31.08.2015 Views

AIR POLLUTION – MONITORING MODELLING AND HEALTH

air pollution – monitoring, modelling and health - Ademloos

air pollution – monitoring, modelling and health - Ademloos

SHOW MORE
SHOW LESS

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

Analytical Model for Air Pollution in the Atmospheric Boundary Layer<br />

Analytical Model for Air Pollution in the Atmospheric Boundary Layer 13<br />

Run Recursion c(x, y, z, t) Run Recursion c(x, y, z, t)<br />

depth (10 −7 g<br />

)<br />

m 3<br />

0 8.68 4.49 0 3.65 2.10 1.31<br />

1 10.45 3.66 1 7.28 2.23 1.47<br />

1 2 10.21 3.65 6 2 7.78 2.20 1.46<br />

3 10.21 3.65 3 7.78 2.20 1.46<br />

4 10.21 3.65 4 7.78 2.20 1.46<br />

5 10.21 3.65 5 7.78 2.20 1.46<br />

0 8.62 2.01 0 5.77 3.07 2.27<br />

1 6.60 2.30 1 6.73 3.33 1.65<br />

2 2 7.09 2.28 7 2 6.61 3.32 1.65<br />

3 7.09 2.28 3 6.60 3.32 1.65<br />

4 7.09 2.28 4 6.60 3.32 1.65<br />

5 7.09 2.28 5 6.60 3.32 1.65<br />

0 5.58 6.38 3.72 0 6.38 3.94 2.76<br />

1 10.76 7.97 4.51 1 6.40 4.91 1.91<br />

3 2 9.84 7.88 4.50 8 2 5.99 4.87 1.91<br />

3 9.79 7.88 4.50 3 5.97 4.87 1.91<br />

4 9.79 7.88 4.50 4 5.97 4.87 1.91<br />

5 9.80 7.88 4.50 5 5.97 4.87 1.91<br />

0 10.73 0 5.01 2.83 2.08<br />

1 15.32 1 5.73 3.05 2.20<br />

4 2 15.23 9 2 5.65 3.04 2.19<br />

3 15.24 3 5.64 3.04 2.19<br />

4 15.24 4 5.64 3.04 2.19<br />

5 15.24 5 5.64 3.04 2.19<br />

0 7.14 4.11 2.47<br />

1 9.11 4.77 3.78<br />

5 2 6.96 4.46 3.73<br />

3 6.65 4.47 3.73<br />

4 6.65 4.48 3.73<br />

5 6.68 4.48 3.73<br />

Table 2. Pollutant concentrations for nine runs at various positions of the Copenhagen<br />

experiment and model prediction by the 3D-GILTT approach with time dependent eddy<br />

diffusivity.<br />

To perform statistical comparisons between GILTT results against Copenhagen experimental<br />

data we consider the set of statistical indices described by Hanna (1989) and defined in by<br />

• NMSE (normalized mean square error) = (C o−C p ) 2<br />

C o C p<br />

,<br />

• COR (correlation coefficient) = (C o−C o )(C p −C p )<br />

σ o σ p<br />

,<br />

• FA2 = fraction of data (%, normalized to 1) for 0, 5 ≤ C p<br />

C o<br />

≤ 2,<br />

• FB (fractional bias) = C o−C p<br />

0,5(C o +C p ) ,<br />

• FS (fractional standard deviations) =<br />

σ 0 −σ p<br />

0,5(σ 0 +σ p ) , 51

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!