/
EPA-450/2-77-013
JUIV 1977
USER'S IKIANUAL
FOR
SINGLE-SOURCE (CRSTER) MODEL
U.S. ENVIRONMENTAL PROTECTION A8ENCV
Office of Air and waste management
Office of Air Quality Planning and standards
Research Triangle ParK, North Carolina 27711
-------
EPA-450/2-77-013
USER'S MANUAL
FOR
SINGLE-SOURCE (CRSTER) MODEL
Monitoring and Data Analysis Division
Source Receptor Analysis Branch
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Air and Waste Management
Office of Air Quality Planning and Standards
Research Triangle Park, North Carolina 27711
July 1977
-------
This report has been reviewed by the Office of Air Quality Planning and
Standards, Environmental Protection Agency, and approved for publication.
Approval does not signify that the contents necessarily reflect the
views and policies of the Environmental Protection Agency, nor does
mention of trade names or commercial products constitute endorsement or
recommendation for use.
Document is available to the public through the National
Technical Information Service, Springfield, Virginia 22161.
Publication No. EPA-450/2-77-013
n
-------
EPA-450/2-77-013
SUPPLEMENTAL INFORMATION
TO THE
USER'S MANUAL FOR SINGLE SOURCE (CRSTER) MODEL
A. Computer Program Modifications
Make the following program line modifications to Appendix A of
the User's Manual and check your computer program for conformance.
1. In line MET03040 change ' DAY ',12. to ' DAY ',F4.0, .
2. Delete lines CV002000 and CV002100.
2
3. Insert the following lines :
IMET=1 CV002710
IPTZ=0 CV002720
IF (ELEV(I7,I8).LT.PELEV)GADJ(I8,I7)=0.0 CV022910
4. Remove statement numbers 152, 816, and 515 respectively from
lines CV013200, CV016400 and CV027700.
5. Remove IHC(4), from line CR000800.
6. Remove IHC, and 6,13,18,24, from line CR002700.
7. Delete lines CR008600, CR068800, CR068900, CR071300 and
CR071400.
8. Insert a comma after:
TABLE' in line CR045500,
'/' in line CR061400,
\T_ in line CR061800 and
C0|_ in line CR067000.
9. Remove ,KSTL from lines CV027700 and CR000100.
10. Remove TX3(2),TX4(2), from line BL000200.
To overcome compiler incompatibilities associated with various
computer systems, additional modifications may be required in the
areas of variable/array initialization and DO statement incrementa-
tion parameters.
2
Presently, the model does not consider receptors below ground
elevation of the plant to be at the plant elevation, as described
under section 2.4. on page 2-21 of the User's Manual. The third
insertion under item 3 in the above modifications will cause the
model to be compatible with the noted text on page 2-21. This is
the only modification that can be considered a technical correction
to the model; in a few cases the correction may cause a change in
estimated concentrations.
-------
B. Input Form Modification
The format of the PREP INITIALIZATION CARDS form at the top of
Figure C-l, PREPROCESSOR INPUT DATA CODING FORMS, page C-3, should
be modified to be compatible with the format described in TABLE 4-1,
PREPROCESSOR INITIALIZATION CARD FORMAT, on page 4-11 of the User's
Manual.
C. Clarification of the Highest 50 Concentration Tables
Users should note that the highest 50 maximum daily concentration
table allows only the highest of the 180 24-hour average receptor
concentrations on each day to be a candidate for entry into the
table. However* the highest 50 concentration tables for 1-hour,
3-hour and variable hour time periods allow all of the 180 receptor
concentrations for the specified time period to be candidates for
entry into their respective tables.
If one desires a high 50 maximum daily concentration table where
all of the 180 receptor concentrations are candidates for entry,
then the variable time option should be used. To accomplish this,
the namelist variable IVT should be set equal to 24 and program
line CV011600 on page A-13 should be modified to read:
*OR.IVT.EQ.12.0R.IVT.EQ.24) CV011600
D. Random Number Alternative
The preprocessor program requires a random number subroutine not
supplied with the program. The subroutine calculates numbers to
be used in the development of the hourly randomized vn'nd flow
vectors. If a random number subroutine is not available on a user's
computer or if the user desires to circumvent the possible problems
stated in the '*NOTE:' on page 4-11 of the User's Manual, the
attached set of numbers may be substituted. These numbers were used
to derive the randomized wind flow vectors for the test case. These
numbers were computed by the Sperry Rand Corp. random number generator
subroutine,RANDU, installed on EPA's Univac 1110 computer; they are
commonly used in EPA's applications of the model. The random number
seed was 65549.
Each data line of the attachment contains the day number and 24
integers. The positions of the integers(l-24) identify the hours
of that day for which the integers are to be used in the derivation
of the randomized wind flow vectors. Using the numbers will require
the user to effectively modify the preprocessor program such that
the given integers get transferred into the integer variable IRAND
in line MET01870 to properly coincide with the KHR hour of the IDY
day.
-------
1 528776971580736851814644
2 066401304595314825698912
3 408413032862561435121351
4 002356020117965473131362
5 981469693223249230116426
6 079760262290596b69256060
7 2749703U4876501988562659
8 352919954431386324767248
9 900803784851894758053409
10 574929880657900221739625
11 487878656914981570380053
12 238012267523225303690194
13 327384175705415018402542
14 872460684421751060970114
15 462494361164847071568734
16 031386626156645905340036
17 123871458609028623312253
18 433796327611595933943570
19 119072661583445782481520
20 371817924920746753193832
21 343415319469304228030123
22 654396162827119830914957
23 414435319308434769851114
24 031236172812317226669973
'25 468959058652760405542414
26 692280707860060495820U82
27 565525145674018025002500
28 141861033080295652241522
29 169528787374759841713682
30 96l7426916o3842765872b52
31 826571960820183023035207
32 721310304800330026039656
33 697937009050706428217437
34 59020046879422oll9067279
35 110962138097719606930002
30 036u3Q32496G900971963601
37 2740218382200396oo253oll
38 29025468!.,725o45:>87294978
39 459546595807060723917306
40 380152014558301261130995
41 93450953145o0379204l9015
42 070610727955269901575000
43 3385367u069125282694o929
44 893106143505314349003320
45 460029407720619770315227
46 706421500209262000095040
47 956853942896790463680358
48 06585l95259828>:o33824670
49 777814482231015037482769
50 332032418547ot?1181o42312
51 270527696176240021959004
52 4022020000605722:b603760
53 04400494810030,>o7520<)391
54 0243579b6!j3009n10 :966427
5b 76;Jt50103b'-''(:'J . '76JJM 15911
56 89219767n070-,4952lj214088
57 5246947o3544250ol0789598
50 21769H7793532c?^')Ob230921
59 33014304103^496307143969
-------
60 3321829628H1998959333374
61 933449097112500240729843
62 323798327819007763706259
63 497020131040727388723431
64 4401419b0835984537380281
65 044145492524615353091464
6d 865015050207362406532494
67 622068244854795638845018
68 372646946090684899002112
69 343415319469304228030123
70 654896162827119830914957
71 414435319308434769851114
72 031236172812317226669973
73 468959058652766405542414
74 692280707860060495820062
75 565525145674018025002500
76 141861033880295652241522
77 169528787374759841713682
78 961742b916638427653725L>2
79 826571960820183023035207
80 721310304800830026039656
81 697937009050706428217437
82 59020046879422611^067279
83 1109b2138097719606980862
'84 036630324966908971963631
85 274021838220839660253511
86 298254688725645587294978
87 459546595807866723917306
88 380152014558301261130995
89 934509531456837928419815
90 078610727955269^81575008
91 338536760691252826946929
92 893186143585314349803320
93 460029487728619770315227
94 766421580209262080095848
95 95685394289079046368(1358
96 005851952598288633824670
97 777814482231015037482769
98 3320324185476t>1181o42312
99 278527695176248021950004
100 482282808068572225683760
101 044OH4948180388675200391
102 0243b 795653009010(1960427
103 76658103564359V708885911
104 892197676070049525214088
105 52469476354M250618789598
106 21769877935329^.>05230921
107 330143041030496307143969
108 3321829628M1998959333674
109 933449097112500240729843
110 323798327til900776370b259
111 49702013184072738!l723431
112 440141950835984537330281
113 O441'ib492524615353091464
114 8b501b050207362466532494
115 6220682448b479b638a4b018
116 372646946090684899002112
137 343415319469304228030123
118 654890102827119830914957
119 414435319308434709851134
-------
120 031236172812317226669973
121 468959058652766405542414
122 692280707860060495820082
123 565525145674018025002500
124 141861033880295652241522
125 169528787374759841713682
126 961742691663842765872552
127 826571960820183023035207
128 721310304800830026039656
129 697937009050706428217437
130 590200468794226119067279
131 110962138097719606980862
132 036630324960908971963681
133 274021838220839666253611
134 298254688725645587294978
135 459546595807860723917300
136 380152014558301261130995
137 934509531456837928419815
138 078610727955269981575008
139 338536760691252826946929
140 893186143585314349803320
141 460029487728619770315227
142 766421580209262080095848
143 956853942896790463688358
144 065851952598288633824670
145 777814482231015037482769
146 332032418547661181642312
147 278527696170248021959004
148 482282808068572225683760
149 044084948180388675200391
150 024357956530090100966427
151 76o581035643b9t>768005911
152 892197676070049525214080
153 5246947o3b442bOul07095(;»0
154 2176907 793b329c>''05230921
155 33014304103o4963o7143969
156 3321029o20»19yo(.>b9365 574
157 93344909711250H240729843
158 323798327819007763706259
159 49702013184072738.'1723431
160 440141950835984537380281
161 044145492524615353091464
162 865015050207362466532494
163 622068244854795638845018
164 372640946090684899002112
165 343415319469304228030123
160 65489blo28271l9030914957
167 414435319308434769851114
168 03123617281231722b6r,9>_>73
169 46895(>0586527Ho4 05542414
170 09228070 78t>OUo0495820 082
171 56552L.145674 0180250 02500
172 1418610338H02(:>5652241522
173 169520707374759841713602
174 96l74269l6:>3M7658725b2
175 826571960020-1^02.3035207
176 721310304800030026039650
177 697937009050706428217437
178 59020ii4687942L'6l 19067279
179 1J0962138097710600980062
-------
180 036630524966908971963681
181 274021030220839600253611
182 290254600725045587294978
183 459b4659b80706t> 723917306
184 38015201451.8301261130995
185 934509531450837928419815
186 078010727955269981575008
187 33853676069125282t>946929
188 893186143585314349803320
189 400029487728619770315227
190 766421580209262080095848
191 956853942896790463688358
192 065851952598283633824670
193 777814482231015037482769
194 332032418547061181642312
195 27852769617o2400219b9004
196 482282808068572225603760
197 044084948180388675200391
198 0243b79b6b30H9010096b427
199 766581035643599768885911
200 892197676070049525214088
201 524694763544250618789598
202 217698779353299905230921
203 330143041033496367143969
204 332182962801990959333374
205 933449097112500240729843
206 323798327819007/03706259
207 49702013184072738.3723431
208 440141950835984537300281
209 044145492524615353091464
210 06501505020736246b532494
211 622068244854795638045010
212 372640946090684899002112
213 343415319469304228030123
214 654890162027119830914957
215 4144353193084347698511 l 4
216 03123ol72012317226oo9973
217 468959050ob27oo405042414
218 692200707000060495020002
219 56552514b674018025002bOO
220 1410610330b0295o5224l522
221 169520707374759041713002
222 9617426916o38427650725b2
223 026571960020103023035207
224 7213l03040000300260396bo
225 697937009050706420217437
226 59020;)4607c>422bll90o7279
227 1.1 0962130097 719606900862
220 03663032496o90097lt>o3681
229 274021838220039060253611
230 29825460. J72564bb07294978
231 45954oL»950070oo723917306
232 380152014558301261.130995
23:> 934509531456837928419815
234 078610727955209^01575008
235 330530760691252826^46929
236 093106143505314349003320
237 460029407720619770315227
238 76t>421500209262080^95040
239 95o053l)420967904o3b00350
-------
240 065051952590208633624670
241 777814482231015037482769
242 332032418547661101642312
243 278527696176248021959004
244 402282800060572225683760
245 044084948180300675200391
246 0243b795653009010096o427
247 766581035643599768885911
248 892197676070049525214088
249 524694763544250618709590
250 217698779353299905230921
251 3301430410334963671439o9
252 332182962081998959333374
253 933449097112500240729843
254 323798327819007763706259
255 497020131840727380723431
256 440141950835984537380201
257 044145492524615353091464
258 8b501505020736246o532494
259 622068244854795638045018
260 372646946090684099002112
261 343415319469304228030123
262 654896162027119830914957
263 414435319300434769051114
264 031236l72812317226b69973
265 4689590586527604055424m
266 692200707060060495820082
267 565525145674018025002500
268 141861033800295652241522
269 169520707374759841713682
270 961742691663842765872552
271 826571960820103023035207
272 721310304800030026039056
273 697937009050706420217437
274 590200468794226119067279
275 110962138097719600900862
27b 036&3032496U908971963601
277 274021830220839000253611
278 290254688725645507294P78
279 459540595807000723917306
280 300152014550301261130995
281 934509531456037920419015
282 078610727955269t>0l575000
283 338536760091252026946929
284 893186143505314349003320
285 460029407728619770315227
286 76u421580209202080095840
207 9560539420967904o3o80350
20.3 Oo58bl952rj90200o33024670
209 777014402231015037402769
290 332032410547601101642312
291 278527o9617o248021Q59004
292 48228200d0605722256a37t>0
293 044084948180380075200391
294 024357950530090100960427
295 76o501035o4359<)7680d5911
296 892197o7607004952b214008
297 524694763544250618709598
298 2l7698779353299L'Ob"230921
299 3301430410334963o7143969
-------
300 332182962881998959333374
301 933449097112500240729843
302 323798327819007763706259
303 497020131840727388723431
304 440141950835904537380281
305 044145492524615353091464
306 865015050207362466532494
307 622008244854795638845018
308 372646946090684899002112
309 343415319469304228030123
310 654896162827119830914957
311 414435319308434769851114
312 031236l72812317226o69973
313 46895905865276o405542414
314 692280707860000495820082
315 565525145674018025002500
316 141861033830295652241522
317 169528787374759841713682
318 961742691663842765872552
319 826571960820183023035207
320 721310304800830026039656
321 697937009050706428217437
322 590200468794226119067279
323 110962138097719606980862
324 036630324966908971963681
325 274021838220839666253611
326 298254688725645587294978
327 459546595807866723917306
328 3801520145b83012611309"5
329 934509531456837928419815
330 078610727955269981575008
331 338536760691252826946929
332 893186143585314349803320
33J 460029487728619770315227
334 76o421580209262080095848
335 956653942890790463083350
336 065851952598208633324670
337 7778144822310150374827o9
338 3320324185476oll81642312
339 278527696176248021959004
340 482202000068572225603760
341 044084948180383675200391
342 0243579b65300901009o6427
343 76o581035643599768335911
344 892197676070049525214083
345 52469476354M250618789598
346 2176937 7935329^'.>05230921
347 330143041035496367143969
348 332182962831998959333^74
349 93344909711250 024 072'->H4 3
350 323798327819007763706259
351 497020131840727383723431
352 440141950835984537380281
353 04414b4925246153530914o4
354 8650150b020736246b532494
355 6220o8244054795630^45010
356 372646946090o84899Q02112
357 343415319469304228030123
358 6548961o2827119830914957
359 414435319308434769851114
-------
360 031236172812317226669973
361 468959058652766405542414
362 692280707860060495820082
363 565525145674018025002500
364 141861033880295652241522
365 169528787374759841713682
366 961742691603842765872552
-------
Preface
This publication contains information on and the computer programs
for the Single Source (CRSTER) Model which is based on Gaussian assump-
tions and is applicable to non-reactive pollutants emitted from well-
defined point sources. The Single Source (CRSTER) Model is one of the
atmospheric dispersion models on the User's Network for Applied Modeling
of Air Pollution (UNAMAP) system. The UNAMAP system may be purchased on
magnetic tape from NTIS for use on user's computer system, or UNAMAP may
be accessed through phone lines and time-share computer terminals using
a national teleprocessing network. For information on accessing UNAMAP
contact: Chief, Environmental Applications Branch, Meteorology and
Assessment Division, (MD-80), U.S. Environmental Protection Agency,
Research Triangle Park, MC 27711.
Although attempts are made to thoroughly check out computer programs
with a wide variety of input data, errors are found occasionally. In
case there is a need to correct, revise or update this model, revisions
will be distributed in the same manner as this report. Revisions may be
obtained as they are issued by completing the mailing form on page V. A
user can be assured that the latest version of the Single Source (CRSTER)
Model is on the UNAMAP system.
Comments and suggestions regarding this publication should be
directed to: Chief, Source Receptor Analysis Branch, Monitoring and
Data Analysis Division (MD-14), EPA, Research Triangle Park, NC 27711.
However, technical questions regarding execution of the model may be
handled by telephone call to the Chief, Modeling Support Section, Source
Receptor Analysis Branch in Durham, NC at 919-541-5335 or, using FTS,
629-5335.
-------
Acknowledgements
Although many individuals assisted with time and energy in the
preparation and review of this report, notwithstanding the extensive
modification, testing and validation of the basic model, the credits for
truly significant contributions belong primarily to Russell F. Lee,
Connally E. Mears, Jerome B. Mersch, Gerald K. Moss and Phillip L.
Youngblood.
The extensive assistance, under contract No. 68-02-2506 of Peter H.
Guldberg, Joseph P. Meyers and Kenneth W Wiltsee, under the direction of
Paul Morgenstern, Walden Division of Abcor, Inc., Wilmington, MA, in
writing this Manual is also gratefully acknowledged.
-------
Chief, Environmental Applications Branch
Meteorology and Assessment Division (MD-80)
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
I would like to receive future revisions to the (l&&i't> Manual
the. Single. Source (CRSTER) Model.
Name
Address
ZIP
Telephone (Optional)
-------
CONTENTS
List of Tables ix
List of Figures x
1. MODEL OVERVIEW 1-1
1.1 BACKGROUND AND PURPOSE 1-1
1.2 SCOPE AND USE 1-1
1.2.1 Applications 1-1
1.2.2 Limits of Use 1-2
1.3 BASIC DEFINITIONS AND CONCEPTS 1-3
1.3.1 Basic Definitions 1-3
1.3.2 Concepts 1-4
1.4 SYSTEM DESCRIPTION 1-5
1.4.1 Preprocessor Program 1-5
1.4.2 Single Source (CRSTER) Model 1-7
1.5 SUMMARY OF INPUT DATA 1-7
1.5.1 Preprocessor Program 1-7
1.5.2 Single Source (CRSTER) Model 1-9
1.6 MODEL OUTPUT 1-11
2. TECHNICAL DISCUSSION 2-1
2.1 MODEL FORMULATION 2-1
2.2 METEOROLOGICAL DISPERSION ASSUMPTIONS .... 2-1
2.2.1 The Gaussian Plume Equation 2-2
2.2.2 Basic Assumptions 2-2
2.2.3 Dispersion Coefficients 2-3
2.2.4 Wind Speed Determination 2-4
2.2.5 Effective Stack Height - Plume Rise . 2-4
2.2.6 Limited Mixing 2-8
2.2.7 Treatment of Wind Direction and
Relation to Receptor Network 2-11
2.2.8 Urban - Rural Considerations 2-13
2.3 DEVELOPMENT OF METEOROLOGICAL DATA BY
PREPROCESSOR 2-13
2.3.1 Hourly Mixing Heights 2-14
2.3.2 Hourly Stability Classification . . . 2-16
2.3.3 Wind Direction 2-18
2.3.4 Wind Speed 2-18
2.3.5 Temperature 2-20
2.3.6 Missing Data 2-20
2.4 TERRAIN CONSIDERATIONS 2-21
2.5 RECEPTOR ARRAY 2-21
vii
-------
CONTENTS (continued)
Section Title Page
2.6 CONSIDERATION OF SOURCE CHARACTERISTICS . . . 2-23
2.7 SOURCE-CONTRIBUTION MODE 2-25
2.8 AVERAGING PROCESS AND RANKING PROCEDURE . . . 2-25
2.9 LIMITATIONS 2-27
2.9.1 Steady State 2-27
2.9.2 Terrain Adjustment 2-30
2.9.3 Mixing Height 2-30
2.9.4 Calm Winds 2-30
2.9.5 Other Limitations 2-31
3. DATA REQUIREMENTS AND OUTPUT 3-1
3.1 PREPROCESSOR PROGRAM 3-1
3.1.1 Input Data Requirements 3-1
3.1.2 Output Information 3-3
3.2 SINGLE SOURCE (CRSTER) MODEL 3-4
3.2.1 Input Data Requirements 3-4
3.2.2 Output Information 3-8
4. USER'S GUIDE 4-1
4.1 INTRODUCTION 4-1
4.2 PREPROCESSOR PROGRAM 4-3
4.2.1 Description 4-3
4.2.2 Control Language and Data Deck Setup . 4-4
4.2.3 Input Data Description 4-9
4.2.4 Output Data Description 4-14
4.2.5 Diagnostic Messages 4-16
4.3 SINGLE SOURCE (CRSTER) MODEL 4-16
4.3.1 Description 4-16
4.3.2 Control Language and Data Deck Setup . 4-28
4.3.3 Input Data Description 4-32
4.3.4 Output Data Description 4-40
4.3.5 Diagnostic Messages 4-51
5. REFERENCES 5-1
APPENDIX A - Program Source Listings A-l
APPENDIX B - Example Single Source (CRSTER) Model Runs ... B-l
APPENDIX C - Input Data Forms C-l
APPENDIX D - Applications of the Single Source (CRSTER)
Model to Power Plants: A Summary D-l
APPENDIX E - Sensitivity Analysis of the Single Source
(CRSTER) Model E-l
APPENDIX F - Validation of a Single Source Dispersion Model . F-l
-------
LIST OF TABLES
Table Page
2-1 Wind Speed Profile Exponent 2-5
2-2 Definition of Symbols Used In Briggs' Plume
Rise Equations 2-7
2-3 Modified Gaussian Plume Equations Used In
The Single Source (CRSTER) Model 2-10
2-4 Insolation Class As A Function Of Solar
Altitude For Cloud Cover <_ 5/10 2-17
2-5 Stability Classification Criteria 2-19
4-1 Preprocessor Initialization Card Format 4-11
4-2 Preprocessor Mixing Height Data Card Format . . . 4-12
4-3 Data Record Format For NCC Magnetic Tapes Of
Morning And Afternoon Mixing Heights 4-13
4-4 Preprocessor Output File Record Description . . . 4-15
4-5 Preprocessor Fatal Error Messages 4-17
4-6 Preprocessor Informative Messages 4-20
4-7 Mandatory Name List Variables 4-35
4-8 Optional Name List Variables 4-36
4-9 Name List Variables Required For Specified
Options 4-38
4-10 Single Source (CRSTER) Model Fixed Format
Input Card Descriptions 4-41
4-11 Time Period Identifiers And Corresponding
Hours Of The Day (Local Standard Time) 4-48
4-12 Source Contribution Table Output By The
Single Source (CRSTER) Model 4-50
4-13 Single Source (CRSTER) Model Fatal Error
Messages 4-53
4-14 Single Source (CRSTER) Model Non-Fatal
Error Messages 4-55
4-15 Single Source (CRSTER) Model Informative
Messages 4-56
4-16 Julian Day To Calendar Day Conversion Chart
For Leap Years 4-58
-------
LIST OF FIGURES
Page
1-1 Schematic Of Meteorological Data Preprocessor . . 1-6
1-2 Schematic Of The Single Source (CRSTER) Model . . 1-8
2-1 The Method of Multiple Plume Images Used To
Simulate Plume Reflections In The Single Source
(CRSTER) Model 2-9
2-2 Example of Receptor Network Used In The Single
Source (CRSTER) Model For A South Wind And For
Each Stability Class 2-12
2-3 Determination of Hourly Mixing Heights By The
Single Source (CRSTER) Model Preprocessor .... 2-15
2-4 Illustration Of The Method For Terrain Adjustment
In The Single Source (CRSTER) Model 2-22
4-1 Procedure For Using The Single Source (CRSTER)
Model 4-2
4-2 Preprocessor Program Flow Diagram 4-5
4-3 CRSV Program Flow Diagram 4-22
4-4 Subroutine CRS Flow Diagram 4-24
4-5 Subroutine SIGMA Flow Diagram 4-29
4-6 Subroutine BEH072 Flow Diagram 4-30
4-7 Input Data Deck Setup For The Single Source
(CRSTER) Model 4-33
C-l Preprocessor Input Data Coding Forms C-3
C-2 Single Source (CRSTER) Model Input Data
Coding Form C-5
C-3 Single Source (CRSTER) Model Terrain x
Elevation Coding Form C-6
C-4 Single Source (CRSTER) Model Stack
Data Coding Form C-7
-------
1. MODEL OVERVIEW
1.1 BACKGROUND AND PURPOSE
The Single Source (CRSTER) Model is a computer program designed to
simulate atmospheric dispersion processes for the purpose of calculating
ambient concentration levels of atmospheric contaminants. The model,
originally known as MX24SP, was developed by the Meteorology Laboratory of
the Environmental Protection Agency in 1972. Since that time numerous
modifications and revisions have been added to the computer program to
increase its utility. The purpose of this manual is to document fully the
present version of the Single Source (CRSTER) Model computer program.
1.2 SCOPE AND USE
1.2.1 Applications
The Single Source (CRSTER) Model is appropriate for application to a
wide variety of air pollution problems. It has been utilized primarily in
simulating the behavior of stack effluents from combustion sources. Al-
though designated as the Single Source (CRSTER) Model, the computer program
offers the capability of considering up to 19 stacks simultaneously which
are located at a common site.
The types of application for which the model is well-suited includes:
Stack design studies
Combustion source permit applications
Regulatory variance evaluation
Monitoring network design
Control strategy evaluation for SIPs
Fuel (e.g., coal) conversion studies
Control technology evaluation
Design of supplementary control systems
New source review
Prevention of significant deterioration
The model has been successfully applied previously to these types of prob-
lems. Additional uses will be developed as new issues arise.
1-1
-------
1.2.2 Limits of Use
Successful application of the Single Source (CRSTER) Model is dependent on
recognition of limitations imposed by mathematical assumptions associated
with the model, the availability and form of meteorological data, and the
structure of the computer program. The restrictions which these limitations
impose may be categorized into source, pollutant, and meteorological factors.
Source factors which restrict the application of the model include a
limitation to hot, bouyant stack plume emissions. In particular, the
behavior of momentum-dominated plumes (i.e., cool, high velocity emissions)
are not mathematically represented in the model. The model also does not
simulate the behavior of plumes which are dominated by complex aerodynamic
effects due to flow fields in the vicinity of the stack, nearby tall buildings
or topographic features. Furthermore, only emissions from chimney stacks are
simulated thereby eliminating ill-defined and/or fugitive type emission sources.
Because the computer program considers all stacks at a plant to be co-located,
significant physical separation of stacks can produce erroneous results. Con-
sequently, stack emissions at other sites are not included in the computer
program. Furthermore, the contribution to ambient pollutant concentrations
from background sources is not included in the Single Source (CRSTER) Model;
this must be handled separately.
The Single Source (CRSTER) Model does not consider gravitational effects
and chemical transformation of plume constituents; neither does it incorporate
any depletion mechanisms, e.g., rainout, washout and dry deposition. It assumes
that the pollutants exhibit the dispersion behavior of non-reactive gasses.
The general types of meteorological data required as input to Single Source
(CRSTER) Model can be classified as data on winds, temperature, mixing height,
and sky conditions. The model is unique in this regard due to the last category.
Sky condition parameters (i.e., sky cover and ceiling height) are routinely
reported only by continuously manned meteorological stations. The input data
base must, therefore, be in the proper format and the Preprocessor program
(Section 1.4.1) is the most effective means to produce this.
1-2
-------
1.3 BASIC DEFINITIONS AND CONCEPTS
1.3.1 Basic Definitions
• Air Pollutants - An air pollutant is a substance added to the
atmosphere that causes a deviation from its mean composition and
is present in sufficient quantity to have an adverse effect.
The types of pollutants that are commonly considered in dispersion
modeling applications are those for which National Ambient Air
Quality Standards (NAAQS) have been established. Currently these
are sulfur oxides (S02), total suspended particulates (TSP), car-
bon monoxide (CO), non-methane hydrocarbons (HC), nitrogen dioxide
(N02) and photochemical oxidants (0^). S02 and TSP are emitted
primarily by stationary sources, while transportation sources are
the principal contributors to CO, HC and N02 emissions. Ox are
not directly emitted, but are formed by chemical reactions in the
atmosphere involving HC and N02 emissions and sunlight.
. Dispersion Model - A dispersion model is a mathematical descrip-
tion or representation of the meteorological transport and turbu-
lent diffusion processes that occur in the atmosphere. Generally,
such a model relates pollutant concentrations for specific re-
ceptors and averaging times to emissions from pollutant sources.
This relationship is a function of meteorological conditions and
the spatial relationship between sources and receptors. Thus the
input data requirements for a dispersion model include: meteoro-
logical data, source data and receptor information.
• Stationary Pollutant Emission Sources - Stationary sources are
those for which emissions can be identified with a plant or an
area that remains geographically fixed. The principal emissions
from stationary sources are due to external fuel combustion in
boilers, industrial processing operations, and solid waste in-
cineration.
• Point Sources - A point source is a large, identifiable stationary
source causing emissions of any pollutant at a rate greater than
a defined limit, e.g., 100 tons per year. For a given point
source, the data required as input to a dispersion model are
average emission rates, and physical stack parameters such as
stack height, diameter, exit velocity and exit temperature.
. Receptors - A receptor is a point in space at which the ambient
air quality is being determined. For a given receptor, the data
required as input to a dispersion model are its coordinate loca-
tion relative to that of the pollutant source.
1-3
-------
• Averaging Time - This is the time interval of specific length
over which variations in pollutant concentration at a receptor
are averaged. Because instantaneous concentrations at receptors
exhibit wide temporal variations, time averaged concentrations
provide a more convenient characterization of pollutant levels
at a receptor. The averaging times for dispersion models are
designed to be consistent with the NAAQS and commonly include
the following: 1-hour, 3-hour, 8-hour, 24-hour, and annual
averages.
• Meteorological Data - The meteorological data required for input
to a dispersion model characterize the transport and turbulent
diffusion properties of the atmosphere. The parameters which
are commonly used to characterize these processes are: wind
direction, wind speed, atmospheric stability and mixing height.
Wind direction determines the direction of movement of the plume,
i.e., its advectiori . Wind speed affects the initial dilution
of the pollutant as it is emitted from the stack. Atmospheric
stability determines the rate of turbulent diffusion of the plume
as it moves downwind. Mixing height determines the depth of the
atmosphere through which pollutants can be dispersed in the
vertical. The National Weather Service (NWS) collects most data.
• Advection - Advection is the process of transport of an air parcel
by the velocity field of the atmosphere. Advection is represented
in the model by the azimuth direction of the wind.
* Diffusion - Diffusion in the atmosphere involves mass
exchange between regions in space. Diffusion in the lower
atmosphere is dominated by eddy exchange due to turbulent air
movements, the magnitude of which is generally relatable to
atmospheric stability.
• Turbulence - Turbulence is a state of fluid flow in which the in-
stantaneous velocities exhibit irregular and apparently random
fluctuations, and can in practice only be described by statistical
properties.
• Atmospheric Stability - Stability is an atmospheric property
which characterizes the thermodynamic structure of the atmosphere
in terms of sustaining disturbances. Stability is commonly de-
fined by comparison of the actual temperature lapse rate with the
"dry adiabatic" lapse rate and can be broadly classified into
classes of stable, neutral, and unstable.
1.3.2 Concepts
The Single Source (CRSTER) Model presents a specific set of input data re-
quirements, and yields output data consisting of pollutant concentrations for
a specific averaging time and receptor location. The input data requirements
1-4
-------
can be broadly classified as source factors, site factors, and meteorological
factors. Source factors are related to the location and operating character-
istics of the pollutant emission sources, while site factors include the
effect of terrain and the location of sensitive receptors relative to the
emission sources. Meteorological factors represent the dispersion properties
of the lower atmosphere at any particular time in terms of the joint occur-
rence of specific conditions of atmospheric stability, mixing depth, and
winds.
The Single Source (CRSTER) Model is composed of two parts, namely, a
plume rise model for estimating the effective release height of the pollutant
plume from a point source, and a diffusion model to calculate the downwind
dispersion of the plume. Both of these mathematical models attempt to repre-
sent the actual processes that occur in the atmosphere, under a simplifying
set of assumptions.
The rise of an emission plume above its source height often accounts
for a significant reduction in related ground-level concentrations. A sub-
model in the Single Source (CRSTER) Model calculates the rise of hot, buoyant
plumes under varying meteorological conditions. Representation of the trans-
port and turbulent diffusion of a source plume is accomplished by a Gaussian
plume model. This model provides a representation of the time-averaged
spatial distribution of pollutant concentrations downwind of a continuously
emitting point source. The rate of expansion of the plume is characterized
by a series of empirical dispersion coefficients which are functions of
atmospheric stability and downwind distance from the source.
1.4 SYSTEM DESCRIPTION
1.4.1 Preprocessor Program
The preprocessor program is executed independently and prepares meteoro-
logical data in the tape input format required by the Single Source (CRSTER)
Model. The basic function of this program, as shown by the schematic in Figure
1-1, is to transform surface and upper air meteorological information obtained
from the National Climatic Center (NCC). The operations performed by this
program include:
1-5
-------
r
z: «•
I— l LU
>- K-
1— eC 0
M Of. -Z.
_i ui - o i—
O _l Z 3C
i—i i—i i—i C9
3 =£ X h-i
I— O i—i UI
1-6
-------
• Calculate hourly values for atmospheric stability from meteoro-
logical surface observations,
• Interpolate twice daily mixing height data to hourly values,
• Reformat other meteorological data to be compatible with Single
Source (CRSTER) Model input requirements.
A detailing of the methodologies employed is presented in Section 2.3.
1.4.2 Single Source (CRSTER) Model
A functional schematic of the Single Source (CRSTER) Model is presented
in Figure 1-2. As shown in this schematic, the model accepts the preprocessed
meteorological data tape directly as input, as well as emission source,
receptor site terrain, and program control specifications data. The program
produces printouts and an optional tape of estimated concentrations. Details
of the methodologies employed are presented in Section 2.2.
1.5 SUMMARY OF INPUT DATA
1.5.1 Preprocessor Program
The input data requirements for the Preprocessor program consist of
two categories of meteorological data - hourly surface observations and
twice daily upper air observations. Hourly surface observation data used
are:
Wind direction
Wind speed
Dry bulb temperature
Sky cover
Cloud ceiling height
These data can be obtained by request from the NCC by specifying a magnetic
tape of hourly surface observations in card deck 144 format for the specific
year of analysis and surface observation stations.
Daily upper air observation data input to the Preprocessor are:
• Early morning mixing height
• Afternoon mixing height
1-7
-------
00
z
o
s: —i i—i
•=c o i—
Q; o; ec
C3 I— O
O Z >->
a: o LL.
a. o i—i
o
UJ
0_
OO
—I O LU __l
o a: i— uj
z: ^ oo a
>—• o Q; o
oo oo cj E:
CM
i
a;
Q
O
QL
LU
t—
OO
O
LU
CtL
Z3
O
oo
OO
O UJ
i—i o <:
i/o a: i—
oo => <:
i-. O Q
i- <:
i— i—
00 <
Q
a:
o z
I— i-.
a. =c
UJ C£
c_> Q:
LU LU
o
1-8
-------
The mixing heights are based on NWS upper air soundings at 1200 GMT and 0000
GMT, respectively. These data also can be obtained by request from the NCC
in either magnetic tape or tabular form by specifying "twice daily mixing
heights" for the specific year of analysis and upper air observation
stations.
The surface and upper air stations should be selected for their mete-
orological representativeness of the general area being modeled. Generally
this corresponds to the stations closest to the point sources being modeled
and in the same climatological regime (e.g., coastal, mountainous, plains).
Meteorological data from 1964 are frequently used as input to the Single
Source (CRSTER) Model, although data from any year can be obtained as long
as surface data for each hour are available. The year 1964 is convenient
because it is the last and most recent year for which routinely reported
surface observations are transcribed by NWS on an hourly basis for data
analysis purposes. Although NWS stations still report observations on an
hourly basis on WBAN Form A, only every third hour is keypunched for com-
puter entry and this is not adequate for the Single Source (CRSTER) Model.
Thus, there is a significant increase in the cost of obtaining a magnetic
tape of hourly surface observations from the NCC for the years subsequent
to 1964.
The output of the Preprocessor consists of hourly values of wind speed,
flow vector, randomized flow vector, stability class, mixing heights and
ambient temperature. The flow vector is numerically the 180° complement
of the meteorological wind direction.
1.5.2 Single Source (CRSTER) Model
The input data requirements for the Single Source (CRSTER) Model consist
of four categories of data:
• Meteorological
• Receptor site
• Source emissions
• Program control parameters
1-9
-------
Each of these data categories is discussed separately below.
a. Meteorological Data. All meteorological data required by the Single
Source (CRSTER) Model are contained on the magnetic tape output by the Pre-
processor.
b. Receptor Site Data. The Single Source (CRSTER) Model estimates con-
centrations for each hour at a circular field of 180 receptors, defined by:
(1) five downwind ring distances and (2) 36 radials (one for every 10° of
azimuth). Although the 36 azimuths are fixed, selection of the five ring
distances is made by the user and warrants judicious consideration since the
choice of locations for receptor sites can significantly affect the concentra-
tion estimates.* Once the ring distances have been chosen, terrain elevations
for the receptor coordinates can be obtained from topographic US6S quand-
rangle maps of the area (scale 1:24,000) available from the U.S. Geological
Survey, Washington, D.C. 20242.
c. Source Emissions Data. The emission parameters required by the Single
Source (CRSTER) Model for each point source are:
• Source elevation
• Average stack parameters, for each stack
- pollutant emission rate
- stack gas exit velocity
- stack gas temperature
- stack exit diameter
- physical stack height
If desired, monthly instead of annual average values can be specified for the
pollutant emission rate, stack gas exit velocity and stack gas temperature.
The Single Source (CRSTER) Model permits data for up to 19 different point
sources to be input, but considers all of them to be at the same geographic
location.
*
First approximations for the receptor ring distances can be made with the
aid of the PTMAX program (see Section 2.5).
1-10
-------
d. Program Control Parameters and Options. Program Control input data
required by the Single Source (CRSTER) Model consist of parameters which direct
the execution of the program and the use of optional model features. All
control parameters have automatic default values (see Section 4.3.3.a), except
for the urban/rural indicator which must be specified.
• Urban/Rural Indicator - this parameter specifies whether rural
or urban mixing heights are to be used in the model calculations.
For the urban case, only unstable and neutral stability conditions
are considered, i.e., stable conditions are calculated as if
neutral below the mixing height lid.
• Output Tape Indicator - this parameter specifies whether a mag-
netic tape of all hourly calculated concentrations is to be output.
• Elevation Indicator - this parameter specifies whether the point
sources and receptors are located in flat or uneven terrain.
• Day Indicators - these parameters identify days of the year for
which concentrations are to be calculated.
* Meteorological Output Indicator - this parameter specifies whether
the hourly meteorological data used in the model calculations will
be printed out.
• Variab1e Averaging T1me - this parameter specifies the optional
concentration averaging times available in the model calcula-
tions.
• Monthly Emissions Indicator - this parameter specifies whether
average monthly emission rates will be input for each point source.
* Monthly Stack Gas Exit Velocity Indicator - this parameter speci-
fies whether monthly stack gas exit velocities will be input for
each point source.
• Monthly Stack Gas Temperature Indicator - this parameter specifies
whether monthly stack gas temperatures will be input for each
point source.
• Source Contributions Indicator - this parameter specifies whether
separate source contribution tables will be printed out in lieu
of concentration tables.
1.6 MODEL OUTPUT
The output generated by the Single Source (CRSTER) Model can be classi-
fied into the following four categories:
• Printouts of input data
• Printouts of concentration tables
1-11
-------
• Printouts of source-contributions
• Magnetic tape of calculated concentrations
The type of output produced under each of these categories is described
briefly below and in more detail in Section 4.3.4.a.
Printouts of the input data by the model include the source emission
information, meteorological data for each day concentrations are calculated,
receptor elevation data, and program control parameters and options.
The concentration tables printed out by the model include tables for
each averaging time of the highest and second-highest concentrations at
each receptor point, and tables of the 50 highest concentrations for the
entire year for each averaging time (except the annual).
One optional output of the model is the form of a magnetic tape of
all 1-hour pollutant concentrations calculated during a given model
execution. Another optional output of the model is printouts of the source-
contributions of individual point sources to up to 20 receptor points for
each averaging period, in lieu of concentration tables.
1-12
-------
.CHNICAL DISCUSSION
,.1 SINGLE SOURCE (CRSTER) MODEL FORMULATION
The Single Source (CRSTER) Model is designed to calculate the contributions
from multiple elevated stack emissions at a single plant location to ambient air
quality levels, defined in the same time scales as the National Ambient Air
Quality Standards. The program calculates concentrations for an entire year
and prints out the highest and second-highest 1-hour, 3-hour and 24-hour, as
well as annual mean concentrations at a set of 180 receptors surrounding the
plant. The Single Source (CRSTER) Model is based on a modified form of the Gaussian
plune equation which uses empirical dispersion coefficients and includes
adjustments for plume rise, limited mixing height and elevated terrain. Pol-
lutant concentrations are computed from measured hourly values of wind speed
and direction, and estimated hourly values of atmospheric stability and mixing
height. This chapter discusses the technical basis for the model as currently
formulated.
2.2 METEOROLOGICAL DISPERSION ASSUMPTIONS
The Single Source (CRSTER) Model is based on a modified version of the
Gaussian plume equation. This model assumes a continuous emissions source,
steady-state downwind plume, and a Gaussian distribution for concentrations
of pollutants within the plume in both the crosswind and vertical directions.
Plume rise is estimated using equations proposed by Briggs [1-3] for hot,
buoyant plumes. As the plume expands due to eddy diffusion, it is diluted
and transported downwind by the mean wind. The rate of expansion is char-
acterized by a series of empirical dispersion coefficients which are depen-
dent on the stability of the atmosphere, as determined in studies made by
Pasquill [4] and Gifford [5], and reported by Turner [6,7]. The modifications
made to the basic Gaussian plume equation in the Single Source (CRSTER) Model
include the following:
• Trapping of the plume between the top of the mixing layer
and the ground surface,
2-1
-------
• Uniform vertical mixing of the plume in the mixing U.
beyond a critical distance,
• Exclusion of any ground-level impacts from plumes releasev
above the mixing layer.
Each of the above factors are discussed separately in the sections below.
2.2.1 The Gaussian Plume Equation
The Gaussian plume equation for a continuous emission source gives the
local concentration x of a gas or aerosol at a ground-level location (x,y)
by the following expression:
x (x,y) =
exp
TT ayaz u
* O I — •• j
^ y/_
exp
" i /" nf
rt I I
t \ (7 /
(2-1)
where the wind is advecting the plume at a speed u along the x-axis and dis-
persing it along the crosswind and vertical direction with diffusion coeffi-
cients a and a , respectively. The pollutant emission from the source is
at a uniform rate Q and is assumed to be released at an "effective stack
height" H (see Section 2.2.5). It is assumed that complete reflection of the
plume takes place at the earth's surface, i.e., there is no atmospheric trans-
formation or deposition at the surface (see Section 2.2.6). The concentration
X is an average over the time interval represented by a and a . The Single
Source (CRSTER) Model calculates short-term concentrations and uses these
directly as 1-hour average concentrations without consideration of plume
history, i.e., each 1-hour period is completely independent. Equation (2-1) is
valid for any consistent set of units, however those commonly employed are x in
(g m" ), Q in (g s" ), u in (m s ), and x, y, H, a and az in (m).
2.2.2 Basic Assumptions
The Gaussian plume equation (2-1) is a solution to the simplified con-
servation of mass equation assuming nonzero wind speed and constant eddy
diffusivities along the principal axes. The use of a single wind speed and
2-2
-------
constant direction in the Gaussian plume equation reflects the assumption
that the horizontal wind field is homogeneous, and that the effects of di-
rectional wind shear in the atmospheric boundary layer are not considered.
The effects of surface friction in reducing wind speeds near ground-level
are taken into account, however, by the Single Source (CRSTER) Model (see
Section 2.2.4).
The assumptions incorporated in the Gaussian plume equation and the
Single Source (CRSTER) Model can be summarized as follows:
• Steady-State Conditions - ideal gas, continuous uniform emission
rate, homogeneous horizontal wind field, representative hourly
mean wind velocity, no directional wind shear in the vertical,
infinite plume, no plume history.
• Pollutant Characteristics - the pollutant emitted is a stable
gas or aerosol which remains suspended in the air and partici-
pates in the turbulent movement of the atmosphere, none of the
material is removed as the plume advects and diffuses downwind
and there is complete reflection at the ground.
• Gaussian Distribution - the pollutant material within the plume
takes on a Gaussian distribution in both the horizontal cross-
wind and vertical directions, described by empirical dispersion
parameters a and a .
2.2.3 Dispersion Coefficients
The empirical dispersion coefficients, a and a , used in the Single
Source (CRSTER) Model are those suggested by Pasquill [4] and Gifford [5],
and reported by Turner [6,7]. Values for cr and a are represented as a
function of downwind distance from the emissions source and the stability of
the atmosphere. These values are representative for a sampling time of up
to about one hour and were developed based on aerometric measurements taken
in open, level to gently rolling country. Implementation of these param-
eters in the Single Source ,(CRSTER) Model is accomplished by using piece-
wise equations (dependent on stability) which approximate the dispersion
curves between specific downwind distances.
2-3
-------
Atmospheric stability is determined indirectly from the amount of
incoming solar radiation at the surface (insolation), and the wind speed.
Pasquill suggested a six category classification scheme from A for extremely
unstable to F for moderately stable, based on the range of these two param-
eters. Because solar radiation is not a widely measured parameter, Turner [6]
developed an objective classification method based on cloud cover, ceiling
height, and solar elevation (see Section 2.3.2). The Single Source (CRSTER)
Model Preprocessor calculates the stability classification by this method for
each hour from the recorded meteorological observations.
2.2.4 Wind Speed Determination
The wind speed required for input to the Single Source (CRSTER) Model is
considered to be representative of the conditions throughout the vertical
height interval in which the plume is dispersing. The wind at the stack
elevation is commonly used as an approximation to this condition. Because
the wind is generally measured near 7 meters by the National Weather Service
(NWS), an adjustment is made in the model by the following power law relation-
ship:
u = UQ (h/7)p (2-4)
where
u = hourly wind speed at stack height (m s )
u = hourly wind speed near 7m above the ground (m s~ )
h = stack height (m)
p = wind profile exponent
The profile exponent p is a function of stability and has the values given
in Table 2-1. The adjusted wind speed is used by the model to calculate
plume rise and dilution.
2.2.5 Effective Stack Height - Plume Rise
The effective height of emission (H) used in the Gaussian plume
equation (2-1) is defined as the sum of the physical stack height (hg) and
the plume rise (Ah). Estimates of plume rise are required to predict
2-4
-------
TABLE 2-1
WIND SPEED PROFILE EXPONENT
Pasquill Stability Class Wind Speed Profile Exponent, P
A = extremely unstable 0.10
B = moderately unstable 0.15
C = slightly unstable 0.20
D = neutral 0.25
E = slightly stable 0.30
F = moderately stable 0.30
2-5
-------
the dispersion of continuous gaseous emissions possessing buoyancy. The
rise of emission plumes above their source release height often accounts
for a significant reduction in related ground-level concentrations.
Plume rise in the Single Source (CRSTER) Model is estimated using equa-
tions proposed and later modified by Briggs [1,2,3]. These equations are based
on the assumption that plume rise depends on the inverse of the mean wind speed
and is directly proportional to the 2/3 power of the downwind distance from
the source, with different equations specified for the neutral -unstable con-
ditions and the stable conditions. Only the final plume rise as predicted
by Briggs is used in the Single Source (CRSTER) Model. Briggs' plume rise
equations are detailed below, where all symbols are defined in Table 2-2.
• For unstable or neutral atmospheric conditions, the downwind
distance of final plume rise is x-? = 3.5 x*, where
x* = 14 F5/8, when F < 55 m4 s"3 (2-5)
x* = 34 F2/5, when F >_ 55 m4 s"3. (2-6)
The final plume rise under these conditions is
Ah = 1.6 F1/3 (3.5 x*)2/3 u'1. (2-7)
For stable atmospheric conditions, the downwind distance of
where
t or sae amosperc conons,
final plume rise is x^ = TT u s-1/2,
s = g 86/8Z T"1. (2-8)
The plume rise is
fzA [F/(u s)]1/3s for windy conditions (2-9)
|5 F1/4 s"3//8, for near-calm conditions (2-10)
For the final plume rise under these conditions, the smaller
of the values estimated by (2-9) and (2-10) is used.
2-6
-------
TABLE 2-2
DEFINITION OF SYMBOLS USED IN BRIGGS' PLUME RISE EQUATIONS
Symbol Definition Units
_2
g gravitational acceleration 9.8 m s
d stack inside diameter at top m
4 -3
F buoyancy flux parameter m s
[g vs (d/2)2 (Ts - T/TS)]
x* distance at which atmospheric turbulence m
begins to dominate entrainment
Ah plume rise above stack top m
x downwind distance from the source m
T ambient air temperature °K
T stack gas temperature °K
u mean wind speed from stack top to plume top m s~
v stack gas exit velocity m s"
99/92 vertical potential temperature gradient °K m~
from stack top to plume top
_p
s restoring acceleration per unit vertical s
displacement for adiabatic motion in the
atmosphere, a stability parameter
2-7
-------
The final plume rise given by these formulae does not take cognizance
of "negative" buoyancy due to cold plumes, or aerodynamic effects from
flow fields around the stack or nearby tall buildings and prominent terrain.
The final plume height used by the Single Source (CRSTER) Model does not
follow changes in terrain height,as described further in Section 2.4.
2.2.6 Limited Mixing
Turbulent mixing and vertical diffusion of a plume is often limited
by the existence of a stable layer of air aloft, i.e., an inversion layer.
The effects of limited mixing (or plume "trapping") on plume dispersion
are incorporated into the Single Source (CRSTER) Model by the assumption that
the plume is completely reflected at the mixing height (L), as well as the
ground. Since multiple reflections are possible as shown in Figure 2-1,
trapping is simulated using the method of multiple images proposed by
Bierly and Hewson [8]. In this procedure, each reflection is represented
by an "image plume" from an imaginary source with a "stack height" equal
to the vertical distance travelled by the plume "edge" to the point of
ground reflection. The reflections between the mixing height (L) and the
ground can be represented by the convergent infinite series of Gaussian
plume terms given in Equation (2-11), see Table 2-3. Several simplifications
are used in the Single Source (CRSTER) Model in implementing this concept.
Because the infinite series in Equation (2-11) converges rapidly, the summa-
tion is limited to the sum of -k to +k terms until the additional contribu-
tion of the next'two terms [-(k+1) and (k+1)] is less than 0.01 snf3. This
means the relative concentration is calculated as a sum of a series of terms.
When the addition of the next term in the stries contributes to an increase
of less than 0.01, the summation is stopped. In any case, k is limited to a
maximum of 45. Also beyond the downwind distance where a = 1.6 L, such
reflections result in a nearly uniform vertical distribution of concentra-
tion. Thus, an appropriate simplification is introduced in the computations,
see Equation (2-12). Another assumption is that whenever the plume center-
line (H) is above the mixing height (L) at a given receptor location, there
is no contribution from the plume at that receptor.
2-8
-------
A \V plume \
2L-H V
Mixing Height (L)
-------
o
CO
I
CM
I
I
CM
CM
CM
CO
CM
CM
-|CM
0.
X
O)
-------
2. TECHNICAL DISCUSSION
2.1 SINGLE SOURCE (CRSTER) MODEL FORMULATION
The Single Source (CRSTER) Model is designed to calculate the contributions
from multiple elevated stack emissions at a single plant location to ambient air
quality levels, defined in the same time scales as the National Ambient Air
Quality Standards. The program calculates concentrations for an entire year
and prints out the highest and second-highest 1-hour, 3-hour and 24-hour, as
well as annual mean concentrations at a set of 180 receptors surrounding the
plant. The Single Source (CRSTER) Model is based on a modified form of the Gaussian
plume equation which uses empirical dispersion coefficients and includes
adjustments for plume rise, limited mixing height and elevated terrain. Pol-
lutant concentrations are computed from measured hourly values of wind speed
and direction, and estimated hourly values of atmospheric stability and mixing
height. This chapter discusses the technical basis for the model as currently
formulated.
2.2 METEOROLOGICAL DISPERSION ASSUMPTIONS
The Single Source (CRSTER) Model is based on a modified version of the
Gaussian plume equation. This model assumes a continuous emissions source,
steady-state downwind plume, and a Gaussian distribution for concentrations
of pollutants within the plume in both the crosswind and vertical directions.
Plume rise is estimated using equations proposed by Briggs [1-3] for hot,
buoyant plumes. As the plume expands due to eddy diffusion, it is diluted
and transported downwind by the mean wind. The rate of expansion is char-
acterized by a series of empirical dispersion coefficients which are depen-
dent on the stability of the atmosphere, as determined in studies made by
Pasquill [4] and Gifford [5], and reported by Turner [6,7]. The modifications
made to the basic Gaussian plume equation in the Single Source (CRSTER) Model
include the following:
• Trapping of the plume between the top of the mixing layer
and the ground surface,
2-1
-------
• Uniform vertical mixing of the plume in the mixing layer
beyond a critical distance,
• Exclusion of any ground-level impacts from plumes released
above the mixing layer.
Each of the above factors are discussed separately in the sections below.
2.2.1 The Gaussian Plume Equation
The Gaussian plume equation for a continuous emission source gives the
local concentration x of a gas or aerosol at a ground-level location (x,y)
by the following expression:
X (x,y) =
exp
(2-1)
where the wind is advecting the plume at a speed u along the x-axis and dis-
persing it along the crosswind and vertical direction with diffusion coeffi-
cients a and a , respectively. The pollutant emission from the source is
at a uniform rate Q and is assumed to be released at an "effective stack
height" H (see Section 2.2.5). It is assumed that complete reflection of the
plume takes place at the earth's surface, i.e., there is no atmospheric trans-
formation or deposition at the surface (see Section 2.2.6). The concentration
X is an average over the time interval represented by a and a . The Single
Source (CRSTER) Model calculates short-term concentrations and uses these
directly as 1-hour average concentrations without consideration of plume
history, i.e., each 1-hour period is completely independent. Equation (2-1) is
valid for any consistent set of units, however those commonly employed are x in
(g m" ), Q in (g s" ), u in (m s" ), and x, y, H, a and az in (m).
2.2.2 Basic Assumptions
The Gaussian plume equation (2-1) is a solution to the simplified con-
servation of mass equation assuming nonzero wind speed and constant eddy
diffusivities along the principal axes. The use of a single wind speed and
2-2
-------
constant direction in the Gaussian plume equation reflects the assumption
that the horizontal wind field is homogeneous, and that the effects of di-
rectional wind shear in the atmospheric boundary layer are not considered.
The effects of surface friction in reducing wind speeds near ground-level
are taken into account, however, by the Single Source (CRSTER) Model (see
Section 2.2.4).
The assumptions incorporated in the Gaussian plume equation and the
Single Source (CRSTER) Model can be summarized as follows:
• Steady-State Condition " gas, continuous uniform emission
rate, homogeneous nor ' field, representative hourly
mean wind velocity, no direc^iv..^ wind shear in the vertical,
infinite plume, no plume history.
• Pollutant Characteristics - the pollutant emitted is a stable
gas or aerosol which remains suspended in the air and partici-
pates in the turbulent movement of the atmosphere, none of the
material is removed as the plume advects and diffuses downwind
and there is complete reflection at the ground.
• Gaussian Distribution - the pollutant material within the plume
takes on a Gaussian distribution in both the horizontal cross-
wind and vertical directions, described by empirical dispersion
parameters a and a .
2.2.3 Dispersion Coefficients
The empirical dispersion coefficients, a and az, used in the Single
Source (CRSTER) Model are those suggested by Pasquill [4] and Gifford [5],
and reported by Turner [6,7]. Values for cr and a are represented as a
function of downwind distance from the emissions source and the stability of
the atmosphere. These values are representative for a sampling time of up
to about one hour and were developed based on aerometric measurements taken
in open, level to gently rolling country. Implementation of these param-
eters in the Single Source JCRSTER) Model is accomplished by using piece-
wise equations (dependent on stability) which approximate the dispersion
curves between specific downwind distances.
2-3
-------
Atmospheric stability is determined indirectly from the amount of
incoming solar radiation at the surface (insolation), and the wind speed.
Pasquill suggested a six category classification scheme from A for extremely
unstable to F for moderately stable, based on the range of these two param-
eters. Because solar radiation is not a widely measured parameter, Turner [6]
developed an objective classification method based on cloud cover, ceiling
height, and solar elevation (see Section 2.3.2). The Single Source (CRSTER)
Model Preprocessor calculates the stability classification by this method for
each hour from the recorded meteorological observations.
2.2.4 Wind Speed Determination
The wind speed required for input to the Single Source (CRSTER) Model is
considered to be representative of the conditions throughout the vertical
height interval in which the plume is dispersing. The wind at the stack
elevation is commonly used as an approximation to this condition. Because
the wind is generally measured near 7 meters by the National Weather Service
(NWS), an adjustment is made in the model by the following power law relation-
ship:
u = UQ (h/7)P (2-4)
where
u = hourly wind speed at stack height (m s )
u = hourly wind speed near 7m above the ground (m s~ )
h = stack height (m)
p = wind profile exponent
The profile exponent p is a function of stability and has the values given
in Table 2-1. The adjusted wind speed is used by the model to calculate
plume rise and dilution.
2.2.5 Effective Stack Height - Plume Rise
The effective height of emission (H) used in the Gaussian plume
equation (2-1) is defined as the sum of the physical stack height (hg) and
the plume rise (Ah). Estimates of plume rise are required to predict
2-4
-------
TABLE 2-1
WIND SPEED PROFILE EXPONENT
Pasquill Stability Class
Wind Speed Profile Exponent, P
A = extremely unstable
B = moderately unstable
C = slightly unstable
D = neutral
E = slightly stable
F = moderately stable
0.10
0.15
0.20
0.25
0.30
0.30
2-5
-------
the dispersion of continuous gaseous emissions possessing buoyancy. The
rise of emission plumes above their source release height often accounts
for a significant reduction in related ground-level concentrations.
Plume rise in the Single Source (CRSTER) Model is estimated using equa-
tions proposed and later modified by Briggs [1,2,3]. These equations are based
on the assumption that plume rise depends on the inverse of the mean wind speed
and is directly proportional to the 2/3 power of the downwind distance from
the source, with different equations specified for the neutral-unstable con-
ditions and the stable conditions. Only the final plume rise as predicted
by Briggs is used in the Single Source (CRSTER) Model. Briggs1 plume rise
equations are detailed below, where all symbols are defined in Table 2-2.
• For unstable or neutral atmospheric conditions, the downwind
distance of final plume rise is x* = 3.5 x*, where
x* = 14 F5/8, when F < 55 m4 s"3 (2-5)
x* = 34 F2/5, when F >_ 55 m4 s"3. (2-6)
The final plume rise under these conditions is
Ah = 1.6 F1/3 (3.5 x*)2/3 u"1. (2-7)
• For stable atmospheric conditions, the downwind distance of
final plume rise is x^ = TT u S"V2, where
s = g 90/3z T"1. (2-8)
The plume rise is
("2.4 [F/(u s)]1/3, for windy conditions (2-9)
|5 F1/4 s~3/8, for near-calm conditions (2-10)
For the final plume rise under these conditions, the smaller
of the values estimated by (2-9) and (2-10) is used.
2-6
-------
TABLE 2-2
DEFINITION OF SYMBOLS USED IN BRIGGS' PLUME RISE EQUATIONS
Symbol Definition Units
_2
g gravitational acceleration 9.8 m s
d stack inside diameter at top m
4 -3
F buoyancy flux parameter m s
[9 vs (d/2)2 (Ts - T/Ts)]
x* distance at which atmospheric turbulence m
begins to dominate entrainment
Ah plume rise above stack top m
x downwind distance from the source m
T ambient air temperature °K
T stack gas temperature °K
u mean wind speed from stack top to plume top m s~
v stack gas exit velocity m s~
99/9Z vertical potential temperature gradient °K m
from stack top to plume top
_2
s restoring acceleration per unit vertical s
displacement for adiabatic motion in the
atmosphere, a stability parameter
2-7
-------
The final plume rise given by these formulae does not take cognizance
of "negative" buoyancy due to cold plumes, or aerodynamic effects from
flow fields around the stack or nearby tall buildings and prominent terrain.
The final plume height used by the Single Source (CRSTER) Model does not
follow changes in terrain height,as described further in Section 2.4.
2.2.6 Limited Mixing
Turbulent mixing and vertical diffusion of a plume is often limited
by the existence of a stable layer of air aloft, i.e., an inversion layer.
The effects of limited mixing (or plume "trapping") on plume dispersion
are incorporated into the Single Source (CRSTER) Model by the assumption that
the plume is completely reflected at the mixing height (L), as well as the
ground. Since multiple reflections are possible as shown in Figure 2-1,
trapping is simulated using the method of multiple images proposed by
Bierly and Hewson [8]. In this procedure, each reflection is represented
by an "image plume" from an imaginary source with a "stack height" equal
to the vertical distance travelled by the plume "edge" to the point of
ground reflection. The reflections between the mixing height (L) and the
ground can be represented by the convergent infinite series of Gaussian
plume terms given in Equation (2-11), see Table 2-3. Several simplifications
are used in the Single Source (CRSTER) Model in implementing this concept.
Because the infinite series in Equation (2-11) converges rapidly, the summa-
tion is limited to the sum of -k to +k terms until the additional contribu-
tion of the next' two terms [-(k+1) and (k+1)] is less than 0.01 snf3. This
means the relative concentration is calculated as a sum of a series of terms.
When the addition of the next term in the series contributes to an increase
of less than 0.01, the summation is stopped. In any case, k is limited to a
maximum of 45. Also beyond the downwind distance where a = 1.6 L, such
reflections result in a nearly uniform vertical distribution of concentra-
tion. Thus, an appropriate simplification is introduced in the computations,
see Equation (2-12). Another assumption is that whenever the plume center-
line (H) is above the mixing height (L) at a given receptor location, there
is no contribution from the plume at that receptor.
2-8
-------
Mixing Height (L)
FIGURE 2-1
THE METHOD OF MULTIPLE PLUME IMAGES USED TO SIMULATE PLUME
REFLECTIONS IN THE SINGLE SOURCE MODEL
2-9
-------
Q
UJ
OO
O
OO
c5
OO
CM
UJ
-u
UJ C£
UJ -^
OO
00
CO
CM
I
CM
CM
N
r-|cVJ
Q.
X
O)
H3
•O
re
—i
v|
I
CM
Q. Q.
X X
Ol CD
cy
M
to o-
D
t=
M
t)
1 t=
^sl 0
n M ii
X XX
2-10
-------
Hourly values of the mixing height, for both rural and urban regimes,
are estimated by the Preprocessor program by the method described in
Section 2.3.1. The Single Source (CRSTER) Model assumes the mixing height
is constant above any receptor location whose elevation is above plant base
elevation and below the top of the stack. Terrain adjustments are discussed
in detail in Section 2.4.
2.2.7 Treatment of Hind Direction and Relation to Receptor Network
The Single Source (CRSTER) Model calculates pollutant concentrations for
a radial network of 180 receptor points, as shown in Figure 2-4. This array
consists of five downwind distances along each 10° azimuth direction (see
Section 2.5). The Single Source (CRSTER) Model calculates concentrations at
these receptor points relative to the azimuth wind direction as determined
by the Preprocessor program. This wind direction includes a random variation
(-4 to +5°) superimposed on the value recorded by the NWS. Off-centerline
concentrations are computed for those receptors which fall within the stability-
dependent widths as defined in Figure 2-2. For example, under extremely un-
stable conditions (Class A), off-center!ine concentrations are computed up to
±50° from the plume centerline, whereas for very stable conditions (Class F),
the concentrations are calculated only to ±20°. The computation of off-
centerline concentrations requires a value for the lateral distance of the
receptor (y) in Equation (2-1). This value is estimated fay the arc length
along the ring for the angular difference between the randomized wind direction
and the receptor azimuth. Thus,
y -arc*= r (a - 6) (2-14)
where
r = ring distance of receptor point
9 = azimuth of receptor point
a = azimuth of randomized wind direction.
Concentrations at off-centerline receptors within the limits illustrated in
Figure 2-2 are computed except when the exponential y term in Equation (2-1)
-77
is essentially zero, i.e., less than 2 x 10 .
* The difference between using arc length and straight line is quite small.
2-11
-------
-ou
•* A.D
-40°
s «"30° .. D
X \ V .-r-20° E.F +
+OU
+40°> 1
+30° , 1 y
1
Hourly
Measured Wind
Direction
FIGURE 2-2
EXAMPLE OF RECEPTOR NETWORK USED IN THE SINGLE SOURCE (CRSTER)
MODEL FOR A SOUTH WIND AND FOR EACH STABILITY CLASS
2-12
-------
2.2.8 Urban-Rural Considerations
The principal difference between dispersion coefficients in rural and
urban environments is associated with the occurrence of the nocturnal, ground-
based temperature inversion. On calm, clear nights, radiational cooling can
produce such an inversion, and hence stable atmospheric conditions, in a rural
environment. Such inversions do not occur, though, in urban areas, due primarily
to the influence of a city's larger surface roughness and the release of stored
heat from structural surfaces, i.e., the urban heat island effect. Thus, stable
atmospheric conditions do not occur near the ground in urban areas on calm, clear
nights.
The Single Source (CRSTER) Model accounts for these effects in both the
choice of dispersion coefficients and mixing heights. If an urban applica-
tion is indicated, stability categories E and F default to category D for the
purpose of determining a and a (see Section 2.3.2). Separate sets of hourly
mixing height data, for urban and rural environments, are input to the model
and it chooses between these, depending on the conditions indicated (see
Section 2.3.1).
2.3 DEVELOPMENT OF METEOROLOGICAL DATA BY PREPROCESSOR
The Preprocessor program prepares surface and upper air meteorological
data for input to the Single Source (CRSTER) Model. As discussed in Sections
1.4 and 1.5, the program is applied to two meteorological data bases. The
first of these is the set of hourly surface observations in card deck 144
format, while the second is the set of twice-daily mixing heights based on
upper air observations. Both data bases are obtained from the National
Climatic Center.
Because the 24-hour period for the Single Source (CRSTER) Model starts
with the 1-hour period ending 0100 LST (Local Standard Time), whereas NWS
data begin each day with the observation reported at 0000 LST, the Prepro-
cessor skips the first hourly record of data in a year when reading NWS
meteorological tapes. Sequential reading of the remaining data automatically
makes this adjustment for each succeeding day. Meteorological data for the
2-13
-------
last hour in a year are assumed equal to that for the next to the last hour,
with the exception of the mixing height, which is actually estimated for
2400 LST on the last day of the year.
2.3.1 Hourly Mixing Heights
Hourly values of the mixing height are determined by the Preprocessor
from: 1) twice daily estimates of mixing height; 2) the local standard time
of sunrise and sunset; and 3) hourly estimates of stability. The first are
based on the method of Holzworth [9] and are available from the National
Climatic Center (NCC), Asheville, NC. The second are computed by the Pre-
processor from input data on the date, latitude, longitude, and time zone,
using well known earth-sun relationships (e.g., see Sellers [10]. The
third are discussed in Section 2.3.2. Two different interpolations are
used by the Preprocessor; one is for application to rural sites while the
other is for urban locations. Both sets of values are input to the Single
Source (CRSTER) Model, which chooses between them depending upon the option
specified by the user.
The method by which hourly mixing heights are determined is depicted
schematically in Figure 2-3. The procedure uses values for the maximum
mixing height (MAX) from the previous day (i-1), the computation day (i)
and the following day (i+1) and for the minimum mixing height (MIN) for
days (i) and (i+1). For urban sites between midnight and sunrise under
neutral stability (i.e., Class D), the interpolation is between MAX. , at
sunset and MAX.. at 1400 1ST. Under stable conditions (i.e., Class E or F),
the value for MIN^ is used. During the hours between sunrise and 1400 LST, if
the stability was classified as neutral in the hour before sunrise, the earlier
interpolation between MAX- -, and MAX- is continued; if the hour before sunrise
was classified as stable, the interpolation is between MIN. and MAX^. For
the period 1400 LST to sunset, the value for MAX, is used. During the hours
between sunset and midnight under neutral stability, the interpolation is
between MAX.. at sunset and MAX^-j at 1400 LST the next day; if the stability
is stable, the interpolation is between MAX- at sunset and MIN.+-, at midnight.
2-14
-------
CD
Z
i—i
O
o
I
o
I—I
I
cu
o
jj
oo
I—
CD
I—I
LU
CD
X
o
o
1H9I3H
1H9I3H
-------
For rural sites between midnight and sunrise, the interpolation is
between MAX., at sunset and MAX. at 1400 LSI. During the hours between
sunrise and 1400 LSI, if stability was classified as neutral in the hour
before sunrise, the earlier interpolation between MAX. , and MAX. is con-
tinued; if the hour before sunrise was classified as stable, the interpola-
tion is between 0 and MAX^. For the period 1400 LSI to sunset, the value
for MAX. is used. During sunset to midnight, the interpolation is between
MAXj at sunset and MAXi+] at 1400 LSI the next day.
2.3.2 Hourly Stability Classification
One of seven stability classes is determined from meteorological data
for each hour by the Preprocessor. The first six of these categories (1-6)
correspond to Pasquill's classifications (A-F). The seventh category cor-
responds to the 'dashes' in Pasquill's original classification [4] and repre-
sents the existence of a strong, ground-based nocturnal temperature inversion
and non-definable wind flow conditions. The Single Source (CRSTER) Model
restricts changes in stability to one class per hour, and in the urban mode,
treats categories 5, 6 and 7 as category 4 for the purpose of estimating
o and a . In the rural mode for hours classified as stability 7, the
Single Source (CRSTER) Model does not compute concentrations (i.e., does
not permit the plume to reach the ground).
Initially, the Preprocessor determines the hour angle of the sun and
the times of sunrise and sunset from the day number, longitude, and time
zone to permit differentiation of daytime and nighttime cases by the
method of Woolf [18]. For daytime cases, the appropriate insolation
class is selected by means of the Turner [6] objective method using cloud
cover, ceiling height, and solar elevation as indicators. This method
assigns net radiation indices, using the criteria shown in Table 2-4, for
cases where the total cloud cover <_ 5/10. If the cloud cover > 5/10, but
less than 10/10 (overcast), the insolation class is reduced by one cate-
gory when the ceiling height is between 7,000 and 16,000 ft. and by two
categories for ceilings less than 7,000 ft. For a cloud cover of 10/10,
the insolation class is reduced by one category when the ceiling height is
greater than 16,000 ft. and by two categories for ceilings between 7,000 and
2-16
-------
TABLE 2-4
INSOLATION CLASSES AS A FUNCTION OF SOLAR
ALTITUDE FOR CLOUD COVER ^5/10*
Solar elevation Insolation Net Radiation
angle (a) class Index
0° 5/10 cloud cover, see Section 2.3.2
2-17
-------
16,000 ft. For ceilings below 7,000 ft and 10/10 cloud cover (i.e., over-
cast), a net radiation of 0 is defined and neutral stability is specified.
With the exception of the 10/10 low cloud cases, the net radiation index is
never reduced below 1, or "weak". The final stability category is selected
from Table 2-5 and Turner's insolation classes.
2.3.3 Wind Direction
Hourly data on wind direction input to the Preprocessor are tested for
calms (coded as 0° by the NWS; note that a north wind is coded as 360°),
and if present, the wind direction from the previous hour is substituted.
If a calm occurs for the first observation of the year read by the Preprocessor,
a default direction specified by the user is used instead.
The Preprocessor converts all wind directions to flow vectors by shifting
input wind directions by 180°. This change is made because data reported by
NWS indicate the direction from which the wind is blowing, whereas the Single
Source (CRSTER) Model assumes input wind data specifying the direction towards
which the wind is blowing. These flow vectors then are randomized by adding
a random integer number of azimuth degrees between -4° and +5°. This pro-
cedure is applied to remove the directional bias introduced into the NWS data
because they are reported to only the nearest 10°. The randomization gives
the flow vector an equal probability of occurring anywhere within the 10°
sector and so incorporates the natural fluctuations of this parameter.
2.3.4 Wind Speed
Hourly wind speed data are converted from the NWS reporting units of
(kts) to the units of meters per second (m s ) used in the Single
Source (CRSTER) Model. The multiplicative conversion factor is 0.51444.
A/ind speeds
the model.
knots (kts) to the units of meters per second (m s ) used in the Single
The multiplicative conversion factor is 0.51
Wind speeds below 1.0 m s (calms included) are raised to 1.0 m s -jn
2-18
-------
0
O TJ
i— 13
\ O
c
i—i
QJ
re
a
OT
cu
•p
re
s-
a>
-o
o
01
c
o
i — i — c\jcvic\jc\Joororo
•g^
CU QJ 00
U Q-4->
re i/i o
M- C
O i— CM
i— CMro«3-LOi£>r».ooo>i— i— i—
vl A|
2-19
-------
2.3.5 Temperature
Hourly ambient temperature data are converted from the NWS reporting
units of degrees Fahrenheit (°F) to the units of degrees Kelvin (°K) used
in the Single Source (CRSTER) Model.
2.3.6 Missing Data
Meteorological data are input to the Preprocessor in two forms: (1)
hourly NWS observations are read, from magnetic tape, one record per hour,
and (2) daily minimum and maximum mixing heights are read from punched
card input. Data may be missing from either of these sources for certain
hours, days, or even months. The Preprocessor, however, performs checks
for missing data records for only the magnetic tape of surface data. Be-
cause similar checks are not performed on the mixing height data, it is the
user's responsibility to ensure that the information on these cards is com-
plete and properly ordered.
Data checks on the meteorological data tape begin with the station
identification number contained in each hourly record. If this number ever
fails to match the station identification number specified by the user, an
error message is printed and the program terminates. Next, the year, month,
day and hour numbers contained in each data record are checked for consis-
tency in their order and completeness. If one or more hourly records are
found to be missing, or are out of order, then the program terminates after
printing appropriate error messages. Because the data check procedures
used in the Preprocessor can only sense missing data when the entire record
is absent, if only one meteorological parameter is missing, e.g., wind speed,
the Preprocessor will not flag such blank data entries and the blanks will
be interpreted as zeros. It is recommended, therefore, that the user per-
form additional screening for missing data either manually or by other
computer programs.
2-20
-------
2.4 TERRAIN CONSIDERATIONS
The Single Source (CRSTER) Model is an uneven terrain model that takes
into account certain changes in ground elevation between the point of source
emissions (the plant) and the surrounding 180 receptor points. The method
used in the Single Source (CRSTER) Model for making terrain adjustments is
shown in Figure 2-4. This schematic shows that the height of the plume
centerline (H) is lowered by an amount equal to the distance of the elevated
receptor point above the ground elevation of the plant. The terrain adjust-
ment made for any one receptor point does not affect concentrations at any
other receptor point. When the height of a receptor is above the shortest
plant stack height, then plume impaction on surrounding terrain is possible
and the model terminates processing after printing an error message. Also,
the model considers receptors below the ground elevation of the plant (e.g.,
receptor Rl in Figure 2-4) to be at the plant elevation.
Figure 2-4 also illustrates the mixing height assumption. This permits
calculations to be made using Equations (2-11) through (2-13), i.e., without
adding a vertical displacement term. This method of treating terrain adjust-
ments assumes ground based receptors and is not equivalent to simply including
a vertical coordinate term z in the Gaussian plume equation (e.g., Equation
(3.1) in Turner [7]). That method would not imply any changes in terrain
elevation at all. Rather, the value of z would specify the height at which
the receptor point would be "floating" in the air, and reflections of the
plume at the ground close to the stack, caused by elevated terrain, would not
be simulated.
2.5 RECEPTOR ARRAY
The Single Source (CRSTER) Model predicts concentrations for each hour
at a circular field of 180 receptors, as shown in Figure 2-2. The receptor
array is defined by: (1) five downwind ring distances and (2) 36 radials
(one for every 10° of azimuth). Although the 36 azimuths are fixed, selection
of the five ring distances is made by the user and warrants judicious consid-
eration since the choice of locations for receptor sites can result in major
2-21
-------
Mixing Height
Rl
Mixing Height
TERRAIN TREATMENT
WITHIN MODEL
in iiiii/fliii/rn / ifi 11 minh
Note: R1-R5 are receptor points at 5 ring distances.
FIGURE 2-4
ILLUSTRATION OF THE METHOD FOR TERRAIN ADJUSTMENT
IN THE SINGLE SOURCE (CRSTER) MODEL
2-22
-------
differences in the evaluation of impact on air quality. Once the ring dis-
tances have been chosen, terrain elevations can be obtained from a topo-
graphic USGS quadrangle map of the area (scale 1:24,000 is suggested) avail-
able from the U.S. Geological Survey, Washington, D.C. 20242.
Any one of a number of procedures can be used for preliminary deter-
mination of receptor ring distances in a Single Source (CRSTER) Model analysis,
The UNAMAP programs and the PTMAX* program in particular have been developed
to calculate the distance to the maximum 1-hour concentration from a single
point source for each of 49 stability/wind speed combinations. The model
equations used in PTMAX are the same as those in the Single Source (CRSTER)
Model except that PTMAX does not account for elevated terrain or for limited
mixing. The use of PTMAX does provide first order approximations to the
selection of receptor distances which can be refined through successive
applications of the Single Source (CRSTER) Model.
2.6 CONSIDERATION OF SOURCE CHARACTERISTICS
The emission source parameters which are input to the Single Source
(CRSTER) Model are:
• Plant elevation
• Stack parameters, for each stack
- pollutant emission rate (Q)
- stack gas velocity (vs)
- stack gas temperature (Ts)
- stack exit diameter (d)
- physical stack height (hs)
PTMAX is available through EPA's User's Network for Applied Models of Air
Pollution (UNAMAP) and has been documented by Khanna [11].
2-23
-------
The plant elevation is used in the analysis of uneven terrain by the model
(see Section 2.5). Effectively, the plant elevation is taken as the ref-
erence plane and all receptor elevations, plume height and mixing heights
are calculated relative to this plane.
The stack parameters are used by two components of the Single Source
(CRSTER) Model. The emission rate Q for a stack is used directly in the
modified Gaussian plume equations (2-11) and (2-12). The stack gas parameter
vs, T , and d are used in Briggs' plume rise equations (2-5) through (2-10)
to calculate the bouyancy flux parameter F. The estimate of plume rise Ah
is then added to the physical stack height hg to obtain the effective stack
height H used in the modified Gaussian plume equation (2-11).
The Single Source (CRSTER) Model provides for input of different values
for Q for each month. Since variations in emissions are often due to changes
in a plant's operating conditions, which in turn affects the values for v and
T , the model allows monthly values to be input for these stack parameters
as well. For each 1-hour concentration computed by the Single Source (CRSTER)
Model, the values of Q, v and T used are those corresponding to the month of
the year which contains the 1-hour period. Thus, an execution of the model
for an entire year will match up the monthly stack parameters with the appro-
priate month of meteorological data to provide a realistic simulation of actual
conditions.
The Single Source (CRSTER) Model assumes one geographical source location
but provides for specifying up to 19 individual stacks at the one plant site.
For the case of multiple stacks, each stack is considered separately and
calculations made based on its individual stack parameters. The program
totals the impacts of each individual stack at each of the 180 receptor points.
2-24
-------
Thus, concentrations output for each receptor point represent the impacts
from all stacks. In order to obtain the individual contributions from each
stack to concentrations at a receptor point, it is necessary to run the
Single Source (CRSJER) Model in source-contribution mode, as discussed below.
2.7 SOURCE-CONTRIBUTION MODE
When the source-contribution mode is specified for the Single Source
(CRSTER) Model, the concentrations output for each averaging time are reported
to show each stack's contribution to each receptor point. The model generates
a source-contribution table for each day processed. The receptor points for
which source-contributions are desired must be specifically identified in the
model input parameters by a paired number giving the ring distance and the
azimuth coordinates of the receptor. The number of such receptors is limited
to 20, though a different set of 20 may be specified for each averaging period.
Due to this restriction, the source-contribution mode is normally requested in
a model execution subsequent to review of the output from a multi-stack model
execution. The model will output source contributions for only those receptors
and averaging periods so identified. Thus, the user has complete control over
the output produced in this mode. A detailed discussion of the source-
contribution output and its interpretation is given in Section 4.3.4.a.
2.8 AVERAGING PROCESS AND RANKING PROCEDURE
The basic time increment for the Single Source (CRSTER) Model is one hour,
and 1-hour concentrations are computed for each of the 8,760 hours in a given
year (8,784 hours in a leap year). The 1-hour values are averaged to obtain
concentrations for longer averaging periods. The Single Source (CRSTER) Model
reports pollutant concentrations for 1-hour, 3-hour, 24-hour, and annual
averaging periods. In addition, calculations can be selected for a variable
averaging period from any one of the following: 2, 4, 6, 8, or 12 hours.
Concentrations for each averaging time are computed for discrete, non-
overlapping time periods, i.e., running averages are not computed. Thus, the
model can report for each day:
2-25
-------
• Twenty-four 1-hour average concentrations corresponding to the
periods 0000-0100 1ST, 0100-0200 LSI, ..., 2300-2400 LSI.
• Eight 3-hour average concentrations corresponding to the periods
0000-0300 LSI, 0300-0600 LSI, ..., 2100-2400 LSI.
• One 24-hour average concentration corresponding to the period
0000-2400 LSI.
Similar fixed time periods are used for the variable averaging period.
Concentration maxima are output by the Single Source (CRSTER) Model in
several different forms, as shown in Appendix B. An optional output is a
magnetic tape containing all of the 1-hour, 24-hour and annual average con-
centrations calculated by the model. This data tape can serve as a convenient
input to other analysis programs (see Section 3.2.2). The printed output
consists of the following:
. A table of the highest concentration at each of the 180 receptors, for
each averaging period. For a given averaging time, this table is ob-
tained by ranking, at each receptor point, the concentrations estimated
for the entire year and then selecting the highest value from each of the
180 rankings. Examples are the tables on pages B-15, B-16, B-18, B-20
and B-22 (in Appendix B).
• The maximum highest concentration at any receptor, for each averaging
period. For a given averaging time, this maximum is obtained by se-
lecting the maximum of the 180 entries in the above table of highest
concentrations. Examples are the concentrations listed on the second
line of the above mentioned tables in Appendix B.
• A table of the second-highest concentration at each of the 180 receptors,
for each averaging period except the annual. For a given averaging time,
this table is obtained by discarding the highest value at each of the
180 receptors and selecting the second-highest value from each of the
180 rankings. Examples are the tables on pages B-17, B-19, B-21 and B-23.
• The maximum second-highest concentration at any receptor, for each aver-
aging period except the annual. For a given averaging time, this maximum
is obtained by selecting the maximum of the 180 entries in the above table
of second-highest concentrations. Examples are the concentrations listed
on the second line of the above mentioned tables in Appendix B.
• A ranking of the 50 highest concentrations for the entire year, for each
averaging period except the annual. Examples are the tables on pages
B-24 through B-27.
Second-highest concentrations are output since the National Ambient Air
Quality Standards (NAAQS) for short-term averaging times are not to be
exceeded more than once per year. The maximum second-highest concentration
2-26
-------
at any receptor is especially relevant since EPA's Guidelines [12] specify
that this value should be used to determine compliance of a plant with
NAAQS on all time scales shorter than the annual average.
It should be noted that the maximum second-highest concentration at
any receptor for a given averaging time would not be necessarily second in
the ranking of the 50 highest concentrations for the entire year. More
likely it would be farther down the ranking of the 50 highest. This factor
serves to distinguish the difference in the procedure by which these values
are computed. The maximum of the second-highest concentrations is obtained
by discarding the highest concentration at each receptor site and selecting
the highest value among those remaining. The 50 highest concentrations for
an entire year are obtained by combining all concentrations (regardless of
receptor) in one ranking. Thus, the second concentration in the ranking of
50 is also most likely a highest concentration at one of the 180 receptors
(but not the maximum highest concentration).
2.9 LIMITATIONS
This section discusses the limitations associated with the assumptions
inherent in the Single Source (CRSTER) Model.
2.9.1 Steady-State Assumptions
The Gaussian plume equation is representative of steady-state conditions
and a homogeneous atmosphere. Its use is less valid when emissions, wind
speeds, directions, local turbulence and atmospheric stability are changing
rapidly with time or distance. This steady-state condition translates into
the following source and meteorological factors which must be considered:
• Continuous uniform emission rate
• Homogeneous horizontal wind field
• Representative hourly mean wind speed and direction
• No directional wind shear in the vertical
• Constant eddy diffusivities
• No plume history
• No material depletion or atmospheric transformation
2-27
-------
Most pollutant sources exhibit significant diurnal variations in their
emission rates. Thus, a monthly mean emission rate as used by the model may
not be representative for computing short-term concentrations.
The assumption of a homogeneous horizontal wind field is an idealized
situation. Uneven terrain generally complicates the transport and diffusion
of a plume by altering the speed and direction of the air flow. This assump-
tion also is unrealistic for time periods when large shifts in either direc-
tion or speed occur, as might be associated with the passage of a weather
front or the onset of a local lake or sea breeze circulation. Furthermore,
at sufficiently large downwind distances, synoptic scale motions, will cause
the mean wind direction to vary. Since the model is very sensitive to
changes in wind direction, results may be in error for time periods or over
distances during which these conditions occur.
Wind direction is generally found to veer (turn clockwise) and increase
in speed with increasing height (in the northern hemisphere at extratropical
latitudes) due to the effects of friction at the earth's surface. The amount
of veering in direction that occurs varies and is related to the roughness
of the surface, the thermal wind component (horizontal temperature gradient)
and the stability of the atmosphere (vertical temperature gradient). Over
smooth terrain, such as the great plains, the degree of veering from this
wind shear effect is on the order of 10°. Over average terrain with small
changes in elevation and with some trees and shrubs, the amount of veering
with height is about 15° to 20°. Over rough terrain, quite hilly or moun-
taneous or with numerous buildings and tall vegetation, the amount of veer-
ing with height can be as much as 40° to 50°. The thermal wind component
may act in either a clockwise or counter-clockwise direction and may be suf-
ficient in some cases to cause backing (turn counter-clockwise) of the wind
direction with height. The Single Source (CRSTER) Model assumes no direc-
tional wind shear in the vertical. Thus, errors in pollutant concentrations
estimated by the model, due to using a nonrepresentative wind direction,
will increase with increasing surface (terrain) roughness.
The values of the dispersion coefficients a and az in the Gaussian plume
equation are assumed to vary only in the downwind direction, i.e., they are
assumed invariant with height and crosswind distance. This condition is not
2-28
-------
necessarily satisfied in the atmosphere, particularly in the vertical direction,
thus necessitating the use of spatially representative parameters. The
Pasquill-Gifford estimates of dispersion coefficients are based on measure-
ments taken in open, level to gently rolling country and so may not be repre-
sentative of coastal areas or those with significant terrain features. Also,
since the field measurements were taken in the surface boundary layer of the
atmosphere, the derived dispersion coefficients may be less representative
at the heights of tall stacks (i.e., above 100 m). The dispersion coeffi-
cients are valid only for downwind distances greater than 100 m from the
plant. Beyond a few km downwind, the estimates are based on limited data and
so may be less accurate. Use of the Pasquill-Gifford dispersion coefficients
for a finite number of stability "states" or categories represents an approxi-
mation to the continuous changes which occur in the atmosphere.
The assumption of no plume history in the Single Source (CRSTER) Model
eliminates consideration of the cumulative effects from consecutive 1-hour
periods and does not allow for changes in the wind with time to affect the
path taken by the plume downwind. The average ambient concentration at a
receptor site may be significantly influenced by emissions released from the
plant during previous periods.
The assumption of no atmospheric transformations or depletion of the
plume constituents limits the validity of the Single Source (CRSTER) Model
to certain pollutants and conditions. Of the five directly-emitted criteria
pollutants, nitrogen dioxide and non-methane hydrocarbons participate in the
most rapid and complex chemical reactions in the atmosphere, and thus cannot
be simulated with this model. The conversion of sulfur oxides in the atmo-
sphere is highly variable, with oxidation rates in the range of 0,5% to 10%
per hour [13]. Thus, estimations of ambient S02 concentrations can be simu-
lated with the Single Source (CRSTER) Model as long as the travel time
to a downwind receptor (equal to the distance x divided by the wind speed u)
is short relative to the half-life of S0~ in a particular situation.
2-29
-------
Suspended participate matter can also be simulated with the Single
Source (CRSTER) Model when the size distributions do not include a significant
number of particles above 20 ym in diameter. Above this limit, a particle
generally has an appreciable settling velocity and so the use of the model
would introduce errors into the predicted concentrations.
Since carbon monoxide is gaseous and inert, it satisfies the modeling
assumption but generally is not associated with point emission sources.
2.9.2 Terrain Adjustment
Elevated terrain can significantly affect the downwind trajectory of a
plume by disturbing the wind flow field. Also, large scale eddies can form
in the lee of a hill or ridge resulting in increased turbulence and diffusion.
The Single Source (CRSTER) Model does not adjust plume flow vectors and dis-
persion for such terrain effects; the only terrain adjustment made in the
model is to lower the plume centerline to account for a receptor location
above that of the base elevation of the plant (see Section 2.4). Thus, the
reliability of the Single Source (CRSTER) Model is more limited for receptor
points which lie in significantly uneven terrain.
2.9.3 Mixing Height
The Single Source (CRSTER) Model assumes negligible concentration whenever
the effective stack height H exceeds the mixing height L at a given receptor
location. When this condition occurs, the plume is assumed to enter and remain
in the elevated stable layer and no contribution from the plume at that receptor
is computed. When these conditions occur and H > L, the Single Source (CRSTER)
Model may underestimate pollutant concentrations by assuming no source contri-
bution.
2.9.4 Calm Winds
Pollutant concentrations estimated by the Gaussian plume equation used in
the Single Source (CRSTER) Model are inversely proportional to average wind
speed (see Table 2-3). This relationship implies that concentrations will
2-30
-------
become infinite as the wind speed approaches zero, which is clearly not the
case. Low or calm wind speeds violate the assumption made in solving the
differential equation of diffusion that dilution dominates diffusion
in the downwind direction. For this reason, the model cannot simulate
the no-wind case of isotropic diffusion from a point source or diffusion
under low wind conditions, i.e., u < 1.0 ms . Thus, hourly mean wind speeds
below this value are increased to 1.0 ms to preclude an invalid application
of the model.
When wind speeds are less than 1.0 ms" , but greater than calm, the
measured wind direction is used. However, when calm conditions occur there
is no measured wind direction. In that case, the model uses the wind direction
from the previous (non-calm) hour in the dispersion calculations. Problems
can arise if a series of consecutive calm hours occur, because then the model
will assume a single, persistent wind direction for the duration of the calms.
Such an assumption of directional persistence may cause the model to overestimate
pollutant concentrations.
2.9.5 Other Limitations
The diffusion of pollutant emissions from an elevated point source is
often less than ideal due to aerodynamic effects. These effects can result
from the interaction of the wind with the physical structure of the plant,
upwind terrain, or from a low stack gas exit velocity. Such interaction
can retard, or in the extreme case, prevent plume rise. The extreme case
is commonly referred to as plume "downwash". With downwash, the effluent
is brought downward toward the ground into the wake of the plant, from which
point it diffuses as though emitted very close to the ground. Retardation
of plume rise and downwash can significantly increase the resulting impact
of a source. The Single Source (CRSTER) Model does not consider these
aerodynamic complications. Small sources, in particular, are likely to have
low stack heights and low gas exit velocities. Therefore, these sources, as
a class, are more likely to be influenced by aerodynamic effects.
2-31
-------
In addition, model estimations for low-level sources may be inaccurate under mete-
orological conditions that include low mixing heights. When the estimated mix-
ing height for a given hour is near the elevation of a low level source, the
emissions may be released either below or above the mixing height. The first
condition will result in a high concentration impact, while the second will
have no impact on ground-level concentrations. Due to the fact the hourly
mixing height is just an estimate, the model may simulate the wrong condition
and so make either a large over or underestimation of the impact of a low level
source. The user is therefore advised to review the meteorological and source
emissions data printouts to identify the occurrence of such conditions, and
where found, to flag the estimated concentrations as possibly being inaccurate.
When a plant configuration has multiple stacks, the separation between
stacks can act as an initial plume spread in the crosswind direction, reducing
downwind concentrations. This effect is at a maximum for wind directions
normal to a line of stacks, and is nearly zero when the wind is parallel to
a line of stacks. The Single Source (CRSTER) Model does not account for the
effects of stack separation since all stacks are assumed to be at the same geo-
graphical location. Thus, the model may somewhat overestimate concentrations
from multi-stack plants whenever stack separation is significant relative to
the crosswind dispersion coefficient a .
The Briggs plume rise formulas used in the Single Source (CRSTER) Model
(see Section 2.2.5) are for hot, buoyant plumes. Some small industrial sources
do not emit plumes with temperatures significantly in excess of those of
ambient air. Therefore, a simulation of these sources with the Single Source
(CRSTER) Model may tend to overestimate receptor concentrations since plume
rise will be underestimated. Briggs [1] has developed equations which can be
used to calculate plume rise for non-buoyant plumes, however, they are not
included in the Single Source (CRSTER) Model.
2-32
-------
3. DATA REQUIREMENTS AND OUTPUT
This chapter discusses the input data requirements for the Preprocessor
and Single Source (CRSTER) Model programs, and the output they produce. Input is
described by the type of information required, its source, limitations, res-
trictions, options and form. Program output is described by the information
provided, its form and ultimate usage. Specific information on input and out-
put format for these programs is given in Section 4.
3.1 PREPROCESSOR PROGRAM
The Preprocessor program is designed to directly utilize meteorological
information available from the National Climatic Center (NCC). Data obtained
from other sites can be utilized if the information satisfies the input
requirements described below and the formats given in Section 4.
3.1.1 Input Data Requirements
a. Surface Station Number. This number identifies the National Weather
Service (NWS) surface observation station for which hourly meteorological data
is input to the Preprocessor. This number is the WBAN station identification
number, not the WMO block identification number which is typically used by the
NWS. The Survey of TD-1440 [14] tabulates WBAN station numbers for all surface
stations for which hourly meteorological data are available from NCC. The
surface station selected for analysis of the plant should be representative of
the site. The station number is a five-digit integer, input in card form.
b. Year of Surface Data. This input parameter identifies the year during
which the meteorological data was observed, hereafter referred to as the "y^ar
of record". Only the last two digits of the year are used,with the input in
card form.
C- Latitude of Surface Station. This input parameter gives the geographic
latitude coordinate for the surface station in degrees (north of the equator).
The directory of NWS Offices and Stations [15] lists the latitude for each
station. The input is in card form.
3-1
-------
d. Longitude of Surface Station. This input gives the geographic lon-
gitude coordinate for the surface station in degrees (west of Greenwich). The
directory of NWS Offices and Stations [15] lists the longitude for each station.
The input is in card form.
e. Time Zone. This input parameter is a code for the time zone in which
the surface station is located. The standard time zone for a given station
can be determined from the nationwide map given in any telephone directory.
The standard values for this parameter are: 05 = Eastern, 06 = Central, 07 =
Mountain, 08 = Pacific. The value corresponds to the number of time zones west
of Greenwich Mean Time. The input is in card form.
f. Number of Days in the Year. This input variable gives the number of
days in the year of record. The value input must be 365,or 366 for a leap
year. The input is in card form.
g. Random Number Seed. This input variable is an integer, chosen ran-
domly, with a value between 1,000 and 9,999,999,999. This initiates the wind
direction randomization. In the program this variable is multiplied by 10,000
prior to entry into the random number generator. The input is in card form.
(,. Hourly Surface Observations. These data are the "Hourly Surface Ob-
servations in Card Deck 144 Format" available from NCC for the year of record.
The Preprocessor performs checks for missing data from this file but can
only sense when an entire record (one hour of data) is missing. Thus, if
data for only one parameter are absent, the Preprocessor will not flag such
gaps in the input data. It is recommended, therefore, that the user perform
independent integrity and completeness checks. The input of this data is in
magnetic tape form.
3-2
-------
i. Mixing Height Data. These data input the nocturnal (minimum) and
afternoon (maximum) mixing height for each day in the year of record. The
information required for each day includes:
• Upper Air Station Number - this number is the WBAN station
identification number identifying the NWS upper air observa-
tion station used to calculate mixing heights. The List of
Upper Air Stations [14] tabulates such WBAN numbers. The
station must be representative of the plant site.
• Year - the last two digits of the year of record for the
mixing height data.
• Month - the month number corresponding to a given set of
mixing heights.
• Day - the calendar day number corresponding to a given set
of mixing heights.
• Nocturnal Mixing Height - the minimum mixing height for a given
day calculated from the 1200 GMT upper air sounding on that day.
• Afternoon Mixing Height - the maximum mixing height for a given
day calculated from the 1200 GMT upper air sounding on that day.
The input of mixing height data is in card form where one card is
required for each day of the year of record. The first card of this file must
contain mixing height data for December 31 of the preceding year, while the
last card of the file must contain data for January 1 of the year succeeding
the year of record. The twice daily mixing height cards must be punched
from information available from NCC, on either magnetic tape or in printed
tabular form, for an appropriate NWS upper air observation station in the
year of record (see Section 4.2.3).
3.1.2 Output Information
Two types of output are generated by the Preprocessor program: (1) a
magnetic tape of hourly meteorological data for the year of record, and (2)
printed output of program diagnostic messages. The output tape serves as
direct input to the Single Source (CRSTER) Model program as the hourly values
of wind speed, flow vector, randomized flow vector, stability and temperature
are structured in the format required by the model. The printed output in-
forms the user of any inconsistencies or missing records in the output tape
as a result of deficiencies in the hourly surface observation data input to
the Preprocessor. A description of the error message directory for the Pre-
processor program and recommended user action are given in Section 4.2.4.
3-3
-------
3.2 SINGLE SOURCE (CRSTER) MODEL
3.2.1 Input Data Requirements
a. Heading Data. This information is input to label the model output
and generally contains the plant name and the air pollutant being modeled. The
input is in card form.
b. Comments (Optional). Comments are input in card form and are printed
at the beginning of the model output. Any number of comment cards may be
input to the model, but the last card must be totally blank (a delimiter) and
is required regardless of whether any comment cards are input.
c. Namelist Data. A variety of namelist data required by the model are
input in card form. Since namelist input is a free format, there is no fixed
number of cards. These data are itemized below:
Parameter
Name
IUR Urban/Rural Indicator - this parameter specifies whether the
urban or rural option is selected for analysis of the plant
site. The information is used by the model to select one of
the two sets of mixing heights on the preprocessed hourly
meteorological data tape. Values for this variable are re-
stricted to "1" (rural) and "2" (urban).
ITAP Output Tape Indicator (Optional) - this parameter specifies
whether a tape of calculated concentrations is to be output.
Values are restricted to "0" (no) and "1" (yes) with 0 as the
default value.
IPTZ Elevation Indicator (Optional) - this parameter specifies
whether the plant and receptors are located in flat or uneven
terrain. Values are restricted to "0" for the flat terrain
case (plant and terrain elevations set to 0) and to "1" for
the uneven terrain case (plant and terrain elevations and de-
limiter card must not be input with 0 as the default value).
DAY Day Indicators (Optional) - these parameters consist of an
array of 366 numbers corresponding to the calendar day in
the year of record. Concentrations are not calculated for
each day assigned a value of "0" and are calculated for
each day assigned a value of "1". The default value is "1"
for all 366 positions, but for non-leap years, the 366th
position must be set to "0".
RNG Receptor Ring Distances - these input data consist of five
receptor ring distances selected for analysis.
3-4
-------
Parameter
Name
IMET Meteorological Output Indicator (Optional) - this parameter
specifies whether the hourly meteorological data for each
day used in the model calculations will be printed. Values
are restricted to "0" (no output) and to "1" (output) with a
default value of "1".
ISS Surface Station Number - this number identifies the NWS sur-
face station observations input to the model. This number
must be the same as the "Surface Station Number" input to
the Preprocessor.
ISY Surface Station Year - this parameter identifies the year of
record for the surface meteorological data input to the model.
This year must be the same as the "Year of Surface Data" input
to the Preprocessor.
IVS Upper Air Station Number - this number identifies the NWS
upper air station observations used in computing mixing heights
input to the model. This number must be the same as the "Upper
Air Station Number" input to the Preprocessor.
IUY Upper Air Station Year - this parameter identifies the year of
record for the mixing height data input to the model. This year
must be the same as the "Year" input to the Preprocessor.
IVT Variable Averaging Time (Optional) - this parameter specifies
the length of the concentration averaging period used in the
model calculations. The values, in hours, are restricted to
"0" (no variable period), or one of the following: "2", "4",
"6", "8", and "12". The default value is "0".
IQCK Monthly Emissions Indicator (Optional) - this parameter speci-
fies whether separate monthly emission rates will be input for
each stack. Values are restricted to "0" (no) and "1" (yes,
monthly emissions for each stack must be input). The default
value is "0".*
QSSN Monthly Emission Rates (Optional) - These data consist of the twelve
monthly emission rates for each stack at the plant and must be
input when this option is selected.
IVCK Monthly Stack Gas Exit Velocity Indicator (Optional) - this para-
meter specifies whether separate monthly stack gas exit velocities
will be input for each stack. The values are restricted to "0"
(no) and "1" (yes, monthly stack gas exit velocities for each
stack must be input). The default value is "0".*
VSSN Monthly Stack Gas Exit V&1ocities (Optional) - these data consist
of the twelve monthly stack gas exit velocities each stack and
must be input when this option is selected.
ITCK Monthly Stack GdS Temperature Indicator (Optional) - this para-
meter specifies whether separate monthly stack gas temperatures
will be input for each stack. Values are restricted to "0" (no)
and "1" (yes, monthly stack gas temperatures for each stack must
be input). The default value is "0".*
*
Under default conditions, the annual value for this parameter, input with the
stack data, is used.
3-5
-------
Parameter
Name
TSSN Monthly Stack Gas Temperatures (Optional) - these data consist of
the twelve monthly stack gas temperatures for each stack at the
plant and must be input if this option is selected.
ISC Source Contributions Indicator (Optional) - this parameter speci-
fies whether separate source contribution tables will be generated
in the model output for selected receptors. Values are restricted
to "0" (no) and "1" (yes, source contribution receptors for each
selected averaging period must be input). The default value is "0".
SCI 1-Hour Source Contribution Receptors (Optional) - these data consist
of the distance and azimuth coordinates of each receptor for which 1-hour
source contributions will be printed. The receptor coordinates are
restricted to the standard model 180-receptor array. No more than
20 receptors can be selected for source contribution output. (Dis-
tances are in kilometers and azimuth range from 1 to 36).
SC3 3-Hour Source Contribution Receptors (Optional) - these data consist
of the distance and azimuth coordinates of each receptor for which
3-hour source contributions will be printed. The receptor coordi-
nates are restricted to the standard model 180-receptor array. No
more than 20 receptors can be selected for source contribution
output.(Distances are in kilometers and azimuths range from 1 to 36).
SCN Variables Averaging Period Source Contribution Receptors (Optional) -
these data consist of the distance and azimuth coordinates for each
receptor for which source contributions for the variable averaging
period will be printed. The receptor coordinates are restricted to
the standard model 180-receptor array. No more than 20 receptors
can be selected for source contribution output.(Distances are in
kilometers and azimuths range from 1 to 36).
SC24 24-Hour Source Contribution Receptors (Optional) - these data con-
sist of the distance and azimuth coordinates of each receptor for
which 24-hour source contributions will be printed. The receptor
coordinates are restricted to the standard model 180-receptor
array. No more than 20 receptors can be selected for source con-
tribution output.(Distances are in kilometers and azimuths range
from 1 to 36).
SCAN Annual Source Contribution Receptors (Optional) - these data consist
of the distance and azimuth coordinates of each receptor for which
annual source contributions will be printed. The receptor coordi-
nates are restricted to the standard model 180-receptor array. No
more than 20 receptors can be selected for source contribution
output. (Distances are in kilometers ?rd azimuths range from 1 to
36).
3-6
-------
d. Plant Elevation (Optional). This parameter inputs the elevation of
the plant site being modeled in feet above mean sea level (MSL). This param-
eters is required if namelist variable IPTZ=1.
e. Receptor Elevations (Optional). These data are requred input if the
"uneven terrain" option (ITPZ=1) is selected and consist of the receptor ele-
vations (in feet above MSL). The default values are 0. A blank delimiter
card is required to terminate this input file.
f. Stack Data. These data consist of individual stack information for
a maximum of 19 stacks. The data for each stack is presented on two cards
with the first card in each providing a stack identification, and the second
card providing data on the following stack parameters:
• Emission Rate - this variable specifies the average pollutant
emission rate for a given stack to be used in the model calcula-
tions for the year of record. If "Monthly Emission Rates" are
input, this variable can be left unspecified.
• Stack Height - this variable specifies the physical height of a
given stack above the plant elevation.
• Stack Diameter - this variable specifies the diameter of the
opening at the top of the stack.
• Stack Gas Exit Velocity - this variable specifies the average
stack gas exit velocity for a given stack to be used in the model
calculations for the year of record. If "Monthly Stack Gas Exit
Velocities" are input, this variable can be left unspecified.
• Stack Gas Temperature - this variable specifies the average stack
gas temperature for a given stack to be used in the model calcu-
lations for the year of record. If "Monthly Stack Gas Temperatures"
are input, this variable can be left unspecified.
Two blank delimiter cards are required to terminate this input file.
g. Preprocessed Hourly Meteorological Data. These data are a magnetic
tape file of hourly values of wind speed, flow vector, randomized flow vector,
stability class, mixing height and ambient temperature output by the Prepro-
cessor.
3-7
-------
3.2.2 0 u tput Infor mation
The output generated by the Single Source (CRSTER) Model program is in
two forms: (1) printed input data and calculated concentration values, and
(2) a magnetic tape (optional) containing 1-hour, 24-hour and annual average
concentrations. The printed output includes tables for each averaging period
of the highest and second-highest concentrations at each receptor point, and
tables of the 50 highest concentrations for the entire year for each averaging
period (except the annual). The contents of these tables and how they are
produced is detailed in Section 2.8. If specified, source contribution tables
for each averaging period will be output, in which case, the printing of all
other concentration tables will be suppressed. These outputs are illustrated
and described in detail in Section 4.3.4.
Each of these model outputs have potential application in the analysis
of air quality impact. The maximum annual concentration and the maximum
second-highest concentration for short term averaging periods are especially
relevant in determining compliance of a plant with National Ambient Air Quality
Standards (NAAQS), and if necessary, to determine required source reductions
to achieve NAAQS. The tables of highest and second-highest concentrations
for each averaging period can be used to plot concentration isopleths on maps
of the area surrounding a plant, providing information on the spatial varia-
tion of maximum concentration levels. These concentration isopleths can also
be overlaid on population density maps to obtain measures of total population
exposure to different pollutants and levels. The ranking of the 50 highest
concentrations for each averaging period can be used to determine the number
of times in a year that ambient concentrations exceed a particular level (e.g.,
one of the NAAQS) in the vicinity of a plant. The concentration data written
on the output tape provide a complete record of model predictions for each
receptor point and so can be used as input to various analysis programs. For
example, the concentrations at a given receptor can be rank ordered by size
and their frequency distribution characterized using standard techniques [16].
Such frequency distributions can be used for comparison with distributions of
measured data for the same receptor site in model validation studies (see
3-8
-------
Appendix F). The concentration data can also be ordered in time for a given
receptor to provide a time series of model predictions for comparisons with
measured data or statistical trend analysis. Finally, the source contribution
tables allow the impact of individual stacks to be determined. Such informa-
tion is required when it is necessary to examine alternative control strate-
gies for reducing plant emissions toward achieving compliance with NAAQS.
Individual stack contributions are also required in evaluations of proposed
new facilities with regard to Prevention of Significant Deterioration regula-
tions.
3-9
-------
4. SINGLE SOURCE (CRSTER) MODEL USER'S GUIDE
4.1 INTRODUCTION
The procedure for modeling using the Single Source (CRSTER) Model involves
two data processing steps. First, a Preprocessor program prepares the meteo-
rological data in the hourly format required for input to the Model. Second,
these data and appropriate source emissions and stack data for the plant are
input to the Single Source (CRSTER) Model and predicted concentrations obtained.
Figure 4-1 is a flow diagram depicting the modeling procedure.
The Preprocessor program generates a magnetic tape of hourly meteorolo-
gical data for the Single Source Model. The input required by the Preprocessor
are hourly meteorological surface observations and twice daily mixing heights.
In the process of generating data for the Single Source (CRSTER) Model, the
Preprocessor also performs several quality assurance data checks.
The tape of hourly meteorology, output by the Preprocessor program, can
be used with the Single Source (CRSTER) Model to model any plant for which the
selected combination of upper air station and surface station data is applicable.
Thus, output should be stored permanently whenever possible for future appli-
cations.
The Single Source (CRSTER) Model calculates concentrations for an entire
year from a plant with up to 19 individual stacks and prints out the highest
1-hour, 3-hour, 24-hour, as well as annual average, concentrations at a set of
180 receptors, surrounding the plant. In addition, concentrations can be ob-
tained for a variable averaging period which can be selected as 2, 4, 6, 8 or
12 hours. The model can optionally produce output for individual source con-
tributions at selected receptor points for each time interval in a day. It
can also generate a magnetic tape file of hourly, daily and annual concentra-
tions at each receptor point.
4-1
-------
Initialization
Card and Twice
Daily Mixing
Heights
Hourly
Surface
Meteorology
Preprocessed
Hourly
Meteorology
Program
Options,
Receptor and
Source Data
Single
Source
Model
Modeling
Results
Hourly
Concentrations
(Optional)
FIGURE 4-1
PROCEDURE FOR USING THE SINGLE SOURCE (CRSTER) MODEL
4-2
-------
Input requirements for the Single Source (CRSTER) Model include the
meteorological data tape output by the Preprocessor program, receptor ring
distances and terrain elevations, and individual stack data. The stack data
items required are source emission rate, stack height, diameter, exit velocity
and exit temperature.
The following sections describe the Preprocessor and Single Source (CRSTER)
Model programs in detail; their input and output and the requirements for their
execution on a UNIVAC 1100 EXEC 8 Operating System. Instructions are also
given for executing the programs on an IBM System/360 Operating System.
4.2 PREPROCESSOR PROGRAM
4.2.1 Description
The Preprocessor program generates a magnetic tape of hourly values for
the meteorological parameters required by the Single Source (CRSTER) Model,
namely wind speed, wind direction, mixing height, stability class and tem-
perature. The input required by the Preprocessor are: (1) hourly National
Weather Service (NWS) observations of surface wind speed, wind direction,
temperature, cloud cover and ceiling height in magnetic tape format; and
(2) daily minimum and maximum mixing heights in punched card format, as de-
termined from NWS 1200 GMT upper air temperature surroundings, using the
methods of Holzworth [9]. The Preprocessor, in addition to generating the
meteorological data for the Single Source (CRSTER) Model, performs checks for
missing data on the tape of hourly surface observations and prints diagnostic
messages for any discrepancies that are detected. Quality checks are not per-
formed on mixing height data input to the Preprocessor, and so it is recom-
mended that the user review such data independently.
Data for each day are read by the Preprocessor and processed one hour at
a time, using methods described in Section 2.3. The cloud ceiling height,
wind speed, ambient temperature, and sky cover data are used to classify the
atmospheric stability for each hour. The wind speed is converted from the input
units of knots to meters per second, as required by the Single Source (CRSTER)
4-3
-------
Model. A flow vector (indicating plume direction) is calculated from hourly
mean wind direction data. In addition, a randomized flow vector is computed
to account for the natural turbulent fluctuations of the wind, not reflected
in the hourly mean observation. Hourly mixing heights for both rural and
urban conditions are derived from the morning (daily minimum) and afternoon
(daily maximum) mixing height data.
The Preprocessor program is written in FORTRAN V language for execution
on a UNIVAC 1100 Series Operating System and is compatible with most FORTRAN
IV compilers on other types of computers. A program source code listing for
the Preprocessor is included in Appendix A. Figure 4-2 is a flow diagram of
the functions performed by the Preprocessor.
4.2.2 Control Language and Data Deck Setup
a. Control Language Requirements. The following runstream illustrates
the Executive Control Language (ECL) required to execute the Preprocessor
program on the RTCC UNIVAC 1100:
@RUN,priority Jobid,account,userid,time
@ASG,A prog-file
G>ASG,A cd!44-file
0USE 8,cdl44-file
@ASG,CP met-file
G»USE 9, met-file
@XQT prog-file.PREP
card input deck
where:
priority = job priority
jobid = six-character job identification
account = user account number
userid = user identification code
time = time requirement for executing the job
prog-file = name of program file containing the Preprocessor
absolute element
cd!44-file = the file name assigned to the input data file of
hourly meteorology in card deck 144 format
met-file = name to be assigned to the output file of combined
hourly surface and upper air meteorological data
card input deck = the initialization card followed by the set of
mixing height cards.
4-4
-------
Start
d>
Read
Initial-
ization
Card
Skip First
Record on
Input
Tape
Read
Record for
Hour 1 on
Input Tape
7
Read First
Three Mixing
Height Cards
Calculate Time
of Sunrise
and Sunset for
day I DAY
Calculate 24
Random Numbers
for Random
Flow Vector
Check data
for Continuity
Print
Diagnostics
Message
Initialization
Card
Mixing
Height Cards
Stop
FIGURE 4-2
PREPROCESSOR PROGRAM FLOW DIAGRAM
4-5
-------
Transfer
Mixing
Height
Variables
Convert
Mind Speed
from Knots
to m/sec
.
Convert Ambient
Temperature
from
°F to °K
i
Determine
Wind
Direction
Calculate
Flow
Vector
Calculate
Randomized
Flow Vector
t
! Determine
! Stability
Class
Calculate
Urban
and Rural
Mixing Height
_, /
/
ead Mixing
fHeight Data
for
Next Day
Mixing Height
Cards
FIGURE 4-2 (Continued)
PREPROCESSOR PROGRAM FLOW DIAGRAM
4-6
-------
No
No
IHR=IHR+1
Read
Input Tape
for Next
Hour
Output
Record for
Day
I DAY
Yes
Write "All
Records
Have Been
Processed" ,
1
Use Data
from Previous
Hour for
Final Hour
Stop )
FIGURE 4-2 (Continued)
PREPROCESSOR PROGRAM FLOW DIAGRAM
4-7
-------
The following IBM Job Control Language (JCL) is required to compile, link-edit
and execute the Preprocessor program on an IBM System/360 Operating System:
//jobno JOB (account),'name',TIME=time
//EXEC FORTGCLG,COND=(4,LT)
//FORT.SYSIN DD *
source deck
//GO.FT08F001 DD DSN=CD144,UNIT=2400,VOL=SER=xxxxxx,DISP=OLD,
//LABEL=(,NL),DCB=(RECFM=FB,LRECL=80,BLKSIZE=800)
//GO.FT09F001 DD DSN=met-file,UNIT-2400,DISP=(NEW,KEEP),
//VOL=SER=yyyyyy,DCB=(RECFM=FB,LRECL=775,BLKSIZE=7750)
card input deck
/*
where:
jobno = job number
account = accounting information (system dependent)
name = programmer's name
time = estimated CPU time requirement
source deck = the Preprocessor program source deck on cards
xxxxxx = volume serial number of the tape containing
the hourly meteorological data file in card
deck 144 format
met-file = name to be assigned to the output meteorological
data file
yyyyyy = volume serial number of the tape to receive the
output data file
card input deck = card input file consisting of the initialization
card followed by the set of mixing height cards.
This example assumes that the input will be on a nonlabeled tape with
a block size of 800. This is representative of the form in which the data is
issued by the National Climatic Center.
The output file in this example is also on tape with a blocking
factor of 10 records per block.
b. Djita Deck Setup. The data card deck required for input to the
Preprocessor must be set up as follows:
4-8
-------
Preprocessor initialization card
Mixing height card for December 31 of the year preceding
the year of record
Mixing height card for January 1 of the year of record
Mixing height card for January 2 of the year of record
• Mixing height card for December 30 of the year of record
• Mixing height card for December 31 of the year of record
• Mixing height card for January 1 of the year following
the year of record
The Preprocessor initialization card contains vaules for data items that must
be initialized for each run. The mixing height cards contain the morning and
afternoon mixing heights for the corresponding day. The methodology for cal-
culating hourly mixing heights from the twice daily mixing heights involves
interpolations using the afternoon mixing height from the preceding day and
both mixing heights from the following day. For this reason, the mixing
heights for the last day of the year preceeding the year of record must be
included as the first mixing height card and the mixing heights for the first
day of the year following the year of record must be included as the last
mixing height card. If these data are not available, the data for the first
and last days of the year of record can be substituted, respectively. The
contents and formats of the Preprocessor initialization card and mixing height
cards are described below.
4.2.3 Input Data Description
a. Card Input. Card input requirements for the Preprocessor consist of
an initialization card followed by a set of mixing height cards. Coding forms
for preparing the Preprocessor card input data are included in Appendix C.
The Preprocessor initialization card contains information to initialize
the following items:
4-9
-------
NWS meteorological surface station number
Year of record
Latitude of the surface station
Longitude of the surface station
Time zone of the surface station
Number of days in the year
Initial value for generating random numbers
Table 4-1 is a description of the format of the initialization card.
The set of mixing height cards is composed of one card for the last
day of the year preceeding the year of record, one card for each day of the
year of record, and one card for the first day of the next year. Each card
contains the NWS upper air station number, the date and values for the morning
and afternoon mixing heights. Table 4-2 is a description of the format for
the mixing height cards. The twice daily mixing height cards must be punched
up from either magnetic tape or printed tabular data for an appropriate NWS
upper air station in the year of record, which can be purchased from the
National Climatic Center (NCC), Asheville, North Carolina. This tape stores
one day of data per 34-character record using the format shown in Table 4-3,
with 10 records to a block. Note that the afternoon mixing height column
numbers on the NCC tape are not the same as those required on the Preprocessor
mixing height data cards.
b. Tape Input Requirements. A magnetic tape containing card images of
the hourly meteorological data in "Card Deck 144 format" required by the
Preprocessor can be purchased from the NCC. These data can also be purchased
as 8,784 punched cards. Each tape data file contains one card image record
for each hour. The format of these records is described in the Card Deck 144
WBAN Hourly Surface Observations Reference Manual [17]. Data on this file
used by the Preprocessor includes the station number, year, month, day, hour,
cloud ceiling height, wind direction, wind speed, dry bulb temperature, and
sky cover. When ordering tapes to be processed on a UNIVAC computer, a
blocking factor of one record per block should be specified.
4-10
-------
TABLE 4-1
PREPROCESSOR INITIALIZATION CARD FORMAT
Card Columns
1-5
6-7
8
9-18
19-28
29-30
31-34
35-44
Format
15
12
F10.1
F10.1
F2.0
14
F10.Q
Description
NWS Surface Station WBAN Number
Year of Surface Data
Blank
Latitude of the Surface Station (degrees
hundredths )
Longitude of the Surface Station (degrees
hundredths)
Time zone in which the Surface Station
is located:
05 = Eastern
06 = Central
07 = Mountain
08 = Pacific
Number of days in the year of record (365
non-leap years; 366 for leap years)
Random Number Seed *
to
to
for
NOTE: The user is cautioned with regard to the random number generator
used in the preprocessor program. The subroutine called in this program is
entitled RANDU and is provided by Sperry Rand Corporation for use on the
Environmental Protection Agency's Univac 1110. There is no computer code
available, therefore, requiring the user to contact his systems personnel
about a suitable alternative. Because the same random number generator will
not be used by all users, the randomized flow vectors may differ when com-
paring preprocessor file results from two different computers.
4-11
-------
TABLE 4-2
PREPROCESSOR MIXING HEIGHT DATA CARD FORMAT
Card Column Format Description
1-5 15 NWS Upper Air Station WBAN Number
6-7 12 Year of record (last two digits)
8-9 12 Month
10-11 12 Day
12 IX Blank
13-17 F5.0 Morning Mixing Height (m)
18-30 13X Blank
31-35 F5.0 Afternoon Mixing Height (m)
4-12
-------
TABLE 4-3
DATA RECORD FORMAT FOR NCC MAGNETIC TAPES OF
MORNING AND AFTERNOON MIXING HEIGHTS
Record
Positions
1-5
6-7
8-9
10
11-12
13
14-17
18-20
21-23
24
25-28
29-31
32-34
Format
15
12
12
11
12
Al
14
13
13
Al
14
13
13
NWS Upper Air Station WBAN Number
Year of record (last two digits)
Month
Season (1 = Dec. -Feb. , 2 =
Mar. -May, 3 = June-Aug., 4 =
Sept. -Nov.)
Day
Type Code (P = precipitation,
C = morning average temperature plus
5°C is less than 1200 GMT surface
temperature, M = missing)
Morning Mixing Heights (m)
Morning* Surface Wind Speed (m s~ )
Morning Wind Speed Averaged from Surface
through Mixing Height (m s~1)
Type Code (P = precipitation,
C = afternoon average temperature is
less than 1200 GMT surface temperature,
M = missing)
Afternoon Mixing Height (m)
Afternoon** Surface Wind Speed (m s )
Afternoon Wind Speed Averaged from
Surface through Mixing Height (m s~1)
* 0200-0600 LST
** 1200-1600 LST
4-13
-------
If data for an NWS surface observation station have not been processed
into tape format by the NCC, the data can be coded from printed copies of the
WBAN Form A observations, also available from NCC. Instructions for coding
the data in Card Deck 144 format are given in the Card Deck 144 Reference
Manual [l?]. Only the items required by the Preprocessor need to be coded.
4.2.4 Output Data Description
The output tape from the Preprocessor consists of a sequential file con-
taining a file identification record followed by one record for each day in
the year.
The file identification record contains the year of record for the sur-
face meteorological data, the surface station identification number, the year
of record for the mixing height data, and the upper air station identification
number.
Each of the daily records contain the year, month, and the Julian day
followed by 24 values of stability class, wind speed, temperature, flow vector,
randomized flow fector, and rural and urban mixing heights. Because the last
record on the input tape corresponds to the 23rd hour of the last day, the
data for that hour is also used for the 24th hour on the last record on the
output tape.
All records on the output file are written with an unformatted FORTRAN
write statement. The output files, therefore, are machine dependent and
cannot be directly accessed by the Single Source (CRSTER) Model on systems different
from the computer creating the file.
The four parameters on the file identification record and the year, month
and stability class on each of the daily records are stored as FORTRAN integer
varibles. All other values on the daily records are FORTRAN real number
variables. Table 4-4 is a description of the arrangement of the variables on
each of the daily records.
Assuming no fatal error messages, the entire Preprocessor file is printed
and successful run completion is indicated by the message:
ALL RECORDS HAVE BEEN PROCESSED
4-14
-------
«tf-
LU
CQ
<
1-
z
O
p
t_l
CJ
to
1 1 1
o
8
1 1 1
I-H
LL.
1-
e
§
ce.
o
to
to
LU
o
fg
Q.
LU
O-
(O
s-
o
u.
tv
jQ
•^
S-
ID
Ol
S
IB
•^
S-
lO
>
0
c
o
•^
•p
CO
o
CL.
O)
Q.
£*
OJ
«
•r"
(13
i
it)
i.
O
o
ac.
CO
0)
to i—
10 10
r— >
O
z? 1
•r- Q.
r— tO
10 C
•»-» •!-
tO ^E
.C
C CM CM
O
O O
fO (O
>> i. S-
s
CD
—1 LU _|
LU Z LU
Q£ t~H Q£
a
i — LU
>- 1— LU
Q ** tO
t*'* r™*
CM LO
ro i i
^1- 00
CM
^— x
o
CO
(U
3
^~,
(11
S-
CT.
*-*
cn
^
CO
o
>
s
r-*
LU
0)
N
•r*
C
IT3
r^
CM
O
ia
<
S
Q£
U_
ro
CM
i
0
o
1
"« C
s- o -o
S -o s. S
cn aj 3 O
X -P O
cn T- co -C <4-
C E 0
•!!**.£
C 3 fl) <*-
O 3 -P
CM 10 jt cn
^- > ^J -r-
CM T3 0)
£* t— cn
to co c
CM at a. -r-
3 ^ 3 X
c > - — cn
O i—
'!/) f 0) •!- fc-
p *^% ^ «••
fc- O/ ^ — J
a> -i- i— TJ s-
E ai « t-
•o o .c
cn +-> ai -P
4-15
-------
4.2.5 Diagnostic Messages
Two types of diagnostics may be generated by the Preprocessor program
namely, fatal error messages and informative messages.
Fatal error messages are printed when an inconsistency is detected in the
input data, causing the program to stop execution. Table 4-5 is a list of all
possible fatal error messages and the corrective action that should be taken.
Informative messages are printed when anomalous data items are detected.
No user action is necessary in response to an informative message unless the
assumptions made by the program adversely effect the results. Table 4-6 is a
list of the informative messages in the Preprocessor program.
4.3 SINGLE SOURCE (CRSTER) MODEL
4.3.1 Description
The Single Source (CRSTER) Model is a computer program that applies a mod-
ified form of the Gaussian plume equation in calculating the contributions to
ambient air quality levels from a single plant with up to 19 individual stacks.
The model calculates concentrations for an entire year (calculations can
optionally be restricted to only selected days) and prints out the highest
1-hour, 3-hour, 24-hour, as well as annual average, concentrations at a set
of 180 receptors surrounding the plant. In addition, concentrations can be
obtained for a variable averaging period which can be selected as 2, 4, 6, 8
or 12 hours. The Single Source (CRSTER) Model predicts pollutant concentrations
for each hour at a series of 5 ring distances and at 36 wind direction azimuths
(every 10°), based on hourly values of wind speed, direction, stability class
and mixing height. The model can optionally produce output for individual
stack contributions to concentrations at selected receptor points for any
averaging period. The model can also generate a tape file of hourly, daily
and annual concentrations at each receptor point.
4-16
-------
TABLE 4-5
PREPROCESSOR FATAL ERROR MESSAGES
ID DOES NOT MATCH IN RECORD # 1 ID ON TAPE IS n ID REQUESTED IS k
*****DATA IS MISSING. PLEASE CORRECT INPUT FILE*****f
Description: This message is printed if the surface station identification
entered on the Preprocessor initialization card does not
match the station number on a record in the hourly meteoro-
logical data file. The value i is the number of the record
within the file, n is the station number on the record, and k
is the station number requested.
Action: If n = 1, check the surface station number input on the
Preprocessor initialization card and the input meteorological
data file to be sure that the correct file was input.
If n > 1, the input file is bad. A listing of the file
should be inspected to determine what corrective action is
necessary.
YEAR IS i INSTEAD OF j IREC = n
Description: The year i on record number n of the meteorological data
file did not match the year j input on Lhe Preprocessor
initialization card.
Action: If n = 1, check to be sure that the correct year was entered
on the Preprocessor initialization card and that the correct
input file was used.
If n > 1, the data file is in error. A listing of the file
should be inspected to determine what connective action is
necessary.
If n > 8,760, check to be sure the number of days entered
on the Preprocessor initialization card is correct.
"This line appears on fatal error messages for ID, YEAR, MONTH, DAY, HOUR
4-17
-------
TABLE 4-5 (CONTINUED)
PREPROCESSOR FATAL ERROR MESSAGES
MONTH i DOES NOT AGREE WITH LOOP j IREC = n
Description: The month i on input record n of the meteorological data
file is out of sequence. The month should have been
number j.
Action: If i = 2 and j = 3, check to be sure the number of days
entered on the Preprocessor initialization card was 365 for
a non-leap year. If i = 3 and j = 2, check to be sure that
the number of days entered on the Preprocessor initializa-
tion card was 366 for a leap year. If the number of days
input was incorrect, enter the appropriate value and rerun
Preprocessor. Otherwise, inspect a listing of the input
fuel for missing or extraneous records. If the file is out
of order, sort the file in ascending order keying on
columns 1-13.
DAY i DOES NOT AGREE WITH LOOP j IREC = n
Description:
Action:
The day i on input record n of the meteorological data file
is out of sequence. The day should have been number j.
Inspect a listing of the input file for missing or extrane-
ous records. If the file is out of order, sort the file in
ascending order keying on columns 1-13.
HOUR i DOES NOT AGREE WITH LOOP j IREC = n
Description: The hour i on record n of the input meteorological data
file was out of sequence. The hour expected was hour j.
This message is printed for hours 1 through 23, for which
the value on the input record should equal the value of
the index of the hourly 'DO1 loop.
Action: Inspect a listing of the input file for missing or extrane-
ous records. If the file is out of order, sort the file
in ascending order keying on columns 1-13.
4-18
-------
TABLE 4-5 (CONTINUED)
PREPROCESSOR FATAL ERROR MESSAGES
ERROR: MISSING HOUR LOOP VALUE = j WHILE VALUE ON RECORD n IS i
Description: This message is printed if the last hour of the day
being processed (j = 24) is not the data for the
first record of the next day (i = 0).
Action: Inspect a listing of the meteorological data file
for missing or extraneous records. If the file is
out of order, sort the file, keying on columns 1-13.
4-19
-------
TABLE 4-6
PREPROCESSOR INFORMATIVE MESSAGES
THE CHARACTER x IS NOT ALLOWABLE . CLOUD COVER DEFAULTS TO 10.
Description:
The cloud cover on a record in the input meteorological
data file was value x. The only valid values for cloud
cover are 0,1,2,3,4,5,6,7,8,9 or -. The program assumes
a value of "- " which is interpreted as a 10 tenths cloud
cover.
STABILITY = i j r n
Description:
A stability class was assigned an invalid value (i). The
value j is the wind speed index, r is the net radiation
index number, and n is the number of the input meteorolog-
ical data file record being processed.
4-20
-------
The Single Source (CRSTER) Model is written in FORTRAN V language for
execution on a UNIVAC 1100 series computer and is compatible with most FORTRAN
IV compilers on other types of computers. The Single Source (CRSTER) Model is
composed of a main program (CRSV), three subroutines (CRS, BEH072, and SIGMA),
and a block data common storage area. The source code for each of these pro-
gram elements is listed in Appendix A.
The CRSV program is responsible for reading and checking the card input
data, and checking that the correct preprocessed meteorological data file
was assigned to the run. If any errors in the input data are detected, a
diagnostic message is printed and the program execution is terminated.
Otherwise, the program calls the CRS subroutine to perform the dispersion
modeling and output the results, after which control is returned to CRSV and
execution is terminated. Figure 4-3 is a flow diagram of the CRSV program.
The CRS subroutine calculates hourly concentrations at each receptor,
derives the information described previously, and outputs summary tables and
source contribution tables. Figure 4-4 is a flow diagram of the CRS subroutine
described below. First, the SIGMA subroutine is called by CRS to calculate an
estimate of the horizontal (a ) and vertical (a ) dispersion coefficients for
J
each combination of distance and stability class. These values are stored
in an array for use in the dispersion calculations. The program then reads
the hourly meteorology for each day and calculates the concentration for each
stack at each receptor point. Plume rise estimates are computed by the BEH072
subroutine. The concentrations from each stack are summed to form plant
totals which are averaged over each 1-hour, 3-hour and 24-hour time interval.
If a variable averaging time was specified (2, 4, 6, 8 or 12 hours), the
average concentrations for each time interval of that length are also computed.
As calculations for each day are completed, the daily concentrations are added
to the annual total, and the average concentrations are checked for maxima.
If requested, the individual stack contribution to the concentrations at se-
lected receptor points are listed. When the last day in the meteorological
data file has been processed, the summary tables for each averaging period are
printed and control is returned to the CRSV program. If source contributions
have been requested, these summary tables are not printed.
4-21
-------
Start
Read Headingy
Card, Com-
ment Cards,
and Hamelist/
Cards
C)ard Input
Deck
Are
Specified
Receptors
Valid
Read Plant
and Terrain
Elevation
Cards
>,
f
Convert
Elevations to
Meters and
Check for
Errors
y
Set Terrain
Elevation =
Plant
Elevation
Yes
Terrain
Elevation
Tables
Print
Terrain
Elevation
Tables
FIGURE 4-3
CRSV PROGRAM FLOW DIAGRAM
4-22
-------
Stop
Print
Diagnostic
Stack Data
Tables
/ Read /
/ Slack A*
/ Cards /
[ Card Input
Deck
Print
Stack
Input
Data
Determine
Days to be
Processed
Subroutine
CRS
FIGURE 4-3 (Continued)
CRSV PROGRAM FLOW DIAGRAM
4-23
-------
No
( Start CRS )
\
i
Initialize
Arrays
<
Call <
to Cor
ay ar
Vah
1
5IGMA
npute
id az
jes
/Loop on \
Days /
\ (1=1,366) /
'
/Read Hourly /
/ Net. Data L
1 for Day I r
^ SuhrnuHno
SIGMA
/ Met. \^
,( Tape \
1 (Unit 9) I
Yes
/ Is
X"E]
I
\. Yes 1 P^nt
i ^>. »> Meteorological
J/ \ Data for
^ L.^p'
/ Loop on \
\ (J-1.24) /
\
Call B
to Co
Effec
Plurr,e H
for Ho
'
i
EH072
mpute ,. Subroutine
t\ve BEH072
eights
ur J
i
00 0
FIGURE 4-4
SUBROUTINE CRS FLOW DIAGRAM
4-24
-------
Loop on
Receptors
(K=l,180)
Loop on
Stacks
(L-l.NS)
no
Calculate
Concentration
from Stack L
at Receptor K
Add Concentration
to 1-Hour
Total at
Receptor K
I Add Concentration
to Source
Contribution
Arrays
Add to
24-Hour
Total
for Receptor K
FIGURE 4-4 (Continued)
SUBROUTINE CRS FLOW DIAGRAM
4-25
-------
0 0
Add 1-Hour
X's to
3-Hour Totals
Print
Source
Contributions
Calculate
Averages
and Check
for Maxima
Print 1-Hour
and 24-Hour
Maxima
for Day I
Yes
Add 1-Hour
x's to Totals
for Variable
Averaging Period
Output
180 x's
for Hour J
Output
180 x's for
Day I
FIGURE 4-4 (Continued)
SUBROUTINE CRS FLOW DIAGRAM
4-26
-------
Output Table
of Annual
Mean x's
Output Highest
and Second
Highest Tables
for each
Averaging Period
Highest and
Second Highest
Tables
Rank 50 Highest
x's for each
Averging
Period
Output Tables
of 50 Highest
X's
"Output
180 Annual
Mean x's
FIGURE 4-4 (Continued)
SUBROUTINE CRS FLOW DIAGRAM
4-27
-------
The SIGMA subroutine computes a and o for a given downwind distance and
-------
Start
SIGMA
i
Determine
Appropriate Equa-
tion for Given
Stability Class
and Distance
1
f
Calculate
Gz
\
f
Calculate
ay
i
RETURN
FIGURE 4-5
SUBROUTINE SIGMA FLOW DIAGRAM
4-29
-------
Start
BEH072
Calculate:
Volumetric Flow
Rate,Buoyancy
Flux, Heat
Output
Calculate
Unstable-Neutral
Plume Rise
Calculate
Stable
Plume Rise
Calculate
Effective
Plume Height
RETURN
FIGURE 4-6
SUBROUTINE BEH072 FLOW DIAGRAM
4-30
-------
where:
priority = job priority
jobid = six-character job identification
account = account number
userid = six-character user identification code
time = execution time requirements
prog-file = name of the program file containing the Single
Source (CRSTER) Model absolute element
met-file = name of the preprocessed meteorological data file
output-file = name of the file to receive the hourly concen-
trations
card input data = input data card deck.
The following Job Control Language (JCL) is required to compile, link-edit
and execute the Single Source (CRSTER) Model on an IBM System/360 Operating System:
//jobno JOB (account),'name',TIME=time
//SSM EXEC FORTGCLG,COND=(4,LT)
//FORT.SYSIN DO *
Single Source Model Source Deck Optional,
//GO.FT08F001 DD DSN=out-file,UNIT=2400,VOL=SER=xxxxxx, -| required
//DISP=(NEw,KEEP),DCB=(RECFM=VBS,LRECL=720,BLKSIZE=720Q)_T only if
//GO.FT09F001 DD DSN=met-file,UNIT=2400,DISP=OLD, ITAP=1
// VOL=SER=yyyyyy
//GO.SYSIN DD *
card input deck
/*
where:
jobno = job number
account = required accounting information
name = programmer's name
time = estimated CPU time requirements
out-file = name of the output file to receive the hourly
concentrations
xxxxxx = volume serial number of tape to receive the
hourly concentrations
met-file = name of the preprocessed meteorological data file
= volume serial number of the tape containing the
preprocessed meteorological data file.
4-31
-------
In this example, the input meteorological data file and the output
file on unit 8 are assumed to be tape files. If these data files are to be
stored on other devices, appropriate changes must be made to the JCL listed
above.
b. Data Deck Setup. The data card deck required by the Single Source
(CRSTER) Model is dependent on the options requested by the user. In general,
the input deck must be set up as shown in Figure 4-7 and summarized below:
• Plant Title Card
• Comment Cards (as many as required)
• Delimiter Card (All blanks)
• Namelist NAM1 Cards
• Plant Elevation Card
Receptor Elevation Cards (36 cards;
one for each direction)
Optional, required
only if IPTZ=1
• Delimiter Card (All blanks)
• Stack Data Cards (two cards for each stack; the first card is a
stack identification card; the second card contains the necessary
stack data). Up to 19 pairs of stack data cards can be input.
• Delimiter Cards (2) (All blanks)
Examples of decks set up for a standard Single Source (CRSTER) Model run and
for a source contribution run are included in Appendix B. A description of
the format and contents of each card type is given below.
4.3.3 Input Data Description
The data required to execute the Single Source (CRSTER) Model is contained
on two files. A card file must be input that contains source and receptor
information, and the meteorological data tape generated by the Preprocessor pro-
gram must be input for the appropriate upper air and surface station combination.
a. Card Input Requirements. The card deck required for input to the
Single Source (CRSTER) Model consists of seven types of cards:
• Heading Cards
• Comment Cards
• Delimiter Cards
• Namelist Cards
• Plant Elevation Cards
• Receptor Elevation Cards
• Stack Data Cards
4-32
-------
Blank
Blank
Stack Data
Cards
Blank
Card
Receptor
Elevation
Cards
Plant
Elevation Card
Namelist
Cards
Blank
Comment
Cards
Title
Card
Optional, required
only if IPTZ=1
Optional
FIGURE 4-7
INPUT DATA DECK SETUP FOR THE SINGLE SOURCE (CRSTER) MODEL
4-33
-------
The heading card contains the plant name in columns 1-24 and pollutant type
in columns 25-32. Information on this card is printed on the first line on
each page of output.
A set of comment cards follows the heading card. As many comment
cards can be included as needed. Each comment consists of 80 alpha-numeric
characters. A non-blank character must be included in one of the first four
columns. Comment cards are optional. The comments are listed on the first
page of printed output.
A blank card serves as a delimiter for card types that are optional
or that may be input in various quantities. One delimiter card must follow
the last comment card. If no comment cards are input, the delimiter card
must follow the heading card. (A delimiter card must also follow the last
receptor elevation card if IPTZ=1, and two delimiter cards must follow the
last stack data card.)
A set of namelist cards must be included after the comment delimiter
card. Table 4-7 is a list of mandatory variables that must be entered in the
namelist. Other namelist variables that can be included to select model op-
tions are listed in Table 4-8. Table 4-9 is a list of namelist variables that
must be assigned values if certain options in Table 4-8 are selected.
The set of namelist cards must be structured as follows:
&NAM1
name,=v, .name^^,... ,name.=v.
name.+1=vn+r...
...,name =v , & END
where:
name. = name of namelist item i
v. = value to be assigned to variable name.
The first column of each namelist card must be blank. The literal &NAM1
designates the beginning of the namelist input, and must start in column 2
of the first namelist card. The literal &END designates the end of the
4-34
-------
TABLE 4-7
MANDATORY NAMELIST VARIABLES
Variable
Name
IUR
RNG
ISS
ISY
I US
IUY
Type
Integer
Real
Integer
Integer
Integer
Integer
Description
1 = rural mixing heights to be used
2 = urban mixing heights to be used
Array of five receptor ring distances (km)
Surface station number for meteorological data
Year of record for surface meteorology (last 2 digits)
Upper air station number for mixing height data
Year of record for mixing height data (last 2 digits)
4-35
-------
TABLE 4-8
OPTIONAL NAMELIST VARIABLES
Variable Type
Description
Default
Value
ITAP Integer
IPTZ Integer
DAY
Integer
IMET Integer
IVT
Integer
0 = No tape output is to be generated
1 = Hourly and daily concentrations are to be
output to tape
0 = Flat terrain case; plant and terrain
elevations will be set to 0
1 = Plant and terrain elevations will be input
Array with dimension 366. Positions 1-366
correspond to the day of the year. Each day
assigned a value of 1 will be processed. For
non-leap years, DAY(366) must be set to 0.
0 = hourly meteorological data will not be
printed
1 = hourly meteorological data will be printed
for each day
Variable averaging time period. Only averaging
periods of 2,4,6,8 or 12 hours are permissible.
The default of 0 results in no variable aver-
aging period.
IQCK Integer 0 = Monthly emission rates will not be entered.
1 = Monthly emission rates will be input for
each stack (Namelist array QSSN must be
assigned the emission rate values - see
Table 4-9).
IVCK Integer
0 = Monthly stack gas exit velocity values will
not be input
1 = Monthly stack gas exit velocity values will
be input for each stack. (Namelist array
VSSN must be assigned the exit velocity
values - see Table 4-9),
366*1
4-36
-------
TABLE 4-8 (Continued)
OPTIONAL NAMELIST VARIABLES
Variable Type Description
ITCK Integer 0 = Monthly stack temerature values will not be
Default
Value
0
input
1 = Monthly stack temperature values will be
input for each stack (the namelist array
TSSN must be assigned the stack temperature
values - see Table 4-9).
ISC Integer 0 = Source contribution tables will not be
generated
1 = Source contribution tables will be printed
for days assigned a value of 1 in the DAY
array and the receptors assigned to the
namelist arrays SCI, SC3, SCN, SC24 and/or
SCAN (see Table 4-9).
4-37
-------
TABLE 4-9
NAMELIST VARIABLES REQUIRED FOR SPECIFIED OPTIONS
Option
Specified
Required
Namelist Variable
Type
Description
IQCK=I
IVCK=1
ITCK=1
ISC=1
ISC=1
ISC=1
and
IVT>0
ISC=1
ISC=1
QSSN
VSSN
TSSN
SCI1"
SC3f
SCNf
SC241
SCAN1
Real Twelve monthly emission rate
values (g s-1) for each stack.**
Real Twelve monthly exit velocity
values (m s~l) for each stack.**
Real Twelve monthly temperature values
(°K) for each stack.**
Real One distance-direction pair* for
each receptor to be printed in the
1-hourly source contribution table
(maximum of 20 pairs).
Real One distance-direction pair* for
each receptor to be printed in the
3-hourly source contribution table
(maximum of 20 pairs).
Real One distance-direction pair* for
each receptor to be printed in the
variable averaging time source
contribution table (maximum of
20 pairs).
Real One distance-direction pair* for
each receptor to be printed in the
24-hourly source contribution
table (maximum of 20 pairs).
Real One distance-direction pair for
each receptor to be printed in the
annual source contribution table
(maximum of 20 pairs).
F\
A distance-direction pair is the distance (km) and direction (tens of
degrees) of one of the 180 receptors surrounding the plant. The distance
must be one of the five ring distances assigned to the namelist array RNG.
The direction can have a value from 1 to 36.
**
*
Input is an array of 240 numbers, with positions 1-12 corresponding to the
monthly stack parameters for stack 1, positions 13-24 for stack 2, etc.
one of the namelist variables SCI, SC3, SCN, SC24 and SCAN is required
if ISC=1, although all can be specified. (See 4-49)
4-38
-------
name!1st input and must follow the last value assigned. If necessary, &END
can be entered on a separate card, starting in column 2. If a namelist
variable is an array, the values can be assigned to the array as follows:
array = v-j ,v2,.. .»vn,
where:
array = the name of the array
V; = the value to be assigned to array position i. If n is
less than the dimension of the array, only positions 1
through n are assigned values. If a series of contiguous
array positions are to be assigned the same value, n*v can
be entered where n = the number of repetitions of the value v.
(e.g., RNG = 0.5,1.0,2.5,3,5.5 assigns the five values to the five positions of
the array RNG; DAY=2*0,1,50*0,1,314*0 assigns 0 to all positions in the array
DAY except for positions 3 and 54, which are assigned a value of 1).
Discreet array positions can be assigned values by entering:
array (i) = v
where:
array = the name of the array
i = position within the array
v = value to be assigned to the array position i.
(e.g., DAY(3) = 1 will assign a value of 1 to position 3 of the array DAY).
After the last namelist card, the plant elevation and receptor ele-
vation cards must be entered. The plant elevation card contains the elevation of
the plant (feet above MSL) in columns 1-10. This card 1s not input if
the value of IPTZ=0. If a value of IPTZ=1 was entered in the namelist cards,
the plant elevation card must be input.
Thirty-six receptor elevation cards follow the plant elevation card,
corresponding to the 36 azimuths located every 10 degrees, starting with 10°
east of due north. Each receptor card contains the direction (tens of
degrees) in columns 1-2, followed by five receptor elevations (ft MSL)
4-39
-------
corresponding to the five ring distances assigned to the namelist variable
RNG. If the value of IPTZ=0, the receptor elevation cards must not be input.
If a value of IPTZ=1 was entered in the namelist input, the receptor elevation
cards must be input.
A delimiter card must be input following the last receptor elevation
card. If no plant and receptor elevation cards were input (IPTZ=0) then the
delimiter card must not be used.
After the receptor elevation delimiter card (if required), two cards
must be input for each stack that is to be processed. Up to 19 stacks can be
processed by the Single Source (CRSTER) Model. The first stack card is a stack
identification card containing a descriptive name for the stack (e.g., Federal
Power Commission number). The second stack card contains the emission rate
(g s ), stack height (m), stack diameter (m), gas exit velocity (m s" ) and
stack temperature (°K). If the option for entering monthly values of emission
rates (IQ.CK=1), exit velocity (IVCK=1) or temperature (ITCK=1) was specified
in the namelist input, the corresponding field on each stack data card may be
left blank. Table 4-10 contains a description of the format for each of the
fixed format card types. Coding forms for preparing the data for keypunching
are included in Appendix C.
b- Japejnput. A tape file of hourly meteorological data generated by
the Preprocessor program (see Section 4.2.4) must be input to the Single Source
(CRSTER) Model. The file should contain data for the most representative com-
bination of surface and upper air stations. The NWS surface and upper air
station numbers, and the year of record of the data on the tape must correspond
to the values assigned to the namelist variables ISS, ISY, IUS, and IUY (see
Section 4.3.3.a).
4.3.4 Output Data Description
The Single Source (CRSTER) Model generates printed output of the input
data and modeling results, and an (optional) output file of hourly, daily and
annual concentrations at each receptor point.
4-40
-------
TABLE 4-10
SINGLE SOURCE (CRSTER) MODEL FIXED FORMAT INPUT CARD DESCRIPTIONS
Card Type
Heading Card
Comment Card
Comment Delimiter Card
Plant Elevation Card
Receptor Elevation Card
Receptor Elevation
Delimiter Card
Stack Identification
Card
Stack Data Card
Stack Delimiter Cards
Contents
Card Columns
1-24
25-32
1-80
1-80
1-10
1-2
3-10
11-20
21-30
31-40
41-50
51-60
1-80
1-80
1-10
11-20
21-30
31-40
41-50
51-60
1-80
Format
6A4
2A4
20A4
SOX
F10.4
12
8X
F10.0
F10.0
F10.0
F10.0
F10.0
SOX
20A4
F10.4
10X
F10.2
F10.2
F10.2
F10.2
SOX
Description
Plant Name
Pollutant
Comments (first four
columns must contain at
least 1 non-blank
character)
Blank
Plant Elevation (ft MSL)
Direction (tens of
degrees)
Blank
Receptor Elevation (ft
MSL) for ring 1
Receptor Elevation (ft
MSL) for ring 2
Receptor Elevation (ft
MSL) for ring 3
Receptor Elevation (ft
MSL) for ring 4
Receptor Elevation (ft
MSL) for ring 5
Blank
Stack identification
Emission Rate (gm/sec)
Blank
Stack Height (m)
Stack Diameter (m)
Stack gas exit velocity
(m/sec)
Stack Temperature (°K)
Blank
4-41
-------
a. Printed Output. The printed output generated by the Single Source
(CRSTER) Model can be classified into the following categories:
• Card input data listing
• Meteorological data listing
• Modeling results for a standard run
• Modeling results for a source contribution run
A card input data listing is generated by each Single Source (CRSTER) Model.
run. The card input data listing consists of a run identification page,
a receptor data page, and one or more stack data pages.
The run identification page contains a listing of the comment cards
and namelist options. Pages B-5 and B-29 (Appendix B) are examples of run
identification pages. The first line on the page is a page heading printed on
every page of output from the Single Source (CRSTER) Model. This is followed
by the list of comment cards and a comparison of the upper air and surface
meteorological data sources and year of record on the input tape with those
specified in the namelist input; for example:
THIS IS A SINGLE SOURiCE (CRSTER) MODEL EXAMPLE RUN.
THIS RUN ILLUSTRATES THE USE OF THE FOLLOWING OPTIONS:
* SOURCE CONTRIBUTION
* RUN FOR A SINGLE DAY
* UNEVEN RECEPTOR TERRAIN
* NO HOURLY OUTPUT TAPE
* RURAL MIXING HEIGHTS
* VARIABLE AVERAGING TIME
CINCINNATI SURFACE
DAYTON UPPER AIR
MET FILE REQUESTED
STN NO. YR STN NO. YR
SURFACE 93814 64 93814 64
UPPER AIR 93815 64 93815 64
The program then prints options selected by the user in the namelist input,
and the value assigned to each of the 366 positions in the DAY array that
designates which days are to be modeled; for example:
4-42
-------
PLANT LOCATION: RURAL
NO TAPE OUTPUT
0 VALUES REQUIRE MONTHLY INPUT
V VALUES REQUIRE MONTHLY INPUT
T VALUES REQUIRE MONTHLY INPUT
DAY— 0000000000 0000000000 0000000000 0000000000 0000000000
0000000000 0000000000 0000000000 00 00000000 0000000000
0000000000 0000000000 0000111111 11 11000001 1111111110
0000000000
0000000000
0000000000
0000000000
0000000000
0000000000
0000000000
0000000000
0000000000
000000
0000000000
0000000000
0000000000
0000000000
0000000000
0000000000
0000000000
0000000000
0000000000
0000000000
0000000000
0000000000
The receptor data page tabulates the five receptor ring distances,
the plant elevation, and the 180 receptor terrain elevations. Elevations are
listed in both feet and meters. Pages B-6 and B-30 are examples of receptor
data pages, and a partial listing of the format is shown below:
RING DISTANCES(KM)=
.90
1.50 2.00 3.80 6.20
PLANT ELEVATION (FEET ABOVE SEA LEVEL)— 492.0
RECEPTOR ELEVATIONS (FEET ABOVE SEA LEVEL)
DIRECTION RINGH1 RINGH2 RIN6«3 RINGS'* RIN6B5
1
2
3
4
540.0
550.0
525.0
490.0
500.0
550.0
615.0
720.0
470.0
575.0
625.0
640.0
510.0
660.0
710.0
720.0
460.0
460.0
460.0
540.0
PLANT ELEVATION (METERS ABOVE SEA LEVEL)— 150.0
RECEPTOR ELEVATIONS (METERS ABOVE SEA LEVEL)
RING81 RING82 RINGW3 RING84 RIN6H5
164.6
167.6
160.0
149.4
152.4
167.6
187.5
219.5
143.3
175.3
190.5
195.1
155.4
201.2
216.4
219.5
140.2
140.2
140.2
164.6
Information output on stack data pages consists of a stack identifi-
cation list followed by an input stack data list. The format of the stack data
list depends on whether or not monthly values were input for emission rate,
exit velocity and/or temperature. If monthly values were input for any one
of these parameters (i.e., IQCK=1, IVCK=1 or ITCK=1), twelve lines of print
are generated for each stack, listing the monthly values. Pages B-7 and B-8
are examples of the monthly stack data, and a partial listing of this format
is shown below:
4-43
-------
STACK
10NTH
JAN
FEB
MAR
APR
MAY
EMISSION RATE
(GMS/SEC)
36.2200
36.2200
36.2200
36.2200
36.2200
HEIGHT
(METERS)
83.20
83.20
83.20
83.20
83.20
DIAMETER EXIT VELOCITY TEMP
(METERS)
3.05
3.05
3.05
3.05
3.05
(M/SEC)
19.42
19.42
19.42
19.42
19.42
(DEG.K)
428.00
428.00
428.00
428.00
428.00
VOLUMETRIC FLOW
(M**3/SEC)
532.74
532.74
532.74
532.74
532.74
If single, annual values for stack parameters are input instead, one line of
print is generated for each stack as illustrated in the example on page B-31
and shown below:
STACK MONTH EMISSION RATE HEIGHT
(GMS/SEC) (METERS)
1 ALL 36.2200 83.20
DIAMETER EXIT VELOCITY
(METERS) (M/SEC)
3.05
19.42
TEMP VOLUMETRIC FLOW
(OEG.K) (M**3/SEC)
428.00 141.89
Following the card input data listings, the Single Source (CRSTER)
Model prints the hourly meteorological data for each day processed (i.e., each
day corresponding to a position in the namelist array DAY that was assigned a
value of 1). The listing of the hourly meteorology is composed of ten lines
of print. The first line lists the year (yy), month (mm) and Julian day (ddd)
of record as follows:
JYR = yy IMO = mm JDAY = ddd
.-1
The next five lines list 24 values of stability class, wind speed (ms ), am-
bient temerature (°K), flow vector (degrees) and randomized flow vector
(degrees), respectively. The last four lines of the hourly meteorology listing
contain the hourly mixing height values (m) for rural (HLH1) and urban (HLH2)
cases. Examples of the meteorological data listing are illustrated on pages
B-8 through B-13 and shown below:
4-44
-------
JYR=64 IMO= 5 JOAY=125.
ISTAB= 666665,i»33221212293«56777
AWS= 2.1 2.6 2.1 2.1 2.1 2.6 1.5 3.1 1.1 3.6 i'.l 2.6 3.6 2.1 3.6 2.6 3.1 3.1 2.6 2.1 1.5 1.0 1.0 1.0
TEMP- 289. 288. 287. 286. 286. 286. 287. 290. 292. 29*. 297. 299. 299. 299. 300. 299. 300. 299. 298. 291. 291. 289. 287. 28(1.
AFV= 270. 320. 330. 330. 10. 350. 50. 20. 20. 30. 20. 320. 310. 350. 310. 360. 310. 330. 330. 330. 350. 10. 310. 310.
AFVR= 267. 322. 335. 331. 8. 351. 53. 21. 23. 29. 23. 320. 313. 351. 315. 1. 310. 327. 333. 327. 319. 12. 311. 3J8.
HLH1= 2010. 2032. 2051. 2077. 2099. 86. 362. 639. 915. 1192. 1168. 1715.
2021. 2298. 2298. 2298. 2298. 2298. 2298. 2291. 2287. 2279. 2271. 2261.
HLH2= 261. 261. 261. 261. 261. 310. 585. 829. 1071. 1319. 1561. 1808.
2053. 2298. 2298. 2298. 2298. 2298. 2298. 2018. 1563. 1079. 595. 111.
The Single Source (CRSTER) Model calculates pollutant concentrations
for 1-hour, 3-hour, 24-hour, and annual* averaging periods. In addition,
calculations are made for a variable averaging period (optional) which can be
selected as 2, 4, 6, 8 or 12 hours by assigning a value to the IVT namelist
variable (see Section 4.3.3.a). Modeling results printed out by a standard
run of the Single Source (CRSTER) Model include the following:
• A listing of daily maximum 1-hour and 24-hour concentrations
• A table of the annual mean concentrations at each receptor
• A table for each averaging period, listing the highest con-
centration at each receptor and the maximum of these values
• A table for each averaging period (except the annual average),
listing the second-highest concentration at each receptor
point and the maximum of these values
• A table for each averaging time period (except the annual
average) of the 50 highest concentrations for the entire
year.
The daily maximum 1-hour and 24-hour concentrations are listed for
each day that is processed. If the meteorological data are being printed
(i.e., MET=1), the maximum 1-hour and 24-hour concentration for each day are
printed following the meteorology for that day, as follows:
MAX HOURLY M A X 2 4 - H 0 U R
c
125 18.457 2.426167-03 32
DAY RATIO CONCENTRATION DIRECTION DISTANCED) HOUR CONCENTRATION DIRECTION DISTANCED)
•IOK 1A.UR7 9.ii?ft1ft7-03 32 .90 12 1.314483-04 £
Note that all "annual average" concentrations output by the model represent
averages taken over only those days for which the model was run. Thus, only
if an entire year of data is processed will these be true annual averages.
4-45
-------
The examples on pages B-8 through B-13 illustrate this format. If the meteo-
rological data are not printed (i.e., MET=0), the heading for the daily maxi-
mum 1-hour and 24-hour concentrations is printed once on each page, followed
by one line of print for each day. Up to 50 lines are printed on each page.
After the maximum 1-hour and 24-hour concentrations are printed for
the last day processed, the table of annual mean concentrations is printed.
Page B-14 is an example of the annual mean concentration table. The first line
o
on this page (after the heading) lists the maximum annual concentration (g m~ )
at any receptor, and the direction (tens of degrees) and distance (km) of its
location. This is followed by a table of annual mean concentrations (g m )
at each of the 180 receptor points. A partial listing of this format is given
below:
1.52284-06
2.15679-06
2.52828-06
2.46767-06
2.69925-06
4.78657-06
1.32464-05
2.41952-05
4.97701-05
3.16181-05
6.93091-06
2.19752-05
3.37759-05
3.52931-05
3.28854-05
1.13684-05
3.59051-05
5.86661-05
6.91794-05
1.79495-05
9.61468-06
1.85926-05
2.09605-05
2.94097-05
2.14584-05
MAXIMUM MEAN CONC= 6.9179-05 DIRECTION 4 DISTANCE= 3.8 KM
ANNUAL MEAN CONCENTRATION AT EACH RECEPTOR
RANGE .9 KM 1.5 KM 2.0 KM 3.8 KM 6.2 KM
DIR
1
2
3
4
5
The highest and second highest concentration tables have the same
format for each averaging period. The first line after the page heading
o
contains the yearly maximum highest or second-highest concentration (g m~ ) at
any receptor; the receptor location (direction in tens of degrees, and dis-
tance in km) and time interval within the day* of the maximum. The table of
highest (or second-highest) concentrations at each of the 180 receptors follows,
This table contains 11 columns, one identifying the receptor azimuths and five
*
A time interval within the day is not listed for the 24-hour concentration
tables since there is just one 24-hour period per day.
4-46
-------
pairs of columns, one for each of the ftve receptor ring distances. Each
pair of columns contains- one column of concentrations and one identifying the
Julian day and associated time period. The Julian day and time period iden-
tifier are enclosed in parenthesis and separated by a comma. Table 4-11 is a
list of time period identifiers for any given averaging period. Examples of
highest and second-highest concentration tables for 24-hour, 8-hour, 3-hour
and 1-hour averaging periods are given on pages B-15 through. B-22. The basic
format is summarized below:
YEARLY MAXIMUM 24-HOUR CONC= 2.4790-04 DIRECTIONS 4 DISTANCE: 3.8 KM DAY=129
RANGE .9 KM
DIR
2.2800-05 (125)
1.4440-05 (126)
1.6584-05 (126)
3.4162-05 (142)
4.1920-05 (142)
HIGHEST 24-HOUR CONCENTRATION AT EACH RECEPTOR
1
2
3
4
5
6
2.5603-05 (142)
1.5 KM
3.9212-05 (125)
8.5751-05 (126)
1.4062-04 (126)
2.0398-04 (145)
1.6454-04 (145)
7.4144-05 (143)
2.0 KM
5.6554-05 (126)
1.2960-04 (126)
1.8687-04 (126)
1.7518-04 (126)
1.5672-04 (145)
9.6394-05 (143)
3.8 KM
7.1670-05 (126)
1.4925-04 (127)
1.8580-04 (127)
2.4790-04 .(129)
7.9992-05 (144)
6.4913-05 (143)
6.2 KM
4.5615-05 (126)
8.7575-05 (127)
8.5478-05 (126)
1.0550-04 (126)
7.9473-05 (145)
4.1824-05 (130)
The fifty highest concentrations ranked in descending order are printed
for each averaging period. Pages B-23 through B-26 are examples of tables of
fifty highest concentrations for the 24-hour, 8-hour, 3-hour and 1-hour averag-
ing periods, respectively. The tables list the Julian day number, concentration
(gnf ), receptor direction (tens of degrees) and distance (km), and time period
identifier (see Table 4-11). A sample of the format is given below:
4-47
-------
TABLE 4-11
TIME PERIOD IDENTIFIERS AND CORRESPONDING HOURS OF THE DAY
(LOCAL STANDARD TIME)
Time Period
Identifier #
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
1-Hour
0-1
1-2
2-3
3-4
4-5
5-6
6-7
7-8
8-9
9-10
10-11
11-12
12-13
13-14
14-15
15-16
16-17
17-18
18-19
19-20
20-21
21-22
22-23
23-24
2-Hour
0-2
2-4
4-6
6-8
8-10
10-12
12-14
14-16
16-18
18-20
20-22
22-24
-
-
-
-
-
-
-
-
-
-
-
-
Averagi
3-Hour
0-3
3-6
6-9
9-12
12-15
15-18
18-21
21-24
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
ng Period
4-Hour 6-Hour 8-Hour
0-4 0-6 0-8
4-8 6-12 8-16
8-12 12-18 16-24
12-16 18-24
16-20
20-24
- - -
- - -
-
- - -
- - -
- - -
-
-
- - -
- - -
- - -
- - -
- - -
_
- - -
-
_
- - -
12-Hour
0-12
12-24
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
4-48
-------
MAXIMUM DAILY CONCENTRATIONS
DAY 24-HOUR CONCENTRATION DIRECTION DISTANCE
129 2.4790-04- 4 3.80
140 2.3068-04 7 6«20
145 2.0398-04 4 1-50
141 1.9818-04 20 2.00
126 1.8967-04 4 3.80
127 1.8588-04 3 3.80
130 1.8459-04 7 6.20
128 1.7147-04 4 3.80
142 1.6583-04 15 1.50
131 1.6547-04 10 3.80
134 1.5318-04 7 6.20
146 1.4285-04 26 2.00
125 1.3145-04 2 3.80
144 1.2320-04 4 3.80
149 1.2135-04 15 3.80
143 9.6394-05 6 2.00
147 9.1085-05 3 3.80
132 8.6754-05 30 6.20
133 8.5918-05 7 6.20
149 5.6766-05 21 6.20
A source contribution run of the Single Source (CRSTER) Model gene-
rates a source contribution table for each day processed. The source con-
tribution table is printed after the meteorological data for that day. If
several days are run, the source contribution tables and meteorological data
are interspersed. If the meteorological data output was suppressed for the
run (IMET=0), only the source contribution tables are printed. The source
contribution tables list individual stack contributions to the total concen-
tration at selected receptor points for each averaging time period. Different
receptor points can be specified for each averaging period, and source con-
tribution listings will be output for only those averaging periods for which
receptor points have been input. The receptor points are specified in the
namelist arrays SCI, SC3, SCV, SC24 and SCAN (see Section 4.3.3.a). The
concentrations for each time period within the day are printed in the order
that they are calculated, and so at first glance the output as shown in
Table 4-12 may be confusing. One line is printed for each combination
of receptor, averaging period, and time period identifier. The data printed
on each line are the averaging period (INT), the Julian day number (DAY), the
time period identifier (PER), the receptor identification (R/DR, i.e., ring
number/direction), ten stack concentrations (gnf ) and the total concentra-
tion. If more than ten stacks were modeled, a second line is printed con-
taining the remaining stack concentrations. Pages B-32 through B-34 illus-
trate a source contribution table for a single day using four receptor points
for each standard averaging period and a variable averaging period of 8 hours.
4-49
-------
-I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I. I I I I I I I I I I
*i cccNjojipo^-covor^ofOfocrfO^cccjip^-muoaD-^fOLrojcvJo* r^t-a-o^tctT^^r- jioh-iooocjcrr^vOr^-a-fO
f— in*H<\jin.3-»-ih-f-Lncco*cccK")oc" o^«Hoin*DvO;±^«-Hr~r~-v0rOf^»-iK>r--.r\ors-ro\cinou»-i«-i3-cc*-»cooc\jj'5
o :rc\j~cjo^rcT'roro:rir>crooC'tz;of^fOvcC'HC\ju"}in.3'\co^toa:^axoif>=f3-rj7 oi^nr^cj'ipojfor^-coh-
UJ OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOGOCGGGOGG
fO O OOOOOGOOOOOOOOGOOOGOOOOOOOGGOOOOOGGOOGOOOOOOOOOOOG
» CC CCOOOOOCCOOOCOOOOOOCCOOOOOOOOOCOCOOOOOCOGOGOCCOGCO
* 3 oooooooooooooooooooooooooooooooooooooooooooooooooo
£ O •
X in OGGOOGOGOGOGGOGOGOGOOGOOOGGOGGGOOOOOOGOOGOQGOOGOOO
I O>
UJ •*
Q l/l UJ OOOOOOOGOOGOOOOGOOOOOOOOGGOOGGOOGOOOOGOOOO^GOCOGOO
fZ I— U OCOOOCOOOOOOOOOOOOCOOOOOOOOOOOOCOOOOOOOOaCOOCGOCCO
K •-" j cc GoooooooocoooGoooooaoooaoooooGGOocoooooooaccoooooo
*• ~ D OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOG
-—• •• in OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOGOOOOOOOOOO
UJ K UJ
1_ >-• i < 3 UJ ocoooooooooooooooooooooooooaoooooooaoooooooooooaoc
r~ 3 o u ccooaoooooocaoooooooooooooooooooooooocooooooooococ
*—' o in cc ocoacoaoooooooooooaooaoooocooooooooooooooooocoooco
^-* D OOOOOOOOOOOOOOOOOOOOOGOOOOOOOOOOOOOOOOOOOOOOOOOOOO
cc LL o • • «•
UJ >-> O m OOOOGOOGGOOOOGOOGGOOOGGOOOOOGOGOOOGOGOOOOOOOOOOOOO
o <
^ o r^
O u OCOOOOOCOGOOOCIOOOOOCOOOOOOOOOOOOOOOOOOOOOOCCOOOOOO
CO u O OOOGGOOOOOOOOOGOOOOOOOOOOOOOOOOOOOOOCOOOOOOCCOOOOO
UJ CC OOGOCGOGOOOOOGOGOOOOOOOOOOaOOOOOOOOOOOOOOGOOCOOOOO
UJ in o OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOGOOOOOOOOOOOOOO
_| N O
(£ £ I/) OOGGOOOGOOGOOGGOOOOOOGGOOOOOOOOOOOOOOOOOOOOOOOOOOO
>- UJ UJ OOOOOOOOOOOOOOOOOOOGOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
UJ >-< J O OOOOOOCOOOOGOOOOOOOOOOOOGOOOOOO'OOOOOOOOOOOOOOOOOOC
T" 2 CO CC GOOOOOOOOOOOOOGOOOOOOOCGGOOOOOOOOOGOOOOOOOJ2OOOOOOO
U-. 3 < O OGGGGGOGOGGGGGOGGGGOOGOGOOOOOGGOOOGGOOGGGodGGGGQOO
I- O
Z (/> OOGOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOGOOOOOOOOOO
« o
OJ >-
i— ca
^. j_ Lo
Z3 >-i 3 UJ OOOOOOOOCOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOCOCCC
UJ Q_ Z CD u oooooooaooaooooooooooooooooooooooooooooaooooGocooc
I I— UJ •" CC OOOCGOOOGOOOOOOOOOOCOOOOOOOOOOOOOOOOOOOOOOOOOCOOOa
rr\ —i L ID OOOOOOOOOOGOOOOOOGOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
5: o I- o
!*• *-^ 2 'T oooooooooooooooooooooooooooooooooooooooooooooooooo
UJ (j om cjd-coin coo^^ro irt«c
—I a- OJ^( tOOU.-l.-l COrt'Hrt tOCJ
CO UJ II III) I I I I II
-ld-OOOOOOOO*-linOJ=frOOOOt"-.O
CQ D I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I
i — i _j uj tninoo^-*\0o\cc7rr>ovr-~:3-cjir>Fooini-ir^cooNP~'Oot/}\oin'Ococ\j (Ajn oroojor^-^j'd-iDr^-cjrofooo^^o^H
cf _i u o^ui^K)a^aaj^>ocoin^c\jci-oocortfOT-icr'Ooii-cvic\ivO'OtocEis-I-i3-ioeovOin(>r-CT'CO(Oo/evior^
i _ o cc oc><\)>o
^Z Q. o ocoLOo
g o >.
o in ^toKi^
C_> I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I
2 i"J ----i-^i->-~---
^ o
=; ct
~ 3
oo o
in
.
^ i i r i i i i i i i i i i i t i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i
cc uj
u
uj ce
_J Z)
a. o
z i/i
<
UJ
UJ
•s.
<
.
<
_1
Q-
4-50
-------
b. Tape Output. An output tape can be generated by a standard run of
the Single Source (CRSTER) Model containing hourly, daily and annual concentra-
tions for each receptor point. The tape is generated if a value of 1 is assigned
to the namelist variable ITAP (see Section 4.3.3.a). Each record on the file
contains 180 concentration values.written with a FORTRAN unformatted WRITE state-
ment. The 180 values correspond to the concentrations along each of the 36 azi-
muths, at each ring distance as follows:
Values 1-36 correspond to Ring 1
Values 37-72 correspond to Ring 2
Values 73-108 correspond to Ring 3
, directions 1-36
, directions 1-36
, directions 1-36
• Values 109-144 correspond to Ring 4, directions 1-36
• Values 145-180 correspond to Ring 5, directions 1-36
The records are arranged in the file as follows:
• 1-hour concentrations for Hour 1 of Day 1
• 1-hour concentrations for Hour 2 of Day 1
1-hour concentrations for Hour 24 of Day 1
24-hour concentrations for Day 1
1-hour concentrations for Hour 1 of Day 2
• 1-hour concentrations for Hour 24 of Day 2
• 24-hour concentrations for Day 2
1-hour concentrations for Hour 1 of Day N
• 1-hour concentrations for Hour 24 of Day N
• 24-hour concentrations for Day N
• "Annual mean" concentrations
where N is the number of the last day processed.
4-51
-------
Records are output only for days corresponding to positions in the
namelist array DAY that were assigned a value of 1 (see Section 4.3.3.a). The
"annual mean" concentrations on the last record are averages based only on
data from these days, i.e., are not true annual averages unless the entire
year is processed. The record of annual mean concentrations is not output
for source contribution runs.
4.3.5 Diagnostic Messages
Three types of diagnostic messages may be generated by the Single Source
(CRSTER) Model, namely, fatal error messages, non-fatal error messages and
informative messages.
Fatal error messages are printed when an error is detected that causes
the program to stop execution. This is normally due to an error with some
input data. Table 4-13 is a list of the possible fatal error messages and
the corrective action that should be taken.
Non-fatal error messages are printed when a data item does not conform
with input specifications, but the program makes an assumption or uses default
values, allowing execution to continue. Table 4-14 Is a list of the non-fatal
error messages that may be output by the Single Source (CRSTER) Model. If a
non-fatal error message is printed, the user must determine if the assumptions
made by the Single Source (CRSTER) Model were acceptable. If not, the input
data item in question should be corrected and the model rerun.
Informative messages are printed when the user has selected an option
that differs from the default. Table 4-15 is a list of the informative messages
that may be generated by the Single Source (CRSTER) Model. No user action is
required in response to an informative message.
Lastly, Table 4-16 is a Julian day to calendar day conversion chart. Users
who experience significant problems in executing the Single Source (CRSTER) Model
may receive technical assistance by telephoning the Chief, Modeling Support Section,
Source Receptor Analysis Branch in Durham, NC at 919-541-5335 or, using FTS,
629-5335
4-52
-------
TABLE 4-13
SINGLE SOURCE (CRSTER) MODEL FATAL ERROR MESSAGES
BAD RECEPTOR REQUESTED FOR SOURCE CONTRIBUTION — END RUN
Description: The source contribution option was requested (ISC =1),
but a value was entered for a distance in one of the SCI,
SC3, SCN, SC24 or SCAN arrays that did not correspond to
any of the five distances in the RNG array.
Action: Check all input distances and correct any that are
incorrect.
NO MATCH FOR SURFACE/UPPER AIR REQUEST
MET FILE REQUESTED
STN NO. YR STN NO. YR
SURFACE issi isyi iss isy
UPPER AIR iusi iuyi ius iuy
Description: The requested surface station number (namelist item ISS
with value iss) or year (namelist item ISY with value isy)
or upper air station number (namelist item IUS with value
ius) or year (namelist item IUY with value iuy) did not
match the corresponding value on the input meteorological
data tape (values issi, isyi, iusi and iuyi, respectively),
Action: Correct any namelist items that did not match the values
on the tape, or check to be sure that the correct meteor-
ological data file was specified in the ECL.
INVALID IUR: iur, ABORT RUN
Description: The value iur was input for namelist item IUR. Only the
values 1 (for rural) or 2 (for urban) can be used.
Action: Enter the correct value for IUR.
4-33
-------
TABLE 4-13 (Continued)
SINGLE SOURCE (CRSTER) MODEL FATAL ERROR MESSAGES
*** MAXIMUM OF 19 STACKS EXCEEDED — RUN TERMINATED
Description: Data for more than 19 stacks were entered.
Action: Reduce the number of stacks by combining similar stacks,
if possible.
*** RECEPTOR HEIGHT GREATER THAN HEIGHT OF LOWEST STACK - RUN TERMINATED
Description: Receptor elevations cannot exceed the elevation of the
top of the lowest stack (stack height plus the terrain
elevation at the source).
Action: Change any receptor elevations that exceed the height of
the lowest stack to be equal to that value, and rerun.
4-54
-------
^_
1—
1
^>
LU
CO
cC
|—
^)
1 1 1
Sa-
to
y
o:
o
fv*
LU
5ji
l~™
^£
U-
t
Z
O
1
LU
O
i
^-^
o:
LU
t—
to
<*x
0
LU
CJ
o:
o
to
LU
I
C!3
•^
I — t
to
C 1
o
x> o
QJ
^ ^D
ro
C
QJ . .
= t
1 — +•>
ro 3
> O
O) QJ
J= Q.
l"
• O
i— C
So
O <4-
<4- °
O QJ
QJ i—
3 fQ
i— >
ro
> ro
QJ
"O E
O) 3
C V)
•i- ro
to <—
re QJ
•a
^ ^5
jQ 5»I
>> QJ
i — O
E s-
O 3
0
E to
(O
O QJ
a. en
• 3
I— to -a o
LU -r- ••- QJ
to i— i— X
QJ ra QJ
E =>
C ro C to
E -r- QJ
3
QJ V) E
LU -E ro •!-
=5 1— S 4->
_J
^> • •
C.
Q. 0
1 — -M
1—1 (-1
•r—
Q S-
I-H 0
0
i — i
C
1
•a c
QJ O
S- 0
Q)
•M -0
c c
O> ro
3 C
ro rO
> £
QJ QJ
g"* 4-*
^_
^M)
ro
r-* (4—
•lu'
S-
O 0
0 M-
c^»
O OJ
3
QJ •—
3 ro
r- >
rO
> ra
rO to
"O c
OJ 3
c to
ai to
•i- ro
(/I
to i —
ro OJ
•a
QJ Q
O ~^
>*, QJ
r- O
c s-
O 3
0
c to
ra
O QJ
r-j en
I— c
Q- ••—
l-H tO
E QJ •
O QJ JZ C
-(-> h- 0
0 T- -r-
l— +->
+-> • 3
1— to -a o
LU •!- -i- QJ
tO i— i— X
QJ rO QJ
c: ro c to
C •!- QJ
3
QJ to c
LU .C rO 'r—
^ 1 — S +J
>
C
M O
i fr.
a. •)->
1—1 Q.
•i—
Q S-
I-H O
—I V)
^> f~^
1— 1
OJ
1 ^~
a> JD
3 03
^— *r—
(0 S-
> ra
>
a>
J= 0
I— C
• o
^ <+-
o
^.
O QJ
3
CO i—
^O 5*
A
^d- ro
CM to
o"H
3
>4- W1
O to
(O
QJ
3 i—
r— QJ
ra "O •
> o c
S 0
ra T-
QJ +J
XI O 3
QJ t- O
C 3 QJ
CT> O X
•i- tO QJ
i/l
to QJ to
rO i— QJ
CT> 3
QJ C E
_Q -r- -r-
tO 4->
>> C
i— QJ 0
C -E 0
0 H-
-o
C C
ro • ro
0 T3
I— r^ -a
>• ra o
K-l > -1-
C S-
E M- QJ
QJ CL
CD +-> to
•i— ra QJ
O 3 E
h- 4-> T-
(/) - — 4->
1— -r- C
LU i — — cn
tO QJ C
* ra QJ en
c c s- ra
QJ S-
QJ -(-> QJ
JC C >
LU I — QJ rO
3
«=C
> C
0
> X>
HH Q.
•r—
O S_
1-1 o
_l to
«< QJ
> Q
l-H
tO
LU
£»I
I-H
H-
C
2T
o
1 — 1
h-
— 1
t—
LU
_J
LU
ce
0
i—
a.
LU
o
LU
DC
1
1
~^"
O
1— t
1—
5
LU
i
LU
1—
^^
e^
1
Q.
•^
«^
3Z
1—
to
to
LU
_J
Zi
o
I—
<^
^>
LU
LU
Qi
O
l-
Q.
LU
O
LU
O
•(->
•
i— C
(O T-
3 ro 4->
O~ i- C.
QJ S- ro
QJ •—
S- 4J Q.
O
+J 4J QJ
Q. C J=
QJ ro •!-»
O i—
QJ Q. C
S- ra
QJ -E
rO JC •*->
4-3
S- S-
O C QJ
>t- ro S
J= O
E +-> r-
O
•r- i. E
-M QJ -r-
rO 3t ro
> 0 S-
QJi— S-
•— QJ
QJ to •!->
ra
•r~ 4^
ro 1- ••-
S- 0 2
J- •)->
QJ Q. to
+-> Ol +->
O E
QJ QJ *r-
J= S- O
+-> Q.
ro
to S-
4-> -4-> O
QJ ro •<->
{/) Q_
c ai
>>T- O
i— ro QJ
r— S- S-
ra i-
0 QJ M-
•^ 4-> O
ra QJ S-
E J= OJ
O +-> J3
•(-> E
3 4- 3
ro -i- E
i— C QJ
QJ O -E •
T3 •!- +J C
0 4-> 0
SI ra to v-
OJ QJ ra
O i — C >
S- QJ QJ
3 QJ (—
O •!-> 3 QJ
tO E ^±
ra ,0 c
QJ •—>••-
i— a. ra
en i-
E QJ QJ 4-
•r- J= JT CU
to +-> i— •«->
• •
E
o
.J3
a.
L.
0
QJ
O
4-55
-------
TABLE 4-15
SINGLE SOURCE (CRSTER) MODEL INFORMATIVE MESSAGES
Q VALUES REQUIRE MONTHLY INPUT
Description: A value of 1 was entered for namelist Item IQCK, specifying
that monthly emission rates are to be used. The nantelist
array QSSN should have been assigned 12 emission rate
values to each stack.
V VALUES REQUIRE MONTHLY INPUT
Description:
A value of 1 was entered for namelist item IVCK, specifying
that monthly stack gas exit velocity values are to be used.
The namelist array VSSN should have been assigned 12 exit
velocity values for each stack.
T VALUES REQUIRE MONTHLY INPUT
Description: A value of 1 was entered for namelist item ITCK, specifying
that monthly stack gas exit temperatures are to be used.
The namelist array TSSN should have been assigned 12 values
for each stack.
MET DATA WILL NOT BE PRINTED
Description:
A value of 0 was entered for namelist item IMET specifying
that the hourly meteorological data for each day is not to
be printed.
4-56
-------
TABLE 4-15 (Continued)
SINGLE SOURCE (CRSTER) MODEL INFORMATIVE MESSAGES
N GREATER THAN 45
X = x, YD = A , H = h, TOT = t, SUM = s
Description: This message is printed when the number of iterations in
calculating multiple eddy reflections exceeds 45. The
value x is the ring distance (m) at which the calculation
is made, Ay is the crosswind distance (m) of the receptor
from the plume center line, h is the effective height (m)
of the stack adjusted for terrain, t is the current eddy
reflection calculation, and s is the sum of all eddy re-
flection calculations. No further iterations are made,
and the current value of s is used to calculate the con-
centration.
DAY -- n NO CALCULATIONS PERFORMED - NO DIRECTIONS SELECTED
Description: This message is printed only when no calculations are made
for an entire 24 hour period. This condition can only
occur if the atmospheric stability class on the tape for
the entire day was stability class 7.
4-57
-------
*
o:
s
>-
Q_
2
<_>
o
»—(
10 oo
I LU
<1- >
*z
uj o
CO
? I
CVJCMCXJCMOslCVJCJCMOslCMCOC^
O O O O f
CSJOJCVJOJCSJCVJCMPJCVJOI
CMCXJCMOJCvJCMCMCMCMCMOOOO
a)cbobcbboc6cbCT>CT>o>(7>dicrvd^CTiCTiCT)Oooooooooor—i—r—
r— i— r— •— i— I— r— r— r— i— i— i— r— r— •— I— t— CMCSICVJCMCVJCMO-JCMOJCSJCSJCvJCSJCVJ
CMC\JOJOJCMCMCXIC\IC\lC\JrOOO
CVJ
CO
_ O
CVICMCMCMCSJCVJCSlCMCMCJtn
"^ ---------- 0000*3-'
r^-r— ^.— r^-i— r^r— r—i— CVJCVJCSJCMCVJOJOJOJCSicJOOOO
__ _ _ t—~ f\j f^o *^- _ __
CT>aicr>cr>cntncncrioooooooooOr—•—i—r—^-»—i—i—i-^r—CM
CO
CMCMCSJOJC\JCNJC\JCSJC\JOJ
CVJCVJCJCJCMCVJCMCMOJCMCOCO
CVJCMCMCVJC\JCVJ<\IC\I
-------
5. REFERENCES
1. Briggs, G.A., Plume Rise. AEC Critical Review Series TID-25075, 1969.
2. Briggs, G.A., Some Recent Analyses of Plume Rise Observations, paper
presented at the 1970 International Air Pollution Control Conference,
Washington, DC.
3. Briggs, 6.A., "Discussion of Chimney Plumes in Neutral and Stable
Surroundings", Atmospheric Environment, 16: 507-510, 1972.
4. Pasquill, F., Atmospheric Diffusion, D. Van Nostrand Company, Ltd.,
London, 2nd Edition, 1974.
5. Gifford, F.A., "Uses of Routine Meteorological Observations for Esti-
mating Atmospheric Dispersion", Nuclear Safety, 2: 47-51, 1961.
6. Turner, D.B., "A Diffusion Model for an Urban Area", Journal of Applied
Meteorology. 3; 83-91, February 1964.
7. Turner, D.B., Workbook of Atmospheric Dispersion Estimates, Office of
Air Programs, Environmental Protection Agency, Publication No. AP-26,
Revised, 1970.
8. Bierly, E.W. and Hewson, E.W., "Some Restrictive Meteorological Condi-
tions to be Considered in the Design of Stacks", Journal of Applied
Meteorology. 1: 383-390, March 1962.
9. Holzworth, G.C., Mixing Heights, Wind Speeds, and Potential for Urban
Air Pollution throughout the Contiguous United States, Environmental
Protection Agency, Publication No. AP-101, Division of Meteorology,
Research Triangle Park, NC, January 1972.
10. Sellers, W.D., Physical Climatology, U. of Chicago Press, 1965.
11. Khanna, S.B., Handbook for UNAMAP. Walden Division of Abcor, Inc.,
Wilmington, MA, March 1976.
12. Guideline on Air Quality Models and Associated Data Bases (Draft),
Source Receptor Analysis Branch, Environmental Protection Agency,
Research Triangle Park, NC, February 1977.
13. Levy, A., Drewes, D.R., and Hales, J.M., S02 Oxidation in Plumes: A
Review and Assessment of Relevant Mechanistic and Rate Studies, EPA
Publication No. EPA-450/3-76-022, Research Triangle Park, NC,
September 1976.
14. Survey of TD-1440 and List of Upper Air Stations, available from the
National Climatic Center, Asheville, NC 27711.
5-1
-------
15. National Heather Service Offices and Stations. 17th Edition, January
1977, available from the National Weather Service, Silver Spring, MD
20910.
16. Larsen, R.I., A Mathematical Model for Relating Air Quality Measurements
to Air Quality Standards, Environmental Protection Agency, Office of
Air Programs Publication No. AP-89, Research Triangle Park, NC,
November 1971.
17. Card Deck 144 MEAN Hourly Surface Observations Reference Manual 1970.
available from the National Climatic Center, Asheville, NC 27711.
18. Woolf, H.M., On the Computation of Solar Elevation Angles and the
Determination of Sunrise and Sunset Times, NASA Technical Memorandum
NASA TM X-1646, Washington, D.C., September 1968.
5-2
-------
APPENDIX A
PROGRAM SOURCE LISTINGS
A-l
-------
Listings of the FORTRAN source code for the Preprocessor and Single
Source (CRSTER) Model are contained in this section. The various program
elements begin on the following pages:
Program Element Page Number
Preprocessor Programs A-3
Single Source (CRSTER) Model
Main Program A-ll
Subroutine CRS A-18
Subroutine SIGMA A-44
Subroutine BEH072 A-49
Block Data A-53
A-2
------- An error occurred while trying to OCR this image.
------- An error occurred while trying to OCR this image.
------- An error occurred while trying to OCR this image.
-------
oooooooooooooooooooooooooooooooooooooooo
THojrod-in\or^cooo«-ic\i»o j-.m \o.r- co o o .H CM ro ;± in o,r- co CT> o .H.CM 10 * tn \o.r>> to a- o
cMCMcMCMCMCMCMCMCMionroroioroiorotorod-d-d- * ^ ^-d-d-d-d-ininin. ininminin, min«o
^«H«H^»H«-l«-l.-lt-l»-l»-<»-l«-l«-l«-l»-l,-l»-l«-l«-l*-l»-l«H«H«-l»-l»-l«H»-f«-l»-<»-l iH.iM T-| » o
CC CO t-i I I- I-.»H O ~
Oi-< »w DC X'Q-ID -- ~*
H-X CJ \O H- »-« 2 ~- ^t "3
~O ^LJ (/) UJX~ »H —
>-»-i o->o:o I ^ a: ~ CD
o«. o a: •-• i- I ~~DUCD HH
l-« »H 13 » I/) CD O ~- > M O
• >• ~o o:o»- ZCOCD oo t->
»-( n^i-t Ul UJ
^» —* ^ -~ ^,»^»-~OliJ 2U» -~Z O. •
o -O«-l O O <-l CM Ul l/)O»-IUJ rO~ «.«>I-I
in - ro » •
-XQ-- .»» •. . • a: z < d-^H tc
oh-'i- ~ o o \- a: o o o o i-a:o •-• z u. .-i «H a «• it ui
JH'iH Z Xi-« i-«-i-« a:-r3 »-« HH r^ HH,H.UJ
~
A-6
-------
oooooooooooooooooooooooooooooooooooooooo
»H(\irod-tnvor^coo^o^c\itod-invor-co(T>o»-ic\itod-invDh-cooo»Hc\irod-ir)vor>-cocNo
vO\O\OvOvOvOvOvOvOr>-^^-f-h~^h-^-h-^-COCOCOcOCOcOoOCOCOcOCT1OsOxCrvO^OsCTvCT>OsOsO
»H^,H^^«^«H^^^T^^*M^«H^iHt^^^rHrH^«H^^^«H«H^^'r-l«--l»-lT-l«-l«-l«-H«-IC\J
oooooooooooooooooooooooooooooooooooooooo
l-l-l-l-l-l-l-l-l-»-l-»-t-t-»-l-l-»-l-l-HI-l-»-»-l-l-l-l-l-l-l-HI-l-H»-l-HH
UiUJUJUJUJUlUlUJUJUJUJUJLUUJUJUJLdUJUJUJUJUJUJUJULUJUJUJUJUIUiUJLUUJU^
cc
o
I
1/5
LJ
X
UJ
uj
lL
u. z
O
LJ
in
> •-« t-
5 _l • UJ
O
U.
•
1/5 r
< 51
_i
llJ <
2: o
< s
1/5
1/5
o
UJ
>
O
_i
LL
K)
I
~
CC
(T
>
U.
O
*H
CM
o
O O
CM H-
.o o
»-.cs
O-*
ocn
l/5
i/)
.
Q Q-O -t- t-
.MZI --
+ -D • 1/5
oxi-CMH-
«-i -M IO O
#ZQJIZ
»-i i-H < ~ U.
+ ui u. •)«•
ox\oQ
o 5: m ui
UJ
10
X
ct
UI
*-
UI
2:
UJ»-
COCC
O
OD.
I-UI
oo:
vDQ
^nuii/5
»-i/)Z
1/5O
o < t-t
rs t-> H- x
Z II 1/5 *
i— in inu
UJ ~*-tkJ
in'~ 1/50 or
ot/55:»zt-i
in»ii_iooQ
t-ir-i — OD
s:
O
ct
u.0
> • U..O
U.XOX
. • i— cf o. M
O-»-«lxO
»H CD Q m z
#.M z «.o
-• o.< t— 1-«
aro:.o»-
x. uj •-< •. <
D<
z •
< ct
• in
#_IOOD COU_OZU.X — -~ O
ojujtvju
_J_IQ_#
uiosin
z ui o
u.Zl/)Xft(/5
.
HO (TQ.Q:i/5
zou.ui5iujt-"
.-
U M.O Zh-ZinZ>-Hl/)UJt--<
***#•***
oo**## #**
xo t-ni-iD
i-soQQo
_llDi-iX_l
<
O
*
oo*
II>-« J-
etui
I-* Q
•)«••»•
oo* *
coo^* *
^Hr-lU O
fl
X
Q
<
o
o
CVJ
ZZITJII
t-«i-i»-(Z
5I5:-^in
ceor.nu.
uJUjn_j
»-f-X<
uiui
Q Q
**
#*o
**^
OUCJ
A-7
-------
ooooooooooooooooooooooooooooooooooooooo'o
«-tc\jrod-ir)vor^coo^oT-o^Hc\itod-invo. r^-aooo
ooooooooo^^«n.H^«H«H,H,--t«--icMCMC\iCMOJCMOJCMCMCMiorOfOfOiOiOfOfOfOfO.d-
CMCMCMCMCMCMCMCMCMCMCMCMCMCMCMCM. CVJCMCMCMCMCMCMCMCMCMCMCMCMCMCMCMCMCMCMCMCMCMCM C\J
oooooooooooooooooooooooooooooooooooooooo
LdlilUJLUIxLLJLJUJUJlULUUJLULULiJLULJLULdUJUJljJLUUiLaUJL^
2: 2: 2:2:2:2:5:5:5:5:2:2:5:5:5:2:2:2:2:2:2:2:2:2:2:2:2:5:2:2:2:2:2:5:2:2:2:5:5:2:
CJ
UJ
o:
n
X
Q
CC
Q 10
z in
i- - I
in — •
z a: 3-
o x CM
cj — «-i «H ^ o •—
* a: + » — o x
-* X I— H t- to -»
— >- ^ in in in in
-~ in a H ii ii t- H
CD LL *-> 1/5 — — ~ I
a< <, nxx^-tDX
— # ctHM^in x
_i in to I >• -J.i- HO in o —
CDLLCM o x ct:t/)i/)i-i i/i a> •)<•
2_J OvD a < — i^ ^ — l-c\J~
IH o cM.h- a: CM x • — ^ LJ uj»-io<
O'TH O CM OH II II ill «H < H LJ ••'»-« • I
H~-CD HO xx o, ii -cc • H H a: i- a:oo»n
HH O OH QQ ZQ .•-«•>- CCCDCD SX X HCPCL
< Ct -^ H O CD << .M Z - H X .• . CD X Z
LJ I/) I-H O —iCD >->•*-! —' '•>- Z _1 • H — *H LJ Q CD d" X
in CD — r»-«-io O»H •. • _i CD <.< i ^: _i CD • in. LULL
CELL Z If) • • »H ^«. • to H-LUi-«r-< H LUX"
inCMIICD "ZOZLLLL. »ZOZO. •X-H» >-CJCO(JOO>- CJ 1^ Uf^VQCCLULiJLLl "-_J«HI •— CJ • • — X
LU < LL -t-i 5CM CM n O O ^ n CM M CM ^ < M CL CL CL UJ Ct _) >• IO 10 H H LU CC C£ CC CC H M
A-8
-------
oooooooooooooooooooooooooooooooooooooooo
^cMrod-invor~coc^o^cMfOd-in\oNco(^o^.cMrod"inor^aD(>o«HCMfOd-tn\of^coa>o
d-d-d-dd-dd-dd-inir>inininininintntnvO'OvOvOvOvOvOvO\ovOh-r--rvi^r^Fs>r^r^r^h-co
CMCMCMCMCMCMCMCMCMCMCMCMCMCMCMCMCMCMCMCMCMCMCMCMCMCMCMCMCMCMCMCMCMCMCMCMCMCMCMCM
oooooooooooooooooooooooooooooooooooooooo
»-h-(-l-l-l-l-»-t-KI-H-t-»-»-f-l-l-t-l-f-»-l-t-l-l-»-»-»-»-l-l-l-(-»-l-»-l-»-l-
UJ_UJUlUJUlUJlUUJLdUJUJUJUJUJUJLdUJUJUJUlUJUiUJLUUJUJUlUJUJUJUJU^
x^^5rsr^5:x^^^5:2:5:^^^^2:^S2:2:^2:2:225:^2:^^^2:2:2:^2:s
UJ
CD a
<£ UJ
_J UJ
a a
a: 1/5 LJ
— z> *-" a:
~ o •• uj
• — x t£ a
d- — «-i ct
«H • x a o
+ d- i- >-<
(/) «-« d- •• >-
1/5 + CM _i <
»- (/) Max
I I/) UJ UJ _l
— • H U O UJ X
— d- I z. t-> z: -
• I-H •• O £t
d- t/7 a: >
CM ^ or u,
— .. o o ••
NJ-X u_>
— 13 l-» U- 10
or o •• z < t/5
xi->- o •> t-
XZ< i_l-£X«.
+I-HQ HS:Q:
1/5 Q: •-• <.UJl/5
1/5QL- 5TH-I-
i- x o:-.-
I >_(_ OQ^H
• _J 2 U_UJ>-
d- •-( o -z. uj.<
CM < 5: ii-< a. a
— o I-H .UJ d- 10 ••
-^ •• JLLJCM •• >
•>«• o; o; n H K <
— o < u_ujr)i/5Q
^ u_ uj _)ocic:_j
2: >- oa.x»«-
U- O >- n »- 2T I- ,-1 "Z.
< a^.< •• o >- o
x- t/)
r-l«-H5:h-O II <2 «• I-H ^
s^- -~ZCD Q .• _i»-i5:a:
u-XtorHO z a z><.o< ——
<_J*»s:'^ uj o UKDCUJOO
XXttO tO - C£ O _JZU->-tnvO
, 11 II-ID uj t- CM x OUJ«H_I -«inin
-~--vO«<_>«(— CM>+ OUV •••.
o or ce X.CD UJ.LJ 2: >-d-o u or _i ~ o o
f^-XX <£TZO<«- U.UJ O UJ 10 >- _J O *-!•»-«
ioi£:Mi-_;Q:. «5:Q oo •-' oro^>-<< — — —
^»xLLOcti---O 5: ULL H- < Z H- K H
x xx i- xax xx _io xx uj^-o<
-------
oooooooooooooooooooooooooooooooooooooooo
CD CO
CM CM
O 0
f— f—
LJ LJ
2: 2:
•*•«•
CM
»
*-|
n
I-H
*
*-*
^
cv:
*
t-H
n
^
*
—v
-3
•.
H
^•-
x
_J
X
*-*
**f*
*
or
>
LL
*
>
U-
*.
CL
2:
LU
H-
».
Q
LJ
Ul
CL
1/5
— "*• •
0-CL
f- o
in o
- _i
fc_j ^
rt ^-.
Q
UJ
co co
CM CM
0 0
f— h~
LU LJ
5: 5:
X
_J
X
•k
ct
>
u.
*h
>
LL
<
fr.
D.
5:
LU
t-
•>
Q
LU
LJ
CL
1/5
».
h-
1/5
^
^
,_i
>-
<
O
fr.
Z
0
5:
i
».
or
<
Ul
>-
I-H
1 1 1 rTi
- LU C7*
Z
i-i. UJ
co
CM
0
1 —
LU
5:
1/5
1/5
1-
•h
or
1/5
h-
•h
rH
>-
<
D
»
y-
<
Q
*.
^
0
5:
_1
*.
or
<
LU
>-
|_4
x^
o
in
in
1^
LU
CO
CM
O
^—
LU
2:
f—
1/5
v
«-.
O
V0
in
HH
LU
CO CD CO O^
CM CM CM CM
0000
f— H~ 1 — H-
LJ Ul LJ LJ
2: 2: 5: 2:
-**.
Cvi
fr
«-i
n
i-i
».
^*
d-
CM
*.
V)
II
"O
«.
*-*
"^
fe.
»-l
^_*>
X
_J
X
*--
•**
^
or
>
LL
^
>
LL
•^
CL
2:
LU
H-
^
a
LU
0.
U5
^-« «-x
O O
r- co
in in
Of^ *--
O r*
• i^ hH t^
LJ
LJ LJ
CMCMCMCMCMCMCMCMCMlOlOtOrOtOfOtOlOIOtOtOtOlOtO
ooooooooooooooooooooooo
1 — h~l — fr~( — f— 1 — 1 — 1 — t~t — l~t~t-~t — 1 — 1 — \— }— 1 — I — 1 — (—
UJUJUJUJUJU1LULULJLJLJLUUJUJLJLJUJUJLU-LJUJUJLJ
2:2:2:2:2:2:2:2:2:2!5:2:5:2:5:5:5:5:5:2:5:5:2:
•-« in z LJ
2: HH — on
^ ^- ^« ~- •>
• M ~ d- » «. CC
«•;*•-»*•- LJ
CMl/5 ».i-i » * > *
i-i i-i II--*UJO
U » II * Z) 0 II
-LJ LJIIO*-! >
•• OCL QTOLULJ<£O D D
«H — j— or i— i i-i o
••V •• » i-i .LL. UJ _J Z
xo xz --_io <
^i«^ OO~C\J-«-|-l-l HH
\ d- 0 «-l d- »-i «• CM Z) X -I
CV LL. • — Q>-i*-CMt-iCL3 — ID
t- 1 * o IT) *-» ^ •• i— i «. Z — ~D
^ — »-l I-H — •.«•!— 1 •• •
-IILL - II CL- -LU
u. LJ - --uoai-to_j -
o •• r-
O.LJ^HZO < O OILULJUJ Z - LL
*H>-U_-CMX»- O-i-iCM-- 1/5 O II CM*
U-D- -tOin-l-Z -n»CM_i-i|v.
CMZ-Z^I»HLLCM< 1/5- -CMH-il-2: {£. t--LL
— t— II— 1 ^ — — HH t/5 1— It/5— *— 1 — - CM in — o^^QUJori — or-^o- — CO.*^LJ
CM 1 — LL.^.-I—I LL CM *— i O \ — ^Z^-ID^or— LU • - i— •
— t— — — x — — t— i— iuJ>-5:ox*LJi— i H- »H 1/51—1 or
ini/5MU.incMinxX3 - z
^ - — -* f^r 1 , 1 f^ f-*.
*^- J QL LJ O- LJ
H 1- 1- LJ 1- K »- 1- X 1- H J- h- 1- H H or »- 1/5 1- H -
to
o
K
LJ
5:
^~
^^
X
d-
^
t-l
I-H
^^
d-
CM
^
•-
II
f-
1/5
N^
t-
LL
<£
••
\
^^
X
•H
<^
o
•
d-.
LL
•^
31-
CJ
«.
»
II
CL
2:
LU
H-
»•
X
—*
X
,H
•k
1-<
«
d-
LL.
^^
d-
CM
•.
fr-
it
Q
LU
LJ
a
1/5
1-
<
to
o
h-
to
o
K-
LU LU
2:
fr.
X
—*
X
,-)
•.
0
•
in
U.
CM
»
X
vO
\
0^
X
«-|
»
0
in
LL
•^-
CM
T-l
•.
fr.
II
,— |
X
_)
X
fr.
\
*-*
X
«-l
^
o
•
d-
LL
*^
d-
CM
^
fr-
. II
or
>
LL
X
T-(
2:
-^
-^'
X
r-\
^
o
•
in
LL
CM
1-i
fr.
X
vO
X
—
X
«-{
fr.
o
•
tn
LL
f\\
*.\l
«H
••
II
to
o
1-
LJ
2:
^
fr.
Q
UJ
1/5
CO
LJ
u
o
or
CL
z
UJ
LJ
CO
LU
«t
X
1/5
Q
or
o
u
LU
or
1
1
-
in in
CM
o
CO
in o
A-10
-------
ooooooooo
ooooooooo
«HCMiod-in\ois-coos
ooooooooo
ooooooooo
ooooooooo
ooooooooooooooooo
ooooooooooooooooo
O«HCMlOd-in\Ofs-cocTO«-tCMlOd-invO
rH»-ti-lr-(.-(«-l>-l«-lT-lT-ICMCMCMCMCMCMCM
ooooooooooooooooo
ooooooooooooooooo
oooooooooooooo
oooooooooooooo
f*- co CT* CD tH CM *o d" in \o f** co cr* c?
CMCMCMtOlOtOtOtOIOlOlOtOtOd-
oooooooooooooo
oooooooooooooo
o
o
CJCJCJO. CJ, CJ. CJCJCJCJ-CJCJCJCJCJCJCJCJUCJUCJOCJCJOCJCJUCJCJOCJCJCJ
UJ
Q
a:
CJ
o
o:
CD
t—I
CO
I
I
_l
UJ
o
CM
z
LJ
5:
o
u
vO
to
*.
in
LU
LU
*.
in
z
o
Q CJ
O I
Z O
•n CM
.< >-
CC.LU
o: u
iij or
H-Z)
o
Z 1/5
LJ
> -z.
LU O
Z n
O I/)
I- Z
I LU
cc 2:
LU i-i
H- Q
I/)
cc
o
CM
—
LL
>
in
z
LU
Q
•• o
O CM-LU
CM — Z
— z *-»
d- < _l
CM CJ ••
z en cj
O i-i CO
CJ •• >-i
CM in
o
CM
LU —
_J o
H- CM
H CO
»->
<*-» *.
O I-
CMLU
— 21
CO i-i
o
d"
CM
z
en
co-~
h- o
«. «. o
• «-(
IO O —
«H d" CJ
— CM LU
o — or
5: z _i
CO CO _J
». LO -
~ > ~
vO •• O
to -^ o
- O CM
in d- —
— CM ^
I — CO
LU Z <
_J CO_J
LU LO _I
- o <
* * *
~ o
O CM
CM *-
— o
Q. O
» d-
O CM
CM O
— CO
Z H+
Z ••
O —
o o
«. CM
O Z
tH C>
» 10
O i-<
CM ••
to o
Z CM
O —
o ro
•• U
~ LO
ct
CL
o
o ~ —
CO CM O
•• — CM
*~ CM
OX CM
CM vO
U ~
LO «H
•• X
o
4-
CM
•- o
o ~-
CM O
— in
«H (O
— —
a: ~>
>• Q
"3 <
CD
O O
CM CM
Z CJ
o co
O i-i
CM ~
O rH
X O
CM
— CO
Z vO
a to
LO ~
•. o
^ CM
u ••
> CM
u
LU X
i-i O
•. I—I
CD -
z z
CC LO
•• LO
>- H
o z
- CO
(VJ CO
I- >
CL - Z
i-i Z <
- CO CJ
D- co en
< <3 -
I- - d-
>-< I— CM
•• > CJ
CC I-H CO X
CJ -~
LOrH
vD CO
^ t-i J— I
TH CO rH
" < d-
IO _1 ~-
U _J O
CO <.LU
- — a:
< X <
5T *
CO d- — Z
X
o
X
o
»
o
fc
O '
^
»-f
X
CO
z
«. <
(M I
t- I
CL
cc -
a. ••
CM »-( <
O — O
_) co ro CO
< •• U t-i
~- •* co »
CM » «-l
» — rH -~ CO
CJ T-I ~- TH -" •
LU — X CM CJ I-
O ^ CO — LU Z
- CJ < (J CC <
- CO _J LU _l I-
O ^H
^— **"*
«• I/)
O t-i -
o -> o z
51 CM
o —
O Q
z: z
o o
CJ X
O Z
d- O
— 2:
r-< 21
CJ O
CO CJ
*
^ o _ . _
Ud- i-i >- Z o • •• » cj _J
OCM XOCJ* rHCL»O<
i-i>>-r-li-iCOO *.- LU
>o ZDUXCOIO--CLCJ
nCMCMXi-iLO>eOX(-XLUZ
•• "••((•I- - «-LU^:>LUOCOLU
cocMccco>-r-i_;_j<2:2:- _j
Z«-lLUi-icnuLUCOOi-iCO ••<
CLOLULU •• ••<<<<-«2:(-\o_ico _i-~
•-*»<-»•.(-•> H-d-
-^ZOrHI-ll^ I-H<
«H O II I- II 1-0
-. vo i- r-i m u in u CM
~"~~Z •> »—i ~* t—i vD ^- """
o cj < in in i \n i-
LULU_l — Oi-ioini
CCCCCLQvO~'ls-i-i>
-J -J < -H II CM CC
_J_IQLUOXO\OXO
^-~' Q
t- t- DC
Z M <
< _J CJ
K <
Z> D ^
_! O Z
_J <
O Ct _J
CL i-i CD
<
LU
z
- - o
\O QQ
- CJ O
-LU LU
LU X-O
<. CD _J
CM Z O
x -. LL
I- I- CO t~
•> Z K CO Z
«H < i-i Q LU
X _l Z CC 2:
t- CL Z> < O
-» • (J (J
O * Z '-
«-< rH O I- O
rH I l-l Z O
•• »-« CO LU »H
vD ~- LO 2: ••
— i- i-i 2: in
LU < 2: O —
I- 5: LU CJ Q
CC O O LU
3 LL - < CC
* LU
o cc to
*
CJ
*
CJ
A-ll
------- An error occurred while trying to OCR this image.
-------
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
^Hcvjtod-ir)vors-coa'o^c\iiod-tnvo. r^aDCT'Or-tcjiod-invDh-coc^o^-icMtod-invor^coa^o
COfflCOCOOsOOOOOOOOOOO*-l«-t>>>>>>>>>>>>>>>>>>>»>>>>>>>>>>>>>>>>>>
.U.O U O U O O U O O.U O O.U O.U U U U UUOOUUOOUtJUOOUUUUUUOU
CO
•. •
X O
CM UJ
X •
CM »-
«.. M >
•• «. I—I
Ct X •
>- CM • cr
I/) O
• in i- •
>- >- O i-i Z — — — vO
D D Z - UJ - • • , —
i-it-ixs z o o o •
-• «• 2 ^- s: :D LJO
in I/) I- •• O Ct -O O •
D ID in CM O I- M I- K O
Ht-l-l-l t- ^.^-^|_>(_
«...•... _j cr »--•••• Q.HH-I
.»-< i-t X X O O UJ I- I- I- — t-i UJ • I-
> > fO CM CC CD l/)ODD--~ I/) Ct Ul
D 13 in «. »- < — a_ a o_ a h- o in
•-<»-< \ m ~ 2 "Is- »222UJ<-l»«
» » » l-( CM O » O •• I-t hH l-l I— C0»^ ••
t-t i—i Q » i—t (j • 2cO » 2 » » • •
I/) (/I U •> » IO- UJ CM>->->-i-->-r3ino:< - \o
irtinot-icLcccDi- so .. ooomctd'-i-t \o
,»H t-iLJUJ*' ZDCt •• UJSI^^ S n UJ • •• K)
••••QcoxoarDOUJ LU— D I — DCE UJ--
I/) LO -~ — — CO «H _I~UJ~UJ'~UJ'~O^^H_IO D ^n
in in u_ * <. ~ •• ~~ •• ~ _i d--in • ^t(-iTMi-*^(i-4^t |_^>c\J < ~3
«. » D»-iinoOoO« > CO^COH-UJ CODCODCODCO-J. OO »^n > t-t
1-1 »-i m^ctco'-'co'-'O ^CL* • o_^o»'O«'C3»._j. ••Nor~- »
>• XUJ-XUJ--I— «•!— LU CT vOI— vO-LUCL ~>
»-4 HHi-ix--UJ~O-^OCC n UJOU)-* i-in»-i/)H-i/ii-K- ^
t-t n \i-i-
in i/) i-» » ct M t-t »a:t-i o:CLcc:h-Q:< t-lcr:^cc^Q;^ct:_J s<3UOh- _i3_Js_is_isooo_ict:o>_i a
.i-t t-"5:>- tX323ZOt-i< —I CL»t-s <^-<^<^<^ .» o ^-l»-0-~ >,-l>.-t>«H>OHOO>»OJ-~> -»
IO lT)«-» l-«-<_ICM_l«-ICMZ • O • D • \0 Z» • • «UJ»»ZOT-tCMZ CM
IO rOXOCCin»Q.«CL»Oi-i OZOOOO t-^OOO>0l— 02:1— »— t-H»«l^t-i CM
COOCO^-Zt-ii-iO O OCO UJ Ul UJcO Ul UJ Ul UJ _l_l OOlf) If)
^O*-tH >UJ-<»-CL^t-i5:t-(5:i-i5:>H2:t-ii-H5:H-ii-iK5: n H-
t-i t-icr:«-Q- ~cc^(r---i-Hcco^-ct---ct:-'i--ict
SlD3U.fO" HHt^LLt^Ll-tM3LLini-iLLt-^LLt-iSi-(|j_HHLL'--«U,i-iU.i--iU.t-«t-iU.»--iO3U.i--i3
* * D *
toin UJOTH cMtoj-in \oh-cMtoaDin r- (M»H
ion *oo oooo OO«H*-I*-I,-I ^H |^(>.
coco *co
------- An error occurred while trying to OCR this image.
-------
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
^cMio^tf>oi^cocr>o»HCMtOd-Ln\Dr>-coc^o»HC\itOd-LnvOh-cooNo«Hc\jroj-invDi^coaNo
vDvo\ovOvDvo\o\Dvor~f^r^r>-r^f^-r^r>~-r>-r^cococococococooooocoo^criONoc3Nos(T>osoxoo
^^^,H»H^^^^rH^^^TH^^«H^^^^»^»H^^«H«-l,-l*-l«-H«--l«-l«-4i-l*--l.--li-l«-l»HC\J
OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO CKO O O O O O
> > > > > >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>
uu.o.uoououuoouo.oouoooooooouo.oooououuoouuo
Cl
.LU
Q
LU
UJ
.u
X
UJ
LO
*
o
.< —
H- 10
I/) Z
O LO
r-l H-
*>.
LL ~-
O LO — .
2 d- LO
2: — O < LU
O D LO CT> O n
~- LU s: > d- c\j • .(-
> ££ l-l •• «•< l-i-i
UJ LU X 01" O • » H- • O 0
_ia~<— utO'-i-zt < LO a. o
LUi-i(O5:»Q:ZCMOI Q "Z. ~Z_ _J
Q-CO h^Q»O— •O*-'* O O d-.»-i uj
12: inujo^OQooK-LU-- ^m» •-" in >
-^o «-*h-outo«.^H -i
~>LU LU HHOlOOZ-0 LL.O- LOI-Z CM 5T — t- t- CJ X
•^co 1-5:1- in — xoi-HLOtt i-< I-HUJ ZH-I MU
> »-«oiujiia_o»H«' >-O ~ toi/)Oo —
UJ o ar*LUOZOX'-to ZLO^: —loiOLOeoo >-IOCJI--
_l H- 3*H-OOZ:--» LUZO X~ZZ^H<3 _JZ»H + (3 — _I2^I> _J
LU ~ -^* ~ LUO^-LU Q--< -~ h-r-lO-.^, II I«.«-CJII^ILU-«-H X
~* LOCO O»ZOtt «•—>•_! HOK- CO Z • *-• «H »-l •— f— «H«H»H-~«>^-l»H — h-
M LUCJ C\J*^(\fOOOO« OLO C> OOLOIIIICV) ZM.II*C\1O IIIICM Z
-* O»H «xct: •U_oh~»-i-- ^10 od- 5ILUC/)c\irOt-ic\JOC\jrO — i-iLJh-cjrOnf-o
"3. LJCK-- •— IO I— «-tCMLL,L00U«k-vO"' »>HH-ii-i«.mSt-i»-i- •H-mtn>--''~(-~'»- ~-IOOLU»-<»-I>-I trtorocxit-iuLUtninn cr
«t->-iCOliJLOLO lOtn-'OLU
~3t- i-zzs; ZQOQSIXZ i-s:i-h-z >-i_iminoi-LLLnin~on_jir>tnoh-tjL
DZK»-iM"-tri— <-«O»-ir:oQ>cD<
z *LU tr xz a: z a:
tocM."-«\ooto ct o D-OONOO< «-i(—o ioc\j< in i—
vD\£>*THvoi^- * i*- *oc>o-** in # in mtn* in*
<_> uo
A-15
------- An error occurred while trying to OCR this image.
------- An error occurred while trying to OCR this image.
-------
00
o
LU
Z
i—i
r—
^>
O
CO
^>
LO
Q
O
O
o;
o
oo
oo
oooo
o o o o
i-l CM fO d-
o c o o
oooo
0000
tr tr ct a:
u u o u
O O O 0 O O
000000
in vo r- co o o
O O O O O i-l
o o o o o o
000000
0:0:0:0:0:0:
u o o o u u
000000
o o o o o o
1-1 CM io d- in o
o o o o o o
o o o o o o
0:0:0:0:0:0:
- co 0s o
^ ^ i-» CM
oooo
oooo
o: or o: o:
u u u o
o o o o o o
o o o o o o
iH CM ro d~ tn vD
CM CM CM CM CM CM
O O O O O O
o o o o o o
cr o: o: o: or o;
o u u o o co cr o 1-1 CM
CM CM CM ro ro to
o o o o o o
o o o o o o
0:0:0:0:0:0:
U (J U U O O
oooooooo
oooooooc
fOd-invOOcCCTo
rOrorOtorOrorOd-
ooooooco
oooooooo
0:0:0:0:0:0:0:0:
CJO(_)UOOO
O
CM
-
CM
O
U
o
CM
d- *-
CM O
— CM
O > Q
CM LU LL -
Z M " «• O
< _l ~ CM
U - d- —
1/5 O CM LL
* I-H > »
— * LL —
o ^^ << in
CM in •• —
•"•^ ^a*1 *-«i L -
d- o: d- Z
CM CL CM LU
CJ >-i ~ O
o
o
o.
CM
ro
Z
O
in
D
M
I/)
_
»- 1-1
in -
^ o
" CM
or —
•— O
I/) U
o: x
u CM
•^
LU o
z u
>-H X
>-
O Z
Q O
o: 2:
m z:
D o
I/) U
•• O LU "•
— d- r- O
O — •• CM
CM Z — —
— < d- LU
Z U CM _l
o in — i-
1/5 - in *-<
H-t ~ 3 f-
-. o < «.
•*-* ^- ^ ^-»
o o
CM d" d- CM
— CM CM —
IO O — 1/5
<_> in co t— t-i ••
i/j ^ < « ^ ~
o
CM
to
LU
* o in o
— d- t-i CM
o — •• ~-
CM Z »-t CL
— O >- X
iH 1/5 —
O - ~3 CM
CM —
— o Z d-
Z d- O CM
Z i-l ^ X
O U O _l
u in u x
* * *
O
CM
e
o
in
CD
z
cr
->
a
<
x
*
(O CM
ro c\l
r-\ fO CM
0-
O
O
_J
«.
vO
CL
in
^
vO
ro
ro
LJ
CL
— o
o ••
00 —
«H m
o
in
CM
O
CO
ro ~- ^- co ~
^ CM
dLJ
-• a
~ <
o
00
— ~i in
d- «H ••
~ x &
o: < ro
< - LU
^ — CL
O O <
•• CO t-
— ^H e
o -- ••
CO Z —
•H Z) O
— X CO
X < T-4
2: a in
Q - CM
~ o X
O CO U
CO
o
CM
LU
I
o
in
X CM
— CM
O •** X
CO O t-
^n in -
vD
vD
to
C
D
vD O
\O CO
ro ^
C CM
D O
-
in -
» ^
vD O
rO co
~" r-t
QD —
O CM
in Q
CM vH
fO (\J
O CM «H
«H rO CM
CM
rO
ro
LU
C-
.-I X
— Q
in —
CM ••
—r—
X
*
O
in
o
d-
o
ro
•• cc
CM -
CM h-
iH O
ro CM
O C--
ro **
O «-( O
- IO CM
CO - -
•• O O>
r^ ro «H
o CM
CM d-
^H vD
CM -
•• in
o ••
CM d-
CT- CC
CM »-l
CO
CM
vO &• CO
- CM i-l
in - -
•> cc r^-
d- CM «-i
CO O
1-1 d-
0s
ro
a- ro
— X ••
o
CO
"3
in
2:
-3
in
CM rO \D
^ > —
> it: r-l
HH «. X
X — H-
Q O *
- in —
— — co
o rO \O
co > ro
X —
O CM ~>
CM - ••
» O ^
CM CO O
«-l i-l 00
— •" ^H
O CD —
d- < O
CM m >
> — 2: —
- h-
— »-H CM
vO Z —
CM Z) 2:
x x 2:
o o ••
* * *
X
u
*
> O Q
»i in x
_L ^** ^H
0x2;
- > o
O " X
CO —
»-l O Z
— in o
> ~- 'z
HI X 2:
X > O
Q X O
* *
— ». o »•
O CO —
CM *-< O
* — ' CO
CM i-4 «H
^H X ~
^ < O
O S >
d- ~3 s:
CM - "3
> >- -
^ < —
— DO
o co
CM CM «H
- * —
CM o: CM
«-l LJ «-H
— O X
o LU <
d- I- ^
CM Z ~3
O t-i »
* -K-
_ — — ro
00 i-l X
CO CO — <
«-l .H CM 2:
~- — X LU ~3
CM CM «H CL ••
X U •• < —
X > *H h- i-l
"3 ^ .H — o:
^ ^ ~ ^ >.
• • — X
o o o *•» <
CO 00 » 1-1 21
T-f 1-1 • *- Q
—' **• o I—I —'
CM O X X -
O > 5T O -~
X 2: 5: - cr
O
CM
•• CM
CO -
- CM
vO UT
CM
- CM
CM
co
ro
CM
in
in
•• o
ro
tO
in
CM
\o in
to CM
X
ro
ro
\D
ro
- X
• CO
o ••
•• CO
\o in
I-H ro
d- rO
CM iH
in d- ro in
ro CM i-i ro
ro in
•-i ro
o o
iH ••
I o
O T-l
CM iH
I «.
— o z ^H
o co Z LU >H
oo «-i •• d. —
r-4 — t- < Z
•" CM X (- »-<
I Q < O D.
X > 5T — <
5: 2T Q K
7) ^ « LU —
O O < LU —
CO CO 2: _J r-4
»H f-i X < ~-
-- > vO
Q Q < 1-1 CM
X > K D 1-1
^ ^ < O X
~3 ^ O LU 0
* * *
d-
CM
03
ro
o ••
« -d-
I -
•> d-
• ^
o ro
d- »•
I CM
in
*
vD
ro
».
d-
*
vD
ro
«k
ro
*
vD
ro
CM
*
vO
rO
in d-
ro
ro
CM
d- to CM d"
to CM in to
CM d-
i-i ro
d- ro
I-H ro
CM
CM
rO OJ
fO CM
ro
ro
ro CM
•-< ro
i-l O
CM i-l
CM «H O CM
ro CM 1-1 ro
O CM
«-i ro
CM
ro
o
ro
o o
CM -
- CO
»H CM
O> CO
- CM
CO -
O ••
in »-*
I x
X CL
X O
I- O
Q _l
Q Q
CM
iH h-
cc \o
l-< •.
•• in
^s. •„
«H d-
ro
i-H O CT i-l
tO CM •• ro
^ ^ CO *•
O 0^ •• O
ro in r^ ro
- ro
03 -
•• o
h- IO
O CO vD CT1
CM «H - CM
•• •• in ••
CO f> •• CO
CM ^ d- CM
- CM
in -
•• CO
d- CM
vO vO
- CM
in ••
- in
d- CM
•• CM
in ••
iH i-t
r- vO to r^
CM «H •• CM
•• » CM ••
vD in •• vD
CM iH iH CM
d- vO
in d- ^o in
CM iH IO CM
to d-
•• CM
CM -
•• ro
»H CM
ro in
1-1 ro
d- rO in d-
CM i-l tO CM
CM d-
iH ro
* *
rO CM d- ro
CM «-l fO CM
* # * *
ro r>-
- CM
CM -
- vO
•-I CM
X -
03 in
D CM
in ••
d-
< CM
K ^
< to
Q CM
#
A-18
-------
vD OC -
\o co r^ ^ ^ ^ ^
•• •• o to a^ in «H
in r^ cr>
» •• T-I (\j m in \o
CM ^- •• «-t «H rH t-l
\O CO d" •• * * *
«. •. o O \O CM CO
«-! to «-i CJ ro in «-HCM«.«.*- j-j/5
incOOCOd-O^D ZH-I
••».»H»-tl<)in\D OQ
COO«'«-l»Hi-t^ <_>
in eo «H •• » » » in*
*-oh-roCT-in cr #
h-CTv'H'-ilOcl-vD O ^«S*
mr^-.^i^H^H^H a; o i/)
fc^O^^-^^XLJ 'X. • OC
vDCOOOCJCOd-OM I OQLJ
inf^-«H«HtOd-\OCO K)<(-
^^•.rHr^^t^r-IO I— LJ
mr»-o^»»*»»h- Q u) 2:
inh-ainrHr-rocr 2 o
«.«.«.^(iod-vor^ i/) < \o_i
d-\oeo»n»H*H'-iT-(Z t-t
ini^a>»--.«-«-o cra:^
».».«.d-OvOCMCOi-« DO
ro in r^ «-i ro d- \o h- »- o u_ z
mr-CT>«-i»-i«-i«-««-i< x i-» ->
«»^«.«.^^^a: LO M
CM d- >o ro o in «-t r- >- — LJ < uj i/>
mt^O'HCMd-vor^- ~z. d-_i 2:02 ••
^•.•.vH^lrHrH^ljJ CMCDO >•
»-itoin--»»-- u *< »-io i/)
mr^-c^cMcod-ovo z ^M i/jj_ «. ^
•.-•^r-ICMd-vOr^O "^Ct H-*
OCMd-'-l'-l'-l'-l'-l U — < UJD I/) *\O
mr*-o*»»»» >->a:uj ^i/i\o
^^m^tr-focriniu < 02: »>-n
OvHfO'-ICMd-inr^ >0 vOQCC »-Z5 \OX <».
d-r^Ox'H'-l*-«,_l»-l H-ioD O O vO — O l/)l/) ••••>-MD »H
»«.•.««.»»» H-^-l 0»0.00 d-»fO + L. l/)in »HXWl/) II
COOCMOvOCMCOd- <»««o«O«» -.0^ Z Q<«. II — MHZ >-
d-h-C^'-tCMd-inh- _lTHOOIIOl|00*<--~^U)l/5lflCL
d-\oo^om«-tr>rovO z> n»-i — n ~- M M — — MI-IIIO LJ ~3 1-1^^:0 o
•• -• --OCMd-inr^rOOtJO— • — CM — CM — — •-"'-HCKCM^II'^f— T-|^H|— UJO — Ol/)».«.OOO
~
_
f-^lCM>>>>XlX^lX»-l- < — 3rHCM_IZCMZCM_l — — X
liJ >I(-I(-II-IM-N>-i2:
'liJDOOUDODUQQQC)aHH-3»-i-it-i_ll/)QXQOU)intDi-i
a: LJ
------- An error occurred while trying to OCR this image.
-------
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
o o
~-s
I-H
— •
I
-J
a:
** n
+ —
^"* **H
cvj -
V» ~D
»-( — '
— I
ri _J
-> x
0
CvJ
o
1-1
o o o o
o
o
»-l
,-1
O
(-
cvj
» — o
O> MO
3- — —
10
CO _l •
n cr o
HH II UJ
-— • •
o ~ cvj i—
ro c\j •• uj
0 X ~3 ^
,H l-l — - HH
•~ i ~~
o n _i LL
Q ~> X HH
o
ro
o
.-»
o o o
~*
CO
K)
o
CO.
*.
vO
— 2:
LJ — HH
1- - O.
1-1 <
CC 1-
3 —
~ .. o
rH •• O
• » in
O «-l «H
UJ - -
. — vO
>- i
Q < UJ
HH 51 |-
— (X HH
u, o o:
HH LL 3
CO
to
o
CO
o o o
X
in
ro x
CVJ vO
X
ro O
»— ( •
vO
LL
II CVJ
CD i-H
<
h-
l/) II
HH (VJ
N I
•> 0 _J
X • X
o in
• u. -
d- d- X
U. cvj o
» ^ •
^ » \D
II U.
>- II CVJ
< > «-t
O LL -
~> < X
vD
• - X
•• \ 0
CVJ 0 •
I-H • \D
- in LI.
» LL cvj
II d- ^1
O CVJ -
^. - "
H-l » II
II iH
» CL I
- ^ _J
CVJ UJ I
I-H 1-
^ ^
- •• ^
II X 0
CC ^H •
>- • in
-5 in u.
u. 3-
- j- cvj
\ CVJ -
X * -
— - II
i- n a:
< i/) >
S 3 U-
££ < <
o
u. - -
* *
0
0
m
T-f
o o o o
—
t-l
•—
CD
<
1-
l/l
n
II
_J
1-
10
iJ
--»
LO ^1
CK •
O 0
~- O UJ
X I 0 •
0 II >
• z: u Q
\0 O _J HH
u. < —
CVJ Q_ «_) U.
»H O n t-i
* 0
_J ^
* 0
* »-H
* .H
U
O O O O
—
o
ro
1
UJ
o
•
«-l
—
o
cc
UJ
N
££
<
UJ
-z.
o
H
l/)
-z.
o
n
K
<
o:
K
2
UJ
U
z:
d- O o
CVJ U CO 0
•• »-< ro
^ a: •• I
II O .-4 UJ
a: t- n o
i a. a: •
»-i UJ t-t «-t
U II
^H UJ 0 —
co ce o a:
o cvj f-<
d- h- «-i —
UJ •-!
o in o x
Q LU Q 0
re
D. 0
* 0
* cvj
* »H
O
0 O 0 O 0
•
o
vD
ro
+
£T
>
u.
II
cc —
> a:
U. r>
— ~ *-^ »— i
— ct • -
tr — x in o:
x a: >-• • x
t-i X — UJ >-i
— n a: _i —
I/) — > • X
•3 > LL ct _l
< LL < > X
II < II LL II
tn M tr — x
5 > > LL X
X LL LL n X
o o o
o
>- 0
-J vO
D rH
l-(
Q.-O
< »-
CC
o
> CD
cr
< —
> «-t
•
o t-
1- 0
•
>- -^
>- _l
n 1-
_i in
M ^
CD »
< ~
K cr
to x
•^"* K~l
cr s —
X 0 CD
HH _J <
— _l f-
CL < CO
51 t-n
UJ 1- —
I- 0 —
II 2 LL
I- i-t
O
D
*
*
*
O
0 0
«H
+
_l
1—
to
^
II
in —
** cr
\o x
»-l H-l
^-^
O QQ
1- <
1-
O 10
CD n
O
o
vO
^H
O 0 0
^-1
1
_l
»-
to
^
II
.— •
cr
x
n
— *
CD
<
r-
1/1
t— i
^~
»-i
•
H- 0
.CD ro
• vO
•^^ ^"1
^^
cr o
X 1-
i— i
— 0
CD CD
<
1
to cvj
O M •
CVJ I UJ
vO -J Z
i-l H- •
to cr
O M 13
»- — l-l
O U. LL
CD *— i >-*
in o
rH CVJ
vO vO
»-l .H
CTO O
d-
II
~^
cc
X
n
—
CO
<
H
to
1— 1
^^.
^~
•
K
CD
• '-*
~- cr
CC X
X (-<
HH —
— m
CO < r-
< i- to
K to ^:
l/l H-l ||
HH || _J
~- r- f-
LL i/5 to
HH ii ^
o
ro
vO
^H
0 O 0 0
UJ
5:
D
_J
CL
* UJ
* X
* I-
Q
z: >-
15 CD
O
£C " Q
CD UJ
H-
LU U
X UJ
1- LL
LL
O <
h-
UJ
h- CO
UJ
CD r-
X
I- CD
O CO HH
Z CO ^
O
l/l d- X
UJ u
O O HH
Q H- X
31
UJ 0
Z CD tO
13 Z
_l — O
Q. r- HH
• r-
UJ O O -»
X UJ LU .-I
r- • CC *
— HH K
r^ cr o to
n X *:
>- HH LL ~-
H — o a
HH CD C
_l < UJ O
HH H- CD _J
CO U1 Z M
< HH < 3
i- «~ cr o
to u. _l
HH f— HH
LL UJ
HH I/)
* *
* *
* •*
u o
A-21
-------
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
THCMro*in\cr^eo^o^cMro*in\or^co^ot-HCMioj±invDh-cccy>o*-*cMrod-in>ofs-tDO'o
oooooooooooooooooooooooooooooo
ttcccratrccaictQ^aactQicrctcrttcccrcQCoctrtrQiirorccirQr
OUOUOUOOUUOOOUOOUUOOOOOCJUOUOOO
UJ
u
DC
<.
UJ
CC
O
U.
Ul
m
>-«
a:
LJ
a.
Q
Z
<
M
U
.<-»
a:
O
LL
UJ
CC I
UJ H
f-
1x1 o:
21-0
V U.
o
>
o H
.UJ »-i
Q _J
^-1
CJ CD
o <
• H-
O l/)
Lu U.
o
CC
I/) O
-UJ U.
«-
< a:
^ -^
X 0-
x 5:
Q" ~3
210
Z> •>
O H-
*, *
LL rsl
10 O
Q X
^~ u-
••i. r^
O D
Q »
•• X
< s:
10
Q a
^ »
LL tO
O l/)
UJ
•D
Q-
LU
I
t-
>-
GO
Q
UJ
H"
u
UJ
LL
tl
<
Ul
CD
D
oooooooooo
araQrcrirtrQiocQra:
OUOUUOOOUU
•*
*
x
O
<
UJ
OL
o
LL
to
2
O
<
tr
o
U
Q
UJ
U
U-i^
o —
CL
Q- *
.O *
»- -»
.<
Q
\
—
I/)
CMCLWCL
-1/5ZX
UJUJ
w 5:
CLX
-i tO
>-* (-1
to to
UJ H
l/)U_
l->
II II
to to
»-l I-H
too.
l->
• r»
uj in • r^
i-t 1/5 I O »H O
CO _J H* K || M
x-
2:0
.Ul Q
UJ>
— z) CMQI/)
i-
x 2: 2: 2: s:
t- D => D r>
D Q Q Q Q
NO h-
r- N
Is- h-
Ul
OQ
r> « HH LU
o to ui OL -i
X >-i X UJ M
UJ tO II
CO
o
O CM
»-t CO
eo T^
O
OCD
— \o
»orovO
o • • 10
ro
CL
X
II
tn _i _i in t-n t-
•-< 3
CL O
_ _ Q
z) a < ^ +
z o >
z»-m LL
O < ro II
_i M ~ z a:
-i iii
>-• or a: Q t-< Q 3
O •—i •—i t-i i—i 2
n Q Q ii o u •-"
a:»-« M cc H or >-
!-• — — HH l-l Z
Q U. U. Q 00 O
H-l f-l (-1 (-1 O »-l <_)
o o
rH CM
CO 00
uj
2
oo
h-
tn
O^
CM
in
X
ct
UJ
U.
U.
oo
on
1-1 >
II LL
o —
_J It
< CD
U Z
t-l <
LU
I-
<
_J
D
O
_l
<
O
*
*
#
u
A-22
------- An error occurred while trying to OCR this image.
------- An error occurred while trying to OCR this image.
------- An error occurred while trying to OCR this image.
-------
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
rOrOrOfOrOrorOrOrororOrOrOrOrOrotOrOrOfOrororotOrOrorOrOrOrOrOrOrorOrOrOf^
oooooooooooooooooooooooooooooooooooooooo
UOUOUOUUUUUOOOUUUOUOOUOOCJUOUOCJUUCJUOOUIJCJO
o
<
X
o
o
0
UJ
to
-5
»— t
<*^
T-I
> to
^ 0
If fO
— ro
CD
— h-
T-I
> O
X. CD
\D
O
ro
ro
o:
>-i
«^.
i— i
x
u
ii
o
•— t
—
T-I
X
CM
O
ro
ro
f-
T-l
>
II
O
l-t
— '
TH
>
O
o
ro
ro
O
K
O
CD
0.
t— i
*_*
>-)
X
O
M
->
CD
t— i
— ^
T-I
>
X
ro
0
ro
ro
o
CM
O
CO
O
h-
o
CD
££
i— i
tr
H-t
X
u
•
H- UJ
> •
^ *^
II (T
— ) ^'
CD 1-1
.^ . ^^
»~^ _^_
•^- o
*— t '*•'
> U.
0
o
ro
ro
tr
i— i
cc
X
1— 1
X
II
a:
»-H
-^
CM
X
HH
X
a:
|»H
X
u
M
~-fc
C£
n
•«^
t£
X
1— 1
X
o
T-I
o or
CO O
u.
o
i- to
UJ
0 D
CD _l
••» > vO CL
o: o »-i
n OC O «-
~> D CO *-"
CM O X
cr x o u
~ — X — IK •+•
Ct QL t-i OC rO —
i-t 1-1 x M o a: -~
^ ^- D — UJ CD \D n vO Ct
ox • »-i a: roin~ roint-i
x x ujxva: o ~*^~> ..^^^
s;>-:Ecr: _JUDX f-LO«-iT--N«-O
-~ M — llTHCTCr — — QCDUJi-ii-i>-IIO »-i>-i>-||
o^^.£i;^.ot-ii-iQ;QiUJZ<» n *» o »-!•-•
HHtri-tirco ^-ll-*^<^-T-lcMCM~cr:co roro^ — ceo:
— l-l — (-1 l-(CM~'~'2 ^T-lTHOD'-' Oi-4T-ICDt-«X
cM~'CM^-oxct:cMCM»--I
OQXXHUXOXI-l-in>-'COCOl/)>h-T-lcOCOl/)> II
xxxx — •-• x x z :D ^ ii »-• 3 ii no:
S2:5:s:ou.X2:5:oQ.xu.ooct:xotnooQ:xi
~3~)~3~>CD»-iU~DO-iQD»-iOCD>-iDQi-iu<
O 1-1
o T-ICJXCC CM\O ror-
CM T-I-K--X-CO T-40 THO
o o * * o oo oo
CO CO*** COCO COCO
0 U
h-
>
<
X T-I
Ct K
0 X >
• 4^ ^
ro it II
n »H *-H
^_ ^«. ^M.
> > >
< ^ r-l
CO
0
o
CO
o
1-
o
CD
T-I O
1- -
> 0
1— 1 •
1 r-
TH CD
r- •
> CM
^ J—
M >
C\j ^
^— ^—
> u.
< 1— 1
*H
II
T-I
^
to
1— 1
A-26
-------
oooooooo
oooooooo
THCMrOd-invDNcO
vO ^O xO ^3 ^O *O vO v0
rorororororoioro
oooooooo
ctctctctctctctct
oooooooo
ooooooooooooooooo
ooooooooooooooooo
o^oTHCvjrod-inNOr^coooT-icMrod-in
\o r*- r- ^" r^~ r*" ^^"i^-r^f^-cococococo co-
rororororororororororororororororo
ooooooooooooooooo
ctctctctctctctctctctctctctctctctct
ooooooooooooooooo
oooooooo
oooooooo
vOh-cocTOT-tCMrO
CO CO CO CO CT^ O^1 0^ CT*
rorororororororo
oooooooo
ctctctctctctctct
oooooooo
o o o o o o o
o o o o o o o
d~ in \o r*" co c* o
ro ro ro ro ro ro d-
o o o o o o o
ct ct ct ct ct ct ct
o o o o o o o
X
CD
o
o
UJ
I/)
I
CD
ct
o
U.
LJ
d- >
o >-<
ro +
ro —
o
O O
H *4
*
O Ct
O — I
in o *
\O ~- O OO
THTHr\jct>xoin
II II 1-1 HH < • TH II
>-fSJ •. — \ J_ .)(. IT)
l-HI-H>->X_l>-t-l
i-i i—i i—i • O
d-d-~XXXnvO
T-»THCDOO»-<^'-I
O O ^ M II X II ro
cOCOl/)XXOf-rO
II t-t •-< — ro
.OOCtXILL>O
QQ»-
II X
~3 •"
CD
ro ro
in > >
rH X ^
rO M II
ro — —
CD CD
O i-i i-i
1^- X
o o
ro II
ro ~
CD
o i-i
ro ro
o > >
CD X ^
in \o
O TH
ro to
rO rO
ro
O >
CD X
in
ro
ro
ro
l-l l-l X Q
X I- X X t-i >
*-"rOOOX2I>-5:
X> -IIOOQ-)
d- o *: ui ^ n n i-i n
on n _j ct — ~ n —
IO~-^ •H-iCtCt'~Ct
ro">~>X^i-ii-icti-i
CDCDncM^'*-'-' —
o>-«i-iX2:s:cM^-cM
r- — ~-0>>QQO
roro^»-«>-*>>>
o>>u,xx^s:2:
CDX^MOO~3~3-)
(M
2:
>
l-l
^ X
I- O
> •
i-i d- UJ
II T-I _J
~~ o •
Ct CO X
o
ro
ro
o
ro
ro
(/) f- O
UJ UI CD vO
*-• TH ct _i o ro in
X>K OCOCD~--~
OD> r-<~THTHTHN
II HH-hH ID I— O • M II H
— II II • O >- N -
Ct Q O O UJ l-l.r-l >
1-1 ct ct uj ui 2 in ui • •-«
~- i-« t-i D 3 < «C\J«HTH~-
inooD
or-o>Qor-r-r-i-i»-Hi-«cocoi/5
> —>H»2Z=)X — — II
2TOLLX5TSOOO- LL LL O O Ct
~3CDi—iO"~3~3OO2IOi-ii—iQQi-i
o z
O d- CO O
-------
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooc
^CUtOd-invOl^cCC^OrHO
OOOOOOOOO^^«H«H,H^«H^^rHCg(\J(\JOJWC\JC\10J(\l<\llOrOrOrOrOK>rOrOfOrOd-
d-d-d-d-d-d-d-d-d-d-d-d-d-d-d-d-d-d-d-d-d-d-d-d-d-d-d-d-d-d-d-d-d-d-d-d-d-d-d-d-
oooooooooooooooooooooooooooooooooooooooo
accz&c£ac&aiazccc£C£C£ocoza£KKC£C£0£a:a£C£C£Ka:a:c£teccc£ecccotc£l-
CO >
O n
to +
ro "•
O
o o
H- .-I
*
-O a:
0
CD
\O
\n
O
CO
O
o
ma:
(-1 i-, ^»
— — cc
K)lf)»-»
>
<
x
O
.
o
K)lf)
^ ».
^t^i
ii ri
m o •• c> — —
«- O CM • "3 ~3
X O •• t— O O
> O O _J t-ll-H
•-• »H <.
M ^
~
xx
>-
x><
I-II-HII
ll llr-4
a: i- K
<-i
inincoQo>-i~«HtDx oo CDCD cotoi-tot-i>-(Oi-ino>-i>--l-~~O>>Q
11 n »H ~ x ~i— xx xx xx~-»-<»-<>
>>>
O.tO-i-t
in
O
CO
O
in
o
CO
tOCVI
in in
o o
COCO
fOfO
IO IO
K)
K)
(O
K)
ro
K)
A-28
------- An error occurred while trying to OCR this image.
-------
OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOGOOOOOOO
OOOOOOOOOOOOOOOOOOOOGOOOOOOOOOOOOOOOOOOO
•HCMfO-^LOvO^CCC^OrH C\} (O^tOvOh-COO^O ^H CM fO & tO \O r** CO 0s O *—( CM K)^t>tOvOtwC00p O
CO CO CO CO CO CO CO CO CO C7* 0^ 0"* O1* CT* O^ O^ O^ O1^ O1^ ^3 CD ^3 CD CD ^3 ^D ^D CD CD TH «H »H ^H *H »H »H T*4 ^H vH C\J
d-d-d-d-d-d-d-d-d-d-d-d-d-d-d-d-d-d-d-ininininininininininintninininininintninin
oooooooooooooooooooooooooooooooooooooooo
^^^^^^CL^^^^^^^^^^CC^^fL^tLCL^^^^CL^OLCCCLaitt.CCCLCLU.CL
ooooooooooooooooooooooooooo.ooo.oooooooooo
d-
•
o
rH
LJ
a
x
CM
O
ID
LJ
a
x
CM
d-
•
o
TH
UJ
0_
o o d-
^H ^H •
«. » o
T-I >H <-l
II II UJ
:* ^ Q.
d-
•
o
TH
LJ
D.
o
V)
o
«H
^
X
TH
^
CM
«. «. o o
~ — TH TH
< .< X 1/3 X
» - .. O
UJ UJ CM I-
CM
TH
^
Ul
cc
3
d-
•
o
TH
UJ
D.
TH
^
X
CM
TH
^
d-
0
UJ
0.
TH
^^
Q\
TH ^
•f X
UJ CO
HH ^-^
_J »-
II <
LJ 21
—r f^f
^~ CC
»-i O
_l U.
^.
d"
TH
UJ
"Z.
|— |
1-
0
o
^
in
CM
^
TH
I-H
X
CM
1-1
^
X
CM
to
t— i
•i
X
CM
^
»
,_!
«
^
X
CM
1-
21
f^r
CC
O
lo-
in
T-l
•^ to
CM \O Z
X — O
\- LJ 0
I/) ». |- «-
Z »H l-l ••
.O X CC Q
t-t H I/) 3 • 0
t- — z ~ to tn
• 33 vOO'^O'* \ i— i
coco >oind-inin -~ ••
CO- HH co to O d". • d- UJ— CC
^- CC CO - ^O TH O O^ 21 UJ t)
t_r<- O»«.LJ« t-i2: LO
OZ H"vOvO*\O ^ *™-* O *-H
i-oo ^^u^-o UJ* in--
Of- O LJ UJ Z LJ II 2T UJ to 0-
O CDI-»-»-H»-LJ ~-2T »-
O CD LJ O —. H-I.I-I _Ji-iZ IO "* •" ' •** Oi-i
• — O CD O CC CC. • CC •-• ZIO «HCM >— »
OOOOCCTH-. «33Q3_J l/)OZ «••• >-
CMTHII T^ + lOO O — — Z — — Z O- O — >£ M OQ
-•• — OO— -CM LJOO<00 -IIOUJCCCC O M
THTH3LJI/3CMLJ-' • in in. • in in TH — + 2: o o — —
II II 3 • ~ZTH — •. •O--THMUJO— 33 oin
IM3t-H-~CCCC •»! LJ -« UJH-»-t<< CC.CD \D to
zr-h-t-io i — cc d- n ozz. •zz_Jin2:.ii n i/)i/>x. • —
i-ivOvo-'totocMO-d-oLnt-it-ii/ji-Hi-inin^-o^ n II»-IIOLJO
1- CM CM »H t-4 ~- t-( »O 10 i-i _| _J Z _l _J UJ IO IO U) CC CC D II Z 1- »-
OOOOU,ZCLU.OOU.U.U.lJ.U.U-»-— >-Hi-»i-(i-(i-i3CD
cc
d- vO f>- CO Q. in
d" \o o co & in
CM CM CM CO * tO
#
o
IO
0
o
Q
0
U)
l-l
•h
cc
o
LO
l-l
^
0_
1— 1
^
>.
o
f— 1
*^
CO
p»-
to
^
vO
•^
UJ
1-
cc
3
\o
in
fO
^
1-4
fe.
X
TH
&
CM
^
X
CM
^
to
X
CM
^
»
fO
•
•.
X
CM
^**
H-
<
2:
o
Ll-
CO
f*«-
to
^
II
<
*.
<
^
If 1
2:
to
z
o
o
*^
<-«
1^
d-
vH.
»
\D
*~
UJ
>-
cc
3
TH
+
UJ
z
1— t
1
It
UJ
-I
UJ
ID
2
I—*
1-
^—
o
o
r>-
f**
to
\
»
•b
TH
1— 1
X
TH
^
CM
1-4
^
X
CM
to
l-l
X
CM
•k
»
to
~
».
X
CM
^*
t-
^
2:
ty
LL,
o
u.
in
^>
IO
o
CM
^
v-\
II
N
rvi
LJ 1-4
^
z r>
1-4 vO
»- to
0 0
0 Q
d- vO
d" *O
to to
A-30
-------
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
^cM»od-invo^cocFo^cMtod-tn\of^coo^o«-icMrOd-in\or^co(>o.-4CMfr>d-in\of^cocso
cvicvJwcvifMWcvifMtMiOfOroKjroioroiororo^d-d-d-d-j-d-d-d-d-inintninininirjinininvD
inininioinininininininintntnininioininintninininininininintntnininirmininininin
oooooooooooooooooooooooooooooooooo'oooooo
cc.ctcta:t£ccccocccoLC£0£&ott£ctccc£&cto£C£ctaict&ct&etc£a:a:(rfra:a:uit£c£a.
ooooouuoououooouuou. uo.ou. uo. ouoooouoouoouuu
O d-
LO •
f-
^
o
«H
^
II
^
0
Jt
V
<
•t
UJ
5:
i-^
2
2
O
U
O
TH
*
II ~
i^ d-
~D 0
^ *~i
^^ ii i
^ D.
< *-<
» O
UJ «H U
s: » i/)
"~ X f-
2^0
2 • H
O CM -•
O 1-4 ~
UJ
Q.
^
X
CM
»^
d-
o
»-<
UJ
QL
«-l
O
«-l
^
X
«,
CM
n
10 d-
2 d-
O O^
»-i •• —• «^ •. 0\ »
I- CM\O ^«.*^»
r> x— QOX-N
m i-ui uu-T-i- i/)
t-i • h- l/)l/)«-»-i«- 2
a: ^)i-i n >_ ^) n r-< o
I- XOi h- » •. 1-1 ^ ..!-• n
2 1-1/55 > (£££••<•' h-
O — 2 -~ 1-1 OCJX~DX O
OO5 vOO-^O^ X (/) I/) -"
coujvo vo o d-. • d- w— «-«»cMi£: CM tnct
vO- U NO'-lOO^ SUJ Q-CLi-i< H-t I—
rt.o o • «.ui » -i-< s H->-«.~3 •• 02
,ODH- »-\O\D»vO » »-» »H. I-H . >H x •• X HO
K-O ~<^UJ~-0 UJ» CM»«-C\LIU CM U
U) O O Ul bJ 2 UJ II 2-UJ r- >- >- •> 2: - O
.O CJ CD »- I- +* t- UJ —SI O D K) ^- rO O OU
O O OC — -> 1-1 »-H _J 1-1 2 2— —^ Oi-t t-inZ n CM O ~<_>
. • *»O H OQiCt, •CCt-i 22 ^ICM I— • ••••2 •• •• • OCC
CT>0 OO^I> »S3QS_J l/)O2 »•• I- KXO X »-HCro »D
rH || .«I + nO 0-~~z~^ 2 O O -~ X ^ O> > CM O CM M ^ II OO
«.^. 0 -~.CM UJOO-ir>-^-( «-< «-IMS ~ QC
s - «-r< D_ — UJos:iiJUj»iiJUJUJS:>--t--Ol/5CDCD\l— «-CM •• «H «-2UJ^-iliJ »-irsJOd-X»-i
M, II ~Z.CC. If — O2UJIiJCDUJUJ»-i UJ»->-i<rOQ_d-Ol/)>-<>-*l/)H-it-i||tn~-O^ II II i-il/)UJOUJ-» »• •-! d- U) *-* _i _J 2 J _i uj d- 2 in a: cc Q n 2 t- t- t~ s: »- LJ t- 21 H d- d- 2 — M —
2a:~-t-ct>- (_~^«-~'~.^2: 2f-ouoo.^-»-i >-*a:^22cc:2 2tr:~»-a:
O O'CL Q.2Q-U-OOIXll.U-ll-UiJ-MCOO3:ini/)h-LLQ:Oa:O an OOOOOOD.L.2CL
O.O>»-i i-tii-i.i-i t-i Q t---»-i i-ii-it-i»-ii-i_JQul— i-iHH»-iH-(»-iSO3:|jLS_IUlJu-i»-ti-(i-i
cc a:
-l^-eoa in T-tO^CMlDd-«Oh-COCL
A-31
------- An error occurred while trying to OCR this image.
------- An error occurred while trying to OCR this image.
-------
OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO'O
oooooooooooooooooooooooooooooooooooooooo
^CMrOd-invOr^coOxOrHCjrOd-iT)vOf^co^o^c\lrOd-tnvOf^coONOtHC\ltO.d-invOr>-ccOso
d-d-d-d-d-d-c}-d-d-inininmintr)inir)ininvo^\DvOxOvOvO'Z)vD\Dr^h-r^h-r-r^r^f^r^f^cc
vOvO\OvOvOvOvOOvO\O^O\O*OvOvDvOvOvOvO\O*O\OvOl\OvDvO*OvOvOvOvOvOvOvOOvO>O\0\O\0
OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO CIO O O O O O
o:o:a:crctQ:Qio:Q:D:ctQ:a(ro:Q:Q:QictQ:Q:a:(ra:a:Q:a:D:ctQ:D:Q:Q:a:ctQ:a:a:cr:Q:
OOOOOOUUUUOUCJOOUUUUUOUUOUOCJOOUUOOaCJOUCJOO
a: UJ
h- a:
Z 2 (-1
UJ O Q
U 1-1
•2. \-
o o z
u UJ o
cr »-i
HH h-
D <
a:
o H
l-l Z Z
HO Ul
< M O
ah- z
< o
ce u
h-
z
UJ
u
z a:
>- a r>
< u o
a x
oc 2: cj
-Z) ^ X
O CO — »-
- • I X UJ
- «• h- - O «-4
>- x > z x
\n CM — «-i < »-
^-i to _ix^x_ix»- — —
o 3- x^h-xin oj-
o 3- cr ~ ~ — o: - n inin
oco CCLUO~Q:»Q o»o
o o t/)Duind-^o \o*
OO H Z ^ZoinDCJ «••>
O3 h- I OCMO-< ~- vO X h-• h- h-
—•• Ct *~ --- tt Q UJ~- O<_)"~ «-H»-H
o ^_i_i Q; Q; |_o I h-iuinuis: cetr
o ottt:— >-• »-H h-Zh-x >-«h-xh-cr^: S3
~ 1-1 n .-i — -~ xuj d-z a: M j- < HH ~-
h- — — X C\J \O -»-<^
X XX< X>-CM SZO OJ »-H — — 3 (M O QOOh-
X< << ^ >- ~ u \r> • n ~ tt x>0voxxa:*:iouj0xx ntnm>-ih-h- oz
~UJX uj~»ict~touj — tr-> <_iujd-r^ n — in • ui \o h->-<\a>-_J_i MM
a_jx _jct-»>-ia:d-_iQ:«--xo»« — -~
HH.O .MO: — i-ij-»M-^HH^z)X-- cjd-t-na:— >a: Qh-xrvirsi >>-
'-XMCKX'-'-HCvJ-' X — OJ~ ZZMQUJ~h-5: Z U.nh-2:-MXXII nn
(\JOXIIUXXXXh-UXXOJh-h-HHh-i-ih-5: »Oh-( •-• SI -- LJO>-OrMh-h-CC:^-«-
i-H~-tl/) ~-OCX--U>-"OO>-iQ~5UOOQ:h->-H3U.t\J-iU.OJcO-Oh-t-th-i-iH-«>-iQ^:
13 *#* ****
o »-i in in ;f o o ojd- to
o oiotoio*in inm in
O O^\O\D*vD \O»O vD
CO 03d-^-c*-*d- d-d- d-
A-34
------- An error occurred while trying to OCR this image.
------- An error occurred while trying to OCR this image.
-------
LJ
h-
<
a:
<
~ LJ
-> >
x
in LJ
» i
cj ro
«H HH
LJ ••
a •
i-f ii
» >-
- <
o •
n tr:
f- >-l
< Q
tr •
>- ••
z <\j
LJ >-
a:
<
LU
LJ
I
— «-o
cr: - - in
L.--Z
t-i ~* *. O
ll--»-i
"3 «. » h-
» to in <
— HH »-i a:
O — «- LJ
*-< «_>
X » - Z
2: d- it u
in
o
O
.o
z
02:
I. •
I in
d- u.
O *
5IM
>--<
II
X
<
LJ
U
LJ LJ O
xinmcj
,<••••
SIvOvDI-
Q i— >- 1/)
<- ^«.LJ
•• CVICMX
o rofox
o i-»- a:
j- ••••OO
r- xxzi-
m.in \
**
n. M
II. II. I
in
0
**
^
— »
x
— Q
V) •
~
2 "5
-> HH
1— 1 «^
~- CM
•~ CM *-i
- D I/)
^ Z I-H
CM _l 0
X - LJ
t~ "^ l/l ••
- _l " »
^H -~ >- d-
Z •
o
ur-
in LJ
•- o
- z
r- <:
mi/)
LJLJ
^ Q -3
CM M. it
o«-i
inc\j
LJH —
>-za^
-oo
x»
in
d-
O
QO
— t-
••
•-< o
~- O
o -~
o o
d- »
r^ C
••LJLJ
o
x
LJ
_I
CD
<
>-•
a:
<
ct
l/)O
LU»-
XC-
U u o o "D D < M u u u zo Z"~->-ir--K-a:o**~zi-
LJO~OOt-i»/)5:-~-~'~--»Z)»-lt-«LJLJ<'-
(TII^(OOOMllQ»-tCMlOcJ-XXh-K>-2:C)»-2:i-eOIIMI-K»-i
II
tr
LU
~5
01-1
nin
— O
Ct <
T-IOOCLLJ
003-**
o
O
o
co
O
o
co
CM
o
eo
O
o
co
rOCLLJ
O**
o#*
co*#
o u
A-37
------- An error occurred while trying to OCR this image.
-------
0 0
o o
^ CM
d- d-
CO CO
O O
cr cr
V *— I
D X
fH Q
— n
CM —
0 Z
> ~~)
2: I-H
^ •—
II Z
"•»* ~1
^ — /
in x
^-^ rff
•*-• ^
5: 2:
~3 Q
to
vD
O
CO
o
o
K)
d-
CO
o
cr
o
.~
*:
-3
H-t
~
CM
Q
>
^~
v^
If
—.
Z
~3
i— i
~«
D
j---*
X
51
~3
ooooooooooooooooooooo
ooooooooooooooooooooo
d"invor*-cccFxo«-HCMtod"tn\of"-coc^o»HCMtod"
d-d-d-d-d-d-inmintnininininininvOvCvOvOvo
CCCOCCCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCO
ooooooooooooooooooooo
crcrcrcrcrcrcrcrcrcrcrcrcrcrcrcrcrcrcrcrcr
CJCJCJCJCJCJCJCJCJCJCJCJCJCJCJCJCJCJCJCJCJ
u
LJ -
CM cr
»- 1- o
< - 1-
•• x CL
d- Z »-i LJ
•OX U
CM I-H - LJ
in i- 2: cr
LJ < ^
CL - cr — x
.-1 fO H • O CJ
- HH Z - 1- <
•• •• LJ «H I-H LJ
. II - O • ••
o ii z in cr i-
Z >- O LL LL <
o < cj •• »-H
u Q — ii cr
cr x -3 <
cr D r- LJ
02: o ^ — >
o *: i - z>
X 1 - — LJ
i - -2: I-H x
^ ~ _ V^ "T-* •
•• fc ^ X. .*£. 1 —
•>• «H CM Z)
CM • I-H • I-H cr
>— < m •• •• •• o
«. LL - «-i -~ LL
- CL • ~3
o f- in — 1/5
2: - U5LL 0 Z
13 - LJ ^ ^-t O
~ 2! II X d- X I-H
in I-H LJ o •• < H-
- X 0 I-H - 51 <
i-i< z x ~3 cr
M 2: < 1-
_l 1- Q — Z
•• Q 1/5 Z LJ ~3 LU
— Z I-H -. O O — O
-JOOXUZ ZZ
CM — 0 CM LJ < Z>O
— XXLJ i-He>i/)cr x u
^ h-Ql/)--.Z-- <
~) «.— •.•.Q:^*. x cr
i—-d'ffcr-o>»-- - x ••»-<
U l-H<»t-HH^X \O >-• 1 «-ftl
> LU n cr — x T-H to -^ to n
s; od->-ZLJini-HX •• o cri/5
\£ in\O«-OD_\DX*- «-l d- f— LJLJ
n oo-i-i o-cr M o 1/5 "3~3
-~ vOCOXH-LJCOXO i-Hd- CO LJ OHHI-H
ZLJ^-THUSI-i-ll- 1 -LJX •
~) o o v0 — LJi-'vo^'CL \omm\DZ)CD otnm
HHZ i-cri- — h-LJ'-vo**~-zi-H u^t^-i
*•— * HH LJ LiJ ^ ^~^ LL) ^ CJ ** CD HH HH Ii 1 HH T" (O *^* C5 CD
i-«>-H-i-2:o »-s:LJcrcoiiiii-i- iirHcoco
•^T ~^ t ^ »---» fy' LJ r\f ry' \ i (\r f*i . . ~T t "^ ^^
^_ ^1 '^ *^ LL ^^ Lt- LL *^ Ut- X-J ^^ ^- r^ ^- ^^
z)Ocrcro cro QOL-i-croz~3xoo
I-HO3SLL- •• SLLX- QI-HI-HSUI-HI-HQQQ
"W1 * "T * Ql.
CM d- m vO CL in
\0 vD \O vO ^" d"
o o o 0*0
CO CO CO CO * CO
u
o
o
in
v£J
CO
O
cr
L>
<-H
+
Z
-)
n
—?
^L
~3
I-H
O
0
vO
vO
CO
O
cr
u
—*
to
LJ
~}
I-H
^
cr
LJ
~5
i— i
* —
m
I/I
\Jt
M
=^
I-H
0
O
^
\0
CO
o
cr
o
o
T-t
O
CO
o
t-
o
o
— N
«-H
—•
X
Q
•
LJ
1
•
~*
v
~5
i— t
-b-
21
>
I-H
O
LL
i— i
0 0
o o
CO 0^
V0 vO
CO CO
0 0
cr cr
U (_)
*-.
^
~3
>-i
. —
21
> cr
i— i LU
X "3
U >->
II II
*H CM
x 2:
0 ~3
o
o
o
f**
CO
o
cr
u
^^
1/5
LJ
~3
i— i
* —
CD
Z
or
n
to
X
Q
O O O
000
»-l CM fO
r— ^" t*-
co CO CO
O O O
cr cr cr
u u o
*~.
^
~3
hH
~-
2T
>
— ^~ i— i
^ ^ X
I-H I-H II
Q 0 Z
> > ~3
~3 ~3 —
M II Z
d- in x
~3 ~> Q
O
«-4
O
CO
o
o
d~
f**
CO
o
cr
£_}
*-*
^
~3
1— 1
*-*
D
>
~}
M
^
-3
i— i
-^
D
X
^
~3
O O O
o o o
in \o r^
h- h- h-
co co co
o o o
cr cr cr
CJ (J tj
<\i
X
~ h-
~3 JH
I-H X
— 1-
o
2: o
~3 in
II 0
— \O
Z LJ -
I-H Z ^
^^ > 1 1 , 1
^^ f^ 11 I
1-1 f- 1-
~7 ~7 • •
21 Z I-H
D o cr
I-H C_) 3
in
o
CO
o o o
000
CC & O
P*~ r*- cc
CO CO co
O O O
cr cr cr
CJ O O
^
•• IO
d" I"**
< «.
(M -
«H II
LJ >
CL <
^ D
^
••
II 2:
CJ ^
Z
O -
^H
cr •
D in
0 LL
X CL
1 0
to -
II
LU
CJ
z
1 — ' <^
in 2: t-
•-! 2! HH
II I-H Q
—I X
- <
— 2: -
_1
•*~~ V_ ««.
*^^ ij
X _1 I-H
Q cr -
^- < ~
— LJ II
CM >- Z
CM - O
0 •• I-H
CO X |-
•• i-H O
vD ~- LU
— K cr
U^rT * *
^. *— *
t- 2; D
I-H cr
cr o
s LL -
*
CM
CM
O
CO
A-39
------- An error occurred while trying to OCR this image.
------- An error occurred while trying to OCR this image.
------- An error occurred while trying to OCR this image.
------- An error occurred while trying to OCR this image.
-------
ooooooooooooooooooooooooooooooooooooo
ooooooooooooooooooooooooooooooooooooo
ooooooooooooooooooooooooooooooooooooo
crcrcrcrcrcrcrcrcrcrcrcrcrcrcrcrcrcccrcrcrcrcrcrcrcrctcrcrcrcrcrcrcrcrcrcr
z
o
ro
UJ
U
z
tn
^ i— i
x a
o
d-
\
X
- Z
tn o
z •-*
O H
<
cr
LJ
u
z
o
o o
o o
o o
o o ro
• o cr
o »H ^
* * I
*-( CM »H
cr cr cr
n n u
CM ro d-
cr cr cr
\o in
cr cr
I —
d- CD
^ I/)
II II
cr cr
cr -
x CM co
-. • z
in d- o
cr 1-1 »-i
«. D. <
d- o cr
cr •• »-
^ d- z
- «H LJ
-^ •-< CJ
•-« - z
— d- O CM
to • u x
> o »-
x CM cr -
» LJ Z) «H
CM 0. O X
cr ^ x i-
— ^ - l
CM -~ IO «-l ~
— •• IO 1-1 O
d- in CM - o in
cr cr ro x in o
* *: ro o vo
~- ~ •• d- X ••
CD CD vO -~ CD \O
< cr
cr n
z
LJ
O
Z
o z
u o
I—I
•> \-
_J <
cr cr
^> t-
o z
X LJ
2: z
•D O
s: u
1-4
x cr
< Z)
— 2: o —
»H • X X
-I * I ~
to x «-H cr
tO^H O
*• vO O
r-
cr
X <
o
o
o
CD
in
CT
CD
••
d-
cr
o o
in •
- LJ
iH _l
II • —
O
o
o
o
l-l O
— *
O ~ r*
cr uo LJ < X LJ LJ < to >
n n 1-2: »->-5:> rox
cr cr cr o z cr cr o o o LL
CD^3LL«-"33LL- • Qn
cr * *
^H to a. CM »n
CM CM * in T-I
ro to * ro rO
IO K> * tO IO
> cr
M ii
•H CM
cr cr
o
o
o
o ro
o cr
«-i x
* I
CM «-H
cr cr
M II
to d-
cr cr
o o
• o
O rH
* *
d- in
cr cr
:*: *:
n M
in vo
cr tr
vo in
tr cr
:*: v
i —
d- m
cr <
^ i/)
ii it
r^ d-
cr cr
CM
• o
d- CO
iH r-4
LL
O. LL
O O
d^ Q
» »H cr
— I-H O
»-HI ^ U
— d- LJ
»-i • tr
> o z
x CM t-i cr
- LJ Z>
CM OL •• H-
CT ^H LJ LJ
-~ ^ •• CL cr to
CM -~ tO < ~ LJ
— » CM •-! t— »H D-
d- in »H «• • <
crcrtoxKLJr-co
V * ro O D Z O
•^-' «.d-CL •-~LJ
CDcDvO~r-0_co_I
Z< — h-3<~i-«Z
trinLJ
-------
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
<™4 CM to d" in ^0 ^- co CT* CD «H CM to d~ in \o h- co cr* CD «~f CM to d~ in ^o r*- co o^ CD «H CM to d" in NO ^* co o** CD
CD
I—I
OO
C£
CO
oo
LJ
o
2
o
00
CD ^^
o o
0 0
in in
D
tr
^
Q
Z
^_
^_
in
i
,
Q
Z
^
1
<;
}—
Z
O
N
I-H
cr
MX
I/I
\Ji
- in.
X- LU
in i-
— ^
j— _j
mo
•;_i
X-'. ^
XiO
•k
X <
— s:
.CD
^; I-H
s: in
CD
»-H LU
0
o
o
in
in
H- 1
V
in
^
•
- >
LU
Q
•
Q
L_
in
i
^~
i-
z
o
N
HH
£t
O.
X.
LU
X
0
1— 1
(—
^
cr
1-
z
LU
O
z.
o
u
LL
O
in
z
0
o
0
in
LU
o
tr
o
in
LU
X
H
;jr
0
Ct
LL
Z.
3
n^
Q.
LU
Z
I-H
1
«t
Z
LU
1 I 1
LU
LU
CO
LU
_J
CD
Z
<:
LU
X
1-
^r .
O
(T
LL
0
o
0
10-
in
I-H
in
^
LU
s:
^
i
Q_
LU
X.
1-
LL
O
y-
CC
O
z
o
CD
LU
X
f-
'
Z
0
I-H
1-
u
LU
i— i
Q
D
Z
I-H
-£••
Z
3
o
ooooooooooooooooooo
ooooooooooooooooooo
ininininininininininininininininininin
LU
in >- u x
LU 1- HH 1-
m n -^ ct
D _l 5: LU Z
•-• ^: x HH
>• CD ~ CL
_j in.< in z
_|. HH J- • O O
< in x 2: >-t
D - O I- 1-
i- IM z ~ ce. < <
u in o 2: CL tr
< 0. ^ 0. - 1-
«-r> ".< _i z
Z. • _J LU
O>l-«»LULUi-< U
i-iuzcr«uo-D z
in Q.LU o. • a: tr o o
in Q H LU oo in u
D»zo.ce;oo< CM «H«HCM
ODLULU. V >- V >- >- V > HH
»-H tr I- LL Z Z LL LU K t- I- f- h- >- J- K- I- >
0 >O I-.OQ< _I_l_J_J_J_J_l_l_IQ
-l-l-l-l-»-l-<
OHZ> Luin ininininininmininD
• in in cc o z
_J • 1- >H OZ- _J_I_J_J-J_I_I_J-J<
_l LU LU Q Z H < 10 _l _l _I _J _I _J _l _J _J 1-
H_)in «-o-=>D D z> z> DO
0Z-H_I C300C3C2C3C30OLU
in< >-iLuujQuzinininininininininx
^ ^" 5 i/) cy o ^r *£. ^T' ^ ^c ^t ^. ^t ^t ^*
a in o z D _i x- >-• o_ o. CL o_ a o. CL CL o_
^Ct2LU- ~r LI oinoD*™*Lxmininininincnininx-
COH- QLUZl— _JLUt-HHHI-HHHI-H. I-HI-HI-HI-H
Q 1
ll 1
LLJ
D
D
Ct
<
Q
z
^
— »-
s: in
*-»
LU
Z X
I-H
L_ •.
cj fvj
LU
ct <
CM CM
0 0
0 0
in in
o
t—
in
in
<
_i
u
X
I-H
_1.
I-H
CO
<
t—
in.
in
t— i
x
ce
o
LL
O
hH
(-
<
D
<3
— LU
2:
— LL
O
Z
o ^~
I-H LU
i- in
LU tr
CC LU
I-H 0-
CM
O
0
in
z
LU
>
HH
CD
LU
X
1-
tr
o
LL
N.
in
LU
i-
<
_]
D
O
_l
<
U
a
z
<
»
z
LU
>
I-H
CD
LU
X
1-
tr
LL
X
K
LU
to
0
o
in
in
i— i
x
in
D
Z
<
O
o
o
in
o
i-
D
LU
1-
21
I-H-
_l
i n
uJ
M
in
LU
cr
ii i
X
s
o
CO
o
o
CD
in
o
o
in
•
z
tr
^j
H-
LU
fy
LU
•
X
^-
Q
^y
<
X-
X
^~
o
LL
Q
LU
to
0
O
in
H
in
V
^
«^
CO
vO
•.
CM
in
CM
•.
o
^
tototoioiotoioa-
oooooooo
oo~oooooo
ininininininmin
CO
in
CM
•
if*\
in
x
^^
X-
X
•**
CD
0
_l CM CM
< d- IO
* in o
d- o o
to • •
o in *H «H
«-l • - * - *
~- CM CM * fO *
^H »H
< 1 •• X - X
*H * IO *
inZODQLULUh-Z ....._. . .
MI-HLU i— s: <<«-».«-< + •— D d-d-i-Hi>-iicE: ooc_»i-ii-Hd-»coo»oo
>- cr > H- .CD o _i < i- »- _i _J — co CM o in co or-co
Q-D D oi-Hinh-ocjxxx*: in in ouxuoi- x ox o
tr in t- CD i-wit—.ill iii-
co
D OXLLNOLLMO
in (D t- t-n/nD t-nn o
O «H CM IO
T-l rH rH »-4
-O O U U U.U OOUCJOCJOOCJUUCJUOUUUUUUCJUUU O
A-45
-------
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
d" d" d" d" d" d" d" d" d" in in in. in in in. in in in in~ vo~ vo ^o ^o ^o ^o ^o *o vo ^o r** r*^ ^^ ^* ^^ Is-1^* i^- ^^ r** co
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
• i-r 1-1 ^ 1-1 »-f 1-1.1-1 1-1 »-i n. 1-1 t-i i-i t-i t-t: »-H. t-4 t-r H-I t-t 1-1 1-1 »-f I-H t-v h-i •-(• M i-» >-) KH I-H. 1-1 »H, K<. I-H. »H. H-I. 1-1 1-1
(/)t/)i/)into'tointotntO(/5(/5i/5iotoi/)tot/)i/)iOLOtot/)i/)totot/)(/5i/)tni/)int/)toincni^
.CM
«0
CM
*-l
•
vO"«H
.rH
•• *
U>*
rH
-• X
in *
rH
~ CM
in in
.CM. •
• o\
0 f^
1 rH
X
«- II
U- M
•t-4 U)
d- in
rH rH
d-
d-
O
CM
•
co.rH
rH
•• #•
r- *
«-H
•• X
r- *
«H
•~ «H
o d-
fO •
0. • 1^
CO O «H
1 CM
OX
1 II
O U. N
O t-i t/)
vor^
*-) rH
d-
0^
0
d-
•
O <-H
CM
•• *
0> *
«H
" X
CJ\ *
-
X
CD
O CO
-I 0>
< rH
* ro
vO O
o . •
-.0 0
0 CO CM
ro, • ro *
~-,rH - *
rH
00 I rox
•• *
>- rO rH
H- to ro ro
t-iro — h-
_l • CM vO
O r-l CO. • •
CO CD rH O O
.< — 1 CT>
Oh- X
1- I/) II — II
O X U. N
CD l-'»-i 10
O rH
ro ro
CM
ro
ro
CO
o
•
o
d-
ro *
»- •*
ro
ro x
•• *
ro
ro ro
~>.co
d- d-
o •. •
co O CO
1 0
O X
h- ~-.ll
O U. N
CD •-< to
CM ro
ro ro
rH
r>
0
0
•
rH
*
#
X
#
0
ro
•
o o^
CO 0
rH
O
\-.ll
O N
CD W
d-
ro
CO
in
cr
CM
•
t>
in
•>•
X
— in
CD O
.0 d-
_l rH
.-< CT>
*. •
~ h- 0
o in
d-co *
~ 0 *
•
O rH X
#
>- 1
1- rH
I-H in d-
_l. • rH
O l-i CM • O
CO 03 rH rH CO
< ~ O
OH O
l- U3 II II H-
o x r\jo
CD H- 10 CD
O
d-
CO
in
0
CM
•
r^
in
>
x
~ CD d-
CM.O h-
d- _J O
~.< \0
* CO
Q CM •
CO 0
Ct ro O
o CM ro *
r^ «.•>«•
CM. • rH
Q 0 d- X
1 - *
>- fO rH
1- ro d" t>
t-< ro ~* in
_l ro ro d-
HH • • •
CD CO 0 d-
,< «- 1 ro
1- X
10 II ~-.ll
X U. M
H- "-H l/>
CM rH
d- d-
\0
vO
0
rH
CO
•
0
d-
d- *
-. #
to
d- x
•• *
ro
d- ro
— C^
o o
o • •
CO rH CM
1 ro
0 X
1 — .11
O li_ M
CD I-H 10
O ro
ro d-
vO
d-
fe.
in
d-
•h
in
d-
«^
o
o •
co ro
1
O X
1- ~
0 U.
CD "-I
d-
d-
.o
A-46
------- An error occurred while trying to OCR this image.
------- An error occurred while trying to OCR this image.
-------
C\J
r-.
o
IE
LU
03
O
Q£
CD
3
00
o
o
00
cc:
o
oo
o o o
o o o
«H CM IO
000
000
o o o
LULJ-LU
CD CD CD
o o o o o
o o o o o
d- in vO t^ co
00000
o o o o o
o o o o o
LJ LJ LULU LJ.
co en co co co
• • *
»H X
CJ
+ 2
ooo
ooo
ON O T-l
O «-l «-l
ooo
ooo
LJ.LJ LU.
CO CO CD
d-
in
O O
00
CM ro
«-i ^n
oo
o o
LULU
CO CO
Q
in
in
a
fe
X
d- r-
00 (M
2 2
LU LJ
•5. -5-
LULJ
CM<
in in
LJ or
o LJ
•. CO
LL o
t~ i-
in cj
t~> o
Q
U.
X
LU
Q
I
«.
X
I
*,
LL
I
CD
M
LJ
X
LJ
CJ
LJ
LL
LL
.LJ
2 Q
O 2
in
CC ••
LJ O
CM
CM
t*- in
ON <
CM
1/7
cc •
CO >-
CJ
I- 2
2 LJ
LUCD
5; <
a
O 2
LJI-H
> I-
LJ O
Q LU
K
_l O
LU CC
Q a
o
X (^
LU CM
«-l U 0
r-l 2,1-H
d- CC
CO ••-LJ
I :* x
ON CC. CL
d-.< in
in Q. o
— LU
2
in LJ
— in
a.O
CM - O
H >-
D in
CD D
O O
.< I-I
LU LJ
X CC
»- a
o •
CC CM
LL
>-CD 2
_l 0.0
in _J cc
D O. n
O CC >
I-H O 2
> LJ LU
LJ H-
CC LJ "
CL^: >-
CD
ON X.O
in.u _i
.-I CC O
d- < cc
«-i LJ o
• in LU
IO LU h-
DC LU
O-.CC
•X N CO
LU Q —
CO X
-LUQ
2 •• CM
I-+X f-
I- - O
o r> x
.O «..LU
QTHCO
CO l/i
o *:
to
i-i in 2 2 »
cc < < cc LL
LU I- t- LUO
u. in in 2
LL 2 2 CC 2
i-i O O-ID.O
D CJ CJ t- n
in
in LU LU • n.
i-i x x co >
X h
Ov 2 Q
— < 2
JW.<
LU CC
2.K O
O n
X X 2
CL U <
CC.LU
• ,<.O
0. LU.O
H (/) .
CC.LU _l
cc,<
CD -.O.
2 D ^
CD Q 2
LU
CJ
CC
LU
S
SI
O
U
LL
O
2.O
LU
s: D
I-.LU.
cc in
<. <
a co
LU
Q in.
OOOO
oooo
d- in vO. h-
1-1 TH ^-1 I-I
oooo
oooo
LU LU. LU LU
CO CO CO CO
LU 02
in. LU 2- o
»H in o _J
CC:l-i CJ CD
CC.LJ 2
LJ lALU
2LLJ
^ S:LJ. •
-JO X ^
a _j H
a •
LJ LL X
K LL O
< O >
_i in CD.
^ in, CD
O.LJ 2 Q
_i in. i-i. LU
< >- Q I-
CJ _J LJ i-i
.<.LJ Q
CJ. LJ
ooooooooooooooo
ooooooooooooooo
cOCJNOrHCMiO.d-invOh-cOONO.-iCM
«-l«-tCMCMCMCMCMCMCM.CMCMCMtOlOlO
ooooooooooooooo
ooooooooooooooo
LULULJLULULJLJLJLULULULJ. LULU. LJ
COCDCOCOCOCOCOCOCDCDCOCDCOCOCO
o
in
oooooooo
00000000
to d" in \D 7s- co ON o
fOIOlOtOtOlOtOd-
oooooooo
oooooooo
LJLJLULULULULULJ
COCDDDCDCOCDCOCO
in
-LU.
< o
cc
i- a
.
2
O
l CC
• > 2
^ LU.LJ
CC 2
o 2:. •
>- t-i in
x.o
-£. O 2:
LU I-
2 2.<
0
in
LU
cc
CD
2
X __
CD.LU
n o
LU LJ
X CC CM
tO O
LJ LJ O O
> s: *-i
MO cc
t- in - I.M
cj <
in -
2.UJ
O U
LL
LL
LJ
in
CD
CD,
t-i CM 2
r^ o <
ON ^H LJ
^ _J
CC
m:
•• a
.
2 2
.0 O
to CD in
2 in
o _i
O M 2
CC,< O
cc in
t- LL
<-•« in
CJ CM
- O _1
cc 2 z> in
• < < ~~3 LJ
I- CD — _l
. .. ~* .m
3 in cc o n
CD \- ^H CC
Q CD-I3 in <
2 I-H LJ >
< CC 2 I
CO \-
in
cc
LJ
I-
LJ
•~ X
in
CC.LJ
LJ.U
»- 2
LJ.<
5: h-
— in
i-<
»- Q
CD CC —
n.O -g.
LJ LL S
X —
J- .
LJ XLJ
51. CD O
O i-i CC
-I-LJO
a xo
in
LULU
> 51 LL
i-iO O
I- _J
u a i-
LJ o
LL LJ a
LL > I-
LJ n O
I- O
LJ
O
CC
o
o
in
.o in
.-o: cc
LL LJ
t-
LJ LJ
— ws;
fO i-i —
* ^ cc
* in x
o CCLJ
LJ LJ 2: LJ
in —
cc^:
LJ
I- CD
LJ LJ
S. Q
5: a <
^. ^ - i
^ ^^^ r^
* _i in
:E.LJ.<.»-I
" in 2 o
K-f H
x o: LL l-
ID <
-1LJ LL
LL 51 O LU
Z5 in
>- _l Lu i-i
cj a,u cc
< _J < LU
• >- < H- s:
.o 2 in r)
• Q
Q
o in
CC HH
LL X
CM
a
»-
o
.o
< LJ
2 LL <.
M Li
LL LJ X CO LL Q a
LL LL X
X t- X
3 _i in _i
LL X 51 LJ i-i LJ
X X X LL D 0 Q
I- LJ
X CC
IDO
i-i t-
LJ <
X CC
.LJ
• *: a
• o s:
• < LJ
LJ I- H
cc in
< in
in < CD
LJ CJ
co in cj
< >- <
i-i X I-
cc a in
is a in
a x i-
2
o —
LJ in
in cc
x LJ
s: H
".LJ
»-i CC
CJ LU
O K
_J LJ
LJ 51
X X
LJ CJ
<
in i-
< in
CD
LU
X. Q
CJ w
< in.
h- 2
in i-i
_1 CO
— I-I Q
CJ =3 LU
LU 0 2
in in M
X < LL
to a LU
* Q
* LL
s: o in
— ..LJ
ON en
LJ o in
H CM.<
CC.LJ CJ
CD
3 <
o a. •
_l 2
LL LJ O
LJ KH
cj in in
I-H CC
CC «. LJ
H- — a
LU en in
5: in i-i
D.< Q
-I _J
O CJ CJ
> «- I-I
cc
in >- LJ
< h- x
CD n a
_i in
*: M o
u m 2:
t- t- <
in in
< CQ-CJ
in in in
in in in
< < <
_i _i _J
CJ CJ O
CD CD CO
< <.<
»- h- H
in in in
000
in in in
a a a
in in in
i—< i—11—i
rH CM tO
in
LL in
.CJ CJ CJ
------- An error occurred while trying to OCR this image.
------- An error occurred while trying to OCR this image.
-------
ooooooooooooo
ooooooooooooo
«-!CMrO;frtn\Oh-cOO^o*-iCMrO
CM CM CMCMCMCMCMCMCMrOtOrOrO
«-««-«»H«H»-4«-(»-l«-l»-lTH«-l*H«-l
OOOOOOOOOOOOO
Ul LJ Ui LU.LJ UJ LJ.LJ UJ.LJ LJ LULU
LJ
O
H
to
Q
H
<
H
X
O
UJ
X
LJ
>
1-1
•
f~
O
LJ
LL
LL
LJ
LI
H
<
•
X
0
LJ
X
LJ
>
H
O
LJ
LL
LJ
_l
<
^
X'l— (
X LL
i
LJ UJ
Q H
',<
o
O
o
LL «H
X X
_i LL
LJ H
a to
l-l
00
CM
«H U.
~ X
L __ I
I- LJ •
I/)D o _i+_i + a
I-H o ^H D -^
a. n ro.ti • ro o a. o a. it
I O _J X _J X 2
5T-X.O X O < < LL Ct
X X J- X.ll t- U II U II HO
— _i _i t/) H a
LL.LJOLJXO x LLMLUZ
i-fQ-OClXO X XOOiLiJ
CO O O «H
CM CM ro ro
O U
A-52
-------
ooooooooooooooo
ooooooooooooooo
_ I «V t WN _k IJ>* . X\ B^. »«rt X^. ^-^ I Al ^n _fe Irt
ooooooooooooooo
ooooooooooooooo
._
CD CD CD CO CCTCD CDCDCDCDCDCDCDCDCO
o
o
CO
o
o
CO
a:
o
o:
o
CO
CM
d-
X
CM
X.
i-
CM
CM
X
o
CM
CM
•. o
o CM-UJ
CM — Z
~ Z n
d-
tr CM in
•• d- Ct
O CM 0-
CM.O 1-1
~ U) -.
z •-• —
Z * O
O — d-
o o —
•• CM Z
o z cj
O I/)
LO -
•• ~- IO
\0
^^
»-t
X
CO
vO
IO
Q
X
o
•k 3-
M CM
O I-
I- -
«. o
— d-
o -~
CM d-
— CM
ro u
o un
tn ••
O <-l •• O
o »•
> CM
X O
O d-
<3 CM
I-H >
< X
Q
> o
t-i CM
I/) CM
u
o
_1
CO
•Q- O
.< d-
HCM
• t-t O
* #
X
CM
o
u
X
z
o
o —
CM Z
"- U
»H CO
I/) ^
l-i o
- d-
O IO
«H O
o •-
CM ->
^- o
Z d-
< —
ZrM
o o
U I/)
^. .£.
X
•
O
*
o
d-
o
*
o
d-
o
d-
o
d-
o
d-
X
X Z
o <
•• o
o in
CM
CJ
(ft
X
o
X
I/)
-Z. O
•• X
X
O X
•• o .
O •> H-I Z
•. o ~ o
^ X ^ CO
>- * H- U -
< \D > > IO
Q vO t-i i-t (_) X
IO •• •• CO O
CM X Q. X -. X
CC < K 0 U CO
L|J Q I-H I-H I/) »-H
H- »- h- h- I- t-
z < < < < <
MQ D QQO
X
o
*
in
*
o
in
X
££
Q.
Lul
Z
H- Q
< Z
Q UJ
A-53
-------
APPENDIX B
EXAMPLE SINGLE SOURCE (CRSTER) MODEL RUNS
B-l
-------
Listings of the input and output data for two example runs of the Single
Source (CRSTER) Model are contained in this section. The first example is a
standard model run while the second example is a source-contribution run. These
examples begin on the following pages:
Example Listing Page Number
Standard Model Run
Input Data B-3
Output Data B-5
Source-Contribution Run
Input Data B-28
Output Data B-31
B-2
------- An error occurred while trying to OCR this image.
-------
oooooooooooooooooo
oooomooooinoooooooo
oooc\icviin\o\ocM»-(inr--c\iine\jinoooco
oooooooooooooooooo
oooooooooooooooooo
oooooooooooooooooo
ooooinmooiomoooooooo
in
o
•
to
vD
•
to
in «-i
to o^
• •
to in
oooooooooooooooooo
o o o o in moooinmooooooo
CVJ
•
to
oo
h-
•
o
to
CO
•
to
d-
CM
OOC3OOOOOOC3OOOOOOOO
oinininmooooooooinoooo
»HOJ(MC\l(MOJC\JC\lCOCMCMlOfOtOfOtOlOtO
»H CM tO 3-
U U O O
\- \- \- H-
(/)!/)(/)(/)
B-4
------- An error occurred while trying to OCR this image.
-------
in cMCMCM>O3-roinininininininininininin*inminr%»cMrk-in*f~-fOinvOinK>
a •_•••••• • «••«••«•.•••.••••••••••••••«
o CD oood-c\ivooiot^o^cro>t>ovCT'a>o^o>K)o>ovooeoooovmooiocor^(>coc^(MvD
• Z d-Jd-iOinj-.H.H.-lr-l.-l.-lrH.-l.-li-lr^^^H.-l.^infOd-r-'-tintOd-invO'HtT'COCOd-
O I-H 4 .H .H .-4 .H CM >H .H -I .*
in at
«H -~
_l
UJ
> d-
u a
-J CD
I 2
I < 1-1 '4CMCMCM.H.H Ul
UJ > to totoin«-i'Hcor^r^ioa'incMa)inr~oco.-icMinocoa-aiOxocoioococoooroioioioiOfO
< < 2 d-f^-ooor^ioiod-oO'-ior^^r-vDvor^cro^-Hinrod-d-roiod-d-d-d-a-d-d-d-
-UJ >-l .H*H«-I^H«-4.-li-l«-I.-t»-«CMCM»HCM'-t'-l'H«H.-ICMCMCM.H«H.-1.-l^«-».-*.H.H*-t.-l.Hi-».H
in in o:
cc
ui ui
> i-
o ui CM * \o in in to \o to r- r- f- a CM to in t^ in in .H to o in * r- t^ CM CM CM r^ £0 CM CM CM 10 CM CM CM
CD S It • ••••.••.••••.«•.••••.••.«.•••...«.•«•••»•••
< — CD cMr-r-a*r^r^o.-iaicoiocMCsiococoooo»Hd-c>ooioooo
z
I/) I/) M
or z cc
UJ .O
i- f
ui i-
s < ^
— •> a
uj CD
z _J z
O UJ l-l
1-1 Q:
\- ce
< o
> i-
uj .a
_i ui
.UJ O
.UJ
i- ce
o
CM
* in oooooooooooooooooooooooooooooooooooo
vo a « ..... . •_••.• ...........................
CD ooooooooooooooooooooooinooooinoooooooo
Z «DO>Od-OCOCMCMCMCMCMCMCMCMCMCMCMCMOCMCMCMin>OvOCM^inr-CMinCMinCMOOO
o >-i *.* * in in * r- Nr-h-.NNr-N r»N.N N.N ^ r- in * * in h* in * d- in in h- «o o «o *
CD . . C£
CO CM
. • O
to d- ~
_J 3- oooooooooooooooooooooooooooooooooooo
ui a ......._............-...«........ ..t^« .......
> CD oooomooooooooooooooooooooooooooooooo
O UJ Z rt vO ^ CM- in O f^ CM CM CM CM CM CJ CM CM CM CM CM CM »H O» ^ >O iO CO vO O F^ CO O O> CO t^ h- CO CO
o _i 1-1 mvONp^d-d-J-h-r^r-r^Ni^r-r~i^i^r-r~vDin>od-d-d-d'd-d-*d-d-**d-j-d-
I cc
CM I 1
— .UJ
_j in
UJ to oooooooooooooooooooooooooooooooooooo
o > ui a • ......... ..........................
in ui ~> CD oininooominooooooooooooooininooininoooooooo
• _i o z
^4 CD .•-!
< < cc
-Ul
in i-
o ui
O UJ UJ CM OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
. > u. a .....................................
o — CD ooinooooininmooooooooooooininoooiniflooooooo
CD z oin.ncMcoincoininin.-ivociNCMvOiocMinr^ocMoinin\ovo>o>or->o>o>ONvo>o>o
< w w in in >o N o * >o r~ r^ * * * d- * a- a- a- * * * * * *
z cc
II h- O
-~ h 1 t-«
31 UJ t-
^ U. .<. rH OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
— — •> a ....................................
in ui CD ooinoooooinininmoooooooinirjininooooooooinoooo
.u z _J z * ^) CM o> t> co i^ r^ in in in in CM ^i o o o rt r^ in in in in >o >o
o o UJ w
Z ill CC
< i- cc
i— < o
in > i-
,1-1 UJ Q- Z
D _i .ui o
u o >-i
CD ui i-
Z I- CC CJ rt rt «H«H rt ^»H -H »H .-4 CM CM CM CM CM CM CM CM CM CM IO IO IO IO IO IO IO
« Z UJ
cc < cc
B-6
-------
in
UJ
i- o
o
CO
o
u. ~
o
U UJ
Ct V.
K IO
UJ *
51 *
D s:
(MCJM(\J(\J(VICVICVICM(M(MCM
CO CO CO CO CO CO CO
CO CO
UJ
UJ
0. .
z: IB
UJ UJ
I- Q
oooooooooooo
oooooooooooo
oooooooooooo
oooooooooooo
oooooooooooo
oooooooooooo
COCOCOCOCOCOCOCOCOCOCOCO
CMCM(M(M(\JCMCMCMCM(M(MCM
CVI(MCMCM(M(MCMCMCVJC\J(\JCM
UJ
in
o
u —
o o
-I UJ
UJ in
> £
O^ CO CO CO ffl CO CO CO CO CO CO CO CO
rH CVJ (M (VI CM N (M CVJ CTO^OONCJ'OOO>'
o
f- I/)
I CC
IB UJ
UJ UJ
OOOOOOOOOOOO
CVICMCMCMCMCMCMCMCMMCMCM
CO CO C CO CO CO CO CO CO CO C CO
oooooooooooo
oooooooooooo
oooooooooooo
lOrOIOfOIOIOIOIOIOIOtOrO
UJ
_l
UJ
•-ICM 10 d-
.o o o u
in
i/)
o
ct u
^
O X
i/i z:
in o
UJ
oooooooooooo
oooooooooooo
CMPJC\IC\JPJC\I(\JC\ICMC\JC\JCM
CMCMCMCMMCMNCMCMCMCMCM
OOOOOOOOOOOO OOOOOOOOOOOO
oooooooooooo oooooooooooo
oooooooooooo oooooooooooo
(M CM CM CM CM CM CM CM CM CM CM CM «H -H •
lOKiroiooiOroioionioiO
oooooooooooo
i/j ui in (ft
I I I I
I I I I
n it a a
o u o o
'(-t- t- K
i/) in i
UJ
Of
o
0.
UJ
u
UJ
(£
•z.
o
z m ct a: >-Z_)IDQ.I->O
< UJ < 0. -z_m>a.i->o
J.< Q. < D
- - --
_
DO
-51
UJ U O UJ
< WO Z Q
B-7
-------
s
0
_l
U.
u u
1-1 UJ
K X
Ul IO
2: *
13 *
| 2!
O —
>
—
CL *
UJ (0
1- UJ
Q
«•*
3-
1-
1-4
I/)
N
1- 2:
1-1 «—
X
UJ
cc
UJ —
i- in
UJ t£
z: uj
. ^ H*
i-. UJ
Q Z
«
1- -»
I t/)
ID tt
M i, i
+* (jj
.UJ 1-
I LU
s:
*"
ii i
LU
i-
^
cc ~
(J
Z Ul
O 10
'i-i X
1/3 (/)
I/) S
l-i ID
S —
.UJ
I
1-
z
0
•5.
^
t>
t-
in
d-d-d-d-d^d-d-d-d-d-d-d-
oooooooooooo
CO CO CO CO CO CO CO CO CO CO CO CO
COCOCMCMCMCMCOCMCMCMCOCM
OOOOOOOOOOOO
oooooooooooo
ininminmininininininin
oooooooooooo
d-d-d-d-d-d-d-d-d-d-d-d-
lOIOlOlOIOlOIOIOIOiOlOlO
cocococococococococococo
lOIOlOIOIOlOIOIOIOIOlOlO
lOIOlOlOIOlOIOIOIOiOlOlO
o>o>o>o>o»cra-aa-o-(j>a
ininmininin. ininininmin
oooooooooooo
CO CO CO CO CO CO CO CO CO CO CO CO
lOIOlOlOIOlOIOIOIOiOlOlO
d-d-d-d-d-d-d-d-d-d-d-d-
CMCOCMCMCMCMCMCOCMCOCMCO
OOOOOOOOOOOO
oooooooooooo
o o o o o o o o o o o o
COCOCOCOCOCOCOCOCOCOCOCO
d-d-d-d-d-d-d-d-d-d-d-d-
lOIOlOlOIOlOIOIOIOiOlOlO
d-d-d-d-d-d-d-d-d-d-d-d-
IOIOIOIOIOIOIOIOIOIOIOIO
"Z. m cc cc >-z_J030_t->o
< UJ < 0- in d-
TH CM IO IO
...
in O ff- o d-
- • 0*> sO
CM CO IO
...
.CM 03 o o m
• o d- d-
IOIO|OIO«»««
in d- co TH
• • • d" 03 o TH
THTHO^OTHh-CMCOTH
• o in in TH CM TH
CO CO IO IO
.....
. . . co TH d- in
• CF?d-d-coinin
"lO CO IO IO TH CO TH
THosa'oocMOOO
. •OsCMCMO*vP-~THf'-
• CMCMlOlOrHCMlOO
TH CO TH TH
...
CM «H r-- o 10 «. • • •
IO CO TH CO f- 03
... CO TH TH
co 03 d- o a>
• O IO CM- • • • •
IO CM 0"* d" O* CO
IO 0^ CM d-
, «... 03 CM CO O
IO TH CM 0 K) CO CO
• a> co co
d- co ». • • •
CM co in co
lOTHood-ioojinoo
• O» CM CM CM CM
IO CM
, ... \O CO O CO
d-in^-oiocoo^d-o^
.. co in in co 10 co
»H CM CM CO
invOOSod-O^cOd-co
• co in in 0* cj* *o o^
COCMIOIOOCOCOCM
CO CO CM
O3 ^H \O O CO ....
. co »H r- co d- co
0 CM CM CM
• " " CM CM CO
O3 *"4 ^O O ^H
. .COtOIO....
in cocoioiod-cod-co
co in a* 03 &•
,«H ...ococoro
.HvO*-lf-OinCMCM CM
>- • CO IO IO
< COCMlOlO»«»»
Q CO CO d- CO
~) ... (O 0^ vO 0**
03\OCOOCOO(\1COCO
'in .«cococococo co
CM CO IO IO
II ....
O ...o-nd-lO
2^ vD *H CT* O ^ »H CM *D in
i-i «cor^03oocoo
CO CM CO CO CM CM CM
d- II
O CD II II II II
II «I II 0. II CC TH CM
VI/1BUIU.L I _J
"Di-i^l~- in
CO CO
TH CM IO IO
...
IO TH C^ O TH
• a 03
CM CM 10
* • •
IO TH 0 OCO
•OIO IO
d-10
IO THTH 0 IO
•o d- d-
d* IO
...
IO >D O O CO
. «0 CM TH
IO IO ....
O CO >O TH
• . . d- o 03 in
C003OOO03O03TH
• O IO IO TH CM TH
d- 10
« . • TH r- TH r-
, CM TH o O ^ CO tO CO CJ1
• oiotoiood-in
IO IO TH CM TH
covOooacoTHinio
•OCOTHCOP~^d"
IO IO TH O TH O
TH CM TH TH
...
cM\or--oco • • • •
COCO 03 0 CM CO
... CO TH
CO TH 03 O CM
• C^d-d* • • • •
IO CM IO «H IO TH
0 d- CO d-
10 THIO o r- co CM
• o> d- 10
IO CM .««.••
d* co co co
. • • • *• in to in
»O O T-T"O- Ifi IO «H ^- *H
• O1* O CM C\J
»-tcj ro
•. • • IO CO CM CO
*moocMcoif)o^ito
* 0s O *"^ *^ ^
THCM IO CO CM
inTHo*ocoo3coTHco
• coiocaojinTHin
COCMIOIOCjTHTHTH
CM CM CM
...
\Q O- O* O O ....
•COlOIOd-COTHCO
CO TH TH TH
... CO CO CM
in >o co o 10
. .cQIOtO.***
vO CMCOIOIOTHCOTHCO
CM d- in TH m
TH ...OjTHTHTH
II\D030^O^-CMCO CO
>- • CO IO CM
< CMCOIOIO««*«
Q 0"* CO TH CO
~3 ...d-inTHin
OTH^OCOCOTHTHTH
in -COCOCOCOCM CM
IO CM IO IO
II ....
O ...vOCT'THCM
z! ^" o 03 o in in o1* *H T^
HH .COd-d-OvJCOTHO*1
THCM IO IO CO TH TH
d-.ll
>o ca ti ii ii ii
II < II O- II CC TH CM
FV ^- irt y > > x X
^ in 3! UJ LL. tL. 1 i
~3*-i^H*<- vO
< CM
Q TH
in 03 * o 10
• 0- TH TH
CM CO
in TH 10 o vc
• a to co
CM CO
in •o 10 o cc
• a in d-
CM CM
d- 03 in o TH
* CT1 *O
IO OJ rO
• • •
m \o ID o o1
• 0s vO iJn
CM fO
d1 *H CO O vC
• a -H
10 c\j
...
10 TH r o a-
«O CO TH
IO CO
...
d- f~ o o ec
. O IO CM
.OiO
d- CO 0 0 03
. o 10 co
03 IO
...
d- CM TH 0 0^
. o in d-
\O IO •
IO
. . « in
10 CM o o d- TH
• O CM CO TH
1^* IO
.
. . . TH
• o d- 10 O"1
f- IO
10 r- o o >c o
• O IO CM CT
*o 10 r^
...
d- coo o ec •
• O^ ^t" fO CO
r~-co o
03
d- r- r~ o d-
• O> d- d- •
r- co 03
CM
. . . d-
. . 0s IO CO
* CM •
LO
. . . . d-
IO 03 CM O CO CM
. • 0"> CM CO
IO CM
. . . K)
d- o o o in 03
coco
in TH co o r- 0s
. CO TH CM
CM CM X>
....
03 O O O IO •
. CO TH TH fO
TH C\J 03
CO
... TH
r- O CO O TH
. . co 10 10 •
f- THCM CO
CO O-
«H . . . CC
II 03 TH O O CO TH
>- • CO IO IO
< CO CO •
O 10
~^ ... IO
r^ in o o co cr*
in . o> ro TH TH
TH CM
II •
o . . . ^
Z! 03 10 10 a
CM CO IO IO TH
d- II
O CO II II II
ii < ii a. ii cc -H
cc i- in z: > > i
>- in x uj u. u. -i
"> T-I < K < < X
B-8
------- An error occurred while trying to OCR this image.
------- An error occurred while trying to OCR this image.
------- An error occurred while trying to OCR this image.
-------
N in to
.H CM
CM CM
vO tn CM
• 0s
i-< CM
^
in in in
CM CM
.
d- in oo
• O
CM CM
>
to in 1-1
* o
•-< to
.
IO •-! to
« O
to ro
.
tO \O d"
. o
CM IO
CM m d-
• o
<-l IO
.
CM 0 to
« O
to ro
CM >O CM
ro ro
.
CM «H iH
. o
d- to
.
IO «-J 0
• o
d- 10
IO ^ t~
. o*
ro CM
•
3~ «n ro
• 0*^
CM CM
.
in in 0s
* CO
^H CM
vO in ro
« CO
.H CM
.
N in 10
. CO
.-1 CM
•
N 0 d-
• .00
ro «H CM
^
«H .
ii N o d-
v «oo
<. —I CM
0
o .
r- o in
in • co
.-1 CM
n
o •
21 r*- o d~
t-t .00
•H CM
d- n
.O CO II
II < II O.
- l/l 3 UJ
"5 t-i < t-
o d-
CM CM
o ^
CM CM
O O"*
m
o d-
. .
o d-
in in
to ro
e
0 CM
to
. .
o cr>
vO in
IO IO
• •
O CM
•H «-l
O O
ro tO
IO IO
. .
0 vO
vo in
to 10
. .
o in
CM CM
• •
o to
CM CM
0 CM
. •
o in
vO *JO
* .
o d-
N N
o co
in d-
* .
o o
CM CM
. .
o m
CM CM
0 0s
ro CM
. •
O »H
ro ro
. •
o d-
\Q
IO
• •
0 CM
>o
to
. .
O T-l
0
to
• «
o m
>o
to
n
II Ct
> >
U. LL
< <
....
o o d- to
-O 00 N CM
in r- in CM
ro o in o
CM d- d- in
to oo to \D
i-l -H »H
\o o in N
O 0^ i-* O
.....
o o in d"
d- >O oo O
oo o> co in
iH iH
....
co 0* in o*
•H •-! m i-i
CM CM
....
v£> d- in d-
N ro CM ro
to o d" G
CM CM
...»
O> d- in d-
to ro o to
iH O i— 1 O
CM CM
c d- -H d-
o to >O rO
00 o o
rt CM CM
• • • *
d- d- -H d-
r- to ^
UJO
o cf>
z •
rf
1-
l/)
»-*
o
z
o
K-t
J>—
o ^
UJ CM
CC
M
Q
ZlO
O 0
KH t
I- O>
Ct 00
1- TH
Z CM
UJOO
0 •
Z -H
o
-" O
l-a
^ .
Ct CO
>- IO
.< d-
D iH
If) «H d- O \C
• a* CM -<
d- CM
vO vO d- o to
CO CM
\o «H in o CM
CM CM rO
vO «-l >O O O
ro CM
...
in «H \o o co
• o> in d-
to CM ro to
d" \D 0^ O d"
• o tn in
CM CM ro ro
...
IO
« O^ rH
CM CM
...
d1 «^ 0"* O 0^
. • o in d-
d- CM
ro -H o o CM
• o m in
in ro
...
ro rt oo o d-
• o> in in
IO CM ....
rt d- rO o
. • « CM in N «-i
CM CM
• • « CM IO O d-
d- CM «H
d-Nd-Od-iOto-OO
^ CM in o *o co
• • •
rO f** CM G in • • • •
• orororocMtOro
in ro a- o CM r»-
K) 0 m CO
d- CM O O *H
• o ^ ^ • • • •
>O ro d- CM o oo
oo in d- CM
... CM O^ d" O^
ro vO N O CO
. • o in d-
d- CM .....
tn o \o o
. . . N d- in d-
ro^oinoiHi-io^ioo4
• a* ro 10
d- CM
... . in o ro o
rO^Od'OinOd'Nd'
• O> CM CM 0> CM C7>
CM CM
d-iHCMONOoroo
•OCMiHCOd-CMd-
CM CM d- a CM a>
....
in'-l'HO*-! * * • .
•OCVJCMOOIOO
CM CM d- d- CM d-
in o^ CM o
... 1-1
in >H CM o d-
• • 0^ r-) *— 1 • • • •
d- to CM o 0 ro o
d- o d- CM d-
It vO *H i-< O «H «-4
>- • 0s to ro
< CM CM ....
Q O O IO O
O ...vOd-CMd-
'in • o to ro in
CM CM
II ....
2!NinCMOOCMrOCMin
i— i • CJ* CM *H N CO CM CO
•H CM O CD II II II II
II. < II Q- II Ct ,-1 CM
>-(/)3UJU-U 1 _l
"O t-i < *— 1 CO
4 .
ct o**
v d-
< d-
O *H
in ^ d- o in
If) CM rH iH
in-i d- o CM
d- CM —i «H
in >o in o K)
. CT> O o
ro CM >H>H
d- «H vO o oo
. cr> o a
d- CM iH
...
if) >o co o d-
. 0^ CO CO
IO CM
d- »H a* o o
in CM
...
d- N (3s o to
• 0** 00 CO
N CM
...
rO vO vD O 0^
. 0^ »H O
d- CM 1-1 -H
d- co CM o ^
• o in d-
a to
...
d- ro iH o iH
. • o d- d-
.H ro ....
** &• d- d- oo
roromoNd-coinCM
• o in d- «H i-) 1-1
o ro
• • * rO d- o d-
d-CMroovo^vOvOin
. 'Oind-CMood-vO
00 tO 1-1 rH -H
d-Ni-IONNtONO
• omd-ff vocMro
oo rO 0s oo ro o
i-4 iH i-H
d"tO*HOO^ • • • •
•o>oinoiororo
O IO CO vO 0^ *O
«H N 00 «H CO
d- to o o oo
..omd- • « • •
0^ to d" CM CT1 i-f
vD vo in as
* * • in co o N
d-NOOOCO *-ii-l»H
• &• in d-
N CM .....
co CM in CM
. . . d" vO CM vO
d" CM \O~O d" IO CO 0^ 00
• cr> in in IH IH
. N CM
• . « i-l CM «H CM
ro\Od*oorO\OOsvO
•C>d-d-iHoONco
d- CM ^ -H
d-vOIOOO\DCMOCM
. .o^tnino\oiH*o
d- CM d- co N co
i-l iH .H
...
invDlOON • « • •
•O>CMiHvOCMOCM
io CM in \o IH \D
IO CO N 00
... iH i-l i-l
\O vO CM 0 d-
• •O'd-d-**"*
in CM CM in CM o CM
d" O \O *"* ^O
i-l ...rooONco
Hini-iroorOiHi-i »H
> • a d- d-
< d- CM ....
D in CM O CM
~y ... in ^ i— i *o
in^OIOOlOCMoONOO
If) .O^IOlOiHiH r-t
IO CM
II ....
O •••d"*COOO
Xin-ld-olOOd-i-HCM
i-H "OCMCMCMvONN
d- CM 1-1 1-1 in
d- II
vO CD II II II II
II, < II CL II Ct .H CM
£t(-l/12:>>I X
>-l/)3UIU-U._J —1
"5 •-*< 1- < < X I
ct
D
o
X
1
d-
CM
x
2-
*"
_J
ct
-^
o
I
X
<[
2:
~*
•Si
:£
UJ O
o in
• in
< d-
D i-l
in IH a o oo
• 00 CM ^
tO CM IO to
d" vC) CT1 O N
• CO •-( O
ro CM to ro
• CO iH O
to CM ro ro
...
\O «H O O S
. CJ* 0s CO
IO CM CM CM
...
in iH i-r o o
• 0" O O
CM CM rO IO
...
d- 1-1 •& o CM
ro CM CM CM
...
ro >0 co o N
. CT1 O CT*
CM CM IO CM
...
ro «-i a o d-
• CT* CO CO
IO CM CO CM
tO i-l 00 O N
• 0*- d- ro
d- CM CM CO
...
CM vO O O N
• O \O If)
rO ro CM CM •
... K)
CM sO 00 O CM d"
IO CM CO CM
• • • o
«H \D CO O CM C\J
• O CM CM CM
CJ CM CM CM •-<
CM vD co o ro rO
rO CM CM CM O
• • •
CM »H vO O O «
d- CM CM CM CO
• • •
CM «H in o -i
• o^1 in in *
CM CM CM CM CO
\D
• • • m
ro «-( d- o 3-
• O*1 ro ro
d- CM CM CM •
• • • in
d- ^H CM o f^ ro
, • CT\ o 0s
OJ CM CM «H
• * • d-
in o o o o ro
• CT^ «H «H «-t
^H CM CM CM
\O O h- o co *>•
• CO *H O vD
•H CM CM CM CO
• • *
f*- O CO O O^ •
• CO »H O f1-
«H CM CM CM vO
CO
• • • *— 1
\O «-t O O ^O
• • CO *H O •
\0 CM CM CM CM \D
^ vO
•H • • • CO
II vO *£> O O CO »H
>- • CO -H O
< CM CM CM CM •
Q in
in t-« o o o co
in * O1* o^ co ^H
d- CM -4 .H
II .
o ... in
2: d- N ro o CT« vo
*— * * O* 00 N 00
in CM .H .H .-i
d- II
O CO II II II
ii < ii 0. ii ce 1-1
ct i- i/) 2: > > i
>- U) 3 UJ U. U 1
"3 >-"< t- < < I
B-12
-------
...
CO CM CO
h- o in
co in io
•-4 iH
r- r- r-
co io co
iH «-l ,-1
in CM o
r- io f-
CO -H O
•H .H i-l
...
d- f- o
f- d- io
co a d-
iH «H
...
CM CM 0
f- \O O
CO N CO
•H i-l
...
CM r^ CM
r- N f-
co in co
«-l iH
CM CM CM
r>- cr r-
CO IO CO
** i-l
CM CO CM
r- i> o
CO CM CO
•H iH
...
CM CO CM
CO CM CO
*-> i-l
...
CM CO CM
N N f-
CO CM CO
•H i-l
...
CM CO CM
N N l>
CO CM CO
-H iH
...
in in r>-
in \o co
*O CO MD
i-l iH »-l
II
x
_i
X
cc
o
I
1
d-
CM
X
^
^"
_l
CC
^
o
T~
X
<
Si
s
UI O
U 0
z «
< CM
1-
in
»-*
D
Z
O
1-
U vO
UI CM
CC
Q
Z d-
O o
i-* 1
1- 0
< co
CC d-
K CO
Z CM
kl d-
U «
Z .-1
O
(J
CC
O *->
I
5
^
*-•
UI 0
o cr
z .
^
i^-
U1
l-l
Q
Z
O
l-l
K
O CM
UI CM
CC
l-i
O
Z 10
O o
«-• 1
4 IO
CC iH
1- IO
z in
U CO
Z-H
o
, io
...
d" i-l •— 1 O O
. o io io
in io
• • •
& W OJ O CO
• O h- \O
(^ ro
"^SSS
*O ro
• • •
IO \O O"1 O tO
• QN t*~ r*-
d- CM
...
d- N i> o o
• CT IO CM
•£> CM
...
d- CM d- o co
« cr CM CM
O CM
...
io »o io o in
. cr in «H
IO CM
...
• CO iH i-l
CM CM
in i-i co o io
• CO vO
CM CM IO
...
. co \o in
i-l CM IO IO
in »H \o o i-i
. « co in in
t>- CM CM IO IO
^>
i-l ...
II vO i-* CO O d"
>• « CO d- d-
< CM CM IO IO
Q
"D ...
\Q *Q CO O IO
in * co PO io
CM CM IO IO
II
o ...
z: vO in cr o co
•H CM IO IO
d- II
o K) rO
^1 rtrt -1
ssss
O f^ r-4 m
^1 »H «H »H
....
cr co
(T
Deo
0-H
X
2
i£
•— '
UIO
.0 in
z .
< «H
h-
l/)
1-1
0
Z
o
.*-(
1-
oin
UI
cc
•i-i
o
ZK)
oo
t-t 1
I-.-I
4 *H
"CC CM
Hf-
Z d-
UI vH
O »
Z.H
O
u
o in
i-in
CC 00
vH
o co
d- CO CO CO
...
d- *-< 1-1 o io
« cr oj co
d- CM CO CO
...
d- .-i co o cr
• O"1 O O**
d- co CM -i
* • •
d- -H d- o d-
• 0" cO CO
in co - o -H
« cr in in
in co *-< *-<
...
d- «-i co o co
« cr in d-
in CM -H -H
...
d- CM cr o co
. cr CM <-i
f- CM «-irO«Hto«-i
o r- o r~-
• • • ,-f «H
ro ^ in o \o
• 0"' OJ *H • • • •
J- f\J ^ 4_1 ^) (\J ^ (\J
0"^ O 0*\ O
• • • o f- O f-
.* *o in o j- ,H ^H
• o^ ro ro
d"OJ«H*H * • • •
OJ 0 OJ 0
• • • \o o \o o
^" <~t ro o \o vH r^- «— t f^-
• 0^ OJ *H i*H *H
d- CM -H -i
• . . CO O CO O
• CT'-l'-tCMI>'CMf"~
CM CM "H «H r* »H
d->OCMOcod-Od-o
'CM CM rt CM l> CM O
»^ iH
...
d-\OCMOd- • « « «
• crininoooo
CM CM (O O \O O
IO O IO O
d-
II J- ^C -H O CO -H ,H
>• « cr in d-
< d- CM ....
Q CM 0 CM 0
~> ...cromo
in >o *H o d* ^j* ^ v»4 r—
in • cr d- d- .H in
d- CM
n ....
O ...COvOCMvO
x *o o »H o o m o in >o
in . cr io io in r^- »H ^*
»H CM .H r-l
d- II
\O CD II II II II
ii < it a ii cc i-i CM
cci-i/i2:>>i x
Vl/)3UJU.U 1 _l
"D>-"<|-< CO
• • •
\o *-4 oj o r^
• CO «H O
OJ OJ OJ OJ
• • •
\o »-i ro o m
ro OJ «H «-*
* • •
\O *H Cf O C"
• cO o 0^
ro OJ OJ r*
* • •
in \o ^ cf vD
• cO 0s CO
K) CM iH -H
...
d~ »-i in o vo
. co o cr
in CM CM rt
...
d- .-i >OO co
• CO iH O
d- CM CM CM
...
d" *O »O O »H
« CO •-! •-!
d- CM CM CM
...
d- o in o io
« cO o o
CM CO CM CO
...
d- <3 d- o co
« CO l> O
CM CM »H iH
...
d- \O in o d-
. co cr cr
ro OJ «-* *H
. * • •
& *•< m o ^~
• co o o
•3-OJ OJ OJ
• • •
•3" *o in o o**
• co in d~
"d- oj oj oj
*^£?ro
\0 OJ OJ OJ
* • •
d-i^-^- o 3-
in oj oj c\j
• "• •
d- o in o r^
«co d- io
in CM CM CM
...
d- MD in o d-
. -co in in
d- CM CM CM
...
d" «~i in o o
. CO CM CM
IO CM CM CM
...
d~ i— i in o m
. . GO iH »H
K) CM CM CM
d-rtgoo
IO CM CM CM
...
d- «-i vD o d-
• co d- d-
d- CM CM CM
...
d° O co o \0
• . CO IO CO
cr io CM CM CM
d-
^ ...
ii in •-! co o o
> « CO CM CO
< IO CM CM CM
0
-) ...
d- >O CO o o
in « co co CM
IO CM CM CM
II
o ...
rm rt cr o >o
i-i « co d- K)
IO CM CM CM
d- II
>o m ii M
II < II Q- II CC
cci-ins: > >
>- l/l 3 UI U. U.
Ti-t < K < <
• • • «
rO ro ro o^
ij1 T^ O"^ O
co ^H co r>-
*-t
,4 o «-I o^
cC' vD co in
cO o co I*1-
^
S 2 S S
co o co CD
,_!
• • • •
r~- oj p*"- o**
m r*" in in
co Q> co cO
• • • •
m in in in
J- oj j- oj
CO O*1 CO O**
• • • •
ro >N FO r^
rO -H ro «H
cO O^ cO Q"1
*-i r^ ^H r^-
OJ ^ W ^
O ^H O ^
• • • •
co h- co t1-
^ 0s r** c*
* • • •
\o r^ d- r>-
CO «H ro ^)
TH
• • • •
d- O d- r--
f^* O* I*1* 0^
• • • •
CM in d- in
\o o io o
l-l
II II
•H CM
I I
_l _l
I I
cc s:
^
UI 0
O O CM
Z «
I < vD
1-
l in
d- a
CM
Z
0
X 1-
O .-I
.< .UJ CO
cc
^ 1^
Q
z in
O 0
1- CO
< d-
CC £
1— >O
Z f-
Ul vC
0 •
z in
o
o
cc
O CO
O i-l
X
_l Z
^
cc —
UI o
D O CM
Z »
O < vO
1-
x iy>
I-H
Q
X
z
< o
l-l
s: i-
Ul CM
CC
l-H
Q
Z d-
O o
>-" 1
l~ \D
< \O
CC IO
1- d-
Z 10
UI IO
z *
o
o
o cr
i-i m
t- i-H
-------
2:
ID
O
o voinininin
UJ OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
in I I i I i I i I I i i I i I I I I i i i i i i I i i i i i i i I i i i i
Z *
o OD
M • OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
in KI I i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i
1/1 s:
>-< ^:
z ii
UJ hi oiOd-ir)ood-o
i- ii uj* o>ininroinovoinococotoiorfcoj-a'ooo>\oo>o
z z or~-r-o>a3d-oiO'-icovO>O'Hiotoror-'-'Osio«od-otoioco fin <-«
i- "-< <. •
o t- PJ
_
_i LU o
o a: w
0. 1-1 K-
Q <
CL
t- vOlf)ir)
in Z oooooooooooooooooooooooooooooooooooo
o UJ I I I I r i i I I i i I I I I I I i i i i i i I I I I I I I I i I I I I
t us: Nd-c\i«H>-ir^a>eoc\jcovDro\c>Dtococ\jinir5>O'Ocoir)r-c\ir~-covDOrH»HvOir)iocrd-
a z^ in>Dir)oeoir)^cMwc\jd-o^or~a3tnK5coiO{\jvDvOi^i^MrtoiocMcoif>d-d-voa'i>
r- o -i>-ioio3iinc\i\ooh-<-iiod-^icooovoooh-moinod-
^i oio cocvi^H
D II _l
a: <-> <.
z D
UJ O Z oooooooooooooooooooooooooooooooooooo
_l (j Z I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I
0.
ui 2: «
< X UJ
z < o
s: z
_J
a
B-14
-------
K>
*
*
2T
t\i minind-inind-d-mj-inininind-i/ivDininj-j-mifiinininininininuiinirifiinm
• OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
vO I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I
o
Ul .H
in n oo mj-d-d-mind-d-d-d-ind-toind-ifiininmd-d-d-inind-d-inmmifJininind-mm
\ >- • oooooooooooooooooooooooooooooooooooo
s < to i i i i i i i i i i t i i i i i i i i i i i i r i t i i i i t i i i i i
o a oincootMioiovoint^rtiotoainioccd-t^coa-ocr^o-ioor^d-tMOPjcjd-ioi^
h-<\iccavr'-ir^or^-d-r-r^3-CTi
•• vOOinr^-c'O>eoNinir)3-iod-co
1/1 2: -H3-ood-a>*ioiorH\o<-i^Hinvo
I- ^ • ....... • .....
>-i r~'H<-ic\if~>o>-i>-i^t»HcO'-icrco
z co
z>
n
o
•H l|
I/) LJ
l/l O — ----
s: <
M o
D i-o
Q. * OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
LJ CJ I I I I I I I I I I I I I I I I I I t I I I I I I I I I I I I I I I I I
;t o
uj
cj CL
o
i/) ii I
z o
.. o <
I- "-" LJ
Z t-
< O (-
(- UJ <
o a
_J M Z
o <-"
CL t-
o a:^:
UJ « OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
o t-i i i i i i i t i i i i r i i i i i i i i i i i I i i i i i i i i i i i i
"z. t\j^H<\jooj-i-iooi^d-r~-<\jroc>K)j-mioad-'Hh-
O ^imvOOsU1^-COC\lvOd-OC\JIOd-COtOrHf--U1C\)C><\J
o c\ir^oiO3-^i^ooooois-inio»-imi^osoovOiomc\i
II O
o i
I
Z Z
no
a: u co
LJ -
oooooooooooooooooooooooooooooooooooo
i i i i i i i i > i i i i i i i i i i i i i i i i i i i i i i i i i i i
-
B-15
-------
to
*
*
X
ID
„ _ ~~
1-1 • OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
Z tO I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I
o
-o
a: CM
t-l iH
< II
< i
Q
(J V ^ «^ •^f ^ >^ «^ 1^ ^ w 1_* ^ ^ ^f ^ ^ ^> «^ ^ ^ ^ ^ ^ w ^ ._- ^ ,^ ^r ^f ^, -_. ^ ^^ ^ ^ ^
u 2:
i/) ¥ 03 md-d-j-iniDiS-d-md-mininininin^Dininvoinininininmininininiflinininmin
X • OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
s: co ro I I I I I I I i I i I I I i I I i I I I I I I i I i I I I I I i i i I I
ID • '
10
o cn ct
*-**-! O
1/1 Q I-
(/) Q. --------- --- -------- „ --- . ----- „„ -----
t-i liJ inir)r-tf)3-onoortooeococotococavO>OvOvD'HpjMc\ic\jc\j\oinc\i<\)c\ic\jK5to
s: d- O CiJCJCMi-a-d-^d-iOrOtOto^-d-a-d-d-d-d-d-d-d-d-d-d-d-d-J-d-PJKjtOd-d-d-d-
fr" y *^ •»^ *—• • *>^ »^- «^ *** •— • «-^ i- • "-^ >- • «— • «-* »fc- — • "^ «— • ^* ^^ '—• *-* -^- ^* -^ ^P» ^-" ««^ -*^ '-' •*- «— ' ^^ «— •*• • *-^
n io ind-d-d-d-inininvOiniririnininoi^vOvOvoinininininininirivOvDinvOinininin
Z (-> • oooooooooooooooooooooooooooooooooooo
O < o \oini^o
Q. co Z tMPJCJCJ
ain inmmd-
• D • oooooooooooooooooooooooooooo
II O •H I I I I I I I I I I I I t I I I I I I I I I I I I I
,u
z
o
o
O LJ
Z I I
13 I ID
[Ed- "-1
CM I
UJ
_j 5; Q
a. D z
z z o cMCMd-
LU < t/)
s CT> lOvomvOvOvOvOvOh-Nr^r-r^OvOf^o^Nf-r^vOvomminioovocot^vor^^Ovof^t^
• oooooooooooooooooooooooooooooooooooo
o i i i i i i i i r i i i i i i i i i i i i i i i i i i i i i i i i i i i
uj z r*r-iO'^Nh-onvO'Hinr-d-'H»Hinc\id-or^ir)ff'h-CMO^iocod-ir)vO'H3-oinoo
2; O
< o
z uj vocMd-c\icod-oc\joof^oco>O'Hcoir>oro>oc\i^coco(\JOooooNtr>c\joor--.-i
tn
)— uj
"Z. >- «3
< _1 Z
_1 CE <
o_ < o:K
UJ M
>- Q
B-16
-------
CM 3-3-3-3-3-&3-3-f3-3-3-=t3-3-tnin3-=t=t=t-3-3-3-3-3-tr>(nin=t3-3-3-&3-^
• oooooooooooooooooooooooooooooooooooo
IO vO I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I
* ior^pJt-cooio\cma-oj-in(Dvor-coocoo(\jr-cripji^d-vOCT>'-iir^
* od-r-a3(\iir)r-in3-CM>oioin\oincon>O'-ivOroc>(\jc\ii^ocror-c\jd-r1jt--(\)»Dc\j
5: .H^ir>ovo°CT'CC>0(Cc\jroc\itca'.d-oocM3-«-H-ioto>oa)CMi-i-ico«H3-ir)'HintocDC\id-o>O'-ic\ij-oo^-cofOin
s: .... ................... . ............
15 ^CM^CMCM--lir>CM-HCMCMCMCM"H'-ICC>-l«-t.-IK>CM*-l>-I'-ICMCMa>r--CO«~ICMCM.-l'H'-l'H
in
K
n CM
o II _
>- O CMCMCMCMCMCMCMCMCMCMIOCMCMCMCMCMCM'HCMCMCMCMCMCMCMCMCMCMIOCMCMCMCMCMCMCM
t- •"
** cc «.»«»»».».»»»»«»»«..».»»«»».««»»»»».
_j LJ >OvOtf>OTioo.**-i^ir>cccococMCMCMvO'H'-<'-i'-ivO>OvOvO>ocMvOCMCMCMCMir5ir!ir5
«t a. PjMrawd-d-d-roioiod-^-a-d-d-a-d-d-d-d-d-d-d-j-a-d-d-d-a-ioioiod-WPJOJ
"*D 7~ »H »H ^4 ^H i^ j I i \ ^H «— I *— I »H ^H I ^ ^ ,_( ^ ^^ ^ ^ ^ ^ t j ^ ^ ^ ^^ ^ ^ ^ ^ ^ ^^ ^ ,_j ^
0 u ^ ------------- . — . ------------- --------
(£ t-> OD d-d-d-d-d-d-d-d-d-d-J-J-d-d-d-d-ind-d-a-d-d-J-d-d-d-d-irjind-d-d-d-J-d-i*
t-i >~ . OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
of ro I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I r I
o a-
LJ ^i
to ii
X >-
•s: <
O Q
2
_l II it
1/5 liJ
1/5 u o
n 2 • OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
2 < CM I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I
in tc o*ir5r-o»Hh-cococoin— 'i
n o o^^ir~i^ovjtr>d-c\i'-ic\j\D'H
D H- >ocovD
LJ ^rOrf
O O
CM UJ
CM ce
o
tn ii I
Z U
o <
Z I- CM CM CM CM CM CM CM CM CM CM CM CM CM CM CM CM CM CM CM CM CM CM CM CM CM CM CM CM CM CM CM CM CM CM CM CM
< O H
(_ u ovO(Dooio
• o ..... ...... .
in O «HCMfO^d-CMrtJ>OCMrOCM.H«-lCMCM'-iCMinrOt-l>-ICMI<5K>CM
a
D
II O
z z- T
D .0 CO
CMCMCMCMCMCMCMCMCMCMCMCMCMCMCMCMCMCMCMCMCMCMCMCMCMCMCMCMCMCMCMCMCMCMCMCM
Lul £t I-
_| 3 (/) .>.>>^>>.^..»^>>. >>>.».>>>>>>>>>...>
Q- O UJ ir5\Ov£>CMCMCMCMrO'H'-ICMCMCMCMCMCMCMCM>OOvD>O>OfOrOrOIOl/1ir)ir)ir)inmiDir)
S X I CMCMCMd-d-^a-d-rOrOd-d-d-d-d-d-d-
x co >-i^ -- -~ --------------
U I
X • OOOOOOOOQOOOOOOOOOOOOOOOOOOOOOOOOOOO
2 I I I I I I I I T I I I I I I I I I I I I I I I I I I I I I I I I I I I
uj 's:
s: 1-1
.-lf--CMlOCMCMCM«-l
Z >- O
< _J Z
_i • Q
B-17
-------
^C>JCMK5CMCMCMCMCMCMCM'-ltO«-ICMCMCVI«-l«-llOrO.-lTH.-l.-IOJ(\jrOIOlOt'">IOcMrOCMCM
12 ii
D
«• O
in w
(- a:
1-4 tlj
Z £L
<
\£ ^ ^^ ^^ ^f «_ ^* *_» ^ ^* ^, ^r >^ ^, ^, ^^ ^, ^^ ^, ^^ ^^ .^ *_, -_- s_- -_- ^. ,^ ^. ._- ^^ ^ ^^
CM d-d-d-d-^-iOd-a-j-d-d-mind-d-inind-j-d-a-inmind-d-minminind-
• oooooooooooooooooooooooooooooooooooo
>0 I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I
-
in
* oo a-d-a-d-d-j-d-d-i-^-d-d-ct
*4 • OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
n >o i i i i i i r i i i i i i i i i i i i i i i i i i i i i i i i i i i i i
LJ
in
•s.
I-" UJ <
Z -
l^ Q (-
i/> Q.O
U (tt. OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
S 3- UN I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I
LJ iij ^-vocj'OococO'HcvjinTHocoincMto.d-^-iovj.Htnioinvor^od-ococjd-or-ocMO
o: i
v
n i i
Z U ............... ..... ................
O < *-l (M IO W IO ^H N vO ^1 W WK) ^ CM O> W IO »H K> tM r^ rt vO O> CT> * KJ CM «H CJ in W >O «-i N CO
1-1 IJ
CM .»»..»<•««
15 t t- \oinr-ine\it\jrooooocjcoeocotoin'H>OvD\D\oiOvoiorot\jcvicvi(\jc\jc\icM(\jioio
_J CO Z cgwcjd-d-rf^a-ioiOrOd-^-d-d-d-d-i-d-^-d-d-d-d-a-d-d-d-to TOJO d- * d- d- d-
Q. CO Z
. oin d-d-d-d-*d-d-m\oind-inmd-invDvDinvDiod-d-in*d-ininifivo>o>OvOinininin
a- u • oooooooooooooooooooooooooooooooooooo
r4 I I I I I I I I I I I I I I I I I I I I t I I I I I I I I I I I I I I I
cc
g
u i
o co
I O
X CMCMtMCMCVIWCMCJCMWCMNCMCVlCV) «M.
U
_| s Q ...».»...«....^......««.»»>^^>»»...»
Q, D Z vOinCVIvDiniOIOt\lrOO^COCOCOCOCO\DvD^^IK)IOlOvOvOvOvOvOIOPJ(MCvlWCUIOlO
£ 2: O ojrjj-CJd-d-d-d-d-rOiOd-d-d-d-d-d-d-d-d-d-d-d-d-d-d-d-d-d-iond-d-d-d-d-
UJ < in
• oooooooooooooooooooooooooooooooooooo
Q t I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I
uj z coo>*ff>
z: o r-covD<\J
< u oininco
z lu j-r^a-^i
!/) ............... ..... . ..... ..........
\- UJ -t CM d- CM «-4 O rt t\J IO (\l rt CO IO CM .H IO CM «H IO O> K5 IO CJ CM
Z > ID
< _J Z
_J (T <
Q, < o:a:»HCMiO;±in>Of^coo*o>-tcMrod-inO'HCMio:3-invO
UJ w »H •* rt ^1 ^ ^ ^ «-l -H «H CJ CM CM CM CM C\) CM CVJ CJ CM IO IO rO IO IO IO IO
V Q
B-18
-------
cvj d-d'd-d-d-.d'd'd'd-.^-d'd-.d-.d-.d-d-md'd-d-d-cl-d'd-.^d'd-d-d-d'd'.d-d-.^d'd'
• oooooooooooooooooooooooooooooooooooo
IO vO I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I
z
in
'ifl II
£ 8
^ I-H
13 a:
a
• oooooooooooooooooooooooooooooooooooo
in 10 I I I I i i I i i i I I i I I i I i I I i i I i I I I I i i I I I i I I
O 1
LJ C
I/)
D II
LU
z o
o z. z
1-4 < *
CO H-
i/iw o:o__ ______
f-« I-H O * OOOOOOOOOOOOOOOOOOOOOI
Z: O I-CM I I I I I I I I I I I I I I I I I I I I I
LU O-
LU
* o
II I
~Z. O
C\J O «f
O 1-1 LU
I/) I-
u t-
LU W «^ *^ ^^- *-• ~-r «-* •*-• ^-- ^^ -WX •-- »— ^- «-• ^-* ^-» ^^ ^^ *-• ••—• «— ^- '-' •—' *— *-• ^^ «-rf -~~ 1—• »—• »--
D- ^ Z
CM U • OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
O Z CM I
ct in o* «-i o i
II O CM IO l> >H (
z T
O IO
o
ct o
LU I t-
_l IO I/)
Q- LU
X l-l
LU I
Z « oooooooooooooooooooooooooooooooooooo
•• D I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I
LU 2: ..--.— - - . - - -
2! i-1
< X
•z. <
I- LU
•z •>- o
«t -J Z
_l cc <
Q. < a:
LU
B-19
-------
oooooooooooooooooooooooooooooooooooo
I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I
-
ii
c
o
LJ
CL
-J o invO»oin:rf^o*oococococ\jcocoir>'-i»O'-t'--i'H\o\ovCvOvOiO(\jc\ic\j(\jcr><\jioif>
< KI (\JCOC\Jd-d-d-tOfOd-na-3-d-^-d-3-d-dpd-d--3-*d-d-d-d-d-d-iOrOiOrOOO^-j-t\j
G) || ^ ~^ •— ' ^ •— ^^ *_* u'*-'^'*-'*^*^*^^^ V ^* «_* «^ ~_r — • ^— w ^^ ^ ._- ^_ , ^^ ^, ^^ ^, ^^ ^^ ^^ ^, ^^ -_-
>-
IE < co d-d-d-d-d-d-d-^d-a-3-d-d-d-d-d-ind-d-^d-d-^-d-d-d-d-i-^-jfd-***^-*
1-1 O • OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
< 10 I I I I I i i I i I i i I i i I i r i i i i i i i i i i i i i i i i i i
O CO
iij «
(/) tO
X
5:
ID II
UJ
• • (J
in 2
a:
o
a
in *-i LJ
in cto
>-* . oooooooooooooooooooooooooooooooooooo
^ ii r cj i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i
uj 2 o ov0mc\j^c\)rrcMi~-c\icotrir'-f-r^c'0'>!O'-ii'}d-«ir)a>vOd-^or^oooc\jro 2 i
_l > o
o_ tr — — -
z>in ^-d-d-a-d-d-d-inind-j-d-d-d-cj-d-inininj-d-d-ct-d-d-d-d-in^oinintnj-d-ind-
|| O « OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
U I ^ I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I
2 i
O ID l
o
tr
3
o
i
I h-
K) I/)
LJ
Z I
3 ID
i
LJ
_i s; Q
a D 2
s: S O
LJ < l/l
x o
< u
z LJ
i- LJ
2 >- IB
a. <
UJ
>-
oooooooooooooooooooooooooooooooooooo
i i i i i i i i r i i i i i i i i i i r i i i i i i i i i i i i i i i i
--
B-20
-------
oooooooooooooooooooooooooooooooooooo
t i i i i i i i I I I I i i i i i t t I t i i i i i i I i i i i i I I I
in»OCvIOCT'd'COCCIOIOfHr^'-(rOOCJviHlOcOlOCOOr^vCincM'-»C'C\JvOir)d-\O«-lCOvD
-
n
-z.
_
< to
a:
a: D CO d-ioiOiOioa-d-iOiOtotOiOd-iOtOd-d-d-d-d-d-iOd-d-d-d-d-d-d-d-d-d-dd-d-d-
l-H O • OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
< x ro r i i i i i i i i i i i i i i i i i i i i i i i i i i i t i t i i i i i
O CJ
uj ^-
in -H
x it
2: >•
o <
Q
• •
in
Z «-i
in ii
in LJ o
l-l o . OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
z- z CM i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i
LJ <
t- cc
i/) o
•-< t-
o a.
UJ
u
in u
CM cc
o
in i
n o
z <
Z •-<
i- o <
o .ui
_l tE "Z.
D. I-
IO CC * OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
O t- »H I I I I I I I I I I I I I I I I I I I I I I I I I I t I I I I I I I I I
i z
O3 U
a- o
CO Z
in o
« o
CM
ct
D
o
II X
Z O I
z> z ^
U
_i a: in «•«»»».«»».»..»»».»».«««».«..»«.«..»
Q. D LJ in \o CM CM CM cvi CM CM -( -H CM CM CM CM CM CM CM CM * vo \o «o vo ro 10 10 10 to in in in in in in in in
s: O I CMCMd-d-d-d-d-d-tOiOd^d-d-d-J-d-d-d-d-d-d-d-d-d-d-d-d-d-CMCMCMCMCMCMCMCM
U «H I
X ' OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
•• D I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I
UJ 2:
's. i-t
< x
z < o co in ••< ort co o -i CM co n CM o CM «-i o to to vo »H to in co co 10 j- >H in r-- co
t- Ul
Z V ID
< -I Z
-i a: <
Q- < £C(r
LJ >-"
B-21
-------
CJ d-d-d-d-d-d-d-d-d-d-d-d-d-d-d-^-d-d-d-d-d-d-d-d-^-d-a-d-d-d-j-d-d-d-d-d-
• OOOOOOOOOOOOOOOOOOOOGOOOOOOOOOOOOOOO
vo I i i i i i i i i i i i i r i I i i i i i i i i i i i i i i i i i i i i
-
z in
Z? ^H
>- a:
1-1 O
-i i
<
1/1 CO d-d-iOiOd-a-d-iOrOtOiOd-
d" • OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
*+ 10 I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I
in
x
Z < (Md-wcJd-d-d-iOtOiOcfd-
" W OV ---- • — • ----- — ---- ~ — • ---------- • ------ • -- —
in >-i t-
(/) Q CLo JiOioroiOiOd-d-d-iOrOd-d-rOioa-d-ind-d-d-iOd-d-j-d-d-d-ind-d-J-^a-J-d-
l-l UJ . OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
X O CM I I I I I I I I I t I I I I I I I I I I I I I I I I I I I I I I I I I I
II O
z <
O liJ
kH
PJ K- I-
O O <
I/) UJ
C£ Z
-•• i-t O
_l I Z CMd-Wd-^d-d-^iOd-iOd-d-J-d-d-d-d-d-d-d-d-d-J-cJ-d-d-d-K) t i i t r i i i i i r i i i i i i i i i i i i i i i i r i i i i i i i i t
a:
r>
o
II X
O I
z •-!
o
o
(-
IT UI
o 111
Z O I
OX ID
«H i
_i z: Q
0. o z
5: S O
X X UJi;
UJ < I/)
•• o
uj z
oooooooooooooooooooooooooooooooooooo
r i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i
o
o
LJ
10
< -I
_i tc
a. <
UJ M «-i.-i.-i.-i.H.-i«-i»H»-i.-ic\ic\jc\(t\j<\Joje\jc\j<\jwnfOrr>K>fOioio
>-
B-22
-------
IO
•*
*
2.
ID
z
_J
OIO.-lK>vOCMrO«K>
hH
Q
->
•• o o *f^d-od-iOKd-inor^voca*in
I/) 14 UJ CM «H^IC\l >H K5CJ
h- i- o:
>-i < i-.
Z Ct Q
3 I-
Z UJ
o u
11 Z Z
in o o
i/i u >-«
_
1-1 i- oooooooooooaoooooooo
< Z I I I I I I I I I I I I I I I I I I I I
Q uj
o
s: z
D o
r o
CM w
O 'X. a
«/) < 3
r o
.. x
2 *
< CM
_
o >-
0. <
B-23
-------
N
s:
5
O
o
cr
u
CL
PJ CM PO t\l CM CV10J CM
-------
*
2:
s: Q
ID O
*-l
•• ct
in uj
t- a
•-*
2 UJ
D s:
O UJ
I/) O
tr Z Z oooooooooooooooooooooooooooooooooooooooooooooooooo
>-< o < ininintooooa>cooowa'tf}ir)o^a3ir>o£o<\jco(\joa>oeoa>0^C'i/iinorvjcoincocoir>o 0-03 HrtlOt\JIOCJ^I IO»Ht\llOvOIO\OC\l CJIO
< •-«
a: a
o e
UJ U
in o
V Z 2
s: o o
(D (J n
»-
•• >- o ^miod-oiomcMiod-iDinm^iooinorod-d-ifiioinod-h-d-ompj^oio
W_)UJ IO CMIOIOIO«H«HIOrt»H ^IIOCM^I ^l»Ht\l,t\IC\JCMIO (MIO^HPO
I- IE £C
• HOW
i f °
Z ro
.o
i-< r z
i/i r> o
(/) Z 11
t-4 »-» I—
S X ^
ui < a: (Od-d-a-d-d-^d-d-d-d-d-d-d-d-^d-d-d-d-J-d-J-d-d-J-d-d-d-d-a-d-rfiS-J-d-^-a-^-^-d-d-d-s-^d-d-d-d-d-
£ I— OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
Z I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I
UJ ^^h-^
o
2
o
o
CM
o o:
w 3
o
i
o v ioinw(>
a. < d- CJ d- CM d- CJ CM CM CVJ CVJ d- d- d- t\l CM 10 d- K) CM d- d- cvl d- d- K) d- IO 10 10 d- d- d- W d- d- PJ {VI CM d- rO CVJ d- CM d- d- W d- d- d-
a:
X
UJ
UJ
«t
B-25
-------
I
a.
4
X
Ul
I
z
*
X
(D
in
i-i
Z IT
r> ^i
>- i
i-
hH
_l
<.
O LJ
U
M in <
< r
O
»«i
I-
H «-< «-t,CM CMCMIO «HK) •H^H^IK) |O>H«-4CM CM^HK) .H.HCMCM CM«H^H-H
I- >- CC
!-. _l W
2 CC Q
D D
O
Z I
O
i-i r
(/> =)
10 S
*-( hH
2: X
UJ H IO rt CM CM rO rO - CM IO CM CM CM CM CM CM >O K5 K) in in CM in C- CM CM CO 1O CM in in CO CO vO CM
-------
o
o
LU
h-
OO
oi
o
LU
an
o
OO
1 1 1
LJ-J
1
C3
oo
LU
^~
LU
O
Z
;D
o£
o
1 — 1
H-
OO
Oi
Z
o
o
LU
o
C£
— »
0
( x\
V t
LU
_J
Q-
5
X
LU
•?*•
<
rv
0
Lu
- or a. in i-
i- z < or < i-
tOLJOQUIh-XCD
CM or in >-t i- CD z
Of % ^^ i_ . I,! ^^ LJ LJ
1J ^J h— UJ f— K- 1 M
u) — r> _i a: r).LJ o
ill fY\ tr\ ^*i r\ T ^f
LJJ CD \D \J U- -L ^.
UJ x M z i- i- or
O 1- Or M D.D CD LJ
or i- in uio z >
D in z o • w <
O LJ O <.LJ >- X
ini-o or_iMLJLJ
< or or 5i —i o
.LJ or ui o zo CD < or
_J 1- 0 LL LU 0 _I < LL 1-1
Of f\ f^ ^^ *T ^f LJ f^ ^f
UJ U. <^ -L. ^. •"* QL ^.
zoo z UJ or or D
z M _J O-D z o z> LJ
or i-1 *-• D-
< KQ.
UJ Z O
_i in D z
D. M or z z
s I-H o
< 1/5 in o H
X »-H •-• Z >-
UJ X X •-•<
d-
vO
II
——
_J
1— 1
•k
in
^i
CO
fO
0^
II
in
«. , «. • co n
. • d- d- II -
d- - » I- d-
•• CO CM > \D
«. in •• • • M n a
• • »ro\o»>-Z
d-rHd- - --OinUJ
O. • »lOtOd- •• ••
• K*l f\t ~ ~ ,f\ ^. fit
n t CM m ^ MJ w CM
•^•.••COCMIO^-I •
• in. • • •«>covo
lO •tOfOvO'-<'O ••
^*~)^«>«.^CT1CO
o - • • • o n «
• •CMCMCMiiinn
^cM-»->-in*-
• •• «COCM<»-<0
CMinCM»«O-»
• _ __ |^ tf* » (—* f\t
^ . • * n ) \u ^ tj v ^
tr> 1-1. • » » o u ••
• ^CM • • n i- in
*> •^^H^-IU.UJ •
•, t-i • ». » < s: «-«
rH «.»HCOCMI-»-« «•
••in •• • •.••— i ^CT*
•• CT>. • •tOvO*«-t •
•H i-H • «-l CM ll.tl iH || 0
s:. n n M u d- z n IM u
-------
ooooooooooooooooooo
OOOOOlfiOOOOtDOOOOOOOO
co
CM
d-
o o o
• • •
CM in in
\o i-« o
d- d- d-
ooooooooooooooooooo
ooooooooooooooooooo
CM
d-
CO
CVJ
CO
C\J
CO
•
n
10
ooooooooooooooooooo
oooooininoommoooooooo
in
o
•
to
vO
•
to
in rH
to &
to in
ooooooooooooooooooo
o o o o o in in o ooinmooooooo
CM
•
to
CO
h-
•
o
to
•
d-
CO
•
to
d-
CM
ooooooooooooooooooo
ooinmminooooooooLnoooo
«-» CM rO d-
^ ^ OJ X. «H X. CO
OOJO •(_) •<_) •
------- An error occurred while trying to OCR this image.
-------
in CMt\it\j\o:i-K5u'>Lflinininininir>inininin;»'ininmr-ojr-in;j-r~roinvoin»-iocT''o
a • •••«•••••••.•«««••••«•••«««..••••••
O CD OOOd-C\J\OCTCTC*C7'lTONO>OOOC>C>tOOvCJ'COCOOOOlf)COfOCOr^OsCOC'CMvO
O l-l JCMCM(\JC\J(\JCVJC\l(\JCVJ<\ICMCMCVl-ltHCVI-4<-l>H
in or
uj it
_j co
I < 1-4 ^CJC\IC\l-4.-l.-lH<-l ui
ui > to fOK)in-*«-icof^r^iooNin<\iocO'-i(\iiriinor-fOc\jr--i^i\ic\iiororoK->rorr>
_) O A • »•.»...••«••••..••••.•••..•••••••..••
OD CD ioinotncovocoeo»oina>«-ioo>od-f^>oa).3-a>oocoioococoooroiororoioio
< < z *r^i>O't^Nioiod-aD^ior^'Hr^»o>o. r^cro^Hrtinrod-a-iotod-d-i-d-a-i-d-a-
UJ 1-1 .4tH«H<-l'H>H-l«-liH.-l<-l<-l*-l«-l
(A I/) IT
ct
Ul UJ
> h-
o ui CM
m 5: a
< — CD
Z
I/) >-> •H^«-IC\JC\J«-I<-I'-I'H»-I.-I<\|THCM«-I«-IC\I.-I^«-IC\J{M«
2 a:
rr
ui o
t- 1-1
LJ H
2 < «n
— > Ji
ki CD
z _i z
O UI *-<
w Ct
t- IE
< O
> I-
ui a
-I UJ
ui c_>
ui
o
CM
. tfj oooooooooooooooooooooooooooooooooooo
,CD ooooinooooooooooooooooooooooooooooooo
o ui z ^ivD«-ic\imor-cjcjc\ie\jc\ic\jc\ic\i<\jpac\ic\i«-(a>^ivoo>Dh-cocTocoh-r^a3a3
o _i i-« invor~r-d-*d-.r~r-r-r^r^r-i^f^t^h-Nr- ui a ................. ........ ...........
in ui > CD ommoooinmooooooooooooooininooininoooooooo
. _i o z
•-< 03 >-"
< < a:
UJ
in i-
o uj
G* UJ UJ CM OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
« > IL n ................ ........ ............
o — CD oomoooomifiinooooooooooootninooominooooooo
m z o m »-i CM o ro CM in N o c\j o in in -o o >o •o r~ >o \o Df>-^ind-d-*d-\ovDinr-ininh-\o3-'or-r^*d-d-**d-d-*d-d-d-d-d-d-
z tt
II I- O
— UJ 1-1
Z UJ I-
X U. < <-< oooooooooooooooooooooooooooooooooooo
- - > a ....... ........... . .................
in UJ CD oomoooooininininoooooooinmininooooooooinoooo
uj z _j z
o o uj •-<
Z w a:
< t- £E
i- < o
in > i-
f« UI Q. Z
o _i uj o
ui u i-i
ID uj i—
z (- a: o rt .H -H ^ 'H <-i -t ^ rt ^i CM c\j CM
------- An error occurred while trying to OCR this image.
-------
I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I
lil OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOQ
fO <_) OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
* CC OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
* r> oooooooooooooooooooooooooooooooooooooooooaoooooooo
z o ..................................................
\ 10 OOOOOOQOOOOOOOOOOC3OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
ID
a>
(/) LJ 000000000000000000000000000000000000000000^30000000
I— U OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOCO
I-H ;t a: OOOOOOOOOGOOOOOOOOOOOOOOOOOOOOGOOOOOOOOOOOOOOOOOOO
2T ID OOOOOOOOOOOOOOOOOC5OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
r> o .................. « ..................... ... .......
• • (/) OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
>- 10
I- UJ
1-H (_> CO
_i tr
< 3 UJ OCOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
D O U OOOOOOOOOOOOOOOOOCPOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
O I/) £t OOOOOOOOOOOOOOOOOCPOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
3 oooooooooooooooooooooooooooooooooooooooooooooooooo
ct u. o ......... ......... ....... . ..... ...................
>-" O l/l OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
<
o t^-
LJ OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
c> o oooooooooooooooooooooooooooooooooooooooooooooooooo
UJ a; oooooooooooooooooooooooooooooooooooooooooooooooooo
l/l D OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
X O ... .............. ... ....... ............... ........
"5. (fl OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
ID
o
l/l
h- UJ UJ OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
*-l _J O OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
Z m tr oooooooooooooooooooooooooooooooooooooooooooooooooo
~D < D oooooooooooooooooooooooooooooooooooooooooooooooooo
I- O . ............ . ............. .. .....................
~Z 10 OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
o "z:
*-> o
i/i >H m
(/I I-
•-* D UJ OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
5: CD o oooooooooooooooooooooooooooooooooooooooooooooooooc
UJ "-< cr oooooooooooooooooooooooooooooooooooooooooooooooooo
IT 3 OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
I- o ...... ...... . ...... .. ........ .....................
"Z. (/) OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
o
U O1T) (\ld-COlf) ODO»*K5 in\O
d" C\I-H IOC\J"-4-H C\J>-I'-(«H fOCM
UJ II I I I I I I I I II
pj U UJ OOOOOOOOOOOOOOd-CMOOOOOOOOOOOOOJlDinCOOOOOOOOOr^h-lDlOOOOOvOK)
o a: u oooooooooooooocvi;toooooooooooor*-a>vO!>oooooooococ\JC\J3-°ooo.Hai
(S> Z) Q^ OOOOOOOOOOOOOOrOOOOOOOOOOOOOO^cjd-oooor~\o
I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I
_
o o:
a D
o
10
o
cc
o
to
I I I I I I I I I I t I I I I I I I I I I I I I I I I I I t I I I I I I I I I I I I I I
'
Z> I I I I I I I I I I I I I I I I I I I I I I I I I I I I » I I I I I I I I I I I I I I I I I I I I
UJ a: ~
Q. o
2: 10
x
UJ CC ^C\JtO^^CvJtO*^00'O^rtC^tO*^C\l^4-^(\/tO^^COlO^^IC\IIOJ-rtOOIO3-^H(\Jlr)d-^C\J'OJ-<-IC\JtO3-»^CO
.*
UJ
s: ir
< LU
-Z. Q.
I— >- MMMM
-------
O O CM «H »H O CTi-c-d-o»-'inj-o^r*-rocjrwd-cococO'-<
»— ^,H«Hin:±^or^K)a3«HK)cooir)^inf^(Tico*ooiH^o»jrOf^or^*Hin\o\Q^\o
co^ir>^^^o^^co^ifih-^^^ifluiwio^^K)^^.^r^fOK}^co^^^»c^incc^^
td ooooccoooooooooooooooooooooooooooooooooooooooooooo
ro o oooooooooooooooooooooooooooooooooooooooooooooooooo
* ££ OOOOOOCOCOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
* 33 OOOOOOOOOOOOOOOOOOOOOOOOOOOOQOOOOOOOOOOOOOOOOOOOOO
2: o . ...... .... ....... • ............................ ...
\ U) OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO Q^O OOOOOOOO
5;
o
t>
in LJ ooooooooooooooooooooooooooooooooaoaooooooooooooooo
f— C> OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
»-« J- Ct OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
~2. D oooooooooooooooooooooooooooooooooooooooooooooooooo
D O ........ • ........ • .............. .... ..... .........
• • LO OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
> I/)
t- LJ
•-» O (O
-J IX
< D LJ oooooooooooooooooooooooooooooooooooooooooooooooooo
15 O O OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOGOCD
O LO Ct OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
D oooooooooooooooooooooooooooooooooooooooooooooooooo
a: u. o ................ « ..... •• ..... • • ...................
t-t O C/l OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOQOOO
<
•
O h-
-z.
UJ OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
(J U OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
UJ CL OOOOOOCOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
t/) ID OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
X O ......... « ....... . .......... .... ...................
'S. I/) OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
(S>
I/)
K UJ UJ OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOGOO
t-t _J <_> OOOOGOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
2 CD CL OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
Z3 «t ID OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
I- O ..................................................
^ I/) OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
O "Z.
t-t O
(/)>-> If)
10 I-
t-l D UJ OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOCO
2; 03 cj oooooooooooooooooooooooooooooooooooooooooooooooooo
UJ t-t CC OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
o: D oooooooooooooooooooooooooooooooooooooooooooooooooo
\- o ...... ......... . ..... ....... ............ ..... .....
2 in oooooooooooooooooooooooooooooooooooooooooooooooooo
o
u oin COCMO too^r^vo cod-ion
UJ II I I I I I I I I I I -I I
OJ -ioooooind-croooo'_->ooo
tjn Z) CC (TvtHOOOOOOOOOOOOlDCOd'h-OOOOOOOOOOOOCOfOvOvOOOOOd-vOCOcnoOOO "tOOO
O ~D r^OOOOOOOOOOOOO»Hr^O>OOOOOCOOOOOOO^-4-fHiHOOOOI*-O^d-»HOOOO ,OOO
• • 10 O ..................................................
t- l/l ffl«HOOOOOOOOOOOOOJ3-C\l\OOOOOOOOOOOOO*O«-irO'-IOOOO'-<r-ir!ooort
IE t\iTHc\jc>tnoo^a'Ki
in cooa3^cor-c3co(X3d-
o ....
i/) ^io^ivoio«Hor--K5^H
r-Lnd->oioc\j h-omt^-cj'-noi-CT'voinin'Hoiod-
OOrU«-»«-Ht-( <\jC\/^(^«-*.-t^*--
^ ^
< o
_)
Q. t-
B-33
-------
UJ oooooooooooooooooooooooooooooooooooooooooooooooo
ro o oooooooooooooooooooooooooooooooooooooooooooooooo
* Ct OOOOOOOOOOOOOOOOOOCJOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
* D OOOOOOOOOOOOOOOOOOOOOGOOOOOOOOOOOOOOOOOOOOOOOOOO
X O ..«.«..«.«....•.•••.•.«.........................
X l/l OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
to
a>
U) LJ OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
I— o oooooooooooooooooooooooooooooooooooooooooooooooo
*-• 3- ce ooooooooooooooooooooooooooooooooooooooooooooocoo
2 13 OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
D O • . ....... • ..... ......... ..... .... .......... .....
•• 10 OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
> (ft
I- UJ
l-i O CO
_l CE
< z> UJ ooaoooooooooooooooooaooooooooooooooooooooooooooo
Z> O o cooooooooooooooooooooooooooooooooooooooooooooooo
o to ix oocooooooooooooooooooooooooooooooooooooaoooooooa
o oooooooooooooooooooooooooooooooooooooooooooooooo
ce LL o ......... ........... ............................
l-H O 10 OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
»-i m
10 (-
•-" r> UJ oooooooooooooooooooaooooooooooooooooooooooooooao
~SL CD O OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
UJ i-i ce ooooooooooooooooooooooooooooooooooooooooaooocooo
(E 3 OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
t- o .... ........ . .......... .........................
"Z. I/) OOOOOOOOOOOOOOOOOOGOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
o
u ot^coco toovoio
UJ I I I I I I I I I I I I I I I I I I I I I I I I
CM o UJ ^roeONOooooooooooood-a-inoooooooooooo. cocMooroo^iflJ-eo»Hir)li-r^a'*'H
O ce. u rjrt^icoooooooooooooo^^-roj-ooooooooooood'i^ioo^Hr^mo^-cor-h-O-d-tnco
(/) D ir r--or^-Hooooooooooooir)oocMooooooooooooinot\lrOh-cjtMir)ini^d-in>Oroinco
o 3 *oo>'HoooooooooooowoN*Hoooooooooooococomo\o<\jinr^d-iooaimCT>eovO
.. tn o
(- 10
H- rO ^OOO*-4«H*H^H»-I^I^H»H^H*-l^lTHiHOOO«HiHOiH^I^H^*^tCslTH^I»HT-*^HO O"O OOOOOOOOOOO
o i i r i t I i i i i I t i i i i i i i i i i i i i i i i i i i i i i i I I I t i i i i i i i i i
_i uj ajcovCt\ji»-r--c\jffln.Hco:j-ioc\jtnro(y«-ivO'-i;j-Noc\ifOinNioeor'-eo
_i u a>OKfl*iriina)coc\J3-^i
o a: r^mort^-pj^a>d-coos
a. D a>ovDvOtoj-'H^irw
o
I I I I I I I I I t I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I
UJ ^H\OC\lvOONO^^-|'
u od-coa>^-tH
D ^^uico*inin
10 >OC>JrHvOT^Cjr-
covor^crococr
'H OOOO<-IOO«Hir)ff>a>^i>ocM
UJ ce
Q. O
~s. in
•x.
UJ
UJ
"
.<.
CL
C\JCMCMCM<<<<
B-34
-------
APPENDIX C
CODING FORMS FOR CARD INPUT TO THE PREPROCESSOR
AND SINGLE SOURCE (CRSTER)
MODEL PROGRAMS
C-l
-------
The coding form in Figure C-l can be used to prepare the initializa-
tion card and mixing height cards for input to the Preprocessor programs.
The coding form in Figure C-2 can be used to prepare the heading,
comment and namelist input cards for the Single Source (CRSTER) Model.
Figure C-3 is a coding form for preparing the plant and receptor
terrain elevation for input to the Single Source (CRSTERjj Model.
Figure C-4 is a coding form for preparing the stack input data required
for input to the Single Source (CRSTER) Model.
C-2
-------
PREP INITIALIZATION CARDS
Surface
Station
Number
— i — i—
i-
o-o
S- 0
ro U
•o c
C 0
3E4J
U
4-1 <1)
t/> t-
1
C
(O
M
Surface Station
Latitude
(deg)
— r-i — > — r-
Surface Station
Longitude
(deg)
— i — i — i — • —
i ' i
i —
1 *
0>
o
•r-
1—
M-
O
in
S- >)
(U at
.0 Q
3
Random Number
Seed
! !
I ' j
i 1
I 1 ;
MIXING HEIGHT DATA CARDS
Upper
Air
Station
Number
i
2345
i
i
i
! '
j .
1
t
' '
' i
i
S-
tO
-
6
7
1
§
8 9 10 11
I'Z
1
I
I
I
/
/
/
1
I
1
1
/
1
/
/
/
/
/
/
/
/
/
/
/
i
1
1
1
1
1
1
/
t
1
t
Z-
2
2
2
2
/
/
£
3
i
5
4
7
ft
a
?
/
Z
3
i
5
6
7
B
9
o
1
i
3
y
i Morning i
c Mixing ;
* Height
03 (m)
i
12 | 13 14 15 16 17 18 19 20 21 22
|
j
1
I
!
i '
1 '
i
i '
j
i
>
'
23 24 25 26 27 28 29 30
! !
i !
: 'i
1 '
\ i
i l
1 ' j i
1 i
i
i
i ; i
',
i !
i
! t
' ' «
'
'
After-
noon
Mixing
Height (m)
31 32 33 34 35
; ;
I
i i
i i '
i , '
, : > •
i '
\
t
36 37 38 39 40 41 42 43 44 45 46 47 48 49 50
' , ;
I
!
! ; :
( i
• •
j
"-' i ' :
, __L i
i 1
,
i
;
1
! !
i t
1
1 1
.
.
j
FIGURE C-l
PREPROCESSOR INPUT DATA CODING FORMS
C-3
-------
MIXING HE
Upper Air
Station
Number
i
i
2345
I 1 i
: i
| ;
i , i
'
,
1
|
' '
1
1
:
i
2345
IGH1
i.
to
C
o
8 9
,
:
1
TA C/
j
>>
IO
Q
10 II
j
\RDS (Contir
Morning ;
Mixing
Height
(m)
j
lued)
12 13 14 15 o 17 18 19 20 21 22 23 24 25 26 27 28 29 30
1
!
;
i
1 '
i
i ;
1
\ \
\ !
i '
i
1
i
t
1
, i i
! ' .
i ' ' '
1 j 1
I ' 1
t i
1 :
\ , t
i ! -
1 i !
i ! i
, i i i
'
i ,
! ' i
' <
1
I
' i
1
; i
: , i !
( ' i
': ! i
|
j * '
I ' i
> '
|
1 ,
'
After-
noon
Mixing
Height(m)
31 32 33
34 35
| i
j ,
| '
I i
1
i
j I
' , i
, I
j '
i • !
i
i |
i
36 37 38 39 40 41
! ' ; i
i • '
i '
i
i
! | '
i !
1 ' <
| 1 , i
i ; : i
E ' . 1
1 ' 1
; i i i
• ! . I
! 1
1
42 43
I
j
I
44
45 46 47 48 49 50
i
! '
i
i
1 '
! i
i
1 i i
1
7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 4
42 43
44
i
| |
i
i I :
i >
i
i !
!
1
', J
i
,
1 ! ! p
! 1
| i i j
1 ' i
' 1
' ! I i
45 46 47 48 49 50
FIGURE C-l (CONTINUED)
PREPROCESSOR INPUT DATA CODING FORMS
C-4
------- An error occurred while trying to OCR this image.
-------
PLANT ELEVATION CARD (ft MSL)
RECEPTOR ELEVATION CARDS
E
C
•r
4-
t
Q
1
C
1
1
1
1
I
1
1
I
1
1
Z
2
2
Z
2
2
2
^
6-
2
i
3
J
J
3
J
3
5
j
j
L>
2
Elevation (ft MSL!
Ring 1
||
I
•?
4
SI
6
7
8
9!
o
1
Z
5
4
5
&
1
i •
; i
1
81
g
P
1
2.
•z
4
S
6
7
,
!
3' !
?'
9
!• ;
2: I t
i i
<}.
5
6!
RECEPTOR
2
3
4 5
E
= 11 ii i> ii
i •
: i .
' ''
i ' !
! '
i
i
' ]
i I
,
.
Ring 2
Ring
3
Ring 4
, i- •- - •= _j "i :" j :- \- •. .3 .= :j ]• : - -,4 -, -/, > * -t
.
i
i '
, i
|
• !
- :
1
1 !
!
' 1
i '
1 !
[ :
,
.
:
i
!
i
: l!
;
1
;
j ,
; j
,
i
l
i
i
i ! j
' i ' ;
1
.
1 :
•
! 4
j
I
1
1
1
|
J
!
!
LEVATION DELIMETER CARD (one blank card marks the end
"""T
7 3 9 10 It t; !„
;
'i
' '
' i '
, i <
1 i i
i
3f recept
.
.
r
.
<
i
Ring 5
M - J -
'
i
!
i
jr elevati
>
•
,
.
!
1
on iftfwt)
I
FIGURE C-3
SINGLE SOURCE (CRSTER) MODEL TERRAIN ELEVATION CODING FORM
C-6
-------
STACK IDENTIFICATION CARD
Stack id
STACK DATA CARDS
Emission Rate
(gm/sec)
—
> i
i i
;
i
! i
|
1 ;
.
9
.
•
•
4
.
.
.
•
,
.
.
.
,
.
t
,
Stack Height
Blank | (m)
1
Stack Diameter
(m)
Exit Velocity
(m/sec)
Temperature
_ - •_ j M _- i . j ' u ,; - > jj --
<—
*-
*-
—
•*—
\
!
'
Stack id
Stack
Stack id
Stack
Stack id
Stack
Stack id
Stack
Stack id
1
i
1
1
\lf
T
; i
'
' ', . ;
data
data
data
data
iii'
'
1
, 1
i ' !
• • 1
1 :
i : '
| 1 1
i ' '
: ! .'
1 , , ,
! i ' i
! i
• ' 1
i
1 i
i
.
.
*
.
.
•
.
.
I '
•
; j
.
,
, ,
,
* • i
•
(
'.
', ;
1 '
I 1 '
1
I
!
|
i ' i
;
i '
i
1 ;
I
, -
t
.
.
.
.
. '
.
.
i ;
i • I
.
§
' •
.
,
.
i
•
,
,
i
.
1 ' i i
i i j
,
, '
1 i
! ! • :
.
.
.
.
.
.
.
.
1 , i !
.
! ,
.
.
I ' '
.
.
.
.<
*
*
i
.
i
• 1
'
.
I
,
p
i :
.
.
.
.
*
.
4
.
.
.
.
.
,
*
.
•
.
.
*
STACK DELIMETER CARDS (2 blank cards mark the end of stack input)
1 ~ i « T
r
... |
"—— — ^* —
— — * ' "" '" '• u a ' ** ij " -' — _ ^ : -* -;
FIGURE C-4
SINGLE SOURCE (CRSTER) MODEL STACK DATA CODING FORM
C-7
-------
APPENDIX D
APPLICATIONS OF THE SINGLE SOURCE (CRSTER)
MODEL TO POWER PLANTS: A SUMMARY
by
J.A. Tikvart and C.E. Mears
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
from
Proceedings of the Conference On
Environmental Modeling and Simulation
EPA 600/9-76-016 July 1976
D-l
-------
APPLICATIONS OF THE SINGLE SOURCE (CRSTER) MODEL TO POWER PLANTS: A SUMMARY
Joseph A. Tikvart*
Connally E. Mears
Source Receptor Analysis Branch
Office of Air Quality Planning and Standards
U.S. Environmental Protection Agency
Research Triangle Park, N.C.
*0n Assignment from the National Oceanic
and Atmospheric Administration (NOAA)
For the last three years the Environmental Pro-
tection Agency has conducted a series of atmospheric
dispersion model studies of power plants. These
studies have considered the impact of approximately
700 utility power plants whose generating capacity is
25 megawatts or greater. Included in these studies
are (1) dispersion model estimates of SOj concentra-
tions downwind from each power plant, (2) validation
of the Sincfi-e Source Model with data for several typi-
cal power plants and (3) a sensitivity analysis of
this model. The results of these studies have been
used effectively in a number of energy/environmental
policy considerations. This paper summarizes the
findings of the various studies.
Introduction
Shortages in the availability of low-sulfur fossil
fuels have been given national prominence. These
shortages are particularly significant to utility
power plants for two reasons: (1) power plants typi-
cally use large quantities of fossil fuels and (2)
many of the State Implementation Plans (SIPs) require
severe reductions in sulfur dioxide emissions from
power plants which burn fossil fuels. The shortage of
low-sulfur fuel necessitates the elimination of unduly
stringent SIP control regulations, where this can be
done without endangering air quality standards. The
fuel shortage has also led to legislation which em-
powers the Federal Energy Administration to require
that specific power plants switch from oil or gas to
coal. This switch to coal, however, cannot be allowed
to result in a threat to air quality standards. Fur-
thermore, to meet the Clean Air Act requirement for
attainment and maintenance of acceptable air quality,
it may be necessary to revise the SIPs for selected
source categories, including power plants. The power
plant studies summarized in this paper support actions
like those noted above.
Estimates of the air quality impact caused by
.power plants are major components of these studies. A
dispersion model is a commonly used technique for re-
lating pollutant emissions to ambient air quality. It
is a mathematical description of pollutant transport,
dispersion,and transformation processes that occur in
the atmosphere. The Single Source (CRSTER) Model is
the primary dispersion model applied in all the power
plant studies discussed in this summary paper.
Due to severe time constraints and the fact that
models like the Single Source Model are widely applied
and considered state-of-the-art, the accuracy of this
model was not analyzed in the initial phase of the
power plant studies. However, some analyses of the
Single Source Model have been recently completed and
others are continuing. These include validation
studies, sensitivity analysis and model improvement.
Following sections of this paper discuss (1) the
Single Source Model, (2) power plant studies in which
it is applied, (3) evaluation of the model through
validation and a sensitivity analysis, and (4) appli-
cations to energy/environmental policy considerations.
Single Source (CRSTER) MODEL
The Single Source (CRSTER) Model is a Gaussian
plume model. It is based on the dispersion coeffi-
cients and equations described by Turner and on tfie
2
plume rise equations described by Briggs . The model
is essentially the same as that discussed by Hrenko
et al . It is designed to estimate concentrations
for averaging times of 1 hour, 24 hours,and 1 year
due to sources at a single location. The concentra-
tions are estimated for a circular array of receptor
sites which are located so as to approximate the
downwind distances at which the highest concentra-
tions are likely to occur.
The model estimates concentrations for each hour
of a year, based on wind direction (in increments of
10 degrees), wind speed, Pasquill stability class,
and mixing height. Meteorological surface data for
1964 are frequently used in the power plant studies,
although, with the proper data, any year could be
used. The reasons for the routine use of 1964 mete-
orological data are (1) data from earlier years do not
have an adequate resolution of wind direction, and
(2) data from subsequent years are not readily avail-
able on an hourly basis. Mixing height data are from
the upper air observations made at selected National
Weather Service stations. Hourly mixing heights are
estimated within the model by use of an objective
interpolation scheme. Decay of the pollutant between
source and receptor is ignored.
To simulate the effect of elevated terrain in
the vicinity of plant sites, a terrain adjustment
procedure is used. This procedure,decreases the
effective plume height by an amount equal to the
difference in elevation between the plant site and
the specific receptor site. The model then uses the
adjusted plume height in estimating concentrations at
that receptor. In those cases where terrain features
are found to be greater than the effective plume
height of the plant, the Single Source Model is not
apolied.
D-2
-------
Power Plant Studies
Purpose and Limitations
The power plant studies have considered the
impact of approximately 700 utility power plants whose
generating capacity is 25 megawatts or greater. The
studies may be divided into three parts. These are
analyses for (1) the feasibility of compliance exten-
sions in 51 selected Air Quality Control Regions
(AQCRs), (2) the feasibility of oil-to-coal conversions
at selected power plants and (3) the general impact of
power plants on ambient SO,, concentrations in 128
AQCRs. In all cases the studies are primarily con-
cerned with estimates of the maximum 24-hour concen-
trations of S0?. This averaging time and this
pollutant are the critical ones for which power plants
must meet primary National Ambient Air Quality
Standards (NAAQS). The second study is the only one
which considers particulate concentrations. Also, in
those cases where it is estimated that neighboring
power plants could contribute concentrations which add
to those caused by the plant under consideration, an
interaction analysis is performed.
All source data used in the power plant studies
are taken from the Federal Power Commission (FPC Form
67) for base years of 1971-,P.r 1972. In those cases
where emissions are projected to 1975, appropriate
4
data are taken from "Steam Electric Plant Factors" .
Emissions data are based on average monthly oper-
ations for each month of the year; such monthly data
are the limit of detail routinely available from the
FPC. A power plant could quite possibly operate at
near-maximum rated capacity for 24 hours, which
would not be apparent from the monthly data. If
these operations were coincident with days of poor
dispersion conditions, the estimated maximum concen-
trations could be significantly low. Thus, two sets
of emission conditions are routinely considered. One
is the nominal load case in which average hourly
emission rates are used; they are assumed to be con-
stant, except for variations by month. The other is
the maximum load case where emissions and plume rise
are based on the plant continuously operating at 95
percent of rated capacity. Both sets of emissions
data are considered and the one which results in the
highest estimated concentrations is used.
It should be noted that any use of these studies
must recognize the inherent limitations resulting from
the data and procedures used in the modeling effort.
Before final judgment on the control of specific
plants is made, other factors, not addressed in these
studies, should be considered. These include: the
impact of other sources in the area, projected growth
in the area, measured air quality data, known or sus-
pected downdraft or fumigation problems, unique nearby
terrain features, nearby land use patterns and popu-
lation distributions, more specific operational data
for the plant, impact of new units, specific meteoro-
logical studies for the area, and additional studies
or findings by other investigators.
Compliance Extension Studies
In 1972 a study by EPA on the aggregate demand
created by the SIPs for low-sulfur coal was conducted.
This study indicated a nationwide potential deficit of
about 100 million tons/year of such coal by 1975.
The deficit was considered most acute in 12 states
with high coal consumption rates. One means to alle-
viate the deficit would be to selectively reduce the
requirements for low-sulfur coal in those cases where
a higher sulfur coal could be used without endangering
the NAAQS.
An initial modeling study of SO. emissions in
several AQCRs had been conducted. This study showed
that some of the large power plants could be temporar-
ily allowed to burn coal at 1970 sulfur levels with-
out threatening the 24-hour NAAQS. Based on the
results of this study, it was decided to consider
selected power plants in 12 states which are heavily
dependent on coal. This involved a total of approxi-
mately 200 power plants in 51 AQCRs.
The study ' finds that at approximately 55 per-
cent of the plants considered, some relaxation of
emission limitations is possible. Relaxation could
result in increasing the average allowable percent
sulfur content of fuel from approximately 1 percent
sulfur content to 2 percent sulfur content at the
plants considered. Thus, the projected deficit in
low-sulfur coal could be eliminated.
Fuel Conversion Studies
The compliance extension studies discussed in
the preceding section had been conducted prior to the
overall oil shortage and energy crisis which became
apparent in late 1973. The oil shortage initiated a
second study of selected power plants on the U.S.
7 8
East Coast. In this second study'1 , fuel conversion
from oil to coal for selected boilers within specific
plants is analyzed to evaluate the impact on SO,, and
particulate concentrations. Increased SO^ emissions
due to fuel conversions at 16 of 43 plants considered
are estimated to result in concentrations from the
plants alone which exceed the 24-hour NAAQS. Seven
of the plant conversions are estimated to result in
concentrations from the plants alone which exceed the
24-hour particulate NAAQS. The analysis indicates
that in some cases partial conversion from oil to
coal at selected power plants appears to be a viable
option for alleviating the East Coast oil shortage.
Studies of Power Plants in 128 AQCRs
Further studies
9,10
of about 400 power plants dis-
tributed throughout the U.S. have been conducted in 1974
and 1975. The purpose is twofold: (1) to complete,
on a national basis, analyses of the threat of large
emitters of S02 to the NAAQS and (2) to add to the
overall analysis of the power plant industry being
conducted by governmental agencies and industry
itself. Thus, a base for further analyses is devel-
oped and is available if additional decisions must be
made concerning general EPA policy on compliance
extensions or fuel use options for power plants. Of
these 400 additional plants it is found that nearly
20 percent currently may exceed, by themselves, the
24-hour SO, air quality standards.
Evaluation of Model
Validation Studies
To determine the validity and overall accuracy
of the Single Source Model, validation studies have
been performed for the Canal, Paradise, Philo, Stuart
and Muskingum River power plants. The Canal Plant
is located in Massachusetts along Cape Cod Bay. The
1? 13
Paradise Plant ' is located in Western Kentucky.
The other three plants are located in Southern
D-3
-------
14 15
Ohio ' . In all cases, hourly variations in SO-
emissions are determined for each plant. These
emissions are then used with hourly meteorological
data which are representative of transport and dis-
persion in the vicinity of the plant. These data are
input to the model and 1-hour, 3-hour, 24-hour, and
annual concentration estimates are made for the sites
at which air quality monitors are located. The esti-
mated and the observed concentrations are then sub-
jected to several statistical comparisons. These in-
clude comparisons of highest and of second-highest con-
centrations and comparisons of observed and estimated
concentration frequency distributions.
As shown in Table 1, the model generally tends to
underestimate the highest and the second-highest 24-
hour average concentrations. This is also true for
3-hour average concentrations. However, 1-hour
averages are equally divided between overestimates and
underestimates. In cases where surrounding terrain is
nearly as high as the stack top (see the Philo Plant
in Table 1), the model overestimates concentrations
for all averaging times. It should be noted that most
dispersion models comparable to the Single Source Model
are not truly applicable in the vicinity of such sig-
nificant terrain features.
Table 1. Comparison of Observed and Estimated
Concentrations
1 -Hour Average Concentrations
Plant
Cinal
Stuart
Sampling
Station
1
2
3
4
1
2C
3
4
5
6
7C
Hisklngum 1
River 2
Philo
'Obser
6Est1m
3
4
1
2
3
4
'5
6
2nd
Oa
435
553
446
575
685
685
1022
750
495
980
325
857
786
996
735
525
735
745
665
575
565
Highest
Eb
253
174
446
427
1372
814
565
515
823
595
976
980
1304
873
465
1295
945
4049
1945
1279
2359
ved concentrations with
ated concentration.
HK
0
438
618
732
638
857
1014
1153
883
565
1053
435
925
786
1179
786
893
891
917
695
675
595
jhest
E
283
179
509
479
1393
948
1022
541
1219
693
1000
1083
1310
933
645
1639
1059
4593
1981
1344
2482
subtracted backgr
24-Hour Average Concentrations
2nd Hi.
0
66
36
77
63
259
63
181
79
63
147
69
133
131
165
109
132
67
127
62
87
121
ound.
jhest
E
16
9
38
4
149
75
91
45
57
69
73
81
82
73
45
133
86
471
165
222
282
Highest
0
75
46
83
75
277
159
225
83
77
195
77
170
137
227
115
133
110
132
158
94
138
I
29
11
39
16
161
98
102
49
75
83
120
97
91
74
47
147
104
541
220
226
356
cSamplers were 1n operation for less than half the year.
In the comparison of observed and estimated fre-
quency distributions, disparate results are found.
There is considerable variation in comparisons from
site-to-site and plant-to-plant. However, agreement
improves for frequency distributions which include all
monitoring sites around a particular plant. As shown
in Figure 1, all but the few highest and lowest con-
centration percent!les are accurately estimated for
the distributions which include all sites.
Until further studies become available, it may
be concluded from these validation studies that the
Single Source Model is generally accurate within a
factor of two. This is not surprising since this
accuracy is widely accepted for such point source
models. However, an important element is identifica-
tion of the tendency to underestimate, rather than
overestimate, concentrations for averaging times
associated with NAAQS. This tendency undercuts the
position of those who contend that sueh models are
overly conservative when used in determining emission
control requirements. It also places an added burden
on pollution control officials to ensure that an envi-
ronmental threat is not understated.
PERCENTAGE OF CONCENTRATIONS
GREATER THAN INDICATED VALUE
J. M. STUART PI.RMT
O'i's^RIBUTlSN l^Cfl"^ HOUR
SG2 CCNCENfRflTICNS flV flLL
eieflsu*EC
LESS THAN INDICATED VALUE
Figure 1. Stuart Plant Cumulative Frequency Distr-
bution for 24-Hour SO- Concentrations at All Stations.
Sensitivity Analysis
To further understanding of the behavior of the
Single Source Model, a sensitivity analysis has
been conducted. Specifically,this analysis examines
the impact of variations or errors in the input data
on the concentration estimates produced by the model.
Thus, it identifies the model parameters which have
the greatest influence on concentration estimates.
In the analysis the incremental change in pre-
dicted concentration is determined for an incremental
change in input. A case study approach is used with
the three Ohio power plants noted above. The analysis
is limited to the maximum estimated 24-hour concentra-
tion, since this is generally considered to be the
most important averaging time for power plants with
regard to primary air quality standards.
Both source parameters and meteorological param-
eters are considered. The source parameters are (1)
stack height, diameter, gas exit velocity, and gas
exit temperature, (2) emission rate and its monthly
variation and (3) terrain adjustment. The meteorolog-
ical parameters considered are mixing height, wind
speed, ambient temperature and stability class. With
the exception of stability class, each parameter is
varied by a factor of +_ 5, +_ 10, and +_ 25 percent
while all other parameters are held constant.
From the analysis summarized in Tables 2 and 3,
it is found that for sources with relatively short
stacks, for example the Philo Plant which has stacks
about 300 feet high, a percent change in any stack
parameter results in at least that percent change in
the maximum 24-hour concentration. For sources with
relatively tall stacks, for example the Stuart Plant
which has stacks about 800 feet high, a lack of such
sensitivity is found. Stability class, a meteorolog-
ical parameter, is found to be a highly sensitive
D-4
-------
factor for all plants, since this parameter can take
on only six discrete values. The importance of
parameters such as wind speed and mixing height varies
depending on the meteorological conditions that result
in highest concentrations for a plant. In all cases,
the percent change in the maximum 24-hour concentra-
tion is less than the percent change in these meteoro-
logical parameters. Tables 2 and 3 indicate percent
changes in maximum 24-hour concentrations for positive
variations in source and meteorological parameters.
Comparable changes in concentration can also be shown
for negative variations in these parameters.
Table 2. Percentage Change From Base Case—Maximum
24-Hour Concentrations Due to Variations in Source
Related Parameters.
>4n Parameter
Paraneter \^
Stack hclnht (m)
Stack tenp (°C)
Exit velocity(m/s)
Stock diameter(m)
Terrain AClJ !m)
En>lssions{;m/sec)
Mil
v
t 5
-2
-4
-5
-11
1
5
M n fj ui
River
+10
-5
-8
-9
,7
3
"
+25
• 11
-17
-19
-30
12
25
* 5
-6
-4
-6
-11
5
5
Philo
+ 10
-12
- 8
-10
-20
9
10
+25
-27
-18
-23
-43
24
25
t 5
-2
-2
-2
-3
1
5
Sli.sr
•HO
-5
-4
-3
-6
1
10
t
+ 25
-11
- 7
-7
-15
3
25
Table 3. Percentage Change From Base Case— Maximum
24-Hour Concentrations Due to Variations in Meteoro-
logical Parameters.
\l>ercent Vana-
^^tlon In
^s^arameter
Parameter ^\^
Mixing heloht (m)
Uind speed (m/s)
Ambient temp (°C)
Stability class*
itellnijui
River
+ 5
n
3
1
-
+10
0
5
2
+25
0
9
6
-43
Philo
t 5
0
4
1
-
+10
0
7
2
-
+25
0
21
6
-4B
Stuai t
+ 5
-3
-2
1
-
+10
-5
-3
2
+25
-11
-9
5
-47
"biased by +1 Stability Class
The sensitivity of the maximum estimated concen-
trations to changes in meteorological data sets is
also determined. Three data sets are used with each
set of source data. Changes in maximum concentration
from the base case which are shown in Table 4, range
from an increase of nearly 50 percent to a decrease of
almost 30 percent. Inherent in the change of maximum
concentration are the effects of the wind direction
and the variability of wind direction. These are not
considered individually in the sensitivity analysis.
However, wind direction and its variability, which are
a function of the meteorological conditions peculiar
to each data set, play a major role in the percent con-
centration changes shown in Table 4. This illustrates
the importance of a meteorological data set which is
as representative of transport and dispersion in the
vicinity of the plant as possible.
As a result of this analysis it can be concluded
that: (1) the sensitivity of model estimates to accu-
racy in the input parameters varies from source to
source; (2) accuracy in the source parameters
becomes more critical as the stack becomes shorter;
(3) errors in individual meteorological parameters,
with the exception of stability class, result in some-
what smaller errors in estimated concentrations; (4)
the cumulative errors in meteorological parameters,
which result from the use of data from an unrepresent-
ative site, can cause substantial errors in estimated
concentrations.
Table 4. Percentage Change From Base Case—Maximum
24-Hour Concentrations Due to Variations in the
Meteorological Data Sets.
Surface/iliver ni r
Data <,ct
Huntington/.'luntlntjtrDn
Coluiihm/n.iyton
Cincinnati/Dayton
Musk inguin
Rwer
-
47. B
11.6
Plllo
-28.4
-
-5.8
Stuart
-19.4
36.0
-
Model Improvement
As a result of the model validation and the sen-
sitivity analysis, studies to improve the Single
Source Model are being undertaken. Two specific areas
under investigation are (1) the use of other stability
classification and dispersion parameters which may
allow better estimates of plume dilution and (2) the
use of more precise information on the stack param-
eters which affect plume rise. Also, additional
analyses are being undertaken to evaluate the accu-
racy of hourly concentration estimates for various
meteorological regimes. The goal is to assess the
need for better data inputs or more precise algorithms
in the model. Based on these studies, improvements
in the model will be considered.
Applications of Power Plant Studies
Limitations on the model and its application in
the power plant studies have been noted. Even with
these limitations, the power plant studies are of
value for use in generalized analyses which assess
the overall effect of some plan of action for the
utility industry. These studies have been used effec-
tively in a number of energy/environmental policy con-
siderations.
The Clean Fuels Policy is an EPA program to
encourage some states to eliminate unnecessarily
stringent control regulations in their SIPs and there-
by alleviate the shortage of low sulfur coal. The
power plant studies demonstrated the potential use-
fulness of such a policy and helped to indicate those
SIPs where unnecessarily stringent regulations might
exist.
The power plant studies were used in early analy-
ses of proposed oil-to-coal conversions. They were
useful in indicating the types of sources which were
good candidates for conversion and specifically indi-
cated several plants that were poor candidates.
These studies have been used for roughly assessing
the allowable percent sulfur coal which could be used
in oil-to-coal conversions required under the Energy
Supply and Environmental Coordination Act. They will
serve as a basis for more detailed subsequent analy-
ses.
In the development of EPA policy on tall stacks
and meteorological control systems, the power plant
studies were used frequently. They were used to
analyze alternatives for limitations on stack height
D-5
-------
Increases. They allowed the frequency and amount of
emission reductions that would be required by meteoro-
logical control systems to be compared, for various
categories of power plants, to permanent control re-
quirements.
The power plant studies have been the basis for
analyses in support of a viable S0? control strategy
for Ohio. They were used as justification for exist-
ing regulations in the 1974 Ohio S02 hearings. They
were used as an initial base in developing EPA Region
V's current proposed regulations for Ohio . They
have also been used by Region IV in the development
and revision of SIPs applicable to power plants lo-
cated in the Southeastern United States.
Industry has used the power plant studies in
1R
statements to the U.S. Congress on options for con-
trol of SO,. These studies have also been used in
evaluating the impact of proposed legislation to pre-
vent significant deterioration of air quality.
Based on the demand for the reports resulting
from such power plant studies, it is logical to con-
clude that other regulatory agencies and industrial
groups are using these studies. In most cases, they
are being extended by more detailed analyses. It
appears that these studies will continue to play an
important role in the development of regional and
national environmental policies which affect utility
power plants.
Acknowledgments
The authors wish to recognize the major contri-
butions of their co-workers to these power plant
studies. Major contributions were made by D. Barrett,
W. Freas and R. Lee under the overall direction of
H. Slater. Special recognition is also due to those
individuals who performed the bulk of the work under
contract to EPA. These include: P. Morgenstern and
L. Morgenstern of Walden Research Division of Abcor,
Inc.; R. Koch of GEOMET, Inc.; and M. Mills and
R. Stern of GCA Corporation. Thanks are also due to
Mrs. B. Stroud who diligently prepared this manu-
script.
References
6. Morgenstern, P., et al , "Modeling Analysis of
Power Plants for Compliance Extensions in 51 Air
Quality Control Regions," J. Air Poll . Control
Assn., Vol. 25, No. 3, '
1. Turner, O.B., "Workbook of Atmospheric Dispersion
Estimates." Office of Air Programs Publication
No. AP-26. Superintendent of Documents, Govern-
ment Printing Office, Washington, D.C., 1970.
2. Briggs, G.A., Plume Rise, U.S. Atomic Energy
Commission, Division of Technical Information,
Oak Ridge, Tennessee, 1969.
3. Hrenko, J., D.B. Turner, and J. Zimmerman,
"Interim User's Guide to a Computation Technique
to Estimate Maximum 24-Hour Concentrations from
Single Sources," Meteorology Laboratory, Environ-
mental Protection Agency, Research Triangle Park,
N.C., 1972 (Unpublished Manuscript).
4. National Coal Association, "Steam Electric
Factors," Washington, D.C., 1973.
5. Morgenstern, P., "Summary Report on Modeling
Analysis of Power Plants for Compliance Exten-
sions in 51 Air Quality Control Regions." Publi-
cation No. EPA-450/3-75-060. Prepared by Walden
Research Division of Abcor, Inc., under Contract
No. 68-02-0049. Environmental Protection Agency,
Research Triangle Park, N.C., 1973.
7. Morgenstern, L., "Summary Report on Modeling
Analysis of Power Plants for Fuel Conversion."
Publication No. EPA-450/3-75-064. Prepared by
Walden Research Division of Abcor, Inc. under
Contract No. 68-02-1377. Environmental Protection
Agency, Research Triangle Park, N.C., 1975.
8. Morgenstern, L., et al, "Air Quality Modeling
Analysis of Power Plants for Fuel Conversion."
APCA Paper No. 75-33.6, Boston, Mass., 1975.
9. Morgenstern, L., "Summary Report on Modeling
Analysis of Selected Power Plants in 128 AQCRs
for Evaluation of Impact on Ambient SO. Concen-
trations, Volume I". Publication No. EPA-450/3-
75-062. Prepared by Walden Research Division of
Abcor, Inc., under Contract No. 68-02-1484.
Environmental Protection Agency, Research
Triangle Park, N.C., 1975.
10. Koch, R., "Summary Report on Modeling Analysis of
Selected Power Plants in 128 AQCRs for Evaluation
of Impact on Ambient SOp Concentrations, Volume
II-" Publication No EPA-450/3-75-063. Prepared
by GEOMET, Inc., under Contract No. 68-02-1483.
Environmental Protection Agency, Research Triangle
Park, N.C., 1975.
11. Mills, M., "Comprehensive Analysis of Time--
Concentration Relationships and the Validation of
a Single Source Dispersion Model." Publication
No. EPA-450/3-75-083. Prepared by GCA Corporation
under Contract No. 68-02-1376. Environmental
Protection Agency, Research Triangle Park, N.C.,
1975.
12. Klug, W., "Dispersion from Tall Stacks."
Publication No. EPA-600/4-75-006. Environmental
Protection Agency, Washington, D.C., 1975.
13. Enviroplan, Inc., "A Comparison of Predicted
and Measured Sulfur Dioxide Concentrations at
the Paradise Power Plant in 1969." Draft Report
No. 1, prepared under Contract No. 68-01-1913.
Environmental Protection Agency, Washington,
D.C., 1975.
14. Mills, M., and R. Stern, "Model Validation and
Time—Concentration Analysis .of Three Power
Plants." Final Report prepared by GCA Cor-
poration under Contrary No. 68-02-1376, Environ-
mental Protection Agency, Research Triangle
Park, N.C., 1975.
15. Lee, R., M. Mills, and R. Stern, "Validation
of a Single Source Model." Paper presented at
the 6th NATO/CCMS International Technical
Meeting on Air Pollution Modeling, Frankfurt/
Main, Germany, FR, September, 1975.
16. Freas, W. , "Sensitivity Analysis of the Single
Source Model." Office of Air Quality Planning
and Standards, Environmental Protection Agency,
Research Triangle Park, N.C., 1976 (Unpublished
Manuscript) .
17. Environmental Protection Agency, "Technical
Support Document: Development of a Sulfur
Dioxide Control Strategy for the State of Ohio,
Volume 1." Chicago, Illinois, September, 1975.
18. Environmental Research and Technology, "An
Evaluation of Sulfur Dioxide Control Require-
ments for Electric Power Plants." Report pre-
pared for Edison Electric Institute, New York,
N.Y., April, 1975.
D-6
-------
APPENDIX E
SENSITIVITY ANALYSIS OF THE SINGLE SOURCE (CRSTER) MODEL
by
Warren P. Freas and Russell F. Lee*
U.S. Environmental Protection Agency
Presented At
Seventh International Technical Meeting On Air Pollution
Modeling and Its Application, September 1976
*0n assignment from National Oceanic and Atmospheric Administration,
U.S. Department of Commerce.
E-l
-------
SENSITIVITY ANALYSIS OF THE SINGLE SOURCE MODEL
Introduction
The purpose of this paper is to describe a sensitivity analysis
of the Single Source (CRSTER) Model, a model which has been used
extensively by EPA to estimate the air quality impact of fossil
fueled steam-electric power plants and selected industrial emission
sources.
The question of model "validity" or "reasonableness" is
addressed in some recent validation studies of the model for selected
2 3
power plants. ' A sensitivity analysis of the model provides the
natural complement to these validation studies. Whereas a validation
study assesses the impact of model formulation on calculated concen-
trations, a sensitivity analysis takes the model formulation as
given and examines the impact of errors in the input data on the
model calculated concentrations. The objective of a sensitivity
analysis, therefore, is to identify the critical model input para-
meters—those variables which have the greatest influence on model
predictions. Knowing the critical model parameters assists the
model user in assigning proper priorities in the data collection and
quality assurance procedures and in understanding the complex inter-
actions within the model.
In order to complement the model validation studies, a case
study approach is selected which models the same three Ohio power.
plants, Muskingum River, Philo, and J. M. Stuart.
E-2
-------
Model Description
The dispersion model is a Gaussian plume model which calculates
hourly and daily SOp concentrations for an array of 180 receptor
locations, calculates maximum daily SCU concentrations for a year,
and identifies the meteorological conditions associated with these
maxima.
A ground plane displacement procedure is used in the model to
simulate the effect of elevated terrain in the vicinity of the plant
sites. This procedure consists of adjusting (decreasing) the height
of each plume by an amount equal to the difference in elevation between
the plant site and the average surrounding terrain.
An annual average S02 source strength and monthly variation factors
must be specified.
Meteorological inputs to the model consist of hourly surface
observations of wind speed, wind direction sector, temperature, total
cloud cover, and twice daily mixing heights. The data are input into
a preprocessor program which in turn writes a tape containing hourly
values of stability index, mixing height, temperature, wind speed,
flow vector (wind direction plus 180°), and randomized flow vector.
The randomized flow vector is equal to the flow vector minus four
degrees plus a random number between 0 and 9 degrees. The preprocessor
output tape is read by the Single Source Model which performs the
actual concentration calculations.
E-3
-------
Plant Descriptions
The physical characteristics of each plant listed in Table
1 are those reported in the 1973 Federal Power Commission (FPC)
Form 67, "Steam-Electric Plant Air and Water Quality Control Data
for the year ended December 31, 1973."
Some of the stack heights listed below for the Philo and Muskingum
3
plants differ from those reported in the referenced validation study.
It has been determined that modifications/replacement of some of the
stacks at these plants are not reported in the 1973 FPC data. The
stack heights listed in the validation study represent the current
plant configuration (these values were obtained directly from the
utility), whereas the stack heights in the present analysis reflect
plant configuration prior to 1973.
Source Data
The source strength and monthly variation factors are calculated
from data contained in the 1973 FPC Form 67. The data used in the
calculations consists of the monthly fuel consumption and average
sulfur content data. The values for the source strengths and monthly
variation factors are listed for each plant in Table 2.
The adjustments for elevated terrain in the vicinity of each
4
plant are those used in previous modeling studies of these plants.
The terrain adjustments, which represent the difference in elevation
between the plant sites and the average of the surrounding terrain,
are forty-nine meters for Muskingum River, thirty meters for Philo,
and sixty-seven meters for Stuart.
E-4
-------
TABLE 1. PLANT CHARACTERISTICS
Characteristic
Stack Height (m)
Diameter (IT)
Velocity (m/s)
Temperature (°K)
Number of Boilers per stack
Plant
MusUnnum River
Stack Stack Stack Stack
1235
91 91 131 251
4.5 4.5 4.8 6.7
13.4 13.4 22.3 16.8
430 430 416 428
1121
Phllo
Stack Stack Stack
456
46 46 55
5.7 4.5 2.6
6.3 10.1 28.9
458 458 430
2 2 1
Stuart
Stack
1-3
244
6.0
23.2
375
1
Source: 1973 FPC Form 67
TABLE 2. PLANT EMISSIONS DATA
Source
Parameter
Source
Strength
(9 /s)
Monthly
factors:
Jan. -June
July-Dec.
PLAM
Musklngum River
Stack Stack Stack Stack
1235
1446 1414 2952 4025
All Stacks
1.04 .99 .92 .89 1.04 .96
1.06 1.03 1.13 .74 1.19 1.04
Pnflo
Stack Stack Stack
4 5 6
586 566 601
All Stacks
.92 1.09 .86 .87 1.06 1.08
.97 1.02 .98 1.08 1.10 .96
Stuart
Stack Stack Stack
1 2 3
1381 1261 1559
All Stacks
.92 .95 .79 .79 .79 1.07
1.11 1.17 1.06 1.17 1.10 1.12
Source: 1973 FPC Form 67
E-5
-------
Meteorological Data
The sources for the meteorological data used in this analysis are
listed below.
Table 3. METEOROLOGICAL STATIONS
Plant
Muskingum
Philo
J. M. Stuart
Surface Observations
Airport
Huntington, W. Va.
Columbus, Ohio
Cincinnati , Ohio
Upper Air Data
Airport
Huntington, W. Va.
Dayton, Ohio
Dayton, Ohio
Year
1964
1964
1964
Methodology
Model sensitivity is defined as the partial derivative of the
calculated concentration with respect to the parameter, p, evaluated
at some value for p. That is,
9P
Po
For the purposes of numerical simulations, model sensitivity may be
defined as "the incremental change in the calculated concentration
resulting from the incremental change in input." These two defini-
tions suggest the general approaches available for conducting a
sensitivity analysis, analytical and numerical. Either method may
be used to investigate both aspects of model sensitivity: the
E-6
-------
sensitivity of model predictions to (1) variations in the magnitude
of each of the input variables and (2) the collective variations in
all the input variables.
Because of the complex interactions of the meteorological and
source variables found in this particular model, it is not possible,
nor desirable, to examine the model sensitivities analytically.
Analyses of this type have been performed for other Gaussian models,
however. ' ' This analysis concentrates on the first aspect of
model sensitivity; that is, variations in the magnitude of the input
variables related to both the source and the meteorology.
Source Related Parameters
The source related parameters examined in this analysis include:
1. Stack height (m)
2. Stack gas temperature (°C)
3. Stack gas exit velocity (m/s)
4. Stack diameter (m)
5. Emission rate (g/s)
6. Terrain adjustment (m)
7. Monthly variation factors
For the first six source-related parameters, systematic changes
are introduced in each of the individual model parameters. Conceptually,
this approach can be viewed as simulating events such as consistent
instrument or technician bias. The actual numerical procedure consists
E-7
-------
of varying each of the initial source parameter values by factors of
+5, +10 and +25 percent of the initial value while holding all the
other parameter values constant. For the seventh parameter, the
monthly variation factors, only the effect of no monthly variation
(i.e. a constant emission rate for the year) is simulated. Inter-
actions from changes in combinations of these parameters are not examined.
In the analysis of the source parameters, computational savings are
achieved by limiting the model calculations to the meteorology for
the "worst 20 days" of the year.* These days are chosen on the basis
of the calculated daily maximum 24-hour average concentrations from
the "base case" simulations.
Meteorological Parameters
The meteorological parameters included in the analysis are:
1. mixing heights (m)
2. wind speeds (m/s)
3. ambient temperature (°C)
4. stability class
5. randomized flow vector
6. surface/upper air data sets
*The phrase "worst 20 days" is used throughout this report to denote
those days used in the source parameter analysis. The distribution
of daily maximum 24-hour concentrations is examined for a "clean
break" occurring near the top 20 days. The actual number of days
selected varies from 22 days at Philo to 26 days at Stuart.
E-8
-------
The analysis methodology for the first four meteorological parameters
parallels that used for the source parameters. Systematic variations
introduced in each of the meteorological parameters simulate the effect
of errors resulting from factors such as observer and instrument bias
or consistent trends in the meteorological variables. These variations
are applied to each parameter individually; the effect of interaction
is not examined.
In contrast to the analysis of the source parameters which uses
only the meteorology from the "worst 20 days" in the simulations, the
simulation procedure for the meteorological variables uses all 366
days of the year. This latter procedure eliminates the possibility of
having the actual model calculated maximum concentration not fall on
one of the "worst 20 days" due to the variations introduced in the
meteorology related variables. The values of the parameters, wind
speed, mixing height, and ambient temperature are biased by factors
of +5, +10, and +25 percent. Stability is biased by +1 class, with
the added restriction that stability class cannot be less than 1 or
greater than 7.
The simulation procedures developed for the randomized flow
vector include: (1) substituting a new random number generator
(Univac's RANDU) and nine different starting numbers, and (2) eliminating
the random number generator, i.e. using the average flow vector.
E-9
-------
In an effort to assess the effect of changes in all of the
meteorological variables, simulations are conducted for each plant
using the surface/upper air data sets from the other two plants. The
advantages of this ad hoc procedure over explicit numerical methods
are (1) ease of operation, (2) minimal additional computer requirements,
and (3) and identification of the effects of using different upper air
and surface wind stations. With this procedure, however, it is not
possible to explicitly identify the range of variation in each of the
input variables.
Results
The results of the sensitivity analysis for the Single Source
Model for the three Ohio plants are presented in tables, input/output
response graphs, and cumulative frequency distribution plots. Primary
emphasis is placed on the maximum 24-hour concentration of SO- in the
graphical presentation of the results. This averaging time is the
critical one for which power plants must meet primary National Ambient
Air Quality Standards (NAAQS). The effects on the 1-hour and annual
average concentrations are noted in the tables, however.
To provide a basis for comparison, the results from the base case
simulations are summarized in Figure 1, the cumulative frequency dis-
tributions for the model calculated maximum 24-hour concentrations. As
modeled, the Muskingum River plant has two short 91 meter stacks, one 131
meter stack, and one 251 meter stack. Unstable conditions are associated
with the base case maximum 24-hour concentration at this plant. The
E-10
-------
three short stacks and elevated terrain considerations at the Philo
plant yield a base case maximum 24-hour concentration which is
associated with near neutral stability and persistent wind direction,
about 11 hours with the winds within +10 degrees of direction. At
Stuart, a plant with three tall stacks (244 meters), the base case
maximum 24-hour concentration is associated with unstable, light wind
conditions with some limited mixing.
Source Related Parameters
The results of the sensitivity analysis for the source related
parameters are summarized for all three plants in Tables 4 and 5 and
input/output response curves, Figures 2 and 3. Apparent discontinuities
in these response curves arise whenever, as the parameter values vary
from their initial levels, the day associated with the maximum 24-hour
concentration changes. Naturally, the meteorological conditions for
this new maximum day differ somewhat from the original maximum day, thus
producing a slightly different response.
An examination of the tables and graphs reveals, that for the
Muskingum River and Philo plants, the maximum 24-hour concentrations
are quite sensitive to changes in the values of the stack parameters,
i.e. a unit change in the value of an individual stack parameter yields
at least a unit change in the maximum 24-hour concentration. At the
Stuart plant, however, a unit change in the value of any of the stack
parameters yields less than a unit change in the maximum 24-hour
concentration.
E-ll
-------
TABLE * PERCENT/T.E CHANf.E FPO" F«SE CASE "W-lIM 24-HOUR
C01CENTPATIPNS PUE TC VARIATIOSS IN
SOl'RCE RfLATED PARAMETER1;
N. Percent
N. error
^v in
N. value
Parameter < >,
Stack heicht (n)
Stack tenp ("C)
Exit velocity
(i»/s)
Stack dianeter(n)
Terrain AOJ (n-)
Emissions (o/s)
Muskinoun
River
5
-2
-4
-5
-9
1
5
10
-5
-8
-9
-17
3
10
25
-11
-17
-19
-30
12
25
PMlo
5
-6
-4
-6
-11
5
5
in
-12
-8
-10
-20
9
in
25
-27
-18
-23
-41
24
25
Stuart
5
-2
-2
-2
-3
1
5
10
-5
-4
-3
-7
1
10
25
-11
-7
-7
-15
3
25
>v Percent
N. error
\1n
^s^al ue
Parameter ^x^
Stack heiqht (m)
Stack temp (°C)
Exit velocity
(m/s)
Stack dtameter(m]
Terrain «OJ (n)
Emissions (|H5 H.lghtl
Ho Oin^o^lled
flow Vector
HPW ffj|n>1r)nl
tiii^pf fcnernter
(WIOU)
Hew Surftce/
»rr!
Annual
Mean
-3.6
4.4
O.I
-1.3
-33.2
"at
24-hr
•6.5
-2.4
-3.1
-13.0
36.0
Mai
1-hr
-10.0
0.0
0.4
-6.8
-14. (
E-12
-------
The data reveal that, for variations in stack diameter, the
percentage change in the maximum 24-hour concentration is about
double that found for exit gas velocity. The relationship between
exit velocity and stack diameter is found in the model expression
for volumetric flow rate:
VF = ocVsD2
Where a is a constant term, V is the stack gas exit velocity, and
D is the stack diameter. Given this relationship and remembering the
analytical definition of model sensitivity, i.e., the partial derivative
of concentration with respect to the given parameter, one expects the
sensitivity of the model calculated concentrations to variations in
stack diameters to be twice that of stack gas exit velocity.
Except for the Philo plant, maximum 24-hour concentrations are
not sensitive to variations in the value of the terrain adjustment,
because the amount of the adjustment is quite small relative to the
height of the stacks.
From the formulation of the model, we know that a unit change
in emissions results in a unit change in the model calculated con-
centrations. Also, maximum concentrations occur on the same day as
the base case maximum concentrations, only the concentration changes.
The seasonal factors simulate the effect of monthly variations
in the emissions rate resulting from either differing load factors
or varying sulfur content of the fuel. The effect of changing these
factors, therefore, is to change the yearly distribution of emissions.
E-13
-------
The effect of uniform emissions throughout the year, i.e. no seasonal
variation, is shown in Table 5.
Meteorological Parameters
Tables 5 through 9 and Figures 4 through 8 summarize the sensitivity
analysis results for the meteorological parameters.
The maximum 24-hour concentrations are very insensitive to
variations in mixing heights at the Muskingum River and Philo plants.
Using urban mixing heights rather than the base case rural mixing
heights, the calculated maximum 24-hour concentration is unchanged
at Philo, decreases by 2 percent at Stuart, and increases by 11 percent
at Muskingum River as compared with the base case (Table 5).
At the Muskingum River and Philo plants, increasing the average
wind speed produces higher 24-hour concentrations. However, at the
Stuart plant which has three tall stacks, increasing (decreasing)
the average wind speed decreases (increases) the maximum 24-hour
concentration, i.e., the reverse of the relationship at the other
two plants (Figure 5). The maximum concentration from a point source
is affected by the wind speed in two ways, higher wind speeds
dilute the plume and also reduce the plume rise, reducing the dis-
tance to the highest concentration. These two effects combine to
produce the higher concentration at some critical wind speed. Thus,
the maximum concentrations may either increase or decrease when wind
speed increases.
E-14
------- An error occurred while trying to OCR this image.
-------
Because the ambient temperatures are significantly lower than the
stack gas temperatures, for a fixed change in ambient temperature
there is small percentage change in plume buoyancy and thus a small
percentage change in plume rise. Understandably, for the range of
values considered in the analysis, the calculated maximum 24-hour
concentrations are insensitive to variations in ambient temperature.
Lowering the stabilities by one class (one step more unstable)
yields higher concentrations and raising stability by one class
lowers concentrations. As Table 6 illustrates, the 24-hour concen-
trations are very sensitive to a bias of one stability class.
Using a different randum number generator (RANDU) and nine
different starting numbers, the greatest change from the base case
maximum 24-hour concentration is +14% at Muskingum River, +0.8% at
Philo, and -13% at Stuart (Table 5).
When the average flow vector is used in place of the randomized
flow vector, the model calculated maximum 24-hour concentrations are
7% higher at Muskingum River, 10% higher at Philo, and 3% lower at
Stuart (Table 5). Thus, the inclusion of the randomized flow vector
in the model can either decrease or increase the calculated 24-hour
concentrations. For example, when the wind blows directly toward
the receptor, concentrations will always decrease when the randomized
flow vector is added. However, if the wind blows 10° to the left of
the receptor, the addition of the randomized flow vector can either
decrease or increase the concentrations.
E-16
-------
The sensitivity of the model calculated concentrations to the
selection of individual surface/upper air data sets is determined by
running the model for each plant with the meteorological data sets
from the other two plants. Because two of these data sets share a
common upper air station, it is also possible to factor out the effect
of different surface wind stations on calculated concentrations.
The surface/upper air stations associated with each plant are
Huntington/Huntington (HTG/HTG) at Muskingum River, Columbus/Dayton
(CM/DAY) at Philo, and Cincinnati/Dayton (CVG/DAY) at Stuart.
Cumulative frequency distributions of the daily maximum 24-hour
concentrations for all three data sets at each plant are presented in
Figures 6 through 8. The shapes of all the distributions are quite
similar except for some divergence at the upper percentile values.
Table 10 ranks the model calculated concentrations for the three
averaging times from each of the meteorological data set simulations.
Table 10. MODEL CALCULATED CONCENTRATION RANKINGS
Surface/
Upper Air
HTG/HTG
CMH/Day
CVG/Day
Muskingum River
Annual Max. Max.
Avg. 24-hr. 1-hr.
232
3 1 3
1 2 1
Philo
Annual Max. Max.
Avg. 24-hr. 1-hr.
1 3 2
3 1 3
2 2 1
Stuart
Annual Max. Max.
Avg. 24-hr. 1-hr.
233
3 1 1
1 2 2
E-17
-------
For all three plants, the highest 24-hour concentration and the
lowest annual average are associated with the CMH/DAY data sets.
Comparing the results from the CMH/DAY and CVG/DAY data sets indi-
cates that the Columbus surface wind data always give the maximum 24-
hour concentration and the lowest annual average concentration. The
maximum 24-hour concentration is a result of a single day with extremely
strong wind persistence, about 11 hours with the winds within +10 degrees
of direction.
A further examination of the concentration rankings reveals that
the HTG/HTG data always give the lowest value for the maximum 24-hour
concentration. This appears to be due to a greater degree of fluctuations
of the wind direction from hour to hour in the Huntington data.
Conclusions
As a result of this analysis it can be concluded that: (1) the
sensitivity of model estimates to errors in the input parameters
varies from source to source; (2) errors in the source parameters
becomes more critical as the stack becomes shorter; (3) errors in
individual meteorological parameters, with the exception of stability
class, result in somewhat smaller errors in estimated concentrations
than errors in source parameters; and (4) the cumulative errors in
meteorological parameters which result from the use of data from an
unrepresentative site, can cause substantial errors in estimated
concentrations.
E-18
-------
REFERENCES
1. Hrenko, J., D. B. Turner and J. Zimmerman. "Interim User's
Guide to a Computation Technique to Estimate Maximum 24-hour
Concentrations from Single-Sources," unpublished manuscript,
U. S. Environmental Protection Agency (1972).
2. Mills, M. T. and F. A. Record, "Comprehensive Analysis of
Time-Concentration Relationships and the Validation of a Single
Source Dispersion Model," Publication No. EPA 450/3-75-083,
prepared by GCA/Technology Division under Contract No. 68-02-1376,
Environmental Protection Agency, Research Triangle Park, N.C.
(1975).
3. Mills, M. T. and R. W. Stern, "Model Validation and Time-
Concentration Analysis of Three Power Plants," Publication
No. EPA 450/3-76-002, prepared by GCA/Technology Division under
Contract No. 68-02-1376, Environmental Protection Agency,
Research Triangle Park, N.C. (1975).
4. Morgenstern, Paul, "Summary Report on Modeling Analysis of
Power Plants for Compliance Extensions in 51 Air Quality Control
Regions," Publication No. EPA 450/3-75-060, prepared by Wai den
Research Division of Abcor, Inc. under Contract No. 68-02-0049,
Environmental Protection Agency, Research Triangle Park, N.C.
(1973).
5. Roth, P. M., et. al., "An Examination of the Accuracy and
Adequacy of Air Quality Models and Monitoring Data for Use
in Assessing the Impact of EPA Significant Deterioration
Regulations on Energy Developments," prepared by Systems
Applications, Inc. for Greenfield, Attaway, and Tyler, Inc.
under Prime Contract to Office of General Council, American
Petroleum Institute, Washington, D.C. (1975).
6. Thayer, S. D. and R. C. Koch, "Sensitivity Analysis of the
Multiple-Source Gaussian Plume Urban Diffusion Model,"
Proceedings of the Conference on Urban Environment and Second
Conference on Biometeorology, Philadelphia, Pa. (1972).
7. Hilst, G. R., "The Sensitivities of Air Quality Predictions
to Input Errors and Uncertainties," Proceedings of the
Symposium on Multiple Source Urban Diffusion Models, Environ-
mental Protection Agency, Research Triangle Park, N.C. (1970).
E-19
-------
LIST OF FIGURES
1. Cumulative frequency distributions for model calculated max
24-hour concentrations at all three plants—base case.
2. Sensitivity of the yearly max 24-hour concentration to
changes in stack gas exit velocity.
3. Sensitivity of the yearly max 24-hour concentration to
changes in stack height and terrain adjustment.
4. Sensitivity of the yearly max 24-hour concentration to
changes in mixing height.
5. Sensitivity of the yearly max 24-hour concentration to
changes in wind speed.
6. Cumulative frequency distributions of daily max 24-hour
concentrations at Muskingum River—three meterological data sets.
7. Cumulative frequency distributions of daily max 24-hour
concentrations at Philo—three meteorological data sets.
8. Cumulative frequency distributions of daily max 24-hour
concentrations at Stuart—three meteorological data sets.
E-20
-------
U I
_l«
ulh-
OZ
!2
O U»
s
oo
si
?» uj
O
§|
-------
FIGURE 2. SENSITIVITY OF THE YEARLY rAX 24-HOUR CONCENTRATION^
TO CHANCES IN STACK GAS EXIT VELOCITY
i
| 20-
8 <0
5"
'"
T
I I I
KEY: MUSKINGUM Q
PHILO A
STUART G
FIGURE 3. SEI.SITIVITY OF THE YEARLY KAX 21-HOU* CONCENTRATION
TO CHANGES IN STACK HEIGHT AND TERPAIN ADJUSTMENT
i -20
£ -25
00
KEY: MUSKINGUM O
PHIIO A
STUART Q
I I
•10 !5 -20 IS 10 5 0 5 10 15 20 25 30
PERCENTAGE CHANGE IN INPUT VARIABLE
•30 -25 30 -IS 10 5 0 5 10 IS 20 2S 30
PERCENTAGE CHANGE IN INPUT VARIABLE
FIGURE 1. SENSITIVITY OF THE YEA'UY MAX 2"i-HOUR CONCENTRATION
TO CHANGES IN KIX1NG HEIGHT
25 20 -IS 10 S 0 5 10 IS 20 25 30
PERCENTAGE CHANGE 11 INPUT VARIABLE
FIGURE 5. SENSITIVITY OF THE YEARLY PAX 21-HOUR CONCENTRATION
TO CHANGES IN HIND SPEED
KEY: MUSKINGUM O
PHILO A
STUART Q
JO -25 20 15 10 5 0 S 10 IS 20 IS 30
PERCENTAGE CHANGE IN INPUT VARIABLE
E-22
-------
ggggg g 8 | |gSS83 S S
c" ••* 'NOuvbiN3DNCo *M w wnmxvw aaivwusi
O*
5l
X M
§5
I e i
Esl
III 111
$ >
s S
5
'i1
*»
!S2
3 |
s;
*!
Q ui
n 0
si
i'l
'- s
i
i
£
s ggggssg g
u. 4 « r^ w iJ * r» r^
sssgsa s s s
^
E-23
-------
APPENDIX F
VALIDATION OF A SINGLE SOURCE DISPERSION MODEL
by
Russell F. Lee*
U.S. Environmental Protection Agency
and
Michael T. Mills and Roger W. Stern
GCA Corporation
from
Proceedings of the Sixth International Technical Meeting
On Air Pollution Modeling and Its Application NATO/CCMS
N. 41 September 1975
*0n assignment from National Oceanic and Atmospheric Administration,
U.S. Department of Commerce.
F-l
-------
1. INTRODUCTION
BACKGROUND
Estimates of SOp concentrations downwind from large power plants
are urgently needed to guide environmental and energy related policy
decisions. Most mathematical dispersion models for the prediction of
SOp concentrations provide estimates over averaging times which are
either very short (one hour or less) or very long (seasonal or annual).
For example, the plume parameters given by Turner and developed
principally from earlier work of Pasquill, Cramer and Gifford are based
on experimental data, most of which were collected for averaging times of
10 to 30 minutes. Power law relationships, by which concentrations from
point sources are linked to averaging times, are generally considered to
be valid only over averaging times that range from a few minutes to
perhaps one or two hours. The National Ambient Air Quality Standards
for S02 in the U. S., however, include an averaging time of 24 hours.
A method currently favored for estimating 24-hour concentrations is to
average concentrations that have been predicted for the component one-hour
periods. To date, very few sets of field data have been available to
test the adequacy of this estimation technique.
PURPOSE OF STUDY
The purpose of this paper is to describe validation studies of an
Environmental Protection Agency dispersion nicdcl. The model was
designed to estimate 1-hour and 24-hour ground-level concentrations
caused by emissions of a single source. Emphasis is placed on the
24-hour value.
F-2
-------
The analytical procedures are designed to parallel, to the extent
2 3
practicable, those used by Klug and Montgomery, et al. , in their
analyses of TVA data.
MODEL DESCRIPTION
The EPA Single-Source Model (CRSTER) is a Gaussian dispersion
model developed by the EPA Division of Meteorology. The procedure
calculates maximum daily S0~ concentrations for a year, identifies
the meteorological conditions associated with the maxima, and calculates
hourly and daily concentrations for an array of receptor locations.
The concentrations are calculated for 180 receptor positions situated
at each of 36 directions from the source and at five different distances.
The model handles from 1 to 19 sources but assumes all are at the same
physical location.
Meteorological inputs to the model consist of hourly surface
observations of wind speed (knots), wind direction sector (tens of
degrees), temperature (°F), total cloud cover (tenths), and twice daily
mixing heights (meters). The format for these data is that used by the
National Climatic Center for punched cards of hourly surface observations,
The data are input into a preprocessor program which, in turn, writes a
tape containing hourly values of stability index, mixing height,
temperature, wind speed, flow vector (wind direction plus 180°), and
randomized flow vector. The randomized flow vector is equal to the
flow vector, plus a random number between -4 and +5 degrees. The
randomization procedure prevents excessive concentrations in azimuths
F-3
-------
which are exact multiples of ten degrees. The preprocessor output
tape is then read by the single source model which performs the actual
concentration calculations.
The preprocessor program generates hourly mixing heights from the
twice daily mixing height measurements according to the interpolation
4
scheme for rural areas described in the Interim User's Guide . Hourly
stabilities are determined according to the system given by Turner
employing Pasquill's classification scheme with the addition of a
stability class 7 (G) for which the assumption is made that the plume
does not reach the ground. Wind speeds (u ) measured at instrument
height (h ) (assumed 7 meters for most weather stations) are adjusted
by means of a stability dependent power law (u = u [h/h I*3) to correspond
to values one would expect at the stack top (h). Plume rise is calculated
on an hourly basis using the method of Briggs ' ' . If the plume rise
calculation indicates that the plume height will be above the mixing
layer, then a zero concentration is specified. If the plume height is
below the top of the mixing layer, the presence of the top of the layer
is accounted for by introducing image plumes (see Turner ) to satisfy
the essentially zero flux conditions at ground level and at the top of
the mixing layer.
Source strengths input to the single source model may posses various
degrees of temporal resolution. In the seasonal version of the model,
an annual average SO- source strength is specified along with monthly
variation factors. In addition to the monthly factors, the diurnal
version of the model employs hourly emission variation factors for
F-4
-------
each month of the year. A modified version of the model, used in
this validation study, allows actual hourly source strengths to be
utilized.
2. PLANT DESCRIPTIONS
CANAL PLANT
The Canal Plant and Surroundings
The Canal Plant is located on the south side of the Cape Cod
Canal about 1.6 kilometers from the entrance on Cape Cod Bay
(Fig. 1). The surrounding terrain is gently rolling with ele-
vations generally below 60 meters above mean sea level. The
highest elevations in the area are about 90 meters above sea
level in the western end of the Cape. Most of the area is covered
with scrub pine forests and low vegetation.
Data for the study were collected in 1971. During that year,
the plant consisted of a single oil-fired unit with a generating
capacity of 560 megawatts. The top of the stack was about 91 meters
above grade and 5.6 meters in diameter. The main power plant struc-
ture to the north of the stack totally enclosed the turbine generator
and boiler. The roofs of the turbine and boiler rooms were about
30 meters and 59 meters above grade respectively. Stack and boiler
data are given in Table 1.
Overview of Monitoring Program
SOp concentrations were measured at four locations on a continuous
basis with Ultragas S02 Analyzers manufactured in Germany by H. Wosthoff.
These instruments measure sulfur dioxide by the increase in conductivity
F-5
-------
o.
to S
C O
O b
CO -o
O HI
O w
r-l 3
00 0)
C P -O
•H cd
O v-4
.C en en
W C M
o
•X3 i-l 4J
C JJ 4)
A at ^i
f-l > O
05 M 3
M QJ 4J
a) co
CO 4J
CO .i-l
CO .U Cfl
rt C >
0) 1-4 C
4J CO •!-(
co C O
Cd CO OH
CD o
*J
H-4 <1) CU
O .C co
•u C
a. o
CO >4-4 3
s o o*
•3
00
F-6
-------
LO
O
CO
I-H
D;
o
{=
10
O)
ro
to
c
ro
O
CJ
to •—
00
O
ro to
4«>
to
0
•r—
O.
^
o
ra LO
CO
-*
fQ *3"
4— J
to
u
E
Z3
cn
d
^~
to
^
to CM
00
0
ID i —
00
s_
4-)
i-
ro
to
to 3
-* 0
O <4-
•(-> 1 —
ID
O
4->
to
il
0)
U
lO
S-
to
-C
O
i — to CD
LO """ LO
«3" tD CTl CO LO *J-
CO • CM CO i — CM OO
CM ttf- , —
i — cr» r^-v oo to oo
CO • • LO CM tD CM
i — CM LO CO lO «d-
00 • • LO CM tO i —
LO ^J- »3- , — , —
i — r^. co if) i — r>.
LO • • CM i — CD CO
CM tO *d- «J- LO ^J-
CM
i — to LO O ID OO
LO • • co ^3- r~- «^-
CM r^. oo itf" oo i--
CM
«* o CM oo ^: o cn
•vl- • • 1 — u , — OO
CM 1C CM OO lO to *J-
CM a;
*" ^ SI
o
LO '
LO CM
•* OO
1 — LO
to co
^J- CM
0 CO
«d- co
CM i —
CTO^i- *- -r- S- O x-^ 2
s S- i — tO O-^--> 4-J 2!
+-1 s- >>+-> o o a) >,2 ^ CT>
^: O) ^ ro jQ to c 4-> — - S_ r- ro
Cn-U-r- S_ 40 oj >r_ QJ ro S-
(UEOQ.O too O>
-C:ro>— E 1- EQ.ro OJ +J ro
•t-QJOJ S-O) 3ro+J cn
0 _Q -r- S- ^ "c
"3 EX 0) to ra
+•* 3 tO > i — r—
^ 2: z: =£0.0.
F-7
-------
of an acidified hydrogen peroxide solution and have a full scale
reading of 0.4 ppm. The instruments do not conform to the reference
method for sulfur dioxide or to any of the specified equivalent
o
methods . They have, however, been extensively studied and one
comparison noted a correlation coefficient of 0.99 with the West-
q
Gaeke method . The instruments used provide a continuous real-
time chart trace and a tape printout giving date, time, and average
concentration over consecutive 30 minute periods. The locations of
the SOp monitors with respect to the Canal Plant are given in Figure 2
and Table 2.
Meteorological Data
Bendix-Friez Aerovanes were used to provide local wind speed
and direction data. Through July 1971, the principal source of wind
data was the Aerovane mounted on a 12.2 meter mast located on the
58.8 m boiler-room roof. After July 1971, wind data were obtained
from a second Aerovane installed on a 44 meter tower near the top of
Telegraph Hill, approximately 5 kilometers south-southeast of the Canal
Plant. Wind data were used to identify the samplers not affected by the
plant, for background determination, and for the dispersion modeling.
Mixing heights were derived from the Nantucket Island Rawinsonde data.
STUART PLANT
Stuart Plant and Surroundings
The J. M. Stuart plant is located in Southwestern Ohio on the
Ohio River, about nine kilometers southwest of Manchester, Ohio,
F-8
-------
a
e
o
o
U-l
c
o
n]
o
o
00
c
•H
3
O
tfl
3®
.H
*J O
tH (1)
rt x
c -w
n3
O X
x>
01
J= (0
J-» C
o
U-l -H
O u
to
f, 4J
u co
4-J
QJ CN
^ o
CO W
(U
M
3
00
•H
F-9
-------
Table 2. SULFUR DIOXIDE MONITOR STATIONS
Plant No.
Canal
Stuart
Muskingum
Philo
1
2
3
4
-
1
2
3
4
5
6
7
-
1
2
3
4
-
1
2
3
4
5
6
-
Station
Distance
Name (km)
Top of stacks
Boone
Brudysville
Bentonvil le
Manchester
Maysville
Rectorville
Somo
Top of stacks
Beverly
Hackney
Rich Valley
Cal dwell
Top of stacks
Philo
Fox Run
Irish Ridge
Duncan Falls
Salt Creek
Indian Run
Top of stacks
4.7
2.3
1.4
2.0
-
2.4
6.6
13.4
8.7
3.8
8.4
5.0
-
5.3
4.3
8.3
19.6
-
1.7
4.8
5.0
1.3
6.0
4.2
-
Bearing
(degrees)
119
138
224
312
-
35
15
28
49
279
156
220
-
140
40
35
35
-
174
166
235
343
25
334
-
Elevation above
stack base
(m)
10
4
40
20
91
115
85
121
-7
-4
115
115
244
64
82
101
128
251
3
2
99
12
26
63
81
F-10
-------
and four kilometers east of Maysville, Ohio (Figure 3). It is located
near the center of the river valley. The ridges on either side of the
valley are about 115 meters above the valley floor and 700 meters from
the power plant. The 244-meter stacks rise 130 meters above the sur-
rounding countryside. The data were collected during the one year period
from January 1, 1973 to December 31, 1973. The plant consisted of four
identical coal fired boilers with a capacity of 610 megawatts each.
However, one boiler was under repair during the entire year so that the
total generating capacity was only 1830 megawatts. The yearly average
generating load was 1318 megawatts, or 72 percent of available capacity.
Stack and boiler data are given on Table 1.
Overview of Monitoring Program
The monitoring network consisted of seven sulfur dioxide monitoring
stations (Fig. 4 and Table 2). The instruments were electrolytic con-
ductivity devices manufactured by Leeds & Northrup Company. The sample
was obtained by passing ambient air, taken from five feet above ground
level, through an absorption column along with an absorption solution.
Data were taken continuously and listed every hour. Electrical cali-
bration tests were performed weekly for zero and half-scale operations.
Overall calibration tests were made every six months at 0.2 ppm using
the permeation tube method whose accuracy is traceable to the U. S.
Bureau of Standards.
F-ll
-------
CD AY TON
'COLUMBUS OZANESVILLE
-PHILO PLANT
MUSKINGUM
PLANT
J.M. STUART PLANT
KENTUCKY
PITTSBURGH
O
1HUNTINGTON
WEST VIRGINIA
KILOMETERS
0 50
[ I I i I f
'CP Stale Capital
Fiqure 3. Map of Ohio and surrounding states showing location of
J. M. Stuart Plant, Philo Plant, and Muskingum River
Plant
F-12
-------
Figure 4. Sketch of the J. M. Stuart Plant area showing locations
of seven automatic SO^ monitoring stations
F-13
-------
The monitor at Station 2 was moved to Station 4 on March 10, 1973,
and the monitor at Station 7 was discontinued on June 17, 1973. Therefore,
no data are available for Station 2 for nine months, Station 4 for three
months, and Station 7 for six months. There were some additional hours
of missing data due to loss of electrical power, periods of calibration
and maintenance, and system failures.
Fuel Analysis
Each barge of coal from a specific vendor was sampled during the
unloading process. All samples were analyzed. In the process of deter-
mining the caloric value of the coal by bomb calorimeter, the bomb
washings were titrated using tetrahydroxyquionone (THQ) to determine
the acid content which indicates the sulfur level. The procedure, the
THQ colorimetric method, has been shown to be in agreement with the
standard ASTM method. It is used as a typical procedure by the Dayton
Power and Light Company and the American Electric Power System, and
the Ohio Power Company.
Meteorological Data
The meteorological instrumentation at the J. M. Stuart Plant
consisted of a Bendix-Friez wind speed and direction device mounted
40 meters above the ground on the coal stacking tower. The plant
wind data were used to identify samplers not affected by the plant plume
to aid in determining background. The meteorological data collected
F-14
-------
at the National Weather Service (NWS) Station at Cincinnati, Ohio
were used in the model. Mixing heights were derived from rawinsonde
data collected at the NWS station at Dayton, Ohio.
MUSKINGUM PLANT
Muskingum Plant and Surroundings
The Muskingum Plant is located in Southeastern Ohio, on the
Muskingum River, about six kilometers northwest of the town of
Beverly. Figure 5 indicates the location of the plant, the SOp
monitoring sites, and the surrounding towns. The plant is in the
river valley about 500 meters from the valley walls which rise about
75 meters above the valley floor. The two 251-meter stacks are 640
meters apart and extend about 185 meters above the surrounding
terrain. During 1973 the plant consisted of five coal fired units
with a total capacity of 1467 megawatts (Table 1).
Overview of the Muskingum Monitoring Program
Four sulfur dioxide monitoring stations made up the monitoring
network (Fig. 5 and Table 2). The monitoring stations were established
in 1969 to monitor the ambient changes when the new stacks were
installed10.
Meteorological Data
There were two wind monitoring stations. Both used Friez Aerovane
wind speed and direction devices. One was located at Beverly mounted
33 meters above ground, and the other at the Hackney S02 monitoring
F-15
-------
[RICH VALLEY #3
'CENTERVILLE
'HACKNEY #2
RT 76
MUSKINGUM PLANT
STACK I
'STACK 2
KILOMETERS
012345
Figure 5. Sketch of the Muskingum Plant area showing locations of
four automatic SO. monitoring stations
F-16
-------
station, on a 22-meter mount. The data from Hackney were used for
selecting the "background" stations, but Beverly data were used when
the Hackney data were missing.
Weather data used in the modeling were obtained from upper air
and surface data collected at the NWS station at Huntington, West
Virginia. (See Figure 3).
PHILO PLANT
Philo Plant Description
The Philo plant is a 457 megawatt facility located in eastern
Ohio on the Muskingum River in the town of Philo, about 11 kilometers
to the southeast of Zanesville, Ohio (Figs. 3 and 6). The plant is
roughly 500 meters from the valley walls to the east and west, although
the valley widens to the north. The three stacks are approximately
82 meters high and rise about 10 meters above the surrounding terrain.
During 1974, the period of this study, the plant consisted of five
coal-fired boilers feeding into three stacks (Table 1).
Overview of the Phij_o Monitoring Program
The monitoring system in 1974 was made up of six Malloy contin-
uous S02 samplers, located as shown in Fig. 6 and Table 2. The
monitoring system maintenance and data acquisition were performed
by the Environmental Research and Technology Company in Lexington,
Mass. The instruments were calibrated every six months in Lexington
and zeroed every night by computer.
F-17
-------
Figure 6. Sketch of the Philo Plant area showing locations of
six automatic S02 monitoring stations
F-18
-------
Data were recorded for all of 1975 except for the following:
Station 1 First 91 days of year
Station 4 First 91 days of year
Station 6 Second 91 days of year
Meteorological Data
There were three meteorological stations:
1. Irish Ridge Upper - elevation 140 meters above plant base,
(50 meters above ground). Measured wind speed and direction,
temperature, and temperature difference between Irish Ridge
Upper and Irish Ridge Lower.
2. Irish Ridge Lower - elevation 104 meters above plant base,
(11 meters above ground). Measured wind speed and direction,
and temperature.
3. Duncan Falls - elevation 14 meters above plant base,
(6 meters above ground). Measured wind speed and direction.
The wind instruments were Climet Anemometers.
The winds from Irish Ridge Upper were used to determine which sites
to use to estimate background. When Irish Ridge Upper data were missing,
Irish Ridge Lower data were used. When that data were also missing, Duncan
Falls data were used. Surface wind and stability data from the National
Weather Service Station at Columbus, Ohio were used in the dispersion
model. Mixing heights were derived from upper air data collected at
Dayton, Ohio.
F-19
-------
3. DATA PREPARATION AND PROCESSING
Meteorological Data
The sources of the meteorological data are shown on Table 3.
Hourly surface observations from airport log sheets were keypunched
on to cards.
Table 3
SOURCES OF METEOROLOGICAL DATA
Plant
Surface Observations
Airport
Year
Upper Air Data
Airport
Canal
J. M. Stuart
Muskingum
Philo
Quonset Point, R. I. 1971
Cincinnati, Ohio 1973
Huntington, W. Va. 1973
Columbus, Ohio 1974
Chatham, Mass.
Dayton, Ohio
Huntington, W. Va.
Dayton, Ohio
The surface observations included:
t station
• date and time
• ceiling height
• ambient temperature
• wind direction
• wind speed
• percent cloud cover
F-20
-------
Table 4
AVERAGE MONTHLY PERCENT SULFUR CONTENT OF FUEL
Month
January
February
March
April
May
June
July
August
September
October
November
December
Canal
2.0
1.9
2.1
1.9
2.1
2.1
2.1
2.0
1.9
0.9
1.0
0.9
Stuart
1.8
1.6
1.8
1.7
1.8
1.6
1.5
1.5
1.5
1.5
1.8
2.1
Muskingum
4.9
4.8
4.8
4.5
4.7
5.0
4.7
4.7
4.3
4.6
4.5
4.4
Philo
3.9
4.8
4.7
4.4
3.3
3.2
2.6
3.?
3.2
2.4
2.6
3.7
F-21
-------
Table 5. 1-HOUR CONCENTRATION DISTRIBUTION STATISTICS
FOR MEASUREMENTS AND MODEL VALIDATION RUN
(ymg/m3)
Sampling
Plant Station
Canal
Stuart
Muskingum R.
Philo
1
2
3
4
1
2d
3
4
5
6
7d
1
2
3
4
1
2
3
4
5
6
Ninety-fifth
Percentile3
Mb Pc
25 < 1
14 < 1
18 < 1
15 < 1
140 <10
80 <10
74 26
53 <10
28 <10
48 <10
33 <10
27 <10
57 <10
130 <10
72 <10
50 <10
37 <10
47 <10
27 <10
35 80
118 20
Ninety-ninth
Percenti lea
M
101
72
18
31
270
445
200
180
80
135
102
150
270
350
200
170
163
163
190
134
253
P
6
< 1
2
< 1
400
180
240
130
<10
120
30
160
150
210
160
98
222
920
88
555
650
Second
Highest
M
435
553
446
575
685
685
1022
750
495
980
325
857
786
996
735
525
735
745
665
575
565
P
253
174
427
81
1372
814
565
515
823
595
976
980
1304
873
465
1295
945
4049
1945
1279
2369
Highest
M P
433
618
732
638
857
1014
1153
883
565
1053
435
925
786
1179
786
893
891
917
695
675
595
283
179
479
377
1393
948
1022
541
1219
693
1000
1083
1310
933
645
1639
1059
4593
1981
1344
2482
aPercentile values given in terms of cumulative percent of concentration
less than given values.
Measured concentrations with.subtracted background.
cPredicted concentrations.
Samplers were in operation for less than half the year. Data not
included in subsequent analyses.
Corrected 3/24/77
F-22
-------
Table 6. 24 HOUR CONCENTRATION DISTRIBUTION STATISTICS
FOR MEASUREMENTS AND MODEL VALIDATION RUN
(yg/m3)
Ninety-fifth
Percent! lea
Sampling
Plant Station
Canal
Stuart
Muskingum R.
Philo
1
2
3
4
1
2d
3
4
5
6
7d
1
2
3
4
1
2
3
4
5
6
Mb
32
15
17
15
83
46
50
40
31
42
45
32
55
98
52
45
35
44
41
23
65
pc
4
<1
4
<1
55
28
36
24
5
21
23
32
32
31
24
29
39
143
47
81
107
Ninety-ninth
Percent! lea
M
52
28
46
44
245
160
110
63
52
135
69
100
100
130
95
134
60
92
60
78
121
P
14
6
18
2
128
52
75
41
50
46
60
69
80
58
41
139
69
368
111
207
217
Second
Highest
M
66
36
77
63
259
63
181
79
63
147
69
133
131
165
109
132
67
127
62
87
121
P
16
9
38
4
149
75
91
45
57
69
73
81
82
73
45
133-
86
471
165
222
282
Highest
M
75
46
83
75
277
159
225
83
77
195
77
170
137
227
115
133
no
132
158
94
138
P
29
11
39
16
161
98
102
49
75
83
120
97
91
74
47
147
104
541
220
226
356
Percentile values given in terms of cumulative percent of concentrations
less than given values.
Measured concentrations with subtracted background.
Predicted concentration.
Samplers were in operation for less than half the year. Data not
included in subsequent analyses.
F-23
-------
Plant Emission Data
The hourly emission rate of SO,, was computed from the fuel
consumption rate and the sulfur content of the fuel. The hourly fuel
consumption rate of each of the boilers was summed to obtain the plant
fuel consumption rate. The plant fuel rate was then multiplied by the
sulfur content and an appropriate conversion factor to obtain the hourly
sulfur dioxide emission rate from the plant.
Air Quality Data
The air quality data required for this study were the concentrations
of sulfur dioxide (SCL) attributable to the plant being studied. The
samplers measured concentrations, not only from the plant being studied,
but from other sources as well. The portion of the concentration due to
those other sources is referred to as background in this study. Since
there is no accurate way of knowing how much of the total concentration
is due to the power plant, this portion of the concentration must be
estimated.
The portion of the concentration due to the plant, for each hour,
at each sampler, was estimated, first, by estimating the background con-
centration and, second, by subtracting the background concentration
from the concentration measured at each sampler. One background
concentration was assumed to be valid for all receptors for one hour.
The background concentration was taken to be the average of the con-
centrations from all samplers that are not located with + 45° of the
downwind direction. The downwind direction is based on the wind
direction recorded by the wind vane at the plant. When the wind observation
indicated calm, or was missing, the last recorded wind direction was
F-24
-------
used, with one exception. This exception occurred when a sampler
recorded a concentration greater than 0.1 ppm. In that case the downwind
direction was taken as the direction from the plant to that sampler.
Quality Control
Particular attention was given to quality control. Emissions, air
quality, and meteorological data were each keypunched with the date and
time preceding the measured values. All punched cards were verified.
A computer program was used to check for missing hours, cards out of
chronological order, input data out of range, and unreasonable changes
between consecutive data values. The keypunched data were read onto
magnetic tape, with each record prefixed by a plant code to prevent the
unlikely mixing of data from different plants. All programs which
modified the data had internal checking routines to assure that the
correct data was read. The output from each program was spot checked
by manual calculations.
4. RESULTS OF THE MODEL VALIDATION
Figures 7 through 22 show log-normal plots of cumulative
frequency distributions of (1) measured concentrations, (2) measured
concentration minus the estimate of "background," and (3) the model-
calculated concentration for the same location. Rather than presenting
data for all the sampling stations, data are shown for two sites for
each plant. In each case, the data presented are from sites where the
model gave the best and the poorest predictions, respectively. Tables 5
and 6 show the measured and predicted concentrations at each sampler for
selected points on the frequency distributions.
F-25
-------
A comparison of the frequency distributions of the model cal-
culations and the observed one-hour concentrations shows that the
model predicts the upper percentile fairly well, but significantly
under-predicts most of the remainder of the distribution. A
similar effect occurs in the frequency distributions of the 24-hour
concentrations. Part of the underprediction may be due to sampler
2
errors, as Klug speculates , since many of the lower concentrations
are measured near the threshold of the sensing device. Also, much
of the low concentration end of the distribution does not represent
pollution from the plant at all, but rather differences between the
estimated background and the actual background at the sampler. For
example, if three samplers upwind of the plant recorded concentrations
10, 20 and 24 yg/m3, the "background" would be considered the average
of the upwind stations, in this case 18 yg/m3. This "background" is
subtracted from each concentration recorded at that hour, so that,
in this case, there is one negative concentration, and two positive
concentrations of 2 and 6 yg/m3, respectively. Corresponding model
predictions would, quite correctly, be zero.
Not all the underprediction can be explained by the above
hypotheses, however. Several sampling stations show large under-
predictions for high observed concentrations. For example, on the
frequency distribution of one-hour concentrations at the Muskingum
River sampling station 3, the 98th percentile measured value is
well over 200 yg/m3, while the model prediction is on the order of
60 yg/m3. As Mills, et al. noted, an error of this size could
occur if the actual stability differs from the stability as calculated
by the computer program.
F-26
-------
In the U.S.A., short term (24 hours or less) National Ambient
Air Quality Standards, for SCL and particulates, are given in terms of
concentrations "not to be exceeded more than once per year." Thus, the
second highest concentrations are of concern. The ratios of the model
prediction to the measured (less background) of the second highest one-
hour concentrations range from .3 to 3.0, except for two Philo samplers
where the ratios are 4.2 and 5.4. These two samplers were located at
elevations 18 meters below and 18 meters above the stack tops, respectively.
The geometric mean of the ratios is 1.2. The corresponding ratios for the
24-hour concentrations range from .2 to 2.7, except for one sampler at
the Philo plant where the ratio is 3.7 and one at the Canal plant where the
ratio is 0.06. The geometric mean of the ratios is 0.7.
The model tends to underpredict the 24-hour concentrations,
and slightly overpredict the one-hour concentrations. The degree
to which the model overpredicts or underpredicts is a strong function
of the plant being studied. For example, at the Canal plant, the
second highest one-hour concentrations are underpredicted at three
of the four sites, while the corresponding 24-hour concentrations
are underpredicted at all four sites. The ratios of predicted to
measured concentrations (without background) at the Canal plant range
from .3 to 1.0 for 1-hour concentrations, and from less than .1 to
.5 for 24-hour concentrations, with geometric means of .6 and .2
respectively (Figure 7). At Philo, however, the second highest
one-hour concentrations are overpredicted at all six sites, and
the corresponding 24-hour concentrations at five of the six sites.
F-27
-------
Table 7
Geometric Means of the Ratios of Predicted to
Measured (less background) Second-Highest Concentrations
Plant 1-Hour 24-Hour
Canal .60 .20
Stuart .95 .59
Muskingum River 1.01 .51
Philo 2.79 2.06
All Plants 1.23 .68
F-28
-------
The ratios of the predicted to measured concentrations range from
1.3 to 5.4 for one-hour concentrations, and from 1.0 to 3.7 for
24-hour concentrations. The geometric means of the ratios are
2.8 and 2.1 respectively.
At the Stuart and Muskingum River plants, the second-highest
one-hour concentrations are underpredicted at about as many sites as
they are overpredicted. The corresponding 24-hour concentrations are
underpredicted at all 11 sites. The ratios of predicted to measured
second highest one-hour concentrations range from .5 to 2.0 with a
geometric mean of 1.0. The corresponding ratios for 24-hour concen-
trations range from 0.4 to 0.9, with a geometric mean of 0.6.
The correlation coefficients between the ratio of the predicted
to measured (less background) and distance are given in Table 8. The
negative correlations of the ratios of predicted to measured with
distance indicates a tendency for the model to underpredict more at
greater distances. The correlation is highly significant at Stuart,
but not statistically significant at the other plants.
The negative correlations between the ratios and distances are
12 13
consistent with theory and observations which indicate that beyond
a few kilometers, the plume width should not be proportional to distance
to the 0.9 power as used in this model, but some smaller power.
F-29
-------
Table 8. CORRELATION COEFFICIENTS OF RATIOS
(predicted/measured) TO DISTANCE
Averaging
Time Plant
1
1
1
1
1
24
24
24
24
24
Canal
Stuart
Muskingum R
Philo
All Data
Canal
Stuart
Muskingum R.
Philo
All Data
Correlation
-.41
-.91
-.78
.09
-.29
-.24
-.56
-.82
.33
-.19
95% Confidence
Interval
- .98 to
-.994 to
-.995 to
- .78 to
- .66 to
- .98 to
- .97 to
-.996 to
- .66 to
- .59 to
.91
-.14
.72
.84
.19
.94
.64
.67
.90
.29
Significant to
5% Level
No
Significant to 1%
No
No
No
No
No
No
No
No
F-30
-------
5. CONCLUSIONS
Based on the analyses of the performance of the Single Source
model at four large power plants, several conclusions can be made.
The model predicts the upper percent!le of the frequency
distribution of one-hour concentrations and of the corresponding
distributions of 24-hour concentrations acceptably well. Concen-
trations over the remainder of the frequency distributions are
significantly underpredicted. Part of this underprediction is an
apparent one, resulting from the technique used to estimate the con-
centrations. Another portion of the underprediction is probably due
to measurement errors when concentrations are near the threshold level
of the instrument. Obviously there are also uncertainties in speci-
fying the appropriate wind direction, wind speed, and stability class.
The following specific conclusions are made.
• The second highest one-hour concentrations are predicted
within a factor of 2.0 at two-thirds of the sampling sites.
The geometric mean of the ratios of the model prediction to
the measured (less background) for all sampling sites is 1.2.
The most significant overpredictions occurred around the Philo
plant, where the model is less likely to account properly for
the terrain influences. These influences are very pronounced
because the effluent is emitted from stacks which are little
higher than the terrain features themselves.
• The second highest 24-hour concentrations tend to be under-
predicted by the model, with the ratio of predicted concentration
to measured concentration ranging from about .2 to 2.7 at
about 90 percent of the sites.
F-31
-------
• The degree to which the model over or underpredicts con-
centrations is strongly dependent on the factors relating to
the plant and its surroundings. For example, at the Canal
plant, the model significantly underpredicts the second
highest one-hour concentrations at three sampling sites out of
four, and the second highest 24-hour concentrations at all
four sampling stations. At the Philo plant, however, the model
overpredicts the second highest one-hour concentrations at all
six sites, and the second highest 24-hour concentrations at
five of the six sites. At Stuart and Muskingum River, the
second highest one-hour concentrations are underpredicted at
three sites out of five and two sites out of four, respectively.
The corresponding 24-hour concentrations are underpredicted at
all the Stuart sites, and three of the four Muskingum River sites.
t The model tends to underpredict at greater distances, a
tendency which is highly significant at one plant. This
12 13
is consistent with theory and observation in that, at greater
distances, the plume width increases with.distance to a
power rather smaller than the 0.9 power as is assumed in
this model.
F-32
-------
p
6
r>
6
w M
z
2 o
£3
a: Ro
^ o
O O
z
o
to
cr
3
UJ
3:
2y
5m
O
UJ
<
~J
o
o
O CO
1111! I
I I I
I I I
00 O p CD O
O0> «0 f- <• *>
O
en
en
S CO
Z -J
UJ <
2 o >
^°
o O UJ
N O ^_
o <
to cr o
SI
_ z
<
U- X
38
?i^
UJ
10 O
a:
LJ
CO
d
o
1-1
4J
03
4-1
W
t/i
d
o
•r-i
4J
cd
>-i
4J
d
QJ
U
d
o
o
o
CO
V-i
3
o
CM
W
d
o
•r-l
4-1
4J
W
O
d
cu
QJ
>
•H
4-J
I
cu
>-i
3
W)
F-33
-------
o
«f ^
CC 3 2
Z ^
UJ o
O O eg
O o
CE Q *
— X O
o Q:
UJ
o
UJ
v-' »
< < o
1 UJ -
UJ
1 I i I i
T~I t i i I T
Z^s
Q Q 3»-
UJ UJ r> <
cr a: Q: _i
CO tO ••£
-------
tO I 6 8 t 9 S ?i
* _ 1 1 I J .1 1 1
Ul
Z 3
o £
z y
u t-
o <
o: y
UJ tt.
c
o
•H
4-J
3
43
•H
O CO
C
w
nj co
^H C
3 O
a -H
C C
nJ cu
iH O
ex c
o
*J O
t-l
03 cv
3 O
•U C/D
CO
(-1
(U
3
cn
tor
F-35
-------
tO I 6 8 L 9 S h £
ew/on
f> I e 9 L 9 s
-—-Li i i i i
e
_L
V. UJ
z o
O _J
g*
|a
5$
z§
°i
o ~
^H
!SK
Z UJ
UJ H
O <
cr uj
LJ £t
Q. O
tO V 6 8 L 9 S
o
14-1
o
•H
4J
3
X>
^H
M
4-1
W
•H
C
0) C
3 O
cr -H
0) 4-1
J-i CD
M-l 4-1
U)
0)
> *->
iH (t)
4J
t\J CO
rH C
3 O
a -H
3 4-1
O nj
C C
a) a)
.H O
d. C
o
4J O
3 O
4-1 CO
CO
M
• 3
S O
• I
13
C7)
NOUOU1N33N03
F-36
-------
C0 I 6 8 t 9
EW/90 NOI IUyiN33N03
I)I|68t3Sli £
<• 1 I I I I 1 J _ L
fl\
o
•H
4J
3
X>
•H
CO
•H
a
c
c
J-i O
> *->
•H cn
4-1
8 w
3 C!
o o
•H
4-1 4J
Cu C
0)
VJ 0
a> c
> o
•H O .
a o
d c/3
oo
a i-<
•H 3
^! O
cn ,12
,or
£
^
CD
F-37
-------
NOIlUUiN33N03
l-l
o
c
o
t-l
iJ
to
•H
-a
o
0)
3 w
•H CO
4-1
(fl iJ
S w
3 d
o o
a. c
aj
w o
OJ C
> o
•H U
S O
3 to
00
C H
•H 3
4«! O
W 4=
3 I
CVJ
13
cr
F-38
-------
{0 I 6 9 /L 9
ew/on
TI i ^
0 I 6 9 L 9
I 1 I I i
J-J
o
c
o
JD
•H
CO
>> CO
O iJ
C W
CU
Z3 4-1
cr cc
(U
0)
>
rd 4-1
rH C
3 QJ
e u
3 C
O O
O
4J
C CN
03 O
vH CO
a
1_!
O 3
rH O
a>
F-39
-------
C0 I 6 9 L 9 S Ti
to uj
Z 3
O -I
g*
|a
5$
z§
°1
o 5
u_ 2
°<
UJ J
o >-
<
H Q:
z uj
U H
0 <
o: uj
UJ CK
Q. O
tot'
68L 9
T
s
~l i i r~~i i i
£0 I 6 8 L 9 S 1)
>or
Cm
F-40
-------
d
o
rt
4-)
w
Z «
g
t— o
I- ->
Z 3 o
UJ -* ~
Z UJ !•>
U
oc 5
820
T W
DC
O <7i
U
oc
s:
i n i T
i r
ro
Z
O
^3 2 ^
z ^ z
1/1
o
o
QC.
«/>
w « <
"- |_ UJ UJrf<
-, Z 5 S«u
8
o o o o o
01 ID r- <0 m
o
10
O
CM
10 o
Z
-,0
UJ
o ^J
oo z
?8
0^1
_
O I
Of.
UJ
a.
4J
CO
c
O
cfl
J-4
U
C
QJ
O
ti
O
U
CM
O
(/I
S-i
3
O
CM
Vi
O
o
•H
4-1
3
J-l
4J
O
c
(U
cr
•r-l
4-1
I
O
uj/8n
0)
1-1
60
F-41
-------
z
O
cc
t—
Z
UJ O
~
z >
U S
< o
U *»
25
>o
f: o
o
Z o
<
O X
UJ
O <*
< £
t; < °-
Z uj
UJ QC
U O o
o:
LD
Q_ CO
O-
O-
O
T I I T
z
o
Zi
Z
5
O
O
<
ex.
Q
UJ
U
<
U
I I I I I I L
I I I I I I I I
8i
o
o>
CO
o>
10
s|
o <
O- QC
^.
O
"°
O
-
<
oo o o
OQ. oo 1^
o
»O
OO co I
«0 LU
U
VO
q
o
q
o
cuj/6n 'NOIlVyiN30NOO
CXI
C
O
V)
W
C
O
S-i
4J
C
-------
ew/on
C0 I 6 8 I 9 S ft E e ,0 I 6 8 L 9 S li £
» _ i i i i i i i t
,01
o
(4-1
c
o
•r-(
AJ
3
XI
o
c c
0) O
d -H
cr ^-1
OJ OJ
0)
•H
4-1
CO
w
C
r-l O
3 -H
6 *->
3 CO
O W
4-J
4J C
C
-------
,0 I 6 8 L 9 S ti
* i i i i i i ,
~ 4.
o
Z
O
ZH
5t
U. 7
o*
UJ*
CD *-
£0:
Z Ul
Ul I-
o <
o: ui
ui 01
Q. o
o J,
,01,
to
2
o
OC t-
o
r
i
t—
oc
o o
UJ U_
oc
U_ -2.
D
UJ ~
> t—
— r>
I— CD
cr *-
-I OC
r> t-
s: to
o ~
o a
CO
•2
O
2 = a
cc U
tr o a »-
2 uj uj a
i^i oc oc _i
V \
4. o
o
2
O
O
OC
o
iC
u
a
CD
to
o
2
2
«i
(/)
Z
2uj
o < _J
* K <
o t- >
* Z
*§S
'|S
o O
« U. z
O «-•
» uiZ
^x
2 P H
5{g
« O uj
S^
N a-
4- *
o
••3
• 3
«
T 9
6
S
6
c
o
•H
4-1
3
XI
•H
M
4-1
(A
•H
O
C C
(1) O
3 -H
CT 4-»
>4-l CO
(U 4-1
> m
•H
4J W
rt c
•H O
2 -H
S w
3 03
U M
C (U
CO cj
•H C
CL. O
O
CN
CO O
3
CO
3
O
CN
CO
F-44
-------
,0 I 6 8 L
* _ i < i i
S ti
e
_L-
?0 I, 6 8 L 9 S
U I 1 11
cw/on
p r e 8 L 9
NOIiUUlN33N03
c
o
3
XI
•H
1-4
4-1
w
•H
O
C
01 CO
3
cr c:
W
•H
4-J 4-1
CO cd
i-H
3 03
B C
3 O
a -H
4-1
4-1 CO
C VJ
CO 4-1
.H C
{X 0)
u
H C
QJ O
> CJ
•H
a!, C
O
B to
W) n
C 3
•H O
^ x:
en i
3
-------
eO I 6 8 £ 9
* [iiii
EW/9n NOIiUyiN33NOa
E 2 eO I, 6 8 t 9 S >i e Z
I1 6 8 t 9 S li C
o
U-l
c
o
•H
•H
M
4-J
CO
•H
*a
o
c
0) CNI
3
cr c
cu o
M -H
U-l 4-J
CO
OJ 4-1
> CO
2 to
a c
3 O
a -H
4-1
4-1 CO
C >-i
CO 4-1
iH C
ex. ai
CJ
)-i C
0) O
> o
CN
o
E en
3
00 >-i
C 3
•H O
^ X
en I
3
-------
,0 I 6 9 d 9 S h £
c . I i i i T i 1 1
0 I 6 8 L 9
c
o
•H
4J C
w o
•H -H
CJ CO
c
0) 4-1
3 rt
cr
CD CO
y> c
M-l O
•H
0) -U
> CO
•H >-i
4J -U
o) C
iH CU
3 O
e c
3 O
O O
CO
r-l
O O
rH -d
•H I
x:
-------
cw/on
tO I 6 8 L 9 S Ti
«• i i i i i i i
C
_L
,01, e 8 L 9 s
: 11 i i i i
V) UJ
2 ID
O _l
z^
o§
u 5
U. 7
°<
UJ.1
o: uj
uj o:
Q. o
c
o
•H
AJ
D
w o
•H -i-l
U (0
c
(U 4-1
3 OJ
cr
0) 0)
^ C
<4-l O
•H
01 4_)
> 03
•H M
4-1 4J
3 O
I C
3 o
u o
CM
c o
OJ W
iH
CX M
O O
CN
0)
(-1
Ml
F-48
-------
References
1. Turner, D.B., Workbook of Atmospheric Dispersion Estimates. U. S.
Environmental Protection Agency, Office of Air Programs Publication
AP-26, Revised 1970.
2. Klug, Werner, Dispersion from Tall Stacks. U.S. Environmental
Protection Agency, Office of Research and Development, Environmental
Monitoring Series, Publication EPA 600/4-75-006, October 1975.
3. Montgomery, T.L., S.B. Carpenter and H.E. Lindley, The Relationship
Between Peak and Mean S02 Concentrations. Conference on Air Pollution
Meteorology of the American Meteorological Society in cooperation
with the Air Pollution Control Association,. Raleigh, North Carolina,
April 5-9, 1971.
4. Hrenko, J., D.B. Turner and J. Zimmerman, Interim User's Guide
to a Computation Technique to Estimate Maximum 24-Hour Concentra-
tions from Single Sources. U.S. Environmental Protection Agency,
Division of Meteorology, October 1972.
5. Briggs, Gary A., Plume Rise. U.S. Atomic Energy Commission Critical
Review Series, TID-25075, National Technical Information Service,
Springfield, Virginia, 22151, 1969.
6. Briggs, Gary A., Some Recent Analyses of Plume Rise Observations,
pp. 1029-1032 in Proceedings of the Second International Clean Air
Congress, edited by H.M. Englund and W.T. Berry, Academic Press,
New York, 1971.
7. Briggs, Gary A., Discussion on Chimney Plumes in Neutral and Stable
Surroundings. Atmospheric Environment 6_, 507-610, July 1972.
8. Federal Register, Volume 36, No. 158, August 14, 1971.
9. Preining, D., et al., Staub-Reinhalt Luft, Vol. 29, No. 11,
November 1969.
10. Smith, M.E., S. Frankenberg, Improvement of Ambient Sulfur Dioxide
Concentrations by Conversion from Low to High Stacks. Journal of
the Air Pollution Control Association, 25(6): 595-610, June, 1975.
11. Mills, M.T., and F.A. Record, Comprehensive Analysis of Time-
Concentration Relationships and the Validation of a Single-Source
Dispersion Model. U.S. Environmental Protection Agency, Publication
EPA-450/3/75-083, 1975.
12. Pasquill, F., Atmospheric Diffusion, second edition, pp. 193-197,
362.364, 1974.
13. Csanady, G.T., Crosswind Shear Effects on Atmospheric Diffusion,
Atmospheric Environment.
F-49
-------
TECHNICAL REPORT DATA
(Please read Inunicnons on lite re', crse before completing)
1. REPORT NO.
EPA-450/2-77-013
2.
3. RECIPIENT'S ACCESSION* NO.
4. TITLE AND SUBTITLE
USER'S MANUAL FOR THE SINGLE SOURCE (CRSTER) MODEL
5. REPORT DATE
July 1977
6. PERFORMING ORGANIZATION CODE
7. AUTHOHIS)
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
U. S. Environmental Protection Agency
Office of Air and Waste Management
Office of Air Quality Planning and Standards
Research Triangle Park, N. C. 27711
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
12. SPONSORING AGENCY NAME AND ADDRESS
Same as item 9
13. TYPE OF REPORT AND PERIOD COVERED
Final Report
14. SPONSORING AGENCY CODE
200/04
15. SUPPLEMENTARY NOTES
16. ABSTRACT
The Singre Source (CRSTER) Model is a steady-state, baussian yrume dispersior
nodel designed for point-source applications. It calculates pollutant concentrations
for each hour of a year, at 180 receptor sites on a radial grid. The hourly concentra-
tions are averaged to obtain concentration estimates for time increments of specified
length, such as 3-hour, 8-hour, 24-hour, and annual. The model contains the concentra-
tion equations, the Pasquill-Gifford dispersion coefficients, and the Pasquill stability
classes, as given by Turner. Plume rise is calculated according to Briggs. No deple-
tion of the pollutant is considered. Technical details of the programming are presented
with complete descriptions of data acquirements and output. Flow diagrams and source
program listings, including subprograms, are given as well as input data forms. Three
papers on application, sensitivity and validation of the model are included as appendice
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
Air pollution
Turbulent diffusion
Meteorology
Mathematical models
Computer models
Sulfur dioxide
Suspended Particulates
b.lDENTIFIERS/OPEN ENDED TERMS
Dispersion
Air Quality Simulation
Model
c. COS AT I I idd/Croup
13. OlbTHIBUT.ON STATEMENT
Release unlimited
19. SECURITY CLASS ('1 his Kepvrll
Unclassified
21. NO. Of PAGES
20. SECURITY CLASS (Thispanel
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
G-1 rU.S. GOVERNMENT PRINTING OFFICE:! 977 -740-110/308
------- |