Use of dispersion model and satellite SO2 retrievals for environmental impact assessment of coal-fired power plants

The aim of this study is to investigate the impact of ten proposed plants along with three operating plants in canakkale province of Turkey where the proposed plants are within very close proximity. The province has the highest capacity of the planned plants and the region is also of interest due to its history, tourism and agriculture potential. Current SO2 pollution was assessed using ground observations and satellite retrievals where the impact of plants was better captured by satellite retrievals. Individual and cumulative impact from proposed and operating plants was simulated by CALPUFF for 2014. The study domain was 150 x 150 km(2), with 1 x 1 km2 cell size. The effect of changing meteorological inputs and domain size were investigated with simulations. Three cases were performed using meteorological inputs: from one surface and one radiosonde station (Case 1), 22 surface and one radiosonde station (Case 2), and 22 surface and two radiosonde stations (Case 3). Case 2 and 3 resulted in higher concentrations and showed larger affected regions than case 1 in all simulations. The cumulative impact of proposed plants indicated national annual and daily limit values were exceeded in Case 2 and 3. Hourly limit values were exceeded in all three cases. Simulations for two selected proposed plants were assessed for plant impact area given in environmental impact area reports. Results indicated the plant impact areas cannot be sufficient to determine the maximum SO2 concentrations in some cases and using single meteorology station data cannot represent the study area, especially regions with complex terrain and land-sea interactions such as canakkale province. Cumulative impact can be underestimated clue to small size of plant impact areas not including other plants. Lastly satellite retrievals arc better capturing the pollution than air quality monitoring stations which are strongly affected by meteorology. (C) 2019 Elsevier B.V. All rights reserved.