Page 181 - IJES Special Issues for AIEC2016

International Journal of Environment and Sustainability, 2016, 5(2): 72-88 81

Figure 6: Stations required to generate O3 concentration maps (RMSE<7 and MAPE<25)

Table 3:

Performance measures for NOx simulations

No. of Optimal station(s) Interpol. RMSE r2 MAPE NSE ACFT
station(s) to be combination that can be method 5.999 0.865 1.020
eliminated eliminated IDW 7.470 0.808 3.792 0.848 0.961
1 11 IDW 9.084 0.619 4.006 0.731 0.984
2 3, 11 UK 10.766 0.439 6.591 0.599 1.025
3 3, 5, 11 UK 11.521 0.461 12.184 0.430 1.022
4 3, 5, 7, 11 IDW 12.119 0.438 15.621 0.459 1.019
5 3, 5, 7, 11, 14 UK 14.834 0.395 19.839 0.438 0.966
6 3 , 5, 7, 11, 14, 16 IDW 14.684 0.312 23.148 0.341 0.977
7 3 , 5, 6, 7, 11, 14, 16 IDW 27.468 0.287
8 1, 3 , 5, 6, 7, 11, 14, 16

The accuracy factor (ACFT) was 1.015, which is 10 and 7 are located in a dense residential area,
considered to be very good performance. The and stations 2 and 3 are in an industrial area. The
remaining 10 stations that are required to agricultural areas are found near station 9, and
produce satisfactory maps of ozone are shown in construction activities are reported near station
Figure 6. These stations could essentially be 14. Stations 1 and 16 are the city outskirts where
considered hotspots of O3 emissions. small scale industries are located.

The primary sources of ground level O3 are auto- 3.2 NOx
mobiles, cement and power plants, construction
activities and biogenic or natural sources. Small The simulation results for NOx concentrations
industries such as paint shops, dry cleaners and are shown in Table 3. For the one station elimi-
bakeries are also known to contribute O3. These nation scenario, it was found that station 11 was
10 stations fall within these areas. The stations the best one with a RMSE=5.999; MAPE=3.792

Science Target Inc. www.sciencetarget.com