KEY CONTACTS FORM

toadspottedincurableInternet and Web Development

Dec 4, 2013 (3 years and 9 months ago)

224 views


i

KEY CONTACTS FORM


Authorized Representative:

Original awards and amendments will be sent to this individual for
review and acceptance, unless otherwise indicated.


Name:



Linda L. Bryant

Title:



Principal Contract and Grant Analyst

Complete Address:

Off
ice of Research Affairs, 200 University Office Building., University of
California, Riverside, CA 92521.

Phone Number:

(909) 787
-
5535


Payee:

Individual authorized to accept payments.


Name:



Bobbi McCracken

Title:



Assistant Accounting Officer

Complete

Address:

Accounting Office, 1201 University Avenue, Suite 1
-
208, University of
California, Riverside, CA 92521.

Phone Number:

(909) 787
-
7207


Administrative Contact:

Individual from Sponsored Programs Office to contact concerning
administrative matters (
i.e., indirect cost rate computation, rebudgeting requests, etc.)


Name:



Linda L. Bryant

Title



Principal Contract and Grant Analyst

Complete Address:

Office of Research Affairs, 200 University Office Building, University of
California, Riverside, CA 9
2521.

Phone Number:

(909) 787
-
5535

FAX Number:


(909) 78
-
4483

E
-
mail Address:

Linda.Bryant@ucr.edu


Principal Investigator
: Individual responsible for the technical completion of the proposed
work


Name:



Gail Tonnesen

Title:

Assistant Researcher

Complete

Address:

CE
-
CERT 022, University of California, Riverside, CA 92521
-
0434.

Phone Number:

(909) 781
-
5676

FAX Number:


(909) 787
-
5790

E
-
mail/Web Address:

tonnesen@cert.ucr.edu
;
www.cert.ucr.edu





ii

Table of Contents


Abstract

................................
................................
................................
................................
...........
1


1.

Objectives

................................
................................
................................
............................
2

2.

Approach

................................
................................
................................
............................
3


2.1

TSSA Tracer Setup

................................
................................
................................
..
3


2.2

TSSA Science

................................
................................
................................
..........
6


2.3

Evaluation of CMAQ TSSA Algorithm

................................
................................
11


2.4

Quality Assurance Project Plan

................................
................................
..............
12


2.5

Project Tasks

................................
................................
................................
..........
12

3.

Expected Results
or Beneftis

................................
................................
............................
13

4.

General Project Information

................................
................................
...........................
14


4.1

Personnel

................................
................................
................................
................
15


4.2

Project Schedule

................................
................................
................................
.....
16


4.3

Management

................................
................................
................................
...........
16


4.4

Realted Research

................................
................................
................................
....
16


4.5

Interactions with Other Institutions

................................
................................
........
16

5.

Resumes of Key Personnel

................................
................................
..............................
17

6.

Current and Pending Support

................................
................................
........................
23

7.

Budget

................................
................................
................................
...............................
25

8.

Budget Justitification

................................
................................
................................
.......
26

9.

Quality Assurance Statement

................................
................................
..........................
28

10.

References

................................
................................
................................
.........................
29







EPA STAR Grant Abstract


Research Category and Sorting Code:

2004
-
STAR
-
F1, Source Apportionment of Particulate Matter

Title:

Development of a Tagged Species Source A
pportionment Algorithm to
Characterize 3
-
Dimensional Transport and Transformation of Precursors and
Secondary Pollutants

Investigators:

Gail S. Tonnesen

Institution:

University of California, Riverside

Project Period:

October 1, 2004, to September 30,
2006

Project Cost:

$ 260,126


Project Summary:


Objectives/Hypothesis

We will develop new algorithms for tracking the contributions of selected emissions sources to the
formation of fine particulates using a
chemical transport model. We hypotheize that the

new algorithm
will provide results that are consistent with data analysis methods for source apportionment, and that the
method will be useful in modeling studies for identifying important emissions categories and developing
emissions reduction strategies

at attain PM air quality goals.


Approach

We will modify a chemical transport model (CTM) to use reactive tracers to track the mass contribution
of selected emissions source categories and source regions. Efforts proposed here will build on our
previous w
ork to implement a tagged species source apportionment algorithm in the Community
Multiscale Air Quality model. In this study, we will explore several new algorithms for updating the
tracer concentrations at each time step in the CTM. We will also extend t
he current algorithm to include
formation and source attribution of organic aerosols. To test the new algorithms, we will perform model
simulations using CTM scenarios that we have previously developed for annual simulations of fine PM
and regional haze fo
r the continental U.S. In the first year of funding, we will perform test simulations
using a 36
-
km resolution CTM for the continental U.S. Results of the source attributions in the CTM will
be compared with results of ongoing work at the Desert Research I
nstitute in which back trajectory
models and chemical mass balance (CMB) methods are being used to perform source attribution
analyses. In the second year of funding, we will extend these results to higher resolution, nested 12
-
km
model simulations for the

western U.S. and the southeastern U.S. There are important uncertainties in
both CTM results and in data analysis methods such as CMB. However, if these very different methods
provide similar results for source attribution, this would provide evidence to
support the use of both types
of methods. If the source attributions from the CTM and the other data analysis methods are inconsistent,
we will perform additional analysis in which data sets are stratified by meteorology conditions to
determine if the CTM
and data analysis methods agree or disagree under different types of meteorology.


Expected Results

Results of this research will include new algorithms for attributing PM at receptor sites to its sources of
precursor emissions. Post
-
processing programs th
at will be developed to visualize the results of the
source apportionment algorithm. These will include processors to allow the use of VIS5D to visualize 3
-
dimensional transport of the tracers fields and software tools to provide bar plots showing source
c
ontributions to speciated PM and selected receptor sites. Evaluation of the tools will include CMAQ
model performance results at the receptor site by comparison with several ambient monitoring networks,
and comparison of source attributions to results of r
esearch currently being carried out at DRI using data
analysis methods.

Supplemental Keywords:
fine particulate matter; source apportionment; air quality; transport
University of California, Riverside


PM Source Apportionment using CMAQ


2

1. Objectives


Transport of secondary pollutants and their precursors has been an issue o
f serious concern for
over 20 years


the Regional Acid Deposition Model (Chang et al., 1987, 1990) was used to
evaluate transport of pollutants both within the U.S. and between the U.S. and Canada. The
Ozone Transport Assessment Group (OTAG) was a multi
-
y
ear study designed to evaluate the
effects of the transport of ozone and its precursors on ozone nonattainment in the eastern U.S.
There is also continuing concern regarding the transport of pollutants and their precursors among
air basins in Southern Cali
fornia, from California to Nevada, and between Mexico and the U.S.
More recently, there has been increased concern regarding transport of fine particulate matter
(PM) both for health effects and for its contribution to regional haze. Despite the great inte
rest
and considerable past efforts devoted to the transport problem, there is still no definitive method
for quantifying the importance of precursors and pollutants that are transported between air
basins or across political borders.


Previous approaches
to evaluating transport have included statistical analyses, data analysis
methods, back trajectory modeling, and sensitivity studies using chemical transport models
(CTM). While each of these approaches have value, they also have important limitations.
Sta
tistical correlation analyses of the regional ozone pattern (
Guinnup and Collom
, 1997;
Schichtel et al., 1998; Husar and Renard, 1997) do not adequately distinguish the effects of
transport versus meteorological correlation. For example, Husar and Renard (
1997) concluded
that their analysis was “too tentative to warrant conclusions at this time.” Data analysis methods,
such as chemical mass balance (CMB) and positive matrix factorization (PMF), are limited by
the availability of ambient monitoring data and
can only be used to evaluate historical episodes.
Back trajectory analyses cannot adequately treat the complex chemistry of PM formation. CTM
sensitivity studies are computationally expensive when applied to evaluate many individual
sources and, because of

non
-
linearity in chemical transformations, results of sensitivity studies
will vary depending on the magnitude of the emissions change used in the sensitivity simulation.


Our objective here is to develop and test new source apportionment algorithms in a

CTM using
reactive tracers to track the chemical transformation and transport of emissions from selected
emissions source categories or regions. The algorithm will include approximately 20 new tracers
or “tagged species” for each emissions source to be e
valuated. The result of the tagged species
source apportionment (TSSA) algorithm will be 3
-
dimensional concentration fields showing the
transport of mass of both primary emissions and secondary products from the selected emissions
sources. A single CTM sim
ulation will include sufficient tagged species tracers to evaluate the
contributions of as many as 20 different emissions sources. The algorithm will be designed to
track the emissions either of a single source or of selected classes of sources, grouped ei
ther by
source category and/or by region. We will also develop visualization tools to show results of the
source apportionment algorithm as the contribution of each source categories to speciated PM
mass at selected receptor locations, and as 3
-
dimensional

animations of transport from the
selected emissions sources. Initial work on this algorithm has been supported by the Western
Governors’ Association through the Western Regional Air Partnership, and this has demonstrated
the potential usefulness of mass t
racking algorithm for evaluation transport of ammonium nitrate
and sulfates. With the additional funding requested here, we propose to make several
improvements to the algorithm and to complete additional evaluation of the method. In
University of California, Riverside


PM Source Apportionment using CMAQ


3

particular, we will te
st several alternative algorithms for updating tracer concentrations at each
time step in the CTM. We will also extend the algorithm to track the contributions of selected
sources to organic aerosols. We will test the new source apportionment algorithms by

comparing
with results of CTM sensitivity simulations in which we will zero out the selected emissions
sources and then compare the sensitivity results to a base case simulation. We will also compare
the new CTM source apportionment algorithm to results o
f data analysis methods currently being
carried out by Greene and Pitchford (2002) under separate funding from WGA.


We hypothesize that the new TSSA algorithm will provide identical results to CTM sensitivity
simulations in which emissions are removed fo
r non
-
reactive PM species, such as elemental
carbon and coarse crustal materials. Sensitivity simulations using these non
-
reactive species will
be used to verify the TSSA transport algorithms. We also hypothesize that results will be
generally similar to C
TM sensitivity results for secondary PM species, but that there will be
important differences for conditions in which the photochemical transformations are non
-
linear.
In particular, the TSSA results will differ significantly from sensitivity results for p
hotochemical
regimes with low VOC to NOx ratios. Further, we hypothesize that the TSSA algorithm will be
more useful than sensitivity studies for assessing the extent to which mass from a particular
emissions source contributes to PM concentrations at dow
nwind receptor sites. However,
because sensitivity studies are needed to assess PM sensitivity to changes in precursor reductions,
we propose that the combined used of the TSSA algorithm and sensitivity studies will continue
to be needed. We propose that t
he TSSA algorithm can be initially used to identify important
emissions sources, followed by sensitivity simulations to test various emissions reduction
strategies.


The results of this study will be useful to air quality managers and scientists both at re
gulatory
agencies and at the regulated industries for evaluating the contribution and importance of
individual emissions sources. Because the TSSA algorithm specifically tracks mass from
particular sources, it will be useful for evaluating the contribution

of an individual source with
relatively small emissions; that is a source, which cannot be evaluated in “brute force” sensitivity
simulations because the source is small compared to other contributors. The new algorithm and
visualization tools will be in
corporated into CMAQ so that scientists and managers at
universities and state and local air pollution control agencies can continue to use and improve
these tools. This research will benefit the public by assisting in the development of more
effective str
ategies for reducing concentrations of ambient air pollutants, leading to more rapid
and economical attainments of air quality goals for PM and haze.



2. Approach


The Community Multiscale Air Quality (CMAQ) model (EPA, 1999; Novak et al., 1998) will be
used as the platform to develop an algorithm that provides a detailed accounting of the transport
and fate of precursors and secondary pollutants. The algorithm will use a full set of tagged
-
species analogues to a key set of model species for each precurso
r source region to be
represented. Transport of the tagged
-
species will be solved using the model’s advection and
dispersion solvers. The chemical transformations of the tagged
-
species will be updated using the
integrated reaction rates calculated in the c
hemical solver. The effects of chemistry, transport and
University of California, Riverside


PM Source Apportionment using CMAQ


4

deposition on the tagged
-
species analogues will be assumed to be proportional the effects of
these processes on the “bulk” concentration of a model speices within a given grid cell at each
model time
-
step. Thus, the TSSA algorithm will achieve computational efficieny by applying the
bulk species transformations to update the tagged
-
species, thereby avoiding solving the chemistry
individually for each tagged
-
species. Initial work on this approach has be
en funded by WGA and
has demonstrated the feasibility of this approach. Initial testing by comparison to “brute force”
sensitivty simulations showed that the TSSA algorithm gave generally similar results to the
sensitivity results; however, there were sign
ificant differences that can be attributed to the effects
of non
-
linear chemical transformations and numerical noise in the sensitivity simulations. The
approach used in the development and important proposed improvments are described next.



2.1

TSSA Trac
er Setup


The CMAQ Tagged Species Source Apportionment (TSSA) algorithm uses

“tagged
-
species”
tracers to track chemical transformations and the movement and chemical conversion of mass
across domain. New tracer species are added to the model for key model
species to represent
mass contributions to each species for several pre
-
defined source categories and for different
source regions in the modeling domain. A user interface is provided to easily select certain
preconfigured combinations of emissions source
categories and source regions. Figure

1
illustrates the source regions treated in the current version of the code, and Table 1 lists the
combinations of region and source category that can be treated. We propose in this study to
improve the user interface
to allow the user to also select individual point sources and to expand
the number of of source regions and map projections that can be used with the preconfigured
emissions source combinations. The tracers are updated at each time step for the entire mode
l
simulation period and are output at the same hourly interval as regular CMAQ output files. The
tracer output files provide three
-
dimensional fields showing transport and transformation of
secondary species. Currently, the tagged species include nitrate

species such as NO
x

(reactive N
family), HNO
3
, PAN, RNO
3
, ANO
3
I, and ANO
3
J. The NO
x

species is defined as the sum of NO,
NO
2
, NO
3
, 2*N
2
O
5
, HONO, and PAN.


2.1.1

Source Region


In setting up the pre
-
defined region to perform TSSA, a unique number is assig
ned to each
source area. The source region can range from one grid cell to multiple grid cells. Figure 1
illustrates the TSSA regions for each state within the WRAP 36 km modeling domain. In this
figure, California is assigned the number 1, Nevada is as
signed 2, etc. The entire source regions
of Canada and Mexico are assigned 91 and 92, respectively.



University of California, Riverside


PM Source Apportionment using CMAQ


5

91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
11
11
11
11
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
11
11
11
11
11
11
11
11
11
11
15
15
15
15
15
15
15
15
15
15
15
15
15
22
22
22
22
22
22
91
91
91
91
91
91
91
91
91
91
91
0
11
11
11
11
11
11
11
11
11
11
15
15
15
15
15
15
15
15
15
15
15
15
15
15
22
22
22
22
22
22
22
22
22
22
91
91
22
22
91
0
0
11
11
11
11
11
11
11
11
11
11
15
15
15
15
15
15
15
15
15
15
15
15
15
15
22
22
22
22
22
22
22
22
22
22
22
22
22
22
0
0
0
11
11
11
11
11
11
11
11
11
11
15
15
15
15
15
15
15
15
15
15
15
15
15
15
22
22
22
22
22
22
22
22
22
22
22
22
0
0
0
0
0
11
11
11
11
11
11
11
11
11
11
15
15
15
15
15
15
15
15
15
15
15
15
15
15
22
22
22
22
22
22
22
22
22
22
22
22
0
0
0
0
44
11
11
11
11
11
11
11
11
11
11
15
15
15
15
15
15
15
15
15
15
15
15
15
15
22
22
22
22
22
22
22
22
22
22
22
0
33
0
0
44
44
11
11
11
11
11
11
11
11
11
11
15
15
15
15
15
15
15
15
15
15
15
15
15
15
22
22
22
22
22
22
22
22
22
22
33
33
33
33
44
44
44
11
11
11
11
11
11
11
11
11
11
15
15
15
15
15
15
15
15
15
15
15
15
15
15
15
22
22
22
22
22
22
22
22
22
33
33
33
33
33
33
33
11
11
11
11
11
11
11
11
11
11
15
15
15
15
15
15
15
15
15
15
15
15
15
15
15
22
22
22
22
22
22
22
22
22
33
33
33
33
33
33
33
11
11
11
11
11
11
11
11
11
11
26
26
26
26
26
26
26
26
26
26
26
26
26
26
26
22
22
22
22
22
22
22
22
33
33
33
33
33
33
33
33
11
11
11
11
11
11
11
11
11
11
26
26
26
26
26
26
26
26
26
26
26
26
26
26
26
22
22
22
22
22
22
22
22
33
33
33
33
33
33
33
33
10
10
10
10
11
11
11
11
11
11
26
26
26
26
26
26
26
26
26
26
26
26
26
26
26
22
22
22
22
22
22
22
22
33
33
33
33
33
33
33
33
10
10
10
10
10
10
10
10
10
10
26
26
26
26
26
26
26
26
26
26
26
26
26
26
26
26
22
22
22
22
22
22
22
22
33
33
33
33
33
33
33
10
10
10
10
10
10
10
10
10
10
26
26
26
26
26
26
26
26
26
26
26
26
26
26
26
26
22
22
22
22
22
22
22
22
22
33
33
33
33
33
33
10
10
10
10
10
10
10
10
10
10
26
26
26
26
26
26
26
26
26
26
26
26
26
26
26
26
22
22
22
22
22
22
22
22
22
22
22
33
33
33
33
10
10
10
10
10
10
10
10
10
26
26
26
26
26
26
26
26
26
26
26
26
26
26
26
26
26
22
22
22
22
22
22
22
22
22
22
22
33
33
33
33
10
10
10
10
10
10
10
10
10
26
26
26
26
26
26
26
26
26
26
26
26
26
26
26
26
26
17
17
17
17
17
17
17
17
17
17
17
33
33
33
33
10
10
10
10
10
10
10
10
10
26
26
26
26
26
26
26
26
26
26
26
26
26
26
26
26
26
17
17
17
17
17
17
17
17
17
17
17
33
33
33
33
10
10
10
10
10
10
10
10
10
39
39
39
39
39
39
39
39
39
39
39
39
39
26
26
26
26
17
17
17
17
17
17
17
17
17
17
17
17
17
59
59
10
10
10
10
10
10
10
10
10
39
39
39
39
39
39
39
39
39
39
39
39
39
39
39
39
39
17
17
17
17
17
17
17
17
17
17
17
17
17
17
59
10
10
10
10
10
10
10
10
10
39
39
39
39
39
39
39
39
39
39
39
39
39
39
39
39
39
17
17
17
17
17
17
17
17
17
17
17
17
17
17
59
10
10
10
10
10
10
10
10
10
39
39
39
39
39
39
39
39
39
39
39
39
39
39
39
39
39
39
17
17
17
17
17
17
17
17
17
17
17
59
59
59
10
10
10
10
10
10
10
10
10
39
39
39
39
39
39
39
39
39
39
39
39
39
39
39
39
39
39
17
17
17
17
17
17
17
17
17
17
17
59
59
59
9
9
9
9
10
10
10
10
10
39
39
39
39
39
39
39
39
39
39
39
39
39
39
39
39
39
39
17
17
17
17
17
17
17
17
17
17
17
59
59
59
9
9
9
9
9
9
9
9
9
9
9
9
9
39
39
39
39
39
39
39
39
39
39
39
39
39
39
39
17
17
17
17
17
17
17
17
17
17
59
59
59
9
9
9
9
9
9
9
9
9
9
9
9
9
39
39
39
39
39
39
39
39
39
39
39
39
39
39
39
75
75
75
75
75
75
75
75
75
59
59
59
59
9
9
9
9
9
9
9
9
9
9
9
9
9
39
39
39
39
39
39
39
39
39
39
39
39
39
39
39
39
75
75
75
75
75
75
75
75
75
59
59
59
9
9
9
9
9
9
9
9
9
9
9
9
9
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
75
75
75
75
75
75
75
75
75
59
59
9
9
9
9
9
9
9
9
9
9
9
9
9
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
75
75
75
75
75
75
75
75
75
75
59
9
9
9
9
9
9
9
9
9
9
9
9
9
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
75
75
75
75
75
75
75
75
75
75
9
9
9
9
9
9
9
9
9
9
9
9
9
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
75
75
75
75
75
75
75
75
75
75
9
9
9
9
9
9
9
9
9
9
9
9
9
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
75
75
75
75
75
75
75
75
75
75
9
9
9
9
9
9
9
9
9
9
9
9
9
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
75
75
75
75
75
75
75
75
75
75
9
9
9
9
9
9
9
9
9
9
9
9
9
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
75
75
75
75
75
75
75
75
75
75
8
8
8
8
8
9
9
9
9
9
9
9
9
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
75
75
75
75
75
75
75
75
75
75
8
8
8
8
8
8
8
8
8
8
8
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
75
75
75
75
75
80
80
80
80
80
8
8
8
8
8
8
8
8
8
8
8
60
60
60
60
60
60
60
20
20
20
20
20
20
20
20
20
20
20
20
20
80
80
80
80
80
80
80
80
80
80
8
8
8
8
8
8
8
8
8
8
8
60
60
60
60
60
60
60
20
20
20
20
20
20
20
20
20
20
20
20
20
20
80
80
80
80
80
80
80
80
80
8
8
8
8
8
8
8
8
8
8
8
60
60
60
60
60
60
60
20
20
20
20
20
20
20
20
20
20
20
20
20
20
80
80
80
80
80
80
80
80
80
8
8
8
8
8
8
8
8
8
8
8
60
60
60
60
60
60
60
20
20
20
20
20
20
20
20
20
20
20
20
20
20
80
80
80
80
80
80
80
80
80
8
8
8
8
8
8
8
8
8
8
8
60
60
60
60
60
60
60
20
20
20
20
20
20
20
20
20
20
20
20
20
20
80
80
80
80
80
80
80
80
80
8
8
8
8
8
8
8
8
8
8
8
60
60
60
60
60
60
60
20
20
20
20
20
20
20
20
20
20
20
20
20
20
80
80
80
80
80
80
80
80
80
8
8
8
8
8
8
8
8
8
8
60
60
60
60
60
60
60
60
20
20
20
20
20
20
20
20
20
20
20
20
20
20
80
80
80
80
80
80
80
80
80
8
8
8
8
8
8
8
8
8
8
60
60
60
60
60
60
60
60
60
60
60
60
20
20
20
20
20
20
20
20
20
20
80
80
80
80
80
80
80
80
80
8
8
8
8
8
8
8
8
8
8
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
80
80
80
80
80
80
80
80
80
8
8
8
8
8
8
8
8
8
8
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
80
80
80
80
80
80
80
80
8
8
8
8
8
8
8
8
8
8
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
70
70
70
70
70
70
70
70
8
8
8
8
8
8
8
8
8
8
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
70
70
70
70
70
70
70
70
8
8
8
8
8
8
8
8
8
8
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
70
70
70
70
70
70
70
70
8
60
60
60
8
8
8
8
8
8
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
70
70
70
70
70
70
70
99
92
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
70
70
70
70
70
70
99
92
92
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
70
70
70
70
70
70
70
92
92
92
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
70
70
70
70
70
70
92
92
92
92
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
70
70
70
70
70
70
92
92
92
92
92
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
70
70
70
70
70
70
92
92
92
92
92
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
70
70
70
70
70
70
70
92
92
92
92
92
92
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
70
0
70
70
70
70
0
92
92
92
92
92
92
60
60
60
60
92
92
92
92
92
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
0
0
0
0
0
0
0
0
92
92
92
92
92
92
92
92
60
60
92
92
92
92
92
92
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
0
0
0
0
0
0
0
0
0
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
60
60
60
60
60
60
60
60
60
60
60
60
60
60
0
0
0
0
0
0
0
0
0
0
0
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
60
60
60
60
60
60
60
60
60
60
60
60
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
60
60
60
60
60
60
60
60
60
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
60
60
60
60
60
60
60
60
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
60
60
60
60
60
60
60
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
60
60
60
60
60
60
60
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
60
60
60
60
60
60
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
60
60
60
60
60
60
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
60
60
60
60
60
60
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
60
60
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
0
0
0
0
0
0
0
0
0
0
0
0
44
0
0
44
44
0
44
44
44
44
44
44
33
44
44
33
33
44
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
59
59
59
59
59
59
59
59
59
59
59
59
59
59
59
59
59
59
59
59
59
59
59
59
59
59
59
59
59
59
59
59
59
59
59
59
5 9
5 9
5 9
5 9
5 9
5 9
5 9
5 9
5 9
5 9
5 9
5 9
5 9
5 9
5 9
7 5
5 9
5 9
7 5
7 5
5 9
7 5
7 5
5 9
7 5
7 5
7 5
7 5
7 5
7 5
8 0
7 5
6 5
8 0
7 5
6 5
8 0
8 0
6 5
8 0
6 5
6 5
8 0
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 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
9 9
9 9
9 9
9 9
9 9
9 9
9 9
9 9
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
0
7 0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
0
0
0
0
0
0
0
0
0
0
0
0
0
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
0
0
0
0
0
0
0
0
0
0
0
0
0
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
0
0
0
0
0
0
0
0
0
0
0
0
0
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
0
0
0
0
0
0
0
0
0
0
0
0
0
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
0
0
0
0
0
0
0
0
0
0
0
0
0
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
0
0
0
0
0
0
0
0
0
0
0
0
0
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
0
0
0
0
0
0
0
0
0
0
0
0
0
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
0
0
0
0
0
0
0
0
0
0
0
0
0
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
0
0
0
0
0
0
0
0
0
0
0
0
0
0
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
0
0
0
0
0
0
0
0
0
0
0
0
0
0
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
4
4
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
0
0
0
0
0
0
0
0
0
0
4
4
0
0
0
4
4
4
4
4
4
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
0
0
0
0
0
0
0
0
0
0
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
91
91
91
91
91
91
91
91
91
91
91
91
91
91
0
0
0
0
0
0
0
0
0
0
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
5
5
11
11
11
11
11
91
91
91
91
91
91
91
0
0
0
0
0
0
0
0
0
0
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
5
5
11
11
11
11
11
11
11
11
11
11
91
91
91
0
0
0
0
0
0
0
0
0
0
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
5
5
11
11
11
11
11
11
11
11
11
11
11
11
11
0
0
0
0
0
0
0
0
0
0
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
5
5
11
11
11
11
11
11
11
11
11
11
11
11
11
0
0
0
0
0
0
0
0
0
0
3
4
4
4
4
4
4
4
4
4
4
4
4
4
4
5
5
11
11
11
11
11
11
11
11
11
11
11
11
11
0
0
0
0
0
0
0
0
0
3
3
3
4
4
4
4
4
4
4
4
4
4
4
4
5
5
5
5
11
11
11
11
11
11
11
11
11
11
11
11
0
0
0
0
0
0
0
0
0
3
3
3
4
4
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
11
11
11
11
11
11
11
11
11
11
11
0
0
0
0
0
0
0
0
0
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
5
5
5
5
5
11
11
11
11
11
11
11
11
11
11
11
0
0
0
0
0
0
0
0
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
5
5
5
5
5
11
11
11
11
11
11
11
11
11
11
11
0
0
0
0
0
0
0
0
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
5
5
5
5
11
11
11
11
11
11
11
11
11
11
11
0
0
0
0
0
0
0
0
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
5
5
5
5
5
11
11
11
11
11
11
11
11
11
11
11
0
0
0
0
0
0
0
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
5
5
5
5
5
5
11
11
11
11
11
11
11
11
11
11
0
0
0
0
0
0
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
5
5
5
5
5
5
5
11
11
11
11
11
11
11
11
11
11
0
0
0
0
0
0
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
5
5
5
5
5
5
5
5
11
11
11
11
11
11
11
11
11
0
0
0
0
0
0
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
5
5
5
5
5
5
5
5
5
5
5
5
10
10
10
10
10
0
0
0
0
0
0
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
5
5
5
5
5
5
5
5
5
5
5
5
5
10
10
10
10
10
0
0
0
0
0
0
1
1
3
3
3
3
3
3
3
3
3
3
3
3
3
3
5
5
5
5
5
5
5
5
5
5
5
5
5
10
10
10
10
10
0
0
0
0
0
0
1
1
1
1
1
3
3
3
3
3
3
3
3
3
3
3
5
5
5
5
5
5
5
5
5
5
5
5
5
10
10
10
10
10
0
0
0
0
0
1
1
1
1
1
1
1
1
1
1
3
3
3
3
3
3
3
5
5
5
5
5
5
5
5
5
5
5
5
5
10
10
10
10
10
0
0
0
0
0
1
1
1
1
1
1
1
1
1
1
2
2
2
2
3
3
3
5
5
5
5
5
5
5
5
5
5
5
5
5
10
10
10
10
10
0
0
0
0
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
5
5
5
5
5
5
5
5
5
5
5
10
10
10
10
10
0
0
0
0
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
2
2
2
2
6
5
5
5
5
5
5
10
10
10
10
10
0
0
0
0
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
2
2
2
2
6
6
6
6
6
6
6
10
10
10
10
10
0
0
0
0
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
2
2
2
2
6
6
6
6
6
6
10
10
10
10
10
10
0
0
0
0
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
2
2
2
2
6
6
6
6
6
6
6
10
10
10
10
10
10
0
0
0
0
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
2
2
2
2
6
6
6
6
6
6
6
6
10
10
10
10
10
0
0
0
0
0
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
2
2
2
2
6
6
6
6
6
6
6
6
6
6
6
6
9
0
0
0
0
0
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
2
2
2
2
6
6
6
6
6
6
6
6
6
6
6
6
9
0
0
0
0
0
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
2
2
2
6
6
6
6
6
6
6
6
6
6
6
9
9
0
0
0
0
0
0
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
2
2
6
6
6
6
6
6
6
6
6
6
6
6
9
9
0
0
0
0
0
0
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
2
6
6
6
6
6
6
6
6
6
6
6
6
9
9
0
0
0
0
0
0
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
6
6
6
6
6
6
6
6
6
6
6
6
9
9
0
0
0
0
0
0
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
6
6
6
6
6
6
6
6
6
6
6
6
9
9
0
0
0
0
0
0
1
1
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
6
6
6
6
6
6
6
6
6
6
6
6
6
9
9
0
0
0
0
0
0
0
1
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
6
6
6
6
6
6
6
6
6
6
6
6
9
9
9
0
0
0
0
0
0
0
1
1
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
7
7
7
6
6
6
6
6
6
6
6
6
9
9
9
0
0
0
0
0
0
0
0
1
1
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
7
7
7
7
7
7
7
7
7
7
6
6
9
9
9
0
0
0
0
0
0
0
0
1
1
1
1
1
1
1
1
1
1
1
1
1
2
2
7
7
7
7
7
7
7
7
7
7
7
7
7
7
8
8
8
0
0
0
0
0
0
0
0
1
1
1
1
1
1
1
1
1
1
1
1
1
2
2
7
7
7
7
7
7
7
7
7
7
7
7
7
7
8
8
8
0
0
0
0
0
0
0
0
0
1
1
1
1
1
1
1
1
1
1
1
1
1
2
7
7
7
7
7
7
7
7
7
7
7
7
7
7
8
8
8
0
0
0
0
0
0
0
0
0
0
0
1
1
1
1
1
1
1
1
1
1
1
1
7
7
7
7
7
7
7
7
7
7
7
7
7
7
8
8
8
0
0
0
0
0
0
0
0
0
0
0
1
1
1
1
1
1
1
1
1
1
1
1
7
7
7
7
7
7
7
7
7
7
7
7
7
7
8
8
8
0
0
0
0
0
0
0
0
0
0
0
0
0
1
1
1
1
1
1
1
1
1
1
7
7
7
7
7
7
7
7
7
7
7
7
7
7
8
8
8
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
1
1
1
1
1
1
1
1
7
7
7
7
7
7
7
7
7
7
7
7
7
7
8
8
8
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
1
1
1
1
1
1
7
7
7
7
7
7
7
7
7
7
7
7
7
7
8
8
8
8
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
1
1
1
1
1
1
7
7
7
7
7
7
7
7
7
7
7
7
7
7
8
8
8
8
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
1
1
1
1
1
1
7
7
7
7
7
7
7
7
7
7
7
7
7
7
8
8
8
8
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
92
92
92
1
7
7
7
7
7
7
7
7
7
7
7
7
7
7
8
8
8
8
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
92
92
92
7
7
7
7
7
7
7
7
7
7
7
7
7
7
8
8
8
8
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
92
92
92
92
7
7
7
7
7
7
7
7
7
7
7
7
7
8
8
8
8
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
92
92
92
92
92
92
7
7
7
7
7
7
7
7
7
7
7
8
8
8
8
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
92
92
0
92
92
92
92
92
7
7
7
7
7
7
7
7
7
8
8
8
8
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
92
0
0
0
92
92
92
92
92
7
7
7
7
7
7
8
8
8
8
8
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
92
0
0
0
0
92
92
92
92
92
7
7
7
7
7
8
8
8
92
92
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
92
0
0
0
0
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
92
0
0
0
0
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
0
0
0
0
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
0
0
0
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
0
0
0
0
92
92
92
92
92
92
92
92
92
92
92
92
92
92
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
0
0
0
92
92
92
92
92
92
92
92
92
92
92
92
92
92
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
0
92
0
92
92
92
92
92
92
92
92
92
92
92
92
92
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
0
0
0
92
92
92
0
0
0
92
92
92
92
92
92
92
92
92
92
92
92
92
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
92
92
92
0
0
0
0
92
92
92
92
92
92
92
92
92
92
92
92
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
92
92
0
0
0
0
92
92
92
92
92
92
92
92
92
92
92
92
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
92
92
0
0
0
0
0
92
92
92
92
92
92
92
92
92
92
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
92
0
0
0
0
0
92
92
92
92
92
92
92
92
92
92
92
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
92
0
0
0
0
0
92
92
92
92
92
92
92
92
92
92
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
92
0
0
0
0
0
92
92
92
92
92
92
92
92
92
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
0
0
0
0
0
0
92
92
92
92
92
92
92
92
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
0
0
0
0
0
0
92
92
92
92
92
92
92
92
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
0
0
0
0
0
92
92
92
92
92
92
92
92
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
0
0
0
0
0
92
92
92
92
92
92
92
92
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
92
0
0
0
0
0
0
92
92
92
92
92
92
92
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
0
0
0
0
0
0
0
0
92
92
92
92
92
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
0
0
0
0
0
0
0
0
0
92
92
92
92
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
0
0
0
0
0
0
0
92
92
92
92
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
0
0
0
0
0
0
0
92
92
92
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
11
11
11
11
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
11
11
11
11
11
11
11
11
11
11
15
15
15
15
15
15
15
15
15
15
15
15
15
22
22
22
22
22
22
91
91
91
91
91
91
91
91
91
91
91
0
11
11
11
11
11
11
11
11
11
11
15
15
15
15
15
15
15
15
15
15
15
15
15
15
22
22
22
22
22
22
22
22
22
22
91
91
22
22
91
0
0
11
11
11
11
11
11
11
11
11
11
15
15
15
15
15
15
15
15
15
15
15
15
15
15
22
22
22
22
22
22
22
22
22
22
22
22
22
22
0
0
0
11
11
11
11
11
11
11
11
11
11
15
15
15
15
15
15
15
15
15
15
15
15
15
15
22
22
22
22
22
22
22
22
22
22
22
22
0
0
0
0
0
11
11
11
11
11
11
11
11
11
11
15
15
15
15
15
15
15
15
15
15
15
15
15
15
22
22
22
22
22
22
22
22
22
22
22
22
0
0
0
0
44
11
11
11
11
11
11
11
11
11
11
15
15
15
15
15
15
15
15
15
15
15
15
15
15
22
22
22
22
22
22
22
22
22
22
22
0
33
0
0
44
44
11
11
11
11
11
11
11
11
11
11
15
15
15
15
15
15
15
15
15
15
15
15
15
15
22
22
22
22
22
22
22
22
22
22
33
33
33
33
44
44
44
11
11
11
11
11
11
11
11
11
11
15
15
15
15
15
15
15
15
15
15
15
15
15
15
15
22
22
22
22
22
22
22
22
22
33
33
33
33
33
33
33
11
11
11
11
11
11
11
11
11
11
15
15
15
15
15
15
15
15
15
15
15
15
15
15
15
22
22
22
22
22
22
22
22
22
33
33
33
33
33
33
33
11
11
11
11
11
11
11
11
11
11
26
26
26
26
26
26
26
26
26
26
26
26
26
26
26
22
22
22
22
22
22
22
22
33
33
33
33
33
33
33
33
11
11
11
11
11
11
11
11
11
11
26
26
26
26
26
26
26
26
26
26
26
26
26
26
26
22
22
22
22
22
22
22
22
33
33
33
33
33
33
33
33
10
10
10
10
11
11
11
11
11
11
26
26
26
26
26
26
26
26
26
26
26
26
26
26
26
22
22
22
22
22
22
22
22
33
33
33
33
33
33
33
33
10
10
10
10
10
10
10
10
10
10
26
26
26
26
26
26
26
26
26
26
26
26
26
26
26
26
22
22
22
22
22
22
22
22
33
33
33
33
33
33
33
10
10
10
10
10
10
10
10
10
10
26
26
26
26
26
26
26
26
26
26
26
26
26
26
26
26
22
22
22
22
22
22
22
22
22
33
33
33
33
33
33
10
10
10
10
10
10
10
10
10
10
26
26
26
26
26
26
26
26
26
26
26
26
26
26
26
26
22
22
22
22
22
22
22
22
22
22
22
33
33
33
33
10
10
10
10
10
10
10
10
10
26
26
26
26
26
26
26
26
26
26
26
26
26
26
26
26
26
22
22
22
22
22
22
22
22
22
22
22
33
33
33
33
10
10
10
10
10
10
10
10
10
26
26
26
26
26
26
26
26
26
26
26
26
26
26
26
26
26
17
17
17
17
17
17
17
17
17
17
17
33
33
33
33
10
10
10
10
10
10
10
10
10
26
26
26
26
26
26
26
26
26
26
26
26
26
26
26
26
26
17
17
17
17
17
17
17
17
17
17
17
33
33
33
33
10
10
10
10
10
10
10
10
10
39
39
39
39
39
39
39
39
39
39
39
39
39
26
26
26
26
17
17
17
17
17
17
17
17
17
17
17
17
17
59
59
10
10
10
10
10
10
10
10
10
39
39
39
39
39
39
39
39
39
39
39
39
39
39
39
39
39
17
17
17
17
17
17
17
17
17
17
17
17
17
17
59
10
10
10
10
10
10
10
10
10
39
39
39
39
39
39
39
39
39
39
39
39
39
39
39
39
39
17
17
17
17
17
17
17
17
17
17
17
17
17
17
59
10
10
10
10
10
10
10
10
10
39
39
39
39
39
39
39
39
39
39
39
39
39
39
39
39
39
39
17
17
17
17
17
17
17
17
17
17
17
59
59
59
10
10
10
10
10
10
10
10
10
39
39
39
39
39
39
39
39
39
39
39
39
39
39
39
39
39
39
17
17
17
17
17
17
17
17
17
17
17
59
59
59
9
9
9
9
10
10
10
10
10
39
39
39
39
39
39
39
39
39
39
39
39
39
39
39
39
39
39
17
17
17
17
17
17
17
17
17
17
17
59
59
59
9
9
9
9
9
9
9
9
9
9
9
9
9
39
39
39
39
39
39
39
39
39
39
39
39
39
39
39
17
17
17
17
17
17
17
17
17
17
59
59
59
9
9
9
9
9
9
9
9
9
9
9
9
9
39
39
39
39
39
39
39
39
39
39
39
39
39
39
39
75
75
75
75
75
75
75
75
75
59
59
59
59
9
9
9
9
9
9
9
9
9
9
9
9
9
39
39
39
39
39
39
39
39
39
39
39
39
39
39
39
39
75
75
75
75
75
75
75
75
75
59
59
59
9
9
9
9
9
9
9
9
9
9
9
9
9
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
75
75
75
75
75
75
75
75
75
59
59
9
9
9
9
9
9
9
9
9
9
9
9
9
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
75
75
75
75
75
75
75
75
75
75
59
9
9
9
9
9
9
9
9
9
9
9
9
9
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
75
75
75
75
75
75
75
75
75
75
9
9
9
9
9
9
9
9
9
9
9
9
9
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
75
75
75
75
75
75
75
75
75
75
9
9
9
9
9
9
9
9
9
9
9
9
9
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
75
75
75
75
75
75
75
75
75
75
9
9
9
9
9
9
9
9
9
9
9
9
9
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
75
75
75
75
75
75
75
75
75
75
9
9
9
9
9
9
9
9
9
9
9
9
9
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
75
75
75
75
75
75
75
75
75
75
8
8
8
8
8
9
9
9
9
9
9
9
9
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
75
75
75
75
75
75
75
75
75
75
8
8
8
8
8
8
8
8
8
8
8
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
75
75
75
75
75
80
80
80
80
80
8
8
8
8
8
8
8
8
8
8
8
60
60
60
60
60
60
60
20
20
20
20
20
20
20
20
20
20
20
20
20
80
80
80
80
80
80
80
80
80
80
8
8
8
8
8
8
8
8
8
8
8
60
60
60
60
60
60
60
20
20
20
20
20
20
20
20
20
20
20
20
20
20
80
80
80
80
80
80
80
80
80
8
8
8
8
8
8
8
8
8
8
8
60
60
60
60
60
60
60
20
20
20
20
20
20
20
20
20
20
20
20
20
20
80
80
80
80
80
80
80
80
80
8
8
8
8
8
8
8
8
8
8
8
60
60
60
60
60
60
60
20
20
20
20
20
20
20
20
20
20
20
20
20
20
80
80
80
80
80
80
80
80
80
8
8
8
8
8
8
8
8
8
8
8
60
60
60
60
60
60
60
20
20
20
20
20
20
20
20
20
20
20
20
20
20
80
80
80
80
80
80
80
80
80
8
8
8
8
8
8
8
8
8
8
8
60
60
60
60
60
60
60
20
20
20
20
20
20
20
20
20
20
20
20
20
20
80
80
80
80
80
80
80
80
80
8
8
8
8
8
8
8
8
8
8
60
60
60
60
60
60
60
60
20
20
20
20
20
20
20
20
20
20
20
20
20
20
80
80
80
80
80
80
80
80
80
8
8
8
8
8
8
8
8
8
8
60
60
60
60
60
60
60
60
60
60
60
60
20
20
20
20
20
20
20
20
20
20
80
80
80
80
80
80
80
80
80
8
8
8
8
8
8
8
8
8
8
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
80
80
80
80
80
80
80
80
80
8
8
8
8
8
8
8
8
8
8
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
80
80
80
80
80
80
80
80
8
8
8
8
8
8
8
8
8
8
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
70
70
70
70
70
70
70
70
8
8
8
8
8
8
8
8
8
8
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
70
70
70
70
70
70
70
70
8
8
8
8
8
8
8
8
8
8
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
70
70
70
70
70
70
70
70
8
60
60
60
8
8
8
8
8
8
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
70
70
70
70
70
70
70
99
92
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
70
70
70
70
70
70
99
92
92
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
70
70
70
70
70
70
70
92
92
92
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
70
70
70
70
70
70
92
92
92
92
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
70
70
70
70
70
70
92
92
92
92
92
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
70
70
70
70
70
70
92
92
92
92
92
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
70
70
70
70
70
70
70
92
92
92
92
92
92
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
70
0
70
70
70
70
0
92
92
92
92
92
92
60
60
60
60
92
92
92
92
92
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
0
0
0
0
0
0
0
0
92
92
92
92
92
92
92
92
60
60
92
92
92
92
92
92
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
0
0
0
0
0
0
0
0
0
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
60
60
60
60
60
60
60
60
60
60
60
60
60
60
0
0
0
0
0
0
0
0
0
0
0
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
60
60
60
60
60
60
60
60
60
60
60
60
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
60
60
60
60
60
60
60
60
60
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
60
60
60
60
60
60
60
60
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
60
60
60
60
60
60
60
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
60
60
60
60
60
60
60
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
60
60
60
60
60
60
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
60
60
60
60
60
60
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
60
60
60
60
60
60
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
60
60
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
0
0
0
0
0
0
0
0
0
0
0
0
44
0
0
44
44
0
44
44
44
44
44
44
33
44
44
33
33
44
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
59
59
59
59
59
59
59
59
59
59
59
59
59
59
59
59
59
59
59
59
59
59
59
59
59
59
59
59
59
59
59
59
59
59
59
59
5 9
5 9
5 9
5 9
5 9
5 9
5 9
5 9
5 9
5 9
5 9
5 9
5 9
5 9
5 9
7 5
5 9
5 9
7 5
7 5
5 9
7 5
7 5
5 9
7 5
7 5
7 5
7 5
7 5
7 5
8 0
7 5
6 5
8 0
7 5
6 5
8 0
8 0
6 5
8 0
6 5
6 5
8 0
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 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
9 9
9 9
9 9
9 9
9 9
9 9
9 9
9 9
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
0
7 0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
0
0
0
0
0
0
0
0
0
0
0
0
0
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
0
0
0
0
0
0
0
0
0
0
0
0
0
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
0
0
0
0
0
0
0
0
0
0
0
0
0
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
0
0
0
0
0
0
0
0
0
0
0
0
0
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
0
0
0
0
0
0
0
0
0
0
0
0
0
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
0
0
0
0
0
0
0
0
0
0
0
0
0
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
0
0
0
0
0
0
0
0
0
0
0
0
0
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
0
0
0
0
0
0
0
0
0
0
0
0
0
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
0
0
0
0
0
0
0
0
0
0
0
0
0
0
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
0
0
0
0
0
0
0
0
0
0
0
0
0
0
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
4
4
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
0
0
0
0
0
0
0
0
0
0
4
4
0
0
0
4
4
4
4
4
4
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
91
0
0
0
0
0
0
0
0
0
0
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
91
91
91
91
91
91
91
91
91
91
91
91
91
91
0
0
0
0
0
0
0
0
0
0
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
5
5
11
11
11
11
11
91
91
91
91
91
91
91
0
0
0
0
0
0
0
0
0
0
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
5
5
11
11
11
11
11
11
11
11
11
11
91
91
91
0
0
0
0
0
0
0
0
0
0
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
5
5
11
11
11
11
11
11
11
11
11
11
11
11
11
0
0
0
0
0
0
0
0
0
0
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
5
5
11
11
11
11
11
11
11
11
11
11
11
11
11
0
0
0
0
0
0
0
0
0
0
3
4
4
4
4
4
4
4
4
4
4
4
4
4
4
5
5
11
11
11
11
11
11
11
11
11
11
11
11
11
0
0
0
0
0
0
0
0
0
3
3
3
4
4
4
4
4
4
4
4
4
4
4
4
5
5
5
5
11
11
11
11
11
11
11
11
11
11
11
11
0
0
0
0
0
0
0
0
0
3
3
3
4
4
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
11
11
11
11
11
11
11
11
11
11
11
0
0
0
0
0
0
0
0
0
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
5
5
5
5
5
11
11
11
11
11
11
11
11
11
11
11
0
0
0
0
0
0
0
0
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
5
5
5
5
5
11
11
11
11
11
11
11
11
11
11
11
0
0
0
0
0
0
0
0
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
5
5
5
5
11
11
11
11
11
11
11
11
11
11
11
0
0
0
0
0
0
0
0
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
5
5
5
5
5
11
11
11
11
11
11
11
11
11
11
11
0
0
0
0
0
0
0
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
5
5
5
5
5
5
11
11
11
11
11
11
11
11
11
11
0
0
0
0
0
0
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
5
5
5
5
5
5
5
11
11
11
11
11
11
11
11
11
11
0
0
0
0
0
0
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
5
5
5
5
5
5
5
5
11
11
11
11
11
11
11
11
11
0
0
0
0
0
0
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
5
5
5
5
5
5
5
5
5
5
5
5
10
10
10
10
10
0
0
0
0
0
0
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
5
5
5
5
5
5
5
5
5
5
5
5
5
10
10
10
10
10
0
0
0
0
0
0
1
1
3
3
3
3
3
3
3
3
3
3
3
3
3
3
5
5
5
5
5
5
5
5
5
5
5
5
5
10
10
10
10
10
0
0
0
0
0
0
1
1
1
1
1
3
3
3
3
3
3
3
3
3
3
3
5
5
5
5
5
5
5
5
5
5
5
5
5
10
10
10
10
10
0
0
0
0
0
1
1
1
1
1
1
1
1
1
1
3
3
3
3
3
3
3
5
5
5
5
5
5
5
5
5
5
5
5
5
10
10
10
10
10
0
0
0
0
0
1
1
1
1
1
1
1
1
1
1
2
2
2
2
3
3
3
5
5
5
5
5
5
5
5
5
5
5
5
5
10
10
10
10
10
0
0
0
0
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
5
5
5
5
5
5
5
5
5
5
5
10
10
10
10
10
0
0
0
0
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
2
2
2
2
6
5
5
5
5
5
5
10
10
10
10
10
0
0
0
0
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
2
2
2
2
6
6
6
6
6
6
6
10
10
10
10
10
0
0
0
0
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
2
2
2
2
6
6
6
6
6
6
10
10
10
10
10
10
0
0
0
0
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
2
2
2
2
6
6
6
6
6
6
6
10
10
10
10
10
10
0
0
0
0
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
2
2
2
2
6
6
6
6
6
6
6
6
10
10
10
10
10
0
0
0
0
0
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
2
2
2
2
6
6
6
6
6
6
6
6
6
6
6
6
9
0
0
0
0
0
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
2
2
2
2
6
6
6
6
6
6
6
6
6
6
6
6
9
0
0
0
0
0
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
2
2
2
6
6
6
6
6
6
6
6
6
6
6
9
9
0
0
0
0
0
0
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
2
2
6
6
6
6
6
6
6
6
6
6
6
6
9
9
0
0
0
0
0
0
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
2
6
6
6
6
6
6
6
6
6
6
6
6
9
9
0
0
0
0
0
0
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
6
6
6
6
6
6
6
6
6
6
6
6
9
9
0
0
0
0
0
0
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
6
6
6
6
6
6
6
6
6
6
6
6
9
9
0
0
0
0
0
0
1
1
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
6
6
6
6
6
6
6
6
6
6
6
6
6
9
9
0
0
0
0
0
0
0
1
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
6
6
6
6
6
6
6
6
6
6
6
6
9
9
9
0
0
0
0
0
0
0
1
1
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
7
7
7
6
6
6
6
6
6
6
6
6
9
9
9
0
0
0
0
0
0
0
0
1
1
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
7
7
7
7
7
7
7
7
7
7
6
6
9
9
9
0
0
0
0
0
0
0
0
1
1
1
1
1
1
1
1
1
1
1
1
1
2
2
7
7
7
7
7
7
7
7
7
7
7
7
7
7
8
8
8
0
0
0
0
0
0
0
0
1
1
1
1
1
1
1
1
1
1
1
1
1
2
2
7
7
7
7
7
7
7
7
7
7
7
7
7
7
8
8
8
0
0
0
0
0
0
0
0
0
1
1
1
1
1
1
1
1
1
1
1
1
1
2
7
7
7
7
7
7
7
7
7
7
7
7
7
7
8
8
8
0
0
0
0
0
0
0
0
0
0
0
1
1
1
1
1
1
1
1
1
1
1
1
7
7
7
7
7
7
7
7
7
7
7
7
7
7
8
8
8
0
0
0
0
0
0
0
0
0
0
0
1
1
1
1
1
1
1
1
1
1
1
1
7
7
7
7
7
7
7
7
7
7
7
7
7
7
8
8
8
0
0
0
0
0
0
0
0
0
0
0
0
0
1
1
1
1
1
1
1
1
1
1
7
7
7
7
7
7
7
7
7
7
7
7
7
7
8
8
8
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
1
1
1
1
1
1
1
1
7
7
7
7
7
7
7
7
7
7
7
7
7
7
8
8
8
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
1
1
1
1
1
1
7
7
7
7
7
7
7
7
7
7
7
7
7
7
8
8
8
8
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
1
1
1
1
1
1
7
7
7
7
7
7
7
7
7
7
7
7
7
7
8
8
8
8
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
1
1
1
1
1
1
7
7
7
7
7
7
7
7
7
7
7
7
7
7
8
8
8
8
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
92
92
92
1
7
7
7
7
7
7
7
7
7
7
7
7
7
7
8
8
8
8
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
92
92
92
7
7
7
7
7
7
7
7
7
7
7
7
7
7
8
8
8
8
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
92
92
92
92
7
7
7
7
7
7
7
7
7
7
7
7
7
8
8
8
8
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
92
92
92
92
92
92
7
7
7
7
7
7
7
7
7
7
7
8
8
8
8
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
92
92
0
92
92
92
92
92
7
7
7
7
7
7
7
7
7
8
8
8
8
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
92
0
0
0
92
92
92
92
92
7
7
7
7
7
7
8
8
8
8
8
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
92
0
0
0
0
92
92
92
92
92
7
7
7
7
7
8
8
8
92
92
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
92
0
0
0
0
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
92
0
0
0
0
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
0
0
0
0
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
0
0
0
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
0
0
0
0
92
92
92
92
92
92
92
92
92
92
92
92
92
92
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
0
0
0
92
92
92
92
92
92
92
92
92
92
92
92
92
92
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
0
92
0
92
92
92
92
92
92
92
92
92
92
92
92
92
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
0
0
0
92
92
92
0
0
0
92
92
92
92
92
92
92
92
92
92
92
92
92
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
92
92
92
0
0
0
0
92
92
92
92
92
92
92
92
92
92
92
92
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
92
92
0
0
0
0
92
92
92
92
92
92
92
92
92
92
92
92
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
92
92
0
0
0
0
0
92
92
92
92
92
92
92
92
92
92
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
92
0
0
0
0
0
92
92
92
92
92
92
92
92
92
92
92
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
92
0
0
0
0
0
92
92
92
92
92
92
92
92
92
92
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
92
0
0
0
0
0
92
92
92
92
92
92
92
92
92
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
0
0
0
0
0
0
92
92
92
92
92
92
92
92
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
0
0
0
0
0
0
92
92
92
92
92
92
92
92
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
0
0
0
0
0
92
92
92
92
92
92
92
92
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
0
0
0
0
0
92
92
92
92
92
92
92
92
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
92
0
0
0
0
0
0
92
92
92
92
92
92
92
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
0
0
0
0
0
0
0
0
92
92
92
92
92
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
0
0
0
0
0
0
0
0
0
92
92
92
92
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
0
0
0
0
0
0
0
92
92
92
92
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
92
92
92
0
0
0
0
0
0
0
92
92
92


Figure 1

Source Area mapping file: the purpose of this file is to pre
-
define source regions
of interest in the modeling domain for TSSA. In this exam
ple, each source
region (state) is distinguished by a unique number.



2.1.2

Source categories


The source categories include initial and boundary conditions, individual types of emissions
sources such as motor vehicles, point sources, area sources, fires,

etc., and “other” sources which
include everything other than all of the above explicit sources. The emissions files are processed
separately to account for each source category. Table 1 illustrates the types of sources that can
currently be included in
TSSA.


By combining the source region and source categories, it is possible to assess pollutant exposure
at a particular receptor site from individual emissions source categories in a specific region. For
example, Figure 2 shows the formation and transpor
t of aerosol nitrate from California mobile
vehicle (MV) emissions during a period with predominatly northerly winds, with nitrate being
transported over the Pacific ocean.


University of California, Riverside


PM Source Apportionment using CMAQ


6

Table 1. Emissions categories that can be tracked with tagged
-
species tracers.




Figure 2

Three
-
dimensional plot of aerosol nitrate, shown in yellow, from CA mobile
vehicle emissions, visualized using VIS5D.

Any type of emission of WRAP domain
*_WRAP
Notes
Source Category
Types
Any sources other than all of the above
OTHERS
Others
Total Emission of any state
ET_*
Mexico Fire
MX_*
RX Fire of any state
RX_*
AG Fire of any state
AG_*
WF Fire of any state
WF_*
Area sources of any state
AR_*
Point sources with SO2 of any state
PS_*
Point sources without SO2 of any state
PN_*
Non
-
Road dust of any state
NR_*
Road dust of any state
RD_*
Biogenic sources of any state
BG_*
Mobile sources of any state
MV_*
Emissions
Boundary Concentration
BCON
BCON
Initial Concentration
ICON
ICON
Any type of emission of WRAP domain
*_WRAP
Notes
Source Category
Types
Any sources other than all of the above
OTHERS
Others
Total Emission of any state
ET_*
Mexico Fire
MX_*
RX Fire of any state
RX_*
AG Fire of any state
AG_*
WF Fire of any state
WF_*
Area sources of any state
AR_*
Point sources with SO2 of any state
PS_*
Point sources without SO2 of any state
PN_*
Non
-
Road dust of any state
NR_*
Road dust of any state
RD_*
Biogenic sources of any state
BG_*
Mobile sources of any state
MV_*
Emissions
Boundary Concentration
BCON
BCON
Initial Concentration
ICON
ICON
University of California, Riverside


PM Source Apportionment using CMAQ


7

2.2

TSSA Science


The CMAQ Chemical Transport Model (CCTM) is a three
-
dimensional regional scale non
-
hydrostatic air quality

model based on the mass conservation equation to simulate transport,
transformation, dry and wet deposition, and aerosol formation of pollutants. The governing
equation can be rewritten in the generalized coordinates where the turbulent flux terms are
ex
pressed with the eddy diffusion theory (EPA, 1999). Through operator splitting, or fractional
time step, modularity is achieved in CCTM and computationally efficient algorithms were
applied to solve each science processes (e.g., advection, chemistry, etc.
). The TSSA algorithm is
implemented into each science process to account for the change in species concentration and
update the tagged
-
species accordingly. Figure 3 illustrates the TSSA implementation in CCTM
science processors. The TSSA routine c
hecks

for mass conservation at each advection time step
and adjusts the mass (renormalize), if needed, to prevent growth of numerical noise. The program
will halt if large errors (greater than 0.01%) are detected during the mass conservation check.


The TSSA in
itialization stage takes place at the beginning of the model simulation. The tagged
species are initialized with concentrations from the model initial condition. The TSSA algorithm
also updates the tagged species at the boundary using concentrations from t
he boundary
conditions input file when there is influx to the model domain. Outflux at the boundaries is
treated as simple removal process.



Figure 3

Flow chart of the TSSA implementation in CCTM


driver.F:
read
tssa
configuration,
ptssa
_init …
do n = 1,
nsteps
tssa
_couple
tssa
_decouple
write
tssa
output
end do
sciproc
.F
Xadv
Yadv
Yadv
Xadv
Zadv
Tssa Adjadv
Hdiff
Tssa
Decouple
Vdiff
Tssa Cldproc
Chem
Tssa
Aero
Tssa
Couple
Hppm
Tssa
adv
update
vppm
Tssa
hdiff
update
Tssa
vdiff
update
smvgear
Tssa
chem
update
TSSA mass
normalization
driver.F:
read
tssa
configuration,
ptssa
_init …
do n = 1,
nsteps
tssa
_couple
tssa
_decouple
write
tssa
output
end do
sciproc
.F
Xadv
Yadv
Yadv
Xadv
Zadv
Tssa Adjadv
Hdiff
Tssa
Decouple
Vdiff
Tssa Cldproc
Chem
Tssa
Aero
Tssa
Couple
Hppm
Tssa
adv
update
vppm
Tssa
hdiff
update
Tssa
vdiff
update
smvgear
Tssa
chem
update
TSSA mass
normalization
University of California, Riverside


PM Source Apportionment using CMAQ


8

Emissions.
Emissions are treated as a simple source that

increments the bulk species and, if a
given emissions source is selected for the TSSA algorithm, the tagged
-
species is also
incremented. Emissions sources that are not selected for the TSSA are added to a tagged
-
species
identified as “other”. This approac
h allows for all emissions to be included in the TSSA
algoirthm so that the total mass of the sum of the tagged species (including “other”) equals the
total concentration of the bulk species for each individual grid cell. The TSSA routine updates
the tagg
ed
-
species concentration at each model time
-
step with the release of emissions. The
tagged
-
species NO
X

is incremented by emissions of all reactive N species, i.e., NO, NO
2

and
HONO emissions.


Horizontal and Vertical Advection.
In the CMAQ CTM, operator s
plitting is used to divide
the advection horizontal and vertical components in the numerical solutions. This is possible
because the mean atmospheric motion is primarily in the horizontal direction and the vertical
motion is usually associated with the in
teraction of dynamics and thermodynamics. The science
processor ADJCON is designed to supplement the advection routines to ensure mass
conservation. The TSSA algorithm uses CCTM’s transport solver to advect each tracer species.
The fluxes calculated at
the upwind and downwind grid cell boundaries are used to update each
tracer species.


Horizontal and Vertical Diffusion.
The atmospheric turbulent diffusion is divided into vertical
and horizontal diffusion. The vertical diffusion mainly represents the t
hermodynamic influence
on the atmospheric turbulence by the air
-
surface energy exchange process, while the horizontal
diffusion represents subgrid scale mixing due to the unresolved wind fluctuations (EPA, 1999).
A semi
-
implicit scheme is implemented in t
he current CMAQ vertical diffusion science
processor. The net change in species concentration at each layer is obtained by solving the tri
-
diagonal matrix with species concentration from the current and the next time steps. Vertical
diffusion of the the
tracer tagged
-
species is solved using the CMAQ diffusion algorithms to
calculate the net mass exchange of each tracer species through the two
-
way mass diffusion
between layers. At the end of each advection time step, the TSSA routine c
hecks for mass
conse
rvation and adjusts mass (renormalize) if necessary. The program will halt if large errors
are detected during the mass conservation check.


Deposition.

Deposition is coupled with vertical diffusion in the CMAQ CTM. The TSSA
tagged
-
species are deposited

the same way as the CMAQ species, with the loss of each tagged
-
species being proportional to the deposition rate of the bulk species and the sum of deposition for
all tagged
-
species, including “other,” summing to equal the deposition of the bulk species.


Chemistry.
Treatment of dispersion and transport processes is relatively straightforward
because it simply involves the redistribution of the tagged
-
species analogously to that of the bulk
species, with the tagged species concentration conserved at each
step. Chemical transformations,
by contrast, require that mass be transferred between tagged
-
species. Our approach is to use the
integrated rates of chemical reactions, which are provided by CMAQ when selecting the “process
analysis” option. The integrate
d reaction rates provide the mass transfer from one bulk species to
another at each chemistry time step. We then update the tracer species according to the changes
in the integrated reaction rates.

University of California, Riverside


PM Source Apportionment using CMAQ


9


In the simplest case for a species with chemical loss bu
t no chemical production, we simply
reduce the tagged
-
species proportionally to the reduction in the bulk species. For example, given
an initial bulk SO
2

concentration of 100 ppb and a SO
2

concentration of 10 ppb from Colorado
point sources, the chemical l
oss of 5 ppb SO
2

during a chemistry time
-
step, the loss frequency of
5%, would apply proportionally to the bulk and tagged species so that their concentrations would
be 95 ppb and 9.5 ppb, respectively, at the end of the time step.


In the more complex ca
se in which there is both chemical production and loss of species
C
j
, the
rate of removal of the bulk and tagged components is calculated by solving a simple ordinary
differential equation:



bulk
j
Total
j
bulk
j
LC
P
dt
dC
,
,
,








Eq(1)


where
P
j

represents chemical production and
L
j

is the loss frequency of
C
j
. The analytical
solution for Eq(1) during the time interval

t

= (

t
n+1

-

t
n

) is:



t
L
t
bulk
j
t
L
j
Total
j
t
bulk
j
e
C
e
L
P
C
n
n








,
,
,
)
1
(
1




Eq(2)


wher
e the first term one the right
-
hand
-
side of Eq(2) is the amount of new production of
j

remaining at
t
n+
1
, and the second term is the amount of
C
tn

remaining at
t
n+
1
. We can account for
differing rates of production of the bulk and tagged components by subs
tituting the individual
tagged components of
C
j

into Eq(2):



C
j,bulk

=
C
j,initial

+

C
j,bondary

+
C
j,S1

+
C
j,S2

+ …+
C
j,Sm



Eq(3)


where
C
j,initial

represents the amount of the initial distribution still remaining,

C
j,bondary

repres
ents
the amount transported in from boundaries, and
C
j,S1

and
C
j,Sm

represent the amount remaining of
chemical production prior to
t
n

from sources
S
1

and
S
m
, respectively. In addition, there is some
new chemical production during

t
:




P
j,Total

=
P
j,S1

+
P
j,S2

+ … +
P
j,Sm




Eq(4)


We can then write Eq(2) for each tagged component to determine the amount remaining at

t
n+1
:



t
L
t
initial
j
t
initial
j
e
C
C
n
n





,
,
1




Eq(5)



t
L
t
boundary
j
t
boundary
j
e
C
C
n
n





,
,
1




Eq(6)



t
L
t
S
j
t
L
j
S
j
t
S
j
e
C
e
L
P
C
n
n









1
1
1
,
1
,
,
)
1
(





Eq(7)

University of California, Riverside


PM Source Apportionment using CMAQ


10



t
L
t
S
j
t
L
t
S
j
e
C
e
L
P
C
n
m
j
m
S
j
n
m









,
,
)
1
(
,
1





Eq(8)


By solving Equations (5) through (8
) for each tagged species at each time step, it is possible to
track the temporal evolution of the contribution to a species concentration due to initial mass,
transport from boundaries, and chemical formation. The production and loss terms needed to
solve

these equations can be obtained by integrating the rates of chemical reactions.


Aerosol Chemistry.

The CMAQ CCTM treats aerosols using a modal approach where primary
particles are divided into 2 fine modes and 1 coarse mode, where each mode is represen
ted by a
log
-
normal size distribution. The key processes for aerosol formation and removal include
nucleation, secondary organic aerosol formation, particle coagulation and condensation growth,
aerosol
-
cloud droplet interaction, and removal by dry deposit
ion and precipitation. At each step,
the CMAQ algorithms where the aerosol species are updated, the tagged
-
species are also updated
with the change being proportional to the change in the bulk species concentration.


CMAQ uses the ISORROPIA thermodynamics

equliblium algorithm for treating the relationship
among gas
-
phase HNO
3
, sulfate and amonnia, aerosol ammonium nitrate and sulfates, and
humidity. We have currently implemented a simplistic approach in which we calculate the net
change in each of the spec
ies involved in the equlibrium at each time
-
step in the aerosol solver.
For example, if there is net conversion of gas
-
phase to aerosol ammonium, the tagged
-
species for
the aerosol ammonium are updated using the relative concentrations of the gas
-
phase amo
nia
tagged
-
species. This approach might work well because the model oeprates on short time
-
steps
on the order of 1 minute. However, as part of this study we propose to test alternative schemes
for treating the mass transfer between gas
-

and aerosol
-
phase
. In particular, we will explore
dynamic approaches within the ISORROPIA scheme in which we estimate the total flux between
gas and aerosol.


2.3 Evaluation of CMAQ TSSA Algorithm


We propose to carry out three sets of activities to evaluate the CMAQ TSSA
algorithm. First, a
comprehensive model performance evaluation will be carried out for the CMAQ simulation for
calendar year 2002 for both the 36
-
km and nested 12
-
km domains. This performance evaluation
is funded by WGA/WRAP and includes evaluation and QA
of the MM5 simulations, emissions
modeling, and comparisons of the CMAQ results to ambient monitoring networks, including
IMPROVE, SEARCH and Speciation Trends Network for PM species, CASTNet and NADP
deposition data, and EPA’s AIRS (or AQS) gas phase ambi
ent monitoring data. Performance
metrics will include mean normalized error and bias, correlation coefficients, time
-
series plots
comparing model to data, and quantile
-
quantile plots of model versus data.


Two additional tasks are proposed as part of this

study. First, we will perform CMAQ sensitivity
simulations in which selected emissions source categories are zeroed out. The CMAQ TSSA
results for source apportionment will then be compared to the results of model sensitivity
simulations to assess the deg
ree of consistency between the two methods. Sensitivity simulations
for non
-
reactive elemental carbon will be included as a reference test, since the TSSA and
University of California, Riverside


PM Source Apportionment using CMAQ


11

sensitivity methods should provide identical results for non
-
reactive species. Other sensitivity
simulations will be planned as part of the modeling protocol under Task 2 below.


As a second task, we will compare results of our CMAQ TSSA and sensitivity simulations with
results from a separate study entitled, “Causes of Haze,” being funded by WGA/WRA
P and
carried out at the University of Nevada’s Desert Research Institute (PIs: Mark Greene and Marc
Pitchford). Deliverables from the DRI project will include the following:




Geographic source areas of emissions that contribute to impairment at each manda
tory
federal and tribal Class I area.



Mass and species distributions of emissions by source categories within each contributing
geographic source area.



The amount of natural and manmade emissions affecting each Class I area.


A detailed plan for comparing
the CMAQ simulation results to the “Causes of Haze” results will
be prepared as part of the model performance evaluation.


2.4 Quality Assurance Project Plan


As the first task in this project, a Quality Assurrance Project Plan (QAPP) will be prepared in
accordance with EPA requirements for quality assurance project plans for modeling (U.S. EPA
QA/G
-
5M, 2002) and North American Research Strategy for Tropospheric Ozone (NARSTO)
Quality Handbook template for modeling projects (NARSTO, 1998). The QAPP will a
lso
describe QA procedures routinely used by the UCR modeling group for meteorological,
emissions and air quality modeling, and data backup, archiving and security procedures.


2.5 Project Tasks


Work proposed here will use as a starting point the code dev
eloped with funding from WGA. All
model input data sets including emissions and meteorology have been developed by UCR and
collaborators under separate funding from the WGA/WRAP. The following additional tasks are
proposed here:


Task 1. Prepare a Quality

Assurance Project Plan.

Task 2. Prepare a modeling protocol to document data sets and model simulations planned as part

of this study.

Task 3: Modify the Models
-
3/CMAQ code to use tagged
-
species tracers to represent gas
-
phase
organic intermediates, and pr
imary and secondary organic aerosols.

Task 4: Test the tagged
-
species approach to evaluate the contribution of transport and chemical
production to speciated PM, including nitrate, sulfate, ammonium, organic aerosols, and
elemental carbon for a CMAQ simula
tion for a calendar year 2002 simulation using the coarse
grid model domain shown in Figure 4.

University of California, Riverside


PM Source Apportionment using CMAQ


12

Task 5: Perform 8 CMAQ sensitivity simulations with selected emissions source categories
zeroed out and compare results of sensitivity simulations with results f
rom the CMAQ TSSA
simulations.

Task 6. Compare results of CMAQ TSSA and sensitivity simulations with back
-
trajectory
modeling and chemical mass balance modeling results from the DRI project.

Task 7: Apply the CMAQ TSSA approach in a nested, 12
-
km resoluti
on model simulation, as
shown in Figure 5, to evaluate effects of grid resolutions on the source apportionment results.

Task 8: Present results at the American Geophysical Union Fall Meeting in December 2005, and
submit manuscripts to peer
-
reviewed journal
s.

Task 9: Submit interim and final reports.





Figure 4. RPO Unified grid to be used in WRAP 2002 coarse grid modeling.

University of California, Riverside


PM Source Apportionment using CMAQ


13



Figure 5. High resolution 12 km nested grid for the WRAP region, shown in blue, to be
used in the WRAP 2002 modeling. Domain siz
e is 207 columns by 186 rows.






3. Expected Results or Benefits


The major result of this effort will be development of a new modeling tools for performing
source apportionment.
By providing a method to characterize the three
-
dimensional dynamics of
ai
r pollutant transport, this project will contribute to an understanding of the role of transport in
air pollution episodes and will provide valuable knowledge to state and local air quality
managers for developing strategies to attain air quality goals. Su
ch knowledge will increase our
confidence in model predictions of air quality attainment strategies and will reduce the risk of
adopting emissions reductions strategies that are ineffective for reducing pollution levels.



The algorithm developed will be m
ade available to the scientific and regulatory modeling
communities. It is expected to be useful for assessing the contribution of inter
-
basin and trans
-
boundary transport to violations of air quality standards. It will be particularly useful for
evaluatin
g the effects on secondary pollutants due to long
-
range transport from large point
sources of NO
x

emissions. Because the algorithm will include a detailed representation of the
budgets of individual NO
y

species, it can be easily extended to represent the c
omplex inter
-
conversion of NO
y

between the reactive forms of NO
x

and the relatively inert forms of PAN,
HNO
3

and RNO
3
.



University of California, Riverside


PM Source Apportionment using CMAQ


14

4. General Project Information


Dr. Tonnesen’s research group is located in the UCR College of Engineering
-
Center for
Environmental Re
search and Technology (CE
-
CERT), which operates approximately 25,000
square feet of laboratory space for the measurement and study of atmospheric pollutants and
environmental modeling
. Dr. Tonnesen’s modeling group at
CE
-
CERT
/UCR has been home for
the past

three years to the Western Regional Modeling Center (WRMC), which supports state
and tribal agencies in the Western United States with visibility modeling needs. The WRMC
project (which has just been renewed for at least a 4th year) provides us with the e
xpertise,
experience, personnel, and resources to efficiently carry out a comprehensive air quality
modeling program.


Dr. Tonnesen’s computer laboratory is designed to process large data sets for air quality
modeling. It includes an SGI Origin 2000 workst
ation and 30 high
-
performance, dual CPU Linux
workstations configured as several small Linux clusters on a private network to facilitate parallel
simulations. These systems include a 24 CPU Athlon 2000MP cluster, an 8 CPU Xeon 2.2GHz
cluster, and an 8 CPU
64
-
bit Opteron 2GHz cluster. The data storage system includes over 20 TB
of disk space configured as RAID5 disk systems. All computers and disk systems are connected
using high speed Gigabit ethernet for efficient simulation and analysis of large datasets.

To
provide maximum data security, the systems are located behind the UCR firewall and an
additional firewall internally within the laboratory. A separate system is used for the project
websites and ftp site to allow project data to be accessed thought the

UCR T1 internet
connection.
Also available through the Internet are the University of California’s supercomputer
facilities, located in San Diego.





Data is also routinely transferred from Dr. Tonnesen’s systems to other organizations using a
variety o
f tape formats and portable hard drives. The data backup/archiving system include 8mm
tape drives and DLT and Super DLT auto loading cartridge system capable of performing
unattended archive/backups of over 1 TB (uncompressed). Key disk systems have hot
-
sw
appable
hard drives with stand
-
by spare drives and redundant power supplies. Copies of critical project
data are backed up to tape and to a redundant RAID5 disk system. The compute clusters and disk
systems are located in a locked, secure room with a dedic
ated climate control system and with
backup air conditioning. Dr. Tonnesen also uses the computer laboratory for air pollution
modeling classes for UCR graduate students and for professional staff from State and Tribal air
pollution agencies. The laborator
y has a full time systems administrator to perform system
backups, maintenance and updates, and Dr. Tonnesen’s group includes a second full time
systems administrator.



4.1 Personnel


The Principal Investigator for this project is Dr. Gail S. Tonnesen, Ma
nager of Environmental
Modeling at the University of California, Riverside, Bourns College of Engineering
-
Center for
Environmental Research and Technology. Dr. Tonnesen received the B.S. degree in Chemical
Engineering from Michigan State University in 1983
, and M.S. and Ph.D. degrees in
Environmental Engineering from the University of North Carolina at Chapel Hill in 1991 and
University of California, Riverside


PM Source Apportionment using CMAQ


15

1995, respectively. Dr. Tonnesen also served as a Peace Corps Volunteer in Zaire, Africa, from
1983 to 1987, and then worked in SIP d
evelopment at the U.S. EPA Regional office in San
Francisco from 1987 to 1988. From 1995
-
1999 Dr. Tonnesen was a National Research Council
Associate and worked with U.S. EPA and NOAA scientists in Research Triangle Park, NC. In
1999, Dr. Tonnesen joined U
CR to develop a regional scale air quality modeling program. Dr.
Tonnesen’s research group uses computer simulation models to study the chemistry and transport
of trace species in the troposphere including both fundamental research and model applications.
Research activities include improving the understanding of chemical processes in the
troposphere, particularly the budgets of the free radical and odd nitrogen species that control
urban to regional scale pollutants; development of process level analysis m
ethods to understand
and explain results of model simulations; investigation of the usefulness of indicator ratios for
assessing ozone sensitivity to precursor emissions and for validating model simulations; methods
to evaluate source impacts and area of i
nfluence of precursor emissions; and investigation of the
effects of algorithms, grid
-
structure and model formulation on the interaction of chemical and
transport processes.


Dr. Zion S. Wang

is a Principal Development Engineer at UCR. He holds a B.S. in A
tmospheric
Sciences from National Taiwan University (1985), an M.S. in Atmospheric Sciences from North
Carolina State University (1991), and a Ph.D. in Environmental Sciences and Engineering from
the University of North Carolina at Chapel Hill (1998). He j
oined UCR in 2001 to work in the
expanding environmental modeling group, where his research includes atmospheric modeling,
fate, transport, and visibility. He is a key staff member of the Western Regional Modeling
Center. Previously, he was an independent
consultant in atmospheric modeling (1996
-
2001), a
postdoctoral researcher at the University of North Carolina (1998
-
2000), and graduate student
(1995
-
96). He was a senior meteorologist at Systems Applications International from 1991 to
1996.



Bo Wang

is a

programmer
-
analyst for UCR’s Western Regional Modeling Center. He develops
automated software tools for model performance evaluations
.

He has built a
3000
-
line air quality
modeling
visualization

tool using
NCAR

graphics.
He has developed code for CMAQ par
allel
version and coded C programs to transform MM5 data to netcdf files for both RIP (NCAR
graphics) and PAVE (IOAPI) visualization tools. He has run and debugged the SMOKE
emissions processor. Mr. Wang holds a M.S. in computer science from the University

of
California, Riverside (2001) and was a Ph.D. candidate in environmental toxicology from 1998
to 1999. He holds an M.D. degree from China Medical University (1994).


CE
-
CERT technical and administrative staff will provide additional support as needed.



University of California, Riverside


PM Source Apportionment using CMAQ


16

4.2 Project Schedules


Activity

Begin

End

1. Prepare Quality Assurance Project Plan

10/01/04

11/1/04

2. Prepare a modeling protocol

10/01/04

12/01/04

3. Modify the Models
-
3/CMAQ TSSQAcode

10/1/04

10/01/05

4. Test Simulations with CMAQ 36
-
km model

3/0
1/05

5/01/05

5. CMAQ Sensitivity Simulations

5/01/05

7/01/05

6. Model comparison to “Causes of Haze” study results

㜯ㄯ〵

ㄲ⼰ㄯ〵

㜮⁃潭灡o楳潮i映䍍䅑⁔ p䄠楮″㘠A湤‱n
-
歭k
浯摥m

ㄯ〱N〶

㤯㌱V〶

㠮⁐牥灡牥慮畳 物灴⁦潲⁰ e爠re癩敷潵牮 氬⁡湤l步

c潤o⁡癡楬a扬攠瑯瑨敲⁃䵁j⁵獥牳r


ㄲ⼰ㄯ〵

㘮⁉湴敲業⁒e灯牴Ⱐ㄰⼰㔮⁆楮慬⁒数潲琬‱〯〶M

㄰⼰ㄯ〵

㄰⼰ㄯ〶



4.3 Management


The Principal Investigator will be responsible for management of project tasks, schedules, and
budgets. CE
-
CERT will provide

administrative support including financial and technical project
administration. CE
-
CERT and the University have carried out numerous Federal research
projects, many of them larger in size and scope than this proposed project. The University has all
admin
istrative and managerial mechanisms in place to assure successful administration of this
project.



4.4 Related Research


Dr. Tonnesen developed the process analysis method for evaluating air quality model simulations
in her doctoral work at the University

of North Carolina. Since 1995, Dr. Tonnesen has worked
with the NOAA/EPA modeling group in Research Triangle Park, NC to further develop and
apply the process analysis for Eulerian air quality models, and to implement this method in the
Models
-
3/CMAQ mode
ling system.



4.5 Interactions with Other Institutions


Several of Dr. Tonnesen’s projects are joint collaborations with other institutions, including
ENVIRON Corporation, the University of North Carolina at Chapel Hill, the University of
California at S
anta Barbara, the U.S. Forest Service, and U.S. Department of Agriculture.

University of California, Riverside


PM Source Apportionment using CMAQ


17

5. Resumes of Key Personnel


Gail S. Tonnesen

Assistant Research Engineer; Manager of Environmental Modeling

CE
-
CERT 022, University of California, Riverside, CA 92521
-
0434 tonnes
en@cert.ucr.edu


Professional Preparation


Michigan State University

Chemical Engineering

B.S., 1983

University of North Carolina, Chapel Hill

Environmental Engineering

M.S., 1990

University of North Carolina, Chapel Hill

Environmental Engineering

Ph.D., 1
995


Appointments


1999
-
present. University of California, Riverside, Bourns College of Engineering
-
Center for
Environmental Research and Technology. Director, Western Regional Air Partnership Regional
Visibility Modeling Center, University of California,
Riverside (2000
-
present).
Assistant
Research Engineer and
Manager of Environmental Modeling (1999
-
present).


1995
-
99. National Research Council Research Associate, U.S. Environmental Protection
Agency, Research Triangle Park, NC.


1987
-
88. Environmental E
ngineer, Air Management Division, U.S. EPA, San Francisco, CA.


1983
-
87. Field Director/Volunteer Leader/Peace Corps Volunteer, Republic of Zaire.


Relevant Publications


Dennis, R.L.; Arnold, J.R.; and Tonnesen, G.S. (2000) One
-
at
-
a
-
Time and Mini
-
Global

Analyses
for Characterizing Model Sensitivity in the Nonlinear Ozone Predictions the U.S. EPA Regional
Acid Deposition Model, in
Mathematical and Statistical Methods for Sensitivity Analysis
, (ed.
Saltelli, A., K. Chan, and M. Scott). John Wiley Publisher
, pp 329

353.

Pierce, T.; Geron, C.; Bender, L.; Dennis, R.L.; Tonnesen, G.S.; and Guenther, A. (1998)
Influence of Isoprene Emissions on Regional Ozone Modeling.
J. Geophys. Res., 103,
25,611
-
25629.

Lents, J.M., et al. (2001) A Study of Peak
-
Load Energy P
roduction Potential and Air Quality
Implications of Backup Generators and Distributed Generation in California. Progress Report #1
to the California Energy Commission under Contract 500
-
00
-
032. July.


Other Significant Publications

Tonnesen, G.S., and Denn
is, R.L. (2000) Analysis of Radical Propagation Efficiency to Assess
Ozone Sensitivity to Hydrocarbons and NO
x
. Part 1: Local Indicators of Odd Oxygen Production
Sensitivity.

J. Geophys. Res.
,
105
,9213

9225.

University of California, Riverside


PM Source Apportionment using CMAQ


18

Tonnesen, G.S., and Dennis, R.L. (2000) Analysis

of Radical Propagation Efficiency to Assess
Ozone Sensitivity to Hydrocarbons and NO
x
. Part 2: Long
-
Lived Species as Indicators of Ozone
Concentration Sensitivity.
J. Geophys. Res.
,
105
, 9227

9241.

Tonnesen, G.S. (1999) Effects of Uncertainty in the React
ion of the Hydroxyl Radical with
Nitrogen Dioxide on Model Simulated Ozone Control Strategies.
Atmos. Environ.,
33
, 1587
-
1598.

Luecken, D.J.; Tonnesen, G.S.; and Sickles, J.E. II, (1999) Differences in NO
y

Speciation
Predicted by Three Photochemical Mechan
isms.
Atmos. Environ.,
33
, 1073
-
1084.

Dennis, R.L.; Arnold, J.R.; Tonnesen, G.S.; and Li, Y. (1999) A New Response Surface
Approach for Interpreting Eulerian Air Quality Models Sensitivities.
Computer Physics
Communication, 117
, 99
-
112.

Synergistic Activit
ies




Director of Western Regional Modeling Center, established by U.S. EPA and Western
Governors’ Association to train state and tribal air quality officials in visibility modeling for
development of plans to protect air quality in and around national park
s and other Class I
areas.



Principal Investigator for NSF
-
sponsored Biodiversity planning grant (
BES
-
0083383)
to
integrate research and modeling approaches from multiple disciplines.


Collaborators

Professor Michael Allen, University of California, Riversi
de

Professor Edith Allen, University of California, Riverside

Professor Julian Allison, University of California, Riverside

Professor Gene Anderson, University of California, Riverside

Dr. J. R. Arnold, U.S. Environmental Protection Agency

Mr. Ralph Morris
, ENVIRON Corp., Novato, CA

Dr. Robin Dennis, National Oceanic and Atmospheric Administration (NOAA/EPA)

Professor Harvey Jeffries, University of North Carolina, Chapel Hill

Professor William Jury, University of California, Riverside

Dr. James Lents, Unive
rsity of California, Riverside

Dr. Greg Yarwood, ENVIRON Corp., Novato, CA


Academic Advising

Dr. Tonnesen is advising two Ph.D. students beginning fall semester, 2001.


Graduate and Postgraduate Advisors

Prof. Harvey Jeffries, University of North Carolina
, Chapel Hill, NC.

Dr. Robin Dennis, National Oceanic and Atmospheric Administration, RTP, NC.


University of California, Riverside


PM Source Apportionment using CMAQ


19

Zion S. Wang

Principal Development Engineer

CE
-
CERT 022, University of California, Riverside, CA 92521
-
0434 zsw@cert.ucr.edu


Education


B.S., Atmospheric Sci
ences, National Taiwan University, 1985.

M.S., Atmospheric Sciences, North Carolina State University, 1991.

Ph.D., Environmental Sciences and Engineering, University of North Carolina, Chapel Hill,
1998.


Professional and Academic


2001
-
present.

University

of California, Riverside, College of Engineering
-
Center for
Environmental Research and Technology. Staff Research Associate in atmospheric
modeling, including fate, transport, and visibility modeling. Key researcher for the
Western Regional Air Partnershi
p Regional Modeling Center.

1996
-
2001.

Independent consultant in atmospheric modeling. Projects included a 3
-
year
investigation into how vertical mixing affects downstream ozone formation and
implementing a vertical mixing method based on turbulent transil
ient theory into the
SARMAP Air Quality Model (SAQM). As part of the Hong Kong Territorial
-
Wide Air
Quality Project, implemented the Model for Aerosol Reacting System (MARS) into the
SAQM. Assisted EPA scientists in developing a new methodology to obtain v
olatile
organic compound (VOC) reactivity data for use in development of ozone control
strategies based on three
-
dimensional photochemical model. Reviewed several chapters
in the “Science Algorithms of the EPA Models
-
3 community Multiscale Air Quality
(CMA
Q) Modeling System.”


1995
-
2000.

University of North Carolina, Chapel Hill. Postdoctoral researcher (1998
-
2000) and
Research Assistant (1995
-
96). Implemented Integrated Process Rate Analysis Method
(IPRAM), also known as process analysis, into the Urban Ai
rshed Model
-
IV (UAM
-
IV),
the urban Airshed Model
-
V, the SAQM, and the Comprehensive Air Quality Model with
Extensions (CAMx). Also developed the Continuous Process Composition and Source
Receptor (CPCSR) method, which enables detailed source
-
receptor analy
ses to be
performed in Eulerian grid
-
based models.

1991
-
96.

Systems Applications International. Senior Meteorologist. Investigated influence of
geopolitical regions on air quality in the Baltimore area for the Maryland Department of
Environment. Provided t
raining workshop on the use of CAMx.

1989
-
91.

North Carolina State University. Research Assistant.

1990.

Computer Sciences Corporation. Meteorologist.


University of California, Riverside


PM Source Apportionment using CMAQ


20

Selected Publications


“Implementation of Integrated Process Rate Analysis to the SAQM
-
AERO Model” (wit
h Dr.
Harvey Jeffries), prepared for the California Air Resources Board, Sacramento, CA (2000).


“Computation of Volatile Organic Compound Reactivity,” prepared for the U.S. EPA, Office of
Air Quality Planning and Standard, October 1999.


“The Continuous P
rocess Composition and Source Receptor Methodology” (with Dr. Harvey
Jeffries), presented at the A&WMA Conference in St. Louis, MO, June 1999.


“Applications of Integrated Process Rate Analysis to the SARMAP Air Quality Model” (with Dr.
Harvey Jeffries), p
repared for the California Air Resources Board, Sacramento, CA (1998).


“An Overview of the Territory
-
Wide Air Quality Modeling System (TWAQMS) for Hong Kong”
(with others), prepared for the California Air Resources Board, Sacramento, CA (1998).


Compariso
n of Three Vertical Diffusion Schemes in the SARMAP Air Quality Model with the
Integrated Process Rate Analysis Method and the Continuous Process Composition and Source
Receptor Methodology. A Ph. D. dissertation submitted to the University of North Caroli
na at
Chapel Hill (1998).


Role of Atmospheric Stability in the Formation of the Gulf Stream Cloud Bands. A Master’s
Thesis submitted to the Department of Marine, Earth and Atmospheric Sciences, North Carolina
State University, Raleigh.


Role of Atmospheri
c Stability in the Formation of the Gulf Stream Cloud Bands. Presented at the
AMS Conference in Miami, Florida, May 1991.


Collaborators


Gail S. Tonnesen

Harvey Jeffries

Ralph Morris

Fred Vukovich


Graduate and Postgraduate Advisors


Harvey Jeffries





University of California, Riverside


PM Source Apportionment using CMAQ


21

Bo Wang

Programmer/Analyst

College of Engineering, University of California

Center for Environmental Research and Technology, Riverside, CA 92521

voice: 909
-
781
-

5680; fax: 909
-
781
-
5790; bwang@cert.ucr.edu


Education

M.S. in Computer Science, University
of California, Riverside,

12/99
-

09/2001.

Ph.D Candidate in Environmental Toxicology, University of California, Riverside, 09/98


12/99
.

M.
S
. in Medicine, China Medical University, 09/9
4



07/97.

M.D., China Medical University, 09/89


07/94


Professio
nal Experience

1.

Programmer Analyst II
,

Regional

Modeling Center, UC, Riverside,
09/2001


present

a.

Build a
6000
-
line Model Evaluation Tool on Linux for EPA's CMAQ and
Environ's REMSAD model system using C++, Perl, Csh, Fortran90 and
Gnuplot. Key parts includ
e a database handling preprocessor, a converter
between different coordinate systems and model domain, a translator to fulfill
the same function as Windows' Excel does on vectors' macros equation, and a
plot
t
ing package to do statistical analysis and draw
time

series and scatter
plots automatically
.

b.

Build a 3000
-
line Air Quality Modeling Result
Visualization

Tool using Ncar
graphics.

c.

Develop Code for CMAQ parallel version, Box Model and Dataset
Processing.

d.

Coded C programs based on Netcdf (network indepen
dent file) library to
transform Meteorology database to netcdf files for both RIP (NCAR graphics)
and PAVE (IOAPI) visualization tools.

e.

Run and Debug Sparse Matrix Operator Kernel Emissions Systems, etc.

2.

Software Engineering Intern
,

Environmental Modeling
Group,
CE
-
CERT,
UC
Riverside,
Summer 2000

a.

Coded on Linux and SGI platform to
do air quality modeling
.

Special
Knowledge for CMAQ model includes BCON, ICON, MCIP, JPROC, ECIP,
and CMAQ chemistry
-
transport model (CCTM).

3.

Research
Assistant
,
Computer Science,

University of California, Riverside, 09/00
-
09
-
01

a.

Implemented the Kleinberg algorithm and the PageRank algorithm (based on
Google's idea) to improve the quality and efficiency of Web Searching Engine

for INFOMINE project
, C++ and Java are used. The impleme
ntation includes
writing a web spider to construct root set and base set according search query,
using Kleinberg and PageRank algorithm to go through these sets to form sub
-
graph
s

which held hub and authority relationship between forward and back
University of California, Riverside


PM Source Apportionment using CMAQ


22

links, an
d doing matrix and vector calculation which leads to conversion. Top
authority or "best link" will finally be returned to users.

4.

Teaching
Assistant
,
Computer Science, U
niversity of California,

Riverside, 12/99


09/01


Taught Compiler lab, Data Stru
cture lab, and Algorithm lab in C and C++


5.

Graduate Student Researcher
,
Environmental Toxicology,

UC, Riverside, 09/98
-
12/99

a.

Did Research on the Function and Pharmacology of a putative Na
-
K
-
2Cl co
-
transporter in the Malpighian Tubules of the mosquito Aed
es Aegypti.

b.

Did lab works on Atmospheric Chemistry and Bio
-
Toxicology Mechanisms.

6.

Research
Assistant
,
Department of Pharmacolog
y
, China Medical University, 07/97
-
07/98

a.

Did Research on the Relationship between Bcl2/P53 protein expression and
lung cancer di
agnosis.

b.

Did research on Pharmacology effects of certain mouse specie
s knockout
genes
.

c.



SKILLS
:

1.

Programming Languages
: C, C++, Visual C++/MFC, Java, Perl, VHDL, Assembly,
FORTRAN, PostgreSQL, Ncarg, HTML.

2.

Programming

Tools
: GDB, DDD, Dmalloc, Xilinx

Tools, Vmware (virtual
machine), CVS.

3.

Platforms
: Linux,
UNIX,
SGI, Windows 98/NT/2000, MS
-
DOS.

4.

Biology Laboratory Techniques
:

Recombinant DNA and Hybridization, Southern
Blotting, and Western Blotting, Cell and
Tissue

Culture, SDS Page Electrophoresis,
S
pectrophotometer, Antibody Purification, PCR, Immuno
-
histo
-
chemistry, etc.



HONORS & AWARD
:


1.

Dean Fellowship
, University of California, Riverside, 08/98


12/
99

2.

Fukuto Fellowship,
University of California, Riverside, 06/30/99

a.

For the best first year gradu
ate student in Environmental Toxicology
Graduate Program

3.

TAships,
University of California, Riverside, 12/99


09/01



University of California, Riverside


PM Source Apportionment using CMAQ


23


6. Current and Pending Support



University of California, Riverside


PM Source Apportionment using CMAQ


24



University of California, Riverside


PM Source Apportionment using CMAQ


25

7. Budget



Itemized Budget for EPA STAR Grant Application
CATEGORIES
YEAR ONE
YEAR TWO
TOTAL PROJECT
A. Personnel
G. Tonnesen — Principal Investigator
17,280
$

17,280
$

34,560
$

Z. Wang — Pr. Dev. Eng.
20,801
$

20,801
$

41,602
$

B. Wang — Assoc. Dev. Eng.
26,610
$

26,610
$

53,220
$

TOTAL PERSONNEL
64,691
$

64,691
$

129,382
$

B. Fringe Benefits
G. Tonnesen (@ 17%)
2,938
$

2,938
$

5,876
$

Z. Wang (@ 22%)
4,576
$

4,576
$

9,152
$

B. Wang (@ 22%)
5,854
$

5,854
$

11,708
$

TOTAL BENEFITS
13,368
$

13,368
$

26,736
$

C. Travel
Out-of-town travel
1,250
$

1,250
$

2,500
$

TOTAL TRAVEL
1,250
$

1,250
$

2,500
$

D. Equipment
RAID5 disk system
8,000
$

-
$

8,000
$

4 dual CPU linux mach. @ $3500/ea.
14,000
$

14,000
$

TOTAL EQUIPMENT
22,000
$

-
$

22,000
$

E. Supplies
Phone, fax, mailing, misc. supplies
500
$

500
$

1,000
$

Computer supplies, data tapes
1,000
$

1,000
$

2,000
$

Computing Services fees
241
$

242
$

483
$

TOTAL SUPPLIES
1,741
$

1,742
$

3,483
$

F. Contracts
None
TOTAL CONTRACTUAL
-
$

-
$

-
$

G. Other
Facilities rental
16,939
$

16,940
$

33,879
$

TOTAL OTHER COSTS
16,939
$

16,940
$

33,879
$

H. Total Direct Costs
(sum of A-G)
119,989
$

97,991
$

217,980
$

I. Indirect Costs
26% of MTDC*
21,073
$

21,073
$

42,146
$

J. Total Project Costs
(sum of H and I)
141,062
$

119,064
$

260,126
$

K. Total Requested
from EPA
141,062
$

119,064
$

260,126
$

*Modified Total Direct Costs: Total Direct Costs minus facilities and equipment
University of California, Riverside


PM Source Apportionment using CMAQ


26

8. Budget Justification


Personnel

The University charges for personnel based on p
ercentage of full
-
time equivalent per month. The
actual number of working hours per month varies. All of the personnel on this project are on full
-
time research appointments, and thus 100% of their salaries can be charged to extramurally
funded research. P
roposed level of effort is as follows:



Annual Salary

Level of effort (% FTE)

Total Cost

Name


Year 1

Year 2


Principal Investigator

$86,400

20

20

$34,560

Principal Development
Engineer

$77,040

27

27

$41,602

Associate Development
Engineer

$53,220

50

5
0

$53,220


No salary escalation is assumed.


Fringe Benefits

For budgeting purposes, benefit rates are 17% for academic personnel (Dr. Tonnesen) and 22%
for staff personnel. These rates are University averages; actual benefit costs will be charged.


Trave
l

The budget provides sufficient funds for project personnel to confer with EPA staff and report
findings. This includes funds for annual meetings of the key investigators to visit EPA to discuss
the project. All travel will be conducted in accordance with

Federal and University policies and
requirements.


Equipment

This proposal includes an equipment budget of $22,000 or approximately 8% of the total project
budget. The equipment budget will be expended in the first year of the project to purchase
addition
al systems:




Four dual
-
CPU linux workstations. Cost is approximately $3,500 each x 4 = $14,000.



$8,000 is budgeted for a RAID5 disk system.


Supplies

The budget includes a moderate amount of funding for printing, copying, mailing, and computer
supplies an
d data tapes in support of the project.


University of California, Riverside


PM Source Apportionment using CMAQ


27

The University instituted a computing services charge in 2002. This charge is based on $20.75
per month per “knowledge worker,” a class that includes researchers, staff, and students. The
charge is pro
-
rated by lev
el of effort.


Subcontractors

None.


Other Direct Costs

Because CE
-
CERT is a permanent off
-
campus facility, federal regulations require us to account
for facilities rental as a direct cost. Facilities rental is charged based on 20.9% of Modified Total
Dire
ct Costs (MTDC). MTDC consists of total direct costs minus equipment, facilities rental,
graduate student partial fee remission/health insurance, and subcontracts over $25,000.


Indirect Costs

The University’s Federally approved indirect cost rate for off
-
campus facilities is 26% of
Modified Total Direct Costs (MTDC).


University of California, Riverside


PM Source Apportionment using CMAQ


28

9. Quality Assurance Narrative Statement



Activities to be performed:

This project will develop new model algorithms and apply these to
analyze transport in model simulations for a ongoin
g regional air quality study. Model input data
sets were collected under the sponsorship of the U.S. Environmental Protection Agency and other
public agencies according to work plans and quality assurance guidelines developed and
approved for each project.

The first task in the project will be the development of a Quality
Assurance Project Plan (QAPP) per EPA guidance.


Study design:

The study consists of algorithm development and modeling model simulations
using model input data sets acquired through previ
ous studies to test the new algorithms. The
second task in the project will be the development of a detailed modeling protocol that described
input data sets and outlines the model simulations to be performed.


Sample handling and custody:

Not applicable.


Analytical methods:

Computer modeling analyses will be performed.


Calibration and performance evaluation:
Model performance will be evaluated and QA’s in
collaboration with ongoing modeling projects at state and federal agencies. The performance of
new a
lgorithms developed under this project will be evaluated by testing for conservation of
mass during the algorithm operation, and by testing for consistency among results of the new
source apportionment algorithm, model sensitivity simulations, and data ana
lysis methods being
carried out at DRI.


Data reduction and reporting:

Entire data sets or representative subsets will be used for
modeling.


Intended uses of data:

The results of the modeling simulations and new algorithms will be a
characterization of ma
ss transport and area of influence of precursor emissions.


Quantitative and/or qualitative procedures for assessment of success:

The project will
considered successful if computational efficient and practicable algorithms are developed to
augment existing

air quality models; if the algorithm results are mass conservative and consistent
with reasonable expectations of the magnitude of transport derived from other studies; and if new
results can be clearly and efficiently conveyed to policy makers.


Peer rev
iew before data collection:

No new data collection is planned. Existing data sets were
collected according to standard operating procedures and quality assurance guidelines that
included peer review. After collection, data sets were analyzed and quality as
surance
-
checked.



University of California, Riverside


PM Source Apportionment using CMAQ


29

10. References Cited


Chang, J. S., P. B. Middleton, W. R. Stockwell, C. J. Walcek, J. E. Pleim, H. H. Lansford, S.
Madronich, F. S. Binkowski, N. L. Seaman, and D. R. Stauffer, The Regional Acid Deposition
Model and Engineer
ing Model,

SOS/T Report 4, in
National Acid Precipitation Assessment
Program: State of Science and Technology, Vol 1
, National Acid Precipitation Assessment
Program, Washington, D.C., 1990.


Chang, J. S, R. A. Brost, I. S. A. Isaksen, P. Middleton, W. R. Stockwell,

and C. J. Walcek, A
three
-
dimensional Eulerian Acid Deposition model: Physical concepts and formulation,
J.
Geophys. Res., 92,

14,681
-
14,700, 1987.


Dennis, R. L., D. Byun, J. H. Novak, K. L. Gallupi, C. J. Coats, M. A. Voukavich, (1996) The
next generat
ion of integrated air quality modeling: EPA’s Models
-
3,
Atmos. Environment, 30
,
1925
-
1938.


Guinnup, D., Collom, B. (1997)

Final Report, Vol. I: Executive Summary

OTAG Air Quality Analysis Workgroup.


McHenry, J. N., F. S. Binkowski, R. L. Dennis, J. S. C
hang, D. Hopkins, (1992) The tagged
species engineering model (TSEM),
Atmospheric Environment, 26A,
1427
-
1443.


Novak, J.H. et al. (1998) Model
-
3: A Unifying Framework for Environmental Modeling and
Assessment,
Proceedings of the 10
th

Conference on Air Poll
ution Meteorology
, pp. 259
-
263.
American Meteorological Society, January 11
-
16, 1998.


Rudolf B. Husar and Wandrille P. Renard (1997) Ozone as a Function of Local Wind Direction
and Wind Speed: Evidence of Local and Regional Transport, Center for Air Poll
ution Impact and
Trend Analysis (CAPITA), Washington University, St. Louis, MO 63130
-
4899. (report available
at: http://capita.wustl.edu/otag/Reports/OTAGWIND/OTAGWIND.html)


Schichtel, B., Husar, R., Poirot, R., Wishinski, P., (1998) Climatology of ozone
synoptic scale
transport in the eastern U.S., Presented at the American Meteorological Society's 78th annual
meeting, Jan. 11
-
16 1998, Phoenix AZ .


US EPA, (1999) CMAQ User Guide, (eds. Byun, D. and Ching, J.), US Environmental
Protection Agency, Office o
f Research and Development, Research Triangle Park, NC 2711.
Documentation available online at:
www.epa.gov/asmdnerl/models3/cmaq.html


Yarwood, G., R. E. Morris, M. A. Yocke, H. Hogo and T. C
hico, Development of a methodology
for source apportionment of ozone concentration estimates from a photochemical grid model,
89th AWMA Annual Meeting, Nashville, TN, June 23
-
28, 1996.