Long Term Change Point Detections in Total Ozone Column over East Africa via Maximal Overlap Discrete Wavelet Transform
1 Department of Physics, University of Nairobi, Nairobi, Kenya
2 Department of Meteorology, University of Nairobi, Nairobi, Kenya
3 Department of Science, Technology and Engineering, Kibbie University, Bungoma, Kenya
Citation: J. W. Makokha, H. K. Angeyo, N.J. Muthama. “Long Term Change Point Detections in Total Ozone Column over East Africa via Maximal Overlap Discrete Wavelet Transform”, American Research Journal of Physics, Volume 2; pp:1-9
Abstract
ABSTRACT: Change point analysis (CPA) for the detection of both natural or artificial discontinuities and
regime shifts in Total Ozone Column (TOC) aids in inferring its influence on regional climate change. Assessment
of temporal variability in climate is complex and requires the utilization climate models that are known to
exhibit autocorrelation which enhances their capabilities in detecting either gradual or abrupt changes in
TOC. Normally, changes in TOC may be associated to instrumentation changes or anthropogenic influence.
This study presents for the first time, the utilization of Maximal Overlap Discrete Wavelet Transform (MODWT)
in performing long term change point detections in TOC over East African from 1978 to 2013. This is because
MODWT is not affected by circular shifting of the input series and also, its multiresolution capabilities allow
for long term change point detections in TOC. Additionally, MODWT automatically separates the trend from the
time series data therefore estimating the autocorrelated trend data. Results show that utilization of MODWT
reveals no interannual change points detected in the TOC measurements over the region, therefore implying
that TOC properties remained relatively constant interannually during the study period. On the seasonal scale,
TOC variability was evident and may be connected to biomass burning as well as the temporal evolution in
precursor emissions such as carbon dioxide (), Methane () and Nitrogen oxides (). Photochemical
oxidation during the December-January-February (DJF) season characterized by elevated temperatures
explains the enhanced variability in TOC. However, dry spells with minimal temperature experienced (during
June-July-August (JJA)) may explain the single or no observed change points during the study period. During
wet season i.e. March-April-May (MAM) and September-October-November (SON) the TOC variability
may be associated with biomass and refuse burning, lightening and extreme rainfall and intrusion of
stratospheric air over the region.
