1. Convective Clouds Lecture Sequence Basic convective cloud types
Severe convection & mesoscale systems
Tropical cloud population
Convective feedbacks to large-scales
Extreme convection
Diurnal variability
Clouds in tropical cyclones

2. The tropical convective cloud population
Robert Houze
University of Washington
The midlevel inflow to the sf region is an imp part of the MCS circulation.
Lecture, Summer School on Severe and Convective Weather, Nanjing, 11 July 2011

3. The tropical convective cloud population
What do we know?
How has technology contributed?
What comes next?
The midlevel inflow to the sf region is an imp part of the MCS circulation.

4. Before Satellites

5. Visual Observation
Cumulonimbus
Cumulus congestus
Small cumulus

6. Radiosonde data in the tropics
“Hot tower hypothesis”
Riehl & Malkus 1958

7. Convective parameterization
GCM grid

8. Satellite Observations: an “inconvenient truth”

9. Large cloud shields
The IR pattern emphasizes the higher cloud tops, which largely determine the global precipitation (and hence latent heating) pattern.

10. Satellite view of the tropical cloud population
Early 1970’s
Satellite view of the tropical cloud population
Explained satellite pictures
Retained the hot tower notion
Included smaller clouds

11. The second “inconvenient truth”
Radars:
The second “inconvenient truth”

12. GATE 1974
40 ships. 12 a/c

13. 4 shipborne scanning digital C-band radars
1974
40 ships!
12 aircraft!
4 shipborne scanning digital C-band radars
40 ships. 12 a/c

14. More Field Projects to Study Convection
BoB 1979
JASMINE 1999
EPIC
TEPPS 1997
(Dashed: No sounding network)
Ground, ship, & airborne cm radars

15. Post-GATE view of the tropical cloud population
MESOSCALE CONVECTIVE SYSTEMS (MCSs)
Hot Tower
STRATIFORM RAIN
GCM grid
Houze et al. (1980)

16. Heating and cooling processes in an MCS
Top heavy
Houze 1982

17. Simplified MCS Heating Profiles
Schumacher et al. 2004
Height (km)
Deg K/day
Convective
Stratiform
The blue curve is applied to the stratiform fraction of the precipitation at the given pixel. Go to next slide.

18. MCS Net Heating Profiles
Height (km)
Deg K/day
0% stratiform
40% stratiform
70% stratiform
Schumacher et al. 2004
70% stratiform precipitation at a pixel yields the blue profile, which has a higher maximum than the 0% or 40 % stratiform profiles. Note also how the increased stratiform percentage yields a stronger vertical gradient of heating at upper levels and hence greater generation of potential vorticity. This shows how an increase in the stratiform rain fraction leads to a stronger circulation at upper levels.

19. Precipitation Radar in Space

20. The TRMM Satellite
Ku-band Radar
Low altitude, low inclination orbit

21. Knowledge of global rainfall before satellites measured rain from space
Global water budget is formed from P, E, and runoff. Getting P accurate is very important to evaluating the residual P-E. The large max of P in the tropics is almost entirely the result of deep convection.

22. Rainfall mapping revolutionized!
Satellites can map precipitation globally.
….This map is derived from a combination of IR, PMW, TRMM PR.
Deeper clouds produce precipitation.
……Tropics mostly deep convection.
….. Midlatitudes partly convective, partly baroclinic.
Combined satellite rainfall July 2000
TRMM plus passive microwave sensors + other

23. TRMM Satellite Instrumentation
λ= 2 cm
Important! PR measures 3D structure of radar echoes
Wavelength—short for radar.
Obtains data at high resolution in a 3D volume.
Allows us to study vertical structure of echoes.
Not trying to determine precip amounts.
92.5 minute orbit; approximately 16 orbits per day.
Revisits specific locations times per 30 day period (dependent on latitude).
Our data set consists of 1648 TRMM 2A23 & 2A25 rain containing orbit files from the monsoon season (June – September) in 2002 & 2003.
The TRMM 2A25 data contains corrected TRMM reflectivity data in 250 m height bins.
The TRMM 2A23 algorithm detects the presence of a bright band and categorizes precipitation as stratiform, convective or other (unclassified).
Kummerow et al, 1998

24. How tropical rain is distributed by cloud size and type

25. 2 Years of TRMM PR data Large Cbs MCSs Small isolated Cbs
Shallow isolated: Pixel separate from other rain certain areas withtop 1.5 km or more lower than climatological 0 deg C level.
Schumacher & Houze 2003

26. How do the environments of these regimes differ?

27. TRMM PR Deep Convective TRMM PR Shallow, Isolated Convective
SST Climatology (July)
July SST
TRMM PR Shallow, Isolated Convective

28. Traditional conceptual view of mean meridional distribution of tropical convection
Simpson 1992

29. Stratocumulus Regime
Moist layer a little deeper, but very dry aloft; no mechanism to remove the cap.

30. Trade Wind Regime
Moist layer a little deeper, but very dry aloft; no mechanism to remove the cap.

31. Indo/Pacific Warm Pool
Marginally conditionally unstable through depth of troposphere. LCL low. Warm moist sfc air can get above LCL and have slight buoyancy to great heights, deep clouds, but relatively weak updrafts.

32. “Trimodal” distribution Johnson et al
“Trimodal” distribution Johnson et al Evidence from TOGA COARE sounding data

33. “Trimodal distribution”
Cu congestus
Small Cb
“Trimodal distribution”
Suggested by
TOGA COARE radiosonde data
Johnson et al. 1999

34. Cloud Radars

35. “Trimodal” DistributionX
MANUS
“Trimodal” Distribution
Frequency of cloud-top height from 9 mm wavelength vertically pointing radar
Histograms of cloud top height from MMCR for (a) day 20-08Z (b) night 8-20Z. Manus. Aug 1999-Oct 2000
35 GHZ, 9 mm wavelenght
Hollars et al. 1999

36. Cloud Radar in Space

37. The A-Train Satellites show the complete MCS
MCSs previously studied in field projects—i.e. in local regions.
Now they can be studied globally.
Furthermore, the anvil clouds can be separated and studied separately.
CloudSat CPR is 3mm wavelength
MODS & AMSR-E are on Aqua
Yuan and Houze 2010

38. MCSs Over the Whole Tropics
(< km2)
(> km2)
Size of MCS is size of cold cloud top. To qualify as MCS, must have large cold cloud top, large rain area, and some heavy rain. This figure shows spatial distribution of active MCSs. Color of each point shows the number of active MCSs found within a circular area with radius of five degrees centered on that point. (a) Small separated MCSs (<12,000km2; i.e., the smallest 25%) in DJF, (b) large separated MCS (>40,000km2; i.e., the largest 25%) in DJF, (c) connected MCS in DJF, (d)-(f) same as (a)-(c) except for JJA.
Yuan and Houze 2010

39. Morphology of MCS anvils in different parts of the tropics
Data from CloudSat mm-Wavelength Cloud Profiling Radar
Normalized by dimension of the MCS rain are
Yuan and Houze 2010

40. Internal structure of MCS anvils shown by CloudSat Cloud Profiling Radar
Africa
Indian Ocean
Yuan, Houze, and Heymsfield 2011

41. Conceptual model of anvil microphysics
convective rain
stratiform rain
graupel
snow
WAVELENGTH: W-band, 3 mm wavelength
THIN ANVILS:
1) LEADING—higher -- comes directly from convective updraft
2) TRAILING—occur at more levels -- moisten a deeper layer
THICK ANVILS:
3) LEADING—stronger-larger particles -- graupel
Cetrone and Houze 2011

43. Five radars on a tiny island
Addu Atoll

44. Non-precipitating Cumulus
Radar Supersite Approach
Will document many aspects of the convective population
HUMIDITY
DUAL WAVELENGTH
Water vapor *
MM-WAVELENGTH
Non-precipitating Cumulus
Anvil cloud
CM-WAVELENGTH
Precipitation
POLARIMETRY
Microphysics
DOPPLER
Air motions

45. Summary & Conclusions

46. Timeline of progress Pre-satellite era Hot towers
Radars in field projects
MCSs, convective and stratiform precipitation regions
Precipitation radar in space
Global patterns—convective, stratiform, shallow
Cloud radars Ground based—trimodality of the population Satellite based—global distributions of MCSs, anvils, ...
Dual wavelength Water vapor

47. What we’ve learned
Spectrum of convective cloud types and sizes covers a wide range of types and sizes of convective entities
Mesoscale systems with stratiform rain
Top-heavy heating profiles
Multimodal size distributions
Shallow isolated cells
Structures of large anvil clouds
Global variability of the population

48. Where we are going
How does convective population project onto larger-scale dynamics?
Latent heating profiles
Radiative heating profiles in anvils of MCSs
Nonprecipitating convective clouds
Relation between humidity field and cloud population evolution
Role of clouds in MJO, ENSO, monsoon, & coupled equatorial waves

49. This research was supported by
End
This research was supported by
NASA grants NNX07AD59G, NNX07AQ89G, NNX09AM73G, NNX10AH70G, NNX10AM28G,
NSF grants, ATM , ATM ,
DOE grant DE-SC / ER-6

50. Extra Slides

51. Internal structure of “thick” anvils shown by CloudSat
Africa
Indian Ocean
Yuan, Houze, and Heymsfield 2011

52. Internal structures of MCS anvils
Data from MCSs seen by ARM W-band radar in Niamey, Niger
Height (km)
WAVELENGTH: W-band, 3 mm wavelength
THIN ANVILS:
1) LEADING—higher -- comes directly from convective updraft
2) TRAILING—occur at more levels -- moisten a deeper layer
THICK ANVILS:
3) LEADING—stronger-larger particles -- graupel
Cetrone & Houze 2011
and also
Yuan et al CloudSat
Reflectivity (dBZ)

53. Internal structures of MCS anvils
Data from MCSs seen by ARM W-band radar in Niamey, Niger
Height (km)
WAVELENGTH: W-band, 3 mm wavelength
THIN ANVILS:
1) LEADING—higher -- comes directly from convective updraft
2) TRAILING—occur at more levels -- moisten a deeper layer
THICK ANVILS:
3) LEADING—stronger-larger particles -- graupel
Cetrone & Houze 2011
and also
Yuan et al CloudSat
Reflectivity (dBZ)