vvEPA
United Mates
Environmental Protection
Air Quality Modeling Technical Support Document: EGU
Mercury Analysis
-------
EPA-454/R-11-008
September 2011
Air Quality Modeling Technical Support Document: EGU
Mercury Analysis
U.S. Environmental Protection Agency
Office of Air Quality Planning and Standards
Air Quality Assessment Division
Research Triangle Park, NC 27711
-------
I Introduction
This document describes the air quality modeling performed by EPA in support of air quality and
mercury deposition assessments related to large stationary point sources that generate electricity.
A national scale air quality modeling analysis was performed to estimate the impact of the sector
emissions changes on future year annual and 24-hour PM2.5 concentrations, 8-hr maximum
ozone, total mercury deposition, as well as visibility impairment. Air quality benefits are
estimated with the Community Multi-scale Air Quality (CMAQ) model. CMAQ simulates the
numerous physical and chemical processes involved in the formation, transport, and destruction
of ozone, particulate matter and air toxics. In addition to the CMAQ model, the modeling
platform includes the emissions, meteorology, and initial and boundary condition data which are
inputs to this model.
Emissions and air quality modeling decisions are made early in the analytical process. For this
reason, it is important to note that the inventories used in the air quality modeling and the
benefits modeling may be slightly different than the final utility sector inventories presented in
the RIA. However, the air quality inventories and the final rule inventories are generally
consistent, so the air quality modeling adequately reflects the effects of the rule.
II. Photochemical Model Version, Inputs and Configuration
Photochemical grid models use state of the science numerical algorithms to estimate pollutant
formation, transport, and deposition over a variety of spatial scales that range from urban to
continental. Emissions of precursor species are injected into the model where they react to form
secondary species such as ozone and then transport around the modeling domain before
ultimately being removed by deposition or chemical reaction.
The 2005-based CMAQ modeling platform was used as the basis for the air quality modeling for
this rule. This platform represents a structured system of connected modeling-related tools and
data that provide a consistent and transparent basis for assessing the air quality response to
projected changes in emissions. The base year of data used to construct this platform includes
emissions and meteorology for 2005. The modeling system treats the emissions, transport, and
fate of criteria pollutants and certain toxics including hydrogen chloride (HCL) and speciated
mercury: Hg(0) (gaseous elemental), Hg(II) (oxidized gaseous), and Hg(p) (particle-bound). This
modeling platform and analysis is described below.
As part of the analysis for this rulemaking, the modeling system was used to calculate daily and
annual PM2.5 concentrations, 8-hr maximum ozone, annual total mercury deposition levels and
visibility impairment. Model predictions are used to estimate future-year design values of PM2.5
and ozone. Specifically, we compare a 2016 reference scenario, a scenario without the boiler
sector controls, to a 2016 control scenario which includes the adjustments to the boiler sector.
This is done by calculating the simulated air quality ratios between any particular future year
simulation and the 2005 base. These predicted ratios are then applied to ambient base year
design values. The design value projection methodology used here followed EPA guidance for
such analyses (USEPA, 2007). Additionally, the raw model outputs are also used in a relative
sense as inputs to the health and welfare impact functions of the benefits analysis. Only model
-------
predictions for mercury deposition were analyzed using absolute model changes, although
percent changes between the control case and two future baselines are also estimated.
A. Model version
The Community Multi-scale Air Quality (CMAQ) model v4.7.1 (www.cmaq-model.org) is a
state of the science three-dimensional Eularian "one-atmosphere" photochemical transport model
used to estimate air quality (Appel et al., 2008; Appel et al., 2007; Byun and Schere, 2006).
CMAQ simulates the formation and fate of photochemical oxidants, ozone, primary and
secondary PM concentrations, and air toxics over regional and urban spatial scales for given
input sets of meteorological conditions and emissions. CMAQ is applied with the AERO5
aerosol module, which includes the ISORROPIA inorganic chemistry (Nenes et al., 1998) and a
secondary organic aerosol module (Carlton et al., 2010). The CMAQ model is applied with
sulfur and organic oxidation aqueous phase chemistry (Carlton et al., 2008) and the carbon-bond
2005 (CB05) gas-phase chemistry module (Gery et al., 1989). Mercury oxidation pathways are
represented for both the gas and aqueous phases in addition to aqueous phase reduction reactions
(Bullock and Brehme, 2002). Mercury estimates from CMAQ have been compared to
observations and other mercury modeling systems in several peer reviewed publications (Bullock
et al., 2008, 2009; Lin et al., 2007).
B. Model domain and grid resolution
The modeling analyses were performed for a domain covering the continental United States as
shown in Figure II-l. This domain has a parent horizontal grid of 36 km with two finer-scale 12
km grids over portions of the eastern and western U.S. The model extends vertically from the
surface to 100 millibars (approximately 15 km) using a sigma-pressure coordinate system. Air
quality conditions at the outer boundary of the 36 km domain were taken from a global model
and vary in time and space. The 36 km grid was only used to establish the incoming air quality
concentrations along the boundaries of the 12 km grids. Only the finer grid data were used in
determining the impacts of the emissions changes. Table II-l provides geographic information
about the photochemical model domains.
Table II-l. Geographic elements of domains used in photochemical modeling.
Map Projection
Grid Resolution
Coordinate Center
True Latitudes
Dimensions
Vertical extent
Photochemical Modeling Configuration
National Grid
Western U.S. Fine Grid
Eastern U.S. Fine Grid
Lambert Conformal Projection
36km
12km
12km
97 deg W, 40 deg N
33 deg N and 45 deg N
148 x 112x 14
213 x 192 x 14
279 x 240 x 14
14 Layers: Surface to 100 millibar level (see Table II-3)
-------
Figure II-1. Map of the photochemical modeling domains. The black outer box denotes the 36
km national modeling domain; the red inner box is the 12 km western U.S. grid; and the blue
inner box is the 12 km eastern U.S. grid.
C. Modeling Time-period
The 36 km and both 12 km modeling domains were modeled for the entire year of 2005. Data
from the entire year were utilized when looking at the estimation of PM2.5, total mercury
deposition, and visibility impacts from the regulation. Data from April through October is used
to estimate ozone impacts.
D. Model Inputs: Emissions, Meteorology and Boundary Conditions
The 2005-based modeling platform was used for the air quality modeling of future emissions
scenarios. In addition to the photochemical model, the modeling platform also consists of the
base- and future-year emissions estimates, meteorological fields, as well as initial and boundary
condition data which are all inputs to the air quality model.
1. Emissions Input Data
The emissions data used in the base year and future reference and future emissions adjustment
case are based on the 2005 v4.1 platform. The emissions cases use different emissions data for
some pollutants than the official v4 platform to use data intended only for this rule. Unlike the
2005 v4 platform, the configuration for this modeling application included mercury emissions
-------
from the National Air Toxics Assessment Inventory and some industrial boiler sector mercury
emissions more consistent with the engineering analysis for the
Industrial/Commercial/Institutional Boilers and Process Heaters NESHAP. Emissions for the
future years for the EGU sector utilized information collected from the utility MACT
information collection request. The information collection request informed existing HCL, NOX,
HG, and PM controls. In addition this data was used to supply HCL removal rates from selected
control technology. Emissions are processed to photochemical model inputs with the SMOKE
emissions modeling system (Houyoux et al., 2000).
The 2016 reference case is intended to represent the emissions associated with growth and
controls in that year projected from the 2005 simulation year. The United States EGU point
source emissions estimates for the future year reference and control case are based on an
Integrated Planning Model (IPM) run for criteria pollutants, hydrochloric acid, and mercury in
2016. Both control and growth factors were applied to a subset of the 2005 non-EGU point and
non-point to create the 2016 reference case. The 2005 v4 platform 2014 projection factors were
the starting point for most of the 2016 SMOKE-based projections. The mercury projections for
non-EGU point sources accounted for emission reductions expected in the future due to
NESHAP for various non-EGU source categories that were finalized or expected to be finalized
prior to the rule proposal including the Boiler MACT, Gold Mine NESHAP and Electric Arc
Furnace NESHAP. The estimated total anthropogenic emissions and emissions for the utility
sector used in this modeling assessment are shown in Table II.2.
Table II.2 Estimated total inventory and EGU sector emissions for each modeling scenario.
Emissions (tons/year)
Scenario Sector VOC NOx CO SO2 PM10 PM2.5
2005 baseline EGU(PTIPM) 40,950 3,726,459 601,564 10,380,786 615,095 508,903
All 17,613,543 22,216,093 83,017,436 15,050,209 13,031,716 4,400,680
2016 baseline EGU(PTIPM) 40,845 1,769,764 691,310 3,577,698 523,504 384,320
All 14,390,421 15,019,836 59,148,384 7,245,595 12,772,091 4,022,846
2016control case EGU(PTIPM) 38,217 1,618,199 656,245 1,220,379 358,165 291,044
All 14,387,792 14,868,270 59,113,319 4,888,276 12,606,752 3,929,570
Scenario
2005 baseline
2016 baseline
2016 control case
Emissions (tons/year)
Sector Hg(0) Hg(ll) Hg(p) HCL
EGU(PTIPM)
All
EGU(PTIPM)
All
EGU(PTIPM)
All
30
64
21
42
5
26
21
33
7
16
2
11
1.6
8.5
0.7
5.9
0.4
5.6
351,592
429,223
74,089
140,638
8,802
75,351
CL2
99
6,409
6,050
6,050
NH3
21,684
3,762,641
36,655
3,897,033
36,982
3,897,360
Other North American emissions of criteria and toxic pollutants (including mercury) are based
on a 2006 Canadian inventory and 1999 Mexican inventory. Both inventories are not grown or
-------
controlled when used as part of future year baseline inventories. Global emissions of criteria and
toxic pollutants (including mercury) are included in the modeling system through boundary
condition inflow.
All mercury emissions from facilities included in the PTIPM sector were removed, or "zeroed-
out" in both the 2005 baseline and 2016 baseline scenarios to provide information about the
contribution of mercury from this sector.
2. Meteorological Input Data
The gridded meteorological input data for the entire year of 2005 were derived from simulations
of the Pennsylvania State University / National Center for Atmospheric Research Mesoscale
Model. This model, commonly referred to as MM5, is a limited-area, nonhydrostatic, terrain-
following system that solves for the full set of physical and thermodynamic equations which
govern atmospheric motions. Meteorological model input fields were prepared separately for
each of the three domains shown in Figure II-l using MM5 version 3.7.4. The MM5 simulations
were run on the same map projection as shown in Figure II-l.
All three meteorological model runs were configured similarly. The selections for key MM5
physics options are shown below:
• Pleim-Xiu PEL and land surface schemes
• Kain-Fritsh 2 cumulus parameterization
• Reisner 2 mixed phase moisture scheme
• RRTM longwave radiation scheme
• Dudhia shortwave radiation scheme
Three dimensional analysis nudging for temperature and moisture was applied above the
boundary layer only. Analysis nudging for the wind field was applied above and below the
boundary layer. The 36 km domain nudging weighting factors were 3.0 x 104 for wind fields and
temperatures and 1.0 x 105 for moisture fields. The 12 km domain nudging weighting factors
were 1.0 x 104 for wind fields and temperatures and 1.0 x 105 for moisture fields.
-------
Table II-3. Vertical layer structure (heights are layer top).
CMAQ Layers
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
MM5 Layers
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
Sigma P
1.000
0.995
0.990
0.985
0.980
0.970
0.960
0.950
0.940
0.930
0.920
0.910
0.900
0.880
0.860
0.840
0.820
0.800
0.770
0.740
0.700
0.650
0.600
0.550
0.500
0.450
0.400
0.350
0.300
0.250
0.200
0.150
0.100
0.050
0.000
Approximate Height
(m)
0
38
77
115
154
232
310
389
469
550
631
712
794
961
1,130
1,303
1,478
1,657
1,930
2,212
2,600
3,108
3,644
4,212
4,816
5,461
6,153
6,903
7,720
8,621
9,625
10,764
12,085
13,670
15,674
Approximate Pressure
(mb)
1000
995
991
987
982
973
964
955
946
937
928
919
910
892
874
856
838
820
793
766
730
685
640
595
550
505
460
415
370
325
280
235
190
145
100
All three sets of model runs were conducted in 5.5 day segments with 12 hours of overlap for
spin-up purposes. All three domains contained 34 vertical layers with an approximately 38m
deep surface layer and a 100 millibar top. The MM5 and CMAQ vertical structures are shown in
Table II-3 and do not vary by horizontal grid resolution. The meteorological outputs from all
three MM5 sets were processed to create model-ready inputs for CMAQ using the MCIP
processor.
Before initiating the air quality simulations, it is important to identify the biases and errors
associated with the meteorological modeling inputs. The 2005 MM5 model performance
evaluations used an approach which included a combination of qualitative and quantitative
analyses to assess the adequacy of the MM5 simulated fields. The qualitative aspects involved
comparisons of the model-estimated synoptic patterns against observed patterns from historical
weather chart archives. Additionally, the evaluations compared spatial patterns of estimated to
observed monthly average rainfall and checked maximum planetary boundary layer (PEL)
heights for reasonableness.
Qualitatively, the model fields closely matched the observed synoptic patterns, which is not
unexpected given the use of nudging. The operational evaluation included statistical
-------
comparisons of model/observed pairs (e.g., mean normalized bias, mean normalized error, index
of agreement, root mean square errors, etc.) for multiple meteorological parameters. For this
portion of the evaluation, five meteorological parameters were investigated: temperature,
humidity, shortwave downward radiation, wind speed, and wind direction. The three individual
MM5 evaluations are described elsewhere (Baker, 2009a, b, c). It was ultimately determined that
the bias and error values associated with all three sets of 2005 meteorological data were
generally within the range of past meteorological modeling results that have been used for air
quality applications.
3. Initial and Boundary Conditions
The lateral boundary and initial species concentrations are provided by a three-dimensional
global atmospheric chemistry model, the GEOS-CHEM model (standard version 7-04-11). The
global GEOS-CHEM model simulates atmospheric chemical and physical processes driven by
assimilated meteorological observations from the NASA's Goddard Earth Observing System
(GEOS). This model was run for 2005 with a grid resolution of 2.0 degree x 2.5 degree
(latitude-longitude) and 30 vertical layers up to 100 mb. The predictions were used to provide
one-way dynamic boundary conditions at three-hour intervals and an initial concentration field
for the 36 km CMAQ simulations. The 36 km photochemical model simulation is used to supply
initial and hourly boundary concentrations to the 12 km domains. Initial and boundary conditions
for the projected future year (2016) 36 km simulations are the same as the 2005 base year. The
first 10 days of the 36 km modeling simulation are not used in the analysis, which is beyond the
number of days necessary to remove the influence of initial conditions on mercury deposition
estimates (Pongprueksa et al., 2008).
Mercury initial and boundary conditions were based on a GEOS-CHEM simulation using a 2000
based global anthropogenic emissions inventory that includes 1,278 Mg/yr of Hg(0), 720 Mg/yr
of Hg(II), and 192 Mg/yr of particle bound mercury (Selin et al., 2007). A comparison of global
mercury emissions by continent for 2000 and 2006 is published in (Streets et al., 2009). Total
mercury emissions from China (and Oceania) total 1,306 Mg/yr in 2000 and 1,317 Mg/yr in
2006 (Streets et al., 2009). Given these consistent emissions estimates from Asia, the 2005
boundary inflow to the 36 km CMAQ domain was not adjusted. Recent research has shown that
ambient mercury concentrations have been decreasing in the northern hemisphere since 2000
(Slemr et al., 2011). Since emissions from China have not appreciably changed between 2000
and 2006, ambient mercury concentrations have been decreasing, and the large uncertainties
surrounding projected mercury global inventories the mercury boundary conditions are the same
for both the 2005 and 2016 simulations.
III. Base Case Model Performance Evaluation
A. PM2.5
An operational model performance evaluation for the speciated components of PM2.5 (e.g.,
sulfate, nitrate, elemental carbon, organic carbon, etc.) was conducted using 2005 state/local
monitoring data in order to estimate the ability of the modeling system to replicate base year
concentrations. The evaluation of PM2.5 component species includes comparisons of predicted
-------
and observed concentrations of sulfate (SO4), nitrate (NO3), ammonium (NH4), elemental
carbon (EC), and organic carbon (OC). PM2.5 ambient measurements for 2005 were obtained
from the Chemical Speciation Network (CSN) and the Interagency Monitoring of PROtected
Visual Environments (IMPROVE). The CSN sites are generally located within urban areas and
the IMPROVE sites are typically in rural/remote areas. The measurements at CSN and
IMPROVE sites represent 24-hour average concentrations. In calculating the model performance
metrics, the modeled hourly species predictions were aggregated to the averaging times of the
measurements.
V H-TT
Speciated PM2.5 Monitors
Figure III-l. Speciated PM2.5 monitors used in the model performance evaluation.
Model performance statistics were calculated for observed/predicted pairs of daily
concentrations. Estimated metrics include bias, error, fractional bias, and fractional error (Boylan
and Russell, 2006). The aggregated metrics and number (N) of prediction-observation pairs are
shown by chemical specie and quarter in Table III-l. The "acceptability" of model performance
was judged by comparing our 2005 performance results to the range of performance found in
recent regional PM2.5 model applications for other, non-EPA studies. Overall, the mean bias
(bias) and mean error (error) statistics shown in Table III-l are within the range or close to that
found by other groups in recent applications (Doraiswamy, 2010; Tesche et al., 2006). The
model performance results give us confidence that our application of CMAQ using this modeling
platform provides a scientifically credible approach for assessing PM2.5 concentrations for the
purposes of this assessment.
-------
TABLE III-l. Model performance metrics for speciated PM2.5 averaged by quarter.
Quarter
N
Mean Observed (ug/m3)
Mean Predicted (ug/m )
Bias (ug/m )
Error (ug/m )
Fractional Bias (%)
Fractional Error (%)
Sulfate Ion
Nitrate Ion
Ammonium Ion
Organic Carbon
Elemental Carbon
Sulfate Ion
Nitrate Ion
Ammonium Ion
Organic Carbon
Elemental Carbon
Sulfate Ion
Nitrate Ion
Ammonium Ion
Organic Carbon
Elemental Carbon
Sulfate Ion
Nitrate Ion
Ammonium Ion
Organic Carbon
Elemental Carbon
Sulfate Ion
Nitrate Ion
Ammonium Ion
Organic Carbon
Elemental Carbon
Sulfate Ion
Nitrate Ion
Ammonium Ion
Organic Carbon
Elemental Carbon
Sulfate Ion
Nitrate Ion
Ammonium Ion
Organic Carbon
Elemental Carbon
1
6,927
6,577
4,752
7,049
6,977
2.5
2.5
1.7
0.6
0.4
2.3
2.3
1.5
0.9
5.5
-0.2
-0.1
0.0
0.4
5.1
1.0
1.5
0.8
0.5
5.2
-6.6
-17.9
6.2
33.3
159.6
42.9
76.7
51.8
61.9
161.5
2
7,248
6,850
4,777
7,219
7,182
3.6
0.9
1.4
0.5
0.7
3.2
1.2
1.4
0.7
4.1
-0.3
0.4
0.3
0.2
3.6
1.2
0.9
0.7
0.4
3.7
-6.0
-24.8
28.5
17.7
124.0
35.8
95.8
58.3
61.1
134.0
3
6,819
6,532
4,576
6,726
6,830
5.1
0.6
1.7
0.5
0.8
4.1
0.5
1.3
0.7
4.8
-0.8
0.0
-0.1
0.2
4.1
1.6
0.5
0.7
0.4
4.3
-13.4
-67.0
10.4
15.6
127.9
38.7
111.7
57.8
59.1
136.9
4
6,372
6,240
4,303
6,386
6,464
2.3
1.6
1.3
0.7
0.6
2.2
1.8
1.3
0.9
5.8
-0.1
0.3
0.2
0.3
5.3
0.7
1.3
0.7
0.5
5.3
1.0
-6.3
31.1
18.5
151.8
36.8
91.1
60.8
55.8
154.0
-------
B. Ozone
An operational model performance evaluation for hourly and eight-hour daily maximum ozone
was conducted in order to estimate the ability of the modeling system to replicate the base year
concentrations. Ozone measurements were taken from the 2005 State/local monitoring site data
in the Air Quality System (AQS) Aerometric Information Retrieval System (AIRS). The ozone
metrics covered in this evaluation bias, error, fractional bias, and fractional error for both daily
maximum 1-hr ozone and daily maximum 8-hr ozone (Boylan and Russell, 2006).
Figure III-2. Ozone monitors used in the model performance evaluation.
The evaluation principally consists of statistical assessments of model versus observed pairs that
were paired in time and space. This ozone model performance was limited to the prediction-
observation pairs where observed ozone exceeded or equaled 60 ppb. This cutoff was applied to
evaluate the model on days of elevated ozone which are more policy relevant. Aggregated
performance metrics by quarter are shown in Table III-2.
10
-------
TABLE III-2. Model performance metrics for daily maximum ozone by quarter.
Month
N
Mean observed (ppb)
Mean predicted (ppm)
Bias (ppb)
Error (ppb)
Fractional bias (%)
Fractional error (%)
Daily peak 1-hr ozone
Daily peak 8-hr ozone
Daily peak 1-hr ozone
Daily peak 8-hr ozone
Daily peak 1-hr ozone
Daily peak 8-hr ozone
Daily peak 1-hr ozone
Daily peak 8-hr ozone
Daily peak 1-hr ozone
Daily peak 8-hr ozone
Daily peak 1-hr ozone
Daily peak 8-hr ozone
Daily peak 1-hr ozone
Daily peak 8-hr ozone
5
7,173
7,180
75
68
68
63
-7
-6
9
8
-9
-9
13
12
6
9,553
9,557
79
71
74
68
-5
-4
9
8
-7
-6
12
11
7
9,522
9,529
81
72
78
70
-3
-1
11
9
-5
-2
14
12
8
8,433
8,437
81
71
77
69
-4
-2
11
9
-6
-4
14
13
9
7,118
7,120
78
70
72
64
-6
-5
10
9
-9
-8
14
13
This model performance is consistent with photochemical modeling used to support other
national regulations (USEPA, 2010).
C. Mercury Wet Deposition
Model estimated weekly mercury wet deposition is compared to observation data to assess model
skill simulating this component of mercury deposition. Mercury wet deposition measurements
are weekly totals taken at sites that are part of the Mercury Deposition Network
(http://nadp. sws.uiuc.edu/MDN/) which operates under the National Atmospheric Deposition
Program. In addition to mercury wet deposition, the network sites also collect rainfall data which
is also evaluated against estimates used by the photochemical model from prognostic
meteorological model output. Previous versions of the CMAQ modeling system has been applied
by other researchers at a continental and regional scale and evaluation has been published
(Bullock et al., 2008, 2009; Lin et al., 2007; Pongprueksa et al., 2008; Vijayaraghavan et al.,
2007).
Model performance is characterized using a variety of statistical metrics common in
photochemical model evaluation journal articles: bias, error, fractional bias, and fractional error
(Boylan and Russell, 2006). These metrics are estimated for total mercury wet deposition and for
rainfall. Performance is best when the metrics approach 0. The fractional bias and error metrics
are bound by ±200%, which would be considered poor performance.
11
-------
12 km domain evaluation
36 km domain evaluation
Figure III-l. Mercury deposition network monitors used for evaluation of the 12 and 36 km
domains.
CMAQ estimates of total mercury wet deposition are paired in space and time with MDN
observations. Model performance metrics are averaged by season and shown in Table III-3 for
the 12 km domain and Table III-4 for the 36 km model domain. Other published mercury
modeling studies show a positive bias for annual total mercury wet deposition (Bullock et al.,
2009; Lin et al., 2007; Vijayaraghavan et al., 2007). This CMAQ application also shows an over
prediction bias except during the summer months in the eastern United States.
An annual mercury modeling application done by ENVIRON and Atmospheric and
Environmental Research for Lake Michigan Air Directors Consortium show seasonal average
normalized bias between 70 and 158% and seasonal average normalized error between 72 and
503% (Yarwood et al, 2003). The model performance shown by EPA is consistent with other
long term mercury modeling applications and often more robust in terms of estimated metrics
that are more stringently paired in space and time before averaging to an annual or seasonal
temporal scale (Seigneur et al., 2006; Vijayaraghavan et al., 2008).
12
-------
TABLE III-3. Model performance metrics for mercury wet deposition and rainfall by quarter.
12 km domain.
12 km Domain
Total Mercury Wet
Deposition
Rainfall
TABLE III-4. Model
36 km domain.
Quarter
N
Observed (ng/m )
Predicted (ng/m )
Bias (ng/m )
Error (ng/m )
Fractional Bias (%)
Fractional Error (%)
N
Observed (mm)
Predicted (mm)
Bias (mm)
Error (mm)
Fractional Bias (%)
Fractional Error (%)
performance metrics
1
753
155
290
135
185
37
80
753
23
23
-0.04
10
3
14
for mercury wet
2
795
255
314
59
240
11
83
795
24
28
3.43
16
5
19
deposition
3
773
317
183
-134
246
-46
97
773
27
31
4.67
23
1
22
4
733
134
209
75
136
28
81
733
27
23
-3.52
11
-3
15
and rainfall by quarter.
36 km Domain
Total Mercury Wet
Deposition
Rainfall
Quarter
N
Observed (ng/m2)
Predicted (ng/m2)
Bias (ng/m2)
Error (ng/m )
Fractional Bias (%)
Fractional Error (%)
N
Observed (mm)
Predicted (mm)
Bias (mm)
Error (mm)
Fractional Bias (%)
Fractional Error (%)
1
783
152
262
110
165
29
82
783
23
21
-1.73
10
1
14
2
822
252
276
24
211
6
84
822
24
24
-0.09
15
2
19
3
797
313
187
-125
235
-40
91
797
27
29
1.87
20
0
21
4
754
133
183
50
114
19
79
754
27
21
-6.16
12
-5
15
13
-------
IV. Post Processing Mercury Deposition
CMAQ outputs hourly wet and dry deposition estimates (kg/ha) in each grid cell of speciated
mercury: HgO, Hg2, and PM2.5 Hg. Hourly outputs are summed to an annual estimate. CMAQ
model estimates of annual total mercury deposition from both 12 km model domains (12EUS1
and 12WUS1) were joined into a single 12 km model file covering the entire continental United
States. Where both the eastern and western 12 km domains intersect in the integrated 12 km
domain, the average of the 12 km eastern US and 12 km western US is assigned to the integrated
12 km domain. Total mercury deposition is defined as the sum of all wet and dry deposition of
elemental mercury, divalent gas-phase mercury, and PM2.5 mercury.
14
-------
V. References
Appel, K.W., Bhave, P.V., Gilliland, A.B., Sarwar, G., Roselle, S.J., 2008. Evaluation of the community multiscale
air quality (CMAQ) model version 4.5: Sensitivities impacting model performance; Part II - paniculate matter.
Atmospheric Environment 42, 6057-6066.
Appel, K.W., Gilliland, A.B., Sarwar, G., Gilliam, R.C., 2007. Evaluation of the Community Multiscale Air Quality
(CMAQ) model version 4.5: Sensitivities impacting model performance Part I - Ozone. Atmospheric Environment
41, 9603-9615.
Baker, K., Dolwick, P., 2009a. Meteorological Modeling Performance Evaluation for the Annual 2005 Continental
U.S. 36-km Domain Simulation. US Environmental Protection Agency OAQPS.
Baker, K., Dolwick, P., 2009b. Meteorological Modeling Performance Evaluation for the Annual 2005 Eastern U.S.
12-km Domain Simulation. US Environmental Protection Agency OAQPS, RTF.
Baker, K., Dolwick, P., 2009c. Meteorological Modeling Performance Evaluation for the Annual 2005 Western U.S.
12-km Domain Simulation, in: EPA, U. (Ed.). US Environmental Protection Agency OAQPS.
Boylan, J.W., Russell, A.G., 2006. PM and light extinction model performance metrics, goals, and criteria for three-
dimensional air quality models. Atmospheric Environment 40, 4946-4959.
Bullock, O.R., Atkinson, D., Braverman, T., Civerolo, K., Dastoor, A., Davignon, D., Ku, J.Y., Lohman, K., Myers,
T.C., Park, R.I, Seigneur, C., Selin, N.E., Sistla, G., Vijayaraghavan, K., 2008. The North American Mercury
Model Intercomparison Study (NAMMIS): Study description and model-to-model comparisons. Journal of
Geophysical Research-Atmospheres 113.
Bullock, O.R., Atkinson, D., Braverman, T., Civerolo, K., Dastoor, A., Davignon, D., Ku, J.Y., Lohman, K., Myers,
T.C., Park, R.I, Seigneur, C., Selin, N.E., Sistla, G., Vijayaraghavan, K., 2009. An analysis of simulated wet
deposition of mercury from the North American Mercury Model Intercomparison Study. Journal of Geophysical
Research-Atmospheres 114.
Bullock, O.R., Brehme, K.A., 2002. Atmospheric mercury simulation using the CMAQ model: formulation
description and analysis of wet deposition results. Atmospheric Environment 36, 2135-2146.
Byun, D., Schere, K.L., 2006. Review of the governing equations, computational algorithms, and other components
of the models-3 Community Multiscale Air Quality (CMAQ) modeling system. Applied Mechanics Reviews 59, 51-
77.
Carton, A.G., Bhave, P.V., Napelenok, S.L., Edney, E.D., Sarwar, G., Finder, R.W., Pouliot, G.A., Houyoux, M.,
2010. Model Representation of Secondary Organic Aerosol in CMAQv4.7. Environmental Science & Technology
44, 8553-8560.
Carton, A.G., Turpin, B.J., Altieri, K.E., Seitzinger, S.P., Mathur, R., Roselle, S.J., Weber, R.J., 2008. CMAQ
Model Performance Enhanced When In-Cloud Secondary Organic Aerosol is Included: Comparisons of Organic
Carbon Predictions with Measurements. Environmental Science & Technology 42, 8798-8802.
Doraiswamy, P., Hogrefe, C., Hao, W., Civerolo, K., Ku, J., Sistla, G., 2010. A Retrospective Comparison of
Model-Based Forecasted PM2.5 Concentrations with Measurements. Journal of Air & Waste Management
Association 60, 1293-1308.
Gery, M.W., Whitten, G.Z., Killus, J.P., Dodge, M.C., 1989. A Photochemical Kinetics Mechanism for Urban and
Regional Scale Computer Modeling. Journal of Geophysical Research-Atmospheres 94, 12925-12956.
15
-------
Houyoux, M.R., Vukovich, J.M., Coats, C.J., Wheeler, N.J.M., Kasibhatla, P.S., 2000. Emission inventory
development and processing for the Seasonal Model for Regional Air Quality (SMRAQ) project. Journal of
Geophysical Research-Atmospheres 105, 9079-9090.
Lin, C.J., Pongprueks, P., Rusell Bulock, O., Lindberg, S.E., Pehkonen, S.O., Jang, C., Braverman, T., Ho, T.C.,
2007. Scientific uncertainties in atmospheric mercury models II: Sensitivity analysis in the CONUS domain.
Atmospheric Environment 41, 6544-6560.
Nenes, A., Pandis, S.N., Pilinis, C., 1998. ISORROPIA: A new thermodynamic equilibrium model for multiphase
multicomponent inorganic aerosols. Aquatic Geochemistry 4, 123-152.
Pongprueksa, P., Lin, C.J., Lindberg, S.E., Jang, C., Braverman, T., Bullock, O.R., Ho, T.C., Chu, H.W., 2008.
Scientific uncertainties in atmospheric mercury models III: Boundary and initial conditions, model grid resolution,
and Hg(II) reduction mechanism. Atmospheric Environment 42, 1828-1845.
Seigneur, C., Lohman, K., Vijayaraghavan, K., Jansen, J., Levin, L., 2006. Modeling atmospheric mercury
deposition in the vicinity of power plants. Journal of the Air& Waste Management Association 56, 743-751.
Selin, N.E., Jacob, D.J., Park, R.J., Yantosca, R.M., Strode, S., Jaegle, L., Jaffe, D., 2007. Chemical cycling and
deposition of atmospheric mercury: Global constraints from observations. Journal of Geophysical Research-
Atmospheres 112.
Slemr, F., Brunke, E.G., Ebinghaus, R., Kuss, J., 2011. Worldwide trend of atmospheric mercury since 1995.
Atmospheric Chemistry and Physics 11, 4779-4787.
Streets, D.G., Zhang, Q., Wu, Y., 2009. Projections of Global Mercury Emissions in 2050. Environmental Science
& Technology 43, 2983-2988.
Tesche, T.W., Morris, R., Tonnesen, G., McNally, D., Boylan, J., Brewer, P., 2006. CMAQ/CAMx annual 2002
performance evaluation over the eastern US. Atmospheric Environment 40, 4906-4919.
USEPA, 2007. Guidance on the Use of Models and Other Analyses for Demonstrating Attainment of Air Quality
Goals for Ozone, PM2.5, and Regional Haze, RTF.
USEPA, 2010. Air Quality Modeling Technical Support Document: Boiler Source Sector Rules (EPA-454/R-10-
006), Research Triangle Park, North Carolina.
Vijayaraghavan, K., Karamchandani, P., Seigneur, C., Balmori, R., Chen, S.-Y., 2008. Plume-in-grid modeling of
atmospheric mercury. Journal of Geophysical Research-Atmospheres 113.
Vijayaraghavan, K., Seigneur, C., Karamchandani, P., Chen, S.Y., 2007. Development and application of a
multipollutant model for atmospheric mercury deposition. Journal of Applied Meteorology and Climatology 46,
1341-1353.
Yarwood, G, Lau, S., Jia, Y., Karamchandani, P., Vijayaraghavan, K. 2003. Final Report: Modeling Atmospheric
Mercury Chemistry and Deposition with CAMx for a 2002 Annual Simulation. Prepared for Wisconsin Department
of Natural Resources, http://www.gvpsvmoth.wi.gov/air/toxics/mercurv/hg X97579601 appB.pdf
16
-------
United States Office of Air Quality Planning and Standards Publication No. EP A-454/R-11-008
Environmental Protection Air Quality Assessment Division September 2011
Agency Research Triangle Park, NC
------- |