Development of air quality indices using multivariate analysis | |
| Author | Rupakheti, Maheswar |
| Call Number | AIT Thesis no.EV-00-9 |
| Subject(s) | Air quality indexes Multivariate analysis |
| Note | A thesis submitted in partial fulfillment of the requirement for the degree of Master of Science |
| Publisher | Asian Institute of Technology |
| Abstract | Three multivariate techniques (principal component analysis, cluster analysis and canonical correlation analysis) were used to analyze two years (1997-98) daily average data on CO (lh), CO (8h), N02 (lh), S02 (lh), 0 3(1h) and PM10 (24h) concentration as well as meteorological data including relative humidity, temperature, wind speed, wind direction, pressure and global radiation (all are lh average), monitored at 13 stations in the city of Bangkok with the purpose to develop new air quality indices. Principal component analyses were applied to the data representing winter, summer, rainy and entire periods at every station. The results of principal component analyses are used to investigate the underlying causes of pollution. It has been found that the first three principal components extracted from the air pollution data are related to automobile exhaust, secondary pollutants and sulfur containing fossil fuel combustion (mostly from the stationary sources) respectively. The first three principal components from meteorological data are related to low wind speed, temperature and solar radiation, dominant wind direction and relative humidity respectively. The meteorological elements however load differently for different seasons. The result of cluster analysis shows that the monitoring stations can be divided into four main groups. Moreover, it is also found that the group produced can be interpreted based on the three principal components attributed to three major air pollution sources (principal components) mentioned above. The canonical correlation analysis determined the relationships between the two different data sets (air pollution data set and meteorological data set). The main relationship is between automobile exhaust pollutants and low speed wind in combination with high temperature. The air quality subindices were computed for different source categories based on the principal components and then those subindices were aggregated to give a single air quality index. To simplify the index computation procedure linear multiple regression equations were developed between the subindices and air pollution parameters. The aggregated air quality index, when compared with other two indices PINDEX (another aggregated index) and Pollutants Standard Index, shows a good agreement in fluctuation pattern with PINDEX and a good agreement with Pollutant Standard Index in description level with pollutant standard index. |
| Year | 2000 |
| Type | Thesis |
| School | School of Environment, Resources, and Development (SERD) |
| Department | Department of Energy and Climate Change (Former title: Department of Energy, Environment, and Climate Change (DEECC)) |
| Academic Program/FoS | Environmental Engineering and Management (EV) |
| Chairperson(s) | Nguyen Thi Kim Oanh |
| Examination Committee(s) | Annachhatre, Ajit P.;Shrestha, Ram M.;Nguyen C. Thanh |
| Scholarship Donor(s) | Queen's Scholarship for the Asian Environment;Development Program (Royal Thai Government) |
| Degree | Thesis (M.Sc.) - Asian Institute of Technology, 2000 |