الفهرس 
Spatio-temporal analysis of air pollutants
AbstractOnly 14 pages are availabe for public view
 |  | from | 128 |  |  |
| | | from | 128 | | |
Abstract
This study has two main objectives. The first is analyzing the spatiotemporal variability of air pollutants in Egypt using monthly averages from the air quality monitoring network during the period 2011 to 2015. Particulate matters (PM10), nitrogen dioxide (NO2) and sulfur dioxide (SO2), measured by the monitoring station network are studied. The second is improving the network of the monitoring stations. Concerning the first objective, three approaches are used, namely; cokriging, spatiotemporal kriging, and hierarchical Bayesian models. Interpolated maps are created using the three approaches. Cross validation is executed between the methods in order to determine the best methodology using Root Mean Squared errors (RMSE). The results indicated that cokriging gives better results especially when the months are correlated. This helps in determining the highly polluted regions besides the highly polluted months taking into consideration the effect of previous situations. For achieving the second objective, throughout this study two approaches for network improvement are adopted; the first depends on the approach proposed by Santacruz et al. (2014) where the simultaneous methods are applied for adding 11 extra stations to the network, measuring NO2, from which 4 stations should measure both SO2, as well as PM10, and only 3 stations to measure SO2 alongside NO2. The second approach consists of two steps; the first includes removing the stations that have the least effect on the increase in standard error from the network, and then in the second step the simultaneous approach is used to add the extra stations to the remaining ones