Degree Type


Date of Award


Degree Name

Master of Science


Geological and Atmospheric Sciences



First Advisor

Xiaoqing Wu


Mechanisms behind the Madden-Julian Oscillation (MJO) initiation are poorly understood, although there are a few theories based on the cloud evolution. The purpose to this study is to use cloud resolving model output to observe MJO evolution by analyzing cloud top information and cloud types. Cloud top data will be used to divide MJOs into stages, and study cloud types and cloud characteristics for each stage during each MJO.

We use a two-dimensional version of the Clark-Hall Cloud Resolving Model (CRM) with large-scale forcing input over the Northern Sounding Array of the Dynamics of the Madden-Julian Oscillation (DYNAMO) field campaign to produce convective outputs such as cloud mass flux, heat and moisture budgets, precipitation, etc. Large-scale forcing data (provided by Colorado State University) showed two periods of stronger forcing, and one period of weaker forcing. This study focuses on the periods of two MJOs linked with the stronger forcing. Observational temperature, moisture, radiation and precipitation biases show comparisons to the two MJOs linked with these stronger forcing periods, and one weaker MJO. The largest biases occurred during periods of weaker large-scale forcing. During periods of enhanced precipitation, convective rain had higher rates, while stratiform rain dominated the domain.

Cloud top height and temperature frequencies were examined to determine the stages of the two MJOs evaluated in the 90-day time frame (October 1 to December 29). Vertical profiles of different variables show the evolution during each of the four stages of the MJO; suppressed, developing, mature, and decay. The suppressed stages show moisture in the lower atmosphere and convection that remained mostly lower than 0˚C. Mostly low and medium level stratiform clouds appear in the suppressed stages for both MJOs. The developing stage reaches into the higher atmosphere as ice water content increases and total cloud mass flux increases with height. Higher stratiform clouds form in this stage, and few convective towers start to form. The mature stage shows peaks in all variables, as convective towers and anvil clouds are dominant at this time and reach to the troposphere. Lower level variables tend to decrease first before higher levels in the decaying stages as notably fewer lower stratiform clouds are present. MJO1 (October MJO) lasts longer than the MJO2 (November MJO), but MJO2 shows greater quantities in all variables. MJO2 has higher clouds present in each stage except for the decaying stage. The connections between the stages can be made to convection and a moistening of the atmosphere. CRM sensitivity test was performed with a 500 m resolution to reproduce MJO convection and clouds on a finer scale. Evolution of the clouds show agreement with MJO initiation forming in a stepwise fashion, and with the 'recharge-discharge' theory.


Copyright Owner

Cora Lynn Virgei



File Format


File Size

64 pages

Included in

Meteorology Commons