Ozone trend and concentration in Doha City: Time
series models versus neural network
Accepted 8th
April, 2018
Adil Yousif
Department of Math, Stat, and Physics, Qatar University, Doha, Qatar.
This study aimed to
investigate the concentration of the Ozone layer
in Doha City, compare between different air
pollutants and test its relationship with the
main meteorological factors. A comprehensive
time series analysis using artificial neural
network technique was conducted, and appropriate
models were determined for future forecast. The
bivariate correlation, as well as regression
analysis, indicated that there were no
significant relationships between the Ozone and
other pollutants. On the other hand, the Ozone
concentration was significantly related with all
meteorological factors. It is concluded that the
Ozone concentration is within Qatar standards
for air pollution. However, there was a linear
trend with a slight increase that needed to be
controlled. ANN outperformed time series models
such as non-seasonal data ARIMA and Holt’s trend
models.
Key words:
Ozone, Qatar, neural network, ARIMA, air
quality.
This is an open access article
published under the terms of the
Creative Commons Attribution
License, which permits unrestricted use, distribution, and
reproduction in any medium, provided the original work is
properly cited.
Cite this article as:
Yousif A (2018). Ozone trend and
concentration in Doha City: Time series models versus neural network. Acad. J.
Environ. Sci. 6(4): 107-112.
