Skip to main content
. Author manuscript; available in PMC: 2018 May 16.
Published in final edited form as: Atmos Chem Phys. 2017 May 8;17:5721–5750. doi: 10.5194/acp-17-5721-2017

Table 2.

(a) Evaluation of the period-mean (1 May–30 June 2010) multi-global model free simulations against the CASTNET observations, only at the sites where 95% of the hourly O3 observations are available. Evaluation of the individual models is summarized in (b). (b) Evaluation of the period-mean (May–June 2010) global model free simulations against the EANET and CASTNET observations. The STEM boundary condition models are highlighted in bold. (c) Evaluation of the period-mean (May–June 2010) multi-global model free simulations against the EANET observations in Japan and South Korea. Evaluation of the individual models is summarized in (b).

(a) Subregion US EPA regions
contained
Number of
sites
Mean bias
(ppbv)
RMSE
(ppbv)

3 BC*
models
8 global
models
3 BC
models
8 global
models
Western US 8, 9, 10 19 −5.68 −2.52 10.37 7.05
Southern US 4, 6 18 11.61 10.24 13.62 11.96
Midwest 5, 7 13 8.03 7.66 9.16 8.67
Northeast 1, 2, 3 17 9.55 10.63 10.28 11.24
All 1–10 67 5.49 6.22 11.11 9.96
(b) Network Number of
sites
RMSE
(ppbv)

CAM-
Chem
EMEP CHASER SNU–GEOS-
Chem
GEOS-Chem
adjoint
RAQMS OsloCTM3 v2 C-IFS
CASTNET 67 13.30 11.61 15.43 15.55 13.48 9.32 11.05 11.00
EANET 11 10.38 9.96 11.39 9.18 11.04 8.60 12.97 10.86
(c) Country Number of
sites
Mean bias
(ppbv)
RMSE
(ppbv)

3 BC*
models
8 global
models
3 BC
models
8 global
models
Japan 8 0.36 1.01 8.77 9.25
South Korea 3 1.14 3.98 8.37 10.51
All 11 0.57 1.82 8.66 9.61
*

BC: boundary conditions.