characteristic structure. The integrated attenuated backscatter (sr−1) multiplied by 10, shown in Fig. 1b, shows a typical value of 0.025–0.03 sr−1 in the region from 22° to 17°S latitude. The value then increases to a maximum of about 0.057 sr−1 at the system center and then decreases again. Beyond 15°S the value decreases to about 0.01, indicating semitransparent cloud. The system, as in all of the systems studied, did not possess a visible eye, as in a fully developed hurricane. Such a system was detected by LITE (Platt et al. 1999) when its orbit passed over the center of Typhoon Melissa, and the enhancements in backscatter on each side of the eye were present and in fact were larger than in the present case. The depolarization ratio in Fig. 1b has a fairly steady value of about 0.38 across the whole system except near the center, where the values follow the changes in integrated backscatter closely to a maximum value of about 0.47. Last, the integrated attenuated backscatter in Fig. 1c peaks at the coldest part of the cloud top, as expected.

The millimeter radar backscatter, or reflectivity Ze, shows backscatter nearly down to the surface, indicating an active system with widespread rainfall. The measured cloud-top altitudes are the same as the CALIPSO altitudes near the system center and follow the same pattern. This indicates that particles large enough to be measured by CloudSat are present. In fact, the calculated effective radius is seen to increase from ∼30 μm at cloud top to 80 μm where the CALIPSO pulse is completely attenuated at a depth of 1–1.5 km, showing that “large” ice crystals are present near cloud top. In Fig. 2 the IWC is in the region of 0.2 g m−3. The reflectivity is the strongest near the cloud center. A narrow peak in cloud particle size also occurs at the cloud center.

The second of the eight systems investigated is shown in Fig. 3. This was a system over the Pacific Ocean warm pool observed on 31 March 2007. The features are broadly similar to those in Fig. 1. The integrated attenuated backscatter again peaks near the cloud center. There is one difference, however; in this system the depolarization ratio Δ actually decreases where the integrated backscatter peaks. Values of Δ also fluctuate away from the system center. However, peaks in Δ away from the center are all coincident with greater laser penetration to lower levels, which will affect the values of Δ.

The CloudSat millimeter radar reflectivity again shows the broad features seen on 15 March 2007. Possibly the only difference is that the effective radius of the cloud particles is somewhat smaller in the case of the 31 March 2007 system, but the effective radius again shows a narrow peak at the system center.

Six other systems were investigated in the CALIOP data where Δ showed either positive or negative behaviors. The lidar properties of interest for all of the eight MCS systems investigated are summarized in Table 1. The subscripts b and e represent the background, or baseline, and enhanced values of the various quantities listed, respectively. The enhanced values are taken as the peak values of integrated backscatter and the peak or minimum values of backscatter depolarization ratio. The baseline values represent the values of quantities along the MCS and away from the areas of enhanced backscatter.

The availability of temperature profiles in the vicinity of the CALIPSO observations allowed an investigation of any penetration of the systems through the tropopause. It was found that all of the MCSs except one did penetrate the tropopause, but by variable amounts and volumes. On average the maximum cloud-top vertical distance reached above the tropopause was 900 m (with an uncertainty of 200 m), but it varied between 1800 and 200 m for the various cases. The only MCS that did not reach the tropopause occurred on 7 April 2007; this MCS was at 30°N over Texas. The width of the MCSs where penetration occurred varied from 0.3° to 4.9°, with a mean of 2°. This is equivalent to about 240 km in width across the system. There was a weak dependence of latitude width on cloud penetration, with a linear correlation coefficient of 0.38.

The type of cloud extending above the tropopause varied between the dense central cloud and apparent cirrus blow-off of lower density. We take the case of 15 March 2007, which is illustrated in Fig. 1. Here the central dense portion protruded above the tropopause, which was at 16.65 km, but the highest parts of the cloud were thinner cirrus-type cloud occurring on the south side of the center, as shown in Fig. 1. Figure 2 shows that the cirrus cloud was not detected by the millimeter radar; thus, the cirrus crystal size was very small, probably 10 μm or less. In the second case illustrated in Figs. 3 and 4, a large part of the central section of the cloud, from 7.2° to 11°N, was above the tropopause, with maximum penetration of 660 m. In this case the penetration contained a large section of the main cloud, with little of the thin cirrus observed in Fig. 1. Penetration of the clouds above the tropopause was also noted by de Reus et al. (2009), who made direct measurements of particle number and size from an aircraft. They found crystals at between 400 and 1800 m above the tropopause above single convective storms in the Darwin region.

3. Discussion

Past work on systems similar to the ones discussed here is fairly sparse. Knollenberg et al. (1993) flew through the tops of large MCSs in northern Australia, including developed cyclones, using particle detectors to measure particle size and number density. They found high number densities of small crystals. In one traverse of Tropical Cyclone Jason the flight passed through the coldest cloud region, with minimum in-cloud temperature of −87°C at an altitude of 16.7 km. The cloud particle size spectrum was continuous between 1 and 50 μm in radius. The cloud extinction and IWC were estimated, for the current CALIOP study, from the corresponding ice crystal number spectrum using a model of spherical particles. The spectrum from 1 to 30 μm where the radar backscatter would not be detected yielded an extinction of 0.37 km−1 as compared with 0.18 km−1 for the larger particles. Thus, in such conditions CALIOP would detect many small particles undetected by CloudSat. The CloudSat returns would be confined to the larger particles over 30 μm in size. The IWC calculated was 5 × 10−2 g m−3 as compared with a maximum of 2 × 10−1 g m−3 measured by CloudSat shown in Figs. 2 and 4, so that conditions were comparable.

Recent work on the microphysics and dynamics of maritime tropical storms is summarized by Heymsfield et al. (2009). Doppler radar observations and simulations indicate that core updrafts can be 7–10 m s−1 at an altitude of 15 km, allowing large particles to reach cloud top. It is also found by Heymsfield et al. (2009) that in clean maritime conditions supercooled droplets swept up in the updrafts will undergo homogeneous nucleation at −38°C, producing a high concentration of small ice particles that can then be lofted to cloud top in the form of droxtal ice crystals (Yang et al. 2003). The presence of many small ice crystals is in agreement with the work of Knollenberg et al. (1993). Thus, such a scenario should also be considered for the current observations, as discussed below.

As discussed previously, it is assumed that the region at and around peak altitude in the CALIOP data represents the most active region with the strongest core updrafts and is defined as the center of the system. It is in these regions that both extinction and also integrated attenuated backscatter reach maximum values. A change in the integrated attenuated backscatter implies a change in the backscatter to extinction ratio k and therefore the normalized backscatter phase function P(π). Calculated values of this quantity for some typical ice crystal habits are shown in Table 2. The closest fit to the peak value of ke in Table 1 is the randomly oriented hexagonal column. The value for randomly oriented solid hexagonal plates is somewhat higher, but it depends on aspect ratio. Takano (2010, personal communication; see also Takano and Liou 1989) calculates the value of P(π) shown in Table 2 for the given aspect ratio. Droxtal ice crystals modeled for a 3-μm size give a value of P(π) of about 1.0, although more calculations are needed to obtain P(π) across the whole small-particle spectrum. There is also evidence that well-formed hexagonal columns and plates can exist near cloud top. Noel et al. (2004) found 10%–20% of such columns existing at every height in a convective system in Florida down to a temperature of −73°C. Hexagonal columns can certainly grow to over 30 μm in size, and at very high ice saturations (Knollenberg et al. 1993; Libbrecht 2005). Hexagonal solid ice crystal plates and spheroids were also observed by Noel et al. (2004) near cloud top at about the same concentration as hexagonal columns, but many irregular particles also occurred. These would not give the peak backscatter phase functions observed in the current study, but a high concentration of droxtals could balance out the effects of the irregular particles to give an overall value of P(π) that is close to the measured values in Table 1.

The depolarization ratio Δ also gives information on particle type and phase, and the peak values of ke in this study are accompanied either by an increase or decrease in Δ. Calculated values of Δ for various cases are fairly sparse, however. When Δ increases, as shown in the cases from 11 March, 15 March, and 4 April 2007 of Table 1, its values lying between 0.45 and 0.55 correspond approximately to values of 0.55 calculated by Takano and Liou (1995) for hexagonal columns and approximately 0.48 found by Noel et al. (2004). Noel et al. show that values of Δ only decrease, as required for the other cases in Table 1, when the aspect ratio decreases below unity, indicating hexagonal plates rather than columns. They show a value of 0.25 for Δ when the aspect ratio decreases to 0.5, but they do not show equivalent backscatter phase functions. Takano (2010, personal communication; see also Takano and Liou 1989) finds Δ to be 0.339 for solid plates for the example shown in Table 2, a lower value that is in agreement with the observations. Values of Δ for droxtals were not available, but it is possible the cases of lower Δ in the peak activity areas might be a function of a greater number concentration of droxtals.

A feature of the precipitating systems analyzed in this study was that the maximum cloud-top altitudes in the CALIOP case agreed closely with the CloudSat values. This is not surprising, however, when considering the size of crystals observed in the top layers of the clouds. The particles observed by CloudSat at cloud top would be over 30 μm in size. Also, de Reus et al. (2009) found crystals up to 400 μm in size when flying over the convective storms in the stratosphere north of Darwin, Australia.

The CloudSat images in Figs. 2 and 4 indicate that crystals grow up to 100 μm at about 1 km below the cloud top in the stronger systems, again in a peak at the cloud center. Outside the peak areas, the values of ke and Δ are similar to what are found in many cirrus clouds. The crystals there are presumably a mixture of various crystal types. It is clear that the conditions in the peak areas are different and are the result of higher updrafts and different nuclei and humidity conditions that lead to many small frozen drops, or droxtals.

4. Conclusions

The current study has shown that there is a region near the centers of large MCSs where the cloud microphysics is different from the surrounding cloud. This region is the most active, with maximum cloud altitude and extinction and with the coldest temperatures. Typhoon Melissa observed in LITE displayed the same properties where the integrated attenuated backscatter peaked on each side of the eye, with values of approximately 1.4–1.6. There is no evidence of an eye at the center of any of the systems in this study, but this might be obscured by the strong attenuation at cloud top (e.g., Heymsfield et al. 2009). Based on past work of Knollenberg et al. (1993) and Heymsfield et al. (2009), among others, there is a likely mix of many small particles and larger ice crystals in the active region. The small particles are likely to be frozen droplets or droxtal ice crystals (Yang et al. 2003), and the larger ones are likely to be hexagonal columns, hexagonal solid plates, spheroids, and irregular particles (e.g., Noel et al. 2004). The CALIOP detects the whole particle spectrum whereas CloudSat detects particles larger than 30 μm in size. This can account for the high extinction at the CALIOP wavelength, where a good proportion of the extinction is due to the small ice crystal spectrum. On the other hand, CloudSat can detect to cloud top also because of the presence of larger ice crystals lofted to cloud top in the strong updrafts.

Within the eight MCSs investigated there were found to be two types of active region, one where the depolarization ratio increased with the increasing integrated attenuated backscatter and one where the depolarization ratio systematically decreased. The latter might indicate a greater preponderance of small droxtals or frozen droplets in the spectrum. However, there is no information at present on the depolarization ratio of droxtals of various sizes.

The nature of the microphysics in these most active regions, at cloud centers, and indeed the regions themselves are obviously very important to the internal dynamics of the convective systems. Further modeling should aim to elucidate the processes occurring and to understand the nature of these active regions. Aircraft passes measuring the crystal spectra would be the most direct method of elucidating the microphysics further; it will be necessary to locate such regions with infrared brightness temperatures and to fly through the cloud tops.

Acknowledgments

Author C. M. Platt thanks NASA Langley Research Center and Colorado State University for financial support; Dr. Stuart Young, Dr. Yongxiang Hu, Dr. Ken Sassen, and the paper reviewers made useful comments and suggestions that have improved the paper substantially.

REFERENCES

  • Austin, R. T., , A. J. Heymsfield, , and G. L. Stephens, 2009: Retrieval of ice cloud microphysical parameters using the CloudSat millimeter-wave radar and temperature. J. Geophys. Res., 114 , D00A23. doi:10.1029/2008JD010049.

    • Search Google Scholar
    • Export Citation
  • de Reus, M. S., and Coauthors, 2009: Evidence for ice particles in the tropical stratosphere from in-situ measurements. Atmos. Chem. Phys., 9 , 67756792.

    • Search Google Scholar
    • Export Citation
  • Dubuisson, P., , V. Giraud, , J. Pelon, , B. Cadet, , and P. Yang, 2008: Sensitivity of thermal infrared radiation at the top of the atmosphere and the surface to ice cloud microphysics. J. Appl. Meteor. Climatol., 47 , 25452560.

    • Search Google Scholar
    • Export Citation
  • Heymsfield, A. J., , A. Bansemer, , G. Heymsfield, , and A. O. Fierro, 2009: Microphysics of maritime tropical convective updrafts at temperatures from −20° to −60 °C. J. Atmos. Sci., 66 , 35303562.

    • Search Google Scholar
    • Export Citation
  • Hunt, W. H., , D. M. Winker, , M. A. Vaughan, , K. A. Powell, , P. L. Lucker, , and C. Weimer, 2009: CALIPSO lidar description and performance assessment. J. Atmos. Oceanic Technol., 26 , 12141228.

    • Search Google Scholar
    • Export Citation
  • Knollenberg, R. G., , K. Kelly, , and J. C. Wilson, 1993: Measurements of high number densities of ice crystals in the tops of tropical cumulonimbus. J. Geophys. Res., 98 , 86398664.

    • Search Google Scholar
    • Export Citation
  • Libbrecht, K. G., 2005: The physics of snow crystals. Rep. Prog. Phys., 68 , 855895.

  • Macke, A., , and M. I. Mischenko, 1996: Applicability of regular particle shapes in light scattering calculations for atmospheric ice particles. Appl. Opt., 33 , 42914296.

    • Search Google Scholar
    • Export Citation
  • Noel, V., , D. M. Winker, , M. McGill, , and P. Lawson, 2004: Classification of particle shapes from lidar depolarization ratio in convective ice clouds compared to in situ observations during CRYSTAL-FACE. J. Geophys. Res., 109 , D24213. doi:10.1029/2004JD004883.

    • Search Google Scholar
    • Export Citation
  • Ou, S. C., , K. N. Liou, , Y. Takano, , and R. L. Slonaker, 2005: Remote sensing of cirrus cloud particle size and optical depth using polarimetric sensor measurements. J. Atmos. Sci., 62 , 43714383.

    • Search Google Scholar
    • Export Citation
  • Pitts, M. C., , L. W. Thomason, , Y. Hu, , and D. M. Winker, 2007: An assessment of the on-orbit performance of the CALIPSO wide field camera. Remote Sensing of Clouds and the Atmosphere XII, A. Comerón et al., Eds., International Society for Optical Engineering (SPIE Proceedings, Vol. 6745), 67450K, doi:10.1117/12.737377.

    • Search Google Scholar
    • Export Citation
  • Platt, C. M. R., 1973: Lidar and radiometric observations of cirrus clouds. J. Atmos. Sci., 30 , 11911204.

  • Platt, C. M. R., 1981: Remote sounding of high clouds. III: Monte Carlo calculations of multiple scattered lidar returns. J. Atmos. Sci., 38 , 156167.

    • Search Google Scholar
    • Export Citation
  • Platt, C. M. R., , D. M. Winker, , M. A. Vaughan, , and S. D. Miller, 1999: Backscatter-to-extinction ratios in the top layers of tropical mesoscale convective systems and in isolated cirrus from LITE observations. J. Appl. Meteor., 38 , 13301345.

    • Search Google Scholar
    • Export Citation
  • Platt, C. M. R., , S. A. Young, , R. T. Austin, , G. R. Patterson, , D. L. Mitchell, , and S. D. Miller, 2002: LIRAD observations of tropical cirrus clouds in MCTEX. Part I: Optical properties and detection of small particles in cold cirrus. J. Atmos. Sci., 59 , 31453162.

    • Search Google Scholar
    • Export Citation
  • Rienecker, M. M., and Coauthors, 2008: The GEOS-5 data assimilation system: Documentation of versions 5.0.1, 5.1.0, and 5.2.0. NASA Technical Report Series on Global Modeling and Data Assimilation, Vol. 27, NASA-TM-2008-104606, 101 pp.

    • Search Google Scholar
    • Export Citation
  • Stephens, G. L., and Coauthors, 2008: CloudSat mission: Performance and early science after the first year of operation. J. Geophys. Res., 113 , D00A18. doi:10.1029/2008JD009982.

    • Search Google Scholar
    • Export Citation
  • Takano, Y., , and K-N. Liou, 1989: Solar radiative transfer in cirrus clouds. Part I. Single-scattering and optical properties of hexagonal ice crystals. J. Atmos. Sci., 46 , 319.

    • Search Google Scholar
    • Export Citation
  • Takano, Y., , and K-N. Liou, 1995: Radiative transfer in cirrus clouds. Part III: Light scattering by irregular ice crystals. J. Atmos. Sci., 52 , 818837.

    • Search Google Scholar
    • Export Citation
  • Winker, D. M., 2003: Accounting for multiple scattering in retrievals from space lidar. 12th International Workshop on Lidar Multiple Scattering Experiments, C. Werner et al., Eds., International Society for Optical Engineering (SPIE Proceedings, Vol. 5059), 128–139.

    • Search Google Scholar
    • Export Citation
  • Winker, D. M., , R. Couch, , and M. P. McCormick, 1996: An overview of LITE: NASA’s Lidar In-Space Technology Experiment. Proc. IEEE, 84 , 164180.

    • Search Google Scholar
    • Export Citation
  • Winker, D. M., , W. H. Hunt, , and M. J. McGill, 2007: Initial performance assessment of CALIOP. Geophys. Res. Lett., 34 , L19803. doi:10.1029/2007GL030135.

    • Search Google Scholar
    • Export Citation
  • Winker, D. M., , M. A. Vaughan, , A. H. Omar, , Y. Hu, , K. A. Powell, , Z. Liu, , W. H. Hunt, , and S. A. Young, 2009: Overview of the CALIPSO mission and CALIOP data processing algorithms. J. Atmos. Oceanic Technol., 26 , 23102323.

    • Search Google Scholar
    • Export Citation
  • Yang, P., , B. A. Baum, , A. J. Heymsfield, , Y. X. Hu, , H-L. Huang, , S-C. Tsay, , and S. Ackerman, 2003: Single-scattering properties of droxtals. J. Quant. Spectrosc. Radiat. Transfer, 79 , 11591169.

    • Search Google Scholar
    • Export Citation
  • Young, S. A., , and M. A. Vaughan, 2009: The retrieval of profiles of particulate extinction from Cloud Aerosol Lidar Infrared Pathfinder Satellite Observations (CALIPSO) data: Algorithm description. J. Atmos. Oceanic Technol., 26 , 11051119.

    • Search Google Scholar
    • Export Citation

Fig. 1.
Fig. 1.

Strong mesoscale convective system observed over North Western Australia on 15 Mar 2007. (a) Latitude–height plot of CALIOP 532-nm attenuated backscatter coefficient; layer boundaries shown represent the limits of integration used in determining γ′(π) and Δ. (b) Plot of integrated attenuated backscatter (sr−1 × 10) (solid line) and integrated depolarization ratio (dashed line) with latitude. (c) Integrated attenuated backscatter plotted against midcloud temperature.

Citation: Journal of Applied Meteorology and Climatology 50, 2; 10.1175/2010JAMC2537.1

Fig. 2.
Fig. 2.

The same system as in Fig. 1, but showing corresponding images from the CloudSat millimeter radar. (top) A repeat of the CALIOP lidar image; (top middle) radar reflectivity (dBZ), (bottom middle) cloud particle effective radius (μm), and (bottom) ice water content (mg m−3).

Citation: Journal of Applied Meteorology and Climatology 50, 2; 10.1175/2010JAMC2537.1

Fig. 3.
Fig. 3.

As in Fig. 1, but for a strong mesoscale system observed in the Pacific warm pool on 31 Mar 2007.

Citation: Journal of Applied Meteorology and Climatology 50, 2; 10.1175/2010JAMC2537.1

Fig. 4.
Fig. 4.

As in Fig. 2, but for the system on 31 Mar 2007.

Citation: Journal of Applied Meteorology and Climatology 50, 2; 10.1175/2010JAMC2537.1

Table 1.

The cloud system date, baseline and enhanced (i.e., γb and γe) integrated attenuated backscatter values (sr−1), baseline and enhanced values of isotropic backscatter-to-extinction ratio (kb and ke) for η = 0.6, baseline and enhanced or diminished values of the integrated depolarization ratio (Δb and Δe), change in depolarization ratio from baseline value (Δe − Δb), and the midcloud temperature (Tt; °C).

Table 1.
Table 2.

Theoretical values of the normalized backscatter phase function P(π) for randomly oriented crystals at 0.55-μm wavelength.

Table 2.
Save