Ozonesonde Classification Using Self-Organizing Maps at Ascension Island and Natal
Open Access
- Author:
- Jensen, Anders Alstrup
- Graduate Program:
- Meteorology
- Degree:
- Master of Science
- Document Type:
- Master Thesis
- Date of Defense:
- April 05, 2011
- Committee Members:
- Anne Mee Thompson, Thesis Advisor/Co-Advisor
Anne Mee Thompson, Thesis Advisor/Co-Advisor - Keywords:
- ozone
self-organizing map
Ascension Island
Natal
shadoz
biomass burning
tropics - Abstract:
- Balloon-borne ozonesondes measure ozone to approximately 30 kilometers, and they reveal how ozone changes with height because of various origins, including anthropogenic sources, meteorological conditions, and stratospheric influence. These ozonesonde profiles can be clustered using self-organizing maps. Self-organizing maps use a learning algorithm to reveal the most prominent data features of a data set. These clusters can then be compared to different meteorological phenomena to determine the effects the meteorology has on different ozonesonde profile types. Analysis of ozonesondes from Ascension Island, located at 7.98S and 14.42W, and Natal, located at 5.42S and 35.38W, using self-organizing maps reveals different ozonesonde profile types that deviate from the average ozone. A 2x2 self-organizing map, which creates 4 clusters, reveals that deviations from the average ozone are both increased ozone resulting from seasonal biomass burning in Africa and convection vertically mixing ozone. Expanding to a 4x4 self-organizing map shows how biomass burning influences the yearly cycle of tropospheric ozone at Ascension Island and displays the seasonality of ozone at both Ascension Island and Natal. Finally, comparing Ascension Island and Natal using a 4x4 self-organizing map of both locations reveals similarities in mid-tropospheric ozone, but shows differences in lower tropospheric ozone due to Ascension Island being closer to biomoass burning from Africa and due to differences in the underlying meteorology.