Satellite atmospheric particulate matter retrieval and mortality prediction to support decisions making
Particulate matter 10 micron diameter (PM10) is one of the key pollutant in assessing air quality. In terms of human exposure, it is the easiest pollutant to come into contact and accounted for 3.7 million premature death worldwide annually. PM10 has been addressed as a major environmental and healt...
Saved in:
| 主要作者: | |
|---|---|
| 格式: | Thesis |
| 出版: |
2017
|
| 主题: | |
| 标签: |
添加标签
没有标签, 成为第一个标记此记录!
|
| id |
my-utm-ep.79589 |
|---|---|
| record_format |
uketd_dc |
| spelling |
my-utm-ep.795892018-10-31T13:00:20Z Satellite atmospheric particulate matter retrieval and mortality prediction to support decisions making 2017 Yap, Xen Quan G70.39-70.6 Remote sensing Particulate matter 10 micron diameter (PM10) is one of the key pollutant in assessing air quality. In terms of human exposure, it is the easiest pollutant to come into contact and accounted for 3.7 million premature death worldwide annually. PM10 has been addressed as a major environmental and health risk by World Health Organization (WHO) where countries around the globe struggle to reduce exposure value to the minimum. Currently, gaps of pollution between ground stations network are left unmonitored. With the advancement of atmospheric aerosol studies using satellite imaging technology, the possibility to solve the gap was found. This study investigates the possibilities of using Moderate Resolution Imaging Spectroradiometer (MODIS); a satellite sensor to retrieve PM10 concentrations. There are three objectives: (i) to retrieve long-term ground level PM10 measurement using MODIS aerosol optical depth (AOD) across Peninsular Malaysia for year 2001 to 2006; (ii) to improve the accuracy of MODIS retrieved PM10 using a statistical model; and (iii) to support decisions making by formulating a satellite outdoor air pollution variance indexes (SODAPVI) in order to predict exposure mortality rate using MODIS retrieved PM10. Cross-validation was used to investigate the relationship between MODIS AOD and ground level PM10 concentrations, followed by mixed effect model enhancement which take into account both random and fixed errors to calibrate MODIS AOD for retrieve PM10 concentrations. The findings demonstrate (i) MODIS AOD has a good correlation of R = 0.6 with ground level PM10 concentrations in most part of Peninsular Malaysia; (ii) direct retrieval of monthly PM10 concentrations from MODIS AOD served quantitative use with root mean square error (RMSE) of ±12.9 μg/m3; (iii) a systematic error (or fixed error) was observed while comparing MODIS AOD product with ground level PM10 data that needs to be treated to reduce bias; (iv) by using the mixed effect model, calibration of the MODIS retrieved PM10 concentrations can achieve an improvement of 50% compared to conventional crossvalidation method; (v) the SODAPVI has the potential to be embedded into decision making processes due to the high similarity to available ground level data obtained from WHO. With these findings, the application of MODIS AOD product has a potential to be extended to air quality and epidemiology studies in Malaysia as well as ASEAN region. 2017 Thesis http://eprints.utm.my/id/eprint/79589/ phd doctoral Universiti Teknologi Malaysia, Faculty of Geoinformation and Real Estate Faculty of Geoinformation and Real Estate |
| institution |
Universiti Teknologi Malaysia |
| collection |
UTM Institutional Repository |
| topic |
G70.39-70.6 Remote sensing |
| spellingShingle |
G70.39-70.6 Remote sensing Yap, Xen Quan Satellite atmospheric particulate matter retrieval and mortality prediction to support decisions making |
| description |
Particulate matter 10 micron diameter (PM10) is one of the key pollutant in assessing air quality. In terms of human exposure, it is the easiest pollutant to come into contact and accounted for 3.7 million premature death worldwide annually. PM10 has been addressed as a major environmental and health risk by World Health Organization (WHO) where countries around the globe struggle to reduce exposure value to the minimum. Currently, gaps of pollution between ground stations network are left unmonitored. With the advancement of atmospheric aerosol studies using satellite imaging technology, the possibility to solve the gap was found. This study investigates the possibilities of using Moderate Resolution Imaging Spectroradiometer (MODIS); a satellite sensor to retrieve PM10 concentrations. There are three objectives: (i) to retrieve long-term ground level PM10 measurement using MODIS aerosol optical depth (AOD) across Peninsular Malaysia for year 2001 to 2006; (ii) to improve the accuracy of MODIS retrieved PM10 using a statistical model; and (iii) to support decisions making by formulating a satellite outdoor air pollution variance indexes (SODAPVI) in order to predict exposure mortality rate using MODIS retrieved PM10. Cross-validation was used to investigate the relationship between MODIS AOD and ground level PM10 concentrations, followed by mixed effect model enhancement which take into account both random and fixed errors to calibrate MODIS AOD for retrieve PM10 concentrations. The findings demonstrate (i) MODIS AOD has a good correlation of R = 0.6 with ground level PM10 concentrations in most part of Peninsular Malaysia; (ii) direct retrieval of monthly PM10 concentrations from MODIS AOD served quantitative use with root mean square error (RMSE) of ±12.9 μg/m3; (iii) a systematic error (or fixed error) was observed while comparing MODIS AOD product with ground level PM10 data that needs to be treated to reduce bias; (iv) by using the mixed effect model, calibration of the MODIS retrieved PM10 concentrations can achieve an improvement of 50% compared to conventional crossvalidation method; (v) the SODAPVI has the potential to be embedded into decision making processes due to the high similarity to available ground level data obtained from WHO. With these findings, the application of MODIS AOD product has a potential to be extended to air quality and epidemiology studies in Malaysia as well as ASEAN region. |
| format |
Thesis |
| qualification_name |
Doctor of Philosophy (PhD.) |
| qualification_level |
Doctorate |
| author |
Yap, Xen Quan |
| author_facet |
Yap, Xen Quan |
| author_sort |
Yap, Xen Quan |
| title |
Satellite atmospheric particulate matter retrieval and mortality prediction to support decisions making |
| title_short |
Satellite atmospheric particulate matter retrieval and mortality prediction to support decisions making |
| title_full |
Satellite atmospheric particulate matter retrieval and mortality prediction to support decisions making |
| title_fullStr |
Satellite atmospheric particulate matter retrieval and mortality prediction to support decisions making |
| title_full_unstemmed |
Satellite atmospheric particulate matter retrieval and mortality prediction to support decisions making |
| title_sort |
satellite atmospheric particulate matter retrieval and mortality prediction to support decisions making |
| granting_institution |
Universiti Teknologi Malaysia, Faculty of Geoinformation and Real Estate |
| granting_department |
Faculty of Geoinformation and Real Estate |
| publishDate |
2017 |
| _version_ |
1747818263354540032 |