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

-------