In this work a theory for moisture interchange between cloud systems and their environment in the tropical atmosphere is developed. Data collected during the intensive operation period of TOGA-COARE (Tropical Ocean and Global Atmosphere-Coupled Ocean-Atmosphere Response Experiment) is then used to study, on a case-by-case basis, the nature of this interchange for different convective regimes. The kinematic characteristics of the studied systems for each of the ten cases are synthesized from Doppler measurements made by the two National Oceanic and Atmospheric Administration WP-3D aircraft, while the environmental conditions were retrieved from balloon soundings made nearby the studied region. For this ten-case sample, a strong correlation between system-top height and tropospheric moisture (in the layer between 1 and 10 km) is found. The analysis indicates that it is tropospheric moisture which is controlling the height of the clouds and not vice-versa. It is found that the role of small clouds is to moisten the environment while deep convective regimes dry their environment.

I want to thank the guidance and support that I received from my adviser Dr. David J. Raymond, who expended with me many hours discussing how convection works in general, and in particular over the tropical regions. I also want to thank the confidence deposited in me by Dr. Odon Sanchez, who first told me about the research opportunities in atmospheric sciences, and for his support that allow me to pursue further graduate studies in this area. I'm grateful to the members of my committee: Drs. Kenneth R. Mischwaner, Steve Schaffer, and Ken Eack for their prompt review of the manuscript and the suggestions they made to improve it. I particularly want to thank my fellow student Clifton Murray for proof reading the many versions of the manuscript, which considerably improved the way it was written. Special thanks goes to Sandy Kieft whose mastery and efficiency of administrative matters, took lots of worries from my head.

A good part of this work could not be possible with out the data provided by John Daugherty from NOAA/NSSL. Most of the data analysis was made using the in-house CANDIS software pioneered by David J. Raymond and further developed by many students here at New Mexico Tech. I also benefited from powerful packages such as Rlab by Ian Searle, and Gri by Dan E. Kelley. All my work was done on a Linux-debian platform.

I especially want to thank my source of continuous inspiration since I remember - my mother, who instilled in me her strong work ethic. My children, I have to thank no only for being a source of inspiration for me, but also for being so patient with me and give up much of our precious time together.

From all people that had helped me, there is one for who I can hardly express my gratitude, for supporting me in so many ways, for being the source of energy and joy that keep me going ... Thank You, Chata - my wife.

This work was supported by the following National Science Foundation grants:
ATM-9311735, ATM-9413289, ATM-9616290, and ATM-0082612.

Contents

• • Title
• Contents
• List of Figures
• List of Tables
• Part : Introduction
• I. Fundamental Equations
  • 1. Multicomponent fluids
  • 2. Standard techniques
    • 2.1 Budget balance
    • 2.2 Standard fixed-control-volume technique
  • 3. Governing equations
    • 3.1 Mass Continuity
    • 3.2 Dry air
    • 3.3 Individual components of water substance
    • 3.4 Total water substance
    • 3.5 Total Mass
    • 3.6 Linear Momentum equation
    • 3.7 Total energy equation
    • 3.8 Entropy
  • 4. Explicit relationships
    • 4.1 Internal energy
    • 4.2 Kinetic energy
    • 4.3 Potential energy
    • 4.4 Bernoulli function
    • 4.5 Entropy
  • 5. Tendency equations
    • 5.1 Water substance
    • 5.2 Total energy tendency
    • 5.3 Entropy tendency
    • 5.4 Getting rid of precipitation
  • 6. Moistening and drying tendency equations
• II. Data Analysis &
  Results
  • 7. Data Sources
  • 8. Soundings
  • 9. Mass Fluxes
  • 10. System Top and Tropospheric Moisture
  • 11. Lateral Export
• Part : Conclusion
  • A. Clear air sounding profiles
    • A.1 Moist energy
    • A.2 Moist Bernoulli function
    • A.3 Moist entropy
    • A.4 Measurement Error
  • B. Radar observations
    • B.1 Radar measurements in meteorology
    • B.2 Doppler Radar measurements
    • B.3 Airborne Doppler Radar
      • B.3.1 Scanning strategy -- FAST
      • B.3.2 Trajectory strategies
    • B.4 Wind field synthesis
      • B.4.1 Error analysis
  • C. Covariance Calculations
  • D. Divergence - Column Correction
• Bibliography

• 5.1.
• Sketch of a typical control volume used in this work.
• 7.1.
• This figure shows the area studied during field phase of TOGA-COARE...
• 8.1.
• Relative location between sounding and the aircraft target area for the case of January 16, 1993. The system was moving toward the east. The background shading is the infrared brightness temperature from the GMS-4 Japanese satellite taken at 2315 UTC. Infrared temperatures are in degrees Celsius. The dashed polygon represents the intensive flux array(IFA), while the target area is enclosed by the rectangle. The aircraft path (thin-solid line) reached the target area approximately at 2300 UTC. The sounding was launched at 2302 UTC by the Shiyan 3 (solid diamond at the most right corner of the IFA and ahead of the system).
• 9.1.
• Interpolation grid for the cases presented here. The vertical height is always 20 km and the horizontal dimensions vary from case to to case in order to acomodate all radar data collected during the period of interest.
• 9.2.
• Histogram of Cartesian speeds for the case of December 15, 1992. This case corresponds to a deep convective system. All speeds from grid points where large geometrical errors (greater than 1 - see text) were not included in the histogram.
• 10.1.
• This figure shows the average vertical mass flux for the ten case studies examined here. They are grouped according to environmental moisture and cloud top height. Cases on the left panel correspond to small systems in dry environments. Tall systems in dry environments are in the central panel while tall systems in a moist environment are in the right panel.
• 10.2.
• Vertical variation of the saturation deficits defined as the differences between saturated and environmental moist energy (left), moist entropy (center), and specific humidity (right), for all missions in this study.
• 10.3.
• Correlation between system-top height and three measures of tropospheric moisture: In the left panel the moisture is represented by the moist-energy saturation deficit, in the central by the moist-entropy saturation deficit, and in the right by the specific-humidity saturation deficit. In each case the saturation deficit is average over a layer extending from 1 to 10 km. The corresponding correlation coefficients for each case are given at the top of each panel.
• 11.1.
• Average detrained mass flux for the ten case studies examined here, grouped according to environmental moisture and cloud top height. Cases in the left panel correspond to small systems in dry environments. Tall systems in dry environments are in the central panel; deep systems in a moist environment are in the right panel.
• B.1.
• Sketch of a radar ray showing one gate.
• B.2.
• Range-ambiguous echos. The nth transmitted pulse and its echos are crosshatched. This example assumes that the larger echo at delay is unambiguous in range but the smaller echo, at delay ... (Figure taken from, Doviak (1984), pg. 45])
• B.3.
• Signals at three different Doppler frequencies that yield , when sampled, the same set of data. These Doppler frequencies are aliases of each other.(figure taken from, Doviak (1984) pg. 45])
• B.4.
• Fore/Aft scanning technique (FAST)
• B.5.
• Horizontal pseudo-dual Doppler coverage for tail Doppler radar using the fore/aft scanning technique(FAST).
• B.6.
• Examples of aircraft trajectories used during TOGA-COARE to survey convective systems. Panel (a) shows a so-called ``stack'' flown by WP3-42 on February 1, 1993. Panel (b) shows part of the mission flown by WP3-43 (dashed) and WP3-42 (solid) on December 12, 1992. The segments were time-coordinated to allow for a more dense sampling. This sampling geometry performed by two aircraft in FAST mode is called quad-Doppler.
• B.7.
• Radial velocity measurements inside a grid volume. The spheres represent radar-sample volumes (i.e. gates). Arrows at the corners represent the interpolated wind velocities.
• B.8.
• Horizontal wind velocities assuming zero vertical particle velocity for the case of December 15, 1992, surveyed by the WP3-42 during TOGA-COARE. The velocities are Earth-relative, but they are shown in the storm reference frame. (a) All velocities inferred. (b) Locations where the amplification factors are bigger than 1. (c) Velocities associated with the locations shown in panel (b). (d) Velocities that remain after thresholding out points where the amplification factors are larger than one.

• 7.1.
• Information about the data sources used in this work. The following are the meanings of the labels. BS = Balloon sounding: SH = R/V Shiyan 3 ; MW = R/V Moana Wave ; XY = R/V Xiangyanghong 5 ; KP = Kapingamarangi. ACR = radar data: BWP means both WP-3D aircraft. AD = Average distance between soundings and aircraft target area. LT = Balloon launching time; RTI = selected time intervals for Doppler analysis in each mission. The Cartesian components of the system velocity were calculated from the values reported in, Kingsmill and Houze (1999)] - they are positive in the east and north directions.
• 8.1.
• The constants in this table were arbitrarily chosen. Other constants used but not listed here, such as the latent heat at the reference points and the heat capacities, were taken from, Emanuel (1994).
• 8.2.
• Ttop is the temperature at the top of the control cylinder ....
• 10.1.
• For each mission, the moisture in the environment is estimated using sounding data ...
• 10.2.
• The first column shows the layer over which the average of the corresponding saturation deficit was taken. The rest of the columns show the correlation coefficient for each correlation.
• 11.1.
• Convective contributions to the budgets of moist energy (ME) and moist entropy (MS), respectively. The implied role of the convective system over its environment according to the analysis presented in chapter is also listed.