Return to search

Nature and characteristics of tropospheric ozone over Johannesburg.

The aim of this thesis is to examine the nature and characteristics of tropospheric ozone
over Johannesburg, South Africa. Ozone, water vapour and meteorological profile data,
which form part of the MOZAIC (Measurement of Ozone and Water Vapor by Airbus In-Service
Aircraft) database for the period 1995 to 2000 were utilized in this study.
The thesis is divided into two main parts. The first part deals with the computation of total
tropospheric ozone. A clear seasonal cycle, with ozone peaking in September and October
is found. It is suggested that the main reason for the spring maximum is biomass burning,
combined with prevailing anticyclonic circulation patterns, which facilitate the build-up of
ozone over the region. Variability in TTO is greatest in January, September and November
and least during autumn and winter (April to July). The lower day-to-day variability in
autumn and winter is a reflection of the more settled weather at this time. Interannual
variability is least in January and April to June. The autumn and winter ozone values are
more consistent and appear to represent background tropospheric ozone loadings on which
the dynamic and photochemical influences of other months are superimposed.
High TTO events (>30 DU) occurred predominantly during September and October.
Enhancements in the lower troposphere occurred mostly in September and seldom lasted
for more than 1-2 consecutive days. It is suggested that these events are most likely due to
effects of local surface pollution sources, either localised biomass burning or urban-industrial
effects. An extended period of enhancement in the 7-12 km layer occurred from
14-17 September 1998 and again on 20 September 1998. The extended duration of this
event suggests that it is due to an STE event. Confirmation of this was given in a case
study of a particular MOZAIC flight on 16 September 1998 from Johannesburg to Cape
Town.
The second part of the thesis deals with the classification of ozone profiles and is used to
find pattern and order within the profiles. TWINSPAN (Two-Way INdicator SPecies
ANalysis), a cluster analysis technique, was used to classify the profiles according to the
magnitude and altitude of ozone concentration. Six distinct groups of profiles have been
identified and their characteristics described.
The HYSPLIT (Hybrid Single-Particle Lagrangian Integrated Trajectory) trajectory model
was used to relate the profiles to the origin of air masses, revealing clearly defined source
regions. The mid-tropospheric peak in summer and the low to mid-tropospheric
enhancement in spring is attributed to continental areas over central Africa and long-range
transport while local sources are responsible for the winter low tropospheric enhancement.
Reduced ozone values are due to westerlies bringing in clean maritime air.
The classification has highlighted three important findings. Firstly, it has emphasized the
pronounced seasonality of ozone profiles. It is evident that seasons are dominated by
particular patterns and by inference, the processes and transport patterns that shape
individual profiles are seasonally dependent. Secondly, the widely recognized spring
maximum in tropospheric ozone has been confirmed in this classification, but a new and
equally high summer mid-tropospheric enhancement due to the penetration of tropical air
masses from continental regions in central Africa has been identified. Thirdly, it is
suggested that the computation of a mean profile and furthermore, extrapolation of trends
based on a mean profile is meaningless, particularly for a location on the boundaries of
zonally defined meteorological regimes. / Thesis (M.A.)-University of Natal, Durban, 2002.

Identiferoai:union.ndltd.org:netd.ac.za/oai:union.ndltd.org:ukzn/oai:http://researchspace.ukzn.ac.za:10413/4027
Date January 2002
CreatorsRaghunandan, Atham.
ContributorsDiab, Roseanne D.
Source SetsSouth African National ETD Portal
LanguageEnglish
Detected LanguageEnglish
TypeThesis

Page generated in 0.0029 seconds