EPA910-R-12-007 Alaska
United States Region 10 Idaho
Environmental Protection 1200 Sixth Avenue Oregon
Agency Seattle WA 98101 Washington
Office of Environmental Assessment October 2012
Evaluation of the Combined
AERCpARE/AERMOD
Modeling Approach
for Offshore Sources
-------
-------
Evaluation of the Combined AERCOARE/AERMOD
Modeling Approach for Offshore Sources
EPA Contract No. EP-D-07-102
Work Assignment 5-17
Prepared for:
U.S. Environmental Protection Agency
Region 10
1200 Sixth Avenue
Mail Code OEA-095
Seattle, WA 98101
Prepared by:
Ken Richmond
Ralph Morris
ENVIRON International Corporation
Air Sciences Group
773 San Marin Drive, Suite 2115
Novato, California, 94998
www.environcorp.com
P-415-899-0700
F-415-899-0707
October 2012
-------
October 2012
DISCLAIMER
The opinions, findings, conclusions, or recommendations expressed in this report are those of
the authors and do not necessarily reflect the view of the U.S. Environmental Protection
Agency, nor does the mention of trade names or commercial products constitute endorsement
or recommendation for use by the Federal Government.
-------
October 2012
PREFACE
On 01 April 2011, the U.S. Environmental Protection Agency (EPA) Region 10 (RIO) approved
the use of the AERMOD dispersion program with output from an overwater meteorological
data preprocessor program to estimate ambient air pollutant concentration impacts at outer
continental shelf (OCS) locations in the Beaufort and Chukchi Seas of the Arctic Ocean.
AERMOD was approved because it contains the necessary options, features and capabilities to
estimate air pollutant concentration impacts from emission sources located in these two seas.
The options and features include the PRIME downwash algorithm, Plume Volume Molar Ratio
Method (PVMRM), and Ozone Limiting Method (OLM). Its capabilities consist of (1) estimating
impacts from point, area and volume sources, (2) accounting for calm conditions, and (3)
calculating design values based on deterministic and probabilistic standards. As an alternative
to the AERMET preprocessor program designed for terrestrial application, the Coupled Ocean
Atmosphere Response Experiment (COARE) air-sea flux algorithm was also approved to
preprocess overwater meteorological data measurements. The COARE algorithm output was
assembled with other meteorological variables in a spreadsheet to form the AERMOD
overwater meteorological input files. EPA's guideline Offshore and Coastal Dispersion (OCD)
model does not contain all these options, features, and capabilities, and the COARE algorithm
to adequately predict ambient concentrations from emission sources proposed in marine
environments.
Building upon its prior approval, RIO initiated two studies in late 2011. The first study modifies
AERMOD to include the platform building downwash algorithm contained in the OCD model.
The bases of the algorithm were wind tunnel experiments conducted by Ronald L. Peterson that
employed scaled models of the Chevron U.S.A West Cameron 28A platform located near
Cameron, LA. The second study that is the focus of this report, codes the COARE air-sea flux
procedure into a meteorological data preprocessor program called AERCOARE. AERCOARE will
read overwater measured hourly meteorological data or Weather Research and Forecasting
(WRF) model predicted hourly meteorological data output from the Mesoscale Model Interface
(MMIF) program. The output from AERCOARE can then be used by AERMOD in a marine
environment.
The work was funded under a subcontract from the University of North Carolina at Chapel Hill
with EPA Prime award EPD07102, Work Assignment 5-17. The EPA RIO Office of Environmental
Assessment (OEA) Work Assignment Manager was Ms. Jennifer Crawford and the RIO Technical
Lead was Mr. Herman Wong. Peer review of the draft document was provided by Dr. Sang-Mi
Lee of the South Coast Air Quality Management District. Her review and comments are greatly
appreciated.
-------
October 2012
CONTENTS
1.0 INTRODUCTION 1
2.0 BACKGROUND 2
3.0 EVALUATION METHODS AND DATA SETS 4
3.1 Overwater Tracer Data Sets 4
3.1.1 Pismo Beach 4
3.1.2 Cameron 9
3.1.3 Carpinteria 12
3.1.4 Ventura 16
3.2 AERCOARE Overwater Data Set Procedures 19
3.2.1 Data for AERCOARE 19
3.2.2 AERCOARE Meteorological Data Assembly Options 20
3.3 Statistical Evaluation Procedures 22
4.0 RESULTS 25
5.0 SUMMARY 50
6.0 REFERENCES 51
APPENDIX A: SENSITIVITY TO ASSUMED MINIMUM MIXING HEIGHTS
APPENDIX B: BOOT PROGRAM OUTPUT
APPENDIX C: DISTRIBUTION CD/DVD
-------
October 2012
TABLES
Table 1. Pismo Beach OCD Meteorological Data 7
Table 2. Pismo Beach Source and Receptor Data 8
Table 3. Cameron OCD Meteorological Data 10
Table 4. Cameron Source and Receptor Data 11
Table 5. Carpinteria OCD Meteorological Data 14
Table 6. Carpinteria Source Parameters 15
Table 7. Ventura OCD Meteorological Data 18
Table 8. Ventura Source and Receptor Data 19
Table 9. Performance Evaluation Statistical Results by Data Set and AERCOARE-MOD
Case 42
IV
-------
October 2012
FIGURES
Figure 1. Pismo Beach 5
Figure 2. Cameron 9
Figure 3. Carpinteria 12
Figure 4. Ventura 16
Figure 5. Scatter Plot of AERCOARE Case 1 versus Observations 26
Figure 6. Scatter Plot of AERCOARE Case 2 versus Observations 27
Figure 7. Scatter Plot of AERCOARE Case 3 versus Observations 28
Figure 8. Scatter Plot of AERCOARE Case 4 versus Observations 29
Figure 9. Scatter Plot of AERCOARE Case 5 versus Observations 30
Figure 10. QQ Plot of AERCOARE versus All Observations 31
Figure 11. QQ Plot of AERCOARE versus Carpinteria Observations 32
Figure 12. QQ Plot of AERCOARE versus Cameron Observations 33
Figure 13. QQ Plot of AERCOARE versus Ventura Observations 34
Figure 14. QQ Plot of AERCOARE versus Pismo Beach Observations 35
Figure 15. QQ Plot of AERCOARE Case 1 versus Observations 36
Figure 16. QQ Plot of AERCOARE Case 2 versus Observations 37
Figure 17. QQ Plot of AERCOARE Case 3 versus Observations 38
Figure 18. QQ Plot of AERCOARE Case 4 versus Observations 39
Figure 19. QQ Plot of AERCOARE Case 5 versus Observations 40
Figure 20. Sigma Plot for All Sites 45
Figure 21. Sigma Plot for Cameron 46
Figure 22. Sigma Plot for Carpinteria 47
Figure 23. Sigma Plot for Pismo Beach 48
Figure 24. Sigma Plot for Ventura 49
-------
October 2012
1.0 INTRODUCTION
ENVIRON conducted an evaluation of the combined AERCOARE/AERMOD (AERCOARE-MOD)
modeling approach for offshore sources using tracer data from four field studies. AERCOARE
processes overwater meteorological data for use by the AERMOD air quality dispersion model
(EPA, 2004a). AERCOARE applies the Coupled Ocean Atmosphere Response Experiment
(COARE) air-sea flux algorithm (Fairall, et. el., 2003) to estimate surface energy fluxes and
assembles these estimates and overwater measurements for subsequent dispersion model
simulations with AERMOD. AERCOARE would supplement AERMET (EPA, 2004b), the overland
meteorological preprocessor for AERMOD, and allow AERMOD to be applied to offshore
sources in a fashion similar to current new source review procedures over land.
The current study assesses the AERCOARE-MOD modeling approach using measurements from
four offshore field studies. The remainder of this report presents: the evaluation datasets,
techniques used to prepare data for AERCOARE, statistical model performance procedures and
the results of the evaluation. The development of AERCOARE was sponsored by the U.S.
Environmental Protection Agency (EPA) under Contract EP-D-7-102, Work Assignment 4-14 and
Work Assignment 5-17.
-------
October 2012
2.0 BACKGROUND
The AERCOARE-MOD approach would update the current regulatory approach for offshore
projects, the Offshore Coastal Dispersion (OCD) model (Chang and Hahn, 1997; DiCristofaro and
Hanna, 1989). OCD has not been updated for many years and does not reflect the latest
scientific advancements found in the AERMOD modeling system including:
OCD does not contain internal routines for processing either missing data or hours of calm
meteorology.
OCD does not contain the regulatory PRIME downwash algorithms (Schulman, L. L. et al,
2000)
The PVMRM1 and OLM2 methods for assessing the new 1-hour NO2 ambient standard are
not included in OCD.
EPA methods recommended for estimating design concentrations associated with the new
24-hour PM2.5,1-hour NO2, and 1-hour SO2 ambient standards must be obtained by post-
processing the OCD output files.
OCD does not contain a volume source routine and the area source routine only considers
circular areas without allowance for any initial vertical dispersion.
Although OCD contains routines to simulate the boundary layer over the ocean, the surface
energy flux algorithms are outdated and have been replaced within the scientific
community by the COARE air-sea flux algorithms.
The current regulatory AERMOD modeling system depends on the AERMET meteorological pre-
processor. AERMET was developed primarily to simulate meteorological processes driven by
the diurnal cycle of solar heating over land. The marine boundary layer behaves in a
fundamentally different manner because the ocean does not respond the same to diurnal
heating and cooling effects. Improvements needed to AERMET-AERMOD for offshore
applications include:
The surface roughness over the ocean varies with wind speed and wave conditions, and is
not a constant. The surface roughness for wind speed is also different than for temperature
and specific humidity.
AERMET uses the solar angle as an indication of the transition between daytime and
nighttime boundary layer regimes. Over the ocean, the stability of the boundary layer does
not respond as a strong function of solar heating, and especially in coastal waters, is driven
more by advection and horizontal differences in sea surface temperature. Unstable
conditions can occur during the night and stable conditions during the day.
AERMET does not explicitly include the effects of moisture in the assumed temperature and
wind speed profiles. The Monin-Obukhov length and convective velocity scale estimated by
1 Plume Volume Molar Ratio Method, used to limit NO-to-N02 conversion based on available ozone.
2 Ozone Limiting Method, used to limit NO-to-N02 conversion based on available ozone.
-------
October 2012
AERMET also do not incorporate moisture effects. The effect of surface moisture fluxes is
typically stronger over the ocean than over land.
The Bowen Ratio method for the latent heat flux in AERMET is overly simplistic. The ratio
between the latent and sensible heat is not a constant.
AERMOD does not contain routines for elevated platform downwash.
AERMOD cannot simulate shoreline fumigation or dispersion affected by non-homogenous
conditions either in space or time.
AERCOARE with the COARE air-sea flux method replaces AERMET by providing a meteorological
input file that is technically more appropriate for marine applications. When AERCOARE
provides the necessary meteorological data, AERMOD can be used to predict overwater
concentration impacts in a manner consistent with new source review procedures over land.
This allows the PVMRM, calms processing, volume source, and design concentration calculating
procedures in AERMOD to be applied to sources located within the marine boundary layer.3
A similar AERMOD-COARE approach was recently approved by EPA Region 10 (RIO) (EPA,
2011b) as an alternative model to OCD for application in an Arctic ice-free environment with
concurrence from the EPA Model Clearinghouse (EPA, 2011a). In that application, the COARE
algorithm was applied to overwater measurements and the results assembled in a spreadsheet.
AERCOARE replaces the need for post-processing with a spreadsheet, provides support for
missing data, adds options for the treatment of overwater mixing heights, and can consider
many different input data formats (Richmond and Morris, 2012).
3 Note the current version of AERMOD does not contain routines for platform downwash or shoreline fumigation.
-------
October 2012
3.0 EVALUATION METHODS AND DATA SETS
The AERCOARE-MOD modeling approach was assessed by comparing predictions to the
observations obtained from four offshore tracer studies: Pismo Beach, CA; Cameron, LA,
Carpinteria, CA; and Ventura, CA. These studies are a subset of the data used to evaluate OCD
(Chang and Hahn, 1997) and more recently, CALPUFF, the model preferred by the Minerals
Management Service (MMS) (now Bureau of Ocean Energy Management (BOEM)) for
permitting within their jurisdiction (Earth Tech, 2006). This section provides the rationale for
the selection of these data sets, describes the data sets, outlines the procedures for the
application of the AERCOARE algorithm, and presents the statistical methods used to compare
AERCOARE-MOD predictions to measurements from the field programs.
3.1 Overwater Tracer Data Sets
The four model evaluation data sets used in the current study were provided by EPA RIO from
the archives supporting development of the MMS (BOEM) version of CALPUFF and OCD Version
4 (DiCristofaro and Hanna, 1989). These studies occur under a wide range of overwater
atmospheric stabilities that might be expected in coastal waters regardless of the latitude. The
tracer measurements in Pismo Beach and Cameron occur in level terrain near the shoreline
downwind of offshore tracer releases. These two studies provide tests of overwater dispersion
without the complications due to air modification over the land or complex terrain. The
Ventura study is similar; however the receptors are located 500 meters (m) to one kilometer
(km) inland from the shoreline, so some air modification may have affected dispersion in this
study. The Carpinteria complex terrain tracer study involved shoreline measurements observed
on a bluff near plume level. The Carpinteria data set had much lighter winds and the transport
distances were less than the other three studies.
3.1.1 Pismo Beach
The Pismo Beach experiment was conducted during December 1981 and June 1982. A depiction
of land use, release point locations and receptor sites are shown in Figure 1 based on the files
from the CALPUFF evaluation archives. Tracer was released from a boat mast height of 13.1 m
to 13.6 m above the water. Peak concentrations occurred near the shoreline at sampling
distances from 6 km to 8 km away. The Pismo Beach evaluation database consists of 31
samples.
-------
October 2012
Pismo Beach, CA
3390-
3885-
o
00
3880-
Q
o
0)
g
N
1 3875-
C
O
3870-
3865-
a
710 715 720 725
UTM East (km) Zone 10N. Datum: NAS-C
^^H
-100
-95
-90
-85
-80
-75
-70
-65
-60
-55
-50
-45
-40
-35
-30
-25
-20
-15
m
Snow/Ice
Tundra
Barren
Wetland
Water
Forest
Range
Agriculture
Urban/Built-Up
Land Use
X Sampler Locations
A Tracer Releases
Figure 1. Pismo Beach
-------
October 2012
Table 1 lists the overwater meteorological data used in the current study. These same data
were also used in previous OCD and CALPUFF evaluations. A description of the data collection
and preparation can be found in the OCD and CALPUFF model evaluation reports with
references to the original field studies.
Examination of the meteorological data in Table 1 reveals several inconsistencies between the
air-sea temperature difference and the virtual potential temperature lapse rate. The virtual
potential temperature lapse rate sometimes indicates a stable boundary layer (positive) when
the air-sea temperature difference is unstable (negative).4 Either there was a low mixed layer
not reflected by the mixing height measurements in Table 1, or one of the measurements is not
representative of the boundary layer profile. We adjusted the air-sea temperature difference to
be at least as stable as indicated by the virtual potential temperature lapse rate to address this
inconsistency in our evaluation. In these instances, the sea temperature was adjusted so the
air-sea temperature difference matched the measured potential temperature lapse rate. The
revised estimates are shown in Table 1
Table 2 shows the source-to-receptor relationships and the release characteristics assumed for
the AERCOARE-MOD simulations. All simulations where performed with a unit emission rate
and without plume rise. Building downwash from the release boat was considered using the
dimensions shown in Table 2. As in the original OCD and CALPUFF evaluations, only peak
concentration predictions and observations for each hour are compared in the current
evaluation. In order to ensure that plume centerlines travelled over the receptor with the
highest observed concentration, a constant westerly wind was assumed and predictions were
obtained at a single receptor located the correct distance east of the release point.
4 OCD contains a dispersion algorithm for very stable conditions that can only be triggered when the measured virtual potential
temperature gradients exceeds 0.04 °C/m. Such conditions are triggered irrespective of all other meteorological data provided
to OCD. In this fashion, this variable can be used to override OCD's normal dispersion algorithms when other evidence suggests
extremely stable conditions have occurred.
-------
October 2012
Table 1. Pismo Beach OCD Meteorological Data.
Date/Time
12/8/81 15:00
12/8/81 16:00
12/11/81 14:00
12/11/81 15:00
12/11/81 17:00
12/11/81 19:00
12/13/81 14:00
12/13/81 15:00
12/13/81 17:00
12/14/81 13:00
12/14/81 15:00
12/14/81 17:00
12/15/81 13:00
12/15/81 14:00
12/15/81 19:00
6/21/82 15:00
6/21/82 16:00
6/21/82 17:00
6/21/82 18:00
6/22/82 15:00
6/22/82 16:00
6/22/82 19:00
6/24/82 13:00
6/24/82 15:00
6/25/82 12:00
6/25/82 13:00
6/25/82 15:00
6/25/82 16:00
6/25/82 17:00
6/27/82 16:00
6/27/82 18:00
Wind
Obs.
Ht. (m)
20.5
20.5
20.5
20.5
20.5
20.5
20.5
20.5
20.5
20.5
20.5
20.5
20.5
20.5
20.5
20.5
20.5
20.5
20.5
20.5
20.5
20.5
20.5
20.5
20.5
20.5
20.5
20.5
20.5
20.5
20.5
Temp
RHObs.
Ht. (m)
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
Wind
Dir.
261
284
275
283
289
305
289
280
301
292
292
296
304
299
321
276
269
261
276
274
268
289
269
269
286
280
286
288
290
287
285
Wind
Speed
(m/s)
2.2
1.6
4.5
5.4
8.6
7.9
5.4
6.1
7.9
7.7
10.9
9.9
5.6
6.1
1.6
4.3
3.8
2.7
3.0
3.7
5.2
3.2
3.9
5.3
5.6
6.5
9.8
9.1
9.5
12.7
10.2
Mix Ht.
(m)
100
100
600
600
700
900
50
50
50
50
50
50
50
50
50
800
800
800
800
700
700
700
600
600
100
100
100
100
100
100
100
Rel.
Humid.
(%)
67
75
74
73
84
81
95
97
92
79
90
88
88
83
70
84
86
87
89
80
78
84
82
84
76
80
82
82
81
93
94
Air Temp.
(K)
287.7
287.5
285.6
286.1
286.0
286.1
285.5
285.3
286.2
287.2
286.4
286.7
286.1
287.7
289.4
287.5
287.3
287.3
286.9
288.6
288.8
287.2
288.1
288.1
288.9
288.5
288.3
288.3
288.4
287.0
287.7
Air-Sea
Temp (K)
1.3
1.2
-0.4
0.0
0.1
0.2
-0.8
-0.8
0.3
1.3
0.4
0.9
0.3
1.1
3.4
1.5
1.4
1.5
1.2
1.7
2.1
1.3
0.9
0.6
2.2
2.6
2.6
2.9
3.2
3.4
3.7
Virt. Pot.
Temp Grad.
(K/m)
0.030
0.030
0.010
0.010
0.010
0.010
0.000
0.000
0.060
0.020
0.020
0.020
0.010
0.010
0.030
0.008
0.008
0.008
0.008
0.005
0.005
0.005
0.010
0.010
0.010
0.010
0.010
0.010
0.010
0.010
0.010
Sigma-
Theta
9.43
12.90
5.60
4.57
2.12
45.00
0.92
2.41
1.89
1.20
1.20
1.78
14.41
45.00
45.00
1.37
2.12
6.84
19.70
6.05
3.32
10.59
27.79
7.46
1.37
1.60
5.48
0.92
1.20
1.09
7.74
Revised Air-
Sea Temp
(K)
1.30
1.20
0.00
0.00
0.10
0.20
-0.80
-0.80
0.35
1.30
0.40
0.90
0.30
1.10
3.40
1.50
1.40
1.50
1.20
1.70
2.10
1.30
0.90
0.60
2.20
2.60
2.60
2.90
3.20
3.40
3.70
-------
October 2012
Table 2. Pismo Beach Source and Receptor Data.
Date/Time
12/8/81 15:00
12/8/81 16:00
12/11/81 14:00
12/11/81 15:00
12/11/81 17:00
12/11/81 19:00
12/13/81 14:00
12/13/81 15:00
12/13/81 17:00
12/14/81 13:00
12/14/81 15:00
12/14/81 17:00
12/15/81 13:00
12/15/81 14:00
12/15/81 19:00
6/21/82 15:00
6/21/82 16:00
6/21/82 17:00
6/21/82 18:00
6/22/82 15:00
6/22/82 16:00
6/22/82 19:00
6/24/82 13:00
6/24/82 15:00
6/25/82 12:00
6/25/82 13:00
6/25/82 15:00
6/25/82 16:00
6/25/82 17:00
6/27/82 16:00
6/27/82 18:00
Rel. Ht.(m)
13.1
13.1
13.1
13.1
13.1
13.1
13.1
13.1
13.1
13.1
13.1
13.1
13.1
13.1
13.1
13.6
13.6
13.6
13.6
13.6
13.6
13.6
13.6
13.6
13.6
13.6
13.6
13.6
13.6
13.6
13.6
Bldg.
Ht. (m)
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
Bldg.
Wid. (m)
20.0
20.0
20.0
20.0
20.0
20.0
20.0
20.0
20.0
20.0
20.0
20.0
20.0
20.0
20.0
20.0
20.0
20.0
20.0
20.0
20.0
20.0
20.0
20.0
20.0
20.0
20.0
20.0
20.0
20.0
20.0
Recep.
Dist.(m) *
6730
6506
6422
6509
6619
7316
6516
6372
6870
6378
6378
6526
6944
6697
8312
6532
6589
6748
6532
6125
6214
6054
6244
6244
6406
6377
6406
6435
6455
6630
6579
1 All releases were simulated with a 270 degree wind direction from a source at (0, 0) and a receptor at (X,0) where X is the
downwind distance with the peak observed concentration. All receptors are in flat terrain with a 1.5m flag pole height
-------
October 2012
3.1.2 Cameron
Figure 2 shows the land use, release points, receptors, and meteorological stations for the
Cameron evaluation data set. Twenty-six tracer samples from the field studies in July 1981 and
February 1982 were used in the evaluation. Tracer was released from both a boat and a low
profile platform, from a height of 13 m. As in the Pismo Beach study, the receptors are located
in flat terrain near the shoreline with transport distances ranging from 4 km to 10 km.
CAMERON, LA
o
I
1
B
N
t:
o
z
3286
C
1 25m
\r
vljlst
ast
xxxxx&x
28a
*****
**>**<
2 If
42/24
**^
*^
466 468 470 472 474 476 478 480 482 484 486 488 490
UTM East (km) Zone 15N, Datum: NAS-C
^^H
Lan
-100
"95 Snow/Ice
-90
5 Tundra
-80
"75 Barren
-70
-65 Wetland
-60
-55 Water
-50
-45 Forest
-40
-35 Range
-30
-25 Agriculture
-20
-(5 Urban/Built-Up
-10
JUse
X Sampler Locations
A Tracer Releases
Figure 2. Cameron
The Cameron meteorological data used in the current analysis are shown in Table 3, and are
based on the OCD and CALPUFF model evaluation data set. The data set contains both very
stable and fairly unstable conditions. As with the Pismo Beach data, there are several hours of
stable lapse rates accompanied by unstable air-sea temperature differences. For example on
February 15,1982 hour 1700, the air-sea temperature difference is -0.8 °C, while the virtual
potential temperature lapse rate is 0.06 °C/m (extreme stability "G" in OCD). Over 10 m, this
virtual potential temperature lapse rate would result in at least an air-sea temperature
difference of +0.5 °C. These contradictory data were resolved using the same methodology as
in the Pismo Beach dataset.
-------
October 2012
Table 3. Cameron OCD Meteorological Data.
Date/Time
7/20/81 14:00
7/20/81 15:00
7/23/81 17:00
7/23/81 18:00
7/27/81 20:00
7/27/81 22:00
7/29/81 16:00
7/29/81 17:00
7/29/81 19:00
2/15/82 16:00
2/15/82 17:00
2/15/82 20:00
2/17/82 14:00
2/17/82 15:00
2/17/82 16:00
2/17/82 17:00
2/17/82 18:00
2/22/82 14:00
2/22/82 16:00
2/22/82 17:00
2/23/82 14:00
2/23/82 17:00
2/24/82 15:00
2/24/82 16:00
2/24/82 17:00
2/24/82 19:00
Wind
Obs. Ht.
(m)
10
10
10
10
10
10
10
10
10
10
10
10
10
18
18
18
18
18
18
18
18
18
18
18
18
18
Temp
RH Obs.
Ht. (m)
10
10
18
18
18
18
18
18
18
10
10
10
10
18
18
18
18
18
18
18
18
18
18
18
18
18
Wind
Dir.
202
210
232
229
176
151
218
240
241
142
134
147
178
195
210
206
193
171
172
182
152
165
143
143
140
156
Wind
Speed
(m/s)
4.6
4.8
4.3
5.1
2.1
4.5
4.6
5.0
5.0
5.7
5.6
5.9
3.3
3.7
4.3
3.5
3.5
5.2
4.7
4.5
4.8
6.2
3.7
3.7
3.5
4.1
Mix Ht.
(m)
800
800
225
225
400
450
420
430
450
200
200
200
200
200
200
200
200
100
100
100
50
80
50
50
50
50
Rel.
Humid.
(%)
63
64
73
74
82
82
69
68
68
89
88
87
93
93
93
93
93
75
76
76
84
88
49
50
50
52
Air
Temp.
(K)
302.4
302.6
303.6
303.7
300.2
300.0
303.0
303.0
303.1
287.4
287.1
287.4
288.8
288.1
288.0
287.7
287.4
290.6
290.6
290.9
291.5
291.2
293.1
292.9
292.9
290.7
Air-Sea
Temp (K)
-2.7
-2.6
-1.4
-1.2
-4.4
-4.5
-2.2
-2.0
-1.7
0.0
-0.8
-0.4
2.1
0.9
0.6
-0.2
-0.7
1.3
0.9
0.8
3.7
2.3
5.0
4.6
4.7
2.7
Virt. Pot.
Temp
Grad.
(K/m)
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.06
0.06
0.06
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.05
0.05
0.05
0.05
Sigma-
Theta
6.39
4.92
4.74
4.74
999.00
999.00
9.59
6.45
9.59
999.00
999.00
999.00
2.46
7.63
3.89
3.78
2.06
2.69
2.41
2.81
0.63
3.21
2.75
3.21
3.26
2.63
Revised
Air-Sea
Temp (K)
-2.7
-2.6
-1.4
-1.2
-4.4
-4.5
-2.2
-2.0
-1.7
0.5
0.5
0.5
2.1
0.9
0.4
0.4
0.4
1.3
0.9
0.8
3.7
2.3
5.0
4.6
4.7
2.7
10
-------
October 2012
Table 4 shows the source and receptor characteristics used in the Cameron tracer simulations.
The platform releases were simulated without downwash and the boat releases assumed a
building height of 7 m and a width (and length) of 20 m. A constant hypothetical wind direction
was assumed and downwind receptor distances were varied to match the downwind distances
of the measurement site with the highest observed concentration for each period.
Table 4. Cameron Source and Receptor Data.
Date/Time
7/20/81 14:00
7/20/81 15:00
7/23/81 17:00
7/23/81 18:00
7/27/81 20:00
7/27/81 22:00
7/29/81 16:00
7/29/81 17:00
7/29/81 19:00
2/15/82 16:00
2/15/82 17:00
2/15/82 20:00
2/17/82 14:00
2/17/82 15:00
2/17/82 16:00
2/17/82 17:00
2/17/82 18:00
2/22/82 14:00
2/22/82 16:00
2/22/82 17:00
2/23/82 14:00
2/23/82 17:00
2/24/82 15:00
2/24/82 16:00
2/24/82 17:00
2/24/82 19:00
Rel. Ht.(m)
13.0
13.0
13.0
13.0
13.0
13.0
13.0
13.0
13.0
13.0
13.0
13.0
13.0
13.0
13.0
13.0
13.0
13.0
13.0
13.0
13.0
13.0
13.0
13.0
13.0
13.0
Bldg.
Ht. (m)
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
7.0
7.0
7.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
7.0
7.0
7.0
7.0
Bldg.
Wid.(m)
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
20.0
20.0
20.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
20.0
20.0
20.0
20.0
Recep.
Dist.(m) *
7180
7400
8930
8710
7020
7859
7820
9780
9950
4834
5762
4526
7000
6985
7400
7260
6950
7095
7070
6955
7769
7245
5669
5669
6023
4786
l.AII releases were simulated with a 270 degree wind direction from a source at (0, 0) and a receptor at (X,0) where X is the
downwind distance with the peak observed concentration. All receptors are in flat terrain with a 1.5m flag pole height.
11
-------
October 2012
3.1.3 Carpinteria
The Carpinteria tracer study was conducted in September and October 1985. Studies were
conducted to examine offshore impacts caused by both interaction with complex terrain and
shoreline fumigation. The current analysis only evaluated the complex terrain data set as the
AERCOARE-MOD approach currently cannot simulate shoreline fumigation.
Figure 3 shows the land use and terrain for the Carpinteria field study. The shoreline receptors
are located on a 20 m to 30 m high bluff within 0.8 km to 1.5 km of the offshore tethersonde
release. Two tracers were released with heights varying from 18 m to 61 m. The tethersonde
was well above the anchor boat and downwash was not considered in the simulations.
CARPINTERIA, CA
3814
3813
o
3812
3811
3810
* 3809
1
3808
3807
3806
268 269 270 271 272 273 274
UTM East (km) Zone 11N, Datum: NAS-C
275
276
100
~95 Snow/Ice
-90
~85 Tundra
-80
~75 Barren
-70
-65 Wetland
-60
-55 vVater
-50
-45 Forest
-40
-35 Range
-30
-25 Agnculture
-20
-15 Urban/Built-Up
10
Land Use
Sampler Locations:
X - Complex Terrain
X Fumigation
Tracer Releases:
A - Complex Terrain
A Fumigation
Figure 3. Carpinteria
12
-------
October 2012
Table 5 displays the meteorological data used in the current simulations and previous
evaluations of OCD and CALPUFF. The winds were very light for most of the releases, especially
considering the wind measurement heights were from 30 m to 49 m. The combined influences
of low wind speeds and the air-sea temperature differences in Table 5 result in cases with
unstable to very stable stratifications. Unlike the Pismo Beach and Cameron data sets, the
virtual potential temperature lapse rates do not contradict the gradient inferred from the air-
temperature difference measurements. One suspect aspect of the data is the constant mixed
layer height of 500 m for the entire data set. In cases where plumes are not trapped under a
strong inversion, CALPUFF and OCD are less sensitive to the mixing height than AERMOD. Thus
uncertainty in the boundary layer height in this experiment may not have been important to
the original investigators.
Table 6 lists the source release parameters used for the AERCOARE-MOD simulations of the
Carpinteria data set. Unlike the Pismo Beach and Cameron simulations, actual wind directions,
source locations and receptor sites were used in the analysis to consider the effects of terrain
elevation on the model predictions. Receptor elevations and scale heights for AERMOD were
calculated with AERMAP (Version 11103) (EPA, 2004c) using 1/3 arc-second terrain data from
the National Elevation Data (NED) set. The peak predicted concentration was compared to the
peak measured concentration for each release.
13
-------
October 2012
Table 5. Carpinteria OCD Meteorological Data.
Date/Time
9/19/85 9:00
9/19/85 10:00
9/19/85 11:00
9/19/85 12:00
9/22/85 9:00
9/22/85 10:00
9/22/85 11:00
9/22/85 11:00
9/22/85 12:00
9/22/85 12:00
9/25/85 10:00
9/25/85 11:00
9/25/85 12:00
9/25/85 13:00
9/26/85 12:00
9/26/85 13:00
9/28/85 10:00
9/28/85 10:00
9/28/85 11:00
9/28/85 11:00
9/28/85 13:00
9/28/85 13:00
9/28/85 14:00
9/28/85 14:00
9/29/85 11:00
9/29/85 12:00
9/29/85 12:00
Wind
Obs. Ht.
(m)
30
30
30
30
30
30
30
30
30
30
24
46
46
46
49
49
24
24
24
24
24
24
24
24
30
30
30
Temp
RH Obs.
Ht. (m)
9
9
9
9
9
9
9
9
9
9
9
9
9
9
9
9
9
9
9
9
9
9
9
9
9
9
9
Wind
Dir.
259.7
235.4
214.1
252.9
220.8
251.1
253.8
230.0
248.4
237.7
163.8
163.8
165.6
175.0
262.0
262.2
155.8
155.8
174.7
177.0
234.5
229.5
215.0
215.0
243.7
238.9
232.7
Wind
Speed
(m/s)
1.3
1.3
2.6
3.1
1.0
1.2
2.4
2.4
2.8
2.8
1.0
1.6
1.0
1.0
3.8
4.0
5.4
5.4
3.2
3.2
1.5
1.5
2.1
2.1
3.4
3.1
3.1
Mix Ht.
(m)
500
500
500
500
500
500
500
500
500
500
500
500
500
500
500
500
500
500
500
500
500
500
500
500
500
500
500
Rel.
Humid.
(%)
78.8
79.0
80.1
80.1
70.6
81.0
92.1
92.1
91.1
91.1
60.3
69.9
90.3
90.4
83.5
81.0
85.1
85.1
84.1
84.1
82.5
82.5
81.7
81.7
86.0
87.8
87.8
Air
Temp.
(K)
289.45
289.95
290.15
290.25
290.55
290.15
289.55
289.55
289.45
289.45
294.35
294.15
294.05
294.55
291.85
291.95
291.25
291.25
291.15
291.15
291.45
291.45
291.65
291.65
291.35
291.25
291.25
Air-Sea
Temp (K)
-1.1
-0.8
-0.7
-0.7
0.5
0.3
1.0
1.0
1.1
1.1
2.8
2.3
2.1
2.7
-0.7
-1.0
-0.6
-0.6
-0.8
-0.8
-0.6
-0.6
-0.3
-0.3
-0.3
-0.4
-0.4
Virt. Pot.
Temp
Grad.
(K/m)
0.00
0.00
0.00
0.00
0.02
0.02
0.02
0.02
0.02
0.02
0.01
0.01
0.01
0.01
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
Sigma-
Theta
26.84
28.41
24.42
32.86
32.13
17.43
7.97
7.97
17.43
17.43
41.67
9.87
26.06
18.37
10.87
11.80
8.92
8.92
10.87
10.87
10.87
10.87
11.80
11.80
18.37
4.97
4.97
Revised
Air-Sea
Temp (K)
-1.10
-0.80
-0.70
-0.70
0.50
0.30
1.00
1.00
1.10
1.10
2.80
2.30
2.10
2.70
-0.70
-1.00
-0.60
-0.60
-0.80
-0.80
-0.60
-0.60
-0.30
-0.30
-0.30
-0.40
-0.40
14
-------
October 2012
Table 6. Carpinteria Source Parameters.
Date/Time
9/19/85 9:00
9/19/85 10:00
9/19/85 11:00
9/19/85 12:00
9/22/85 9:00
9/22/85 10:00
9/22/85 11:00
9/22/85 11:00
9/22/85 12:00
9/22/85 12:00
9/25/85 10:00
9/25/85 11:00
9/25/85 12:00
9/25/85 13:00
9/26/85 12:00
9/26/85 13:00
9/28/85 10:00
9/28/85 10:00
9/28/85 11:00
9/28/85 11:00
9/28/85 13:00
9/28/85 13:00
9/28/85 14:00
9/28/85 14:00
9/29/85 11:00
9/29/85 12:00
9/29/85 12:00
Release
Type1
SF6
SF6
SF6
SF6
SF6
SF6
SF6
Freon
SF6
Freon
SF6
SF6
SF6
SF6
Freon
Freon
SF6
Freon
SF6
Freon
SF6
Freon
SF6
Freon
SF6
SF6
Freon
Rel. Ht.
(m)
30.5
30.5
30.5
30.5
18.3
18.3
18.3
36.6
18.3
36.6
24.4
24.4
24.4
24.4
24.4
24.4
24.4
42.7
24.4
42.7
24.4
39.6
24.4
39.6
30.5
30.5
61.0
UTM East
(m)
270,343
270,343
270,343
270,343
270,133
270,133
270,133
270,133
270,133
270,133
271,024
271,024
271,024
271,024
269,524
269,524
271,289
271,289
271,289
271,289
270,133
270,133
270,133
270,133
270,133
270,133
270,133
UTM North
(m)
3,806,910
3,806,910
3,806,910
3,806,910
3,806,520
3,806,520
3,806,520
3,806,520
3,806,520
3,806,520
3,806,660
3,806,660
3,806,660
3,806,660
3,807,330
3,807,330
3,806,340
3,806,340
3,806,340
3,806,340
3,806,520
3,806,520
3,806,520
3,806,520
3,806,520
3,806,520
3,806,520
1. For some hours releases were from two different heights using different tracer gases. Actual source and receptor locations
were used in the simulations where receptor heights and scale heights were calculated with AERMAP. There was no building
downwash assumed for these simulations.
15
-------
October 2012
3.1.4 Ventura
The Ventura experiment was conducted during September 1980 and January 1981. Land use,
release point locations and receptor sites are shown in Figure 4 based on the files from the
CALPUFF evaluation archives. The tracer was released from a boat mast height of 8.1 m above
the water. Peak concentrations occurred along the closet arc of receptors in Figure 4 at
sampling distances from 7 km to 11 km away. The Ventura evaluation database consists of 17
samples.
VENTURA, CA
3798-
£ 3792-
3786-
3784-
3782-
3780-
X
x
X
X
X
X
X*
X
284 286 288 290 292 294 296 298 300 302 304
UTM East (km) Zone 11N, Datum: NAS-C
Figure 4. Ventura
100
~95 Snow/Ice
-90
-1 Tundra
-80
"75 Barren
-70
-65 wetland
-60
-55 Water
-50
-45 Forest
-40
-35 Range
-30
-25 Agriculture
-20
-15 Urban/Built-Up
10
Land Use
X Sampler Locations
A Tracer Releases
16
-------
October 2012
The Ventura meteorological data used in the current analysis are shown in Table 7. The OCD
and CALPUFF model evaluation data set stabilities ranged from moderately unstable to slightly
stable. As with the Pismo Beach data, there are several hours of stable lapse rates accompanied
by unstable air-sea temperature differences. For example, on September 29,1980 hour 1400,
the air-sea temperature difference is -0.8 °C, while the virtual potential temperature lapse rate
is 0.03 °C/m. These contradictory data were resolved using the same methodology as in the
Pismo Beach and Cameron datasets.
Table 8 shows the source and receptor characteristics used in the Ventura tracer simulations.
The boat releases assumed a building height of 7 m and a width (and length) of 20 m. A
constant hypothetical wind direction was assumed and downwind receptor distances were
varied to match the downwind distances of the measurement site with the highest observed
concentration for each period.
17
-------
October 2012
Table 7. Ventura OCD Meteorological Data
Date/Time
9/24/80 16:00
9/24/80 18:00
9/24/80 19:00
9/27/80 14:00
9/27/80 19:00
9/28/80 18:00
9/29/80 14:00
9/29/80 16:00
9/29/80 18:00
1/6/81 16:00
1/6/81 17:00
1/6/81 18:00
1/9/81 15:00
1/9/81 16:00
1/9/81 18:00
1/13/81 15:00
1/13/81 17:00
Wind
Obs. Ht.
(m)
20.5
20.5
20.5
20.5
20.5
20.5
20.5
20.5
20.5
20.5
20.5
20.5
20.5
20.5
20.5
20.5
20.5
Temp
RH Obs.
Ht. (m)
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
Wind
Dir.
266
281
292
272
272
265
256
264
264
276
283
276
286
277
274
274
242
Wind
Speed
(m/s)
4.1
6.2
6.9
6.3
6.1
3.1
3.3
5.1
5.2
4.0
5.1
4.9
4.7
4.6
4.9
5.8
4.2
Mix Ht.
(m)
400
400
400
400
400
250
100
100
50
50
50
50
100
100
100
50
50
Rel.
Humid.
(%)
72
78
77
80
80
80
76
76
76
60
58
60
87
85
87
65
84
Air
Temp.
(K)
288.3
288.0
288.0
288.0
289.0
290.0
288.7
289.3
289.2
290.3
290.6
290.4
287.6
288.0
288.2
290.1
289.0
Air-Sea
Temp (K)
-2.1
-2.0
-2.1
-1.9
-1.0
-1.0
-0.8
0.0
-0.1
1.6
1.7
1.8
-0.9
-0.5
-0.3
1.4
0.4
Virt. Pot.
Temp
Grad.
(K/m)
0.00
0.00
0.00
0.00
0.00
0.01
0.03
0.03
0.03
0.01
0.01
0.01
0.00
0.00
0.00
0.01
0.01
Sigma-
Theta
8.0
6.5
6.0
4.7
3.6
4.4
5.0
3.9
5.2
21.5
13.1
9.4
3.4
4.8
3.1
11.6
8.5
Revised
Air-Sea
Temp (K)
-2.1
-2.0
-2.1
-1.9
-1.0
0.0
0.1
0.1
0.1
1.6
1.7
1.8
-0.9
-0.5
-0.3
1.4
0.4
18
-------
October 2012
Table 8. Ventura Source and Receptor Data.
Date/Time
9/24/80 16:00
9/24/80 18:00
9/24/80 19:00
9/27/80 14:00
9/27/80 19:00
9/28/80 18:00
9/29/80 14:00
9/29/80 16:00
9/29/80 18:00
1/6/81 16:00
1/6/81 17:00
1/6/81 18:00
1/9/81 15:00
1/9/81 16:00
1/9/81 18:00
1/13/81 15:00
1/13/81 17:00
Rel. Ht.(m)
8.1
8.1
8.1
8.1
8.1
8.1
8.1
8.1
8.1
8.1
8.1
8.1
8.1
8.1
8.1
8.1
8.1
Bldg.
Ht. (m)
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
Bldg.
Wid. (m)
20.0
20.0
20.0
20.0
20.0
20.0
20.0
20.0
20.0
20.0
20.0
20.0
20.0
20.0
20.0
20.0
20.0
Recep.
Dist.(m) 1
9291
9211
10799
9123
9123
9145
8085
7854
7854
7463
7416
7463
7956
7749
7704
7705
6914
1. All releases were simulated with a 270 degree wind direction from a source at (0, 0) and a receptor at (X,0)
where X is the downwind distance with the peak observed concentration. All receptors are in flat terrain with a
1.5m flag pole height.
3.2 AERCOARE Overwater Data Set Procedures
AERCOARE Version 1.0 (12275) was applied to prepare the overwater meteorological data for
the four offshore datasets. Several different options within AERCOARE were evaluated in the
study including the estimation of mixing heights, the use of horizontal wind direction (sigma-
theta data), and limitations on several important variables provided to AERMOD. Further details
are provided in the following discussion.
3.2.1 Data for AERCOARE
AERCOARE uses the COARE algorithm to predict the surface energy fluxes from the overwater
data sets briefly described above. The data necessary for the COARE algorithm depend on the
options employed for estimating the surface roughness, for the treatment of a cool-skin, or
heating of the upper layer of the ocean. The options selected for the evaluation and associated
data are as follows:
Several options are available to adjust the sea temperature to account for the difference
between the skin temperature and the bulk temperature measurement taken at depth from
a buoy or ship. The cool-skin and warm-layer options depend on solar radiation and
downward longwave irradiance input data. Such data were not readily available for the
19
-------
October 2012
current analysis and these options were not selected for the current evaluation. The warm-
layer effects option also needs continuous data over the diurnal cycle that are not available
for the tracer studies. CALPUFF also uses the COARE algorithm and previous studies
concluded model performance was not sensitive to the cool-skin or warm-layer options for
the Pismo Beach, Cameron, Ventura, or Carpinteria data sets (Earth Tech, 2006).
(AERCOARE variable Jwarm = Jcool = 0).
COARE also contains several methods for estimating the surface roughness length, and the
routines can use wave height and period measurement data. The current simulations were
conducted with the default option for a well-developed or deep sea. As with the warm-layer
and cool-skin options, sensitivity tests from previous studies suggest the COARE algorithm is
not very sensitive to surface roughness options, especially in the absence of wave
measurement data. (AERCOARE variable Jwave = 0).
The air-sea temperature difference, overwater relative humidity and the wind velocity drive
the energy fluxes and surface stability routines within the COARE routines. Air-sea
temperature differences were based on the OCD data sets except for the cases discussed
previously where the stable temperature lapse rate data contradicts such observations. In
these instances the air-sea temperature difference was based on the lapse rate applied
from the surface to the temperature measurement height.
Wind speed, air temperature, and relative humidity were taken directly from the OCD data
sets listed in Table 1, Table 3, Table 5, and Table 7. The measurement heights are also listed
in these tables.
Wind direction was assumed to be from the west for the Pismo Beach, Ventura and
Cameron data sets, as simulated receptors were located east of the release points with the
downwind distances appropriate for the peak measurement sites. For Carpinteria, the wind
directions shown in Table 5 were used in the simulations.
Surface pressure was assumed to be 1000 mb. This is the same pressure assumed for
previous evaluation studies with these data sets and the COARE algorithm is not sensitive to
the assumed atmospheric pressure.
The COARE algorithm has a small term that depends on rainfall. No precipitation was
assumed for any of the hours of the evaluation.
The COARE algorithm has a small term for "gustiness" that adds to the momentum fluxes
during light winds caused by large scale eddies. The model evaluation used the COARE
algorithm default for this parameter. (AERCOARE variable defzi = 600 m).
AERCOARE combines surface energy flux estimates from the COARE algorithm with additional
overwater measurements. Such techniques were evaluated using several options as discussed
in the next section.
3.2.2 AERCOARE Meteorological Data Assembly Options
Several different AERCOARE options were considered for preparation of the AERMOD data and
were included as cases in the model evaluation. The options selected for the evaluation and
associated data are as follows:
20
-------
October 2012
The standard deviations of horizontal wind direction (sigma-theta or oe) for the simulations
are based on the measurements shown in Table 1, Table 3, Table 5 and Table 7. One case in
the AERCOARE-MOD simulations excluded such measurements to test the sensitivity of the
predictions to the availability of these data compared to the internal AERMOD algorithm for
prediction of sigma-theta.
Standard deviations of the vertical wind velocity (sigma-w or ow) were not provided to
AERMOD. Such data were not available for Pismo Beach or Ventura and previous studies
have cautioned against the use of such data from the Carpinteria and Cameron data sets.
Sigma-w data were also not used in the previous OCD and CALPUFF evaluation studies.
AERMOD restricts the Monin-Obukhov length (/.) such that ABS (/.) > 1. This restriction
avoids unrealistic extremely stable and unstable conditions during light wind conditions. In
the evaluation simulations, we tested a Monin-Obukhov length of ABS (/.) > 5, as is assumed
by OCD and CALPUFF over water. (AERCOARE variable dlmin = 5 m).
The virtual potential temperature gradient above the convective boundary layer was
assumed to 0.01 °C/m. This variable is used by AERMOD to estimate plume penetration for
plume rise calculations and for the portion of the plume predicted to be above the
convective mixed layer. Plume rise and plume penetration are not applicable to the passive
tracer releases in the current evaluation. (AERCOARE variable dvptg = 0.01 °C/m).
Convective boundary layer heights were assumed to be the same as the observed mixing
heights from field studies when conditions where unstable as indicated by the Monin-
Obukhov length (/. < 0). Two options for mechanical mixing heights (z/m) were considered in
the evaluation:
mechanical mixing heights were calculated from the surface friction velocity using the
Venketram equation in AERMET (Venketram, 1980). The AERCOARE option for
smoothing as in AERMET was not applied because the data in the field studies are not
sequential. In addition, the smoothing does not significantly affect hour-to-hour
variations when the heights are relatively small as they are in these studies. (AERCOARE
variable mixopt = 1)
mechanical mixing heights were also assumed to be the same as the observed mixing
heights in Table 1, Table 3, Table 5 and Table 7. (AERCOARE variable mixopt = 0)
For low winds and smooth surfaces, the Venketram equation results in very small
mechanical mixing heights. The mechanical mixing height is an important variable in
AERMOD and is used as a scaling parameter during the construction of several important
meteorological profiles and the vertical dispersion term (az). The mechanical mixing height
is also in the denominator of the AERMOD equation used to calculate the lateral diffusion
term (ay) during stable conditions. AERMOD requires mixing heights be above 1 m. In this
study we used a minimum mixing height of 25 m (AERCOARE variable zimin = 25 m).
Appendix A provides further discussion on the sensitivity of the results to the assumed
minimum mechanical mixing height.
21
-------
October 2012
Using the techniques and data discussed above, AERCOARE-MOD meteorological data sets
were prepared for each of the four field studies. Five cases were considered using various
combinations of the many possible methods to assemble the data:
Case 1: Require Abs (/.) > 5, use oe measurements, and use the Venketram equation in
AERMETforz/m and require z\m > 25 m.
Case 2: Require Abs (/.) > 5, use AERMOD predicted oe, and use the Venketram equation in
AERMETforz/m and require z\m > 25 m.
Case 3: Require Abs (/.) > 1, use OQ measurements, and observed mixing heights for the
mechanical mixing height (z/m).
Case 4: Require Abs (/.) > 5, use oe measurements, and observed mixing heights for the
mechanical mixing height (zim).
Case 5: Require Abs (/.) > 5, use oe measurements, use the Venketram equation in AERMET
for z/m and require z\m > 25 m, and modify AERMOD to use the Draxler equation for the
ambient lateral dispersion parameter:
1 + 0.9
where x is the downwind distance, u the effective wind speed, and ov is the effective
standard deviation of the lateral wind speed calculated from OQ. This equation is used both
by OCD and CALPUFF. Case 5 was included to remove the sensitivity of the lateral dispersion
term in AERMOD to the mixing height. The CALPUFF evaluations found this equation
performed better than several alternatives that are more similar to the formulation used by
AERMOD (Earth Tech, 2006).
AERCOARE-MOD predictions from the five cases above were obtained for the Pismo Beach,
Cameron, Ventura, and Carpinteria data sets. The same five model option cases were evaluated
in previous studies submitted to RIO and the EPA Modeling Clearinghouse (EPA 2011a; EPA
2011b). The current analysis adds the Ventura field study to the three data sets previously
evaluated. Peak predictions were compared to peak observations using the statistical model
evaluation methods discussed in the following section.
3.3 Statistical Evaluation Procedures
Statistical procedures were applied to evaluate whether the AERCOARE-MOD modeling
approach was biased towards underestimates using the Pismo Beach, Cameron, Ventura, and
Carpinteria overwater tracer studies. In addition, the procedures were applied to examine
which of the five cases for preparing the meteorological data performed statistically better
within a regulatory modeling framework. The procedures are designed to evaluate how well the
modeling approach explains the frequency distribution of the observed concentrations,
especially the upper-end or highest observed concentrations. The analysis also measures the
22
-------
October 2012
model's ability to explain the temporal variability of the observations. Given two unbiased
models, the approach with the least amount of scatter would generally be preferred.
The statistical methods and measures are similar to the techniques applied in the EPA
evaluation of AERMOD (EPA, 2003) with a few changes as will be discussed below.
Quantile-quantile (Q-Q) plots were prepared to test the ability of the model predictions to
represent the frequency distribution of the observations. Q-Q plots are simple ranked
pairings of predicted and observed concentration, such that any rank of the predicted
concentration is plotted against the same ranking of the observed concentration. The Q-Q
plots can be inspected to examine whether the predictions are biased towards
underestimates at the important upper-end of the frequency distribution.
The robust highest concentration (RHC) has been used in most EPA model evaluation
studies to measure the model's ability to characterize the upper end of the frequency
distribution. Note that this can also be accomplished by visual inspection of the Q-Q plots.
The RHC is calculated from:
RHC = cn + (c-cn}ln
where cn is the nth highest concentration and ~c is the average of the (n-1) highest
concentrations. For the small sample size data sets in the current analysis, n was taken to be
10.
Log-log scatter diagrams were prepared to test the ability of the model to explain the
temporal variability in the observations. When the data from all studies are combined, the
combined scatter diagrams can also be used to infer whether the model can explain the
variability between the studies.
Tables of statistical measures and "sigma" plots were prepared using the BOOT (Level
2/2/2007) statistical model evaluation package (Chang and Hanna, 2005). The BOOT
program is an update of the package applied in the CALPUFF evaluation (Earth Tech, 2006).
The BOOT program was applied to provide information regarding bias of the mean, scatter
or precision, and confidence limits using the bootstrap resampling method. The statistics
were performed using the natural logarithm of the predictions and observations. Such
geometric methods are more appropriate than linear statistics when the data exhibit a log-
normal distribution and/or vary over several orders of magnitude. Bias of the geometric
mean is measured from:
MG = e
where c0 and cp are the observed and predicted concentrations, respectively. MG is a
symmetric measure that is independent of the magnitude of the concentration where for a
perfect model, MG = 1 and a factor of two is bounded by 0.5 < MG < 2. Note there are no
zero observed or predicted concentrations in the evaluation data set. The scatter or
precision is measured with the geometric variance:
23
-------
October 2012
VG is similar to the normalized mean square error in linear statistics and measures scatter
about a 1:1 observation-to-prediction ratio. A random scatter of a factor-of-two is
equivalent to VG = 1.6, and VG = 12 would indicated a random scatter equivalent to a
factor-of-five bias.
The BOOT program also provides other descriptive statistics, including the geometric
correlation coefficient and the fraction within a factor-of-two. Importantly, bootstrap
resampling methods are used by BOOT to test whether differences in MG or VG between
the different cases are statistically significant.
The results of the performance evaluation using the methods outlined above are presented in
the next section. Complete output listings from the BOOT program for each dataset and the
combined dataset are attached.
24
-------
October 2012
4.0 RESULTS
AERCOARE-MOD simulations were conducted to predict concentrations from the Pismo Beach,
Cameron, Ventura, and Carpinteria field studies using four different methods for the
preparation of the meteorological data, and for Case 5, the differences caused by an alternative
lateral dispersion term. AERMOD (Version 12060) was applied using default dispersion options
for rural flat terrain for the Pismo Beach, Ventura and Cameron simulations. Complex terrain
was assumed for the Carpinteria data set. Peak predicted concentrations were compared to
peak observed concentrations resulting in a total of 101 paired samples for statistical analysis
with the techniques described in Section 3.3. In order to be independent of the tracer emission
rate, the simulations were performed with a unit emission rate of 1 g/s and the observations
were normalized by the tracer release rate providing concentrations in units of u.s/m3.
Figure 5 to Figure 9 show log-log scatter diagrams for the five cases. Each plot shows the 1:1
and factor-of-2 bounds for the prediction-to-observation ratio. The scatter diagrams for the five
cases are similar with only subtle differences. Most of the differences occur at the upper end of
the frequency distribution primarily populated by the Carpinteria complex terrain data set. In
this region, a couple of the cases over-predict the highest observations. There are also
significant differences between the cases for the mid-range concentrations from the Pismo
Beach and Ventura data sets, but these differences are difficult to pick out from the scatter
diagrams.
Q-Q plots for the combined data set and each of the four individual data sets are shown in
Figure 10 to Figure 14. Each plot shows the differences caused by the four different methods
used to prepare the meteorological data, and for Case 5 the differences caused by an
alternative lateral dispersion term. Figure 15 to Figure 19 show Q-Q plots for each of the five
cases where the results from each field study are compared to one another.
25
-------
October 2012
1000.00
100.00
j| 10.00
0.10
AERCOARE (Case 1) vs Observations
Ventura, Pismo Beach, Cameron and Carpinteria OCD Data Sets
0.01
Case 1: Abs(L)>5, Obs
00, Venk Mech Zi
O.oi
0.10
1.00 10.00
Observed Concentration (ns/m3)
100.00
Figure 5. Scatter Plot of AERCOARE Case 1 versus Observations
1000.00
26
-------
October 2012
1000.00
100.00
AERCOARE (Case 2) vs Observations
Ventura, Pismo Beach, Cameron and Carpinteria OCD Data Sets
Cameron
O Carpinteria
D Ventura
Case 2: Abs(L)>5, Pred o0
Venk Mech Zi
0.01
0.01
o.io
1.00 10.00
Observed Concentration (ns/m3)
100.00
Figure 6. Scatter Plot of AERCOARE Case 2 versus Observations
1000.00
27
-------
October 2012
AERCOARE (Case 3) vs Observations
Ventura, Pismo Beach, Cameron and Carpinteria OCD Data Sets
1 0DD Ofi
100.00
3.10.00 -
1
8
T>
2 1.00 -
1
0.10
n m -
*f
.'''
X
X
,'
/
s
* Pismo
O
u
X
r'
Zameron
larpinteria
Centura
X
x'
X
s
'
'
/
r *
r
X
0*
s
/
,'
*
X
f
^& ^
\
?
1
s
X
X
,
I
r"
L-
>
<
Ğ
ğ
*
p
rj'
.'' ^
ğa
^
X£-
X
Ğ
^P
k ^
x
y
ğ--
kx4
i-r^i
^>^'
+*
k~
x"
JJ^L
\
r ^
^0""
' O
>
O
,'
/
r'
X
X
X
X
X
>
^0 °
1 III
V
X
X
4
X
X
£
A x
' ^
xi-r
t'
'
x
X
X
^
>
'
*
x
Case 3: Abs(L)>l, Obs
00, Obs Zi
0.01 0.10 1.00 10.00
Observed Concentration ((is/m3)
<
100.00 1000.00
Figure 7. Scatter Plot of AERCOARE Case 3 versus Observations
28
-------
October 2012
100.00
I
3.10.00 -
ted Concentration
h-
b
o
0.10 -
n 01
^
,'*
X
'
,'
/
.'
Ventura
Cameron
0<
O
X
X*
r*
larpinteria
/entura
X
x-
X
x
X
x'
^
,
ri
J
^
^
x
4
;'
-A
_ D
O
Q'
i
y
>nb '"'
S* .
X
*
X
s^
*
4
V
4
?
!
r
*
x'
^
4
'
>
<^S
x
*' A AX^
i
*
*
XX/
x'
x
X
r
/
y
'
X
'
X
^
rf*
x'
r
x
^x
x'"
x
/
*
Case 4: Abs(L)>5, Obs
00, Obs Zi
0.01 0.10 1.00 10.00
Observed Concentration (ns/m3)
100.00
X
f*
X
^
X
*
1000.00
Figure 8. Scatter Plot of AERCOARE Case 4 versus Observations
29
-------
October 2012
1000.00
100.00
I
3.10.00
ted Concentration
b
o
.a
0.10
Om -
Fv
f,''
^r ,'
AERCOARE
Ventura, Pismo Beach, C
Cameron
0
*
*
/
''
Carpinteria
/entura
X
i
X
X
X
X
s
s
'
s
X
*f
,'
X
1
*-
X
-* A
/f
**
r
^^
X
x'
X
D
X
x'
rf
i
<
>
(Case 5) vs Observations
ameron and Carpinteria OCD Data Sets
,'
'
A
T Afc ,
^2
*
w
^x
f,
r
4
>Ğ
X
x'
x^
*
-iri^
v v
^
X
-
^
'
)'
4
Ğ
>
'
X
/v
ğxĞ'
x'
x
x
^
x*ğ
f^*'''
'^X'
*
X
Case 5: Abs(L)>5, Obs o0,
Venk Mech Zi, Draxler ay
0.01 0.10 1.00 10.00 100.00 1000.00
Observed Concentration (ns/m3)
Figure 9. Scatter Plot of AERCOARE Case 5 versus Observations
30
-------
October 2012
1000.00
QQ Plot AERCOARE vs Observations
Ventura, Pismo Beach, Cameron and Carpinteria OCD Data Sets
XCase 1: Abs(L)>5, Venk Zi, Obs oO
DCase 2: Abs(L)>5, Venk Zi, Mod oG
* Case 3: Abs(L)>l, Obs Zi, Obs oG
Case 4: Abs(L)>5, Obs Zi, Obs oG
ğ Case 5: Abs(L)>5, Venk Zi, Obs oG, Draxler ay
0.01
0.01
0.10
1.00 10.00
Observed Concentration {jis/m3)
100.00
1000.00
Figure 10. QQ Plot of AERCOARE versus All Observations
31
-------
October 2012
1000.00
QQ Plot AERCOARE vs Observations
Carpinteria OCD Data Set
XCase 1: Abs(L)>5, Obs oO, Venk Zi
QCase 2: Abs(L)>5, Pred oO, Venk Zi
*Case 3: Abs(L)>l, Obs aİ, Obs Zi
Case 4: Abs(L)>5, Obs oG, Obs Zi
ğ Case 5: Abs(L)>5, Obs oG, Venk Zi, Draxler cy
0.01
0.01
0.10
1.00 10.00
Observed Concentration (ns/m J)
100.00
Figure 11. QQ Plot of AERCOARE versus Carpinteria Observations
1000.00
32
-------
October 2012
100.00
QQ Plot AERCOARE vs Observations
Cameron OCD Data Set
0.01
0.01
0.10
1.00
Observed Concentration (|is/m3)
10.00
100.00
Figure 12. QQ Plot of AERCOARE versus Cameron Observations
33
-------
October 2012
100.00 -i
ntration (n$/m3)
h
* C
5 C
J
Predicted Conce
O *
- C
D C
n m -
QQ Plot AERCOARE vs Observations
Ventura OCD Data Set
XCase 1: Abs(L)>5, Obs oO, Venk Zi
DCase 2: Abs(L)>5, Pred oG, Venk Zi
*Case 3: Abs(L)>l, Obs oG, Obs Zi
Case 4: Abs(L)>5, Obs oG, Obs Zi
ğ Case 5: Abs(L)>5, Obs oG, Venk Zi, Draxler ay
ff
X
0.01
/
X
x"
^x
,'-'
J, *
x
x
x'
s/
,'
X*
X
^x
x
ğ
^x
^
X
x
x
X
^^^
s*
/
x
x
x'
^
*
^
X
\?*
,**
1
X;
1
y
B**
/
x^
/Xn
9
x'
85
X
*
^
X
X
i
]
3
f <
H ' 1
u
fii
^
\^^ J+*
^T ^
-^
*
^
x
£
ğ'
X
,''
_
^x
x'
x'
X*
X
p
X
X
x'
x
'
t
x
>
*
f*'
,'
/
J*
X
x
0+
0+
/
fs
X
^ x
x*
r*
x
x
X
''
^^
X
X
X
1
-
.-
0.10 1.00 10.00 100.00
Observed Concentration (ns/m3)
Figure 13. QQ Plot of AERCOARE versus Ventura Observations
34
-------
October 2012
ntration (|is/m3)
r
i S
Predicted Conce
;>
D
n m -
QQ Plot AERCOARE vs Observations
Pismo Beach OCD Data Set
XCase l:Abs(L)>5, Obs 00, VenkZi
DCase 2: Abs(L)>5, Pred C0, Venk Zi
* Case 3: Abs(L)>l, Obs C0, Obs Zi
Case 4: Abs(L)>5, Obs O0, Obs Zi
* Case 5: Abs(L)>5, Obs o0, Venk Zi, Draxler ay
,
rj^f1-
*'
X
X
^
X
x'
X'
^x
^*
0.01
,/
,'
/
f
,
/
,
f
s
f
'
s
'
'
s
z
^x
^
x
0+
^^^^ i i
X
+*
/
x
-,X*
p"
x'
^"
x'
/
X
X
^/
X
--
^
**
X
f
'
f
f
0
4
^*f
s
/
A
,'t T
X
I
d
F
jf\ w
X ' '^
^ Cj
*
?
f
9*
&
I
/
X
rf
X*
/
,
/
'
\
0
X
*
s
^**
x X
p*
*
+ *
*
,'
/
,-'
X
^ ^
/
.''
s
i
f
'
^
,
^
'
*
0.10 1.00 10.00 100.00
Observed Concentration (|as/mB)
Figure 14. QQ Plot of AERCOARE versus Pismo Beach Observations
35
-------
October 2012
1000.00
100.00
0.01
0.01
QQ Plot AERCOARE (Case 1) vs Observations
Ventura, Pismo Beach, Cameron and Carpinteria OCD Data Sets
0.10
1.00 10.00
Observed Concentration {|is/m3)
100.00
1000.00
Figure 15. QQ Plot of AERCOARE Case 1 versus Observations
36
-------
October 2012
1000.00
QQ Plot AERCOARE (Case 2) vs Observations
Ventura, Pismo Beach, Cameron and Carpinteria OCD Data Sets
I
Pismo
A Cameron
O Carpinteria
QVentura
0.01
0.01
0.10
1.00 10.00
Observed Concentration (|is/m3)
100.00
1000.00
Figure 16. QQ Plot of AERCOARE Case 2 versus Observations
37
-------
October 2012
1000.00
QQ Plot AERCOARE (Case 3) vs Observations
Ventura, Pismo Beach, Cameron and Carpinteria OCD Data Sets
* Pismo
A Cameron
O Carpinteria
QVentura
0.01
0.01
0.10
1.00 10.00
Observed Concentration (ns/m3)
100.00
1000.00
Figure 17. QQ Plot of AERCOARE Case 3 versus Observations
38
-------
October 2012
1000.00
100.00 -
m
j* 10.00 -
ted Concentration
D
3
f
0.10
0.01 -
0.
QQ Plot AERCOARE (Case 4) vs Observations
Ventura, Pismo Beach, Cameron and Carpinteria OCD Data Sets
f'
pX
X^
^ Pismo
A Cameron
OCarpinteri
D Ventura
X
X
j/
,''
X
^x
ğ*
-X*
1
3
x
^X
X
x
^
,'
'
Ğ
^
'
r
f*
x'
x'
X
x
x
x **
,p
*
1
1
<\
Lx'
i
D
^
1
A
*
<
ID E'
J
! i
J
B *X<
^r^
x ^
2 A J|
V ^
F
?
>
1
r
r
/
^''
i ^"
l^^
^^
v>
x?s
x
^s>
!x^ "' '
^^
X
X
X
-
.1
*
o'
x
0 -'''/'.'
y\''
/
2
2
^X
X
*
x
x
-
i
01 0.10 1.00 10.00 100.00 1000.00
Observed Concentration (ns/m J)
Figure 18. QQ Plot of AERCOARE Case 4 versus Observations
39
-------
October 2012
1000.00 -i
inn nn
m
3-10.00 -
ted Concentration
D
D
TJ
£
&
0.10
n m -
QQ Plot AERCOARE (Case 5) vs Observations
Ventura, Pismo Beach, Cameron and Carpinteria OCD Data Sets
f'
X
/
* Pismo
A Cameron
OCarpinteri
QVentura
X
f~
/
X
^ ^
^\
X
^x*
x'
3
x
^
x
X
^/
^
X
f
4
x
'
.
*
1 f 1
+*
,'
/
*
*
_J_
^ ^
X
x'
xfr
^xCr ^.
X
x
'
p
r
X
**
4
4,
X
A '
XI '
^
4
,'
+
A'
S
'
/
,''
-
/
^
: .x-
>
X
ğ
g
$f'*
X
^+
x
X*
X
^
*
/
^
/
?
X
^.'"
ff
,'
'
f'
/
,''
X
X
x'
^X*
x
x'
J
>
X
>
X1
^
^
>
-
0.01 0.10 1.00 10.00 100.00 1000.00
Observed Concentration (ns/m3)
Figure 19. QQ Plot of AERCOARE Case 5 versus Observations
40
-------
October 2012
Comparing the Q-Q plots for the combined data set and each of the four field studies, the five
AERCOARE-MOD simulations generally predict the frequency distribution within a factor-of-
two. The predictions tend to be biased towards over-prediction for the highest concentrations
and under-prediction for the lower-end of the frequency distribution. This tendency is most
apparent for the Ventura (Figure 13) and Pismo Beach (Figure 14) data sets. In most instances
higher concentrations are over-predicted using the AERMOD oe estimates (Case 2).
Importantly, AERCOARE-MOD does not appear to be biased towards underestimates for the
higher end of the frequency distribution, regardless of the options examined in this study.
Comparing the optional cases using the Q-Q plots, there is no clear choice for the best method
to prepare the meteorological data. Case 2 using the AERMOD oe estimates seems to result in
over-prediction for the combined data set and each individual data set. Depending on the data
set, the method used to estimate the mechanical mixing height influenced the results. The
observed mixing height seemed to perform the best for Pismo Beach, while the Venketram
estimate worked the best overall. Allowing the Monin-Obukhov length to become very stable
(Case 3) also resulted in severe over-predictions in some instances. Removing the dependency
of the lateral dispersion term on mixing height (Case 5) also improved model performance in
some instances, especially the Carpinteria data set where observed mixing heights appear to be
the most uncertain.
The BOOT program statistics for each data set are summarized in Table 9 where the best
performing modeling approach is highlighted for each statistic and data set. The full output of
the BOOT program is attached in Appendix B. Table 9 also shows the RHC calculated for each
data set and modeling case. For all the data sets and especially the Pismo Beach data set, the
predicted concentrations are more variable than the observations. The Pismo Beach field study
had the poorest paired-in-time model performance and the RHC is significantly over-predicted
by each modeling alternative. Overall, the performance statistics tend to be the best for Case 5
with the modified lateral dispersion term followed by Case 1. The poorest performance usually
was associated with using predicted AERMOD oe estimates (Case 2).
41
-------
October 2012
Table 9. Performance Evaluation Statistical Results by Data Set and AERCOARE-MOD Case.
Data Set
All Data (101
samples)
Ventura, CA (17
samples)
Pismo Beach,
CA (31 samples)
Cameron, LA
(26 samples)
Case
0
1
2
3
4
5
0
1
2
3
4
5
0
1
2
3
4
5
0
1
2
3
4
5
Description
Observations
Abs(L)>5, ObsoO, VenkZi
Abs(L)>5, PredoO, VenkZi
Abs(L)>l, Obs oO, Obs Zi
Abs(L)>5, Obs oO, Obs Zi
Abs(L)>5, Obs oO, Venk Zi, Draxler ay
Observations
Abs(L)>5, Obs oO, VenkZi
Abs(L)>5, Pred oO, VenkZi
Abs(L)>l, Obs oO, Obs Zi
Abs(L)>5, Obs oO, Obs Zi
Abs(L)>5, Obs oO, Venk Zi, Draxler ay
Observations
Abs(L)>5, Obs oO, VenkZi
Abs(L)>5, Pred oO, VenkZi
Abs(L)>l, Obs oO, Obs Zi
Abs(L)>5, Obs oO, Obs Zi
Abs(L)>5, Obs oO, Venk Zi, Draxler ay
Observations
Abs(L)>5, Obs oO, VenkZi
Abs(L)>5, Pred oO, VenkZi
Abs(L)>l, Obs oO, Obs Zi
Abs(L)>5, Obs oO, Obs Zi
Abs(L)>5, Obs oO, Venk Zi, Draxler ay
Geom.
Mean
(Hs/m3)
4.5
4.7
6.8
4.7
4.9
4.6
1.2
1.6
2.4
2.1
2.1
1.4
3.5
3.7
5.9
3.2
3.8
3.3
3.2
4.1
4.2
3.7
3.7
4.1
Geom.
Std.
1.36
1.60
1.73
1.67
1.57
1.53
0.76
1.03
1.37
1.18
1.18
0.88
0.50
1.39
1.45
1.40
1.23
1.33
1.41
1.84
1.87
1.77
1.79
1.70
MG
1.00
0.96
0.67
0.97
0.93
0.99
1.00
0.73
0.50
0.57
0.57
0.87
1.00
0.93
0.59
1.09
0.91
1.04
1.00
0.78
0.76
0.86
0.84
0.76
VG
1.00
3.21
4.93
4.05
3.23
2.60
1.00
1.81
5.28
2.58
2.58
1.41
1.00
6.17
12.90
7.53
4.30
4.80
1.00
3.03
3.60
2.67
2.68
2.58
Geom.
Correl.
Coef.
1.00
0.75
0.72
0.71
0.74
0.78
1.00
0.73
0.62
0.75
0.75
0.77
1.00
0.27
0.04
0.14
0.26
0.35
1.00
0.83
0.81
0.83
0.84
0.84
Frac.
Factor of
2
1.00
0.54
0.47
0.48
0.47
0.55
1.00
0.77
0.59
0.59
0.59
0.88
1.00
0.45
0.26
0.45
0.45
0.42
1.00
0.42
0.42
0.46
0.46
0.46
RHC
(HS/m3)
125
146
311
493
333
117
4
6
20
8
8
4
9
43
55
19
20
30
41
49
53
40
44
36
VG is a measure of geometric variance or scatter, VG = exp(average(ln(Co/Cp)))
MG is a measure of bias about the geometric mean, MG = exp(average((ln(Co/Cp))A2))
RHC = "Robust Highest Concentration" based on top 10 samples
Best performing modeling approach or Case is highlighted in red
42
-------
October 2012
Table 9. Performance Evaluation Statistical Results by Data Set and AERCOARE-MOD Case (Continued).
Data Set
Carpinteria, CA
(27 samples)
Case
0
1
2
3
4
5
Description
Observations
Abs(L)>5, Obs aO, Venk Zi
Abs(L)>5, Pred aO, Venk Zi
Abs(L)>l, Obs aO, Obs Zi
Abs(L)>5, Obs aO, Obs Zi
Abs(L)>5, Obs aO, Venk Zi, Draxler ay
Geom.
Mean
(Hs/m3)
20.1
14.0
24.3
15.0
14.2
15.5
Geom.
Std.
0.93
1.19
1.29
1.50
1.36
0.97
MG
1.00
1.44
0.83
1.34
1.42
1.30
VG
1.00
2.29
2.10
3.95
3.19
1.90
Geom.
Correl.
Coef.
1.00
0.72
0.76
0.66
0.67
0.69
Frac.
Factor of
2
1.00
0.59
0.67
0.44
0.41
0.56
RHC
(HS/m3)
137
172
330
470
329
129
VG is a measure of geometric variance or scatter, VG = exp(average(ln(Co/Cp)))
MG is a measure of bias about the geometric mean, MG = exp(average((ln(Co/Cp))A2))
RHC = "Robust Highest Concentration" based on top 10 samples
Best performing modeling approach or Case is highlighted in red
43
-------
October 2012
Sigma-plots prepared from the BOOT program output are shown in Figure 20 to Figure 24 for
the combined data set and each individual data set. Sigma-plots display MG (bias) plotted
against VG (scatter). The 95 percent confidence limits on MG are also shown based on the
bootstrap resampling techniques applied by BOOT. For the combined data set, Case 2 (AERMOD
OQ estimates) significantly over-predicts observations, has more scatter, and predicts
significantly higher concentrations than the other cases. Examination of the attached BOOT
output listing suggests Case 5 (Draxler oy) has statistically less significant scatter than all the
other cases.
The Cameron sigma-plot in Figure 21 again shows Case 2 has the most scatter (highest VG) and
the BOOT output suggests these differences are significant at the 95 percent confidence level.
All the cases are biased towards over-prediction with Case 3 and Case 4 being the statistically
least biased.
All the Pismo Beach cases in Figure 23 have a significant amount of scatter and do not perform
as well as for the Cameron, Ventura or Carpinteria field studies. Based on a comparison
between Case 3 and Case 4, restricting the Monin-Obukhov length such that Abs (/.) > 5 seems
to improve performance, but often not in a statistically significant manner. This restriction
appears to help for the other sites as well when extremely stable conditions occurred.
The Ventura sigma-plot in Figure 24 again shows that Case 2 has the most scatter (highest VG)
and the BOOT output suggests these differences are significant at the 95 percent confidence
level. All the cases except Case 5 are biased towards over-prediction. Some over-prediction may
be the result of not accounting for enhanced dispersion caused by air modification as the
plumes travel over land since the receptors are located 500 m to 1 km inland.
The complex terrain field study at Carpinteria is the exception to the trends from the other data
sets as shown in Figure 22. Case 2 (AERMOD OQ) predicts significantly higher than the cases with
the observed oe data but in this instance these predictions are less biased overall. Case 1 is
biased towards under-prediction for Carpinteria, but examination of the Q-Q plot and scatter
diagram in Figure 5 and Figure 14 shows this Case's performance is relatively good at the upper-
end of the observed frequency distribution.
44
-------
October 2012
AERCOARE Ventura, Pismo Beach, Cameron & Carpinteria Data Sets
All Blocks as Ln(Co/Cp)
8.
4. -
2. *
1.
0.25
0.50
1.00
2.00
4.00
MG (with 95% conf. int.)
Overprediction Underprediction
Figure 20. Sigma Plot for All Sites
45
-------
October 2012
>
8.
4.
2.
1.
AERCOARE Cameron Data Set
All Blocks as ln(Co/Cp)
0.25
0.50
1.00
2.00
4.00
MG (with 95% conf. int.)
Overprediction Underprediction
Figure 21. Sigma Plot for Cameron
46
-------
October 2012
>
8.
4. -
2. -
1.
AKRCOARFl Carpmteria Data Set
All Blocks as ln(Co/Cp)
0.25
0.50
1.00
2.00
4.00
MG (with 95% conf. int.)
Overprediction Underprediction
Figure 22. Sigma Plot for Carpinteria
47
-------
October 2012
>
16.
2.
1.
F: Pismo Beach Data Set
All Blocks as ln(Co/Cp)
8.
4.
jase 2
rase 1
Dase 4
0.25
0.50
1.00
2.00
4.00
MG (with 95% conf. int.)
Overprediction Underprediction
Figure 23. Sigma Plot for Pismo Beach
48
-------
October 2012
8.
>
2.
1.
AFRCOARE Ventura Data Set
AJ1 Blocks as ln(Co/Cp)
4. -
0.25
0.50
1.00
2.00
4.00
MG (with 95% conf. int.)
Overprediction Underprediction
Figure 24. Sigma Plot for Ventura
49
-------
October 2012
5.0 SUMMARY
ENVIRON conducted this analysis to evaluate the combination of AERCOARE/AERMOD as a
viable regulatory dispersion modeling approach for offshore sources. The proposed alternative
approach bypasses the AERMET meteorological preprocessor using AERCOARE and overwater
meteorological measurements. ENVIRON conducted a model evaluation analysis using data
from four offshore tracer experiments. The conclusions from our analysis are as follows:
The AERCOARE-MOD modeling approach was not biased towards underestimates at the
high-end of the concentration frequency distribution.
The AERCOARE-MOD approach performed better using the observed oe measurements. The
internal AERMOD estimates of OQ resulted in concentrations that were biased towards over-
predictions and often caused statistically significant higher scatter as measured by the
geometric variance (VG).
AERCOARE-MOD predictions were sensitive to the mixing height. An estimate of the
mechanical mixing height based on the friction velocity, as in AERMET, was a better
alternative than using the observed mixing height from the field studies. A portion of this
sensitivity was due to the AERMOD equation for ambient lateral dispersion that depends on
the mixing height. A replacement equation similar to OCD and CALPUFF reduced the scatter
in some of the comparisons.
The AERCOARE-MOD approach was sensitive to assumptions during low wind speed
conditions. Restricting the Monin-Obukhov length such that Abs (/.) > 5 seems to improve
performance by limiting the occurrence of extremely unstable or stable conditions.
The results of current study where data from the Ventura field study was added to the
analysis are consistent with the model evaluation results previously submitted to RIO and
the EPA Model Clearinghouse (EPA 2011a; EPA 2011b).
Based on our analysis, we believe that the AERCOARE-MOD approach is a more suitable
modeling technique than either AERMET/AERMOD or OCD for regulatory simulations of sources
in offshore areas. The combination of surface fluxes predicted by the COARE algorithm and
measured overwater meteorological data is preferred to the conventional application of
AERMET. For the dispersion model, AERMOD is preferred over OCD because of the PRIME
downwash algorithm, the ability to simulate volume sources, and the importance of the
PVMRM algorithm for assessing the 1-hour NO2 ambient standard. AERCOARE-MOD was not
biased towards underestimates in the field studies examined in this study.
50
-------
October 2012
6.0 REFERENCES
Chang, J.C. and K.J. Hahn, 1997. User's Guide for the Offshore and Coastal Dispersion (OCD)
Model Version 5. MMS Contract No. 1435-96-PO-51307, November, 1997. Available
from: http://www.epa.gov/ttn/scram/dispersion prefrec.htmffocd
Chang, J.C., and S.R. Hanna, 2005. Technical Descriptions and User's Guide for the BOOT
Statistical Model Evaluation Software Package, Version 2.0. July 10, 2005. Available
from: http://www.harmo.org/Kit/Download.asp
DiCristofaro, D.C. and S.R. Hanna, 1989. OCD The Offshore and Coastal Dispersion Model,
Version 4, Volume I: User's Guide. MMS Contract No. 14-12-001-30396, November
1989.
Earth Tech, 2006. Development of the Next Generation of Air Quality Models for the Outer
Continental Shelf (OCS) Applications, Final Report: Volume 1. Prepared for MMS,
Contract 1435-01-01-CT-31071, March 2006.
EPA, 2003. AERMOD: Latest Features and Evaluation Results. EPA, OAQPS, Research Triangle
Park, NC 27711, EPA-454/R-03-003, June 2003.
EPA, 2004a. User's Guide for the AMS/EPA Regulatory Model - AERMOD . Publication No. EPA-
454/B-03-001. OAQPS, Research Triangle Park, NC 27771, September 2004.
EPA, 2004b. User's Guide for the AERMOD Meteorological Preprocessor (AERMET). Publication
No. EPA-454/B-03-002. OAQPS, Research Triangle Park, NC 27771, November 2004.
EPA, 2004c. User's Guide for the AERMOD Terrain Preprocessor (AERMAP). Publication No.
EPA-454/B-03-003. OAQPS, Research Triangle Park, NC 27771, September 2004.
EPA, 2011a. Memorandum: Model Clearinghouse Review of AERMOD-COARE as an Alternative
Model for Application in an Arctic Marine Ice Free Environment. From George Bridgers,
EPA Model Clearinghouse Director, to Herman Wong, EPA Regional Atmospheric
Scientist, Office of Environmental Assessment, OEA-095, EPA Region 10, May 6, 2011.
EPA, 2011b. Memorandum: COARE Bulk Flux Algorithm to Generate Hourly Meteorological
Data for Use with the AERMOD Dispersion Program; Section 3.2.2.e Alternative Refined
Model Demonstration. From Herman Wong, EPA Regional Office Modeling Contact to
Tyler Fox, Lead Air Quality Modeling Group, Office of Air Quality Planning and Standards.
April 1, 2011.
Fairall, C.W., E.F. Bradley, J.E. Hare, A.A. Grachev, and J.B. Edson, 2003. "Bulk Parameterization
of Air-Sea Fluxes: Updates and Verification for the COARE Algorithm." J. Climate, 16,
571-591.
Richmond, K. and R. Morris, 2012. Draft User's Manual: AERCOARE Version 1.0. Prepared for US
EPA Region 10, 1200 Sixth Avenue, Seattle, WA 98101, EPA Contract EP-D-08-102, Work
Assignment 5-17, EPA 910-R-12-007, October 2012.
51
-------
October 2012
Schulman, L. L, D.G. Strimaitis and J.S. Scire, 2000. Development and Evaluation of the PRIME
Plume Rise and Building Downwash Model. Journal of the Air and Waste Management
Association, 50, 378-390.
Venketram, A., 1980. "Estimating the Monin-Obukhov Length in the Stable Boundary Layer for
Dispersion Calculations." Bound. Layer Meteor., 19, 481-485.
52
-------
October 2012
APPENDIX A: SENSITIVITY TO ASSUMED MINIMUM MIXING HEIGHTS
53
-------
October 2012
APPENDIX A: SENSITIVITY TO ASSUMED MINIMUM MIXING HEIGHTS
This appendix examines the sensitivity of model evaluation results to the minimum mixing
height allowed by AERCOARE. The mixing height is an important scaling variable in many
different AERMOD algorithms. In Case 1, Case 2, and Case 5, the mechanical mixing height is
calculated using the same Venketram algorithm as employed by the AERMET meteorological
preprocessor. AERMET calculates the mechanical mixing height (z!m) from the friction velocity
(us) according to:
zim = 2300us3/2
In AERMET, and optionally in AERCOARE, the initial estimate is smoothed based on the previous
estimate to allow for residual turbulence from the previous hour. The mechanical mixing height
trends towards zero as the friction velocity or wind speed approach zero. Over water very low
friction velocities occur during light winds. For example for the Carpinteria field study, the
COARE predicted friction velocities during several hours are less than 0.005 m/s, resulting in
mechanical mixing heights less than 1 m. Out of the 36 hours of data, 50 percent are less than
25 m for the light winds observed during this study.
There are both numerical and practical reasons for specifying a minimum mixing height. A
minimum mixing height must be used with the AERMOD model since the variable is used in the
denominator of several equations. For example the AERMOD horizontal dispersion parameter
for ambient turbulence (oy) is calculated from:
max(z, .0.46)
where z, is the mixing height, z is the height of the plume centerline, x is the downwind
distance, ov is plume average standard deviation of the crosswind velocity, and u the plume
average wind velocity. As the mixing height goes to zero, very small plume widths are predicted
and the mixing height must be limited to some small value to keep the equation from becoming
indeterminate. Currently, AERMOD restricts the mixing height to be greater than 1 m.
The above equation and the equation used by AERMOD for the vertical dispersion from sources
near the surface differ from simpler expressions used by CALPUFF, OCD and many other
models, because the authors cited poor performance for the Prairie Grass field experiment.5
The equation above is an empirical fit to the Prairie Grass data set and should be applied with
caution when the variables are well outside those used for the fit. The Prairie Grass field study
is the sole experiment (out of 17) that examined dispersion from a near surface release. In the
other datasets used in the AERMOD model evaluation study, plumes were influenced by
5 Cimorelli, A.J., Perry, S.G., Venketram, A., Weil, J.C., Paine, R.J., Wilson, R.B., Lee, R.F., Peters, W.D., and R.W.
Brode, 2005. "AERMOD: A Dispersion Model for Industrial Source Applications. Part I: General Model Formulation
and Boundary Layer Characterization." J. Applied Meteorology, 44, 683-693.
54
-------
October 2012
downwash or were sufficiently high that the surface dispersion algorithms in AERMOD are not
that important. The minimum mixing height used in the Prairie Grass experiment simulations
was 67 m based on the input files from EPA's website.6 Mixing heights of 1 m to 10 m are well
outside the range used to develop the near surface dispersion algorithms in AERMOD.
In order to test the sensitivity of the model evaluation results to the assumed minimum mixing
height, ENVIRON reran Case 1 with minimum mixing heights of 1 m, 5 m, and 15 m to compare
with the simulations in the main body of this report where 25 m was assumed. This assumption
affects the results from the evaluation of the Pismo Beach and Carpinteria data sets. Winds
during the Ventura and Cameron studies were sufficient to keep predicted mechanical mixing
heights above 25 m for all hours.
Figure A-l shows a scatter diagram where predictions for each assumed minimum mixing
height are compared to the observed normalized concentrations for Pismo Beach and
Carpinteria. Q-Q plots for Carpinteria and Pismo Beach are shown in Figure A-2 and Figure A-3,
respectively. The predictions for Pismo Beach were only slightly affected when mixing heights
were allowed to be lower than 25 m. However, the predictions for Carpinteria were up to three
times higher when the AERMOD default of 1 m was allowed resulting in severe over prediction
at the upper end of the frequency distribution.
ENVIRON recommends a default minimum of 25 m be used to limit mixing heights when the
Venketram algorithm is used for the mechanical mixing height. The AERMOD default limit of
1 m potentially results in very high predictions that are not supported by the tracer data in the
Carpinteria study and is outside the limits used to develop the empirical algorithm AERMOD
employs for the horizontal dispersion parameter.
6 The Prairie Grass AERMOD files can be found at http://www.epa.gov/ttn/scram/7thconf/aermod/pgrass.zip. The
meteorological file used is "PGRSURF.222"
55
-------
October 2012
1000.00
100.00
I
10.00
1.00
AERCOARE (Case 1) vs Observations
Pismo Beach and Carpinteria OCD Data Sets
0.10
*Zimm = 25m
Zimin = 1m
Ozimin = 5m
DZimin = 15m
Case l:Abs(L)>5, Obs
00. Venk Mech Zi
0.10
1.00
10.00
Observed Concentration (ns/m3)
100.00
1000.00
Figure A-l. Scatter Plot of Case 1 for Several Minimum Mixing Heights
56
-------
October 2012
1000.00
QQ Plot AERCOARE vs Observations
Carpinteria OCD Data Set
XCase IrZimin = 25m
DCase la:Zimin = 1m
*Case lb:Zimin = 5m
Case lc:Zimin = 15m
0.10
1.00
10.00
Observed Concentration (ns/m3)
100.00
1000.00
Figure A-2. QQ Plot for Case 1 Carpinteria for Several Minimum Mixing Heights
57
-------
October 2012
>redicted Concentration (}is/m3
h
* C
3 \
Om
XCase l:Zimin =
DCase la: Zimin
*Case Ib: Zimin
Case Ic: Zimin
25m
= 1m
= 5m
QQ Plot AERCOARE vs Observations
Pismo Beach OCD Data Set
= 15m
,''
X
/
0.10
X"
rxx
/
xx"
.x'"
^x
X
/
,'-'
X
**
/
,''
,x'
X
X
X'
X
x'
x'
X
X
'
x
X
1
X
/
X
1
X
X
/
X
jl
^X
x
/
H:
X
x"i /^
..^l
z
/^
2
'' H
x''
r
X
X
x"
x'
X
X
X
x
X
x'
x"'
x^^
^
x
,'
,'
.''
~2_
,''
/
/'
x"
x*'
^/
X'X
J.*
X
^x*
,x''
X
xx
X
,x
X
x^
X
XT
X
X
1.00 10.00 100.00
Observed Concentration (|is/m3)
Figure A-3. QQ Plot for Case 1 Pismo Beach for Several Minimum Mixing Heights
58
-------
October 2012
APPENDIX B: BOOT PROGRAM OUTPUT
59
-------
October 2012
APPENDIX B: BOOT PROGRAM OUTPUT
Boot Program Output for All Data Sets
Combined
experiments = 101
models = 6
the observed data counted as one)
observations = 101
might be multiple observations In each experiment, If the ASTM option Is chosen)
Is only one prediction In each experiment)
observations available for
sampling = 98
might be odd number of observations In each block)
blocks (regimes) = 4
experiments In each block (regime)
31 26 27
1.91
1.7;
HIGH 2nd HIGH
HE
451
60
-------
October 2012
Block 1: Ventura, Ca
MODEL MEAN SIGMA
PCOR
<
DBS. 0.18
n/a
1.18
1.18
Block 2: Plsmo Beach, Ca
MODEL MEAN SIGMA
PCOR
1.24
1.31
1.77
1 .16
1.45
1 .:
BIAS
HIGH 2nd HIGH
16
(N= 31)
BIAS VG CORK FA2 MG HIGH 2nd HIGH
16
61
-------
October 2012
Block 3: Cameron, La
MODEL MEAN SIGMA
PCOR
1.41
1.84
1.43
1.31
1 .'
Block 4: Carpinteria, Ca
MODEL MEAN SIGMA
PCOR
<
DBS. 3.00 0.93
n/a
1.19
(N= 26)
BIAS VG CORK FA2 MG HIGH 2nd HIGH
BIAS
HIGH 2nd HIGH
451
62
-------
October 2012
Student
t
Mean
0 .160
0 .104
0 .061
0 .061
0 .183
0 .117
0 .053
0.082
0.235
0 .113
0 . 073
0 .060
0 .169
0.102
0 .063
0.057
0.164
0 .096
0 .057
0 .055
0 .783
1.431
0 .765
1.324
1.432
1 .211
Mean
0 .171
0 . 057
0 .036
0 .045
0 .207
0 .066
0 .043
0 .045
0 .089
0 .046
0 .018
0 .039
0 .058
0 .027
0.019
0 .018
0 .252
0 .101
0 .060
0 . 066
210
0 .081
0.044
0 .063
0 .203
0.078
0 .044
0 .058
0 .179
0 .039
0 .036
0 .012
0 .193
0 .061
0 .037
1.099
1.114
1 .0;
0
63
-------
October 2012
MGfp 1.009 1.199 2.192 0.095 0.043
VG 1.049 1.460 2.564 0.213 0.083
MG 0.864 1.029 -1.342 -0.059 0.044
MGfn 0.996 1.071 1.778 0.032 0.018
MGfp 1.017 1.180 2.429 0.091 0.038
SUMMARY OF CONFIDENCE LIMITS ANALYSES BASED ON PERCENTILE CONFIDENCE LIMITS
e e e e e
12345
X
D(ln(MGfp)) among models: an 'X' indicates significantly different from zero at 95% confidence
limits
64
-------
October 2012
a a a a a
e e e e e
65
-------
October 2012
Boot Program Output for Pismo Beach
experiments = 31
models = 6
the observed data counted as one)
observations = 31
might be multiple observations In each experiment, If the ASTM option Is chosen)
Is only one prediction In each experiment)
observations available for
sampling = 30
might be odd number of observations In each block)
blocks (regimes) = 1
experiments In each block (regime)
1.24
1.77
1.16
1.:
1.45
1 .:
16
66
-------
October 2012
Conf. limits
Student'
Conf. limits
Student
t
Conf. limits
67
-------
October 2012
SUMMARY OF CONFIDENCE LIMITS ANALYSES BASED ON PERCENTILE CONFIDENCE LIMITS
C C C C C
a a a a a
68
-------
October 2012
a a a a a
e e e e e
In(MGfn) for each model: an 'X' indicates significantly different from zero at 95% confidence
limits
In(MGfp) for each model: an 'X' indicates significantly different from zero at 95% confidence
limits
e e e e e
69
-------
October 2012
Boot Program Output for Cameron
experiments = 26
models = 6
the observed data counted as one)
observations = 26
might be multiple observations in each experiment, if the ASTM option is chosen)
is only one prediction in each experiment)
observations available for
sampling = 26
might be odd number of observations in each block)
blocks (regimes) = 1
experiments in each block (regime)
1.4;
1.42
1.41
1 .!
70
-------
October 2012
Student'
Conf. limits
Student
t
Conf. limits
Conf. limits
Conf. limits
0 .
0 .
0 .
0 .
0 .
0 .
0 .
1 ,
0 .
0 .
0 .
1 .
1 .
0 .
0 .
0 .
0 .
0 .
0 .
1 .
0 .
0 .
0 .
1 ,
1 ,
0 .
0 .
0 .
0 .
0 .
0 .
0 .
0 .
1 ,
1 ,
0 .
0 .
1 .
1 .
0 .
.684
.954
.944
.898
.966
.855
.965
.007
.963
.872
. 985
.001
. 039
.955
.998
.977
.998
.784
.952
.011
.994
.799
.964
.005
.035
.869
.981
.974
.973
.996
.993
.977
.878
.060
.018
. 907
.882
.036
.007
.911
1.029
1 .118
1 .030
1 .014
1.331
0 . 981
1 .010
1 .153
1.321
0.992
1 .003
1 .140
1.334
1 .104
1 .107
1 .072
1 .830
1 .003
1 .052
1 . 259
1.818
1 .015
1.054
1 .246
1 .902
1.137
1.158
1 .180
1 .017
1 .036
1 022
1 .007
1 .228
1.184
1 .114
0 .995
1.236
1.176
1 .110
1 .007
1 ,
0 .
-0 .
-1 .
1 .
_2
1 ,
2 .
1 ,
-2 .
-1 .
2
2
0 .
1 ,
1 ,
2 .
-2 .
0 .
2
2
-1 .
0 .
2 .
2 .
-0 .
1 ,
1 .
-0 .
1 .
1
1 ,
0 .
4 ,
2 .
_2
0 .
3 .
2
1 ,
.771
.847
.665
.580
.611
.623
.174
.261
.571
.306
.371
.101
.687
.753
.968
. 026
.045
.016
.037
.274
.017
.797
.373
.159
.292
.093
.578
.488
.486
.632
024
.129
.463
.248
.859
.307
.524
.195
.347
.758
-0 .175
0 .033
-0 .014
-0 .046
0.126
-0.088
-0 .013
0 .075
0 .120
-0 .072
-0 . 006
0 .066
0 .163
0.026
0 .049
0 .023
0 .301
-0 .120
0 .001
0 .121
0.296
-0 .105
0 .008
0 .113
0 .339
-0 .006
0 .063
0 .070
-0.005
0 .016
0 007
-0 .008
0 .038
0 .114
0 .063
-0 .051
0 .043
0.098
0.055
-0 .043
0 .099
0 .038
0 .021
0.029
0 .078
0 .033
0 .011
0 .033
0 .077
0 .031
0 .004
0 .031
0 .061
0 .035
0 .025
0 .022
0.147
0 .060
0.024
0 . 053
0.147
0 . 058
0.022
0 .052
0 .148
0 .065
0 .040
0.047
0 .011
0 .010
0 007
0 .007
0.081
0 .027
0.022
0 . 022
0 .082
0 .031
0.024
0 .025
0 .
0 .
0 .
0 .
0 .
0 .
0 .
1 .
0 .
0 .
0 .
1 .
1 .
0 .
1 .
0 .
1 .
0 .
0 .
1 .
1 .
0 .
0 .
1 .
1 .
0 .
0 .
0 .
0 .
1 .
1 .
0 .
0 .
1 .
1 .
0 .
0 .
1 .
1 .
0 .
347
0.971
1 .153
1 .334
1.913
71
-------
October 2012
SUMMARY OF CONFIDENCE LIMITS ANALYSES BASED ON PERCENTILE CONFIDENCE LIMITS
C C C C C
a a a a a
e e e e e
e e e e e
72
-------
October 2012
In(MGfn) for each model: an 'X' indicates significantly different from zero at 95% confidence
limits
In(MGfp) for each model: an 'X' indicates significantly different from zero at 95% confidence
limits
e e e e e
73
-------
October 2012
Boot Program Output for Carpinteria
No. of experiments = 27
No. of models = 6
(with the observed data counted as one)
No. of observations = 27
(there might be multiple observations in each experiment, if the ASTM option is chosen)
(there is only one prediction in each experiment)
No. of observations available for
paried sampling = 26
(there might be odd number of observations in each block)
No. of blocks (regimes) = 1
No. of experiments in each block (regime)
2 7
1.19
1 .50
451
74
-------
October 2012
i .6
3.4
Student
t
5.D.
75
-------
October 2012
SUMMARY OF CONFIDENCE LIMITS ANALYSES BASED ON PERCENTILE CONFIDENCE LIMITS
76
-------
October 2012
ln(MG) for each model: an 'X' indicates significantly different from zero at 95% confidence
limits
e e e e e
a a a a a
e e e e e
77
-------
October 2012
Boot Program Output for Ventura
No. of experiments = 17
No. of models = 6
(with the observed data counted as one)
No. of observations = 17
(there might be multiple observations In each experiment, If the ASTM option Is chosen)
(there Is only one prediction In each experiment)
No. of observations available for
parled sampling = 16
(there might be odd number of observations In each block)
No. of blocks (regimes) = 1
No. of experiments In each block (regime)
17
16
1 .
1.18
-0.14 1.41
78
-------
October 2012
79
-------
October 2012
SUMMARY OF CONFIDENCE LIMITS ANALYSES BASED ON PERCENTILE CONFIDENCE LIMITS
80
-------
October 2012
ln(MG) for each model: an 'X' indicates significantly different from zero at 95% confidence
limits
e e e e e
a a a a a
e e e e e
81
------- |