Spatio -Temporal Modelling of Air Pollution in Malta
(2018) MASM01 20181Mathematical Statistics
- Abstract
- Air pollution has become a major worldwide concern due to the high levels of air pollutants emitted from industrial and traffic related activities. Exposure to air pollution has been linked to various negative health effects, ranging from asthma to chronic illnesses. Consequently, analyzing air pollution data has become an essential tool for giving insight about the potential
health effects. Spatial statistics is a field where such analysis is possible by dealing with geo-referenced data, i.e., including information about space and time.
This dissertation focuses on spatio-temporal patterns of air pollution in Malta. The main objective is to interpolate concentrations of the nitrogen dioxide ( NO 2 ) pollutant across the country. Two... (More) - Air pollution has become a major worldwide concern due to the high levels of air pollutants emitted from industrial and traffic related activities. Exposure to air pollution has been linked to various negative health effects, ranging from asthma to chronic illnesses. Consequently, analyzing air pollution data has become an essential tool for giving insight about the potential
health effects. Spatial statistics is a field where such analysis is possible by dealing with geo-referenced data, i.e., including information about space and time.
This dissertation focuses on spatio-temporal patterns of air pollution in Malta. The main objective is to interpolate concentrations of the nitrogen dioxide ( NO 2 ) pollutant across the country. Two models are presented: a standard Kriging model and a complex spatial-temporal model. The first model uses a Universal Kriging (UK) structure to interpolate concentrations at unobserved locations and/or times. The second model consists of a mean field that incor-
porates dependence on geographic covariates together with seasonal and long-term trends; and a residual field having a spatial correlation structure.
The models are applied to a dataset consisting of monthly NO 2 concentrations measured at 99 monitoring sites across Malta in 2014-2016. Geographic covariates such as elevation, population density, and distances to coast, roads and industrial areas are used to explain spatial and temporal variations in the NO 2 concentrations. The cross-validated R 2 of the UK and spatio-temporal models are 0.52 and 0.55 respectively. Reconstructions of NO 2 across Malta reveal interesting seasonal and spatial patterns in air pollution. The models are
implemented using the statistical software R. (Less)
Please use this url to cite or link to this publication:
http://lup.lub.lu.se/student-papers/record/8952155
- author
- Sheikh, Imran
- supervisor
- organization
- course
- MASM01 20181
- year
- 2018
- type
- H2 - Master's Degree (Two Years)
- subject
- language
- English
- id
- 8952155
- date added to LUP
- 2018-06-21 11:37:38
- date last changed
- 2018-06-21 11:37:38
@misc{8952155, abstract = {{Air pollution has become a major worldwide concern due to the high levels of air pollutants emitted from industrial and traffic related activities. Exposure to air pollution has been linked to various negative health effects, ranging from asthma to chronic illnesses. Consequently, analyzing air pollution data has become an essential tool for giving insight about the potential health effects. Spatial statistics is a field where such analysis is possible by dealing with geo-referenced data, i.e., including information about space and time. This dissertation focuses on spatio-temporal patterns of air pollution in Malta. The main objective is to interpolate concentrations of the nitrogen dioxide ( NO 2 ) pollutant across the country. Two models are presented: a standard Kriging model and a complex spatial-temporal model. The first model uses a Universal Kriging (UK) structure to interpolate concentrations at unobserved locations and/or times. The second model consists of a mean field that incor- porates dependence on geographic covariates together with seasonal and long-term trends; and a residual field having a spatial correlation structure. The models are applied to a dataset consisting of monthly NO 2 concentrations measured at 99 monitoring sites across Malta in 2014-2016. Geographic covariates such as elevation, population density, and distances to coast, roads and industrial areas are used to explain spatial and temporal variations in the NO 2 concentrations. The cross-validated R 2 of the UK and spatio-temporal models are 0.52 and 0.55 respectively. Reconstructions of NO 2 across Malta reveal interesting seasonal and spatial patterns in air pollution. The models are implemented using the statistical software R.}}, author = {{Sheikh, Imran}}, language = {{eng}}, note = {{Student Paper}}, title = {{Spatio -Temporal Modelling of Air Pollution in Malta}}, year = {{2018}}, }