United States
           Environmental Protection
           Agency
             G*fice of Air Quahtv
             Plannmg ana Standards
             Research Triangle Park MC 277' 1
EPA-450/4-85-009
July 1985
           Air
EPA
User's Guide for the
Multiple Airshed
(MASH) Model

-------
                                       EPA-450/4-85-009
USER'S  GUIDE  FOR THE MULTIPLE
       AIRSHED (MASH) MODEL
                          By

                       Roy A Paul
                     PEI Associates, Inc.
                 Durham, North Carolina 27701
                   Contract No 68-02-3512
                EPA Project Officer: Keith Baugues
                                   U.S. Environmental Proie"'i0->
                                   Bogion 5, 7 :?>r?r/ i     ,..'"
                                   230 3. Deal-Do,., ^^i,  7; ' , .
                                   Chicago, IL  60604  '  "" ""
             U.S. ENVIRONMENTAL PROTECTION AGENCY
                  Office Of Air And Radiation
             Office Of Air Quality Planning And Standards
             Research Triangle Park, North Carolina 2771 1

                       July 1985

-------
This report has been reviewed by the Office Of Air Quality Planning And Standards  U S. Environmental
Protection Agency, and approved for publication as received from the contractor Approval does not signify
that the contents necessarily reflect the views and policies of the Agency,  neither does mention of trade
names or commercial products constitute endorsement or recommendation for use.
                                      EPA-450/4-85-009

-------
                                   CONTENTS



Figures	    iv
Tables	     v
Acknowledgment	    vi

     1.   Introduction	     1

     2.   Processing Ozone Data 	     6

     3.   Processing Meteorological  Data	    10

     4.   Establishing a File of Instructions	    15

     5.   Operating the Main Program	    19

     6.   Formatting the Outputs	    23

-------
                                   FIGURES







Number                                                                Page



   i     Flow of data  processing  for MASH analysis  	    2



   2     A runstream for processing ozone data  	    7



   3     A runstream for calculating daily  resultant wind directions  .   12



   4     Sample  records  (lines) from the instructions file 	   16



   5     A runstream for printing the  instructions  file	   17



   6     A partial  runstream for  sorting the instructions file ....   18



   7     A runstream for executing  the main MASH program	   20



   8     A runstream for combining  outputs  into a formatted table. .  .   24

-------
                                   TABLES
Number                                                                 Page

   1     Hourly  Meteorological Data Received in 1984 From National
           Climatic  Center	    11

-------
                                ACKNOWLEDGMENT
     The Multiple Airshed  (MASH) model  was  based on  concepts originating
within the Monitoring Data  and Analysis Division  (MDAD),  and was  initiated
under the auspices  of  the  Strategies  and  Air Standards  Division  (SASD).
Both Divisions are  within  the Office of Air Quality  Planning and  Standards
(OAQPS) of the  U.S.  Environmental Protection  Agency  (EPA).   Special recog-
nition is given  to  Dr.  Edwin L. Meyer and Mr.  Gerald  P.  Gipson of MDAD who
specified methods to distinguish different  airsheds and who provided  detailed
technical guidelines on  allocating hourly ozone values  at  specific  monitoring
sites.

     Mr. Thomas R. McCurdy of  SASD originated this project as a part of the
analysis of  risks associated with alternative  National  Ambient Air  Quality
Standards (NAAQS) for ozone.  Dr. David J. McKee (SASD) supervised the  first
phase of the  project.   Mr.  Keith A. Baugues  (MDAD)   supervised  the second
phase and provided  technical  direction.   Mr.  Warren P.  Freas (MDAD) recom-
mended a  system  for combining  data  sets,  which   was  incorporated into the
model.

     The successful  completion of this work  is largely  due to the  excellent
administrative and technical   guidance received  by  the  author from the  above
named people.

-------
                                 SECTION 1
                                INTRODUCTION

     The ozone (03) problem in the United States  is  sometimes  characterized
by long-range transport of 03 and/or its precursors.   In  such  situations,
observed high concentrations of 03 are not  necessarily  attributable to
nearby, obvious potential  sources (i.e., cities).  The  most  desirable way
to determine cause/effect  relationships is  to perform a detailed  case-by-
case analysis.  This is the procedure recommended  and followed in 03 modeling
analyses performed as part of the State Implementation  Planning  (SIP)
process.  However, in regulatory impact analyses  (RIA), it is  often necessary
to consider the impact of  proposed regulations  in  or near dozens  of cities.
Under such circumstances detailed case-by-case  analyses are  not  feasible.
This document describes an automated procedure, the  Multiple Airshed (MASH)
model, in which wind data  and time of occurrence  of  high 03  values are used
to identify the city (or SMSA) whose emissions  are most likely associated
with each observed high concentration of 03.  When an observed 03 concentration
is assigned to an SMSA, it is assumed that  the  particular monitoring site
is located within an airshed whose source area  consists of the city or SMSA
to which the 03 value has  been assigned.
     Multiple airshed (MASH) model calculates design values  for  each of a
series of ozone monitoring sites in the U.S.  Each monitor is assigned to
various airsheds, depending on time of day  and  the resultant wind direction
for each day.  Using a list of high values  for  each  airshed, one design value
is selected for each airshed and monitor combination.   The design value
associated with a particular SMSA can be selected  from  the resulting list
of site-specific design values which the program  generates for each airshed.
Input data for MASH are 1) hourly ozone values  from  monitors,  2) daily

-------
Component
 Data
sources
Utilities
Intermediate
   files
Special
programs
Ozone
data
Meteorological
     data
Instructions
   data
MASH
files
 Main
program
Output
files
Formatting
 program
Combined
 output
  file
       Figure  1.   Flow  of  data  processing for MASH  analysis.

-------
resultant wind directions at selected meteorological  stations,  and  3)  a
coded set of instructions telling how to allocate  hourly  ozone  values  to
various airsheds based on wind direction.
     Air quality data are obtained from the Storage and Retrieval of
Aerometric Data (SAROAD) data base maintained on UNIVAC by  the  National Air
Data Branch (NADB) within Monitoring Data and Analysis Division (MDAD).  A
utility program developed by NADB called NA205 is  used to select  hourly
ozone data from selected monitors over a selected  3-year  period (Figure 1).
These hourly data are placed in a mass storage file.  Program 03-VALID,
developed by PEI Associates, Inc. (PEI) reads this file and applies certain
validity criteria specified by MDAD.  The screened and validated  data  are
placed into a second mass storage file in a format suitable for MASH.
Limits on the number of monitors that can be specified for  NA205  and practical
limits on the size of mass storage files require that a series  of validation
runs and MASH runs be executed to process all ozone monitoring  sites in the
U.S.  Further information on air quality processing may be  found  in Section 2.
     Meteorological data are obtained from a historical data tape produced
by the National Climatic Center (NCC) in Asheville, North Carolina  (Figure
1).  A standard utility program developed by the U.S. Environmental Protection
Agency (EPA) National Computer Center (NCC) called COPIE  may be used to
select 3 years of data from each station during a  3-year  period (the same
time period used for air quality data).  Each 3-year set  of meteorological
data is placed into a separate mass storage file.  Program  WIND-DIRECT,
developed by PEI, is used to read each of these files and calculate a
resultant wind direction for each day of the year.  A resultant wind
direction is the weighted average of each hourly wind vector (wind
direction and speed) over the 9-hour time period 8:00 a.m.  to 4:00 p.m.  If
the meteorological stations only takes measurements every 3 hours, then
these observations occurring within the same 8-hour period  are  used.   The
UNIVAC Editor is used to combine output files from several  runs of WIND-
DIRECT into one wind direction file for use in MASH.  Further information
on meteorological  data may be found in Section 3.

-------
     The instructions file used in MASH is  created  by  the  UNIVAC  Editor
using information from various documents,  guidelines,  and  analyses produced
by MDAD (Figure 1).   The file containing current  instructions  for all
qualified ozone monitors in the U.S.  is MDAD*INSTRUCT-USA.   More  information
on the instructions  file may be found in Section  4.
     The main MASH program processes  air quality  data  by  (a) determining
which airshed is associated with each hourly  value,  (b) saving the highest
value associated with each airshed during each  day,  and  (c)  determining
whether each saved value is one of the four highest  values  for the airshed
(Figure 1).  The particular value selected  as the design value for the
monitor-airshed combination depends upon whether  the monitor is located in
the source area of the airshed (locally generated ozone) or  in an impact area
of the airshed (transported ozone).  If the ozone is locally generated, then
the design value is  the N + 1 highest observation, where n  is  the number of
valid site-years of  data according to Guidelines  for the Interpretation of
Ozone Air Quality Standards, EPA-430/4-79-003.  If  an  observed ozone concen-
tration is attributable to transport  of emissions from a distant  source
area, it is assumed  that it is that source  area's responsibility  to reduce
the observed concentration to the level  of  the  NAAQS.  This  assumption
holds, even if the observed concentration is  the  highest concentration re-
corded at the particular site.  Ordinarily  such a procedure  has little
effect on the design value associated with  the  distant source  area.  This
follows, because usually the higest 03 concentrations  (recorded anywhere)
which are associated with a source area occur in  relatively  close proximity
to the source area.   As such, they are regarded as  locally  generated, and
the "N + 1" rule described above governs selection of  a design value.
However, occasionally, the highest 03 concentration  attributed to a source
area at a remote site may exceed the  N + 1  highest  concentration  at a local
site attributed to the same area.  Under these  circumstances,  the highest
63 concentration at  the remote site which is  attributable  to the  source
area is assumed to be the design value.
     Design values for monitor-airshed combinations  are placed into an output
file.  Outputs from  a series of MASH  runs may be  combined  into one file

-------
using the Editor,  sorted using  the  utility  @SORT, and placed into an easy-to-
read table using program MASH/HEADINGS.   Further  information is found in
Section 6.
     For the convenience of users,  all program source codes pertaining to
MASH have been placed in program file MDAD*PROG.  This file has a write-key
to protect the programs  from unauthorized changes.  Runstreams to implement
MASH programs are  located in file MDAD*RUN, and elements of this file are
described in the following section.  These  descriptions assume a working
knowledge of the UNIVAC  computer, executive control language (ECL), and
the principal utilities  available at the  NCC.

-------
                                  SECTION 2

                            PROCESSING OZONE DATA


     Processing of ozone data begins by retrieving selected data from the

SAROAD data base and then reviewing these data for validity.   Retrieval  of

hourly data is carried out by program NA205 and validity checks are performed

by program 03-VALID.  Both programs may be executed by a single runstream,

exemplified by element NA205-VALID/TEST1 of file MDAD*RUN,  shown in Figure

2.  This runstream is annotated by comments that follow the symbol  )4.j6 on

most lines of the runstream.  Explanation of this runstream follows:

     Line 1:  The run card, identified by @RUN, should contain an appropriate
     run-Id and account number authorized for the user.  In the example, bin
     number 06, user initials RP, and run series 1A are shown on the  run ID.
     Run series 1A matches the last two characters on the alternate print
     file $03$RP1A and the output file TEST1A used in this  run.  Note that
     run series IB (a different runstream) will execute a MASH analysis  that
     uses the ozone data file TEST1A produced by this" run.   Run elements
     following the account number should be specified as shown in the
     example.

     Lines 2-7:  An alternate print file, $03$RP1A, is established  which can
     be maintained in the computer after the run is completed.  Additional
     copies may be obtained by submitting the command shown on line 7 at a
     demand terminal, substituting the number of copies desired for 'I1  and
     substituting PR for RM95PR if better quality printing  is desired.

     Lines 9-15:  A temporary file called "cards" is established which contains
     data selection specifications for program NA205.   Letter 's' on  line 10
     indicates the type of output desired, in this case a standard  data  file
     (SDF) on mass storage.  Lines 11 through 14 are SAROAD select  cards
     wherein specific fields (columns) give different specifications.  These
     cards MUST be in ascending order by SAROAD code so that  subsequent  files
     will be in proper order.  Columns 1 through 12 indicate  the SAROAD code
     for a monitoring site.  Columns 13 through 17 indicate the pollutant code
     (44201 = ozone).  Columns 18 and 19 indicate the method  code (** =  all
     methods).  Column 20 indicates the interval code (1=1  hour averages).
     Columns 21 through 32 indicate the beginning and ending  dates  of a  3-
     year period from which data is to be selected.  In the example,  data from
     the period January 1, 1981 through December 31, 1983,  are to be  selected.
     Columns 33 through 43 are used only if particular AQCR's, counties, or
     States are to be selected.  The user is advised to select certain

-------
 1:®RUN,R/R    06RP1A,ACCOUNT.MDAD/SA3DP,9,200
 2:aSYM.D      PRINT*.               . DELETE FIRST PRINT FILE.
 3:(SCAT,P      *03*RPlA(i-l).         . NEW CYCLE OF ALT. PRINT FILE.
 4:aCOND       ,                      . PROCESS IN ORDER.
 5:aBRKPT      PRINT*/*O34RP1A       . USE ALTERNATE PRINT FILE.
 £:aTIME                             . RECORD RUN DATE Sc TIME.
 7:aSYM,U      *03*RP1A.,1,RM95PR    . 1 PRINTOUT TO MUTUAL BLDG.
 3:a . THE FOLLOWING CARDS SELECT SAMPLE DATA FROM CALIFORNIA:
 9s«ELT,IL     CARDS                 . TEMPORARY DATA FILE
10:3
11i0524&OOO210144201**1310101331231***********
12'. 055320OO310144201 ** 1310101331231 ***********
13!056320O06101442O1**13101O1331231***********
14!057690001101442O1**1310101331231***********
13:a£OF
17:aCOND
13: SCAT, P
19:aCOND
20:aASG,A
21!<3A30,A
25s 8FREE
2&s8ADD,PL
29J8FREE
30SSMAP.N
31SIN
32SLIB
33!END
34:aASG,A
3£:SDELETE,C
37:aCOND
33:SCAT,P
39!3COND
4l:£U3E
42S-4XQT
43:SBRKPT
 SDF-FILE.

 SDF-FILE..F///999
                       •
 SDF-FILE.
 NADB*NADB-ND-INDX.
 NADB*NADB-ND-BULK.
 NADB*NADB-RUN/SDS.
 NADB*NADB-RUN.NAR205  .
 NADB*NADB-RUN.
 NAR205
 PROG.
 PROG.03-VALID,03-VALID
 PROG.
,TPF*.03-VALID
 TPF*.03-VALID
 LIB*PL1.

 SDF-FILE.
 VAL ID03, SDF-FILE.
 TE3T1A.

 TEST1A..F///999

 TEST1A.
 VALIDOS,TEST1A.
 03-VALID
 PRINT*
REUSABLE OUTPUT FILE.
PROCESS IN ORDER.
LARGE FILE SOMETIMES NEEDED.
PROCESS IN ORDER.
OUTPUT FILE FOR NA205
SAROAD INDEX FILE - 1930+
3AROAD DATA FILE — 1980+
FILE OF NADS RUNSTREAMS.
COPY ONE RUNSTREAM.
FREE FILE FOR OTHER USERS.
START NA205 RUNSTREAM.
FILE FOR MASH SOURCE CODES.
. COMPILE PROGRAM.
FREE FILE FOR OTHER USERS.
CREATE AN ABSOLUTE IN TPF*.
COMPILED RELATIVE ELEMENT.
PL1 LIBRARY SUBROUTINES.
END OF MAPPING INSTRUCTIONS.
OZONE DATA FROM PROGRAM NA205.
INTERNAL NAME USED IN 03-VALID.
CLEAR OUTPUT FILE NAME.
PROCESS IN ORDER.
LARGE FILE NEEDED SOMETIMES.
PROCESS IN ORDER.
OUTPUT FOR PROGRAM 03-VALID.
INTERNAL NAME USED IN PROGRAM.
EXECUTE PROGRAM 03-VALID.
END ALTERNATE FILE PRINTING.
END OF ALL PROCESSING.
      Figure 2.   A runstream for processing ozone data,

-------
     monitoring sites rather than certain areas  because  airshed  allocation
     instructions are only available for urban or downwind  sites.   Limit  the
     number of selection cards to 50 or less.  Since  hundreds  of ozone  moni-
     tors have been established,  a series of runs are required to  process all
     ozone data in the U.S.

     Lines 16-20:  An output data file for NA205.  The file name is  deleted
     in case it was used by a previous run,  then recatalogued.   If the  current
     run bombs, the output file may still be examined.

     Lines 21-22:  The SAROAD files containing recent air quality  data  are
     assigned.  Historical data (prior to 1980)  may be obtained  from an
     analagous set of files (NADB*NADB-HD-INDX and NADB*NADB-HD-BULK).   If
     needed, two lines should be  added to assign these two  files to  the run.

     Lines 23-26:  A runstream is copied from an NADB file  and is  dynamically
     started (including execution of program NA205).

     Lines 27-29:  The source code for program 03-VALID  is  compiled  into
     assembly code and a relative element is created  in  TPF$.

     Lines 30-33:  The assembly code routines, including PL1 library routines,
     are mapped so they can be executed.  The resulting  absolute element  is
     placed into temporary file TPF$.   03-VALID  is a  program that  is executed
     infrequently; therefore, computer storage costs  can be reduced  by  saving
     only the source code element between executions.

     Lines 34-35:  The ozone file from NA205 is  reassigned  and given an
     alternate name used in program 03-VALID.

     Lines 36-41:  An output file is established for  program 03-VALID and
     an equivalent internal  name  is provided.

     Line 42:  Program 03-VALID is executed  using files  named  in this run-
     stream.

     Line 43:  End of printing to alternate  print file.

     Line 44:  End of run.
Use of the Runstream:

     1.   Use the UNIVAC editor to copy this  runstream  to  a  new  element of
          the file,  e.g., MDAD*RUN.NA205-VALID/GIPSON3A.

     2.   In this new  runstream,  change the run  card  to show the desired  bin
          number, user initials,  and run series.

     3.   Change the name of the alternate print file,  $03$RP1A,  to  reflect
          the correct  user initials and run series  (e.g.,  $03$JG3A).

-------
4.   Change the name of the valid hourly ozone file, TEST1A, to reflect
     the run series (e.g, OZONE-3A).

5.   Change the SAROAD select cards to reflect the monitors and years of
     data desired.  A limit of 50 cards is recommended for a single run.

6.   Free file MDAD*RUN and start the runstream (e.g., @START
     MDAD*RUN.NA205-VALID/GIPSON3A).

-------
                                  SECTION 3
                       PROCESSING METEOROLOGICAL DATA

     Processing meteorological (met) data involves mounting tapes on UNIVAC,
copying data from tapes to mass storage files, and calculating daily wind
directions.  Table 1 lists cities and dates for which meteorological data
were obtained from the National Climatological Center in 1984.  When the tapes
were delivered to EPA's National  Computer Center at Research Triangle Park,
a foreign tape number was assigned to each one (e.g.,  B10107).
     The second stage is to read data from tape to mass storage by means of
the COPIE processor.  The following statements show how the data may be read
in demand mode:
     @ASG,TO   TAPE1..36N,810107,0
     @ASG,UP   MET-LA.,F///1800
     @US*ER.COPIE,S   CONVER/ANSIF/SDFSEQ.,TAPE1/80/10.192836/131472,MET-LA.
     In the first statement, a met tape is assigned to the run, stating as
equivalent tape name (TAPE1), a tape reading device.(36N), and the foreign
tape name (B10107).  In the second statement, a new mass storage file is
assigned with maximum space of 1800 tracks.  In the third statement, the
COPIE processor is used to copy selected data from tape to mass storage.  Three
directions separated by commas specify the type of operation, the input format,
and the output format.  The type of operation specified is a conversion (CONVER)
from a tape in fixed block format of the American National Standards Institute
(ANSIF) to a sequential standard data file (SDFSEQ).   The input format for
TAPE1 is 80 characters per record with a blocking factor of 10.  The number
of records to be read is 131,472 starting at record 192,836.  The output file
is SDF file MET-LA.
     The next step in processing meteorological data is to use the met data
established on mass storage files to calculate daily resultant wind directions.
A special program, WIND-DIRECT, was written to perform these calculations.
Figure 3 shows a runstream, located in MDAD*RUN.WIND-DIRECT/SAMPLE, that
builds a single file of resultant wind directions from a series of run
                                      10

-------
TABLE 1.  HOURLY METEOROLOGICAL DATA RECEIVED IN 1984 FROM NATIONAL
                          CLIMATIC CENTER
Tape
number
1
2
3
4
5
6
7
8
9
10
11
12
. 13
14
15
16
17
18
19
20
Station
number
14739
94746
14765
14764
14740
14745
93730
14734
13739
14737
14751
13781
93814
93815
94846
94822
14839
14837
23174
23188
City
Boston, MA
Worcester, MA
Providence, RI
Portland, ME
Hartford, CT
Concord, NH
Atlantic City, NJ
Newark, NJ
Philadelphia, PA
Allentown, PA
Harrisburg, PA
Wilmington, DE
Cincinnati, OH
Dayton, OH
Chicago, IL
Rockford, IL
Milwaukee, WI
Madison, WI
Los Angeles, CA
San Diego, CA
Begin
date
1/1/45
1/1/49
1/1/48
1/1/48
1/1/49
1/1/48
6/1/58
1/1/48
1/1/41
1/1/48
1/1/48
1/1/48
1/1/48
1/1/48
11/1/58
1/1/51
1/1/48
1/1/48
1/1/47
1/1/48
End
date
12/31/83
12/31/83
12/31/83
12/31/83
12/31/83
12/31/83
12/31/83
12/31/83
12/31/83
12/31/83
12/31/83
12/31/83
12/31/83
12/31/83
12/31/83
12/31/83
12/31/83
12/31/83
12/31/83
12/31/83
                                 11

-------
 l!SRUN,D/R
 06RPOO/75, ACCOUNT, MDAD/'3ASDD, 60, 500
 '3: SCAT, P
 4:.

 PRINT*/*03*RPOO

 4034RPOO.,1.RM93PR
 PROG.
 PROG.WIND-DIRECT,WIND-DIRECT
,TPF*.WIND-DIRECT
 TPF*.WIND-DIRECT
 LIB*PL1.

 RUN.MET-FILE3
                         DELETE FIRST PRINT FILE.
                         NEW CYCLE OF ALT. PRINT FILE.
                         PROCESS IN ORDER.
                         USE ALTERNATE PRINT FILE.
                         RECORD RUN DATE b TIME.
                         1 PRINTOUT TO MUTUAL BLDG.
                         FILE OF SOURCE CODES.
                                 COMPILE INTO TPF*.
                       . CREATE ABSOLUTE IN TPF*.
                       . RELOCATABLE IS IN TPFs.
                       . LIBRARY OF PL1 ROUTINES.
                       . END MAPPING INSTRUCTIONS.
                       . STATS ON ALL MET FILES.
13:a . REPEAT THE FOLLOWING SECTION FOR EACH MET STATION:
17:@USE
13:aDELETE,C
19:SCOND
20: SCAT, P
21:<5COND
22:aASG,A
23:aiJSE
24:2XQT
25: YEARS- 1981,
26: ALLENTOWN
27:a£D,U
23s ADD
29: EX IT
30: i2 .  REPEAT
31 s eftSC, A
32s«USE
33:<5D£L£TE,C
34:<»COND
33s SCAT, P
36:aCOND
37:@ASG, A
40: YEARS= 19:31
41: SAN DIEGO
42:<*ED,U
43: ADD
44:EXIT
45:<2BRKPT
43SLNP1
4'?:aFIN
                         HOURLY MET DATA FILE.
                         . NAME USED BY WIND-DIRECT.
                         CLEAR THIS FILE NAME.
                         PROCESS IN ORDER.
                         RECREATE SAME FILE NAME.
                         PROCESS IN ORDER.
                         ASSIGN MORE FILE SPACE.
                         NAME USED BY WIND-DIRECT.
                       . EXECUTE PROGRAM.
                 N_3TATION3=01   . INPUT DATA
                                 . INPUT DATA
                       . FILE ALREADY CATALOGUED.
                       . ADD OUTPUTS FROM THIS RUN.
                       . END OF EDITING.
THE FOLLOWING SECTION FOR EACH MET STATION:
 STDS*MET-SANDI.       . HOURLY MET DATA FIL£.
 METDATA,3TDS»MET-SANDI  . NAME USED BY WIND-DIRECT.
                         CLEAR THIS FILE NAME.
                         PROCESS IN ORDER.
                         RECREATE SAME FILE NAME.
                         PROCESS IN ORDER.
                         ASSIGN MORE FILE SPACE.
                         NAME USED BY WIND-DIRECT.
                       . EXECUTE PROGRAM.
                 N_STATIONS=01   . INPUT DATA
                                 . INPUT DATA
                       . FILE ALREADY CATALOGUED.
                       . ADD OUTPUTS FROM THIS RUN.
                       . END OF EDITING.
                       . END PRINTING TO ALT. FILE.
                       . NEXT FILE PRINTED AT RTP.
                       . REVIEW OUTPUTS CREATED.
                       . PRINT ALL LINES.
                       . END OF RUN.
 STDS*MET-ALLEN.
 METDATA,STDS*MET-ALLEN
 WINDS-ALLEN.

 WINDS-ALLEN.

 WINDS-ALLEN. , F///99-?  .
 RESULTS,WINDS-ALLEN
 WIND-DIRECT
1932,19133
        14737
 WINDS-NE2.
 WINDS-ALLEN.
 WINDS-SAND I.
 WINDS-SAND I.
 WINDS-SAND I. ,
 RESULTS , W I NDS-SAND I
 WIND-DIRECT
1932, 19S3
        14737
 WINDS-NE2.
 WINDS-SAND I.

 PRINT*
 PRINT*, l.PR
 WINDS-NE2.
   Figure  3.   A runstream  for calculating  daily  resultant
               wind  directions.

-------
executions.   Each run execution uses one data  file from one met station.

Explanation  of the runstream follows:

     Line 1:  The @RUN card is similar to the  cards used previously,  with
     bin number and user's initials shown in the run-ID.

     Lines 2-7:  Establishes an alternate print file and causes one copy  to
     be printed at the Mutual  building.

     Lines 8-13:  Compiles and maps the program (WIND-DIRECT)  into  an element
     of the  temporary file for the run (TPF$).

     Line 14:  Prints statistics (e.g.,  file size, data created,  and  disc
     usage)  for each meteorological data file  created in 1984.

     Lines 15-29:  Executes program WIND-DIRECT for a sample meteorological
     station, Allentown,  Pennsylvania.  First  the met file for Allentown  is
     assigned to the run  and given an  internal  name used by the program.
     Then an output file  is established and given an internal  name.   The
     program is executed  using these files and the input data  following the
     @XQT statement.  Years of data to use in  calculations are indicated  by
     three 2-digit numbers in  line 25, columns 9-10, 14-15, and 19-20.  Number
     of met  stations (columns  40-41) should always be '!'.  Other characters
     in the  line are to help the user, but are not read by the program.   The
     next line shows the  name  of the city (columns 1-12) and the  ZIP  code
     (columns 20-24) in which  the met  station  is located (Table 1).   After
     the execution, an already catalogued data file is edited  by  adding to
     it the  data file from this run.

     Lines 30-44.  Executes program WIND-DIRECT for a second sample meteo-
     rological station, San Diego, California.

     Lines 45-49.  Ends the alternate  print file, then lists the  contents of
     the combined output  file  in a standard print file.   This  standard file
     is printed at a high quality printer, then the run is ended.

To Use the Runstream:

     1.   Catalogue a file to  receive  all  daily resultant wind directions over
          a  3-year period.  Use of a removable disk pack should ensure that
          it is retained  on the system rather  than being automatically re-
          moved after 60  days  (e.g, @CAT,P RESULTS-USA.,F33,NCC075).

     2.   Copy the runstream to a new  file element (e.g., WIND-DIRECT/1985).

     3.   Use the editor  to change the run card to reflect the desired bin
          number and user's initials.   Change  the name of the  alternate print
          file to reflect user's initials.

     4.   Delete this line or  create an analagous file element (other than
          MET-FILES) that shows actual input files to be used.
                                      13

-------
5.   Change the name of input and output files in lines  15-29 to indi-
     cate the correct met stations.   Indicate the correct 3-year period
     following @XQT, maintaining column alignment.  Indicate the correct
     station name and ZIP code, maintaining column alignment.

6.   Use the correct combined data file name in lines  27 and 47.

7.   Delete lines 30-44, then ditto  lines 15-29 one or more times.   In
     jeach ditto'd section, name a different met file to  use in each
     execution.  Note that the present version of MASH is limited to 20
     met stations.

8.   Exit and free the file.   Start  the runstream (e.g., @START MDAD*RUN.
     WIND-DIRECT/1985).
                                 14

-------
                                  SECTION 4
                     ESTABLISHING A FILE OF INSTRUCTIONS

     The i-nstructions file is used by the MASH program to allocate hourly
ozone values to airsheds.  This file consists of a series of records whereby
each record shows a range of times of day and a range of resultant wind
directions that associate a particular monitor with a particular airshed.
The format of the instructions file, MDAD*INSTRUCT-USA, is shown in Figure 4.
     The first item on each record is the SAROAD code for the monitor
(columns 1 to 12).  The second datum is the area name (county or city) in
which the monitor is located.  This is the name corresponding to the SAROAD
4.-digit area code within the 12-digit monitor code.  Columns 29-32 show the
Federal Information Processing Standards (FIPS) code'for the Standard Metro-
politan Statistical AREA (SMSA) in which the monitor is located.
     The next two items are the names of two meteorological  stations (trun-
cated to eight characters) from which wind data are to be used for this moni-
tor.  If data from only one station is to be used, then the  second name is
filled with asterisks.  If no wind data are to be used, then both station names
are filled with asterisks and all  hourly ozone values will be allocated to one
local airshed named on this record.  Only one instruction (record) should
exist for a monitor if no wind data are used.
     Instructions for each monitor are given in two 2-digit  integers and
two 3-digit integers.  The first pair of integers (columns 52-53 and 55-56)
indicate the range of hours of the day (0 to 24).  The second pair of integers
(columns 58-60 and 62-64) indicate the resultant wind direction for the day
in compass degrees (0 to 360).  Note that the first integer  must always be
less than the second integer in order for the instructions to be processed
properly.  For example, a wind direction range from 270° to  10° should be
represented by two different records with ranges 270° to 360° and 0° to 10°.
In cases where allocation instructions do not cover all possible conditions,
values not covered will be automatically allocated to a dummy airshed called
"discard."
                                      15

-------
          CODED INSTRUCTIONS FOR MONITORING SITES
                                                             Paae 25
Monitor
221210001F01
2212200O3F01
221 39200 1A05
221520003F01
221 79200 1F01
221300005F01
22130O007F01
221330002F01
222100003F01
2221 600 14F01
2221 '-'600 1F01
22225200 1F01
22236000 1F01
222330005F01
2226400 12F01
22 26400 15F01
Monitor
Ci tv
MEDFIELD
MEDFORO
MONTAGUE
NEWBURyPORT
PEPPERELL
PITTS-FIELD
PITTS-FIELD
QUIIMCY
SOMERVILLE
SPRINGFIELD
SUDBURY
TEWH SBURY
WARE
WATERTOWN
WORCESTER
WORCESTER
Monitor MET Station
SMSA 1 2
1120
11 2O
9099
9999
99^9
6320
6320
1120
1120
30OO
1120
4560
.3900
1120
"'240
•"'24O
HARTFORD
****»**<,
HARTFORD
****•»•»**
*#•******
***•»»•»••»*
**#*•*#**
BOSTON
BOSTON
HARTFORD
BOSTON
BOSTON
HARTFORD
BOSTON
WORCESTE
WORCESTE
— -
**•»•»*•»#*
********
******#•»
*********
*•»*#*•*•»*•
********
PROVIDEN
PROVIDEN
********
PROVIDEN
PROVIDEN
********
PROVIDEN
**•»•»-»*»*•
........
Wind Airshed
Hours Direct SMSA
11-15
16-24
0-24
0-24
11-15
0-24
0-24
0-24
0-24
13-24
0-24
13-24
0-24
14-13
13-24
11-15
11-15
13-24
0-24
13-2-1
0-24
16-24
11-15
13-24
0-24
0-24
16-24
11-15
16-24
0-24
11-15
130-230
130-270
0- 360
0-360
160-200
0- 360
0-360
0- 360
0-360
130-250
0-360
130-250
0-360
130-270
130-270
150-210
1 30-230
130-250
0- 360
130-250
0-360
130-250
230-260
130-250
0—360
30— 130
130-250
0- 360
130-250
30-130
0—360
TRANS
TRANS
LOCAL
LOCAL
LOCAL
TRANS
LOCAL
LOCAL
LOCAL
TRANS
LOCAL
TRANS
LOCAL
TRANS
TRANS
LOCAL
TRANS
TRANS
LOCAL
TRAMS
LOCAL
TRANS
LOCAL
TRANS
LOCAL
TRANS
TRANS
LOCAL
TRANS
TRANS
LOCAL
6430
5600
1120
1120
3000
1120
1120
6320
6320
5600
1120
5600
1120
C OWN
5600
3000
6430
5600
1120
5600
1120
5600
3000
5600
1120
1 1 20
5600
•"'240
5600
1120
•5240
Figure 4.  Sample  records (lines) from the  instructions file.
                               16

-------
     Columns 66-70 indicate whether the  airshed allocation in this  record was
based on  ozone precursers  transported from a distant SMSA (TRANS) or  was
generated within the local  SMSA (LOCAL).   Columns 72-75  indicate the  FIPS
code of the  SMSA of the airshed.
     The  instructions file  may be modified by changing one of the records,
removing  records, or adding new records.   The user may need to review the con-
tents of  the file in a easy-to-read table  with appropriate headings.   Such a
table may be generated by  using the runstream MDAD*RUN.PRINT/INSTRUCT shown
in Figure 5.  Change the run - ID to reflect the correct bin number and  user
-ID, then submit an @START  command at the  terminal.
         l:SRUN,R/R
         3: SCAT, P
         4 ; @COND
         6:STIME
         7: .           . NEW CYCLE OF ALT.  PRINT FILE.
                      . PROCESS IN ORDER.
 PRINT*/*03SRP         . USE ALTERNATE PRINT FILE.
                      . RECORD RUN DATE &  TIME.
 *03*RP. , 1, RM9'3FR      . 1  PRINTOUT TO MUTUAL BLDO.
 PROi3.                 . PROGRAM SOURCE CODE FILE.
 PROO.PRINT/INSTRUCT,PRINT/INSTRUCT  .  COMPILE PROGRAM.
 PROG.
,TPF*.PRINT/INSTRUCT
 TPF*.PRI NT/INSTRUCT
 LIB*PL1.

 INSTRUCT-USA.
 INSTRUCT,INSTRUCT-USA.
 PRINT/INSTRUCT
 PRINT*
FREE FILE FOR OTHER USERS.
CREATE ABSOLUTE  IN TPf*.
RELATIVE LOCATED IN TPFS.
PL1 SUBROUTINE LIBRARY.
END MAPPING  INSTRUCTIONS.
FILE TO BE PRINTED.
.  NAME USED  IN PRINT/INSTRUCT.
EXECUTE PRINTING PROGRAM.
END PRINTING TO ALT.  PRINT  FILE.
END OF RUN.
           Figure 5.  A runstream for printing the instuctions  file,
                                         17

-------
     Proper processing  by MASH requires that records  within the instructions
file be in the proper order.   MASH searches for an  instruction with a monitor
code matching a monitor in the ozone file.  If the  program finds a monitor
code higher than  the ozone file monitor, the search is  stopped and a message
is printed stating  that no instruction was found  for  this  monitor.  This
technique works only if records within the instruction  file are sorted in
ascending order of  SAROAD code.
     Within each  SAROAD code, records should be sorted  in  descending order of
transport value (TRANS, then  LOCAL).  Instructions  allocating all  transported
ozone are read and  used before any instructions for the local airshed.  Once
a value is allocated, no other instructions are used  because only one air-
shed is allocated each  hour.   This protocol saves space in the file because
fewer local records are required.  For example, if wind directions 60° to 180°
represent transported 03 and  all  other directions are local, then only one
local record is required for  directions 0 to 360  rather than 0 to 59 and
181 to 360.  Any  wind direction 60° to 180° will  cause  the first instruction
to be activated and the second one ignored.  Any  wind direction outside this
range will cause  the first instruction to be bypassed and  the second instruc-
tion utilized.
     The instructions file may be sorted in proper order by using the partial
runstream MDAD*RUN.SORT/INSTRUCT, as shown in Figure  6.  No change in the run-
stream is required; simply submit @ADD,PL MDAD*RUN.SORT/INSTRUCT in demand
mode.
          2:i2DELETE,C
          3:<2COND
          4:<»CAT,P
           6:fiASG. A
           7:@USE
           3:3USE
INSTRUCT-USA.
*O2*OUT.

*02*OUT.

*O2*QUT.
INPUT.,INSTRUCT-USA.
OUTPUT.,*O2*QUT.
          K>:FILEIN= INPUT
          1 1 = FILEOUT=OUTFUT
          12:MODE=SDF
          1 4: DAT A= ASCI I
          15:KEY=1, 1.2,Si
          16:RECORD=2
          17:i2EOF
             COL  1-12
FILE TO BE SORTED.
CLEAR OUTPUT FILE NAME.
PROCESS IN ORDER.
NEW OUTPUT FILE.
PROCESS IN ORDER.
TO BE USED BY SSORT.
INTERNAL NAME FOR ©SORT.
INTERNAL NAME FOR (2SORT.
CALL PROCESSOR.
ANOTHER INTERNAL NAME.
ANOTHER INTERNAL NAME.
INPUT IS STANDARD FILE.
RECORD SIZE = 132 CHAR.
INPUT FILE CODING.
S=ALPHANUMERIC» A-ASCENDINO.
2000 RECORDS < OR LESS >.
END SORT INSTRUCTIONS.
      Figure 6.  A  partial  runstream for sorting the  instructions file.
                                       18

-------
                                  SECTION 5
                         OPERATING THE MAIN PROGRAM

     Operation of the main program of MASH requires the assignment of four
input data files and two output files.
     Input files for MASH include a file of resultant daily wind directions
from selected meteorological  stations and a file of instructions on how to
allocate hourly ozone values  to various airsheds based on wind directions.
Note that the meteorological  stations named in the instructions file must
match the stations in the wind directions file.   As of this date, file
MDAD*WINDS-USA contains wind  data for 20 meteorological  stations across the
U.S. for all  days from 1981 through 1983.  These stations match the stations
named in file MDAD*INSTRUCT-USA.  Other input files include a  small, temporary
file of run options and a file of valid hourly ozone data from program 03-VALID.
     Outputs are 1) a file of design values for each airshed monitor combina-
tion, and 2)  a file of nonzero hourly values that were not allocated to any
airshed (discards).  The design value file is called a "raw" value file with
one line of data for each design value.  These outputs are to  be combined
with other files and sorted before being placed in a formal table (Section
6).  The "discard" file lists hourly values from a monitor not assigned to a
specific airshed.  These discard values may be useful  in expanding the in-
structions for future runs.
     Operation of the main program is exemplified in element MASH/TEST1 of
file MDAD*RUN as shown in Figure 7.  Functions of this runstream are described
in the comments on each line  and in the following:
     Line 1;   The @RUN card indicates a routine batch run that can operate
     during normal operating  times of the NCC.  The run identification, 06RP1B,
     indicates bin number 6,  user initials RP, and run series  IB.   Note that
     run series IB is intended to use the ozone date file TEST1 from 03-VALID
     created by runs series 1A (Section 2).  The main program  requires about
     65,000 words in core, indicated by 65 on the run card. An appropriate
     account number may be inserted.  Other elements of the run card must
     remain as shown, except  that more than 9 minutes could be required for
     a very large ozone file.
                                      19

-------
 1:<2RUN,R/R
 2:<»3YM,D
 3: SCAT, P
 4:QCOND
 •5: SBRKPT
 06RPIB/65, ACCOUNT,MDAD/3ASDP,9,200
 7:aSYM,U
 3:SA3G,A
 9:eCOPY,A
10:aFREE
13:<2ASi3,A
14:aUSE
13:eASG, A
17:aASC,T
19:N_STATIQNS=
20: YEAR1 = 19:31
2 1 : (2EOF
22:®DELETE,C
23 SECOND
24: SCAT, P
25:SCOND
26:eASG, A
27:@USE
23:@DELETE,C
29:@COND
30: SCAT, P
3 1 : @C OND
32:<§A3G,A
33:@U5E
34:. TIME.
                        1 PRINTOUT TO MUTUAL BLDG.
                        FILE FOR MASH SOURCE CODES.
                        COPY ABSOLUTE TO TPF*.
                        FREE FILE  FOR OTHER USERS.
                        WINDS ®  20 MET.  STATIONS.
                        INTERNAL NAME USED  IN MASH.
                        03 DATA  FROM  NA205.
                        INTERNAL NAME USED  IN MASH.
                        AIRSHED  ALLOCATION  INSTRUCTIONS,
                        . INTERNAL NAME  USED IN MASH.
                        TEMPORARY  DATA FILE.
                        PLACE DATA INTO  TEMP.  FILE.
                              PRINT_WINDS=YES
                           EOL
                        END OF TEMPORARY DATA FILE.
                        CLEAR OUTPUT  FILE NAME.
                        PROCESS  IN ORDER.
                        RECATALOGUE SAME NAME.
                        PROCESS  IN ORDER.
                        OUTPUT FILE ASSIC-NED TO RUM.
 DESIONVALUES,  RAW-VALUESl.
                               INTERNAL FILE  NAME.
TEST 1-DISC.

TEST1-DISC.

TEST1-DISC. ,F///999
DI3CARDS,TEST1-DISC.

MASH
RAW-VALUES 1.

TEST1-DISC.
1 50
PRINT*
                         CLEAR 2ND OUTPUT FILE NAME.
                         PROCESS IN ORDER.
                         RECATALOGUE SAME NAME.
                         PROCESS IN ORDER.
                         ASSIGN WITH LARGER SPACE.
                         INTERNAL NAME USED IN MASH.
                         SET CORE ACCOUNTING FOR MASH.
                         EXECUTE PROGRAM MASH.
                         REVIEW OUTPUT FILE #1.
                         PRINT ALL LINES.
                         REVIEW OUTPUT FILE #2.
                         PRINT SAMPLE — 50 LINES.
                         END PRINTING TO ALT.  PRINT FILE.
                         END OF RUN.
   Figure 7.   A runstream for  executing the main MASH program.
                                   20

-------
     Lines 2-5:   Establishes an alternate print file so  that results  of the
     run may be  reviewed after run completion,  even  if an  error is  encountered.

     Line 6:  The first data printed in  the  alternate print  file will  be the
     date and time of the run.

     Line 7:  Causes one copy of the alternate  print file  to be printed at  the
     Mutual  Building printer.

     Lines 8-10:  Copies an absolute file element of MASH  to the run's
     temporary file (TPF$) for execution.

     Lines 11-16:  Assigns three input files to the  run  and  assigns internal
     names used  by MASH.

     Lines 17-21:  Establishes a temporary data file giving  certain run options
     to the program.

     Lines 22-23:  Establishes two output files and  provides internal  names
     for use in  MASH.

     Line 34: Sets a value in a portion of  the condition  word  associated
     with each run in UNIVAC.  Option 0  was  established  by NCC  to turn on an
     accounting  system during the execution  of  a program.  At the end  of the
     execution (either normal or error)  a statement  is printed  showing the
     standard units of processing (SUPS) and core space  used.

     Line 35: Executes program MASH using the  files named in this  run.

     Lines 36-39:  Prints certain lines  of the  output files  so  the  user may
     check the run's success.

     Line 40: Signals end of printing to the alternate  print file.

     Line 41: End of run.
Use of the Runstream:

     1.   Copy this  runstream to  a  new element  of  the  file  (e.g.,
          MDAD*RUN.MASH/1985-IB).

     2.   Change the run  card to  show the  desired  bin  number,  user  initials,
          and run series.

     3.   Change the name of the  alternate print file,  $03$RP1B, to  reflect
          the correct user initials and run series.

     4.   Change the name of the  valid ozone data  file, TEST1,  to reflect  the
          correct file to be used.
                                     21

-------
Check the name and contents
to ensure that instructions
ozone data file.
of the instructions file, INSTRUCT-USA.,
exist for the monitors named in the valid
Check the name and contents of
to ensure that wind data exist
instructions file.
   the wind directions file, WINDS-USA.
   for all stations referenced in the
Change the runstream options in file RUNINPUT as follows:

(a)  Keep the maximum number of stations (columns 12-13) at 20,  the
     present capacity of MASH.

(b)  Limit the maximum number of monitors to be processed  (line
     19, columns 30-32) in one run to 50, a  limit that keeps file
     sizes and run costs to a reasonable level.

(c)  If a list of wind data is to be printed in the alternate print
     file, maintain the value of print-winds as "yes"  in line 19,
     columns 55-57.  Otherwise change to "no."

(d)  Indicate year No. 1, year No.  2, and year No.  3 of the hourly
     data file by entering three two-digit numbers  beginning in  line
     20, columns 10, 22, and 34.  Other characters  on  lines 19 and
     20 are comments that help maintain column alignment.
Exit and free file MDAD*RUN.
MDAD*RUN.MASH/1985-IB).
  Start the runstream (e.g., @START
                            22

-------
                                  SECTION 6
                           FORMATTING THE OUTPUTS

     When using the MASH model,  a series of valid ozone data files  are created
to incorporate all  ozone monitors in the U.S.   These valid ozone data are
processed by a series of MASH runs that produce a series of design  value files.
The final step is to combine all design value  files into one file,  sort this
file by design value within each airshed, and  print design values in an easy-
to-read table with  column headings.   A runstream for carrying out this process
located in MDAD*RUN.MASH/HEADINGS, is shown in Figure 8.
     Line 1 is a run card similar to other runs shown in this guideline.
Lines 2-7 establish an alternate print file with one copy to be printed at
the Mutual Building.  Lines 9-13 establish a temporary file and add together
all design values into the file.  Lines 14-18  establish an output file for
the UNIVAC sorting  utility, @SQRT.  Lines 21-29 provide instructions to @SORT
for sorting the input file by airshed and by design value.  In lines 30-31 a
sample of the output file, RAW-SORTED, is printed for quality assurance check-
ing.  In lines 32-37 a special program, MASH/HEADINGS, is compiled  and mapped
so it can be executed.  Lines 38-43  establish  an output file for this program.
Lines 44-45 assign  an input file consisting of FIPS codes and names for all
SMSA's, including special names  used only in this program.  The SMSA file is
needed because the  design value  table uses SMSA names as names of airsheds
rather than using SMSA codes.  Line  46 established a name used in MASH/HEADINGS.
Line 47 turns on the accounting  system.  Line  48 executes the program.  Lines
48-49 prints the output file for checking.  Line 51 terminates printing to the
alternate print file.  Line 52 terminates the  run.
Use of the Runstream:
     1.   Use the UNIVAC editor  to change the  bin number, initials, and
          account number in the  run  card as needed.
     2.   Change the name of the alternate print file to match the  user's
          initials.
                                      23

-------
    l:»RUNiR/R
                  06RP5,ACCOUNT,MDAD/3ASDP, 4, 200
                  PRINT*
3: SCAT, P
4: aCOND
5:fiBRKPT
 7:<23YM,U
 3:@ASG,T
 9:©ED, IQ
10:
11:ADD
12:ADD
13:EXIT
14:<2DELETE,C
                  PRINT*/*034RP5

                  *0.3*RP5. , , RM'P'SPR
                  RAW-VALUES.
                  RAW-VALUES.
                                       DELETE FIRST PRINT FILE.
                                       NEW CYCLE OF ALT. PRINT FILE.
                                       PROCESS IN ORDER.
                                       USE ALTERNATE PRINT FILE.
                                       DATE S< TIME OF THIS RUN.
                                       1 PRINTOUT TO MUTUAL  BLDO.
                                       TEMPORARY FILE FOR THIS RUN.
                                       PUT DATA INTO TEMP. FILE.
   16:8CAT,P
   17:<2COND
   13:@A3G, A
                  RAW-VALUES1.
                  RAW-VALUES2.

                  RAW-SORTED.

                  RAW-SORTED.
    31JLNP  1  100
   33:fiPLl,SE
   :-!4:<*MAP,N
   35: IN
   36: LIB
   37: END
   33:<20ELETE
                                     . ADD  DATA  FROM  A  MASH RUN.
                                     . ADD  DATA  FROM  A  MASH PUN.
                                     . END  OF  NEW  DATA.
                                     . CLEAR THIS  FILE  NAME.
                                     . PROCESS IN  ORDER.
                                     . RECREATE  SAME  FILE NAME.
                                     . PROCESS IN  ORDER-.
                                     . OUTPUT  FILE FOR  asORT.
                                     . INTERNAL  NAME  FOR  2SORT.
                                     . INTERNAL  NAME  FOR  (5SORT.
                                     . CALL SORTING UTILITY PROGRAM.
                                     . ANOTHER INTERNAL FILE NAME.
                                     . ANOTHER INTERNAL FILE NAME.
                                     . INPUT  IS  STANDARD  FILE.
                                     . RECORD  SIZE =  132  CHAR.
                                     . FILE CODING'IS ASCII.
                            SORT  BY  AIRSHED SM3A,  THEN DESIGN  VALUE.
                                     . 40OO RECORDS lOR LESS)
                                     .'END  aSORT INSTRUCTIONS.
                                     . CHECK-SORTED FILE.
                                     . PRINT  10O LINES.
              PROG.                  . FILE OF SOURCE CODES.
              PROG.MASH/HEADINGS,MASH/HEADINGS   .  COMPILE.
               RAW-SORTED.
               INPUT..RAW-VALUES.
               OUTPUT..RAW-SORTED
21:<23GRT,ES
22:FILEIN=INPUT
23:FILEOUT=OUTPUT
24:MODE=SDF.
25:RSZ=132
26:DATA=ASCII
27: KEY=i , 4, S, A: 49,5, S, D
23:RECORD=4
                  RAW-SORTED.
   4O:2CAT,P
   4USCOND
   42:@ASG,A
   43:aU3E
   44 : (2A3G , A
   47:<*SETC,0
   50: P1
   5l:«BRKPT
   52:<»FIN
              iTPF*.MASH/HEADINGS
              TPF*.MASH/HEADINGS
              LIB*PL1.

              DV-S1-33.

              DV-S1-33.

              DV-31-S3.
              DESIGNV ALUES •,  DV-31 -33.
              SMSA-NAMES.
              SMSANAMES.SMSA-NAMES. .
              RAWSGRTED,RAW-SORTED. .

              MASH/HEADINGS
              DV-31-33.

              PRINT*
                                       CREATE ABSOLUTE  IN  TPF4.
                                       RELATIVE LOCATED  IN TPF*.
                                       PL1 SUBROUTINE LIBRARY.
                                       END OF MAP  INSTRUCTIONS.
                                       CLEAR THIS  FILE  NAME.
                                       PROCESS  IN  ORDER.
                                       RECREATE THIS FILE  NAME.
                                       PROCESS  IN  ORDER.
                                       OUTPUT FILE FOR  MASH/HEADINGS.
                                         .  INTERNAL NAME  USED.
                                       FILE OF SM3A CODES  ?< NAMES.
                                       NAME USED BY MASH/HEADINGS.
                                       NAME USED BY MASH/HEADINGS.
                                       TURN ON CORE ACCOUNTING  SYSTEM.
                                       EXECUTE HEADINGS  PROGRAM.
                                       CHECK OUTPUT FILE.
                                       PRINT ENTIRE FILE.
                                       END PRINTING TO  ALT.  PRINT FILE.
                                       END OF RUN.
Figure 8.  A  runstream for combining  outputs  into a  formatted table.
                                       24

-------
3.   Changes names of MASH files in lines 11-12 as needed to reflect
     actual file names that were created.  Add more lines if needed.

4.   Exit from the file and start the runstream (e.g., (PSTART
     MDAD*RUN.MASH/HEADINGS).
                                 25

-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing}
1. REPORT NO. 2.
EPA-450/4-85-009
4. TITLE ANDSUBTITLE
User's Guide For The Multiple Airshed (MASH) Model
7. AUTHOR(S)
Roy A. Paul
9. PERFORMING ORGANIZATION NAME AND ADDRESS
PEI Associates, Inc
Durham, NC 27701
12. SPONSORING AGENCY NAME AND ADDRESS
Office Of Air Duality Planning And Standards
US EPA (MD 14)
Research Triangle Park, NC 27711
3. RECIPIENT'S ACCESSION NO.
5. REPORT DATE
July 1985
6. PERFORMING ORGANIZATION CODE
8. PERFORMING ORGANIZATION REPORT NO.
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
68-02-3512
13. TYPE OF REPORT AND PERIOD COVERED
14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
EPA Project Officer: Keith Baugues
16. ABSTRACT
      Multiple airshed (MASH)  model calculates design values for each of a series of
ozone monitoring sites in the U.  S.   Each monitor is assigned to various airsheds,
depending on time of day and  the  resultant wind direction for each day.  Using a list
of  high values for each airshed,  one design value is selected for each airshed and
monitor combination.   The design  value associated with a particular SMSA can be
selected from the resulting list  of site specific design values which the program
generates for each airshed.  Input data for MASH are 1) hourly ozone values from
monitors, 2) daily.resultant  wind directions at selected meteorological stations, and
3)  a  coded set of instructions telling how to allocate hourly ozone values to various
airsheds based on wind direction.
17. KEY WORDS AND DOCUMENT ANALYSIS
a. DESCRIPTORS

18. DISTRIBUTION STATEMENT
b. IDENTIFIERS/OPEN ENDED TERMS

19. SECURITY CLASS (This Report)
20. SECURITY CLASS (This page/
c. COSATI Field/Group

21. NO. OF PAGES
34
22. PRICE |
EPA Form 2220-1 (Rey. 4-77)   PREVIOUS EDITION is OBSOLETE

-------
DATE DUE

-------