EPA/600/8-90/083C
JANUARY 1991
THE REGIONAL OXTOANT MODEL (ROM) USER'S GUIDE
PARTS:
THE CORE MODEL
by
J. YOUNG
L. MILICH
D. JORGE
Computer Sciences Corporation
Research Triangle Park, NC 27709
Contract No. 68-01-7365
Technical Monitor
Joan Novak*
Atmospheric Sciences Modeling Division
Atmospheric Research and Exposure Assessment Laboratory
Research Triangle Park, NC 27711
*On assignment bom the National Oceanic and Atmospheric
Administration, U.S. Department of Commerce
ATMOSPHERIC RESEARCH AND EXPOSURE ASSESSMENT LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
RESEARCH TRIANGLE PARK, NC 27711
-------
EPA/600/8-90/083C
JANUARY 1991
THE REGIONAL OXTOANT MODEL (ROM) USER'S GUIDE
PART 3:
THE CORE MODEL
by
J. YOUNG
L. MILICH
D. JORGE
Computer Sciences Corporation
Research Triangle Park, NC 27709
Contract No. 68-01-7365
Technical Monitor
Joan Novak*
Atmospheric Sciences Modeling Division
Atmospheric Research and Exposure Assessment Laboratory
Research Triangle Park, NC 27711
*On assignment from the National Oceanic and Atmospheric
Administration, U.S. Department of Commerce
ATMOSPHERIC RESEARCH AND EXPOSURE ASSESSMENT LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
RESEARCH TRIANGLE PARK, NC 27711
-------
Mention of trade names or commercial products
does not constitute endorsement or recommendation for use.
u
-------
TABLE OF CONTENTS
Disclaimer ii
Table of Contents iii
List of Tables v
List of Figures viii
1 Overview and Structure of the Core Model 1
1.1 Introduction 1
1.2 General Model Characteristics 3
1.2.1 Physical Processes Within Layers 0,1,2 and 3 4
1.2.2 ROM Chemistry 7
1.2.3 System Components 8
1.2.4 ROM Limitations 11
1.3 Data and File Requirements 11
1.4 Model Execution and Storage Requirements for the IBM 3090 13
1.4.1 Starting the ROM at the Top of an Hour 13
1.4.2 CPU Use 13
1.4.3 Storage Requirements 14
1.4.4 Core Model Input Data: Transfer from the VAX to the IBM 14
1.5 Software Components of the Core Model 15
1.5.1 BIGGAM 16
1.5.2 LILGAM 19
1.6 Subprograms That Maintain the State Vector and Restart the Model 22
1.7 Utility Subprograms 22
1.8 Miscellaneous Subprograms 24
1.9 Core Model Output Data: Transfer from the IBM to the VAX 25
1.10 CONC File Quality Control Procedures 25
2 Core Model Input Files 27
2.1 The Boundary Conditions (BCON) File 27
2.1.1 Opening the BCON File 28
2.1.2 BCON File Records 29
2.2 The B-Matrix (BMAT) File 34
2.2.1 Opening the BMAT File 35
2.2.2 BMAT File Records 36
2.3 The Backtrack (BTRK) File 48
2.3.1 Opening the BTRK File 48
2.3.2 BTRK File Records 49
2.4 The Initial Conditions (ICON) File 54
2.4.1 Opening the ICON File 55
2.4.2 ICON File Records 56
111
-------
2.5 The New Initial Conditions (NEWICON) File 60
2.5.1 Opening the NEWICON File 61
2.5.2 NEWICON File Records 62
2.6 The Progress (PROG) File 68
2.6.1 Opening the PROG File 68
2.6.2 The PROG File Record 69
2.7 The State Vector (RESTRT) File 69
2.7.1 Processing That Takes Place for Normal Model Execution (How the RESTRT File Gets
Written) 70
2.7.2 Processing That Takes Place for Restarting Model Execution (How the RESTRT File Gets
Read) 85
2.8 The Stop Check (STOPCK) File 88
2.8.1 Opening the STOPCK File 88
2.8.2 The STOPCK File Record 89
3 The Core Model Concentration (CONC) Output File 91
3.1 Opening the CONC File 91
3.2 CONC File Records 92
3.2.1 CONC File Header Records 93
3.2.2 CONC File Body Records 97
APPENDICES
A Jackson Structured Programming (JSP) A-l
Al An Introduction to JSP Flow Diagrams A-2
A.2 Examples of Program Inversion and the Use of State Vectors A-4
A2.1 Program Inversion with Respect to Its Input Data Stream A-5
A.2.2 Program Inversion with Respect to Its Output Data Stream A-6
A.3 Reference and Bibliography .-. A-6
B Design and Structure Diagrams for the ROM2.1 Data Files in the ROMNET Region B-l
C Design and Structure Diagrams for the Principal ROM2.1 Subroutines C-l
D Sample Compile and Link Stream for the Uniprocessor ROM2.1 D-l
E INCLUDE Files E-l
F Core Model Error Checking F-l
Index Index-1
IV
-------
LIST OF TABLES
1 ROM chemical species 8
2 HEADIN.EXT variables: descriptions and origins 18
3 BCON record 1 variables 30
4 BCON record 2 variable 31
5 BCON record 3 variable 31
6 BCON records 4 - (4+ICNTBM) variable 32
7 BCON time-step header record variables 32
8 BCON data variables 34
9 B-matrix record 1 variables 38
10 B-matrix record 2 variable 38
11 B-matrix record 3 variable 40
12 B-matrix record 4 variable 40
13 B-matrix records 5 - 8 variables 41
14 B-matrix records 9 - (9+ICNTBM-l) variables 41
15 B-matrix subfile-order record variables 42
16 B-matrix time-step header record variables 43
17 B-matrix data variables, part 1 46
18 B-matrix data variables, Part 2 48
19 BTRK record 1 variables 51
20 BTRK records 2 - 5 variables 52
21 BTRK records 6 - (6+ICNTBT-1) variables 52
22 BTRK time-step header record variables 53
23 BTRK data variables '. 54
24 ICON record 1 variables 57
25 ICON record 2 variable ., 58
26 ICON record 3 variable 58
27 ICON records 4 - (4+ICNTIC) variable 59
28 ICON time-step header record variables 59
29 ICON data variable 60
30 NEWICON record 1 variables : 64
31 NEWICON record 2 variable 65
32 NEWICON record 3 variable 65
33 NEWICON records 4 - (4+ICNTCN) variable 66
34 NEWICON time-step header record variables 67
35 NEWICON data variable , 68
36 PROG record variables 69
37 RESTRT record 1 variables 73
38 RESTRT record 2 variables 74
39 RESTRT records 3 and 4 variable 74
40 RESTRT record 5 variable 75
-------
41 RESTRT records 6 - (6+ICNTSV) variable 76
42 RESTRT HEADIN record 1 variables 76
43 RESTRT HEADIN record 2 variables 77
44 RESTRT chemistry control variables 1 78
45 RESTRT chemistry control variables 2 78
46 BMAT species index variable 79
47 BCON species index variable 79
48 ICON species index variable 80
49 Primary oxidant species flag records variable 80
50 Special species record variables 81
51 Diffusivities conversion factor record variables 82
52 ICON time-step header record variables 82
53 Text pointer record variables 83
54 Row counters record variables 84
55 Model and file time step header records 85
56 STOPCK record variable 89
57 CONC record 1 variables 95
58 CONC record 2 variable 96
59 CONC record 3 variable 97
60 CONC records 4 - (4+ICNTCN) variable 97
61 CONC time-step header record variables 98
62 CONC data variable 99
E-l INCLUDE file DIMENS.EXT E-2
E-2 INCLUDE file REGION.EXT E-2
E-3 INCLUDE file ADVSFL.EXT E-3
E-4 INCLUDE file BCFILE.EXT E-3
E-5 INCLUDE file BGBCFUEXT E-3
E-6 INCLUDE file BGBTFL.EXT E-4
E-7 INCLUDE file BGICCN.EXT E-4
E-8 INCLUDE file BMCOEF.EXT E-5
E-9 INCLUDE file BMFILE.EXT E-6
E-10 INCLUDE file BTFILE.EXT E-7
E-ll INCLUDE file CHEMIN.EXT E-7
E-12 INCLUDE file CHEMSW.EXT E-8
E-13 INCLUDE file CNFILE.EXT E-8
E-14 INCLUDE file CONFACEXT E-8
E-15 INCLUDE file ERRG.EXT '. '. E-9
E-16 INCLUDE file FLNAMS.EXT E-9
E-17 INCLUDE file GTCOEF.EXT E-10
E-18 INCLUDE file HDFMTS.EXT E-10
E-19 INCLUDE file HDSTAV.EXT E-ll
E-20 INCLUDE file HEADBC.EXT E-12
VI
-------
E-21 INCLUDE file HEADBM.EXT E-13
E-22 INCLUDE file HEADBT.EXT E-14
E-23 INCLUDE file HEADCN.EXT E-15
E-24 INCLUDE file HEADICEXT E-16
E-25 INCLUDE file HEADIN.EXT E-17
E-26 INCLUDE file HSTEPS.EXT E-18
E-27 INCLUDE file ICFILE.EXT E-18
E-28 INCLUDE file LGBMFL.EXT E-19
E-29 INCLUDE file LILGSP.EXT E-20
E-30 INCLUDE file LUNITS.EXT E-20
E-31 INCLUDE file LVNAME.EXT E-21
E-32 INCLUDE file NDXPCEXT E-21
E-33 INCLUDE file NROOTS.EXT E-21
E-34 INCLUDE file RKLEVS.EXT E-22
E-35 INCLUDE file ROWSCT.EXT E-22
E-36 INCLUDE file RTCONS.EXT E-23
E-37 INCLUDE file RTSHBC.EXT E-23
E-38 INCLUDE file RTSHBM.EXT E-24
E-39 INCLUDE file RTSHBT.EXT E-24
E-40 INCLUDE file RTSHCN.EXT E-25
E-41 INCLUDE file RTSHICEXT E-25
E-42 INCLUDE file RUNTMS.EXT E-26
E-43 INCLUDE file SPNAME.EXT E-26
E-44 INCLUDE file STOPFL.EXT E-26
E-45 INCLUDE file SUBID.EXT E-27
E-46 INCLUDE file TEXTPT.EXT E-28
E-47 INCLUDE file TILDE.EXT E-28
E-48 INCLUDE file TSHDBCEXT E-29
E-49 INCLUDE file TSHDBM.EXT E-29
E-50 INCLUDE file TSHDBT.EXT E-29
E-51 INCLUDE file TSHDCN.EXT E-30
E-52 INCLUDE file TSHDICEXT E-30
E-53 INCLUDE file TSHDMD.EXT E-30
E-54 INCLUDE file TSHDSV.EXT E-31
E-55 INCLUDE file TSTEPS.EXT E-31
E-56 INCLUDE file UNITIO.EXT E-32
E-57 INCLUDE file file ZADVSL.EXT E-32
Vll
-------
LIST OF FIGURES
1 NEROS, SEROS, and ROMNET modeling domains : 4
2 The ROM vertical layers and their functional features 5
3 The principal components of the ROM 10
4 Conceptual data-flow diagram of the Core Model 15
5 BIGGAM subprograms 17
6 LILGAM subprograms 20
7 Chemical species mapping from BMAT to the Core Model 39
B-l The BCON file B-2
B-2 The BMAT file B-5
B-3 TheBTRKfile B-10
B-4 The ICON file B-13
B-5 The CONC file B-16
C-l Core Model system specification diagram C-2
C-2 Core Model Jackson Structured Design implementation diagram C-3
C-3 RUNMGR C-4
C-4 BIGGAM C-6
C-5 ICPRCS C-7
C-6 RDICON C-8
C-7 BCPRCS C-9
C-8 RDBCON C-10
C-9 BTPRCS C-ll
C-10 RDBTRK C-12
C-ll RDBT C-13
C-12 LILGAM C-14
C-13 BMPRCS C-16
C-14 RDBMAT C-17
C-15 RDMXBM C-18
C-16 CNPRCS C-19
C-17 WRCONC C-20
C-18 RDCONC C-21
Vlll
-------
SECTION 1
OVERVIEW AND STRUCTURE OF THE CORE MODEL
1.1 INTRODUCTION
The initial development of a regional (wlOOOkm) air quality simulation model began in the late 1970's
after the realization that photochemical smog often extended beyond individual urban areas to entire
sections of the United States. Interstate transport of ozone (03) and its precursors was observed during
field programs, especially in the northeast of the country. Since multiday chemical effects and long-range
transport of ozone and its precursors was beyond the scope of the existing urban-scale photochemical
models, the need for an appropriate simulation model to test the effectiveness of particular emission
control strategies on regional and urban airshed ozone concentrations became clear. The Regional Oxi-
dant Model (ROM) has been developed and enhanced over the past eight years by the U.S. Environmen-
tal Protection Agency (EPA) in response to this need. The first generation ROM (ROM1.P) became
operational in 1984. The initial model formulation and algorithm testing is documented in a three-part
volume titled ^4 Regional Scale (1000km) Model of Photochemical Air Pollution: Part 1 - Theoretical For-
mulation; Part 2 - Input Processor Network Design; Part 3 - Tests of 'Numerical Algorithms.\
ROM1.0 was a test case for future production versions of the model. It contained a very condensed
chemical kinetics mechanism, did not treat natural hydrocarbon emissions, terrain effects, or the vertical
mass flux induced by clouds, and used constant rather than dynamic layer depths. ROM2.0 became
operational in 1987, and included a more sophisticated, contemporary chemical kinetics mechanism
capable of treating both anthropogenic and biogenic precursor species. ROM2.0 also corrected many of
the deficiencies and simplifications of ROM1.0, such as using variable layer thicknesses and properly
treating cloud-induced mass flux and terrain effects. The current version of the ROM, ROM2.1, became
operational in 1989, and was developed chiefly in response to the needs of the EPA's Regional Ozone
Modeling for Northeast Transport (ROMNET; EPA, 1990) project ROM2.1 allows relatively simple
changes in the code to increase or decrease the modeling domain's size, and is adaptable to other model-
ing domains in eastern North America. Some other modifications include use of an upgraded chemical
kinetics mechanism, an updated biogenic hydrocarbon processor, and expanded use of meteorology and
1. Lamb, 1983, Lamb, 1984a, and Lamb and Lanialc, 1985, respectively.
1
-------
anthropogenic emissions data. All new features and modifications of the ROM2.1 are documented in
Young et al. (1990). The EPA is continuing to upgrade the ROM; version 2.2 will likely become opera-
tional in 1991.
The ROM was primarily developed in a VAX environment. Although software development proceeded
in tandem with hardware development - with migrations of the code to more powerful machines as they
were installed at the EPA - by the time of the ROM2.0 era, roughly 100 VAX CPU hours were required
for each 3-day simulation. The model was therefore migrated to the much more powerful IBM 3090,
resulting in the data transfer issues that are discussed in Sections 1.4.4 and 1.9. Even on the IBM 3090, the
ROM2.0 required 6 to 8 CPU hours for a 3-day simulation, increasing to around 10 CPU hours for the
ROM2.1. The CPU requirements implied that we could not be assured of one ROM2.1 execution per
night, thus prompting attempts at parallelizing and vectorizing the model. Vectorization was not con-
sidered to be an appropriate strategy for reducing elapsed clock time since it decreased run time by only
20 percent. This small reduction is due mainly to (1) the "short" vectors that result from the chemistry
solver code, and (2) the prodigious amount of indirect addressing in the code. However, parallelization
on the six available CPUs in the IBM 3090 resulted in nearly a four-fold reduction in elapsed clock time
for a model execution. The model we run at the EPA is this parallelized variant of ROM2.1; note that we
are documenting the uniprocessor variant in this User's Guide. We include a sample compile and link
stream for the uniprocessor variant in Appendix D.
This volume of the ROM2.1 User's Guide is intended for the programmers who will install and execute
the ROM, and provides the information needed to understand the operation of the ROM2.1 Core
Model.2 Companion volumes of the ROM2.1 User's Guide describe the preprocessing of the raw
meteorology and emissions input data (Part 1), and the operation of the ROM Processor Network (Part
2). The Processor Network produces the four data files required for Core Model execution: BMAT,
BCON, ICON, and BTRK. Section 1.3 summarizes these files, plus the other four input files and one
output file that must pre-exist, and that the Core Model expects to find. In addition, you will find the
computer storage requirements listed in Section 1.4. We show how we transfer data between the VAX
and IBM, and vice versa, in Sections 1.4.4 and 1.9 respectively. We describe the software structure of the
model in Sections 1.5 to 1.8. Section 2 is an in-depth look at the nine input files, and Section 3 describes
the final product of the ROM - the chemical concentration predictions contained in the CONC file.
Appendix A is a brief tutorial to Jackson Structured Programming and state vectors. We use these con-
cepts in the Core Model code, which consists of 66 subprograms in addition to the .main program,
RUNMGR. Of these, 14 are variable-state subroutines, which we call processes to distinguish them from
2. This volume will also be useful to those people who may wish to maintain and enhance the code.
2
-------
procedures. Procedures execute their code from top to bottom whenever they are invoked. Processes
interact with their calling programs in a critical time sequence, hence their implementation as variable-
state subroutines. The processes maintain variable text pointers in their code that are set when the pro-
cess is suspended and control is returned to the calling program. At the next invocation of the process, it
resumes execution of the code from the point at which it was previously suspended.
Appendix B contains design and structure diagrams for the ROM2.1 data files in the ROMNET region.
Appendix C contains design and structure diagrams for the principal ROM2.1 subroutines. Appendix E
contains descriptions of all the common blocks in all the INCLUDE files, including DIMENS.EXT and
REGION.EXT, which are the common blocks that set the domain- and model-specific variables, and are
shown in Tables E-l and E-2 respectively. Appendix F is an in-depth guide to the Core Model's error-
checking procedures.
1.2 GENERAL MODEL CHARACTERISTICS
The ROM was designed to simulate most of the important chemical and physical processes that are
responsible for the photochemical production of ozone over a domain of 1000 km and for multiple 3-day
episodes up to approximately 15 days in duration. These processes include (1) horizontal transport, (2)
atmospheric chemistry and subgrid-scale chemical processes, (3) nighttime wind shear and turbulence
associated with the low-level nocturnal jet, (4) the effects of cumulus clouds on vertical mass transport
and photochemical reaction rates, (5) mesoscale vertical motions induced by terrain and the large-scale
flow, (6) terrain effects on advection, diffusion, and deposition, (7) emissions of natural and anthropo-
genic ozone precursors, and (8) dry deposition. The processes are mathematically simulated in a three-
dimensional Eulerian model with 3J/2 vertical layers, including the boundary layer and the capping
inversion or cloud layer.3 Horizontal grid resolution is W longitude by 1/6° of latitude, or about
18.5 km x 18.5 km. Current model domains include the northeastern United States and the southeastern
U.S./Gulf Coast area (Figure 1). For each of these domains, the model uses Eastern Standard Time
(EST) in all its calculations.
3. Layer 0, the "hair layer, has a layer thickness that is always Vjo of the thickness of layer 1; the thicknesses of layers 1,2, and 3 vary.
3
-------
45 N
38 N
35N
2BN
Figure 1. NEROS, SEROS, and ROMNET modeling domains for the ROM; dots represent grid cell
corners.
1.2.1 Physical Processes within Layers 0.1.2 and 3
The meteorological data are used to objectively model regional winds and diffusion. The top three
model layers are prognostic (predictive) and are free to locally expand and contract in response to
changes in the physical processes occurring within them. During an entire simulation period, hori-
zontal advection and diffusion and gas-phase chemistry are modeled in the upper three layers. The
bottom layer, layer 0, is a shallow diagnostic surface layer designed to approximate the subgrid-scale
effects on chemical reaction rates from a spatially heterogeneous emissions distribution. ROM pre-
dictions from layer 1 are used as surrogates for surface concentrations. The time scale of output
concentrations is 30 minutes, although typically 1- and 8-hour daytime averages are used for the
analysis of air quality by the public policy sector. Figure 2 shows the ROM layers during the day and
at night, and describes some of their features.
-------
Subsidence Inversion
Ckxid layer
Subsidence Inversion
Trrr—-•.-.;•—. • .•• ••:. •••.•.;.?' :'••:• .':.0fc
Layer Functions
Cloud layer
- downward transport of stratospheric ozone
- upward transport by cumulus clouds
- gas-phase photochemistry
- long-range transport by the free atmosphere
Mixed layer
- gas-phase photochemistry
- turbulence and wind shear effects on
transport and diffusion
- deposition on mountainous terrain
Marine layer
- lake- and marine-layer effects
Surface layer
- subgrid-scale effects on reaction rates;
segregation of fresh and aged pollutants
• ground deposition
- spatial variations in mean concentrations
due to Ine, point, and area sources
Old cloud layer
- downward transport of stratospheric ozone
- dark gas-phase chemistry
Old mixed layer
- transport of aged gas-phase reactants
and products
- darit gas-phase chemistry
Radiation Inversion/nocturnal Jet layer
- transport of aged pollutants and reactants
by the nocturnal )et
- transport of nighttime emissions from
tall stacks and warm cities
- deposition on mountainous terrain
• • '•. Radiation inversion/ - downward transport during jet breakdowns
{ • ' nocturnal jet layer - nighttime shallow mixed layer over.
^ ' urban heat Islands
Old otoud layer
.pd mixed layer
layer
Surface layer
- calms In rural areas
- ground deposition
- reservoir of nighttime emissions of
small, low-level sources
Figure 2. The ROM vertical layers and their functional features.
Layers 1 and 2 model the depth of the well-mixed layer during the day. Some special features of layer
1 include the modeling of (1) the substantial wind shear that can exist in the lowest few hundred
meters above ground in local areas where strong winds exist and the surface heat flux is weak, (2) the
thermal internal boundary layer that often exists over large lakes or near sea coasts, and (3) deposi-
tion onto terrain features that protrude above the layer. At night, layer 2 represents what remains of
the daytime mixed layer. As stable layers form near the ground and suppress turbulent vertical
mixing, a nocturnal jet forms above the stable layer and can transport aged pollutant products and
reactants considerable distances. At night, emissions from tall stacks and warm cities are'injected
directly into layers 1 and 2. Surface emissions are specified as a mass flux through the bottom of
layer 1.
During the day, the top model layer, layer 3, represents the synoptic-scale subsidence inversion char-
acteristic of high ozone-concentration periods; the base of layer 3 is typically 1 to 2 km above the
ground. Relatively clean tropospheric air is assumed to exist above layer 3 at all times. If cumulus
clouds are present, an upward flux of ozone and precursor species is injected into the layer by penet-
-------
rative convection. At night, ozone and the remnants of other photochemical reaction products may
remain in this layer and be transported long distances downwind. These processes are modeled in
layer 3.
When cumulus clouds are present in a layer 3 cell, the upward vertical mass flux from the surface is
partially diverted from injection into layer 1 to injection directly into the cumulus cloud of layer 3. In
the atmosphere, strong thermal vertical updrafts, primarily originating near the surface in the lowest
portion of the mixed layer, feed growing fair weather cumulus clouds with vertical air currents that
extend in one steady upward motion from the ground to well above the top of the mixed layer. These
types of clouds are termed fair weather cumulus since atmospheric conditions are such that they do
not grow to the extent that precipitation forms. The dynamic effects of this transport process and
daytime cloud evolution can nave significant effects on the chemical fate of pollutants. For example,
fresh emissions from the surface layer can be injected into a warm thermal and rise, essentially
unmixed, to the top of the mixing layer where they enter the base of a growing cumulus cloud. Within
the cloud, the chemical processes of ambient pollutant species are suddenly altered by the presence of
liquid water and the attendant attenuation of sunlight. The presence of fair weather cumulus clouds
implies that the atmosphere above the earth's boundary layer is too stably stratified for thermals to
penetrate higher. In this case, the air comprising the tops of these clouds returns to the mixed layer
and is heated on its descent since it is being compressed by increasing atmospheric pressures. Ulti-
mately, the air again arrives at the surface level where new emissions can be injected into it and
ground deposition may occur, and the process may begin again. The time required for one complete
cycle is typically 30 to 50 minutes with perhaps one-tenth of the time spent in the cloud stage.
Within the ROM system, a submodel parameterizes the above cloud flux process and its impact on
mass fluxes among all the model's layers. In the current implementation of the chemical kinetics,
liquid-phase chemistry is not modeled, and thus pan of the effects from the cloud flux processes are
not accounted for in the simulations. Future versions of the chemical kinetics may include liquid-
phase reactions. The magnitude of the mass flux proceeding directly from the surface layer to the
cloud layer is modeled as being proportional to the observed amount of cumulus cloud coverage and
inversely proportional to the observed depth of the clouds.
Horizontal transport within the ROM system is governed by hourly wind fields that are interpolated
from periodic wind observations made from upper-air soundings and surface measurements. During
the nighttime simulation period, the lowest few hundred meters of the atmosphere above the ground
may become stable as a radiation inversion forms. Wind speeds increase just above the top of this
layer, forming the nocturnal jet This jet is capable of carrying ozone, other reaction products, and
-------
emissions injected aloft considerable distances downwind. This phenomenon is potentially signifi-
cant in modeling regional-scale air quality and is implicitly treated by the model, where the definition
of layer 1 attempts to account for it.
Because standard weather observations do not have the spatial or temporal resolution necessary to
determine with confidence the wind fields in layer 1, a submodel within the ROM system was devel-
oped to simulate the nighttime flow regime in layer 1 only. This prognostic flow submodel is acti-
vated only when a surface inversion is present over most of the model domain. At all other times, the
flow in layer 1 is determined from interpolation of observed winds. The nighttime flow regime within
layer 1 is influenced by buoyancy, terrain, warm cities, pressure-gradient forcing, and frictional forces,
all of which are accounted for in the model's flow formulation. Solution of the wind submodel equa-
tions produces estimates of the wind components as well as the depth of the inversion layer for all grid
cells in layer 1.
1.2.2 ROM Chemistry
The chemical kinetic mechanism embedded in the current version of the ROM is the Carbon Bond IV
(CB-IV) set of reactions (Gery etal., 1989). This mechanism simulates the significant reaction path-
ways responsible for gas-phase production and destruction of the constituents of photochemical smog
on regional scales. The mechanism consists of 82 reactions encompassing 35 individual species; these
species are listed in Table 1. The ROM's chemical solution scheme makes no a priori assumptions
concerning local steady states. Therefore, all species are advected, diffused, and chemically reacted in
the model simulations.
The CB-IV contains a standard set of reactions for atmospheric inorganic chemical species, including
Oj, NO, NC>2, CO, and other intermediate and radical species. Organic chemistry is partitioned
along reactivity lines based on the carbon structures of the organic molecules. Nine individual cate-
gories of organics are represented to account for the chemistry of the hundreds of organic molecules
existing in the ambient atmosphere: ETH, an explicit representation of ethene; FORM, an explicit
representation of formaldehyde; OLE, a double-bonded lumped structure including two carbons
(e.g., olefins); PAR, a single-bond, single-carbon structure (i.e., paraffins); ALD2, the oxygenated
two-carbon structure of the higher aldehydes; TOL, the aromatic structure of molecules with only one
functional group (e.g., toluene); XYL, the structure of molecules with multifunctional aromatic rings
(e.g., xylene); ISOP, the five-carbon isoprene molecule; and NONR, a single-carbon organic structure
not significantly participating in the reaction sequence. We include MTHL (methanol) in the mech-
anism for future-year scenarios that require emission control strategies for methanol-powered
vehicles.
-------
TABLE 1. ROM CHEMICAL SPECIES •
Symbol
ALD2
C203
CO
CRES
CRO
ETH
FORM
H202
HN02
HNO3
HO2
ISOP
MGLY
N205
NO
N02
N03
O
Description
High MW aldehydes
Peroxyacetyl radical
Carbon monoxide
Cresol and high MW phenols
Methylphenoxy radical
Ethene
Formaldehyde
Hydrogen peroxide
Nitrous acid
Nitric acid
Hydroperoxy radical
Isoprene structures
Methylglyoxal
Dinitrogen pentoxide
Nitric oxide
Nitrogen dioxide
Nitrogen trioxide
Q3p atom
Symbol
01D
O3
OH
OLE
OPEN
PAN
PAR
PNA
ROR
TO2
TOL
XO2
XO2N
XYL
MTHL
NONR
TRAC
Description
QiD atom
Ozone
Hydroxyl radical
Olefinic carbon bond
High MW aromatic oxidation
ring fragment
Peroxyacetyl nitrate
Paraffinic carbon bond
Peroxynitric acid
Secondary organic oxy radical
Toluene-hydroxyl radical adduct
Toluene
NO to NO2 reaction
NO to nitrate (No; ) reaction
Xylene
Methanol
Nonreactive hydrocarbons
Tracer species
a MW = molecular weight
Three classes of biogenic hydrocarbons are included in a separate natural area source emissions
inventory used by the ROM: (1) isoprene, a molecule principally emitted by deciduous trees, is
treated by the ISOP species in CB-IV; (2) monoterpenes, a class of natural hydrocarbons emitted
principally by coniferous trees, is not treated explicitly in CB-FV. The surrogate monoterpene mole-
cule, a-pinene, which consists of 10 carbons, is apportioned to the existing CB-IV categories as 0.5
OLE, 1.5 ALD2, and 6 PAR; (3) unidentified hydrocarbons (gas chromatography analysis did not
identify specific hydrocarbon compounds) are tentatively treated as 50% terpenes, 45% PAR, and 5%
NONR (Pierce et al., 1990). These unidentified compounds can comprise as much as 40% of the
biogenic hydrocarbons.
1.2.3 System Components
The raw input data to the ROM (refer to Section 1.3) are manipulated by a hierarchical network of
processors that range in function from simple reformatting of emissions data to generating the com-
plex wind fields that drive the atmospheric transport algorithm in the Core Model. These processors
are interconnected by their requirements for and production of data. The ultimate product of the
processor network is a collection of data files that can be categorized into two types: processor files
(PF) and model files (MF). Processor files contain partially processed data required as input to
higher level processors. Model files contain the parameter fields that are transformed into the vari-
8
-------
ables required by the model algorithms; however, they also provide input to a number of higher level
processors. The output of the processors are the four Core Model input data files. The Core Model
is described in detail starting with Section 1.5. Figure 3 shows the principal components of the model,
starting with the raw input data and ending with the CONC file.
The processors are organized into nine distinct hierarchical stages, numbered 0-8. Stage 0 proces-
sors produce output files such as the gridded land use data. Stage 1 processors interface directly with
the preprocessed input data sets, which have, at this juncture, undergone extensive quality control.
Subsequent stages transform the input data into the gridded fields of temporally and spatially varying
parameter values needed by the highest stages of the processing network. Processors at any stage can
interface directly with the B-Matrix compiler, described below, by production of model input files
(MF). This multistage organization is important to the network because it clearly delineates the
sequence of program execution. Processors at the same stage may execute simultaneously. A pro-
cessor at any given stage, however, must wait until all processors from lower stages along its input
data paths have been completed. Formal definition of all data/processor relationships and
automation of processor executions are essential to ensure consistency and validity of model input
files.
The program that serves as the interface between the model input files and the algorithms describing
the governing processes is called the B-Matrix Compiler (BMC) because it functions similarly to a
computer language compiler that transforms high-level language commands into a machine or
algorithm-specific representation. The BMC mathematically combines physical parameters such as
layer thicknesses, air densities, etc., into the complex coefficients required for solution of the gov-
erning equations. These coefficients can no longer be equated with physical quantities; they are
purely mathematical entities related specifically to the form of the finite difference algorithms used by
the ROM.
The core of the ROM system is a set of algorithms that solves the coupled set of finite difference
equations describing the governing processes in each layer of the model. These governing equations
are expressed in a form that allows the chemical kinetics, advection, and vertical flux to be treated
independently. The chemistry module exchanges information with algorithms of the governing
equations via two vectors: (1) a vector that contains the net production rate of each species, and (2) a
vector that contains the net destruction rate. Such design simplifications enhance the flexibility of the
model and are not limited to the interchanges of the chemical mechanism; they apply to all theoreti-
cal formulations of the physical and meteorological processes (i.e., to all the processors).
-------
DC 5^
00 i*-,
[.[ ^
0 (-
O ui
*•
J
UI
iZ
u.
s
u.
Q.
t
U
W
<
CD
<
H
<
O
O
cc;
CD
0)
C
o
o
O
0)
10
-------
1.2.4 ROM Limitations
There are several limitations inherent in the model. Among the most important of these are: (1) the
model is designed to represent only ozone-season (April - October) meteorological conditions; (2)
the ROM does not take into account any aqueous-phase chemistry, (3) cumulus cloud processes are
such that when a cloud is created in a grid cell, it remains there for a full hour (i.e. cloud physics are
not considered), and the cloud is not advected; (4) the ROM actually consists of three two-
dimensional models that are linked,4 and, as such, cannot be expected to model regions that contain
high, complex, mountainous terrain such as the Rocky Mountains or Sierra Nevada; (5) the ROM,
with its current 18.5 km x 18.5 km grid resolution, is not designed to provide detailed information at
local scales that are significantly influenced by local source distributions; and (8) the ROM is cur-
rently configured to run only on domains in eastern North America on the scale of 1000 km. This
regional scale is germane since long-range transport of ozone precursors will have a significant effect
on local ozone concentrations.
We advise users of ROM to interpret its results in terms of analysis of different emission control
strategies on ozone concentration, rather than assuming the results to be an accurate snapshot of a
specific pollution event.
1.3 DATA AND FILE REQUIREMENTS
The ROM system requires five types of "raw" data inputs: air quality, meteorology, emissions, land use,
and topography. Air quality data required by the ROM include initial conditions (1C) and boundary
conditions (BC). The model is initialized with clean tropospheric conditions for all species several (usu-
ally 2-4) days before the start of an episode. The initial condition field has essentially been transported
out of the model domain in advance of the portion of the episode of greatest interest. Upwind lateral
boundary conditions for ozone are updated every 12 hours based on measurements. Other species con-
centrations at the boundaries, as well as all species at the top of the modeling domain, are set to tropo-
spheric clean-air concentrations.
Meteorological data are assimilated by the first stage of preprocessors. These data contain regular hourly
observations from U.S. National Weather Service surface stations (and from similar stations in Canada as
necessary), including wind speed and direction, air temperature and dew point, atmospheric pressure, and
cloud amounts and heights. Twice-daily sounding data from the upper-air observation network are also
included in the meteorological database. Upper-air meteorological parameters include atmospheric
4. The three models represent layer 3, layer 2, and a combined layer 1 and 0.
11
-------
pressure, wind speed and direction, and air temperature and dew point. Finally, both buoy and Coastal
Marine Automated Station data are used; parameters typically reported are wind speed and direction, and
air and sea temperatures.
Emissions data for the primary species are input to the ROM system as well. Most recently these data
have been provided from the NAPAP1985 emissions inventory with 20-km spatial resolution (Saeger et
al., 1989). Species included are CO, NO, NO2, and nine hydrocarbon reactivity categories. Natural
hydrocarbons are also input, including isoprene explicitly, monoterpenes divided among the existing
reactivity classes, and unidentified hydrocarbons.
The Core Model requires eight input files and one output file to exist in order to run; these files are
summarized below. Each of the six principal input files starts with a text file header that describes the
file's contents, followed by a data body. The data body consists of a sequence of time step headers, each
followed by a body of data records relative to that time step. While these files all contain data, note that
the auxiliary input files PROG and STOPCK do not.
Principalfiles
BCON - The boundary conditions file contains the concentrations for each of the 35 chemical
species in a one-cell deep border within the modeling domain. Concentrations are given for
each model layer and for each model time step. We show the file's structure diagram in
Appendix B, Figure B-l.
BMAT - The B-matrix file contains (1) the vertical flux parameters that are required to solve
the B-matrix differential equation, and (2) the emission source terms, and (3) the gridded
meteorology parameters in the three model layers for each model time step (1800 s) necessary
to make adjustments in chemistry rate constants. We show the file's structure diagram in
Appendix B, Figure B-2.
BTRK - The Backtrack file contains the gridded locations (in grid cell coordinates) of the
previous position of a parcel of air that will arrive at a grid cell node at the current time step.
These data are used by the model to compute the horizontal transport We show the file's
structure diagram in Appendix B, Figure B-3.
ICON - The initial conditions file contains the gridded concentrations for the 35 chemical
species in each model layer for the initial model time step. This file is used only at the start of
an episode time-contiguous run, i.e., at the start of scenario 1. It is supplanted by the NEW-
ICON file at the start of each subsequent scenario within the episode.5 We show the file's
structure diagram in Appendix B, Figure B-4.
5. The relationship between scenarios and episodes is that each scenario is 72 hours long, and an ozone episode consists of one or more
contiguous scenarios.
12
-------
NEWICON - The model writes this file at the completion of the last time step of a scenario. It
is used as the ICON file for the continuation of the episode computation, and is a copy of the
CONC file from the last time step of the scenario. NEWICON supplants the ICON file for the
continuation of the episode computation. The CONC file itself could be used to continue the
episode, but the NEWICON file is much smaller; thus, continuing runs do not require retain-
ing access to the large CONC files.
RESTRT - This file constitutes a state vector file, and is used only to restart a model run in the
event of program termination before the end of a scenario. RESTRT contains the data that
correctly reposition the input files and set the text pointers in the code; this ensures the proper
state of the model when you resume a run from the start of any scenario time step prior to the
termination step.
Auxiliary files
PROG - This file is a one line text file that the model writes at the completion of each time
step. Its purpose is to allow you to view the progress of the model; you can "type" out this file
during the model run and see the number of the last completed time step.
• STOPCK - The purpose of this file is to allow you to terminate a model run at the end of a
currently-executing time step. It consists of one line of text enclosed in single quotes, which, if
anything other than 'STOP', permits the model run to continue.
Outputfile
• CONC - The model writes the predicted concentrations that result from its execution to the
CONC file. The data consist of the gridded concentrations of the 35 chemical species within
each of the model layers for each time step (30 minutes) of the model run. We show the file's
structure diagram in Appendix B, Figure B-5.
1.4 MODEL EXECUTION AND STORAGE REQUIREMENTS FOR THE D3M 3090
1.4.1 Starting the ROM at the Top of an Hour
If you want to run consecutive scenarios, the Core Model must be started at the top of an hour, not at
the half-hour. We begin all our model runs at noon EST.
1.4.2 CPU Use
Currently, we are running the ROM Core Model on the EPA's IBM® 3090 computer.6 Running the
ROM requires significant CPU resources; a typical 3-day simulation (72 hours, 144 model time steps)
requires about 9.5 hours of CPU time on the IBM 3090. Approximately 10 percent of this CPU time
6. IBM is a registered trademark of the International Business Machines Corporation.
13
-------
is used in the calculation Of advection; the other 90 percent is used for the gas-phase chemistry cal-
culations. With the parallel processing capability of the IBM 3090, elapsed clock time is approxi-
mately one-quarter of the CPU time.
1.4.3 Storage Requirements
A list of storage requirements for the Core Model follows (1 track = 47,476 bytes):
• 21 tracks of code (approximately 12000 lines of code)
• 24 tracks of libraries
1 track of compile/link JCL
• 1 track of JCL to submit the run
• 47 tracks of the log file
1.4.4 Core Model Input Data; Transfer from the VAX to the IBM
The following procedure is specific to our installation, and may not pertain to you. The Core Model
input data sets are transferred from the VAX to the IBM using the DECnet/SNA Data Transfer
Facility" (DTP).7 For processors that generate Core Model input data sets,8 you will select a flag in
the control cards file that determines whether to write the data in VAX binary or IBM binary format.
We select the IBM format, and invoke the DTP to directly write the ICON, BCON, BTRK, and
BMAT files to the IBM's disks. For further information on transferring data, we refer you to the
DECnet/SNA VMS Data Transfer Facility User's Guide.? The subroutine that translates data to
IBM-readable format is briefly described below.
SUBROUTINE TRNSLT
This subroutine translates VAX format and writes an IBM-format file on VAX Direct Access Storage
Drive (DASD) so that it may subsequently be copied to IBM DASD via the DEC/DTP (Data Transfer
Facility). The translation is necessary because DTP supports only a one-to-one transfer of the data,
which do not map to identical representation of numbers on the two architectures. If the file to be
transferred is ASCII in its entirety, a call to this routine is unnecessary, and the DTP facility can be
used directly. If, however, the file is a combination of ASCII and REAL*4 or INTEGER *4 data, a
7. DECnet is a trademark of Digital Equipment Corporation, Nashua, NH 03061.
8. P02G, P22G, P38G, and P40G.
9. Order number AA-JM75B-TE; Digital Equipment Corporation, P.O. Boot CS2008, Nashua, NH 03061.
14
-------
call to this routine is necessary. This routine does not support any other VAX data structures.
TRNSLT translates VAX F_floating format words to IBM floating point, and ASCII format to
EBCDIC format.
1.5 SOFTWARE COMPONENTS OF THE CORE MODEL
Figure 4 shows a conceptual data-flow diagram of the Core Model; its design reflects our desire to keep
the model as simple as possible so that we can readily maintain and upgrade it. The design is based on the
concept of modularity with respect to the flow of data, and follows the ideas of Jackson (see Appendix A).
The system specification diagram that illustrates the main data flows is shown in Appendix C, Figure C-l.
Most of the subroutines in the Core Model were written as though they were stand-alone programs that
read and write intermediate files during the computation. Following Jackson, these subroutines were
then "inverted," and the intermediate files eliminated. An illustration of our implementation of this
Jackson Structured Design for the Core Model can be found in Appendix C, Figure C-2.
Maintain model
state vector
Management of
model run and
restart capability
(RUNMGR)
Write next scenario's
initial conditions file
time step *
solve horizontal transport
(B1GGAM)
domain row *
I
domain column
1
vertical flux time step
1
model layer *
chemistry time step
1
solve chemistry
I
solve vertical flux
(LILGAM)
*
Figure 4. Conceptual data-flow diagram of the Core Model. Asterisks indicate iterative loops.
15
-------
The Core Model consists of .three principal components: RUNMGR, BIGGAM, and LILGAM. The main
program is RUNMGR (Run Manager); its purpose is to: (1) initiate the run, (2) restart the run if required,
(3) maintain the state vector file (RESTRT), (4) write the progress file (PROG), and (5) examine the stop
check file (STOPCK). Once the required input files are properly opened, RUNMGR invokes BIGGAM,
starting the actual computation process. BIGGAM in turn invokes LILGAM. Because of the large time-
scale differences between the transport processes and the chemical processes in the regional-scale oxidant
modeling problem, we can write the diffusion equation as two processes to solve for a transport
component F decoupled from the chemistry. Then, with the solution for F in hand, we can determine the
chemistry component 7. We show the structure diagram for RUNMGR in Appendix C, Figure C-3.
1.5.1 BIGGAM
BIGGAM solves for the horizontal advection of the chemical species concentrations. The transport is
computed by (1) interpolating the concentration values at backtrack locations,10 and (2) solving the
transport using a quasi-Lagrangian advection scheme that involves determining a Green's function
for the F diffusion equation at spatially-interpolated upwind points. We use a biquintic surface fit for
the spatial interpolation. For a one-cell deep perimeter within the modeling domain boundaries,
BCON file concentrations supplant the interpolated concentrations if inflow conditions exist at that
border of the domain. Once BIGGAM has computed an advection solution component, it invokes
LILGAM to compute the chemistry component and finalize the current time step concentration pre-
dictions. We show the structure diagram for BIGGAM in Appendix C, Figure C-4.
The subprograms specifically associated with BIGGAM are shown in Figure 5, and then are briefly
described below.
INIRUN reads the run stream's control parameter cards and sets up the model run conditions. INIRUN
loads the common blocks in the include file HEADIN.EXT and performs the following operations:
• Sets the vertical flux time step (300 s) and the chemistry solver parameters, and echoes
these data to the run log.
Sets the number (3) and names (03, NO, NOa) of the primary oxidant species in the ROM,
and reports these data to the run log.
Obtains the indices of the primary oxidant species using the utility function INDEXI, and
reports these data to the run log.
• Sets and echoes the upper and lower chemistry time step limits.
• Sets the advection time step variable used in LILGAM equal to the corresponding variable
in HEADIN.EXT.
10. A backtrack location is tbc prior position of a parcel of air that will arrive at a grid cell node at the current time step, where each time
step is 1800 s.
16
-------
BIGGAM
INIRUN
* ICPRCS
* BCPRCS
* BTPRCS
* LILGAM
* RDICON
ORSPIC
* RDBCON
ORSPBC
* RDBTRK
OPICON
PO1CON
OPBCON
POBCON
OPBTRK
POBTRK
* RDBT
RDBT
Figure 5. BIGGAM subprograms. An asterisk implies that the subprogram can be called more
than one time.
• Calls ADATE to obtain and report the execution creation date and time to the run log.
Initializes all the file date and time information to correspond to the model time initial-
ization. These data are used to maintain all the process subroutine's internal clocks.
The variables in the common blocks contained in HEADIN.EXT are used throughout the model
to:
Set upper limits for processing loops,
• verify file headers on the input files, and
provide data to be written to the CONC file header.
For a complete description of these variables and their origin, see Table 2.
ICPRCS is the process interface between BIGGAM and the time step initial conditions data, restruc-
tured for the model data processing; its structure diagram is shown in Appendix C, Figure C-5.
RDICON is the process interface between ICPRCS and the initial conditions data in the ICON or the
NEWICON file; its structure diagram is shown in Appendix C, Figure C-6.
OPICON opens the ICON or the NEWICON file, reads and checks the file's header information, then
writes a summary to the run log.
17
-------
TABLE 2. HEADIN.EXT VARIABLES: DESCRIPTIONS AND ORIGINS
Name
Description
Origin
CDATIN
CTIMIN
SDATIN
STHRIN
TSTPIN
FRSTIN
GRDNIN
SWLNIN
SWLTIN
NELNIN
NELTIN
DLONIN
DLATIN
NCOLIN
NROWIN
NLEVIN
LVNMIN(1)
NSPCIN
SPNMIN(k)
1CNTIN
TEXTlN(n)
DATE/TIME INFORMATION
Creation date of simulation (YYMMDD)
Creation time of simulation (HHMMSS)
Julian start date for model scenario (YYDDD)
Start hour (00 - 23) for model scenario
Tune step size (s) for model scenario
First time step (s) past start time of model scenario
Argument returned from call to DATTIM
Argument returned from call to DATTIM
Read in from run stream
Read in from run stream
Set in assignment statement equal to 1800
Argument returned from call to CLOCK1
GRID DEFINITION INFORMATION
Grid definition name
Longitude (degrees) of southwest corner of grid
Latitude (degrees) of southwest corner of grid
Longitude (degrees) of northeast corner of grid
Latitude (degrees) of northeast corner of grid
Longitudinal grid cell increment (degrees)
Latitudinal grid cell increment (degrees)
Number of columns in domain grid
Number of rows in domain grid
Read in from run stream
Argument returned from call to CELLM
Argument returned from call to CELLM
Argument returned from call to CELLM
Argument returned from call to CELLM
Argument returned from call to CELLM
Argument returned from call to CELLM
Argument returned from call to CELLM
Argument returned from call to CELLM
Number of model layers
Name of model layer /
MODEL LAYERS INFORMATION
Set in assignment statement, equal to NLEVS (DIMENS param-
eter)
Set in assignment statement, equal to LVNAME (BLKMOD
data)
MODEL SPECIES INFORMATION
Set in assignment statement, equal to NSPECS (DIMENS
Number of chemical species in model parameter)
Name of species A: Set in assignment statement, equal to SPNAME (BLKMOD
data)
CONC FILE HEADER DESCRIPTIVE TEXT
Number of text records Computed during the operation that reads in the text records
Descriptive text record n Read in from run steam
POICON positions the ICON or the NEWICON file to the start of the model execution scenario.
ORSPIC generates the expansion list (look-up table) of species names for which the ICON file has
values. These values are mapped to the full Core Model list of species. For ROM2.1, the ICON file
contains the same species list as the Core Model, and thus the mapping table is one-to-one.
BCPRCS is the process interface between BIGGAM and the boundary conditions data, restructured for
the model data processing; its structure diagram is shown in Appendix C, Figure C-7.
RDBCON is the process interface between BCPRCS and the boundary conditions data in the BCON file;
its structure diagram is shown in Appendix C, Figure C-8.
OPBCON opens the BCON file, reads and checks the file's header information, then writes a summary
to the run log.
18
-------
POBCON positions the BCON file to the start of the model execution scenario.
ORSPBC generates the expansion list (look-up table) of species names for which the BCON file has
values. These values are mapped to the full Core Model list of species. For ROM2.1, the BCON file
contains the same species list as the Core Model, and thus the mapping table is one-to-one.
BTPRCS is the process interface between BIGGAM and the backtrack data in the BTRK file, reorga-
nized for BIGGAM data processing; its structure diagram is shown in Appendix C, Figure C-9.
RDBTRK is the process interface between BTPRCS and the data in the BTRK file; its structure diagram
is shown in Appendix C, Figure C-10.
OPBTRK opens the BTRK file, reads and checks the file's header information, then writes a summary
to the run log.
POBTRK positions the BTRK file to the start of an execution scenario.
RDBT performs the FORTRAN open and the file and data read operations on the BTRK file; its
structure diagram is shown in Appendix C, Figure C-ll.
1.5.2 LILGAM
LILGAM computes the remaining components of the species concentrations: those due to vertical
fluxes across the model layer surfaces, Y, and those due to the chemical reactions between the pol-
lutant species, Y' • The value of the concentration recorded at the end of each 1800-sec advection
time step is the product
c-r- Y -Y'
• Every 300 s within the advection time step, the interlayer vertical fluxes are modeled using
a fourth-order Runge-Kutta integration of the system of differential. equations that
describes them. The integration step size is determined by finding the largest negative
eigenvalue of the third-order matrix of coefficients (B-matrix) that quantifies the inter-
layer fluxes. The reciprocal of this eigenvalue is proportional to the smallest time scale in
the system describing the vertical exchange between the model's three layers.
• Within each 300 s y time step, the chemistry component Y ' is computed using a small time
step that varies between 10 s and 60 s. The differential equations that describe the chemi-
cal reactions are solved iteratively, and model the equilibration process of a reactive mix-
ture of gaseous pollutant species. Essentially, the time step is controlled by the stiffness of
the system of differential equations, which in turn is a function mainly of local primary
emissions and sunlight amount. The relative stiffness of the system is determined by com-
paring (at the end of each chemistry time step) the fractional difference in the nitric oxide
19
-------
(NO) species from the previous step and adjusting the time step downward if the
difference exceeds a preset value. NO is used as tbe standard since it is both one of the
primary oxidant species and is involved in the fastest reactions in the system.
We show the structure diagram for LILGAM in Appendix C, Figure C-12. The subprograms specifi-
cally associated with LILGAM are shown in Figure 6, and then briefly described.
LILGAM
« BMPRCS
» CNPRCS
RTPHO
RTSET
HSTEP
PQCOEF
RATED
PQI
CTILDE
» RDBMAT
ORSPBM
« WRCONC
OPBMAT
RDMXBM
POMXBM
I* RDMXBM
I * RDHDBM
I DUMPHD
* RDHDBM
DUMPHD
RDMXBM
1* RDHDBM
I DUMPHD
OPWRCN
Figure 6. LILGAM subprograms. An asterisk implies that the subprogram can be called more
than one time.
BMPRCS is the process interface between LILGAM and the BMAT file data, reorganized for the
LILGAM data processing; its structure diagram is shown in Appendix C, Figure C-13.
RDBMAT is the process interface between BMPRCS and the data on the BMAT file; its structure dia-
gram is shown in Appendix C, Figure C-14.
OPBMAT opens the BMAT file, reads and checks the file's header information, then writes a summary
to the run log.
RDMXBM performs the FORTRAN open and the file and data read operations on the (possibly mul-
tiple) BMAT file; its structure diagram is shown in Appendix C, Figure C-1S.
RDHDBM reads the BMAT file header on each of the BMATs subfiles and echoes the information to
the run log.
20
-------
DUMPHD is a subroutine invoked by RDMXBM to dump the BMAT file header information to the run
log if multiple BMAT files are not correctly ordered.
POMXBM positions the (possibly multiple) BMAT file to the start of the execution scenario.
ORSPBM generates the expansion list (look-up table) of species names for which the BMAT file has
values. These values are mapped to the full Core Model list of species.
CNPRCS is the process interface between LILGAM and the predicted concentration data, reorganized
for writing to the CONC file; its structure diagram is shown in Appendix C, Figure C-16.
WRCONC is the process interface between CNPRCS and the concentrations data in the CONC file; its
structure diagram is shown in Appendix C, Figure C-17.
OPWRCN opens the CONC file, writes the file's header information on the file, and echoes this data to
the run log.
RTPHO determines the layer average photolytic rate constants, adjusted for variations in solar radi-
ation due to cloud cover and solar angle.
RTSET adjusts the values of rate constants for the nonphotolytic reactions for temperature, air den-
sity, and water vapor concentration.
HSTEP determines the integration step size used in GTILDE.
PQCOEF calculates the part of the normalized reaction coefficients that do not depend on the current
chemistry component of the concentration, Y '•
RATED completes the chemistry time step calculation of the reaction coefficients using the current
value of Y'.
PQI calculates the production and decay terms of the chemistry differential equation system for each
chemistry time step. LILGAM then uses these terms in a predictor/corrector fashion to update y' for
the chemistry time step.
GTILDE calculates the interlayer flux concentration component, y, for each of layer 1, layer 2, and
layer 3.
21
-------
1.6 SUBPROGRAMS THAT MAINTAIN THE STATE VECTOR AND RESTART THE MODEL
The subprograms specifically associated with maintaining the state vector file and with model run restarts
are listed below, then briefly described.
RDCONC
OPCONC
POCONC
WRSTAV
RDSTAV
OPSTAV
POSTAV
RDCONC is the process interface that is called for a RESTART run; it reads the CONC file data for the
time step that precedes the first step of the restarted run, and its structure diagram is shown in Appendix
C, Figure C-18.
OPCONC opens the CONC file for a RESTART run, reads and checks the file's header information, then
writes a summary to the run log.
POCONC positions the CONC file, for a RESTART run, to the requested starting time.
WRSTAV opens and writes the state vector data to the RESTRT file.
RDSTAV reads the data on the RESTRT file for a RESTART run, for the time step that precedes the first
step of the restarted run.
OPSTAV opens the RESTRT file for a RESTART run.
POSTAV positions the RESTRT file for a RESTART run to the requested starting time.
1.7 UTILITY SUBPROGRAMS
ADATE
ASORT
CELLM
CLOCK1
CLOCK2
DATTIM
FSKIP1
INDEX1
IOCL
JFHJE2
JFILE5
JFILE6
JULIAN
JUNIT
RDCHAR
22
-------
RDFILE
WRCHAR
WRFILE
ADATE returns the current formatted date and time from the computer operating system by calling the
FORTRAN library subroutines IDATE and TIME.
ASORT sorts an array of character names into alphabetical order.
CELLM returns the Northeast and Southwest corner latitudes and longitudes of one of three regions
(NEROS1, SEROS1, or ROMNET1). It also returns the number of columns and rows in the grid domain
and the grid cell latitude and longitude increments.
CLOCKI returns the time elapsed from the scenario start, and the current time step number.
CLOCK2 returns the current date and time, and the time elapsed from the scenario start.
DATTIM returns a formatted current date and time using subroutine ADATE.
FSKIPI positions formatted or unformatted sequential files.
INDEXI is a function subprogram that returns the position of a character name in a list of names.
IOCL is a function subprogram that returns a clause field in an I/O status word. In the Core Model, this
field is reported to the run log in case of an I/O error to assist you in determining the cause of the error.
JFILEZ is a function subprogram that opens a sequential file and returns the FORTRAN unit number to
which it is attached.
JFILE5 is a function subprogram that opens a sequential file with fixed length records and returns the
FORTRAN unit number to which it is attached.
JFILE6 is a function subprogram that opens a sequential file with FORTRAN carriage control and vari-
able record type to enable the DEC Data Transfer Facility (DTP) to pass binary VAX records to the IBM.
JFILE6 returns the FORTRAN unit number to which the file is attached. This subprogram is specific to
our installation, and may not pertain to you.
JULIAN is a function subprogram that returns a given calendar year, month and day as the Julian date
YYDDD.
JUNIT is a function subprogram that keeps track of the FORTRAN I/O units already assigned in the
model and returns the next available unit number.
23
-------
RDCHAR performs an unformatted read into a character buffer whose length is a passed argument.
RDFILE performs an unformatted read into a numeric buffer whose length is a passed argument.
WRCHAR performs an unformatted write from a character buffer of arbitrary length.
WRFILE performs an unformatted write from a numeric buffer of arbitrary length.
1.8 MISCELLANEOUS SUBPROGRAMS
BLKMOD
NEWICS
PRGSMY
TIMER
CPUTIM
BLKMOD contains the FORTRAN BLOCK DATA initializations for:
• the standard read and write I/O units,
• the process subroutine text pointers,
the global formats that are used to echo file header information to the run log when the files
have been successfully opened,
• the internal (logical) names for all the model files,
the 35 chemical species names used in ROM2.1,
the 3 model layer names,
• the STOPCK file's flag initial value,
• the concentration species values used to represent zero in the model, i.e., 1(H6 ppm.
the values for the basic rate constants,
the 11 photolytic rate constant levels used by RTPHO to determine the integral average photo-
lytic rate constants,
• the arrays of clear sky photolytic rate constants tabulated for the five CBM4.2 photolytic reac-
tions, the solar zenith angle, and the 11 photolytic rate constant levels.
NEWICS writes a copy of the CONC file's concentration data to the NEWICON file at the end of a model
execution scenario.
PRGSMY records summary information of all the subprograms used to create the executable image of the
model on the run log.
TIMER determines elapsed CPU and elapsed clock time using subroutine CPUTIM.
CPUTIM retrieves from a VAX system the CPU time elapsed since its initial call during a program execu-
tion.
24
-------
1.9 CORE MODEL OUTPUT DATA: TRANSFER FROM THE IBM TO THE VAX
The following procedure is specific to our installation, and may not pertain to you. After the model run
completes, we translate the CONC file to VAX binary format and send the file to the VAX through the
DTP. The program that does the translation is CNTRAN.FOR, which also verifies that all the time steps for
a run are present and backs up the untranslated file on the IBM. The data are translated to ASCII and
VAX floating point format. We show the JCL for this program below.
//XHST87SA JOB
-------
O3 > background in rural areas during episodic conditions;
boundary conditions have not significantly impacted the domain;
high-end magnitudes are not significantly > 200 - 300 ppb
- if the run is a base case, check against observed data
- if the run is a control strategy, check for ozone response in the expected direction.
For tracer species12 -
the existence of urban plumes. Note that tracer concentrations should be virtually
identical between runs with different emission control strategies but with the same
meteorology.
12. We perform quality control analyses on the tracer species plots only if the ozone plots indicate that a problem may exist.
26
-------
SECTION 2
CORE MODEL INPUT FILES
The Core Model expects eight files to be provided as input, four of which are produced by the processor net-
work. These four input files are BMAT (Section 2.2), BTRK (Section 2.3), BCON (Section 2.1) and ICON
(Section 2.4). The other four files are NEWICON (Section 2.5), the new initial conditions file that links one
three-day scenario model run with the next; PROG (Section 2.6), the progress file that enables you to monitor
the model execution; STOPCK (Section 2.8), the stop check file that lets you shut down the model during
execution; and RESTRT (Section 2.7), the file that allows you to restart the model run.
We need to draw your attention to two points that apply to the following documentation. First, during the
historical course of ROM development the terms "levels" and "layers" were used interchangeably and entirely
synonymously. For example, one variable is referenced as NLEVS, but we use the term "layers" in this docu-
ment. Second, we focus on the necessary information contained in the code; extraneous text is replaced by
ellipsis marks (. . .).
We refer you to Tables E-l and E-2 for the full list of values of the model- and region-specific parameters that
are contained in DiMENS.Exrand REGION.EXT.
2.1 THE BOUNDARY CONDITIONS (BCON) FILE
The boundary conditions file (BCON) contains chemical concentrations for each of the species required
by the Core Model. Concentration data are provided in each model layer for a perimeter that is one-cell
deep within the modeling domain.
The boundary conditions are used by subroutine BIGGAM during calculation of the advection component
of the current time step's concentration field. If the calculation is for a grid cell near the modeling
domain's boundary, values from the BCON file are substituted for the previous time step's concentrations
at the border cells where inflow exists.
Array dimensions are set by parameter statements contained in the INCLUDE files REGION.EXT and
DIMENS.EXT as follows:
INTEGERS NROWS, NCOLS
PARAMETER ( .... NROWS = 52. NCOLS = 64 )
INTEGER** NLEVS, NSPECS
PARAMETER (NLEVS = 3, NSPECS =35, ... )
27
-------
2.1.1 Opening the BCON File
At the start of the model scenario, BIGGAM calls BCPRCS. BCPRCS calls RDBCON, which opens the
BCON file by calling OPBCON, and which optionally positions the file to the starting time for the
model execution by calling POBCON. OPBCON reads the BCON file header records.
The code segments and the FORMAT statements that demonstrate the steps to open the BCON file
are listed below. FLNMBC contains the internal (logical) names for BCON that point to the actual
file names in the execution run stream. FLNMBC is set in the block data module BLKMOD. JUNIT
is a function subprogram that returns the next available FORTRAN I/O unit number.
INTEGER** ..... UNITBC, ...
CHARACTER*!? , .... FLNMBC, ...
SUBROUTINE OPBCON
1NTEGER*4 IOST, .... JMLE2
LOGICAL** RECFHT. RDONLY
C
PARAMETER ( RECFMT = .FALSE., RDONLY = .TRUE. )
C
C open BCON file
UNITBC = JFILE2 (FLNMBC, RECFMT, RDONLY)
C
FUNCTION JFILE2 (FNAME, RECFMT, RDONLY}
CHARACTER*12 FNAME, FORM, UNFORM, FORMAT
INTEGER** IDEV, IOST, JFILE2, ..., JUNIT,
LOGICAL** RECFMT, RDONLY
DATA FORM / 'FORMATTED ' /
DATA UNFORM / 'UNFORMATTED ' /
IDEV = JUNITO
IF (RECFMT) THEN
FORMAT = FORM
ELSE
FORMAT = UNFORM
END IF
IF (RDONLY) THEN
OPEN (UNIT = IDEV,
& 10STAT = IOST,
& . FILE = FNAME,
& STATUS * 'OLD',
& ACCESS = 'SEQUENTIAL*,
& FORM = FORMAT,
& READONLY)
ELSE
END IF
JFILE2 - IDEV
28
-------
RETURN
END
2.1.2 BCON File Records
The structure of the BCON file conforms to the requirements for all ROM Core Model files, and
therefore contains a standard file header. In addition, the file consists of a data body organized by
time steps, each section of which is headed by a time step header record. Descriptions of the records
containing this information are given below, and Appendix B contains a structure diagram for the
BCON file.
2.1.2.1 BCON File Header Records--
The first four records contain the BCON file header that consists of the variables in the
INCLUDE file HEADBC.EXT:
c
C HEADBC.EXT
C
BCON file header block
C
CHARACTER*80 TEXTBC
CHARACTER'S GRDNBC
CHARACTER** SPNHBC, LVNMBC
REAL** SWLNBC, SULTBC, NELNBC, NELTBC, DLONBC, DLATBC
INTEGER** CDATBC, CTIMBC, SDATBC, STHRBC, TSTPBC, FRSTBC,
& NCOLBC, NROWBC, NLEVBC, NSPCBC, ICNTBC
C
COMMON /CHARBC/ GRDNBC, SPNMBC(NSPECS), LVNMBC(NLEVS), TEXTBC(20)
COMMON /HEADBC/ CDATBC. CTIMBC, SDATBC, STHRBC, TSTPBC, FRSTBC,
& SWLNBC, SULTBC, NELNBC, NELTBC, DLONBC, DLATBC,
& NCOLBC, NROWBC, NLEVBC, NSPCBC, ICNTBC
C
2.1.2.1.1 Record 1- The first record contains character strings of alphanumeric data that
describe the file's contents. The data are first read (unformatted) by subroutine RDCHAR into
the buffer SEG1BF. The data are then read (formatted) into the variables contained in the
common blocks in the INCLUDE file HEADBC.EXT. The code segments and FORMAT
statements for these steps are shown below, and the variables of record 1 are shown in Table
3.
SUBROUTINE OPBCON
CHARACTER*(8 * 13 + * * 5) SEG1BF
C
C read 1st segment
CALL RDCHAR (UNITBC, SEG1BF, IOST)
C
SUBROUTINE RDCHAR (IUNIT, CHBUFF, IOST)
INTEGER** IUNIT, IOST
CHARACTER*^) CHBUFF
READUUNIT, IOSTAT = IOST) CHBUFF
29
-------
RETURN
END
===================
READ(SEG1BF, 1001, IOSTAT = IOST)
& CDATBC, CTINBC, SOATBC, STHRBC, TSTPBC, FRSTBC,
& GRDNBC,
& SULNBC, SWLTBC, NELNBC, NELTBC,
& DLONBC, DLATBC,
& NCOLBC, NROWBC, NLEVBC, NSPCBC, ICNTBC
1001 FORMATC6I8.8, AS. 4F8.3, 2F8.5, 5K.4)
TABLE 3. BCON RECORD 1 VARIABLES
Var
No.
1
2
3
4
5
6
1
8
9
10
11
12
13
14
15
16
17
18
Var
Name
CDATBC
CTIMBC
SDATBC
STHRBC
TSTPBC
FRSTBC
GRDNBC
SWLNBC
SWLTBC
NELNBC
NELTBC
DLONBC
DLATBC
NCOLBC
NROWBC
NLEVBC
NSPCBC .
ICNTBC
Unit
EST
EST
s
s
°W
°N
ew
°N
ew
"N
Data
Type
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Char*8
Real*4
Real*4
RealM
Real*4
Real*4
RealM
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Description
File creation date as MMDDYY
File creation time as HHMMSS
Julian start date of scenario as YYDDD
Start hour of scenario (00 to 23)
Time step size for simulation
Time to first step
Grid definition name
Longitude of southwest corner of grid
Latitude of southwest corner of grid
Longitude of northeast corner of grid
Latitude of northeast corner of grid
Grid cell longitudinal increment
Grid cell latitudinal increment
Number of columns in grid
Number of rows in grid
Number of levels
Number of model species
Number of levels in the simulation
Number of species in the BCON file
Number of text records
2.1.2.1.2 Record 2-r This record contains the list of species names for which the Core Model
computes concentration outputs. The data are first read by subroutine RDCHAR into the
buffer SPNMBF. The data are then copied into the SPNMBC array contained in a common
block in the INCLUDE file HEADBC.EXT. These steps are listed below, and the variable of
record 2 is shown in Table 4.
SUBROUTINE OPBCON
CHARACTER*^ * NSPECS) SPNMBF
INTEGERS .... ISPC
C read the species names record
CALL RDCHAR (UNITBC, SPNNBF, IOST)
READCSPNMBF, 1003, IOSTAT = IOST) (SPNMBC(ISPC), ISPC
1003 FORMAT(6(10(A4))/)
1, NSPCBC)
30
-------
TABLE 4. BCON RECORD 2 VARIABLE
Var
No.
1
Var
Name Unit
SPNMBQ
Data
Type
Char*4
Description
Name of chemical species k t
f A list of chemical species names can be found in Table 1.
2.1.2.13 Record3- This record contains the list of Core Model layer names. The data are
first read by subroutine RDCHAR into the buffer LVNMBF. The data are then copied into the
LVNMBC array contained in a common block in the INCLUDE file HEADBCEXT. These
steps are listed below, and the variable of record 3 is shown in Table 5.
SUBROUTINE OPBCON
CHARACTER*<4 * NLEVS) LEVNBF
INTEGER** , ..., I LEV
C read level names record
CALL RDCHAR (UNITBC, LEVNBF, IOST)
READ(LEVNBF, 1005, IOSTAT = IOST) (LVNMBC(ILEV), ILEV = 1, NLEVBC)
1005 FORNAT(2(10(A4))/>
TABLES. BCON RECORD 3 VARIABLE
Var
No.
1
Var
Name Unit
LVNMBQ,
Data
Type
Char*4
Description
Name of layer L
2.1.2.1.4 Records 4 - (4 + ICNTBO- These records contain descriptive text that was
entered when the file was created by processor P22G. Each record consists of one 80-char-
acter string. The data are read by subroutine RDCHAR into the buffer TEXTBF, which is then
copied into the TEXTBC array contained in a common block in the INCLUDE file
HEADBCEXT. These steps are listed below, and the variable of the records is shown in Table
6.
SUBROUTINE OPBCON
CHARACTER*80 TEXTBF
INTEGER** , .... ITXT
C read file text group
DO 101 ITXT = 1, ICNTBC
CALL RDCHAR (UNITBC, TEXTBF, IOST)
TEXTBC(ITXT) = TEXTBF
101 CONTINUE
31
-------
TABLE 6. BCON RECORDS 4 - (4+ICNTBC) VARIABLE
Var
No.
1
Var
Name
TEXTBQ,
Data
Unit Type
Cbar'80
Description
Text string of n lines
2.1.2.2 BCON File Body Records-
After the BCON file has been opened and the file header read, BIGGAM obtains the BCON data
by calling BCPRCS at the start of each model time step. BCPRCS in turn invokes RDBCON, which
reads (1) the BCON file's time step headers, and (2) each boundary's data by calling subroutine
RDFILE.
2.1.2.2.1 Time Step Header Record- There is one time step header record for each scenario
time step increment on the BCON file. Subroutine RDFILE is called to read four words of
data into the RTSHBC common block. These steps are listed below, and the record's vari-
ables are shown in Table 7.
REAL*4 DATBC. TIMBC, ELPBC, STPBC
COMMON /RTSHBC/ DATBC, TIMBC, ELPBC, STPBC
SUBROUTINE RDBCON
INTEGER** IOST, UDTSH, ...
PARAMETER ( NUDTSH = 4, ...)
C read BCON T.S.N.
CALL ROFILE (UNITBC, NUDTSH, DATBC, IOST)
C
SUBROUTINE RDFILE(IUNIT, NWORDS, BUFFER, IOST)
IMPLICIT NONE
INTEGER*4 IUNIT, NUORDS, BUFFER, IOST
DIMENSION BUFFER(NUORDS)
READdUNIT, IOSTAT=IOST) BUFFER
RETURN
END
TABLE 7. BCON TIME STEP HEADER RECORD VARIABLES
Var
No.
1
2
3
4
Var
Name
DATBC
TIMBC
ELPBC
STPBC
Data
Unit Type
Real*4
EST Real*4
s Real*4
Real'4
Description
Current time step Julian date as YYDDD
Current time step time as HHMMSS
Elapsed time since scenario start
Current time step number on BCON file
32
-------
2.1.2.2.2 Data Records- These records are read for each time step, and contain the boundary
values for species concentrations (by layer) for all modeling domain border cells. Note that
the northern and southern boundaries have data referenced by columns, while the western
and eastern boundaries have data referenced by rows. The code for these steps is listed below,
and the data variables are shown in Table 8.
REAL*4 NORTH, WEST, EAST, SOUTH
C
COMMON /BCFILE/ WEST(NROUS, NLEVS, NSPECS),
& EAST(NROWS, NLEVS, NSPECS).
& NORTH(NCOLS, NLEVS, NSPECS),
& SOUTH(NCOLS, NLEVS, NSPECS)
C
SUBROUTINE RDBCON
INTEGER*4 , .... NUDTSH, ROUWRD, COLURO, ...
C define record sizes
PARAMETER ( NUDTSH = 4,
& COLWRD = NLEVS * NCOLS,
& ROWWRD = NLEVS * NROWS )
C
C read Western boundary conditions species*
C
DO 201 ISPC = 1, NSPECS
CALL RDFILE (UNITBC, ROWWRD, WEST(1,1,ISPC), IOST)
201 CONTINUE
C
C read Eastern boundary conditions species*
C
DO 301 ISPC = 1, NSPECS
CALL RDFILE (UNITBC, ROWWRD, EAST(1,1,ISPC), IOST)
301 CONTINUE
C
C read Northern boundary conditions species*
C
DO 401 ISPC = 1, NSPECS
CALL RDFILE (UNITBC, COLWRD, NORTH(1,1,ISPC), IOST)
401 CONTINUE
C
C read Southern boundary conditions '. . . species*
C
DO 501 ISPC = 1, NSPECS
CALL RDFILE (UNITBC, COLWRD, SOUTH(1,1,ISPC), IOST)
501 CONTINUE
SUBROUTINE RDFILEdUNIT, NWORDS, BUFFER, IOST)
IMPLICIT NONE
INTEGER*4 IUNIT, NWORDS, BUFFER, IOST
DIMENSION BUFFER(NWORDS)
READ(IUNIT, JOSTAT=JOST) BUFFER
RETURN
END
33
-------
TABLES. BCON DATA VARIABLES
Var
No.
1
2
3
4
Var
Name
WEST),L
-------
2.2.1 Opening the BMAT File
At the start of the model scenario, LILGAM calls BMPRCS. BMPRCS calls RDBMAT, which opens the
BMAT file by calling OPBMAT, and which optionally positions the file to the starting time for the
model execution by calling POBMAT. RDBMAT reads the BMAT file header records by calling
RDMXBM.
The code segments and the FORMAT statements that demonstrate the steps to open the BMAT file
are listed below. FLNMBM contains the internal (logical) names for BMAT that point to the actual
file names in the execution run stream. FLNMBM is set in the block data module BLKMOD. JUNIT
is a function subprogram that returns the next available FORTRAN I/O unit number.
INTEGERS ..., UNITBM, ...
INTEGER** NUMBMF
PARAMETER ( NUMBMF = 6)
CHARACTER*12 .... FLNMBM(NUMBMF), ...
SUBROUTINE RDMXBM ( , , )
INTEGER** ..., ISUB, ...
LOGICAL** RECFMT, RDONLY
C
PARAMETER ( RECFMT = .FALSE., RDONLY = .TRUE.,
& RECLEN = NVRBM1 * NCOLS )
C
C open 1st subfile
UNITBM = JFILE6 (FLNMBM(ISUB), RECFMT, RDONLY, RECLEN)
C
=============s====ss===s===ss========ss==ss====s:==s=========s====r============
FUNCTION JFILE6 (FNAME, RECFMT, RDONLY, RECLEN)
CHARACTER*!2 FNAME, FORM, UNFORM, FORMAT
INTEGER** RECLEN, IDEV, IOST, JFILE6, JUNIT
LOGICAL** RECFMT, RDONLY
DATA FORM / 'FORMATTED ' /
DATA UNFORM / 'UNFORMATTED ' /
IDEV = JUNITO
IF (RECFMT) THEN
FORMAT - FORM
ELSE
FORMAT = UNFORM
IF
&
&
END IF
(RDONLY) THEN
OPEN (UNIT =
IOSTAT
FILE
STATUS
ACCESS
FORM
CARRIAGE CONTROL
RECORDTYPE
RECL
READONLY)
= IDEV,
IOST,
FNAME,
•OLD',
'SEQUENTIAL1,
FORMAT,
'FORTRAN',
'VARIABLE',
RECLEN,
ELSE
35
-------
END IF
JFILE6 = IDEV
RETURN
END
2.2.2 BMAT File Records
The structure of the B-matrix file conforms to the requirements for all ROM Core Model files, and
therefore contains a standard file header. In addition, the file consists of a data body organized by
time steps, each section of which is headed by a time step header record. Descriptions of the records
containing this information are given below. Appendix B contains a structure diagram for the BMAT
file.
2.2.2.1 BMAT File Header Records-
The first seven records contain the BMAT file header that comprise the variables in the
INCLUDE file HEADBM.EXT:
C HEADBM.EXT (ROM2.1 - Carbon Bond 4.2 Chemistry)
C
C BMATRIX file header block
C
CHARACTER*80 TEXTBH
CHARACTER*12 MFNNBN
CHARACTER'S GRDNBM
CHARACTER** SPNMBH, LVNMBM
REAL*4 SULNBN, SULTBH, NELTBM, NELNBM, DLONBM, DLATBM
INTEGER** ISUBFL, NSUBFL, FRSTSF, LSSTSF,
& CDATBM, CTINBM, CONFBN, CTMFBM, UDMFBM, UTMFBM,
& SDATBM, STHRBN, TSTPBN, FRSTBN, BMSPRD, BNINDX,
& NCOLBM, NROWBH, NLEVBN, NSPCBM, NMIFBM. ICNTBM,
& NMFBM, NVRBM1, NVRBN2
C
PARAMETER (NMFBM * 4, NVRBM1 = 18 + 3 * NLEVS + 6 * NPOXSP,
& NVRBM2 = 6)
C
C NMFBM number of MIF files used in generating BMATRIX file
C NVRBM1 number of part 1 BMATRIX variables,
C specifically; B12, B15. B21. B23, B32, B33,
C B11S, B11SS, B31S, B31SS, 003FAC, SSONO,
C TTHETA. PPSI2, ZZO. ZZ1. ZZ2, ZZ3;
C RRHO(NLEVS)'s. TTEMP(NLEVS)'s, WWVC(NLEVS)'s;
C and G1S. G1SS, G1FAC, G3S, G3SS, G3FAC, each
C dimensioned by NPOXSP
C NVRBM2 number of species-array BMATRIX variables, (part 2)
C specifically; B11, B22, B31, G1, G2, G3
C
COMMON /CHARBM/ GRDNBM, SPNMBM(NSPECS), LVNMBM(NLEVS), TEXTBM(20>,
& MFNMBM(NMFBM)
C
COMMON /HEADBM/ CDATBM, CTIMBM, SDATBM, STHRBM, TSTPBM, FRSTBM,
& SULNBH, SULTBM, NELNBM, NELTBM, DLONBM, DLATBM,
& NCOLBM, NROWBH, NLEVBM, NSPCBM, NMIFBM. ICNTBM,
& CDMFBM(NMFBM), CTMFBM(NMFBM), UDMFBM(NMFBM),
36
-------
& UTMFBM(NMFBM), BMINDX(NSPECS),
& ISUBFL, NSUBFL, FRSTSF, LSSTSF
COMMON /BMSPRD/ BMSPRD
2.2.2.1.1 Record 1- The first record contains character strings of alphanumeric data that
describe the file's contents. The data are first read (unformatted) into a character buffer that
is then read (formatted) into the variables contained in the common blocks in the INCLUDE
file HEADBM.EXT. These steps are shown below, and the variables of record 1 are shown in
Table 9.
INTEGER** ..., UNIIBM, ...
INTEGER** NUMBMF
PARAMETER ( NUMBMF - 6)
CHARACTER*!2 .... FLNMBM(NUMBMF), ...
SUBROUTINE RDMXBM ( , , )
INTEGER** ..., JFILE6, ISUB, ...
LOGICAL** RECFMT, RDONLY
C
PARAMETER ( RECFMT = .FALSE., RDONLY = .TRUE.,
& RECLEN = NVRBM1 * NCOLS )
C
C open 1st subfile
UNITBM = JFILE6 (FLNMBM(ISUB), RECFMT, RDONLY, RECLEN)
c-
C read subfile headers
CALL RDHDBM (ISUB)
SUBROUTINE RDHDBM (ISUB)
INTEGER** IOST, , , BFLONE, ...
PARAMETER ( BFLONE = 13 * 8 + 6 * *, ...)
CHARACTER*(BFLONE) RECONE
C
C read 1st segment record
C
READ(UNITBM, IOSTAT = IOST) RECONE
C convert character to mixed character & numeric and load HEADBM
C
READ(RECONE, 1001, IOSTAT = IOST)
& CDATBM, CTIMBM, SDATBM, STHRBM, TSTPBM, FRSTBM,
& GRDNBM, SWLNBM, SULTBM, NELNBM, NELTBM, DLONBM,
& DLATBM, NCOLBM, NROUBM, NLEVBM, NSPCBM, NMIFBM,
& ICNTBM
1001 FORMAT(6I8.8, A8, 4F8.3, 2F8.5, 61*.*)
37
-------
TABLE 9. B-MATRIX RECORD 1 VARIABLES
Var
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
Var
Name
CDATBM
CTIMBM
SDATBM
STHRBM
TSTPBM
FRSTBM
GRDNBM
SWLNBM
SWLTBM
NELNBM
NELTBM
DLONBM
DLATBM
NCOLBM
NROWBM
NLEVBM
NSPCBM
NMIFBM
ICNTBM
Unit
EST
EST
s
s
°W
°N
°W
°N
"W
°N
Data
-type
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Char*8
Real*4
Real*4
RealM
Real*4
RealM
Real*4
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Description
File creation date as MMDDYY
File creation time as HHMMSS
Julian start date of scenario as YYDDD
Stan hour of scenario (00 to 23)
Time step size for simulation
Time to first step
Grid definition name
Longitude of southwest corner of grid
Latitude of southwest corner of grid
Longitude of northeast corner of grid
Latitude of northeast corner of grid
Grid cell longitudinal increment
Grid cell latitudinal increment
Number of columns in grid
Number of rows in grid
Number of levels
Number of model species
Number of model input files used to generate
the B-matrix file
Number of text records
2.2.2.1.2 Record 2- This record contains the list of species names for which the Core Model
computes concentration outputs. The data are read directly into the SPNMBM array
contained in a common block in the INCLUDE file HEADBM.EXT. These steps are listed
below, and the variable of record 2 is shown in Table 10.
SUBROUTINE RDHDBM (ISUB)
INTEGER** .... ISPC, ...
C read species names record
C
READUMITBM, 10STAT = IOST) (SPNMBM(ISPC), ISPC = 1, NSPCBM)
TABLE 10. B-MATRIX RECORD 2 VARIABLE
Var
No.
Var
Name
Unit
Data
Type
Description
1
SPNMBM*
Char*4
Name of chemical species k t
f A list of chemical species names can be found in Table 1.
38
-------
2.2.2.1.3 Record 3-- This record contains the expansion list (look-up table) for the 26 species
names for which the BMAT file has concentration values. (Refer to Processor P23G in Part
2 of the ROM User's Guide for an explanation of the need for an expansion list.) These
species names are mapped to the full Core Model list of species names. Mapping is necessary
because the Core Model requires concentration values for species that have neither mea-
sured nor computed data, such as the free radicals. The expansion list tells the Core Model
how to assign concentration values for the full 35 species of the chemistry mechanism from
the 26 species concentrations in the BMAT file. The mapping from BMAT to the Core
Model is shown in Figure 7. The READ and FORMAT statements for this record are listed
below, and its variable is shown in Table 11.
MODEL
Figure 7. Chemical species mapping from BMAT to the Core Model. Species names are
defined in Section 1.2.2.
SUBROUTINE ROHOBM (ISUB)
INTEGER** .... BFLNDX, ...
PARAMETER ( .... BFLNDX = NSPECS * 4,
CHARACTER*(BFLNDX) RECNOX
C
C read the index group record
C
REAOCUNITBM, IOSTAT = IOST) RECNDX
C
39
-------
C convert character to numeric and load HEADBM
C
READCRECNDX, 1003, IOSTAT = IOST) (BHINDX(ISPC), ISPC = 1, NSPCBM)
1003 FORMAT «NSPCBM> 14)
TABLE 11. B-MATRIX RECORD 3 VARIABLE
Var
No.
Var Data
Name Unit Type
Description
1 BMINDX,, Integer*4 Mapping array of the 26 species names whose
values exist on the reduced B-matrix set, with
indices for expansion to the full model set
2.2.2.1.4 Record 4-- This record contains the list of Core Model layer names. The READ
and FORMAT statements for this record are listed below, and its variable is shown in Table
12.
SUBROUTINE RDHDBM (ISUB)
INTEGER*4 .... LEV. ...
C
C read the level names record
C
READUJNITBM, IOSTAT = IOST) (LVNMBN(LEV), LEV = 1, NLEVBM)
TABLE 12. B-MATRIX RECORD 4 VARIABLE
Var
No.
1
Var
Name Unit
LVNMBML
Data
Type
Char*4
Description
Name of layer L
2.2.2.1.5 Records S - 8-- These records contain the list of logical names of the model input
files (MF) that are used to generate the B-matrix data. (The logical names can change from
run to run.) The data are read (unformatted) into the RECMIF buffer and then read
(formatted) into the HEADBM.EXT variables as listed below. The variables are shown in Table
13.
SUBROUTINE RDHDBM (ISUB)
INTEGER** .... BFLMIF. , , IHIF, ...
C
PARAMETER ( .... BFLMIF = 12 + 4 * 8, ... )
CHARACTER*(BFLMIF) RECMIF
C
C read the MIF data records
C
DO 101 IMIF = 1, NMIFBM
READ(UNITBM, IOSTAT = IOST) RECMIF
40
-------
READCRECMIF, 1007, IOSTAT = IOST)
& MFNMBM(IMIF), CDMFBM(IMIF), CTMFBM(IMIF),
& UDMFBH(IMIF), UTMFBM(INIF)
1007 FORMAT(A12, 418.8)
101 CONTINUE
TABLE 13. B-MATRIX RECORDS 5 - 8 VARIABLES
Var
No.
1
2
3
4
5
2.2.2.1
Var
Name
MFNMBMn
CDMFBMrt
CTMFBMn
UDMFBH,
UTMFBMrt
.6 Records 9 -
Unit
EST
EST
Data
TVpe
Char*12
Integer*4
Integer*4
Integer*4
Integer*4
(28 maximum)- These
Description
File name of MIF file n
Creation date of MIF file n
Creation time of MIF file «
Last update date of MIF file n
Last update time of MIF file n
records contain descriptive text that was
entered when the file was created by processor P40G. Each record consists of one 80-char-
acter string. The READ and FORMAT statements for these records are listed below, and
their variable is shown in Table 14.
SUBROUTINE RDHDBM (ISUB)
INTEGER** .... BFLTXT. , , ITXT, ...
C
PARAMETER ( .... BFLTXT =80, ... )
CHARACTER* RECTXT
C
C read header text records
C
DO 201 ITXT = 1, ICNTBM
READCUNITBM, IOSTAT = IOST) RECTXT
C
READ(RECTXT, 1009, IOSTAT a IOST) TEXTBH(ITXT)
1009 FORHAT(ABO)
C
201 CONTINUE
TABLE 14. B-MATRIX RECORDS 9 - {9+ICNTBM-l) VARIABLES
Var
No.
1
Var
Name
TEXTBH,
Data
Unit Type
Char*80
Description
Text string of n lines
2.2.2.1.7 Subfile-Order Record- The purpose of this record is to ensure the correct ordering
of input data when the Core Model reads several B-matrix subfiles. The READ and
FORMAT statements for this record are listed below, and its variables are shown in Table 15.
41
-------
SUBROUTINE RDHDBM (ISUB)
INTEGER* A ..., BFLFLC, . , I SUB
C
PARAMETER ( .... BFLFLC = 4 * 4 )
CHARACTER*(BFLFLC> FLCREC
C
C read subfile header record
C
READ(UNITBM, IOSTAT = IOST) FLCREC
C
C convert character to mineric and load HEADBM
C
RE AD (FLCREC, 1013, IOSTAT = IOST) ISUBFL, NSUBFL, FRSTSF, LSSTSF
1013 FORMAT(4I4)
TABLE 15. B-MATRIX SUBFILE-ORDER RECORD VARIABLES
Var
No.
1
2
3
4
Var
Name Unit
ISUBFL
NSUBFL
FRSTSF
LSSTSF
Data
Type
Integer*4
Integer*4
Integer*4
Integer*4
Description
Ordinal for this subfile
Total number of subfiles for this B-matrix file
First time step count for this subfile
Last time step count for this subfile
2.2.2.2 BMAT File Body Records-
After the BMAT file has been opened and the file header read, LILGAM calls BMPRCS row by row
to obtain the BMAT data. BMPRCS in turn invokes RDBMAT, which reads (1) the BMAT file's
time step headers, and (2) each row of data by calling RDMXBM.
2.2.2.2.1 Time Step Header Record- There is one time step header record written for each
time step increment on the B-matrix file. The following code reads the time step data on the
BMAT file into the time step header common block, RTSHBM, by referencing the first vari-
able of the common block and specifying the number of the block's words to be read. These
steps, and the FORMAT statements for reading this record, are listed below, the record's
variables are shown in Table 16.
REAL*4 DATBM, TIMBM, ELPBM, STPBM
COMMON / RTSHBM / OATBM, TIMBM, ELPBM, STPBM
SUBROUTINE RDBMAT (IOST)
INTEGER** NUDTSH, .... IOST
PARAMETER ( NUDTSH =4 .... )
C
C read BMAT T.S.H.
CALL RDMXBM (NUDTSH, DATBM, IOST)
C
SUBROUTINE RDMXBM (NUORDS, BUFF, IOST)
REAL*4 BUFF
INTEGER** NUORDS, IOST, ...
C
42
-------
C read record
C
READ (UNITBM, IOSTAT = IOST) BUFF
TABLE 16. B-MATRIX TIME STEP HEADER RECORD VARIABLES
Var
No.
1
2
3
4
2.2.2.2.2
Var
Name
DATBM
TIMBM
ELPBM
STPBM
Unit
EST
s
Data
Type
Integer*4
Integer*4
Integer*4
Integer*4
Description
Current time step Julian date as YYDDD
Current time step time as HHMMSS
Elapsed time since scenario start
Current time step number on B-matrix file
Data Records- There are two groups of data records for each B-matrix time step
increment. The first group (BMAT Part 1 variables, Table 17) contains one record of
variables for each model domain row. These variables consist of:
interlayer flux parameters independent of the model chemical species (B12, B13,
B21, B23, B32, B33, BUS, BUSS, B31S, B31SS.QO3FAC, SSONO);
• ambient conditions data for the chemical rate constants adjustments (TTHETA,
PPSI2, ZZO, ZZ1, ZZ2, ZZ3, RRHO, TTEMP, WWVC); and
• volume flux emissions source parameters for the two primary oxidant species
emissions, NO and NO2 (G1S, G1SS, G1FAC, G3S, G3SS, G3FAC).
The second group (BMAT Part 2 variables, Table 18) contains a data record for each model
domain row for each chemical species in the B-matrix reduced set. This set is expanded to the
full model set by means of the look-up table BMINDX that is read from the file header (see
Section 2.2.2.1.3). These variables consist of:
• interlayer flux parameters and volume flux emissions source parameters that are
dependent on the model chemical species (Bll, B22, B31, Gl, G2, G3).
The following parameters from INCLUDE file HEADBM.EXT are used for dimensioning the
common blocks and controlling the read operations:
NVRBM1 = 18 + SxNLEVS + 6xNPOXSP, where NLEVS = 3 and NPOXSP = 3, the
number of primary oxidant species (NO, NO2, and 03);
NVRBM1 is the number of Part 1 B-matrix variables, specifically: B12, B13, B21, B23, B32,
B33, BUS, BUSS, B31S, B31SS, QO3FAC, SSONO, TTHETA, PPSI2, ZZO,
ZZ1, ZZ2, ZZ3; RRHO(NLEVS)'s, TTEMP(NLEVS)'s, WWVC(NLEVS)'s;
and G1S, G1SS, G1FAC, G3S, G3SS, G3FAC, each dimensioned by NPOXSP;
NVRBM2 =6;
43
-------
NVRBM2 is the number of species-array B-matrix variables (Part 2), specifically: Bll, B22,
B31,01, G2,03.
The first group (Part 1) is read into the common block BMFILE by referencing the first
variable of the common block and specifying the number of the block's words to be loaded.
Subsequently, the second group (Part 2) is read into the last variable in BMFILE through an
iteration of all the species on the BMAT file, again by specifying the number of the variables
words to be read. These steps are listed below.
COMMON /BMFILE/
& XB12(NCOLS), XB13(NCOLS),
& XB2KNCOLS), XB23(NCOLS),
& XB32(NCOLS), XB33(NCOLS),
& XBIIS(NCOLS), XBHSS(NCOLS),
& XB31SCNCOLS), XB31SS(NCOLS),
& XQ03FC(NCOLS), XSSONO(NCOLS),
& XG1S(NCOLS, NPOXSP), XG1SS(NCOLS, NPOXSP),
& XG1FAC(NCOLS, NPOXSP),
& XG3S(NCOLS, NPOXSP), XG3SS(NCOLS, NPOXSP),
& XG3FAC(NCOLS, NPOXSP),
& XRHOCNCOLS, NLEVS), XTENP(NCOLS, NLEVS),
& XUVC(NCOLS, NLEVS),
& XTHETA(NCOLS), XPSI2(NCOLS),
& XZO(NCOLS), XZI(NCOLS), XZ2(NCOLS), XZ3(NCOLS),
& AAUNCOLS, 6, NSPECS)
C
SUBROUTINE RDBMAT (IOST)
INTEGER** ..., NUDBM1, NUDBM2, ISPC, IOST
PARAMETER ( ..., NUDBM1 = NVRBM1 * NCOLS, NUDBM2 = NVRBM2 * NCOLS )
C
C read BMAT part 1
CALL RDMXBM (NUDBM1, XB12, IOST)
C read BMAT part 2 (reduced species list)
DO 301 ISPC = 1, BMSPRO
CALL RDMXBM (NUDBM2, AA1(1,1,ISPC), IOST)
301 CONTINUE
Subroutine BMPRCS moves the BMFILE data into another common block, LGBMFL. The
calculations using these data are made in subroutine LILGAM, which references LGBMFL:
COMMON /LGBMFL/
& B12CNCOLS), B13CNCOLS).
& B2KNCOLS), B23(NCOLS),
B32(NCOLS), B33(NCOLS),
B1KNCOLS, NSPECS), B22(NCOLS, NSPECS), B3KNCOLS, NSPECS),
GKNCOLS, NSPECS), G2(NCOLS, NSPECS), G3(NCOLS, NSPECS),
B11S(NCOLS), BIISS(NCOLS),
B31S(NCOLS), B31SS(NCOLS),
Q03FAC(NCOLS), SSONO(NCOLS),
G1S(NCOLS, NPOXSP), G1SS(NCOLS, NPOXSP),
G1FAC(NCOLS, NPOXSP),
G3S(NCOLS, NPOXSP), G3SS(NCOLS, NPOXSP),
G3FAC(NCOLS, NPOXSP),
RHO(NCOLS, NLEVS), TEMP(NCOLS, NLEVS), WVC(NCOLS, NLEVS),
THETA(NCOLS), PSI2(NCOLS),
ZLEV(NCOLS, NLEVS + 1)
SUBROUTINE BMPRCS
C transfer BM data into model ordered arrays and
C produce LILGAM BMAT record
44
-------
DO 301 ISPC - 1, NSPCIN
INDEX = NXSPBMdSPC)
DO 301 ICOL - 1, NCOLIN
B1KICOL, ISPC) = AAUICOL, 1, INDEX)
B22(ICOL, ISPC) = AAUICOL, 2, INDEX)
B3KICOL, ISPC) = AAUICOL, 3, INDEX)
AAUICOL, 4, INDEX)
AAUICOL, 5, INDEX)
AAUICOL, 6, INDEX)
GUICOL, ISPC)
G2UCOL, ISPC)
G3(ICOL, ISPC)
CONTINUE
301
C
C move LILGAM data from BN file organized by col & lev
DO 401 ILEV = 1, NLEVIN
DO 401 ICOL - 1, NCOLIN
RHOCICOL, I LEV) = XRHOUCOL, I LEV)
TEMPCICOL, ILEV) * XTEMPCICOL, I LEV)
WVCdCOL, ILEV) = XUVCdCOL, ILEV)
401 CONTINUE
C
C move LILGAM data from BM file organized by col & primary ox.
DO 501 ISPC = 1, NPOXSP
DO 501 ICOL = 1, NCOLS
G1SOCOL, ISPC) = XG1S(ICOL, ISPC)
G1SS(ICOL, ISPC) = XG1SSCICOL, ISPC)
G1FAC(ICOL, ISPC) = XGIFACdCOL, ISPC)
G3SUCOL. ISPC) = XG3S(ICOL, ISPC)
G3SSCICOL, ISPC) = XG3SS(ICOL. ISPC)
G3FAC(ICOL, ISPC) = XG3FAC(ICOL, ISPC)
501 CONTINUE
C
C move remaining LILGAM data from BM file organized by col
DO 601 ICOL = 1, NCOLIN
812(1 COL) = XB12OCOL)
B13UCOL) = XB13CICOL)
B2UICOL) = XB2UICOL)
B23UCOL) = XB23(ICOL)
B32(ICOL) - XB32(ICOL)
B33(ICOL) = XB33CICOL)
spec.
601
BIIS(ICOL)
B11SS(ICOL)
B31SCICOL)
B31SS(ICOL)
Q03FAC(ICOL)
SSONOdCOL)
THETAdCOL)
PSI2(ICOL) -
ZLEVdCOL, 1)
ZLEVdCOL, 2)
ZLEVdCOL, 3)
ZLEVdCOL, 4)
CONTINUE
XB11S(ICOL)
XBllSSdCOL)
XB31S(ICOL)
XB31SS(ICOL)•
-- XQ03FCdCOL)
XSSONO(ICOL)
XTHETAdCOL)
XPSI2CICOL)
= XZOdCOL)
= XZUICOL)
= XZ2(ICOL)
= XZ3CICOL)
45
-------
TABLE 17. B-MATRIX DATA VARIABLES, PART 1
Var
No.
1
2
3
4
5
6
7
8
.9
10
11
12
13
14
15
16
17
18
Var
Name Unit
B12, s-i
B13, s-i
B21, s-i
B23, s-i
B32, s-i
B33, s-i
BUS, s-i
BUSS, s-i
B31S, s-i
B31SS, s-i
QO3FAQ m-s-i
SSONO, ppm-m-s-i
CIS,-* ppm-s-1
G1SS,* ppm-s-l
GIFAQjt s-i
G3S,pjt ppm-S'l
G3SS,gfc ppm-s-i
G3FAQ* s-i
Data
Type
RealM
RealM
RealM
RealM
RealM
RealM
RealM
RealM
RealM
RealM
RealM
RealM
RealM
RealM
RealM
RealM
RealM
RealM
Description
B-matrix coefficient in column / for:
layer-l/surface-2 flux
layer- l/surface-3 flux
layer-2/surface-l flux
layer-2/surface-3 flux
layer-3/surface-2 flux
layer-3/surface-3 flux
B-matrix coefficient for subgrid scale adjust-
ment in column i for:
layer-l/surface-1 flux
alternate layer- I/surface- 1 flux
layer-3/surface-l flux
alternate layer-3/surface-l flux
Run time subgrid-scale adjustment parame-
ters in column i:
ozone factor
NO surface emissions source strength
emissions source term in layer 1 for pri-
mary oxidant species k
alternate emissions source term in layer 1
for primary oxidant species k
emissions source factor in layer 1 for pri-
mary oxidant species k
emissions source term in layer 3 for pri-
mary oxidant species k
alternate emissions source term in layer 3
for primary oxidant species k
emissions source factor in layer 3 for pri-
mary oxidant species k
continued
46
-------
TABLE 17 (concluded)
Var
No.
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
Var
Name
RHOM
RHOJ>2
RHQw
TEMP/,/
TEMPI>2
TEMPj.5
WVQ(7
WVQ.2
WVQ.3
THETA,
PSI2/
ZLEV.,0
ZLEVW
ZLEV/)2
ZLEV.-.3
Unit
mol-ppm-m-3
mol-ppm-m-3
mol-ppm-m-3
K
K
K
ppm
ppm
ppm
deg
m
m
m
m
Data
Type
Real*4
Real*4
Real*4
Real*4
RealM
Real*4
RealM
RealM
RealM
RealM
RealM
RealM
RealM
RealM
RealM
Description
Rate constants density correction factor in
column i for:
layer 1
layer 2
layer 3
Absolute temperature for rate constants
adjustment in column i for:
layer 1
layer 2
layer 3
Water vapor concentration for rate con-
stants adjustment in column i for:
layer 1
layer 2
layer 3
Solar zenith angle for photolytic rate con-
stants adjustment in column /
Cloud cover correction factor for photolytic
rate constants adjustment in column i
Heights above sea level in column i (used for
rate constant adjustments):
layer 0
layer 1
layer 2
layer 3
47
-------
TABLE 18. B-MATRIX DATA VARIABLES, PART 2
Var
No.
34
35
36
37
38
39
Var
Name Unit
Blli* s-1
B22,-* s-i
B31* s-i
Gljjc ppm-S"1
G2,* ppm-s-i
G3/* ppm-S'i
Data
Type
Real*4
Real*4
Real*4
Real*4
Real*4
Real*4
Description
B-matrix coefficient in column /
species k for:
layer- 1/surface-l flux
layer-2/surface-2 flux
layer-3/surface-l flux
Emissions source term in column /
species k for:
layer 1
layer 2
layer 3
for
for
2.3 THE BACKTRACK (BTRK) FILE
The Backtrack file (BTRK) contains the gridded backtrack locations and the horizontal diffusivities data
for each grid cell, model layer, and time step required by the Core Model. A backtrack location is the
position in grid cell coordinates of the fluid particle that will arrive at a grid cell corner during the next
30-minute advection time step. The backtrack locations and diffusivities are used by subroutine BIGGAM
in the calculation of the advection component of the current time step's concentration field.
Array dimensions are set by parameter statements contained in the INCLUDE files REGION.EXT and
DIMENS.EXT as follows:
INTEGER*^ .... NCOLS
PARAMETER ( .... NCOLS = 64 )
INTEGER*4 NLEVS, ...
PARAMETER (NLEVS =3, ...)
2.3.1 Opening the BTRK File
At the stan of the model scenario, BIGGAM calls BTPRCS. BTPRCS calls RDBTRK, which opens the
BTRK file by calling OPBTRK, and which optionally positions the file to the starting time for the
model execution by calling POBTRK. RDBTRK reads the BTRK file header records by calling RDBT.
The code segments and the FORMAT statements that demonstrate the steps to open the BTRK file
are listed below. FLNMBT contains the internal (logical) names for BTRK that point to the actual
file names in the execution run stream. FLNMBT is set in the block data module BLKMOD.
NVARBT is the number of layer-dependent BTRK variables (4). JUNIT is a function subprogram
that returns the next available FORTRAN I/O unit number.
48
-------
INTEGER** ..., UNITBT, ...
CHARACTER*12 .... FLMHBT, ...
SUBROUTINE RDBT (...)
LOGICAL** RECFMT, RDONLY
INTEGER** ..., JFILE6, RECLEN
PARAMETER ( ....
& RECLEN = NVARBT * NCOLS * NLEVS,
& RECFMT = .FALSE., RDONLY = .TRUE.)
C open file
UNITBT = JFILE6 (FLNMBT, RECFMT, RDONLY, RECLEN)
C
FUNCTION JFILE6 (FNAME, RECFMT, RDONLY, RECLEN)
CHARACTER*! 2 FNAME, FORM, UNFORM, FORMAT
INTEGER** RECLEN, IDEV, IOST, JFILE6, JUNIT
LOGICAL** RECFMT, RDONLY
DATA FORM / 'FORMATTED ' /
DATA UNFORM / 'UNFORMATTED ' /
IDEV = JUNITO
IF (RECFMT) THEN
FORMAT = FORM
ELSE
FORMAT = UNFORM
IF
&
&
&
&
&
&
&
&
&
END IF
(RDONLY) THEN
OPEN (UNIT
IOST AT
FILE
STATUS
ACCESS
FORM
CARRIAGECONTROL
RECORDTYPE
RECL
READONLY)
= IDEV,
= IOST,
= FNAME,
= 'OLD1,
= 'SEQUENTIAL
= FORMAT,
= 'FORTRAN1,
= 'VARIABLE',
= RECLEN,
ELSE
END IF
JFILE6 = IDEV
RETURN
END
2.3.2 BTRK File Records
The structure of the Backtrack file conforms to the requirements for all ROM Core Model files, and
therefore contains a standard file header. In addition, the file consists of a data body organized by
49
-------
time steps, each section of which is headed by a time step header record. Descriptions of the records
containing this information are given below. Appendix B contains a structure diagram for the BTRK
file.
2.3.2.1 BTRK File Header Records-
The first three records contain the BTRK file header that comprises of the variables in the
INCLUDE file HEADBT.EXT:
C HEADBT.EXT
C
C BCKTRAK file header block
C
CHARACTER*80 TEXTBT
CHARACTER*12 MFNMBT
CHARACTER'S GRDNBT
REAL*4 SWLNBT, SULTBT, NELTBT, NELNBT, DIONBT, DLATBT
INTEGER** CDATBT. CTIMBT, COMFBT, CTMFBT, UDMFBT, UTMFBT,
& SOATBT, STHRBT, TSTPBT, FRSTBT,
& NCOLBT, NROUBT, NMIFBT, ICNTBT,
& NMFBT, NVARBT
C
PARAMETER (NMFBT = 6, NVARBT - 4)
C
C NMFBT number of MIF files used in generating BCKTRAK file
C NVARBT number of LEVEL-DEPENDENT BCKTRAK variables,
C specifically; UU's, W's, KKHU's, KKHV's
C
COMMON /CNARBT/ GRDNBT, TEXTBT(20), MFNMBT(NMFBT)
C
COMMON /HEADBT/ COATBT. CTIMBT, SDATBT, STHRBT, TSTPBT, FRSTBT,
& SWLNBT, SULTBT, NELNBT, NELTBT, DLONBT, DLATBT,
& NCOLBT, NROUBT, ICNTBT,
& NMIFBT.
& CDMFBT(NNFBT), CTMFBT(NMFBT), UDMFBT(NMFBT),
& UTMFBT(NMFBT)
2.3.2.1.1 Record I- The first record contains character strings of alphanumeric data that
describe the file's contents. The data are first read (unformatted) into a character buffer that
is then convened into the record variables by a formatted internal read statement. The
ROBT code segments and FORMAT statements for these steps are shown below, and the
variables of record 1 are shown in Table 19.
SUBROUTINE RDBT (...)
PARAMETER (BFLONE = 13 * 8 + 4 * 4, ...
CHARACTER*(BFLONE) RECONE
C
C read header record
C
READ(UNITBT, IOSTAT = IOST) RECONE
C
C convert character to mixed character & numeric
C
READ(RECONE, 1001, IOSTAT = IOST)
& CDATBT, CTIMBT, SDATBT, STHRBT, TSTPBT, FRSTBT,
& GRDNBT, SULNBT, SULTBT, NELNBT, NELTBT, DLONBT,
& DLATBT, NCOLBT, NROUBT, NMIFBT, ICNTBT
1001 FORMAT(6I8.8. A8, 4F8.3, 2F8.5, 414.4)
50
-------
TABLE 19. BTRK RECORD 1 VARIABLES
Var
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
Var
Name
CDATBT
CTIMBT
SDATBT
STHRBT
TSTPBT
FRSTBT
GRDNBT
SWLNBT
SWLTBT
NELNBT
NELTBT
DLONBT
DLATBT
NCOLBT
NROWBT
NMIFBT
ICNTBT
Unit
EST
EST
s
s
°W
•N
°W
"N
CW
°N
Data
Type
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Char'8
Real*4
RealM
RealM
Real*4
Real*4
RealM
Integer*4
Integer*4
Integer*4
Integer*4
Description
File creation date as MMDDYY
File creation time as HHMMSS
Julian start date of scenario as YYDDD
Start hour of scenario (00 to 23)
Time step size for simulation
Time to first step
Grid definition name
Longitude of southwest corner of grid
Latitude of southwest corner of grid
Longitude of northeast corner of grid
Latitude of northeast corner of grid
Grid cell longitudinal increment
Grid cell latitudinal increment
Number of columns in grid
Number of rows in grid
Number of model input files used to generate
the BTRK file
Number of text records
2.3.2.1.2 Records 2 - 5- These records contain the list of model input file logical names used
to generate the BTRK file data; the list records are processed in the same way as record 1.
The variables of these records are shown in Table 20.
SUBROUTINE RDBT (...)
PARAMETER ( .... BFLNIF = 12 + 4 * 8, ...
CHARACTER*(BFLMIF) RECMIF
C
C read the MIF data records
C
DO 101 IMIF = 1, NMIFBT
REWKUNITBT, IOSTAT = IOST) RECMIF
READ(RECMIF, 1005, IOSTAT = IOST)
& MFNMBT(IMIF), COMFBT(IMIF), CTMFBT(IMIF),
& UDMFBT(IMIF), UTMFBT(IMIF)
1005 FORMATCA12, 418.8)
101 CONTINUE
51
-------
TABLE20. BTRKRECORDS2-5VARIABLES
Var
No.
1
2
3
4
5
Var
Name
MFNMBTrt
CDMFBTn
CTMFBTn
UDMFBTn
UTMFBTn
2.3.2.1.3 Records 6 -
Unit
EST
EST
Data
Type
Char*12
Integer*4
Integer*4
Integer*4
Integer*4
(25 maximum)-- These
Description
File name of MIF file n
Creation date of MIF file n
Creation time of MIF file n
Last update date of MIF file n
Last update time of MIF file n
records contain descriptive text that was
entered when the file was created by processor P38G. Each record consists of one 80-char-
acter string. The READ and FORMAT statements in subroutine RDBT for these records are
listed below, and their variable is shown in Table 21.
SUBROUTINE RDBT (...)
CHARACTER*80 RECTXT
C
C read header text records
C
DO 201 ITXT = 1. ICNTBT
READ (UNITBT, IOSTAT = IOST) RECTXT
READCRECTXT, 1007, IOSTAT = IOST) TEXTBT(ITXT)
1007 FORNAT(ASO)
201 CONTINUE
TABLE 21. BTRK RECORDS 6 - (6+ICNTBT-l) VARIABLES
Var
No.
1
Var
Name
TEXTBTW
Data
Unit Type
Char*80
Description
Text string of n lines
2.3.2.2 BTRK File Body Records--
After the BTRK file has been opened and the file header read, BICCAM obtains the BTRK data by
calling BTPRCS at the start of each model time step. BTPRCS in turn invokes RDBTRK, which reads
(1) the BTRK file's time step headers, and (2) each row of data by calling RDBT.
2.3.2.2.1 Time Step Header Record- There is one time step header record written for each
time step increment on the BTRK file. The time step header data are contained in the com-
mon block RTSHBT. RDBT reads the data from the BTRK file by specifying the number of
the block's words to be read (NWORDS) obtained from its argument list. The data are put
into the output buffer DATA, which is then passed back through RDBT'S argument list to
52
-------
RDBTRK. The common block RTSHBT is thereby loaded since the first variable in its address
space was referenced in the call to RDBT and the address space is exactly NWORDS long.
The RDBTRK and RDBT code segments and the FORMAT statements that demonstrate the
processing steps for this record are listed below, and its variables are shown in Table 22.
SUBROUTINE RDBTRK (IOST)
INTEGER** IUDLN1, ...
PARAMETER ( IUDLN1 = *, ...
REAL** DATBT, TIMBT, ELPBT, STPBT
COMMON / RTSHBT / DATBT, TIMBT, ELPBT, STPBT
C read BTRK T.S.H.
CALL RDBT (IUDLN1. DATBT. IOST)
C
SUBROUTINE RDBT (NWORDS, DATA, IOST)
INTEGER** NWORDS, IOST, ...
REAL** DATA
DIMENSION DATA(NWORDS)
READ (UNITBT, IOSTAT = IOST) DATA
TABLE 22. BTRK TIME STEP HEADER RECORD VARIABLES
Var
No.
1
2
3
4
2.3.2.2.2
Var
Name
DATBT
TIMBT
ELPBT
STPBT
Unit
EST
s
Data Records- The
Data
Type
Integer*4
Integer*4
Integer*4
Integer*4
data records
Description
Current time step Julian date as YYDDD
Current time step time as HHMMSS
Elapsed time since scenario start
Current time step number on BTRK file
are obtained from the BTRK file in the same
manner as the time step header records. RDBT reads a block of data IWDLN2 long from the
file into a buffer of the same length. Since the argument list refers to the address location of
the first variable in the data common block BTFILE, IWDLN2 words of file data are
automatically loaded into the common block's address space. The RDBTRK and RDBT code
segments and the FORMAT statements that demonstrate the processing steps for this record
are listed below, and its variables are shown in Table 23.
C NVARBT = 4 is the number of I eve I-dependent BCICTRAK variables,
specifically: XRU, XRV, XRKU, XRKV.
REAL** XRU, XRV, XRKU, XRKV
COMMON /BTFILE/ XRU(NCOLS, NLEVS), XRV(NCOLS. NLEVS),
& XRKU(NCOLS, NLEVS), XRKV(NCOLS,NLEVS)
SUBROUTINE RDBTRK (IOST)
INTEGER** .... IUDLN2, ...
PARAMETER ( .... IUDLN2 = ( NLEVS * NVARBT ) * NCOLS )
C read BTRK
CALL RDBT (IUDLN2, XRU, IOST)
53
-------
SUBROUTINE RDBT (NUORDS, DATA, IOST)
INTEGERS NWORDS, IOST, ...
REAL*« DATA
DIMENSION DATA(NWORDS)
READ (UNITBT, IOSTAT = IOST) DATA
TABLE 23. BTRK DATA VARIABLES
Var
No.
1
2
3
4
Var
Name
XRU/i
XRVi)L
XRKUJlL
XRKV/lL
Data
Unit Type
t Real*4
* Real*4
rad2*-1 Real*4
rad2*-1 Real*4
Description
Backtrack locations:
Longitudinal component, column i, layer L
Latitudinal component, column i, layer L
Horizontal diffusivities:
Longitudinal component, column i, layer L
Latitudinal component, column i, layer L
t Backtrack location units are fractional column numbers.
J Backtrack location units are fractional raw numbers.
2.4 THE INITIAL CONDITIONS (ICON) FILE
The initial conditions file (ICON) contains initial condition concentrations of all the model CBM 4.2
chemical species for every grid cell and each model layer in the model domain. These data are meant to
simulate the relatively clean background conditions that would prevail prior to the onset of an elevated
oxidant pollution episode. The Core Model uses these concentrations to start a multi-day simulation (an
episode). Thus, the ICON file is used only once for each episode modeled. The ICON file is not restricted
to contain data for only one time step, although in practice only one time step is read. In fact, a full
concentration file (the output from the model execution) can be used as an ICON file.
The initial concentrations are used by BIGGAM only at the very first time step of the episode, during cal-
culation of the advection component of the next time step's concentration field. For all subsequent epi-
sode time step calculations, BIGGAM uses the predicted concentrations calculated for the previous time
step.
During data processing, array dimensions are set by parameter statements contained in the INCLUDE
files REGION.EXT and DIMENS.EXT as follows:
INTEGER*^ NROWS, NCOLS
PARAMETER ( ..., NROWS = 52, NCOLS = 64 )
INTEGERS NLEVS, NSPECS, ..., NPOXSP
PARAMETER (NLEVS = 3, NSPECS =35, .... NPOXSP = 3)
54
-------
2.4.1 Opening the ICON File
At the start of the model scenario, BIGGAM calls ICPRCS. ICPRCS calls RDICON, which opens the
ICON file by calling OPICON, and which optionally positions the file to the starting time for the model
execution by calling POICON. RDICON calls RDFILE to read the ICON file header records. The code
segments and the FORMAT statements that demonstrate the steps to open the ICON file are listed
below. Note that all ICON file records have a fixed length equal to NLEVS x NCOLS.
INTEGER** .... UNITIC, ...
CHARACTER*12 .... FLNMIC, ...
SUBROUTINE OPICON
INTEGER** IUDLTH, ..., JFILE5
LOGICAL** RECFMT, RDONLY
PARAMETER ( IUDLTH = NCOLS * NLEVS,
& RECFMT = .FALSE., RDONLY = .TRUE. )
C
C open ICON file
UNITIC = JFILE5 (FLNMIC, RECFMT, RDONLY, IUDLTH)
C
FUNCTION JFILE5 (FNAME, RECFMT, RDONLY. RECLEN)
CHARACTER*12 FNAME, FORM, UNFORM, FORMAT
INTEGER** RECLEN, IDEV, IOST, JFILE5, JUNIT, ...
LOGICAL** RECFMT, RDONLY
DATA FORM / 'FORMATTED ' /
DATA UNFORM / 'UNFORMATTED ' /
IDEV = JUNITO
IF (RECFMT) THEN
FORMAT = FORM
ELSE
FORMAT = UNFORM
END IF
IF (RDONLY) THEN
OPEN (UNIT = IDEV,
& IOSTAT = .IOST,
& FILE
& STATUS
& ACCESS
& FORM
& RECORDTYPE
& RECL
& READONLY)
ELSE
END IF
JFILE5 = IDEV
FNAME,
'OLD',
'SEQUENTIAL'
FORMAT,
'FIXED',
RECLEN,
55
-------
RETURN
END
2.4.2 ICON File Records
The structure of the ICON file conforms to the requirements for all ROM Core Model files, and
therefore contains a standard file header. In addition, the file consists of a data body organized by
time steps, each section of which is headed by a time step header record (even though there may be
only one time step on the file). Descriptions of the records containing this information are given
below, and Appendix B contains a structure diagram for the ICON file.
2.4.2.1 ICON File Header Records-
The first four records contain the ICON file header that comprises the variables in the
INCLUDE file HEADIC.EXT:
C
C HEAD 1C.EXT
C
C ICON file header block
C
CHARACTER*80 TEXT 1C
CHARACTER'S GRDNIC
CHARACTERS SPNMIC, LVNHIC
REALS SULNIC, SWLTIC, NELNIC, NELTIC, DLONIC, DLATIC
INTEGERS CDATIC, CTIMIC, SOATIC, STHRIC, TSTPIC. FRSTIC,
& NCOLIC, NROWIC, NLEVIC, NSPCIC, ICNTIC, IDUM
C
COMMON /CHARIC/ GRDNIC, SPNMIC(NSPECS), LVNMIC(NLEVS), TEXTIC(20)
COMMON /HEADIC/ CDATIC, CTIMIC, SOAT1C, STHRIC, TSTPIC, FRSTIC,
& SULNIC, SUIT1C, NELNIC, NELTIC, DLONIC, OLATIC,
& NCOLIC, NROWIC, NLEVIC, NSPCIC, ICNTIC,
& IDUM(B)
2.4.2.1.1 Record 1- The first record contains character strings of alphanumeric data that
describe the file's contents. The data are first read (unformatted) by subroutine RDCHAR into
the buffer SEG1BF. The data are then read (formatted) into the variables contained in the
common blocks in the INCLUDE file HEADICEXT. The code segments and FORMAT
statements for these steps are shown below, and the variables of record 1 are shown in Table
24.
SUBROUTINE OPICON
CHARACTER*(8 * 21 + 4 * 5) SEG1BF
C
C read 1st segment
CALL RDCHAR (UNITIC, SEG1BF, IOST)
C
=s=s=s==sss=ss=ssss=ssffsssssss=s=s=ss=
SUBROUTINE RDCHAR (IUNIT, CHBUFF, IOST)
INTEGERS IUNIT, IOST
CHARACTER*^) CHBUFF
READdUNIT, IOSTAT = IOST} CHBUFF
56
-------
RETURN
END
READCSEG1BF, 1001, IOSTAT = IOST)
& COATIC, CTIMIC, SDATJC, STHRIC, TSTPIC, FRSTIC.
& GRDNIC,
& SULNIC, SWLTIC, NELNIC, NELTIC,
& DLONIC, DLATIC,
& NCOLIC, NROWIC, NLEV1C, NSPCIC,
& (IDUM(ITXT), ITXT = 1, 8),
& ICNTIC
1001 FORMAT(6I8.8, AS, 4F8.3, 2F8.5, 414.4, 818, 14.4)
TABLE 24. ICON RECORD 1 VARIABLES
Var
No.
1
2
3
4
5
6
1
8
9
10
11
12
13
14
15
16
17
18
19
Var
Name
CDATIC
CITMIC
SDATIC
STHRIC
TSTPIC
FRSTIC
GRDNIC
SWLNIC
SWLTIC
NELNIC
NELTIC
DLONIC
DLATIC
NCOLIC
NROWIC
NLEVIC
NSPCIC
IDUM
ICNTIC
Unit
EST
EST
s
s
°W
CN
°W
°N
°W
°N
Data
Type
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Char*8
Real*4
Real*4
Real*4
Real*4
RealM
RealM
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Description
File creation date as MMDDYY
File creation time as HHMMSS
Julian start date of scenario as YYDDD
Start hour of scenario (00 to 23)
Time step size for simulation
Time to first step
Grid definition name
Longitude of southwest corner of grid
Latitude of southwest corner of grid
Longitude of northeast corner of grid
Latitude of northeast corner of grid
Grid cell longitudinal increment
Grid cell latitudinal increment
Number of columns in grid
Number of rows in grid
Number of levels in the simulation
Number of species in the ICON file
Padding to fill buffer
Number of text records
2.4.2.1.2 Record 2— This record contains the list of species names for which the Core Model
computes concentration outputs. The data are first read by subroutine RDCHAR into the
buffer SPNMBF. The data are then copied into the SPNMIC array contained in a common
block in the INCLUDE file HEADICEXT. These steps are listed below, and the variable of
record 2 is shown in Table 25.
SUBROUTINE OPICON
CHARACTER*(4 * NSPECS) SPNMBF
INTEGER»4 ..., ISPC, ...
57
-------
C read the species names record
CALL RDCHAR (UNITIC, SPNMBF, IOST)
READCSPNMBF, 1003, IOSTAT = IOST) (SPNMIC(ISPC). ISPC = 1, NSPCIC)
1003 FORMAT( (A4) )
TABLE 25. ICON RECORD 2 VARIABLE
Var
No.
1
Var
Name Unit
SPNMIQ
Data
Type
Char*4
Description
Name of chemical species k t
f A list of chemical species names can be found in Table 1.
2.4.2.1.3 Record 3-- This record contains the list of Core Model layer names. The data are
first read by subroutine RDCHAR into the buffer LVNMBF. The data are then copied into the
LVNMIC array contained in a common block in the INCLUDE file HEADICEXT. These
steps are listed below, and the variable of record 3 is shown in Table 26.
SUBROUTINE OPICON
CHARACTER*(4 * NLEVS) LEVNBF
INTEGER** , .... ILEV
C read the level names record
CALL RDCHAR (UNITIC, LEVNBF, IOST)
READ(LEVNBF, 1005, IOSTAT = IOST) (LVNHIC(ILEV), ILEV = 1, NLEVIC)
1005 FORMAT( (A4) )
TABLE 26. ICON RECORD 3 VARIABLE
Var Var Data
No. Name Unit Type Description
1 LVNMIQ, Char*4 Name of layer L
2.4.2.1.4 Records 4- (4 + ICNTTO- These records contain descriptive text that was entered
when the file was created by processor P02G. Each record consists of one 80-character string.
The data are read by subroutine RDCHAR into the buffer TEXTBF, which is then copied into
the TEXTIC array contained in a common block in the INCLUDE file HEADICEXT. These
steps are listed below, and the variable of the records is shown in Table 27.
SUBROUTINE OPICON
CHARACTERS TEXTBF
INTEGER** .... ITXT, ...
C read file text group
DO 101 ITXT = 1, ICNTIC
CALL RDCHAR (UNITIC, TEXTBF. IOST)
TEXTIC(ITXT) = TEXTBF
101 CONTINUE
58
-------
TABLE 27. ICON RECORDS 4 - (4+ICNTIC) VARIABLE
Var
No.
1
Var
Name
TEXTIQ,
Data
Unit Type
Char*80
Description
Text string of n lines
2.4.2.2 ICON File Body Records--
After the ICON file has been opened and the file header read, BIGGAM obtains the ICON data by
calling ICPRCS before the start of the scenario time step iteration. ICPRCS, in turn, calls RDICON
for each row's data.
2.4.2.2.1 Time Step Header Record- There is one time step header record for each scenario
time step increment on the ICON file. Usually there is only one time step on this file, but this
is not an absolute restriction. Subroutine RDFILE is called to read four words of data into the
RTSHIC common block by referencing the first variable of the common block and specifying
the number of the block's words to be read. These steps are listed below, and the record's
variables are shown in Table 28.
REAL*4 DATIC, TIMIC, ELPIC, STPIC. ...
COHMON /RTSHIC/ DATIC, TIMIC, ELPIC, STPIC, ...
SUBROUTINE OPICON
INTEGER** IOST, NUDTSH, ...
PARAMETER (NUDTSH = 4. ...)
C read ICON T.S.H.
CALL RDFILE (UNITIC, NUDTSH., DATIC, IOST)
C
SUBROUTINE RDFILEdUNIT, NUORDS, BUFFER, IOST)
IMPLICIT NONE
INTEGER*4 IUNIT. NUORDS, BUFFER, IOST
DIMENSION BUFFER(NUORDS)
READdUNIT, IOSTAT=IOST) BUFFER
RETURN
END
TABLE 28. ICON TIME STEP HEADER RECORD VARIABLES
Var
No.
1
2
3
4
Var
Name
DATIC
TIMIC
ELPIC
STPIC
Data
Unit Type
Real*4
EST Real*4
s Real*4
Real*4
Description .
Current time step Julian date as YYDDD
Current time step time as HHMMSS
Elapsed time since scenario start
Step number
59
-------
2.4.2.2.2 Data Records- These records contain the chemical species concentrations iterated
over rows. The values are read into the common block ICF1LE by referencing the first
address of the common block for each species and specifying the number of the block's words
to be read. The code for these steps is listed below, and the data variables are shown in Table
29.
REAL*4 ICFILE
C
COMMON /ICFILE/ ICFILECNCOLS, NLEVS, NSPECS)
C
SUBROUTINE RDICON
INTEGER*4 IOST, .... NWDSIC, IROW, ISPC
PARAMETER (..., NWDSIC = NCOLS * NLEVS)
IROW - 1
201 CONTINUE
IF (IROW .GT. NROWIC) GO TO 301
C
C read ICON row
DO 211 ISPC = 1, NSPECS
CALL ROFILE (UNITIC, NUDSIC, ICFILE(1,1,ISPC), IOST)
211 CONTINUE
C
IROW - IROW + 1
GO TO 201
C'
301 CONTINUE
TABLE 29. ICON DATA VARIABLE
Var
No.
1
Var
Name
ICFILE,^
Unit
ppm
Data
Type
Real*4
Description
Initial chemical species concentrations in
column i and layer L for species k
2.5 THE NEW INITIAL CONDITIONS (NEWICON) FILE
The new initial conditions file (NEWICON) contains the predicted concentrations for the last time step
of a scenario, obtained from an execution of the Core Model. NEWICON is a copy of the CONC file data
for each of the 35 chemical species, each grid cell, and each of the model's 3 layers.
When the model starts the execution of the next scenario in a contiguous sequence of scenarios, the
NEWICON file is used in place of the ICON file to provide initial chemical species concentrations for the
first time step.
60
-------
It is possible to use the CONC file generated from the proceeding scenario for initial concentration data,
but since these files are large, data management becomes a serious problem for running a series of appli-
cations. The NEWICON file, containing only one time step of a CONC file, provides a solution by being
much smaller than a CONC file.
During data processing, array dimensions are set by parameter statements contained in the INCLUDE
files REGION.EXT and DIMENS.EXT as follows:
INTEGER*4 NROUS, NCOLS
PARAMETER ( ..., NROUS = 52, NCOLS = 64 )
INTEGERS NLEVS, NSPECS, .... NPOXSP
PARAMETER (NLEVS = 3, NSPECS = 35, .... NPOXSP = 3)
2.5.1 Opening the NEWICON File
The NEWICON file is opened at the completion of a model scenario. The main program, RUNMGR,
calls subroutine NEWICS, which writes the NEWICON file header and data. The structure of NEW-
ICON is identical to that of the ICON file. The code segments and the FORMAT statements that
demonstrate the steps to open the NEWICON file are listed below. Note that all NEWICON file
records have a fixed length equal to NLEVS x NCOLS.
INTEGER** .... UNITNI, ...
CHARACTER*12 .... FLNMNI, ...
SUBROUTINE NEUICS
1NTEGER*4 IUDLTH, .... JFILE5, ...
LOGICAL*4 RECFMT, RDONLY
PARAMETER ( IWOLTH = NLEVS * NCOLS,
& RECFMT = .FALSE., RDONLY = .FALSE.)
C
C open NEWICON file, unformatted, read/write access
UNITNI = JFILE5 (FLNMNI, RECFNT, RDONLY, IUDLTH)
FUNCTION JFILE5 (FNAME, RECFMT, RDONLY, RECLEN)
CHARACTER*^ FNAME, FORM, UNFORM, FORMAT
INTEGERS RECLEN, IDEV, IOST, JFILE5, JUNIT, ...
LOGICAL*4 RECFMT, RDONLY
DATA FORM / 'FORMATTED • /
DATA UNFORM / 'UNFORMATTED ' /
IDEV = JUKI TO
IF (RECFMT) THEN
FORMAT = FORM
ELSE
FORMAT = UNFORM
END IF
IF (RDONLY) THEN
61
-------
ELSE
OPEN (UNIT = IDEV,
& IOSTAT = IOST,
& FILE = FNAME,
& STATUS = 'UNKNOWN1,
& ACCESS = 'SEQUENTIAL1.
& FORM = FORMAT)
END IF
JFILE5 = IDEV
RETURN
END
2.5.2 NEWICON File Records
The structure of the NEWICON file conforms to the requirements for all ROM Core Model files,
and therefore contains a standard file header. In addition, the file consists of a data body organized by
time steps, each section of which is headed by a time step header record; note that the NEWICON file
contains only one time step. Descriptions of the records containing this information are given below.
2.5.2.1 NEWICON File Header Records-
. The first four records contain the NEWICON file header. The data that are written to the
NEWICON file header originate from (1) header information contained in the newly generated
CONC file, and (2) either the ICON or the NEWICON file used to start the scenario. In addi-
tion, (3) NEWICS calls subroutine DATTIM, which retrieves the current date and time from the the
computer operating system. This time stamp becomes the creation date and time for the
NEWICON file. The CONC file header data are contained in the common blocks in the
HEADCN.EXT include file:
c
C HEADCN.EXT
C
C CONC file header block
C
CHARACTER*80 TEXTCN
CHARACTER'S GRDNCN
CHARACTERS SPNMCN, LVNMCN
REAL** SULNCN, SWLTCN, NELNCN, NELTCN, DLONCN, DLATCN
INTEGER** CDATCN, CTIMCN, SOATCN, STHRCN, TSTPCN, FRSTCN,
& NCOLCN, NROWCN, NLEVCN, NSPCCN. ICNTCN,
& CDBMCN, CTBMCN, CDBTCN, CTBTCN,
& CDBCCN, CTBCCN, CDICCN, CTICCN
C
COMMON /CHARCN/ GRDNCN, SPNMCN(NSPECS), LVNMCN(NLEVS), TEXTCN(20)
COMMON /HEADCN/ CDATCN, CTIMCN, SDATCN, STHRCN, TSTPCN, FRSTCN,
& SULNCN, SULTCN, NELNCN, NELTCN, DLONCN, DLATCN,
& NCOLCN, NROWCN, NLEVCN, NSPCCN. ICNTCN,
& CDBMCN, CTBMCN, CDBTCN, CTBTCN,
& CDBCCN, CTBCCN, CDICCN, CTICCN
62
-------
2.5.2.1.1 Record 1- The first record contains character strings of alphanumeric data that
describe the file's contents. Data from the HEADCN.EXT common blocks are written to the
character buffer SEG1BF, which is then written (unformatted) to NEWICON by subroutine
WRCHAR. The scenario start date and time for the NEWICON file are obtained from the
current (last) CONC file date and time. The variable IELPIC, which represents time elapsed
on the ICON (or previous NEWICON) file from the scenario start time, is obtained from the
data stored in common in the HEADICEXT include file (refer to Section 2.4.2.1). This variable
is used to determine the time to the first step on the NEWICON file, FRSTIC. The ICON
time step size, TSTPIC, is simply copied from the HEADICEXT common block to become the
NEWICON time step size. The code segments and FORMAT statements for these steps are
shown below, and the variables of record 1 are shown in Table 30.
INTEGER** ..., UNITNI, ...
SUBROUTINE NEWICS
C CONC header buffers
CHARACTER*(8 * 21 + 4 * 5) SEG1BF
INTEGER** .... IOST, ...
C get creation date for new 1C file
CALL DATTIM (CDATIC, CTIMIC)
C
C get initial time values from last CONC time step
C (Note: Model cannot be started from an ICON file at half-hour)
SOAT1C = IDATCN
STHRIC = ITIMCN / 10000
C
C get elapsed time value from ICON
FRSTIC = IELPIC
C
C write 1st segment
WRITE(SEG1BF, 1001, IOSTAT - IOST)
& CDATIC, CTIMIC, SOAT1C, STHRIC, TSTPIC, FRSTIC,
& GRDNCN,
& SULNCN, SWLTCN, NELNCN, NELTCN,
& DLONCN, DLATCN,
& NCOLCN, NROWCN, NLEVCN, NSPCCN,
& COBMCN, CTBHCN,
& COBTCN, CTBTCN,
& COBCCN, CTBCCN,
& COICCN, CTICCN,
& ICNTCN
1001 FORMATC6I8.8, A8, 4F8.3, 2F8.5, 414.4. 818.8. 14.4)
C
CALL URCHAR (UNITNI. SEG1BF, IOST)
SUBROUTINE URCHAR (IUNIT, CHBUF, IOST)
IMPLICIT NONE
INTEGER** IUNIT, IOST
CHARACTERS*) CHBUFF
WRITE(IUNIT, IOSTAT = IOST) CHBUFF
RETURN
END
63
-------
TABLE 30. NEWICON RECORD 1 VARIABLES
Var
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
Var
Name
CDATIC
CTIMIC
SDATIC
STHRIC
TSTPIC
FRSTIC
GRDNCN
SWLNCN
SWLTCN
NELNCN
NELTCN
DLONCN
DLATCN
NCOLCN
NROWCN
NLEVCN
NSPCCN
CDBMCN
CTBMCN
CDBTCN
CTBTCN
CDBCCN-
CTBCCN
CDICCN
CTICCN
ICNTCN
Unit
EST
EST
s
s
°W
°N
°W
°N
°W
°N
EST
EST
EST
EST
Data
Type
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Char*8
Real*4
Real*4
Real*4
Real*4
RealM
Real*4
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Description
File creation date as MMDDYY
File creation time as HHMMSS
Julian start date of scenario as YYDDD
Start hour of scenario (00 to 23)
Time step size for simulation
Time to first step
Grid definition name
Longitude of southwest corner of grid
Latitude of southwest corner of grid
Longitude of northeast corner of grid
Latitude of northeast corner of grid
Grid cell longitudinal increment
Grid cell latitudinal increment
Number of columns in grid
Number of rows in grid
Number of levels in the simulation
Number of species in the NEWICON file
Creation date of the B-matrix file (BMAT)
from which the CONC file was generated,
as MMDDYY
Creation time of the B-matrix file (BMAT)
from which the CONC file was generated,
as HHMMSS
Creation date of the backtrack file (BTRK)
from which the CONC file was generated,
as MMDDYY
Creation time of the backtrack file (BTRK)
from which the CONC file was generated,
as HHMMSS
Creation date of the boundary conditons file
(BCON) from which the CONC file was
generated, as MMDDYY
Creation time of the boundary conditons file
(BCON) from which the CONC file was
generated, as HHMMSS
Creation date of the initial conditons file
(ICON) from which the CONC file was
generated, as MMDDYY
Creation time of the initial conditons file
(ICON) from which the CONC file was
generated, as HHMMSS
Number of text records
64
-------
2.5.2.1.2 Record 2- This record contains the list of species names for which NEWICON
contains concentration data. The data are first written into the buffer SPNMBF, which is
then written (unformatted) to NEWICON. These steps are listed below, and the variable of
record 2 is shown in Table 31.
SUBROUTINE NEWICS
CHARACTER*^ * NSPECS) SPNMBF
INTEGER*^ .... ISPC, ...
DO 101 ISPC = 1, NSPCIN
SPNMCN(ISPC) = SPNHINCISPC)
101 CONTINUE
C write the species names
WRITECSPNMBF, 1003, IOSTAT = IOST)
& (SPNMCN(ISPC), ISPC = 1, NSPCCN)
1003 FORMAT( 35
-------
2.5.2.1.4 Records 4 - (4 + ICNTCN)- These records contain descriptive text that was copied
to the CONC file header from the model execution runstream control parameters and now
become the NEWICON descriptive text. One 80-character string is written to each record by
the following statements. The variable of the records is shown in Table 33.
SUBROUTINE NEUICS
INTEGER*4 ..., ITXT, ...
C write file text group
DO 101 ITXT = 1. ICNTCN
CALL WRCHAR (UNITNI, TEXTCN(ITXT), IOST)
101 CONTINUE
TABLE 33. NEWICON RECORDS 4 - (4+ICNTCN) VARIABLE
Var
No.
1
Var
Name
TEXTCNn
Data
Unit Type
Char*80
Description
Text string of n lines
2.5.2.2 NEWICON File Body Records--
2.5.2.2.1 Time Step Header Record- There is one time step header record written for the
single time step increment on the NEWICON file. The time step header data is contained in
the common block RTSHCN, whose first variable is used in subroutine WRFILE'S argument
list; note that NCOLS x NLEVS is specified as the number of words to be written. The code
segments for these steps are listed below, and the record's variables are shown in Table 34.
REAL** DATIC, TIMIC, ELPIC, STPIC, ...
COMMON /RTSHIC/ DATIC, TIMIC, ELPIC, STPIC, ...
SUBROUTINE NEUICS
INTEGER** IUDLTH, .... IOST, ...
PARAMETER ( IUDLTH = NLEVS * NCOLS, ... )
C write NI T.S.H.
CALL URFILE (UNITNI, IUDLTH, DATIC, IOST)
SUBROUTINE URFILE (IUNIT, NUORDS, BUFFER. IOST)
IMPLICIT NONE
INTEGER** IUNIT, NUORDS, BUFFER, IOST
DIMENSION BUFFER(NUORDS)
URITEdUNIT, IOSTAT = IOST) BUFFER
RETURN
END
66
-------
TABLE 34. NEWICON TIME STEP HEADER RECORD VARIABLES
Var
No.
1
2
3
4
2.5.2.2.2
Var
Name
DATIC
TIMIC
ELPIC
STPIC
Data
Unit Type
Real*4
EST Real*4
s Real*4
Real»4
Data Records- Each data record
Description
Current time step Julian date as YYDDD
Current time step time as HHMMSS
Elapsed time since scenario start
Step number
contains the concentration data for one chemical
species in one row of the domain grid for the final time step of a scenario execution. The
concentration data are first copied into the ICFILE common block from the initial condi-
tions buffer (common block BGICCN), which is written by subroutine ICPRCS at the end of
each scenario step. At the end of the scenario execution, the BGICCN common block
contains the last time step concentrations computed in the model. These data are written by
referencing the memory address corresponding to the column 1 and level 1 indices of the
array in the common block ICFILE for each species, and by passing the number of words to
be written to subroutine WRFILE. The code segments for these steps are listed below and the
variable for this record is shown in Table 35.
REAL*4 ICFILE
COMMON /ICFILE/ ICFILE(NCOLS, NLEVS, NSPECS)
REAL** BGICCN
INTEGER** TOG1, TOG2
COMMON /BGICCN/ TOG1, TOG2, BGICCNCNCOLS, NLEVS, NSPECS, NROUS, 2)
SUBROUTINE NEUICS
INTEGER** IUDLTH, ISPC, IOST, ITXT, JFILE5,
& INDEX, ICOL, IROU, ILEV
INTEGER** .... IUDLN2, .... ISPC, IOST
C
PARAMETER ( IUDLTH = NLEVS * NCOLS, ... )
C copy body row*
C
IROU = 1
'201 CONTINUE
IF (IROW .GT. NROWIC) GO TO 301
C
C reorder species
DO 203 ISPC = 1, NSPCIN
INDEX = NXSPIC(ISPC)
DO 203 ILEV * 1. NLEVIN
DO 203 I COL = 1, NCOLIN
ICFILEdCOL, ILEV, ISPC) = BGICCNOCOL, ILEV, INDEX, IROU, TOG2)
203 CONTINUE
C
C write NEUICON file
DO 205 ISPC = 1, NSPECS
CALL URFILECUNITNI, IUDLTH, ICFILEd,1,ISPC),IOST)
67
-------
205 CONTINUE
C
I ROW = I ROW + 1
GO TO 201
C
C copy body ............................ end
C
301 CONTINUE
TABLE 35. NEWICON DATA VARIABLE
Var Var Data
No. Name Unit Type Description
ppm Real*4 Initial chemical species concentrations in
column i and layer L for species k
2.6 THE PROGRESS (PROG) FILE
The ROM Core Model requires a long execution time. We have found it convenient to be able to ascer-
tain the last completed time step of a scenario during the progress of a model execution. Although this
can be determined by interactively examining the currently open run log file, the progress file (PROG) is
much smaller, consisting of one line, and therefore is more easily accessed.
2.6.1 Opening the PROG File
At the completion of each scenario time step, the main program, RUNMGR, opens the PROG file to
write the one line time step progress data. After the data are written, RUNMGR closes the file so that
there will be no conflict if a user decides to read it. In order that the computer operating system does
not create new versions of the file every time it is reopened, it is necessary for the file to be opened
with status 'OLD'. This, then, requires that the file exist prior to the start of a model run. The code
segments below illustrate the steps to open the file.
INTEGER** .... UNITPR
COMMON / LUNITS / .... UNITPR
CHARACTER*^ ..., FLNMPR
COMMON / FLNAMS / .... FLNMPR
PROGRAM RUNMGR
C get unit numbers for ... PROGRESS file
UNITPR = JUNITO
C update PROGRESS file
OPEN (UNIT = UNITPR,
& STATUS = 'OLD1,
& ACCESS = 'SEQUENTIAL1,
& FORM = 'FORMATTED',
& FILE = FLNMPR)
68
-------
2.6.2 The PROG File Record
The one line PROG file record consists of text that echoes the scenario date, scenario time, and the
scenario time step number that has just been completed by the model's execution. The code segments
that write this line are listed below, and the variables in the PROG record are shown in Table 36.
INTEGERS HDDATE, MDTIME, .... MDSTEP, ...
COMMON /TSTEPS/ HDDATE, MDTIME, ..., MDSTEP, ...
PROGRAM RUNMGR
WRITE(UNITPR, 1007) MDDATE, MDTIME, MDSTEP
1007 FORMAH3X, 16, 4X, 16.6, 4X, 14)
WRITECLUNOUT, 1009) MDDATE, MDTIME, MDSTEP
1009 FORMAT(/ 5X, 'Progress Completed: DATE/TIME = ', 16, ' / ', 16.6,
& 4X, 'STEP = ', 14 )
CLOSE (UNIT = UNITPR)
C
TABLE 36. PROG RECORD VARIABLES
Var
No.
1
2
3
Var
Name Unit
MDDATE
MDTIME EST
MDSTEP
Data
Type
Integer*4
Integer*4
Integer*4
Description
Julian scenario date as YYDDD
Scenario time as HHMMSS
Scenario step number
2.7 THE STATE VECTOR (RESTRT) FILE
The purpose of the state vector file (RESTRT) is to allow the model run to be restarted in case a previous
execution was stopped before the end of the scenario. A model run may have been stopped by either a
user or the computer operating system (e.g., a queue time-out). The model can then be restarted, begin-
ning at any time step beyond the scenario start but before the scenario step where it was stopped. This
restart feature saves time and computer costs, although in our experience we have needed to use it
infrequently.
The Core Model code consists of 66 subprograms in addition to the main program, RUNMGR. Of these,
14 are variable-state subroutines, which we ca\\processes to distinguish them from procedures. Procedures
execute their code from top to bottom whenever they are invoked. Processes interact with their calling
programs in a critical time sequence, hence their implementation as variable-state subroutines. The
processes maintain variable text pointers in their code that are set when the process is suspended and
control is returned to the calling program. At the next invocation of the process, it resumes execution of
the code from the point at which it was previously suspended.
69
-------
RESTRT saves all necessary data to reset the variable-state subroutines to the exact state they were in
when the previous run was stopped. RESTRT also contains a copy of the HEADIN.EXT common blocks,
which consists of the model input data. In addition to these data, RESTRT saves the following:
the chemistry control parameters set by subroutine INIRUN;
the BMAT, BCON and ICON files' species look-up tables;
species flags indicating which are the primary oxidant species;
species flags for dealing with special species;
diffusivities conversion factors; and
model and file process clock data.
The first card in the user-supplied input run stream control dataset contains a text field, the requested
number of steps to execute, and the starting date and time. Under normal operations, the text field is set
to 'START_RUN < (three embedded blank spaces). To restart a model run, you must set the text field to
•RESTARTJJUN • (one embedded blank space), and also supply the starting date and time from which to
continue the model execution. For example, under normal conditions for model execution, the first card
will appear as follows (refer to Part 4 of the ROM User's Guide):
•START_RUN ' 144 85188 120000
Suppose that the model execution is stopped during step 100, leaving 45 steps to process until the end of
the scenario at 8519112:00. To restart the run, you must change the first card to the following:
'RESTART_RUN ' 45 85190 133000
We recommend that you carefully count the number of time steps remaining until the end of the scenario
prior to restarting the run.
In the event of a restarted run, RUNMGR calls subroutine RDSTAV to open and read the state vector file
and retrieve the data necessary to reset the clocks and text pointers for all the processes (variable state
subroutines). In addition, RUNMGR opens and positions the RESTRT, BCON, BTRK, and BMAT files
to the time step records corresponding to the requested scenario restart time. Finally, RUNMGR calls
subroutine RDCONC to get the concentration data from the CONC file for the time step prior to the
restart time step. These data are then loaded into the initial conditions buffer that BIGGAM reads to
calculate the advection component of the next time step's concentration field (refer to Section 2.4). With
the model thus reinitialized, execution proceeds normally from the restart time step.
2.7.1 Processing that Takes Place for Normal Model Execution (haw the RESTRT file gets written)
2.7.1.1 Opening the RESTRT FUe-
At the completion of the first model scenario time step, the main program (RUNMGR) calls sub-
routine WRSTAV to open the RESTRT file and record the state of the model execution on the file.
70
-------
WRSTAV opens the RESTRT file with default status "formatted," allowing you to edit and change
the file if you wish. The following code segments show the operations that open the file.
FLNMSV is set in the block data module, BLKMOD. JUNIT is a function subprogram that
returns the next FORTRAN I/O unit number not being used by the model execution.
INTEGER** .... UNITSV, ...
CHARACTER*12 .... FLNMSV. ...
SUBROUTINE WRSTAV
INTEGER** JUNIT, .... IOST, ...
C open STATE VECTOR file, formatted, read/write access
C
UNITSV = JUNITO
OPEN (UNITSV,
& FILE = FLNMSV,
& ACCESS = 'SEQUENTIAL1,
& STATUS = 'UNKNOWN1,
& IOSTAT = IOST)
C
2.7.1.2 RESTRT File Records
The structure of the RESTRT file conforms to the requirements for all ROM Core Model files,
and therefore contains a standard file header. In addition, the file consists of a data body orga-
nized by time steps, each section of which is headed by a time step header record. Descriptions of
the records containing this information are given below.
2.7.1.2.1 RESTRT file header records- The first 16+ records constitute the RESTRT file
header.
Just as for the CONC file header, the data that are written to the RESTRT file header
originate from:
• the execution run stream's control cards;
• header information from the BMAT, BTRK, BCON, and ICON files;
• variables set by assignment statements in subroutine INIRUN; and
arguments returned from various subroutines called by INIRUN.
These data are loaded into the common blocks in the include file HEADIN.EXT at the end of
the first scenario time step of the model execution:
c
C HEADIN.EXT
C
C input header information
C
C used to check file headers on BMATRIX, ICON, and BCON files
C and to create file header on RESTRT file
C
71
-------
CHARACTER*80 TEXTIN
CHARACTER*8 GRDNIN
CHARACTERS SPNMIN, LVNMIN
REALS SULNIN, SUIT IN, NELNIN, NELTIN, DLONIN, DLATIN
INTEGERS COATIN, CTININ, SDATIN, STHRJN, TSTPIN, FRSTIN,
& NCOLIN, NROUIN, NLEVIN, NSPCIN, ICNTIN
COMMON /CHARIN/ GRDNIN, SPNHIN(NSPECS), LVNMIN(NLEVS), TEXTIN(20)
COMMON /HEAOIN/ CDATIN, CTIMIN, SOATIN, STHRIN, TSTPIN, FRSTIN,
& SULNIN, SULTIN, NELNIN, NELTIN, DLONIN, DLATIN,
& NCOLIN, NROUIN, NLEVIN, NSPCIN, ICNTIN
The data in the HEADIN.EXT common blocks are used in the creation of the RESTRT file
header. The INCLUDE file HEADSV.EXT contains the common blocks that are loaded with
the RESTRT file header variables.
c
C HDSTAV.EXT
C
C formatted STATE VECTOR file header block
C
CHARACTER*80 TEXTSV
CHARACTER*8 GRDNSV
CHARACTERS SPNMSV, LVNMSV
REALS SULNSV, SULTSV, NELNSV, NELTSV, DLONSV, DLATSV
INTEGERS CDATSV, CTIMSV, SDATSV, STHRSV, TSTPSV, FRSTSV,
& NCOLSV, NROUSV, NLEVSV, NSPCSV, ICNTSV
C
COMMON /CHARSV/ GRDNSV, SPNMSV(NSPECS), LVNMSV(NLEVS), TEXTSV(ZO)
COMMON /HEADSV/ CDATSV, CTIMSV, SDATSV, STHRSV, TSTPSV, FRSTSV,
& SULNSV. SULTSV, NELNSV. NELTSV, DLONSV, DLATSV,
& NCOLSV, NROUSV, NLEVSV, NSPCSV, ICNTSV
2.7.1.2.1.1 Record 1-- The first two records contain character strings of alphanumeric
data that describe the file's contents. The data that have been stored in the common
blocks in INCLUDE file HEADIN.EXT are first loaded into the common blocks in
INCLUDE file HEADSV.EXT. The data are then written (formatted) to the RESTRT file.
The data are written to two records (instead of one record) so that users can easily edit
the file. The code segments that perform these operations are listed below, along with
the WRITE and FORMAT statements. The variables of record 1 are shown in Table 37.
INTEGER** .... UNITSV, ...
SUBROUTINE URSTAV
INTEGERS .... IOST, ...
C prepare header buffer
C
CDATSV = CDATIN
CTIMSV = CTIMIN
SDATSV = SDATIN
STHRSV = STHRIN
TSTPSV = TSTPIN
C
C first data is one time step beyond 1C time step
C
FRSTSV = FRSTIN + TSTPIN
GRDNSV = GRDNIN
72
-------
SULNSV
SULTSV
NELTSV
NELNSV
DLONSV
DLATSV
NCOLSV
NROUSV
NLEVSV
NSPCSV
ICNTSV
SULNIN
SUITIN
NELTIN
NELNIN
DIONIN
DLATIN
NCOLIN
NROUIN
NLEVIN
NSPCIN
ICNTIN
C write STATE VECTOR header segment 1
C
WRITECUNITSV, FMT = 1001, IOSTAT = IOST)
& CDATSV, CTIMSV, SDATSV, STHRSV, TSTPSV,
& FRSTSV, GRDNSV, SULNSV, SULTSV, NELNSV
1001 FORMAT(1X, 2(16, 1X), IS, 1X, 12, 1X, 2(18, IX),
& AS, 1X, 3(F8.3, IX))
TABLE 37. RESTRT RECORD 1 VARIABLES
Var
No.
1
2
3
4
5
6
7
8
9
10
Var
Name
CDATSV
CTIMSV
SDATSV
STHRSV
TSTPSV
FRSTSV
GRDNSV
SWLNSV
SWLTSV
NELNSV
Unit
EST
EST
s
s
ew
°N
°W
Data
Type
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Char*8
Real*4
Real*4
Real*4
Description
File creation date as MMDDYY
File creation time as HHMMSS
Julian start date of scenario as YYDDD
Start hour of scenario (00 to 23)
Time step size for simulation
Time to first step
Grid definition name
Longitude of southwest corner of grid
Latitude of southwest corner of grid
Longitude of northeast corner of grid
2.7.1.2.1.2 Record 2- This record contains the remainder of the first segment
data. The relevant code segments are listed below, along with the WRITE and FOR-
MAT statements. The variables of record 2 are shown in Table 38.
SUBROUTINE WRSTAV
INTEGER** .... IOST, ...
C write STATE VECTOR header segment 2
C
WRITE(UNITSV, FMT = 1003, IOSTAT = IOST)
& NELTSV, DLONSV, DLATSV, NCOLSV,
& NROUSV, NLEVSV, NSPCSV, ICNTSV
1003 FORMATdX, F8.3, 2(1X, F8.S), 5(1X, 14))
73
-------
TABLE 38. RESTRT RECORD 2 VARIABLES
Var
No.
1
2
3
4
5
6
7
8
Var
Name
NELTSV
DLONSV
DLATSV
NCOLSV
NROWSV
NLEVSV
NSPCSV
ICNTSV
2.7.1.2.1.3 Records 3
Unit
°N
°W
°N
and 4--
Data
Type
Real*4
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
These records
Description
Latitude of northeast corner of grid
Grid cell longitudinal increment
Grid cell latitudinal increment
Number of columns in grid
Number of rows in grid
Number of levels in the simulation
Number of species in the RESTRT file
Number of text records
contain the list of species names for which
the Core Model computes concentration outputs. The data that have been stored in the
HEADIN.EXT INCLUDE file common blocks are first loaded into the HEADSV.EXT
common blocks. They are then written (formatted) to the RESTRT file. The data are
written to two records (instead of one record) so that users can easily edit the file. The
code segments that perform these operations are listed below, along with the WRITE
and FORMAT statements. Note that if the number of species exceeds 30, the FORMAT
statement will cause additional records to be written when the last WRITE statement is
executed. The variable of records 3 and 4 is shown in Table 39.
SUBROUTINE WRSTAV
INTEGER** .... ISPC, IOST, ...
DO 101 ISPC = 1, NSPCIN
SPNMSV(ISPC) = SPNMIN(ISPC)
101 CONTINUE
C write the species names records
c
WRITE(UNITSV. FMT = 1005, IOSTAT = IOST)
& (SPNMSV(ISPC), ISPC = 1, 15)
1005 FORMATC1X, 15
-------
2.7.1.2.1.4 Records- This record contains the list of Core Model layer names. The data
that are stored in the HEADIN.EXT common blocks are first loaded into the HEADSV.EXT
common blocks. They are then written (formatted) to the RESTRT file. The code
segments that perform these operations are listed below, along with the WRITE and
FORMAT statements. The variable of record 5 is shown in Table 40.
SUBROUTINE WRSTAV
INTEGER** ..., IOST, ..., ILEV, ...
DO 103 ILEV = 1, NLEVSV
LVNMSV(ILEV) = LVNHIN(ILEV)
103 CONTINUE
C write the level names record
C
RECNSV - RECNSV + 1
WRITECUNITSV, FMT = 1005, IOSTAT = IOST)
& (LVNMSV(IIEV), ILEV = 1, NLEVSV)
TABLE 40. RESTRT RECORD 5 VARIABLE
Var
No.
1
Var
Name Unit
LVNMSVL
Data
Type
Char*4
Description
Name of layer L
2.7.1.2.1.5 Records 6 - (6 + ICNTSV)- These records contain the descriptive text that
was copied to the CONC file header from the model execution run stream control
parameters. One 80-character string is written to each record. The data have been stored
in the HEADIN.EXT common blocks, and are first loaded into the HEADSV.EXT common
blocks. They are then written (formatted) to the RESTRT file using the code segments
listed below. The variable of the records is shown in Table 41.
SUBROUTINE WRSTAV
INTEGER** .... IOST, ITXT, ...
DO 105 ITXT = 1, ICNTSV
TEXTSV(ITXT) « TEXTIN(ITXT)
105 CONTINUE
C write header text records
C
DO 109 ITXT = 1, ICNTSV
WRITECUNITSV. FMT = 1007, IOSTAT = IOST) TEXTSV(ITXT)
1007 FORNATdX, ABO)
109 CONTINUE
75
-------
TABLE 41. RESTRT RECORDS 6 - (6+ICNTSV) VARIABLE
Var
No.
1
Var
Name
TEXTSVn
Data
Unit Type
Char*80
Description
Text string of n lines
2.7.1.2.1.6 HEADIN record 1-- The next two records contain copies of the HEADIN.EXT
common blocks. The data are written to two records (instead of one record) so that users
can easily edit the file. The WRITE and FORMAT statements for this record are listed
below, and its variables are shown in Table 42.
SUBROUTINE WRSTAV
INTEGER** .... IOST, ...
C write HEADIN header record segment 1
C
WRITE(UNITSV, FMT = 1001, 10STAT = IOST)
& COAT IN, CT1M1N, SDATIN, STHRIN, TSTPIN, FRSTIN,
& GRDNIN, SULNIN, SWLTIN, NELNIN
TABLE 42. RESTRT HEADIN RECORD 1 VARIABLES
Var
No.
1
2
3
4
5
6
7
8
9
10
Var
Name
CDATTJM
CTIMIN
SDATIN
STHRIN
TSTPIN
FRSTIN
GRDNIN
SWLNIN
SWLTIN
NELNIN
2.7.1.2.1.7
Unit
EST
EST
s
s
"W
°N
°W
HEADIN
Data
Type
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Char'8
Real*4
Real*4
Real*4
Description
File creation date as MMDDYY
File creation time as HHMMSS
Julian stan date of scenario as YYDDD
Start hour of scenario (00 to 23)
Time step size for simulation
Time to first step
Grid definition name
Longitude of southwest corner of grid
Latitude of southwest corner of grid
Longitude of northeast corner of grid
record 2~ This record is the continuation of the HEA-
DIN.EXT common block's data. The WRITE and FORMAT statements for this record
are listed below, and its variables are shown in Table 43.
SUBROUTINE WRSTAV
INTEGER** .... IOST, ...
C unite HEADIN header record segment 2
C
URITE(UNITSV, FMT = 1003, IOSTAT = IOST)
& NELTIN, OLONIN, DLATIN, NCOLIN, NROUIN,
& NLEVIN, NSPCIN, ICNT\N
76
-------
TABLE 43. RESTRT HEAD1N RECORD 2 VARIABLES
Var
No.
1
2.
3
4
5
6
7
8
Var
Name
NELTIN
DLONIN
DLATIN
NCOLIN
NROWIN
NLEVIN
NSPCIN
ICNTIN
2.7.1.2.1.8
Unit
°N
°W
°N
Chemistry
Data
Type
Real*4
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Description
Latitude of northeast corner of grid
Grid cell longitudinal increment
Grid cell latitudinal increment
Number of columns in grid
Number of rows in grid
Number of levels in the simulation
Number of species in the RESTRT file
Number of text records
control records- The next two records contain the
chemistry control data used by subroutine LILGAM. These data are loaded into the
CHEMIN common block by subroutine INIRUN at the start of a model run. They are
saved in the RESTRT file to preserve the same data values in the event that a restart run
is required. The data are written to two records (instead of one record) so that users can
easily edit the file. The variables of these records are shown in Table 44 and Table 45, and
the code segments that write them to the RESTRT file are listed below.
INTEGER** NCOUT, ISPEC
REAL*4 ATS, GTS, UFRAX, BFRAX, FACTOR, DIVP, DIVQ,
& ULIM, BLIM, FNOLIH
C
COMMON /CHEMIN/ ATS, GTS, UFRAX, BFRAX, FACTOR, DIVP, DIVQ,
& NCOUT, ISPEC(NSPECS), ULIM, BLIM, FNOLIM
SUBROUTINE URSTAV
INTEGER** .... IOST, ...
C write CNEMIN header record
C
WRITEOJNITSV, FMT = 1013, IOSTAT = IOST)
& ATS, GTS, UFRAX, BFRAX, FACTOR,
& DIVP, DIVQ, NCOUT
1013 FORMATdX, 2(F8.2, 1X), 5(F8.5, IX), 15)
URITE(UNITSV, FMT = 1015, IOSTAT = IOST)
£ (ISPEC(ISPC), ISPC = 1, NCOUT), ULIM, BLIM, FNOLIM
1015 FORMATdX, (K.3, IX), 3(E10.3. 1X) )
77
-------
TABLE 44. RESTRT CHEMISTRY CONTROL VARIABLES 1
Var
No.
1
2
3
4
5
6
7
8
Var
Name
ATS
GTS
UFRAX
BFRAX
FACTOR
DIVP
DIVQ
NCOUT
Data
Unit Type
s Real*4
s Real*4
Real*4
Real*4
Real*4
Real*4
Real*4
Integer*4
Description
Advection time step
G-tilde time step
Upper FRAX limit
Lower FRAX limit
Threshold factor to include species in
chemistry time step determination
Chemistry decay term predictor -
corrector combination factor
Chemistry source term predictor -
corrector combination factor
Number of primary oxidant species
Var
No.
TABLE 45. RESTRT CHEMISTRY CONTROL VARIABLES 2
Var
Name
Unit
Data
Type
Description
2
3
4
ISPEQ
ULIM
BLIM
FNOLIM
Integer*4
Real*4
Real*4
Real*4
Index for species k used in chemistry time'
step determination
Chemistry time step upper limit
Chemistry time step lower limit
Chemistry solution accuracy control
parameter
2.7.1.2.1.9 BMAT species index records- These records contain the expansion
list (look-up table) of species names for which the B-matrix file has values. (Refer to
P23G, Part 2 of the ROM User's Guide, for an explanation of the expansion list.) These
values are mapped to the full Core Model list of species names (see Section 2.2.2.1.3).
The WRITE and FORMAT statements for these records are listed below, note that if the
number of species exceeds 30, the FORMAT statement will cause additional records to
be written when the last WRITE statement is executed. The records' variable is shown in
Table 46.
SUBROUTINE WRSTAV
INTEGER** .... ISPC. IOST. ...
C write species ordering header record
C
78
-------
URITE(UNITSV, FMT = 1019. IOSTAT = IOST)
& (NXSPBM(ISPC), ISPC = 1, 15)
1019 FORMAT (IX, 15(13, IX))
WRITE(UNITSV, FMT = 1019, IOSTAT = IOST)
& (NXSPBM(ISPC), ISPC = 16, NSPECS)
TABLE 46. BMAT SPECIES INDEX VARIABLE
Var
No.
1
Var
Name Unit
NXSPBM*
Data
"type
Integer*4
Description
BMAT file index for species k \
\ A list of chemical species names can be found in Table 1.
2.7.1.2.1.10 BCON species index records- These records contain the expansion
list (look-up table) of species names for which the BCON file has values. These values
are mapped to the full Core Model list of species (see Section 2.2.2.1.3). For ROM2.1,
the BCON file contains the same species list as the Core Model, therefore the mapping
table is one-to-one. The WRITE and FORMAT statements for these records are listed
below; note that if the number of species exceeds 30, the FORMAT statement will cause
additional records to be written when the last WRITE statement is executed. The
records' variable is shown in Table 47.
SUBROUTINE WRSTAV
INTEGER** .... ISPC, IOST, ...
WRITECUNITSV, FMT = 1019, IOSTAT = IOST)
& (NXSPBC(ISPC), ISPC =1, 15)
WRITE(UNITSV, FMT = 1019, IOSTAT = IOST)
& (NXSPBC(ISPC), ISPC = 16, NSPECS)
1019 FORMAT (1X, 15(13, IX))
TABLE 47. BCON SPECIES INDEX VARIABLE
Var
No.
1
Var
Name Unit
NXSPBQ
Data
Type
Integer*4
Description
BCON file index for species k t
f A list of chemical species names can be found in Table 1.
2.7.1.2.1.11 ICON species index records- These records contain the expansion
list (look-up table) of species names for which the ICON file has values. These values are
mapped to the full Core Model list of species. For ROM2.1, the ICON file contains the
same species list as the Core Model, therefore the mapping table is one-to-one. The
WRITE and FORMAT statements for these records are listed below, note that if the
79
-------
number of species exceeds 30, the FORMAT statement will cause additional records to
be written when the last WRITE statement is executed. The records' variable is shown in
Table 48.
SUBROUTINE WRSTAV
INTEGER** .... ISPC, IOST, ...
URITECUNITSV, FMT = 1019, IOSTAT = IOST)
& (NXSPIC(ISPC), ISPC = 1. 15)
URITECUNITSV, FMT = 1019, IOSTAT = IOST)
& (NXSPIC(ISPC), ISPC = 16, NSPECS)
1019 FORMAT <1X, 15(13, IX))
TABLE 48. ICON SPECIES INDEX VARIABLE
Var
No.
1
Var
Name Unit
NXSPIQ
Data
Type
Integer*4
Description
ICON file index for species k t
f A list of chemical species names can be found in Table 1.
2.7.1.2.1.12 Primary oxidant species flag records-- These records contain
LOGICAL variables corresponding to the list of model species indices. The variable has
the LOGICAL value of .TRUE, if the species index corresponds to a primary oxidant
species, i.e., NO, NO2, or ©3; otherwise the variable has the logical value of .FALSE..
The WRITE and FORMAT statements for these records are listed below; note that if the
number of species exceeds 30, the FORMAT statement will cause additional records to
be written when the last WRITE statement is executed. The records' variable is shown in
Table 49.
SUBROUTINE WRSTAV
INTEGER*4 .... ISPC, IOST, ...
C write species.control for LILGAM
C
URITECUNITSV, FMT = 1025, IOSTAT = IOST)
& (INBIG3USPC). ISPC = 1, 15)
1025 FORMATC1X, 15CL4, IX))
URITECUNITSV, FMT = 1025, IOSTAT = IOST)
& CINBIG3C1SPC). ISPC = 16, NSPECS)
TABLE 49. PRIMARY OXIDANT SPECIES FLAG RECORDS VARIABLE
Var Var Data
No. Name Unit Type Description
1 INBIG3* Logical *4 Primary oxidant flag for species k
80
-------
2.7.1.2.1.13 Special species record- This record contains the indices for NO,
NO2,03, PAR, TRAC, and NONR from the model species list. These variables are used
in the Core Model for special processing of these species. The WRITE and FORMAT
statements for this record are listed below, and the record's variable is shown in Table 50.
SUBROUTINE URSTAV
INTEGER** .... IOST, ...
WRITE(UNITSV, FMT = 1027, IOSTAT = IOST)
& NOHIT, N02HIT, 03HIT, PARHIT, TRCHIT, NONNIT
1027 FORMATOX, 6(13, 1X))
TABLE 50. SPECIAL SPECIES RECORD VARIABLES
Var
No.
1
2
3
4
5
6
Var
Name
NOHIT
NO2HIT
03HIT
PARHIT
TRCHIT
NONHIT
2.7.1.2.1.14
Data
Unit Type
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Description
Species list index for NO
Species list index for NO2
Species list index for 03
Species list index for PAR
Species list index for TRAC
Species list index for NONR
Diffusivities conversion factor record-- These variables are used in
BIGGAM for the treatment of horizontal diffusion in the advection scheme used in the
model. They must be saved to the RESTRT file because, in the event of a restart, the part
of the BIGGAM code where they are set is bypassed. The WRITE and FORMAT
statements for this record are listed below, and the record's variable is shown in Table 51.
SUBROUTINE URSTAV
INTEGER** ..., IOST, ...
C Write diffusivities conversion factor for BIGGAM
c
WRITE(UNITSV, FMT = 1029, IOSTAT = IOST) RDLNT2, RDLTT2
1029 FORMAT(1X. E13.6. 2X, E13.6)
81
-------
TABLE 51. DIFFUSIVmES CONVERSION FACTOR RECORD VARIABLES
Var
No.
Var
Name
Data
Unit Type
Description
RDLNT2 s-rad-2 Real*4 Longitudinal horizontal diffusivities
conversion factor
RDLTT2 s-rad-2 Real*4 Latitudinal horizontal diffusivities con-
version factor
2.7.1.2.2 RESTRT file body records- At the completion of each model scenario time step
the main program (RUNMGR) calls subroutine WRSTAV to record the state of the model
execution on the RESTRT file. RESTRT thus contains a state vector for each completed
time step.
2.7.1.2.2.1 Time step header record- There is one time step header record for each time
step increment on the RESTRT file. The code references the time step data in the time
step header common block, TSHDSV, and writes it (formatted) to the RESTRT file.
The common block is loaded by the main program, RUNMGR, prior to calling subroutine
WRSTAV. These steps are listed below, and the record's variables are shown in Table 52.
INTEGER** IDATSV, ITIMSV, IELPSV, ISTPSV
c
COMMON /TSHDSV/ IDATSV, ITIMSV, IELPSV, ISTPSV
SUBROUTINE URSTAV
INTEGER** .... IOST, ...
WRITECUNITSV, FMT = 1031, IOSTAT = IOST)
& IDATSV, ITIMSV, IELPSV, ISTPSV
1031 FORMAT(1X, 15, 1X, 16, 1X, 18, 1X, 14)
C
TABLE 52. RESTRT TIME STEP HEADER RECORD VARIABLES
Var
No.
1
2
3
4
Var
Name
IDATSV
ITIMSV
IELPSV
ISTPSV
Unit
EST
s
Data
Type
Integer*4
Integer*4
Integer*4
Integer*4
Description
Current scenario Julian date as YYDDD
Current scenario time as HHMMSS
Elapsed time since scenario start
Step number
82
-------
2.7.1.2.2.2 Text pointers record- This record contains the current time step value for
the 14 variable-state subroutine text pointers. The WRITE and FORMAT statements
are listed below, and the record's variables are shown in Table 53.
C write text pointers
C
WRSVPT * 2
WRITE(UNITSV, FMT = 1033, IOSTAT = IOST)
& BIGHPT, LILGPT,
& BCPSPT, BMPSPT, BTPSPT, CNPSPT, ICPSPT,
& RDBCPT, ROBMPT, RDBTPT, RDCNPT, RDICPT,
& URCNPT, WRSVPT
1033 FORMAT(1X, 14(13, 1X»
TABLE 53. TEXT POINTER RECORD VARIABLES
Var
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
2.7.1.2.2.3
Var
Name Unit
BIGMPT
LILGPT
BCPSPT
BMPSPT
BTPSPT
CNPSPT
ICPSPT
RDBCPT
RDBMPT
RDBTPT
RDCNPT
RDICPT
WRCNPT
WRSVPT
Data
Type
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Description
Text pointer in subroutine BIGGAM
Text pointer in subroutine LILGAM
Text pointer in subroutine BCPRCS
Text pointer in subroutine BMPRCS
Text pointer in subroutine BTPRCS
Text pointer in subroutine CNPRCS
Text pointer in subroutine ICPRCS
Text pointer in subroutine RDBCON
Text pointer in subroutine RDBMAT
Text pointer in subroutine RDBTRK
Text pointer in subroutine RDCONC
Text pointer in subroutine RDICON
Text pointer in subroutine WRCONC
Text pointer in subroutine WRSTAV
Row counters record- In addition to text pointers, some of the subroutines
also maintain variable indices that point to the row value of the grid currently being
executed. These pointers are also maintained in the state vector file. The intent of the
original design concept was to enable the model to be restarted not only at the start of a
time step but also within a time step on a row boundary. However, in ROM2.1 this
feature was not fully implemented. Therefore, ROM2.1 can be restarted only at time step
boundaries. The WRITE and FORMAT statements are listed below, and the record's
variables are shown in Table 54.
C write row counters
C
WRITE(UNITSV, FMT = 1035, IOSTAT = IOST)
& BHPSRU, RDBMRU, RDBTRW, RDCNRU
1035 FORMATdX, 4(13, 1X»
83
-------
TABLE 54. ROW COUNTERS RECORD VARIABLES
Var
No.
1
2
3
4
2.7.1.2.2.4
Var
Name
BMPSRW
RDBMRW
RDBTRW
RDCNRW
Model and file
Data
Unit Type
Integer*4
Integer*4
Integer*4
Integer*4
Description
Row counter in subroutine BMPRCS
Row counter in subroutine RDBMAT
Row counter in subroutine RDBTRK
Row counter in subroutine RDCONC
time step header records- These two records save the process
clock steps of the principal processes (variable-state subroutines) that (1) maintain the
model scenario time (BIGGAM), and (2) manage the ROM's data files (BCPRCS, BMPRCS,
BTPRCS, CNPRCS, and ICPRCS. These data ensure that the files are kept in step with one
another. The WRITE and FORMAT statements are listed below, and the records'
variables are shown in Table 55.
C write scenario time
C
URITE(UNITSV, FMT * 1037, IOSTAT = IOST)
& MODATE, MOTIME, MDELAP, MDSTEP,
& BCDATE, BCTIME, BCELAP, BCSTEP,
& BMDATE, BMTIME, BMELAP, BMSTEP
1037 FORMATOX, 3(15. 1X, 16, 1X, 18, 14, IX))
C
WRITECUNITSV, FMT = 1037, IOSTAT = IOST)
& BTDATE, BTTIME, BTELAP, BTSTEP,
& CNDATE, CNTIME, CNELAP, CNSTEP,
& ICOATE, ICTIME, ICELAP, ICSTEP
84
-------
TABLE 55. MODEL AND FILE TIME STEP HEADER RECORDS VARIABLES
Var
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
Var
Name
MDDATE
MDTIME
MDELAP
MDSTEP
BCDATE
BCTIME
BCELAP
BCSTEP
BMDATE
BMTIME
BMELAP
BMSTEP
BTDATE
BTTIME
BTELAP
BTSTEP
CNDATE
CNTIME
CNELAP
CNSTEP
ICDATE
ICTIME
ICELAP
ICSTEP
Unit
EST
s
EST
s
EST
s
EST
s
EST
s
EST
s
Data
Type
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Description a
Current scenario date for MODEL
Current scenario time for MODEL
Time elapsed since start for MODEL
Current scenario step for MODEL
Current scenario date for BCON file
Current scenario time for BCON file
Time elapsed since start for BCON file
Current scenario step for BCON file
Current scenario date for BMAT file
Current scenario time for BMAT file
Time elapsed since start for BMAT file
Current scenario step for BMAT file
Current scenario date for BTRK file
Current scenario time for BTRK file
Time elapsed since start for BTRK file
Current scenario step for BTRK file
Current scenario date for CONC file
Current scenario time for CONC file
Time elapsed since start for CONC file
Current scenario step for CONC file
Current scenario date for ICON file
Current scenario time for ICON file
Time elapsed since start for ICON file
Current scenario step for ICON file
a All dates are Julian, i.e., YYDDD; all times are as HHMMSS.
2.7.2 Processing That Takes Place for Restarting Model Execution (how the RESTRT file gets read)
The Core Model main program, RUNMGR, calls subroutine RDSTAV to open and position the
RESTRT file. RDSTAV does this by calling subroutines OPSTAV and POSTAV. OPSTAV opens the file
and extracts the header data in the same manner as described above for subroutine WRSTAV. POSTAV
positions the file to the requested scenario date and time. RDSTAV then reads the RESTRT time step
header, the text pointers records (Section 2.7.1.2.2.2), the row counters record (Section 2.7.1.2.2.3),
and the model and file time step header records (Section 2.7.1.2.2.4), used to reset the clocks main-
tained by the principal processes. The code segments for these steps are shown below.
PROGRAM RUNMGR
INTEGER*4 .... IDATE, ITIME, ...
C open and read state vector file
CALL RDSTAV (IDATE, ITIME)
85
-------
SUBROUTINE RDSTAV (I DATE, ITIME)
C open STATE VECTOR file
CALL OPSTAV
C
C position STATE VECTOR file
CALL POSTAV (IDATE, ITIME)
C
C read STATE VECTOR T.S.H.
READCUNITSV, FHT « 1001, IOSTAT = IOST)
& IDATSV, ITIMSV, IELPSV, ISTPSV
1001 FORMATdX, 15, 1X, 16, 1X, 18, 1X, 14)
C read text pointers
READCUNITSV, FMT = 1003, IOSTAT = IOST)
& BIGMPT, ULGPT,
& BCPSPT, BMPSPT, BTPSPT, CMPSPT, ICPSPT,
& RDBCPT, RDBMPT, RDBTPT, RDCNPT, RDICPT,
& WRCNPT, WRSVPT
1003 FORMATdX, 14(13, 1X))
C read row counters
READCUNITSV, FMT = 1005, IOSTAT = IOST)
& BHPSRU, RDBMRU, ROBTRU, RDCNRU
1005 FORMATdX, 4(13, 1X))
C read scenario time
READCUNITSV, FMT = 1007, IOSTAT = IOST)
& MDDATE, MDTIME, MDELAP, MDSTEP,
& BCDATE, BCTIME, BCELAP, BCSTEP,
& BMDATE, BMTIME, BMELAP, BMSTEP
1007 FORMATdX, 3(15, IX, 16, 1X, 18, 14, 1X))
READCUNITSV, FMT = 1007, IOSTAT = IOST)
& BTDATE, BTTIME, BTELAP, BTSTEP,
& CNDATE, CNTIME, CNELAP, CNSTEP,
& ICDATE, ICTIME, ICELAP, ICSTEP
Once the RESTRT file has been opened, positioned, and read, RUNMOR opens and positions the
BCON, BTRK, and BMAT files to the time step records corresponding to the requested scenario
restart time. Finally, RUNMGR calls subroutine RDCONC to obtain the concentration data for each
grid row from the CONC file for the time step prior to the restart time step. RDCONC calls subroutine
RDFILE to read NWDSCN words of data into the CNFILE common block. RUNMGR copies these
data into the initial conditions buffer (common block BGICCN) that BIGGAM reads to calculate the
advection component of the next time step's concentration field (refer to Section 2.5.2.2.2). .With the
model thus reinitialized, execution proceeds normally from the restart time step. The essential steps
are shown below.
86
-------
REAL*4 BGICCN
INTEGER*^ TOG1, TOG2
COMMON /BGICCN/ TOG1, TOG2, BGICCN(NCOLS, NLEVS, NSPECS, NROWS, 2)
REAL*4 CNFILE
COMMON /CNFILE/ CNFILE(NCOLS, NLEVS, NSPECS)
PROGRAM RUNMGR
INTEGERS ..., IDATE, ITIME IROW, ISPC, LEV. ICOL, ...
C open CONC file and check file header with STATE VECTOR file header
CALL OPCONC
C open BCON file and check file header
CALL OPBCON
C open BTRK file and check file header
CALL OPBTRK
C open BMAT file and check file header
CALL OPBMAT
C position CONC file
CALL POCONC {IDATE, ITIME)
C. position BCON file
CALL POBCON (IDATE, ITIME)
C position BTRK file
CALL POBTRK (IDATE, ITIME)
C position BMAT file
CALL POMXBM (IDATE, ITIME)
C copy CONC rows to ICCN file
00 101 IROW = 1, NROUIN
CALL RDCONC (IOST)
DO 101 ISPC = 1, NSPECS
DO 101 LEV = 1, NLEVS
DO 101 ICOL = 1, NCOLS
BGICCN( ICOL, LEV, ISPC, IROU.TOG2) = CNFILEdCOL, LEV, ISPC)
101 CONTINUE
C
COMMON /CNFILE/ CNFILE(NCOLS, NLEVS, NSPECS)
SUBROUTINE RDCONC (IOST)
INTEGER** IOST, ..., NUDSCN, ISPC
C define record sizes
PARAMETER (..., NUDSCN = NCOLS * NLEVS)
87
-------
DO 211 ISPC = 1, NSPECS
CALL RDFILE (UNITCN, NUDSCN, CNFILE(1,1,ISPC), IOST)
211 CONTINUE
C
SUBROUTINE RDFILEdUNIT, NWORDS, BUFFER, IOST)
INTEGERS IUNIT, NMOROS, BUFFER, IOST
DIMENSION BUFFER (NWORDS)
RE ADO UNIT, IOSTAT-IOST) BUFFER
2.8 THE STOP CHECK (STOPCK) FILE
The stop check (STOPCK) file allows you to shut down a model execution before the run-stream-
designated scenario end time. The STOPCK file can be edited during the course of a model run and its
one line of text altered. If you alter the text to "STOP", then, at the end of the current time step, the model
writes a NEWICON file and exits.
2.8.1 Opening the STOPCK File
At the completion of each scenario time step, the main program, RUNMGR, opens the STOPCK file to
read the one line of text. If the text is "STOP", (no leading or trailing blanks), then RUNMGR shuts the
model down. Otherwise, RUNMGR closes the file, and model execution continues until the end of the
scenario prescribed by the run control parameters. The file is closed so that that there will be no
conflict if a user decides to change the file. There is some risk involved if a user does edit the file to
change it. If the model execution is coincidentally at the point where it must open the STOPCK file
to check its text, and a user already has the file open to edit it, the execution will be aborted by the
system. We consider this risk minimal because the fraction of time users may have the file open is
small compared with the elapsing clock time required for one model step. In order that the computer
operating system does not create new versions of the file every time it is reopened, it is necessary for
the file to be opened with status 'OLD'. This, then, requires that the file exist prior to the start of a
model run. The code segments below illustrate the steps to open the file.
INTEGER** .... UNITST, ...
COMMON / LUNITS / .... UNITST, ...
CHARACTER*!? .... FLNMPR
COMMON / FLNAHS / .... FLNHPR
PROGRAM RUNMGR
C get unit numbers for STOP FLAG file . . .
UNITST = JUNITO
88
-------
C check STOP flag
OPEN (UNIT = UNITST,
& STATUS = 'OLD1,
& ACCESS = 'SEQUENTIAL1,
& FORM = 'FORMATTED1,
& FILE = FLNMST)
2.8.2 The STOPCK File Record
The one line STOPCK file record consists of text that controls the continuation of the model run.
The data variable is loaded into the STOPFG common block from BLOCK DATA at the start of a
model run. The variable remains unchanged unless a user edits the STOPCK file and changes it.
Anything other than "STOP" permits the model run to continue. The relevant code segments are listed
below, and the variable in the STOPCK record is shown in Table 56.
CHARACTERS STOPFG
COMMON /STOPFG/ STOPFG
BLOCK DATA BLKMOD
DATA STOPFG / 'GO ' /
PROGRAM RUNMGR
READCUNITST, 1011) STOPFG
1011 FORMAT(A4)
CLOSE (UNIT = UNITST)
IF (STOPFG .EO. 'STOP') THEN
WRITE(LUNOUT, 1013) STOPFG
1013 FORMAU/ 5X, •> > >STOP FLAG IS SET TO: ',A4)
GO TO 401
END IF
401 CONTINUE
C
C write NEW ICON file
CALL NEWICS
C
WRITE(LUNOUT, 1015)
1015 FORMAT(// 10X, ' Model Run Completion from RUNMGR '//)
C
STOP ' '
TABLE 56. STOPCK RECORD VARIABLE
Var Var Data
No. Name Unit Type Description
1 STOPFG Char*4 Flag that can terminate model execution
89
-------
This page is intentionally left blank.
-------
SECTIONS
THE CORE MODEL CONCENTRATION (CONC) OUTPUT FILE
The concentration file (CONC) contains the concentrations predicted by the ROM for each of the 35 chemi-
cal species, each grid cell, each of the model's three layers, and each time step recorded for one execution of
the Core Model. The chemical species list consists of 35 species, 33 of which are the condensed species
required by the Carbon Bond 4.2 chemical mechanism; the other two species are a tracer species used for
quality assurance monitoring and a nonreactive hydrocarbon species.
When the model starts the execution for each scenario, the scenario initial conditions (either the ICON or the
NEWICON file) are used to start the computation for the subsequent steps in the scenario. These initial
concentration data are copied to the CONC file as the "zeroth" step. The first computed step, i.e., one time
step beyond the initial conditions, is counted as step one. Each succeeding step is incremented by one; a
CONC file for a typical three-day scenario will therefore have 145 steps, starting at day 1, hour 12 and ending
at day 4, hour 12.
During processing of this file, array dimensions are set by parameter statements contained in the INCLUDE
files REGION.EXT and DIMENS.EXT as follows:
INTEGER** NROWS, NCOLS
PARAMETER ( .... NROWS = 52, NCOLS = 64 )
INTEGER*4 NLEVS, NSPECS, .... NPOXSP
PARAMETER (NLEVS = 3, NSPECS =35, ..., NPOXSP = 3)
3.1 OPENING THE CONC FILE
The CONC file is opened at the start of the model scenario. Once the model has started, subroutine
BIGGAM calls LILGAM, which calls CNPRCS, which then calls WRCONQ WRCONC finally calls OPWRCN to
open the file (using the JFILES function subprogram) and write its header records. Note that all CONC
file records have a fixed record length equal to NLEVS x NCOLS.
FLNMCN contains the internal (logical) names for CONC that point to the actual file names in the
execution run stream. FLNMCN is set in the block data module BLKMOD. JUNIT is a function subpro-
gram that returns the next FORTRAN I/O unit number not being used by the model execution (refer to
Section 1.7). The code segments and the FORMAT statements that describe the processing steps to open
the CONC file are listed below.
91
-------
INTEGER*4 .... UNITCN, ...
CHARACTER*12 .... FLNMCN, ...
SUBROUTINE OPWRCN
INTEGER*4 IWDLTH, .... JFILE5
LOGICAL*4 RECFMT, RDONLY
PARAMETER ( IWDLTH = NLEVS * NCOLS,
& RECFMT = .FALSE., RDONLY
.FALSE.)
C open concentration file, unformatted, read/write access
UNITCN = JFILE5 (FLNNCN, RECFMT, RDONLY, IWDLTH)
C
FUNCTION JFILE5 (FNAME, RECFMT, RDONLY, RECLEN)
CHARACTER*!? FNAME, FORM, UN FORM, FORMAT
INTEGERS RECLEN, IDEV, IOST, JFILE5, JUNIT, ...
LOGICAL*^ RECFMT, RDONLY
DATA FORM / 'FORMATTED ' /
DATA UNFORM / 'UNFORMATTED ' /
IDEV = JUNITO
IF (RECFMT) THEN
FORMAT s FORM
ELSE
FORMAT = UNFORM
END IF
IF (RDONLY) THEN
&
&
&
&
&
ELSE
OPEN (UNIT
IOSTAT
FILE
STATUS
ACCESS
FORM
END IF
= IDEV,
= IOST,
= FNAME,
= 'UNKNOWN1.
= 'SEQUENTIAL',
= FORMAT)
JFILE5 = IDEV
RETURN
END
3.2 CONC FILE RECORDS
The structure of the CONC file conforms to the requirements for all ROM Core Model files, and there-
fore contains a standard file header. In addition, the file consists of a data body organized by time steps,
each section of which is headed by a time step header record. Descriptions of the records containing this
information are given below, while Appendix B contains a structure diagram for the CONC file.
92
-------
3.2.1 CONG File Header Records
The data that are written to the CONC file header originate from:
the execution runstream's control cards,
• header information from the BMAT, BTRK, BCON, and ICON files,
variables set by assignment statements in subroutine INIRUN, and
arguments returned from various subroutines called by INIRUN.
These data are loaded into the common blocks in the INCLUDE file HEADIN.EXT at the beginning of
the model execution:
c
C HEADIN.EXT
C
C input header information
C
C used to check file headers on BMATRIX, ICON, and BCON files
C and to create file header on CONC file
c
CHARACTER*80 TEXTIN
CHARACTER'S GRDNIN
CHARACTER*^ SPNMIN, LVNMIN
REAL*4 SULNIN, SUIT IN, NELNIN, NELTIN, DLONIN, DLATIN
INTEGER*^ CDATIN, CTIMIN, SOAT IN, STHRIN, TSTPIN, FRSTIN,
& NCOLIN, NROUIN, NLEVIN, NSPCIN, ICNTIN
C
COMMON /CHARIN/ GRDNIN, SPNMIN(NSPECS), LVNMIN(NLEVS), TEXTINC20)
COMMON /HEAD IN/ CDATIN, CTIMIN. SDATIN, STHRIN, TSTPIN, FRSTIN,
& SWLNIN, SULTIN, NELNIN, NELTIN, DLONIN, DLATIN,
& NCOLIN. NROWIN, NLEVIN, NSPCIN, ICNTIN
C
The data in the above common blocks are used in the creation of the CONC file header. The
INCLUDE file HEADCN.EXT contains the common blocks that are loaded with the CONC file header
variables:
c
C HEADCN.EXT
c
C CONC file header block
C
CHARACTER*80 TEXTCN
CHARACTER'S GRDNCN
CHARACTERS SPNMCN, LVNMCN
REAL*4 SULNCN, SWLTCN, NELNCN, NELTCN, DLONCN, DLATCN
INTEGER*4 CDATCN, CTIHCN, SDATCN. STHRCN, TSTPCN, FRSTCN,
& NCOLCN, NROUCN, NLEVCN. NSPCCN, ICNTCN,
& CDBMCN, CTBMCN, CDBTCN, CTBTCN,
& CDBCCN, CTBCCN, CDICCN, CTICCN
C
COMMON /CHARCN/ GRDNCN, SPNMCN(NSPECS), LVNMCN(NLEVS), TEXTCNC20)
COMMON /HEADCN/ CDATCN, CTIMCN, SDATCN, STHRCN, TSTPCN, FRSTCN,
& SWLNCN, SWLTCN, NELNCN, NELTCN, DLONCN, DLATCN,
& NCOLCN, NROWCN, NLEVCN, NSPCCN, ICNTCN,
& CDBMCN, CTBMCN, CDBTCN, CTBTCN,
& CDBCCN, CTBCCN, CDICCN, CTICCN
C
The first four records comprise the CONC file header.
93
-------
3.2.1.1 Record 1-
The first record contains character strings of alphanumeric data that describe the file's contents.
The data are first loaded into the HEADCN.EXT common blocks. They are then written to the
character buffer SEG1BF, which is then written (unformatted) to the CONC file by subroutine
WRCHAR. The code segments for these steps are shown below, and the variables of record 1 are
shown in Table 57.
INTEGER** .... UNITCN, ...
SUBROUTINE OPWRCN
C CONC header buffers
CHARACTER*(8 * 21 + 4 * 5) SEG1BF
INTEGER** .... IOST, ...
C load CONC header
CDATCN = COATIN
CTIMCN = CTIMIN
SDATCN = SCATIN
STHRCN = STKRIN
TSTPCN = TSTPIN
FRSTCN = FRSTIN
GRDNCN = GRDNIN
SULNCN = SULNIN
SWLTCN = SWLTIN
NELTCN = NELTIN
NELNCN = NELNIN
DLONCN = DLONIN
OLATCN = DLATIN
NCOLCN = NCOLIN
NROWCN = NROUIN
NLEVCN = NLEVIN
NSPCCN = NSPCIN
ICNTCN = ICNTIN + 1
C
C set values for creation dates/times of BM, BT, BC and 1C files
CDBMCN = CDATBM
CTBMCN = CTIMBM
CDBTCN = CDATBT
CTBTCN = CTIMBT
CDBCCN = CDATBC
CTBCCN = CTINBC
CDICCN = COAT1C
CTICCN = CTIM1C
C
C write 1st segment
WRITECSEG1BF. 1001, IOSTAT = IOST)
& CDATCN. CTIMCN, SDATCN. STHRCN, TSTPCN, FRSTCN,
& GRDNCN,
& SWLNCN, SWLTCN, NELNCN, NELTCN,
& DLONCN, DLATCN,
& NCOLCN, NROWCN, NLEVCN, NSPCCN,
& CDBMCN, CTBMCN,
& CDBTCN, CTBTCN,
& CDBCCN. CTBCCN,
& CDICCN, CTICCN,
& ICNTCN
1001 FORMATC6I8.8. A8, 4F8.3. 2F8.5, 414.4, 818.8, 14.4)
C
CALL URCHAR (UNITCN, SEG1BF, IOST)
94
-------
SUBROUTINE WRCHAR (IUNIT, CHBUF, IOST)
IMPLICIT NONE
INTEGER*4 IUNIT, IOST
CHARACTER*^) CHBUFF
URITEdUNIT, IOST AT = IOST) CHBUFF
RETURN
END
TABLE 57. CONC RECORD 1 VARIABLES
Var
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
Var
Name
CDATCN
CTIMCN
SDATCN
STHRCN
TSTPCN
FRSTCN
GRDNCN
SWLNCN
SWLTCN
NELNCN
NELTCN
DLONCN
DLATCN
NCOLCN
NROWCN
NLEVCN
NSPCCN
CDBMCN
CTBMCN
CDBTCN
CTBTCN
CDBCCN
CTBCCN
CDICCN
CTICCN
ICNTCN
Unit
EST
EST
s
s
°W
°N
°W
°N
°W
°N
EST
EST
EST
EST
Data
Type
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Char*8
Real*4
Real*4
Real*4
Real*4
Real*4
Real*4
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Description
File creation date as MMDDYY
File creation time as HHMMSS
Julian stan date of scenario as YYDDD
Start hour of scenario (00 to 23)
Time step size for simulation
Time to first step
Grid definition name
Longitude of southwest corner of grid
Latitude of southwest corner of grid
Longitude of northeast corner of grid .
Latitude of northeast corner of grid
Grid cell longitudinal increment
Grid cell latitudinal increment
Number of columns in grid
Number of rows in grid
Number of levels in the simulation
Number of species in the CONC file
Creation date of B-matrix file (BMAT) from
which the CONC file was generated (MMDDYY)
Creation time of B-matrix file (BMAT) from
which the CONC file was generated (HHMMSS)
Creation date of backtrack file (BTRK) from
which the CONC file was generated (MMDDYY)
Creation time of backtrack file (BTRK) from
which the CONC file was generated (HHMMSS)
Creation date of boundary conditions (BOON) file
from which the CONC file was generated
(MMDDYY)
Creation time of boundary conditions (BCON)
file from which the CONC file was generated
(HHMMSS)
Creation date of initial conditions (CONC) file
from which the CONC file was generated
(MMDDYY)
Creation time of initial conditions (CONC) file
from which the CONC file was generated
(HHMMSS)
Number of text records
95
-------
3.2.1.2 Record 2-
This record contains the list of species names for which the Core Model computes concentration
outputs. The data are written to the character buffer SPNMBF, which is then written (unfo-
rmatted) to the CONC file. These steps are listed below, and the variable of record 2 is shown in
Table 58.
SUBROUTINE OPWRCN
CHARACTER*^ * NSPECS) SPNMBF
INTEGER** .... ISPC, .... IOST, ...
DO 101 ISPC = 1, NSPCIN
SPNMCN(ISPC) = SPNMIN(ISPC)
101 CONTINUE
C write the species names
WRITE(SPNMBF, 1003, IOSTAT = IOST)
& (SPNNCN(ISPC), ISPC = 1, MSPCCN)
1003 FORMAT( (A4) )
CALL URCHAR (UNITCN, SPNMBF, IOST)
TABLE 58. CONC RECORD 2 VARIABLE
Var
No.
1
Var
Name Unit
SPNMCN*
Data
Type
CharM
Description
Name of chemical species k t
f A list of chemical species names can be found in Table 1.
3.2.1.3 Record 3-
This record contains the list of Core Model layer names. The data are written to the character
buffer LEVNBF, which is then written (unformatted) to the CONC file. The steps are listed
below, and the variable of record 3 is shown in Table 59.
SUBROUTINE OPWRCN
CHARACTER*^ * NLEVS) LEVNBF
INTEGER** ..., ILEV, .... IOST, ...
LVNMCN(I) = ' V
LVNMCN(2) = ' 2'
LVNMCNO) = ' 3'
C write the level names
WRITE(LEVNBF, 1005, IOSTAT = IOST)
& (LVNMCN(ILEV), ILEV = 1. NLEVCN)
1005 FORMAT( (A4) )
CALL WRCHAR (UNITCN, LEVNBF, IOST)
96
-------
TABLE 59. CONC RECORD 3 VARIABLE
Var
No.
1
Var
Name Unit
LVNMCNL
Data
Type
CharM
Description
Name of layer L
3.2.1.4 Records 4 - (4 + ICNTCN)--
These records contain descriptive text. The first text record is created by subroutine OPWRCN and
contains the model version name. The subsequent records consist of descriptive text to be copied
to the CONC file header. The text data were optionally entered as part of the model execution
run stream (e.g., see Section 4.3 for an IBM run stream). One 80-character string is written to
each record by the following statements. The variable of the records is shown in Table 60.
SUBROUTINE OPWRCN
INTEGER*4 .... ITXT, ...
C
C copy input text records to CONC file
TEXTCN(I) = 'ROM2.1 '
CALL WRCHAR (UNITCN, TEXTCN(I), IOST)
C
DO 103 ITXT = 2. ICNTCN
TEXTCN(ITXT) = TEXTINdTXT - 1)
CALL WRCHAR (UNITCN, TEXTCN(ITXT), IOST)
103 CONTINUE
TABLE 60. CONC RECORDS 4 - (4+ICNTCN) VARIABLE
Var
No.
Var
Name
Unit
Data
Type
Description
1 TEXTCNn Char*80 Text string of n lines
3.2.2 CONC File Body Records
3.2.2.1 Time Step Header Record-
At the start of each model time step, subroutine LILGAM calls CNPRCS, which in turn calls
WRCONC to write the CONC file time step header. There is one time step header record written
for each time step increment on the CONC file. The time step header data are. contained in the
common block RTSHCN. The following code segments illustrate how the data are written to the
97
-------
CONC file. The first variable of the common block RTSHCN is used in subroutine WRFILE'S
argument list, and NCOLS x NLEVS is specified as the number of words to be written. The
record's variables are shown in Table 61.
COMMON / RTSHCN / DATCN, TIMCN, ELPCN, STPCN
SUBROUTINE WRCONC
INTEGER** IULTSH, ..., IOST
C define record Lengths
PARAMETER ( IWLTSN = NCOLS * NLEVS, ... )
CALL WRFILE (UNITCN, IULTSH, OATCN, IOST)
SUBROUTINE WRFILE (IUNIT, (DWORDS, BUFFER, IOST)
IMPLICIT NONE
INTEGER*4 IUNIT, NWORDS, BUFFER, IOST
DIMENSION BUFFER(NWORDS)
URITEdUNIT, IOSTAT = IOST) BUFFER
RETURN
END
TABLE (1. CONC TIME STEP HEADER RECORD VARIABLES
Var-
No.
1
2
3
4
Var
Name
DATCN
TIMCN
ELPCN
STPCN
Data
Unit Type
RealM
RealM
s RealM
RealM
Description
Current time step date as YYDDD
Current time step time as HHMMSS
Elapsed time since scenario start
Step number on the CONC file
3.2.2.2 Data Records-
Each data record contains the concentration data for one chemical species in one row of the
domain grid. Subroutine LILGAM calls CNPRCS after it completes the calculations for the vertical
flux and the chemical reaction components of the concentration for all layers and all species in
one row of the grid. CNPRCS calls WRCONC, which calls WRFILE to write that row of data to the
CONC file for the current time step.
For each time step increment in the CONC file there are NSPECS records for each of the
NROWS domain rows. These data are written by referencing the memory address corresponding
to the column 1 and level 1 indices of the array in the common block CNFILE for each species,
and passing the number of words to be written to WRFILE. The code segments for these steps are
listed below, and the variable for these records is shown in Table 62.
98
-------
COMMON /CNFILE/ CNFILE(NCOLS, NLEVS, NSPECS)
SUBROUTINE WRCONC
INTEGER*4 ..., IUDLN2, .... ISPC, IOST
C
C define record lengths
PARAMETER ( ..., IUDLN2 = NCOLS * NLEVS )
C write CONC row
DO 301 ISPC = 1, NSPECS
CALL URFILE (UNITCN, IUDLN2, CNFILE(1,1,ISPC), IOST)
301 CONTINUE
TABLE 62. CONC DATA VARIABLE
Var Var Data
No. Name Unit Type Description
ppm Real*4 Chemical species concentrations in column i
and layer L for species k
99
-------
This page is intentionally left blank.
-------
APPENDIX A
JACKSON STRUCTURED PROGRAMMING (JSP)
A-l
-------
We wrote much of the ROM Core and Processor codes using the JSP methodology because it allows us to (1)
eliminate the need for intermediate temporary files, (2) it eliminates the need for our knowing exactly when,
and in what order, to call subroutines and functions (this scheduling is accomplished automatically), and (3)
for future model upgrades, its modular format permits us to easily add or delete processors as we find neces-
sary.
A useful JSP transformation is to write a stand-alone program, then invert it. We show examples of program
inversion in Section A.2. Section A.1 gives you some preliminary information about JSP. For a full discus-
sion of JSP, see Jackson (1975).
A.1 AN INTRODUCTION TO JSP FLOW DIAGRAMS
A is a sequence of B followed by C followed by D
A is a selection of either B or C or D
A is an iteration of B zero or more times
A-2
-------
A is a sequence of B followed by C (which is an
iteration of D), followed by E (which is a selection
of either F or G
The next two diagrams show you two common diagramming errors and their corrected versions.
INCORRECT
o o
BCD
INCORRECT
B
D
We next show a simple specification flow diagram.
CORRECT
CORRECT
B
A-3
-------
A.2 EXAMPLES OF PROGRAM INVERSION AND THE USE OF STATE VECTORS
To explain program inversion, refer to the specification flow diagram above. Program P reads file A and
writes to file B. B is then read by program Q, which finally writes file C. If we invert P with respect to its
input data stream, we could make p a subroutine of Q, so that whenever Q requires a record from file B it
calls P. Conversely, Q could be made a subroutine of P, so that when p wants to write a record to file B, it
calls Q. The diagram below shows an example of the inversion of a linear process sequence. Part (a) is a
system specification diagram. Pan (b) shows the inverted programs, and is termed a system implementa-
tion diagram.
X-~^X
^^
a)
SCHEDULER
(main program)
In (b) above, data stream B is replaced by the connecting channel between program P and the Scheduler;
data streams B and C are replaced by the channel between program Q and the Scheduler; finally, data
stream C is replaced by the channel between program R and the Scheduler.
These transformations eliminate the need for actual intermediate files, and establish an automatic
read/write sequence that is useful in system design. However, the program structure of the process con-
tinues to reflect the files' existence. The intermediate files, although nonexistent, contain an implicit
structure reminiscent of the main files' structure.
Sections A.2.1 and A.2.2 show two examples of program inversion. The text pointer TXTPTR allows the
program to always track exactly where it is within itself, and to return to that point when needed. Note
that when inverted, the program becomes a variable-state subroutine.
A-4
-------
A.2.1 Program Inversion with Respect to Its Input Data Stream
Before Inversion:
PROGRAM WRFILE
After Inversion
SUBROUTINE WRFILE (INVAR)
declarations
declarations
OPEN (OUTPUT FILE)
OPEN (INTERMEDIATE FILE)
101 CONTINUE
READ (INTERMEDIATE FILE) INVAR
IF (INVAR .EQ. EOFT GO TO 201
DATA TXTPTR /!/
GO TO (10001, 10002) TXTPTR
10001 CONTINUE
OPEN (OUTPUT FILE)
101 CONTINUE
IF (INVAR .EQ. EOF) GO TO 201
process data
process data
WRITE (OUTPUT_FILE) OUTVAR
GO TO 101
201 CONTINUE
STOP
END
WRITE (OUTPUT_FILE) OUTVAR
TXTPTR = 2
RETURN
10002 CONTINUE
GO TO 101
201 CONTINUE
RETURN
END
A-5
-------
A.2.2 Program Inversion with Respect to Its Output Data Stream
Before Inversion:
PROGRAM RDFILE
declarations
OPEN (INPUT FILE)
OPEN (INTERMEDIATE FILE)
101 CONTINUE
READ (INPUT FILE) INVAR
IF (INVAR -EQ. EOF) GO TO 201
process data
WRITE (INTERMEDIATE FILE) OUTVAR
GO TO 101
201 CONTINUE
STOP
END
After Inversion
SUBROUTINE RDFILE (INVAR)
declarations
DATA TXTPTR /I/
GO TO (10001, 10002) TXTPTR
10001 CONTINUE
OPEN (INPUT FILE)
101 CONTINUE
READ (INPUT FILE) INVAR
IF (INVAR .EQ. EOF) GO TO 201
process data
TXTPTR = 2
RETURN
10002 CONTINUE
GO TO 101
201 CONTINUE
RETURN
END
A.3 REFERENCE AND BIBLIOGRAPHY
Jackson, M. A., 1975. Principles of Program Design. A.P.I.C. Studies in Data Processing 12. Academic
Press, London, United Kingdom. 299 pp.
Jackson, M. A, 1983. System Development. Prentice-Hall International Series in Computer Science.
Prentice/Hall International, Englewood Cliffs, New Jersey. 418 pp.
A-6
-------
APPENDIX B
DESIGN AND STRUCTURE DIAGRAMS FOR THE ROM2.1 DATA FILES
IN THE ROMNET REGION
Refer to Appendix A for the explanation of symbols used.
Also note that boxes drawn with broken lines denote logical records.
B-l
-------
I
1st
SEGMENT
I
I
SPECIES
NAMES
...I...
I
LEVEL
NAMES
...1...
SEG1L NSPECS NLEVS
WORDS WORDS WORDS
|
w
T
SPECIES
(NSPECS)
1
*
LEVEL
(NLEVS)
1
TEXT
SEGMENT
1
TEXT
RECORDS
(ICNTBC)
— J-.-,
NTEXT
WORDS
SEG1L
NSPECS
NLEVS
ICNTBC
NTEXT
NSTEPS
(19 *2) + (5* 1)
35
3
20
20
Figure B-l. The BCON file (page 1 of 3)
B-2
-------
I" '
C* 8
FILE
CREATE
DATE
C* 8
FILE
CREATE
TIME
1
C* 8
SCENARIO
START
DATE
C
» 8
SCENARIO
START
HOUR
C* 8
TIME STEP
SIZE
1
C* 8
TIME TO
FIRST
STEP
C
\t
1 —
C* 8
GRID
NAME
C* 8
SOUTHWEST
CORNER
LONGITUDE
1
C* 8
SOUTHWEST
CORNER
• LATITUDE
1
c* e
NORTHEAST
CORNER
LONGITUDE
C* 8
NORTHEAST
CORNER
LATITUDE
|
C* 8
CELL
LONGITUDE
INCREMENT
C* 8
CELL-
LATITUDE
INCREMENT
A
v*
1 1 1 1
C *4
NO OF
COLUMNS
C*4
NO OF
ROWS
C*4
NO OF
LEVELS
C *4
NO. OF
SPECIES
1
C*4
NO. OF
TEXT
STRINGS
The BCON file (page 2 of 3)
B-3
-------
The BCON file (page 3 of 3)
B-4
-------
§"8
cou.0
*
in
LU CL O.
i-
-------
$
, 1 . , L_ i
C* 8
FILE
CREATE
DATE
C* 8
FILE
CREATE
TIME
C* 8
SCENARIO
START
DATE
"1 ' i 1
C*8
SCENARIO
START
HOUR
C* 8
TIME STEP
SIZE
C* 8
TIME TO
FIRST
STEP
r
\t
C« 12
FILE
NAME
I
C* 8
RLE
CREATE
DATE
| 2
C*8
RLE
CREATE
TIME
C* 8
RLE
UPDATE
DATE
C*8
RLE
UPDATE
TIME
The BMAT file (page 2 of 5)
B-6
-------
CM en
o o
uj cc
a: O
z 3
— in
oo
UJ CC
SI
<£ CM O *- CO tO
,. O Q.
C to cr « — N
T SlJ a x o o
rn O O (O O UJ Ul
™0:b2a.a: DC
z zz m zz z
The BMAT file (page 3 of 5)
B-7
-------
CO
CO
CO
ffi
8
*8
*
*
#
*
o
*
o
*
o
*
8
*g
*
U
*
g
* _
••1
* ou
CO <
* SB
o: u- 3
CO ^J
o 5
01 *? 3
* IU
X *" ^
s§
* eoui
X CO 3
CC CO -j)
£5
S S3
"5
* LU
ff W 3
* IU
OC 03
sg
>
* UJ
« UJ
* S3
>
— | in >
—
§
8
CO
s
C/)
o
5
0
CO
O
CO
O
—
*
8
*
_j
o
#
_J
0
#
*
,
o
* _
CO
°g
—
*
Q.
CO
g
*
0.
CO
g
#
w
X
0
*
a
CO
X
O
*
a.
CO
S
* 67
0. CO
CO X
x o
O i
* o UJ
1 53
85
+ . m
CO 3?
05
*
* UJ
o: co o
r*i _j
C5<
* OUJ
oc < =>
5>
^ JJJ
o:' w 3
05
•v
o: o> S
°§
The BMAT file (page 4 of 5)
B-8
-------
1 1 1
RRHO
TTEMP
WWVC
TTHETA
1 1 [
PPSI2
220
221
222
223
COL *
(NCOLS)
COL
COL
COL
COL
COL
COL
COL
COL
LEV *
(NLEVS)
LEV
LEV
R *4
RRHO
VALUE
R*4
TTEMP
VALUE
R*4
WWVC
VALUE
R*4
TTHETA
VALUE
R*4
PPSI2
VALUE
R*4
220
VALUE
R*4
22t
VALUE
R*4
222
VALUE
R*4
223
VALUE
The BMAT file (page (Tof 5)
B-9
-------
1
\2\
\
C*80
STRING
SEG1L =
NMFBT =
MIFRECL •=
ICNTBT s
NTEXT *
(13*2)* (4*1)
6
(3 » 4 * 2) * NMFBT
20
20
NSTEPS s 144
Figure B-3. The BTRK file (page 1 of 3)
B-10
-------
1
1
C* 8
FILE
CREATE
DATE
1
C* 8
FILE
CREATE
TIME
1
C» 8
SCENARIO
START
DATE
C* 8
SCENARIO
START
HOUR
|
C* 8
TIME STEP
SIZE
1
C* 8
TIME TO
FIRST
STEP
d
V.
1
C* 8
GRID
NAME
"
C* 8
SOUTHWEST
CORNER
LONGITUDE
1 "
C* 8
SOUTHWEST
CORNER
LATITUDE
C* 8
NORTHEAST
CORNER
LONGITUDE
C* 8
NORTHEAST
CORNER
LATITUDE
|
C* 8
CELL
LONGITUDE
INCREMENT
C* 8
CELL
LATITUDE
INCREMENT
>
Cj
C*4
NO. OF
COLUMNS
C*4
NO
OF
ROWS
|
C*4
NO OF
MIF FILES
1
C*4
NO. OF
TEXT
STRINGS
1
C* 12
FILE
NAME
C* 8
FILE
CREATE
DATE
I2
C<
• 8
FILE
CREATE
TIME
|
C* 8.
FILE
UPDATE
DATE
C*8
FILE
UPDATE
TIME
The BTRK file (page 2 of 3)
B-ll
-------
—
-
-
-
-
-
-
-
—
I
KKH2V
KKH1V
KKH3U
KKH2U
KKH1U
*
rg
i
3
3
5
—
—
—
*
d
o
*
d
*
*
#
d
o
*g
*
o
Q
*
o
*
*
d
o
#
o
*5>
°i
—
-
•*
* > UJ
cr « 3
^ <
^ >
+ > Ul
•r ||
* > ui
* 53
S5
* r> ui
X — l
* w "*
" 11
tt 0 =
>
V
* UJ
* UI
•31
^
* UJ
a: ^3
* UJ
V)
d
s 2-
UJ «
UJ ^ *-
1
^ ii 11 n n
a.
UJ c/) £
Z I ^ > QD
2 r z z r
The BTRK file (page 3 of 3)
B-12
-------
1
1st
SEGMENT
SPECIES
NAMES
...I..,
LEVEL
NAMES
I
TEXT
SEGMENT
- - - -,
SEG1L NSPECS NLEVS
WORDS WORDS WORDS
1 1 *---.-.._• [
*
TEXT '
RECORD
(ICNTIC)
SEG1L
NSPECS
NLEVS
ICNTIC
NTEXT
(21
35
3
20
20
•2).(5* 1)
Figure B-4. The ICON file (page 1 of 3)
B-13
-------
1
c* a
FILE
CREATE
DATE
1
C* 8
FILE
CREATE
TIME
1
C* 8
SCENARIO
START
DATE
>
C* 8
SCENARIO
START
HOUR
1
C* 8
TIME STEP
SIZE
1
C* 8
TIME TO
FIRST
STEP
^
V
,
C*8
C* 8
SOUTHWEST
CORNER
LONGITUDE
'
C* 8
SOUTHWEST
CORNER
LATITUDE
1 ' "
C* 8
NORTHEAST
CORNER
LONGITUDE
1 ~
C* 8
NORTHEAST
CORNER
LATITUDE
|
C* 8
CELL
LONGITUDE
INCREMENT
| '
c* e
CELL
LATITUDE
INCREMENT
II
1
C*4
NO. OF
COLUMNS
1
C *4
NO OF
ROWS
1
C*4
NO. OF
LEVELS
T~
C*4
NO OF
SPECIES
—
DUMMY
~
C
* 4
NO OF
TEXT
STRINGS
The ICON file (page 2 of 3)
B-14
-------
1
R * 4
STEP
DATE
1
R *4
STEP
TIME
1
R*4
ELAPSED
TIME
1
R*4
STEP
NUMBER
ROMNET PARAMETERS
NTSH * 4
NROWS = 52
NCOLS = 64
NTSB = NLEVS * NCOLS
The ICON file (page 3 of 3)
B-15
-------
1
1st
SEGMENT
,.--L_,
SEG1L
WORDS
'-"[•--
1
SCENARIO
CONC
FILE
1
FILE FILE
HEADER BODY
1
1 1 *
1 " ' " 11Mb
STEP
SPECIES LEVEL TEXT (NSTEPS)
NAMES NAMES SEGMENT
1 ' ->
, , i^2J
NSPECS NLEVS
WORDS WORDS MODEL
NAME
""I \""
V
TEXT
BODY
* + 1 1
SPECIES LEVEL
(NSPECS) (NLEVS) NTEXT
WORDS
i: .........
*
TEXT
RECORD
(ICNTCN)
C*4 C*4 1 1
SPECIES LEVEL C * 80
NAME NAME MODEL
NAME
VALUE
NTEXT
WORDS
....(....
C* 80
STRING
SEG1L
NSPECS
NLEVS
ICNTCN
NTEXT
NSTEPS
(21
35
3
20
20
145
2)* (5*1)
Figure B-5. The CONC file (page 1 of 3)
B-16
-------
1
C* 8
FILE
CREATE
DATE
|
C*8
FILE
CREATE
TIME
|
C*8
SCENARIO
START
DATE
'
C*8
SCENARIO
START
HOUR
|
C* 8
TIME STEP
SIZE
1"
C* 8
TIME TO
FIRST
STEP
d
(£
D
1
c*e
BOON FILE
CREATE
DATE
C*B
BOON FILE
CREATE
TIME
I
C*8
ICON FILE
CREATE
DATE
1
C*8
ICON FILE
CREATE
TIME
1
C*4
NO OF
TEXT
STRINGS
The CONC file (page 2 of 3)
B-17
-------
1
R*4
STEP
DATE
' NTSH
, WORDS
i
L - _ _
1
1
R*4
STEP
TIME
1
R *4
ELAPSED
TIME
1
R*4
STEP
NUMBER
ROMNET PARAMETERS
NTSH
NROWS
NCOLS
NTSB
4
52
64
NLEVS * NCOLS
The CONC file (page 3 of 3)
B-18
-------
APPENDIX C
DESIGN AND STRUCTURE DIAGRAMS
FOR THE PRINCIPAL ROM2.1 SUBROUTINES
Refer to Appendix A for the explanation of symbols used.
Subprograms are indicated in italics.
T.S.H. = Time Step Header
C-l
-------
Figure C-l. Core Model system specification diagram.
C-2
-------
Figure C-2. Core Model Jackson Structured Design implementation diagram.
C-3
-------
Figure C-3. RUNMGR (page 1 of 2).
C-4
-------
1
READ
CONG
TSH
RDCONC
1
COPY
CONC
TO
ICCN
ROW *
COPY
ARRAY
RUNMGR(page2of2).
C-5
-------
o
SOUTHERN
BC
o
NORTHERN
BC
O
EASTERN
BC
|
O
WESTERN
BC
Figure C-4. BIGGAM (page 1 of 1).
C-6
-------
INIT
1C
TIME
WRITE
BGIC
T.SH
READ
1C
ROW
BUFFER
BGIC
ROW
CLOSIC
1
WRITE
BGIC
FILE
TSTEP *
1
TSTEP
BODY
UPDATE
1C "
TIME
Figure C-5. ICPRCS (page 1 of 1).
C-7
-------
Figure C-6. RDICON (page 1 of 1).
C-8
-------
CLOCK1
*
TSTEP
I
BUFFER
BGBC
RECORD
WRITE
BGBC
REC
UPDATE
BC
TIME
READ
BC
T.S.H
Figure C-7. BCPRCS (page 1 of 1).
C-9
-------
1
RDFILE
CONVERT
TIME
DATA
TYPE
1
WRITE
BC
T.S.H.
TSTEP
TSTEP
BODY
1
WRITE
BC
FILE
READ
BCON
T.S.H.
RDFILE
RDFILE
RDFILE
RDFILE
RDFILE
Figure C-8. RDBCON (page 1 of 1).
C-10
-------
UPDATE
BT
TIME
READ
BT
TSH
Figure C-9. BTPRCS (page 1 of 1).
C-ll
-------
flosr
TSTEP *
1
ROW *
RDBT
Figure C-10. RDBTRK (page l of l).
C-12
-------
WRITE
HDR
READ
RECORD
Figure Oil. RDBT (page 1 of 1).
C-13
-------
CNPRCS
TSTEP
Figure C-12. LILGAM (page 1 of 2).
C-14
-------
1
SET
FRAX
LEV *
NORMAL
RATE
CONSTANTS
LEVEL
BODY
RATED
RATED
PO1
LlLGAM(page2of2).
C-15
-------
READ
BM
T.S.H
I
RDBMAT
INDEX
BMAT
SPECIES
I
ORSPBM
1
INIT
BM
TIME
CLOCK1
PROC
BODY
I
TSTEP *
WRITE
LGBM
T.S.H.
I
TSTEP
BODY
UPDATE
BM
TIME
|
READ
BM
T.SH
I
READ
BM
ROW
ROW *
BUFFER
LGBM
ROW
CLOCKS
I
WRITE
LGBM
ROW
Figure C-13. BMPRCS (page 1 of 1).
C-16
-------
TSTEP
BODY
READ
BMAT
T.S.H
Figure C-14. RDBMAT (page 1 of 1).
C-17
-------
REC
*
(EOF)
Figure C-15. RDMXBM (page 1 of l).
C-18
-------
TSTEP
BODY
READ
ICCN
FILE
Figure C-16. CNPRCS (page 1 of 1).
C-19
-------
1 1 — ' — , ,
CONVERT
TIME
DATA
TYPE
WRITE
CONC
T.S.H.
1
TSTEP
BODY
1
READ
CN
T.S.H.
Figure C-17. WRCONC (page 1 of 1).
C-20
-------
1
OPEN
CONC
FILE
1
POSITION
CONC
FILE
READ
CONC
T.SH
1
PROC
BODY
Figure C-18. RDCONC (page 1 of 1).
C-21
-------
This page is intentionally left blank.
-------
APPENDIX D
SAMPLE COMPILE AND LINK STREAM FOR THE
UNIPROCESSOR ROM2.1
D-l
-------
//GJDCROM JOB (NER1RSMRP,B012),'JORDAN ',MSGCLASS=P, 0000000
// NOTIFY=GJD,TIME=(,45),PRTY=2 0000000
/*JOBPARM LINES=999
//*FTER JKVOFF03
/*ROUTE PRINT RMT378
//VSF2CL PROC FVPGM=FORTVS2,FVREGN=1400K,FVPDECK=NODECK,FVPOLST=NOLIST,
// FVPOPT=0,FVTERM='SYSOUT=*',PGMNAME=ROM21R,PGMLIB='&&GOSET',
// FVLNSPC='3200,(25,6)'
//FORT EXEC PGM=&FVPGM,REGION=&FVREGN,COND=(4,LT),
// PARM='4FVPDECK,&FVPOLST,iOPT(&FVPOPT)'
//STEPLIB DO DSN=SYS1.VSF2COMP,DISP=SHR
//SYSPRINT DD SYSOUT=*,DCB=BLKSIZE=3429
//SYSTERM DD &FVTERM
//SYSPUNCH DD SYSOUT=B,DCB=BLICSIZE=3440
//SYSLIN DD DSN=&&LOADSET,DISP=(MOD,PASS),UNIT=SYSDA,
// SPACE=(&FVLNSPC),DCB=BLKSIZE=3200
//LKED EXEC PGM=IEUL,REGION=768K,COND=(4,LT>,
// PARM='LET,LIST,XREF'
//SYSPRINT DD SYSOUT=*
//SYSL1B DD DSN=SYS1.VSF2FORT,DISP=SHR
// DD DSN=BTLNER1.ROM.LOADLIB,DISP=SHR
//SYSUT1 DD UNIT=SYSDA,SPACE=(1024,<200,20))
//SYSLMOD DD DSN=&PGHLIB(&PGMNAHE),DISP=(,PASS),UMIT=SYSOA,
// SPACE=(TRK,(10,10,1),RLSE)
//SYSLIN DD DSN=&&LOADSET,DISP=(OLD,DELETE)
// DD DDNAME=SYSIN
// PEND
//VSFVCL EXEC VSF2CL,PGMLIB='GJDNER1.ROMNET1.LOADLIB1. 00000300
// FVREGN=4M,FVLNSPC='3200,(250,60)', 00000400
// PARM.FORT='NODECK,NOLIST,OPT(2),CHARLEN(1024)', 00000500
// PARM.LtCED='LET,LIST,XREF' 00000600
//FORT.STEPLIB DD DSN=SYS1.VSF2COMP,D1SP=SHR 00000700
//FORT.SYSLIB DD DSN=GJDNER1.ROMNET1.INCLUDES,DISP=SHR 00000800
// DD DSN=GJDNER1.ROM1.INCLUDES,DISP=SHR 00000800
//FORT.SYSIN DD DSN=GJDNER1.ROM1.FORT(ROM),DISP=SHR 00000900
//LKED.SYSLMOD DD DSN=GJDNER1.ROMNET1.LOADLIB,DISP=SHR 00001000
//LKED.SYSIN DD * 00001100
NAME ROH21R(R) . 00001200
/*
// 00000110
D-2
-------
APPENDIX E
INCLUDE FILES
E-l
-------
DIMENS.EXT and REGION.EXT contain the dimensions and definitions that are used by the ROM, and are tabulated first. The
remaining include files are tabulated in alphabetical order.
TABLE E-l. INCLUDE FILE DIMENS.EXT
Referenced by
Description
BCPRCS, BIGGAM, BLKMOD, BMPRCS, BTPRCS, CNPRCS,
DUMPHD, GTILDE, HSTEP, ICPRCS, INIRUN, LILGAM,
NEWICS, OPBCON, OPBMAT, OPBTRK, OPCONC,
OPICON, OPSTAV, OPWRCN, ORSPBC, ORSPBM, ORSPIC,
POBCON, POBTRK, POCONC, POICON, POMXBM,
POSTAV, PQ1, PQCOEF, RATED, RDBCON, RDBMAT,
RDBT, RDBTRK, RDCONC, RDHDBM, RDICON,
RDMXBM, RDSTAV, RTPHO, RTSET, RUNMGR,
WRCONC, WRSTAV
Model dimensions
Variables
NLEVS
NSPECS
NRCT
NPPRX
NRKL
NPOXSP
Type
Integer'4
Integer*4
Integer*4
Integer*4
lnteger*4
Integer*4
Value
3
35
84
5
11
3
Description
Number of levels in model grid
For Carbon Bond 4.2 Chemistry:
Number of chemical species in model
Number of reactions
Number of primary photolytic reactions
Number of levels at which photolytic rate constant is given
Number of primary oxidant species (for dimensioning
LGBMFL.EXT)
TABLE E-i INCLUDE FILE REGION.EXT
Referenced by
Description
BCPRCS, BIGGAM, BMPRCS, BTPRCS, CNPRCS, ICPRCS,
INIRUN, LILGAM, NEWICS, OPCONC, OPICON, OPSTAV,
OPWRCN, PQCOEF, RDBCON, RDBMAT, RDBT, RDBTRK,
RDCONC, RDICON, RDMXBM, RUNMGR, WRCONC,
WRSTAV
Region-dependent parameters for the ROMNET domain
Variables
GRDNAM
NROWS
NCOLS
Type
Character'8
Imeger*4
Integer*4
Value
'ROMNET
52
64
Description
Name of the region
Number of rows in the model grid
Number of columns in the model grid
E-2
-------
' TABLE E-3. INCLUDE FILE ADVSFL.EXT
Common block Referenced by
Description
ADVSFL
BIGOAM, LILGAM
This is the ADVECTION solution file. An advected row of
BIGGAM is passed to LILGAM.
Variables
Type
Dimension
Description
ADVSFLCU.k) Real*4
(NLEVS, NCOLS, NSPECS) ADVECTION solution for one row
TABLE E-4. INCLUDE FILE BCFTLE.EXT
Common block Referenced by
Description
BCFILE
BCPRCS, RDBCON
The boundary values for each time step for species concen-
trations (by layer) for all modeling domain border cells.
Variables
Type
Dimension
Description
NORTH(i,l,k) RealM
WEST(j,l,k) RealM
EAST(j,l,k) RealM
SOUTH(i,l,k) RealM
(NCOLS, NLEVS, NSPECS)
(NROWS, NLEVS, NSPECS)
(NROWS, NLEVS, NSPECS)
(NCOLS, NLEVS, NSPECS)
Boundary condition concentrations along northern edge
Boundary condition concentrations along western -edge
Boundary condition concentrations along eastern edge
Boundary condition concentrations along southern edge
TABLE E-5. INCLUDE FILE BGBCFL.EXT
Variables
BCW(j,I,k)
BCE(j,l,k)
BCN(i,l,k)
BCS(i,l,k)
Common block
BGBCFL
Type
RealM
RealM
RealM
RealM
Referenced by
BCPRCS, BIGGAM
Dimension
(NROWS, NLEVS, NSPECS)
(NROWS, NLEVS, NSPECS)
(NCOLS, NLEVS, NSPECS)
(NCOLS, NLEVS, NSPECS)
Description
Boundary conditions produced by BCPRCS for use
BIGGAM
Description
Boundary condition concentrations along western edge
Boundary condition concentrations along eastern edge
Boundary condition concentrations along northern edge
Boundary condition concentrations along southern edge
by
E-3
-------
TABLES-*. INCLUDE FILE BGBTFL.EXT
Variables
XV(l,ij)
XKU(l.ij)
Common block
BGBTFL
Type
Real'4
Real'4
Real*4
Real'4
Referenced by
BIGGAM, BTPRCS
Dimension
(NLEVS, NCOLS, NROWS)
(NLEVS, NCOLS, NROWS)
(NLEVS, NCOLS, NROWS)
(NLEVS, NCOLS, NROWS)
Description
BTRK data for BIGGAM produced by BTPRCS
Description
Backtrack locations:
Longitudinal component
Latitudinal component
Horizontal diffusivities:
Longitudinal component
Latitudinal component
TABLE E-7. INCLUDE FILE BGICCN.EXT
Common block Referenced by
Description
BGICCN BIGGAM, CNPRCS, ICPRCS,
LILGAM, NEWICS,
RUNMGR
1C CONC file: time step initial concentrations for BIGGAM
and time step concentrations finalized in LILGAM
Variables
Type
Dimension
Description
BGICCN(i,r,kj,2) Real*4
TOG1
TOG2
Integer*4
lnteger'4
(NCOLS, NLEVS, NSPECS, Time step initial concentrations for BIGGAM and time step
NROWS, 2) concentrations finalized in LILGAM
Pointer to upper half of BGICCN
Pointer to lower naif of BGICCN
E-4
-------
TABLE E-*. INCLUDE FILE BMCOEF.EXT
Variables
BBll(k)
BB22(k)
BB31(k)
BB12
BB21
BB23
BB32
BB33
BB11S
BB11SS
BB31S
BB31SS
Common block Referenced by
BMCOEF HSTEP, LILGAM
Type Dimension
Real'4 (NSPECS)
Real*4 (NSPECS)
Real'4 (NSPECS)
Real'4
Real*4
Real*4
Real'4
Real'4
Real«4
Real'4
Real*4
Real'4
Description
BMAT coefficients from LILGAM for use by HSTEP
Description
B-matrix coefficient in column i for species k for
layer I/surface 1 flux
layer 2/surface 2 flux
layer 3/surface 1 flux
B-matrix coefficient in column i for
layer I/surface 2 flux
layer 2/surface 1 flux
layer 2/surface 3 flux
layer 3/surface 2 flux
layer 3/surface 3 flux
B-matrix coefficient for subgrid scale adjustment in column i for
layer I/surface 1 flux
alternate layer I/surface 1 flux
layer 3/surface 1 flux
alternate layer 3/surface 1 flux
E-5
-------
TABLE E-9. INCLUDE FILE BMFILE.EXT
Variables
XB12(i)
XB13(i)
XB21(i)
XB23(i)
XB32(i)
XB33(i)
XBllS(i)
XBllSS(i)
XB31S(i)
XB31SS(i)
XQ03FC(i)
XSSONO(i)
XGlS(i,k)
XGlSS(i,k)
XGlFAC(i.k)
XG3S(i,k)
XG3SS(i,k)
XG3FAC(i,k)
XRHO(i,l)
XTEMP(i,l)
XWVC(i,l)
XTHETA(i)
XPSI2(i)
XZO(i)
XZl(i)
XZ2(i)
XZ3(i)
AAl(i,6,k)
Common block
BMFILE
Type
Real*4
Real»4
Real*4
Real'4
RealM
Real*4
RealM
RealM
RealM
RealM
RealM
RealM
RealM
RealM
RealM
RealM
RealM
RealM
RealM
RealM
RealM
RealM
RealM
RealM
RealM
RealM
RealM
RealM
Referenced by
BMPRCS.RDBMAT
Dimension
(NCOLS)
(NCOLS)
(NCOLS)
(NCOLS)
(NCOLS)
(NCOLS)
(NCOLS)
(NCOLS)
(NCOLS)
(NCOLS)
(NCOLS)
(NCOLS)
(NCOLS, NPOXSP)
(NCOLS, NPOXSP)
(NCOLS, NPOXSP)
(NCOLS, NPOXSP)
(NCOLS, NPOXSP)
(NCOLS, NPOXSP)
(NCOLS, NLEVS)
(NCOLS, NLEVS)
(NCOLS, NLEVS)
(NCOLS)
(NCOLS)
(NCOLS)
(NCOLS)
(NCOLS)
(NCOLS)
(NCOLS, 6, NSPECS)
Description
Data from BMAT read by RDBMAT to be passed to BMPRCS
Description
B-matrix coefficient in column i for.
layer I/surface 2 flux
layer I/surface 3 flux
layer 2/surface 1 flux
layer 2/surface 3 flux
layer 3/surface 2 flux
layer Surface 3 flux
B-matrix coefficient for subgrid scale adjustment in column /' for
layer I/surface 1 flux
alternate layer I/surface 1 flux
layer 3/surface 1 flux
alternate layer 3/surface 1 flux
Run time subgrid scale adjustment parameters in column i:
ozone factor
NO surface emissions source strength
emissions source term in layer 1 for primary oxidant species k
alternate emissions source term in layer 1 for primary oxidant species k
emissions source factor in layer 1 for primary oxidant species k
emissions source term in layer 3 for primary oxidant species k
alternate emissions source term in layer 3 for primary oxidant species k
emissions source factor in layer 3 for primary oxidant species k
Rate constants density correction factor in column i for layer /
Absolute temperature for rate constants adjustment in column /' for layer /
Water vapor concentration for rate constants adjustment in column i for
layer/
Solar zenith angle for phototytic rate constants adjustment in column i
Cloud cover correction factor for photolytic rate constants adjustment in
column/
Heights above sea level in column i (used for rate constant adjustments):
Layer 0
Layer 1
Layer 2
Layer 3
Buffer containing six species-dependent B-matrix variables:
B11, B22, B31, Gl, G2, G3
E-6
-------
TABLE E-10. INCLUDE PILE BTFTLE.EXT
Variables
XRU(i,l)
XRV(i,l)
XRKU(i.l)
XRKV(i,l)
Common block
BTFILE
Type
Real'4
Real«4
Real'4
Real'4
Referenced by
BTPRCS, RDBTRK
Dimension
(NCOLS, NLEVS)
(NCOLS, NLEVS)
(NCOLS, NLEVS)
(NCOLS, NLEVS)
Description
BTRK fDe block data from the BTRK file read by RDBTRK
to be passed to BTPRCS
Description
Backtrack locations:
Longitudinal component, column i, layer/
Latitudinal component, column i, layer/
Horizontal Diffusivities:
Longitudinal component, column i, layer /
Latitudinal component, column i, layer/
TABLE E-ll. INCLUDE FILE CHEMIN.EXT
Variables
ATS
GTS
UFRAX
BFRAX
FACTOR
DIVP
DIVQ
NCOUT
ISPEC(k)
ULIM
BLIM
FNOLIM
Common block
CHEMIN
Type
Real*4
Real'4
Real*4
Real*4
Real'4
Real»4
Real'4
Integer'4
Integer*4
Real'4
Real'4
Real*4
Referenced by Description
INIRUN, GTILDE, HSTEP, Input control for LILGAM chemistry calculations
LILGAM, OPSTAV, WRSTAV
Dimension Description
ADVECnON time step length (seconds)
VERTICAL FLUX time step length (seconds)
Upper tolerance for FRAX (FRAX = time step tolerance for change in
species value to set the chemistry time step)
Lower tolerance for FRAX
Fraction of (NO + NO^ + 03) to represent lower limit of species
concentration considered in time step computations
Multiplier (0 .LT. DIVP .LE. 1.0) to determine intermediate value of PI
to be used in CORRECTOR chemistry solution per time step
Multiplier (0 .LT. DIVQ .LE. 1.0) to determine intermediate value of 01
to be used in CORRECTOR chemistry solution per time step
Number of species (3) whose concentrations are used to determine which
species are included in chemistry time step compulations
(NSPECS) List of species indices whose concentrations are used to determine which
species are included in chemistry time step computations
Upper limit of chemistry time step length (seconds)
Lower limit of chemistry time step length (seconds)
Upper limit to control UFRAX/BFRAX chemistry computation accuracy
E-7
-------
TABLE E-12. INCLUDE FILE CHEMSW.EXT
Variables
CHEMON
Common block
CHEMSW
Type
Logical* 4
Referenced by
LILGAM, RUNMGR
Dimension
Description
Chemistry on/off flag
Description
= .TRUE. => LILGAM normal run
= .FALSE. => LILGAM chemistry calculations bypassed
TABLE E-13. INCLUDE FILE CNFILE.EXT
Variables
CNFILE(i,l,k)
Common block
CNFILE
Type
Real'4
Referenced by
CNPRCS, RDCONC,
RUNMGR, WRCONC
Dimension
(NCOLS, NLEVS, NSPECS)
Description
Final solution concentrations passe
LILGAM
Description
Solution concentrations for one row
d to CNPRCS by
TABLE E-14. INCLUDE FILE CONFACEXT
Common block Referenced by
Description
CONFAC
BIGGAM, OPSTAV,
WRSTAV
Diffusivities conversion factors for BIGGAM
Variables
Type
Dimension
Description
RDLNT2
RDLTT2
Real'4
Real«4
Longitudinal horizontal diffusivities conversion factor
Latitudinal horizontal diffusivities conversion factor
E-8
-------
TABLE E-15. INCLUDE FILE ERRG.EXT
Variables
NGTILN(k)
Common block
ERRGT
Type
Integer*4
Referenced by
GTILDE, LILGAM
Dimension
(NSPECS)
Description
GTILDE error counts
Description
Count of number of negative GTILDE's computed
TABLE E-16. INCLUDE FILE FLNAMS.EXT
Common block
Referenced by
Description
FLNAMS BLKMOD, DUMPHD, NEWICS,
OPBCON, OPCONC, OPICON,
OPSTAV, OPWRCN, RDBT, RDMXBM,
RUNMGR, WRSTAV
Internal program names used for file operating system
Variables
Type
Dimension
Description
FLNMBC
FLNMBM(m)
FLNMBT
ELNMCN
FLNMIC
FLNMSV
FLNMNI '
FLNMST
FLNMPR
NUMBMF
Character* 12
Character*12
Character'12
Character* 12
Character*12
Character* 12
Character* 12
Character*12
Character* 12
lnteger*4
(NUMBMF)
Internal name of BCON file
Internal name of BMAT subfile m
Internal name of BTRK file
Internal name of CONC Gle
Internal name of ICON file
Internal name of STATE VECTOR file
Internal name of NEW 1C file
Internal name of STOP file
Internal name of PROGRESS file
Maximum number of multiple BMAT subfiles (=6; passed
from common block LUNITS)
E-9
-------
TABLE E-17. INCLUDE FILE GTCOEF.EXT
Common block Referenced by Description
Variables
XO
YO
ZO
Al
A2
A3
Bl
B2
B3
C2
C3
GS1
GS2
GS3
COEFS
Type
Real'4
Real'4
Real*4
Real'4
Real*4
Real*4
Real'4
Real'4
Real'4
Real'4
Real'4
Real*4
Real'4
Real«4
GTILDE, LILGAM Terms in 3 X 3 system of differential equations solved by
GTILDE
Dimension Description
Initial condition (time to) for GTILDE
Initial condition (time to) for GTILDE
Initial condition (time to) for GTILDE
Coefficient in differential equations
Coefficient in differential equations
Coefficient in differential equations
Coefficient in differential equations
Coefficient in differential equations
Coefficient in differential equations
Coefficient in differential equations
Coefficient in differential equations
Inhomogeneous source term in differential equations
Inhomogeneous source term in differential equations
Inhomogeneous source term in differential equations
TABLE E-18. INCLUDE FILE HDFMTS.EXT
Common
block
HDFMTS
Variables
HDALL
HSPN
HLVN
HTEXT
Type
Character*
Character*
Character*
Character*
Referenced by Description
BLKMOD, DUMPHD, NEWICS, OPBCON, OPBMAT, Formats for headers
OPBTRK, OPCONC, OPICON, OPSTAV, OPWRCN,
WRSTAV
Dimension Description
Buffers that contain Fortran formats for echoing file header
information to the log:
488 First segment record log format
48 Species record log format
48 Level record log format
48 Text record log format
E-10
-------
TABLE E-19. INCLUDE FILE HDSTAV.EXT
Common block Referenced by
Description
CHARSV,
HEADSV
OPSTAV, POSTAV,
RUNMGR, WRSTAV
Formatted STATE VECTOR file header block
Variables
Type
Dimension
CHARSV
Description
GRDNSV Character'8
SPNMSV(k) Character'4 (NSPECS)
LVNMSV(l) Character'4 (NLEVS)
TEXTSV(20) Character'80 (20)
Grid definition name
Name of species k
Name of level/
Text description
Variables
CDATSV
CTIMSV
SDATSV
STHRSV
TSTPSV
FRSTSV
SWLNSV
SWLTSV
NELNSV
NELTSV
DLONSV
DLATSV
NCOLSV
NROWSV
NLEVSV
NSPCSV
ICNTSV
Type Dimension
Integer*4
Integer*4
Integer*4
Integer*4
Inieger*4
Integer*4
Real*4
Real'4
Real'4
Real'4
Real'4
Real'4
Integer'4
Integer'4
Integer'4
Integer'4
lnleger'4
HEADSV
Description
Creation date (MMDDYY) of STATE VECTOR file
Creation time (HHMMSS) of STATE VECTOR file
Start date (YYDDD) of scenario
Start hour (00 to 23) of scenario
Size of step (seconds) used in simulation
First time step (seconds past start time of model scenario) in simulation
Longitude (degrees) of southwest corner of grid
Latitude (degrees) of southwest corner of grid
Longitude (degrees) of northeast corner of grid
Latitude (degrees) of northeast comer of grid
Longitudinal increment cell (degrees)
Latitudinal cell increment (degrees)
Number of columns in grid (= NCOLS)
Number of rows in grid (= NROWS)
Number of levels (= NLEVS)
Number of species in simulation (= NSPECS)
Number of text records in header
E-ll
-------
TABLE E-20. INCLUDE FILE HEADBCEXT
Common block Referenced by Description
Variables
GRDNBC
SPNMBC(k)
LVNMBC(l)
TEXTBC(20)
Variables
CDATBC
CTIMBC
SDATBC
STHRBC
TSTPBC
FRSTBC
SWLNBC
SWLTBC
NELNBC
NELTBC
DLONBC
DLATBC
NCOLBC
NRQWBC
NLEVBC
NSPCBC
ICNTBC
CHARBC,
HEADBC
Type
Character'8
Character*4
Character*4
Character'80
Type
Integer*4
lnteger*4
lnteger*4
Integer*4
Integer'4
Integer*4
Real'4
Real*4
Real'4
Real'4
Real'4
Real«4
Integer'4
Integer*4
Integer'4
Integer'4
Integer'4
BCPRCS, OPBCON, BCON file header block
OPWRCN, ORSPBC,
POBCON, RDBCON,
RUNMGR
CHARBC
Dimension Description
Grid definition name
(NSPECS) Name of species k
(NLEVS) Name of level/
(20) Text description
HEADBC
Dimension Description
Creation date (MMDDYY) of BC file
Creation time (HHMMSS) of BC file
Start date (YYDDD) of scenario on BC file
Stan hour (00 to 23) of scenario on BC file
Size of time step (seconds) used in BC file
First time step (seconds past start time of
BCON scenario) on BCON file
Longitude (degrees) of southwest corner of grid
Latitude (degrees) of southwest corner of grid
Longitude (degrees) of northeast corner of grid
Latitude (degrees) of northeast corner of grid
Longitudinal increment cell (degrees)
Latitudinal cell increment (degrees)
Number of columns in grid
Number of rows in grid
Number of levels
Number of species in BCON file
Number of text records in header
E-12
-------
TABLE E-21. INCLUDE FILE HEADBM.EXT
Common block
Referenced by
Description
CHARBM, HEADBM,
BMSPRD
BMPRCS, DUMPHD, OPBMAT, OPWRCN, ORSPBM,
POMXBM, RDBMAT, RDHDBM, RDMXBM, RUNMGR
BMAT file header block
Variables Type
Value •
Description
NMFBM
NVRBM1
Integer*4
IntegerM
NVRBM2 Integer*4 6
4 Number of MIF files used in generating BMAT file
18+3xNLEVS+6xNPOXSP Number of part 1 BMATvariables, specifically: B12, B13, B21, B23, B32, B33,
BUS, BUSS, B31S, B31SS, QO3FAC, SSONO, TTHETA, PPSI2, ZZO, ZZ1,
ZZ2, 723, RRHO(NLEVS)'s, TTEMP(NLEVS)'s, WWVCXNLEVSys; and
G1S, G1SS, G1FAC, G3S, G3SS, G3FAC, each dimensioned by NPOXSP
Number of species array BMAT variables (part 2), specifically: Bl 1, B22, B31,
Gl, G2, G3
Variables
Type
Dimension
CHARBM
Description
GRDNBM Character'8
SPNMBM(k) Character^ (NSPECS)
LVNMBM(l) Character'4 (NLEVS)
TEXTBM(20) Character'80 (20)
MFNMBM(m) Character*12 (NMFBM)
Grid definition name
Name of species*
Name of level /
Text description
File name of MIF file m used to produce the BMAT file
Variables Type Dimension
ISUBFL IntegerM
NSUBFL Integer*4
FRSTSF lnteger*4
LSSTSF lnleger*4
CDATBM lnteger*4
CTIMBM Integer*4
SDATBM IntegerM
STHRBM IntegerM
TSTPBM IntegerM
FRSTBM Inieger*4
SWLNBM RealM
SWLTBM Real'4
NELNBM Real»4
NELTBM Real'4
DLONBM Real'4
DLATBM Real«4
NCOLBM IntegerM
NROWBM IntegerM
NLEVBM Integer*4
NSPCBM IntegerM
NMIFBM IntegerM
CDMFBM(m) IntegerM (NMFBM)
CTMFBM(m) IntegerM (NMFBM)
UDMFBM(m) IntegerM (NMFBM)
UTMFBM(m) IntegerM (NMFBM)
ICNTBM IntegerM
BMINDX(k) IntegerM (NSPECS)
Variables Type Dimension
BMSPRD IntegerM
HEADBM
Description
Ordinal subfile number
Total number of subfiles of BMAT file
First time step on subfile ISUBFL
Last time step on subfile ISUBFL
Creation date (MMDDYY) of BMAT file
Creation time (HHMMSS) of BMAT file
Start date (YYDDD) of scenario
Start hour (00 to 23) of scenario
Time step size (seconds) for simulation
Pint time step (seconds past start time of scenario) on BMAT file
Longitude (degrees) of southwest corner of grid
Latitude (degrees) of southwest corner of grid
Longitude (degrees) of northeast corner of grid
Latitude (degrees) of northeast corner of grid
Longitudinal cell increment (degrees)
Latitudinal cell increment (degrees)
Number of columns in grid
Number of rows in grid
Number of levels
Number of species on BMAT file
Number of MIF files used to generate BMAT
Creation date of MIF file m used to produce the BMAT file
Creation date of MIF file m used to produce the BMAT file
Last update date of MIF file m used to produce the BMAT file
Last update time of MIF file m used to produce the BMAT file
Number of text records on header
Index array for species expansion from reduced BMAT set to full MODEL set
BMSPRD
Description
Number of species in BMAT reduced set representing NSPCBM species.
E-13
-------
TABLE E-22. INCLUDE FILE HEADBT.EXT
Common block Referenced by Description
CHARBT,
HEADBT
Variables Type
NMFBT Integer'4
NVARBT Integer'4
Variables Type
GRDNBT Character'8
TEXTBT(20) Character'80
MFNMBT(m) Character* 12
Variables Type
CDATBT Integer*4
CTIMBT Integer'4
SDATBT . Integer'4
STHRBT Integer'4
TSTPBT Integer'4
FRSTBT Integer'4
SWLNBT Real*4
SWLTBT Real*4
NELNBT Real*4
NELTBT Real '4
DLONBT Real '4
DLATBT Real '4
NCOLBT Integer'4
NROWBT Integer'4
NMIFBT Integer'4
CDMFBT(m) lnteger'4
CTMFBT(m) Integer'4
UDMFBT(m) Integer'4
UTMFBT(m) Integer'4
ICNTBT Integer'4
BTPRCS, OPBTRK, BTRK fiJe header block
OPWRCN, POBTRK, RDBT,
RDBTRK, RUNMGR
Value Description
6 Number of MIF files used in generating BTRK file
4 Number of LEVEL-DEPENDENT BTRK variables, specifically: UU's, W's,
KKHUXKKHVs
CHARBT
Dimension Description
Grid definition name
(20) Text description
(NMFBT) File name of MIF file m used to produce the BTRK file
HEADBT
Dimension Description
Creation date (MMDDYY) of BTRK file
Creation time (HHMMSS) of BTRK file
Start date (YYDDD) of scenario
Stan hour (00 to 23) of scenario
Time step size (seconds) for simulation
First time step (seconds past start time of scenario) on BTRK file
Longitude (degrees) of southwest comer of grid
Latitude (degrees) of southwest corner of grid
Longitude (degrees) of northeast corner of grid
Latitude (degrees) of northeast corner of grid
Longitudinal cell increment (degrees)
Latitudinal cell increment (degrees)
Number of columns in grid
Number of rows in grid
Number of MIF files used to generate BTRK file
(NMFBT) Creation date of MIF file m used to produce the BTRK file
(NMFBT) Creation date of MIF file m used to produce the BTRK file
(NMFBT) Last update date of MIF file m used to produce the BTRK file
(NMFBT) Last update time of MIF file m used to produce the BTRK file
Number of text records in header
E-14
-------
TABLE E-23. INCLUDE FILE HEADCN.EXT
Common block Referenced by Description
Variables
GRDNCN
SPNMCN(k)
LVNMCN(l)
TEXTCN(20)
Variables
Variables
CDATCN
CTIMCN
SDATCN
STHRCN
TSTPCN
FRSTCN .
SWLNCN
SWLTCN
NELNCN
NELTCN
DLONCN
DLATCN
NCOLCN
NROWCN
NLEVCN
NSPCCN
ICNTCN
CDBMCN
CTBMCN
CDBTCN
CTBTCN
CDBCCN
CTBCCN
CDICCN
CTICCN
CHARCN,
HEADCN
Type
Character'8
Character*4
CharacterM
Character'80
Type
Type
Integer'4
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
RealM
Real*4
RealM
RealM
RealM
RealM
IntegerM
IntegerM
IntegerM
IntegerM
IntegerM
IntegerM
IntegerM
IntegerM
IntegerM
IntegerM
IntegerM
IntegerM
IntegerM
CNPRCS, NEWICS, CONC file header block
OPCONC, OPWRCN,
POCONC, RDCONC,
RUNMGR
CHARCN
Dimension Description
Grid definition name
(NSPECS) Name of species k
(NLEVS) Name of level/
(20) Text description
HEADCN
Dimension Description
Dimension Description
Creation date (MMDDYY) of CONC file
Creation time (HHMMSS) of CONC file
Stan date (YYDDD) of scenario
Start hour (00 to 23) of scenario
Size of step (seconds) used in simulation
First time step (seconds past start time of model scenario) on CONC file
Longitude (degrees) of southwest corner of grid
Latitude (degrees) of southwest corner of grid
Longitude (degrees) of northeast corner of grid
Latitude (degrees) of northeast corner of grid
Longitudinal increment cell (degrees)
Latitudinal cell increment (degrees)
Number of columns in grid (= NCOLS)
Number of rows in grid (•= NROWS)
Number of levels (= NLEVS)
Number species in CONC file (= NSPECS)
Number of text records in header
Creation date (MMDDYY) of BMAT file used in simulation
Creation time (HHMMSS) of BMAT file used in simulation
Creation date (MMDDYY) of BTRK file used in simulation
Creation time (HHMMSS) of BTRK file used in simulation
Creation date (MMDDYY) of BCON file used in simulation
Creation time (HHMMSS) of BCON file used in simulation
Creation date (MMDDYY) of ICON file used in simulation
Creation time (HHMMSS) of ICON file used in simulation
E-15
-------
TABLE E-24. INCLUDE FILE HEADIC.EXT
Common block Referenced by Description
Variables
GRDNIC
SPNMIC(k)
LVNMIC(l)
TEXTIC(20)
Variables
CDATIC
CTIM1C
SDATIC
STHRIC
TSTPIC
FRSTIC
SWLNIC
SWLTIC
NELNIC
KELTIC
DLONIC
DLATIC
NCOLIC
NROW1C
NLEVIC
NSPCIC
IDUM(8)
1CNTIC
CHARIC,
HEADIC
Type
Character'8
Character*4
Character*4
Character*80
Type
Integer'4
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Real'4
Real*4
Real'4
Real*4
Real*4
Real*4
Integer*4
Integer'4
Integer'4
Integer'4
Integer'4
Integer'4
ICPRCS, NEWICS, ICON file header block
OPICON, OPWRCN,
ORSPIC.POICON,
RDICON.RUNMGR
CHARIC
Dimension Description
Grid definition name
(NSPECS) Name of species*
(NLEVS) Name of level/
(20) Text description
HEADIC
Dimension Description
Creation date (MMDDYY) of ICON file
Creation time (HHMMSS) of ICON file
Start date (YYDDD) of scenario
Stan hour (00 to 23) of scenario
Size of step (seconds) used in simulation
First time step (seconds past start time of model scenario) on ICON file
Longitude (degrees) of southwest corner of grid
Latitude (degrees) of southwest corner of grid
Longitude (degrees) of northeast corner of grid
Latitude (degrees) of northeast corner of grid
Longitudinal increment cell (degrees)
Latitudinal cell increment (degrees)
Number of columns in GRID (= NCOLS)
Number of rows in GRID (= NROWS)
Number of levels («= NLEVS)
Number of species in ICON file ( - NSPECS)
(8) Padding
Number of text records
E-16
-------
TABLE E-25. INCLUDE FILE HEADIN.EXT
Common block Referenced by Description
Variables
GRDNIN
SPNMIN(k)
LVNMIN(l)
TEXTIN(20)
Variables
CDATIN
CTIMIN
SDATIN
STHRIN
TSTPIN
FRSTIN
SWLNIN
SWLTIN
NELN1N
NELTIN
DLONIN
DLATIN
NCOLIN
NROWIN
NLEVIN
NSPCIN
ICNTIN
CHARIN,
HEADIN
Type
Character'8
Character'4
Character*4
Character*80
Type
Integer*4
lnteger*4
Integer*4
Inleger*4
Integer*4
lnteger*4
Real*4
Real'4
Real'4
Real*4
Real*4
Real'4
Integer*4
Integer*4
Integer'4
lnteger*4
Integer*4
BCPRCS, BIGGAM, Input header information, used to check file headers on
BLKMOD, BMPRCS, BMAT, ICON, and BCON flies, and to create Die header on
BTPRCS, CNPRCS, CONCfile.
DUMPHD, HSTEP, ICPRCS,
INIRUN, LILGAM, NEWICS,
OPSTAV, OPWRCN,
ORSPBC, ORSPBM, ORSPIC,
PQ1, RDBCON, RDBMAT,
RDBTRK, RDCONC,
RDICON, RUNMGR,
WRSTAV
CHARIN
Dimension Description
Grid definition name
(NSPECS) Name of species *
(NLEVS) Name of level/
(20) Text description
HEADIN
Dimension ' Description
Creation date of simulation (YYMMDD)
Creation time of simulation (HHMMSS)
5 -digit JULIAN start date (YYDDD) for model scenario
Start hour (00 TO 23) for model scenario
Step size (seconds) for model scenario
First time step (seconds past start time of model scenario) in simulation
Longitude (degrees) of southwest corner of grid
Latitude (degrees) of southwest corner of grid
Longitude (degrees) of northeast corner of grid
Latitude (degrees) of northeast corner of grid
Longitudinal increment cell (degrees)
Latitudinal cell increment (degrees)
Number of columns in grid (= NCOLS)
Number of rows in grid (= NROWS)
Number of levels (= NLEVS)
Number of species in model
Number of text records in header
E-17
-------
TABLE E-26. INCLUDE FILE HSTEPS.EXT
Common block Referenced by
Variables
HSTEPS(k)
HSTPO3(2)
IND(k)
IND03(2)
HSTEPS
Type
Real'4
Real'4
Integer*4
Integer*4
HSTEP, LILGAM
Dimension
(NSPECS)
(2)
(NSPECS)
(2)
Description
Integration steps (H) for GTILDE computed by HSTEP
Description
Integration step
Integration step for OZONE, two possible conditions
Indicator of success in root calculation from HSTEP
= Invalid root
= 2 =» invalid root
Indicator of success in root calculation from HSTEP
TABLE E-27. INCLUDE FILE ICFTLE.EXT
Common block Referenced by
Description
ICFILE
ICPRCS, NEWICS, RDICON Initial conditions file read by 1CPRCS
Variables
Type
Dimension
Description
ICFILE(i,l,k) Real'4
(NCOLS, NLEVS, NSPECS) Initial concentrations for one row
E-18
-------
TABLE E-2& INCLUDE FILE LGBMFL.EXT
Variables
B12(i)
B13(i)
B21(i)
B23(i)
B32(i)
B33(i)
Bll(i,k)
B22(i,k)
B31(i,k)
Gl(i,k)
G2(i,k)
G3(i,k)
BllS(i)
BllSS(i)
B31S(i)
B31SS(i)
QO3FAC(i)
SSONO(i)
GlS(i,k)
GlSS(i,k)
GlFAC(i,k)
G3S(i,k)
G3SS(i,k)
G3FAC(i,k)
RHO(i,l)
TEMP(i,l)
WVC(i,l)
THETA(i)
PSI2(i)
2LEV(i,l)
Common block
LGBMFL
Type
Real'4
Real«4
Real'4
Real*4
Real«4
Real«4
Real'4
Real*4
Real*4
Real'4
Real*4
Real»4
Real'4
Real'4
Real'4
Real«4
Real'4
Real'4
Real«4
Real'4
Real'4
Real'4
Real*4
Real'4
Real'4
Real'4
Real'4
Real*4
Real'4
Real'4
Referenced by
BMPRCS.LILGAM
Dimension
(NCOLS)
(NCOLS)
(NCOLS)
(NCOLS)
(NCOLS)
(NCOLS)
(NCOLS, NSPECS)
(NCOLS, NSPECS)
(NCOLS, NSPECS)
(NCOLS, NSPECS)
(NCOLS, NSPECS)
(NCOLS, NSPECS)
(NCOLS)
(NCOLS)
(NCOLS)
(NCOLS)
(NCOLS)
(NCOLS)
(NCOLS, NPOXSP)
(NCOLS, NPOXSP)
(NCOLS, NPOXSP)
(NCOLS, NPOXSP)
(NCOLS, NPOXSP)
(NCOLS, NPOXSP)
(NCOLS, NLEVS)
(NCOLS, NLEVS)
(NCOLS, NLEVS)
(NCOLS)
(NCOLS)
(NCOLS, NLEVS + 1)
Description
Intermediate BMAT file read by LILGAM written by BMPRCS
Description
B-matrix coefficient in column /' for
layer I/surface 2 flux
layer I/surface 3 flux
layer 2/surface 1 flux
layer 2/surface 3 flux
layer 3/surface 2 flux
layer 3/surface 3 flux
B-matrix coefficient in column i for species k for
layer I/surface 1 flux
layer 2/surface 2 flux
layer 3/surface 1 flux
Emissions source term in column /' for species k for
layer 1
layer 2
layer 3
B-matrix coefficient for subgrid scale adjustment in column / for
layer I/surface 1 flux
alternate layer I/surface 1 flux
layer 3/surface 1 flux
alternate layer 3/surface 1 flux
Run time subgrid scale adjustment parameters in column i:
ozone factor
NO surface emissions source strength
emissions source term in layer 1 for primary oxidant species k
alternate emissions source term in layer 1 for primary oxidant species k
emissions source factor in layer 1 for primary oxidant species k
emissions source term in layer 3 for primary oxidant species k
alternate emissions source term in layer 3 for primary oxidant species k
emissions source factor in layer 3 for primary oxidant species k
Rate constants density correction factor in column i for layer /
Absolute temperature for rate constants adjustment in column i for layer /
Water vapor concentration for rate constants adjustment in column i for
layer/
Solar zenith angle for photoh/tic rate constants adjustment in column i
Cloud cover correction factor for photolytic rate constants adjustment in
column /
Heights above sea level in column ;' (used for rate constant adjustments)
for layer/
E-19
-------
TABLE E-29. INCLUDE FILE ULGSP.EXT
Common block Referenced by
Description
LILGSP HSTEP, LILGAM, OPSTAV, Contains special SPECIES control for LILGAM
PQ1, WRSTAV
Variables
INBIG3(k)
NOHIT
NO2HIT
03HIT
OLEHIT
PARHIT
TRCHIT
NONHIT
Type Dimension
Logical *4 (NSPECS)
Integer*4
Integer*4
Integer*4
Integer* 4
Integer* 4
Integer*4
Integer*4
Description
.TRUE, for SPECIES NO,
Index for NO
Index for NO2
Index for Oj
Index for OLE
Index for PAR
Index for METH
Index for NONR
NO2, O3, otherwise .FALSE.
TABLE E-30. INCLUDE FILE LUMTS.EXT
Common block Referenced by
Description
LUNITS BLKMOD, DUMPHD, ICPRCS, NEWICS, OPBCON,
OPCONC, OPICON, OPSTAV, OPWRCN, POBCON,
POBTRK, POCONC, POICON, POMXBM, POSTAV,
RDBCON, RDBT, RDCONC, RDHDBM, RDICON,
RDMXBM, RDSTAV, RUNMGR, WRCONC,
WRSTAV
UNIT numbers of input and output files
Variables Type
Value
Description
NUMBMF Integer*4
Maximum number of multiple BMAT subfiles
Variables
UNITBC
UNITBM
UNITBT
UNITCN
UNITIC
UNITSV
UNITNI
UNITST
UNITPR
Type Dimension
Integer*4
Integer* 4
Integer*4
lnteger*4
Integer*4
Inleger*4
Integer'4
lnteger*4
Integer*4
LUNITS
Description
Logical unit number of BCON file
Logical unit number of BMAT subfile
Logical unit number of BTRK file
Logical unit number of CONC file
Logical unit number of ICON file
Logical unit number of STATE VECTOR file
Logical unit number of NEW ICON file
Logical unit number of STOP file
Logical unit number of PROGRESS file
E-20
-------
TABLE E-31. INCLUDE FILE LVNAME.EXT
Common block Referenced by
Description
LVNAME
BLKMOD, INIRUN
Level names for model processors
Variables
Type
Dimension
Description
LVNAME(l) Character'4
(NLEVS)
Name of Ah level in model order
TABLE E-32. INCLUDE FILE NDXPCEXT
Common block Referenced by
Description
NDXSPC BCPRCS.BMPRCS,
DUMPHD, ICPRCS,
NEWICS, OPSTAV,
ORSPBM, ORSPBC, ORSPIC,
WRSTAV
Index lists for SPECIES ordering of input files
Variables
NXSPIC(k)
NXSPBC(k)
NXSPBM(k)
Type
Integer*4
Integer*4
lnteger*4
Dimension
(NSPECS)
(NSPECS)
(NSPECS)
Description
Position of ytth specie in model on ICON file
Position of ton specie in model on BCON file
Position of Alb specie in model on BMAT file
TABLE E-33. INCLUDE FILE NROOTS.EXT
Variables
NCMPLX
NPOS
NDBL
NTRPL
Common block Referenced by
NROOTS HSTEP.LILGAM
Type Dimension
lnteger*4
Integer*4
Inleger*4
Integer* 4
Description
HSTEP roots counts
Description
Number of complex roots of characteristic polynomial computed
Number of positive roots of characteristic polynomial computed
Number of double roots of characteristic polynomial computed
Number of triple roots of characteristic polynomial computed
E-21
-------
TABLE E-34. INCLUDE FILE RKLEVS.EXT
Variables
RKL(1)
NRKL
Common block
RKLEVS
Type
Real'4
Integer*4
Referenced by
BLKMOD, RTPHO
Dimension
(NRKL)
11
Description
Description
Level / for which photolytic rate constant is given
Number of levels in table (from DIMENS.EXT)
TABLE E-35. INCLUDE FILE ROWSCT.EXT
Common block Referenced by
Description
ROWSCT BMPRCS, RDBMAT, RDBTRK,
RDCONC, RDSTAV, RUNMGR,
WRSTAV
Local row counters for I/O processes
Variables
Type
Dimension
Description
BMPSRW
RDBMRW
RDBTRW
RDCNRW
IntegerM
Integer*4
Integer*4
Integer*4
Row counter for BMPRCS
Row counter for RDBMAT
Row counter for RDBTRK
Row counter for RDCONC
E-22
-------
TABLE E-36. INCLUDE FILE RTCONS.EXT
Common block Referenced by
Description
RTCONS BLKMOD, LILGAM, PQ1,
PQCOEF, RATED, RTPHO,
RTSET
Chemistry reaction and rate constants information
Variables
CGA(k,l)
GPRfkJ)
ADTVF(k)
INVADT(k)
RKl(m,l)
RKKl(m)
RKTADV(m)
PQM(m)
CSPRl(m,n,l)
Type
Real'4
Real'4
Real'4
Real'4
Real'4
Real'4
Real'4
Real'4
Real'4
Dimension
(NSPECS, NLEVS)
(NSPECS, NLEVS)
(NSPECS)
(NSPECS)
(NRCT, NLEVS)
(NRCT)
(NRCT)
(NRCT)
(5, 0:90, NRKL)
Description
Advection component of concentration for species k in level /
Chemistry component of concentration for species k in level /
Product of advection component and vertical flux component of
concentration for species k
1 / ADTVF(NSPECS)
Rate constant for reaction m in level /
Rate constant for reaction m
Product of rate constant for reaction m and advection component
and vertical flux of concentration
Product of RKTADV and chemistry component of concentration
for reaction m
Clear sky photolytic reaction rate constant for reaction number
m = 1 - 5, solar angles D = 0 - 90, rate constant level
7=1- NRKL; for Carbon Bond 4.2:
m = 1 corresponds to reaction number 1
NRCT
NRKL
84
11
m = 2 corresponds to reaction number 8
m = 3 corresponds to reaction number 33
m = 4 corresponds to reaction number 34
m = 5 corresponds to reaction number 40
Number of reactions in chemistry solver (from DIMENS.EXT)
Number of levels at which photolytic rate constant is given (from
DIMENS.EXT)
TABLE E-37. INCLUDE FILE RTSHBCEXT
Variables
DATBC
TIMBC
ELPBC
STPBC
Common block Referenced by
RTSHBC RDBCON
Type Dimension
Real'4
Real'4
Real'4
Real'4
Description
BCON file real time step header block
Description
S-digit JULIAN date (YYDDD) on BCON file
6-digit time (HHMMSS) on BCON file
Elapsed time (seconds) since start time for BCON scenario
Step number on BCON file
E-23
-------
TABLE E-38. INCLUDE FILE RTSHBM.EXT
Variables
DATBM
TIMBM
ELPBM
STPBM
Common block Referenced by
RTSHBM RDBMAT
Type Dimension
Real«4
RealM
Real*4
RealM
Description
BMAT file real time step header block
Description
5-digit JULIAN date (YYDDD) on BMAT file
6-digit time (HNMMSS) on BMAT file
Elapsed time (seconds) since start time for BMAT scenario
Step number on BMAT file
TABLE E-39. INCLUDE FILE RTSHBT.EXT
Common block
Referenced by
Description
RTSHBT
RDBTRK
BTRK file real time step header block
Variables
Type
Dimension
Description
DATBT
TIMBT
ELPBT
STPBT
RealM
RealM
RealM
RealM
5-digit JULIAN date (YYDDD) on BTRK file
6-digit time (HHMMSS) on BTRK file
Elapsed time (seconds) since start time for BTRK scenario
Step number on BTRK file
E-24
-------
TABLE E-40. INCLUDE FILE RTSHCN.EXT
Common block Referenced by
Description
RTSHCN
RDCONC, WRCONC
CONC file real time step header block
Variables
Type
Value
Description
IDLT
Integer*4
NLEVSxNCOLS-4
Number of words for padding time step header for file fixed
record length
Variables
RTSHCN
Type
Dimension
Description
DATCN
TIMCN
ELPCN
STPCN
RDUMCN(m)
RealM (IDLT)
RealM
RealM
RealM
RealM
5-digit JULIAN date (YYDDD) on CONC Die
6-digit time (HHMMSS) on CONC file
Elapsed time (seconds) since start time for CONC scenario
Step number on CONC file
Dummy array for padding time step header out to record length
for fixed record length Die
TABLE E-41. INCLUDE FILE RTSHIC.EXT
Variables
IDLT
Variables
DATIC
TIMIC
ELPIC
STPIC
RDUMIC(m)
Common block
RTSHIC
Type
IntegerM
Type
RealM
RealM
RealM
RealM
RealM
Referenced by Description
NEWICS, RDICON CONC file real time step header block
Value Description
NLEVS xNCOLS -4 Number of words for padding time step header for file
record length
RTSHIC
Dimension Description
(IDLT) 5-digit JULIAN date (YYDDD) on ICON file
6-digit time (HHMMSS) on ICON file
Elapsed time (seconds) since start time for ICON scenario
Step number on ICON file
fixed
Dummy array for padding time step header out to record length
for fixed record length file
E-25
-------
TABLE R4Z INCLUDE FILE RUNTMS.EXT
Variables
RUNCPU
RUNCLK
OLDCPU
OLDCLK
KDATE
KTIME
Common block Referenced by
RUNTMS BIGGAM,
TIMER
Type • Dimension
Real*4
Real'4
Real'4
Real'4
Integer*4
Integer*4
Description
RUNMGR, Run timing statistics
Description
Elapsed CPU time for current run
Elapsed clock time for current run
Elapsed CPU time of prior step
Elapsed dock time of prior step
Current dock date (during execution)
Current dock time (during execution)
TABLE E-43. INCLUDE FILE SPNAME.EXT
Common block Referenced by
Description
SPNAME BLKMOD, DUMPHD,
INIRUN, LHX3AM, ORSPBC,
ORSPBM, ORSPIC, PQ1
Spedes names list in model order
Variables .
SPNAME(k)
Type
Characler*4
Dimension
(NSPECS)
Description
Name of ktit species in model order
TABLE E-44. INCLUDE FILE STOPFL.EXT
Common block Referenced by
Description
STOPFG
BLKMOD, RUNMGR
Stop flag
Variables
Type
Dimension
Description
STOPFG
Character*4
Character string:
.EQ. "STOP' «> model run baits at end of current time step
.NE. 'STOP* =•• model run continues
E-26
-------
TABLE E-45. INCLUDE FILE SUBID.EXT
Common block Referenced by
Description
SUBID ADATE, ASORT, BCPRCS, BIGGAM,
BLKMOD, BMPRCS, BTPRCS, CELLM,
CLOCK1, CLOCK2, CNPRCS, CPUTIM,
DATTIM, DUMPHD, FSKIP1, GTILDE,
HSTEP, ICPRCS, INDEX1, INIRUN, IOCL,
JFILE2, JFILE5, JFILE6, JULIAN, JUNIT,
LILGAM, NEWICS, OPBCON, OPBMAT,
OPBTRK, OPCONC, OPICON, OPSTAV,
OPWRCN, ORSPBC, ORSPBM, ORSPIC,
POBCON, POBTRK, POCONC, POICON,
POMXBM, POSTAV, PQ1, PQCOEF,
PRGSMY, RATED, RDBCON, RDBMAT,
RDBT, RDBTRK, RDCHAR, RDCONC,
RDFILE, RDHDBM, RDICON, RDMXBM,
RDSTAV, RTPHO, RTSET, RUNMGR,
TIMER, WRCHAR, WRCONC, WRFILE,
WRSTAV
Internal module description listed at run time by PRGSMY
Variables
SUBDES
SUBDTE
SUBNAM
SUBVER
Type Dimension
Character'80
Character* 12
Character'8
Character'8
Description
Description of module
Date of last update of module
Module name
Module version
E-27
-------
TABLE E-46. INCLUDE FILE TEXTPT.EXT
Common block Referenced by
Description
TEXTPT BCPRCS, BIGGAM, BLKMOD, BMPRCS, BTPRCS,
CNPRCS, ICPRCS, LILGAM, RDBCON, RDBMAT,
RDBTRK, RDCONC, RDICON, RDSTAV, RUNMGR,
WRCONC, WRSTAV
Text pointers for all processes
Variables
BIGMPT
LILGPT
BCPSPT
BMPSPT
BTPSPT
CNPSPT
ICPSPT
RDBCPT
RDBMPT
RDBTPT
WRCNPT
RD1CPT
RDCNPT
WRSVPT
Type Dimension
Integer*4
Integer*4
Integer* 4
Integer'4
Integer*4
Integer*4
Integer* 4
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Integer'4
Description
Text pointer for BIGGAM process
Text pointer for LILGAM process
Text pointer for BCON process (BCPRCS)
Text pointer for BMAT process (BMPRCS)
Text pointer for BTRK process (BTPRCS)
Text pointer for CONC process (CNPRCS)
Text pointer for ICON process (ICPRCS)
Text pointer for RDBCON process
Text pointer for RDBMAT process
Text pointer for RDBTRK process
Text pointer for WRCONC process
Text pointer for RDICON process
Text pointer for RDCONC process
Text pointer for WRITE STATE VECTORS process (WRSTAV)
TABLE E-47. INCLUDE FILE TILDE.EXT
Variables
GTILDl(k)
GTILD2(k)
GTILD3(k)
GTI(k,l)
Common block
TILDE
Type
Real*4
Real*4
Real*4
Real*4
Referenced by
GTILDE, LILGAM, PQ1
Dimension
(NSPECS)
(NSPECS)
(NSPECS)
(NSPECS, NLEVS)
Description
GTILDE values
Description
GTILDE values for LEVEL ONE
GTILDE values for LEVEL TWO
GTILDE values for LEVEL THREE
GTILDE values for ONE ROW
NOTE: DIMENSION GTI(NSPECS, NLEVS)
EQUIVALENCE (GTI, GTILD1)
E-28
-------
TABLE E-4*. INCLUDE FILE TSHDBC.EXT
Common block Referenced by
Description
TSHDBC
POBCON, RDBCON
BCON file time step header block
Variables
Type
Dimension
Description
IDATBC
ITIMBC
IELPBC
ISTPBC
lnteger*4
Integer*4
Integer*4
Integer*4
5-digit JULIAN date (YYDDD) on BCON file
6-digit time (HHMMSS) on BCON file
Elapsed time (seconds) since start time for BCON scenario
Step number on BCON file
TABLE E-49. INCLUDE FILE TSHDBM.EXT
Variables
IDATBM
ITIMBM
IELPBM
ISTPBM
Common block Referenced by
TSHDBM POMXBM, RDBMAT
Type Dimension
Integer'4
Integer*4
Integer*4
lnteger*4
Description
BMAT file time step header block
Description
5-digit JULIAN date (YYDDD) on BMAT file
6-digit time (HHMMSS) on BMAT file
Elapsed time (seconds) since start time for BMAT scenario
Step number on BMAT file
TABLE E-SO. INCLUDE FILE TSHDBT.EXT
Common block Referenced by
Description
TSHDBT
POBTRK, RDBTRK
BTRK file time step header block
Variables
Type
Dimension
Description
IDATBT
ITIMBT
IELPBT
ISTPBT
lnteger'4
Integer*4
Integer*4
Integer*4
5-digit JULIAN date (YYDDD) on BTRK file
^digit time (HHMMSS) on BTRK file
Elapsed time (seconds) since start time for BTRK scenario
Step number on BTRK file
E-29
-------
' TABLE E-51. INCLUDE FILE TSHDCN.EXT
Common block Referenced by
Description
TSHDCN CNPRCS.NEW1CS,
POCONC, RDCONC,
WRCONC
CONC file time step header block
Variables
1DATCN
ITIMCN
IELPCN
ISTPCN
Type Dimension
Integer*4
Integer*4
lnteger*4
Integer*4
Description
5-digit JULIAN date (YYDDD) on CONC file
6-digit time (HHMMSS) on CONC file
Elapsed time (seconds) since start time for model scenario
Step number on CONC file
TABLE E-51 INCLUDE FILE TSHDIC.EXT
Variables
IDATIC
ITIMIC
IELPIC
ISTPIC
Common block
TSHDIC
Type
Integer*4
Integer*4
Integer*4
Integer*4
Referenced by
CNPRCS, ICPRCS, NEWICS,
POICON, RDICON
Dimension
Description
ICON file time step header block
Description
5-digit JULIAN date (YYDDD) on ICON file
6-digit time (HHMMSS) on ICON file
Elapsed time (seconds) since start time for model scenario
Step number on ICON file
TABLE E-53. INCLUDE FILE TSHDMD.EXT
Variables
1DATMD
ITIMMD
IELPMD
ISTPMD
Common block
TSHDIN
Type
Integer*4
Integer'4
Integer*4
Integer'4
Referenced by
BIGGAM, RUNMGR
Dimension
Description
Model time step header block
Description
5-digit JULIAN date (YYDDD) of current model step
6-digit time (HHMMSS) of current model step
Elapsed time (seconds) since start time for model scenario
Step number of current model step
E-30
-------
TABLE E-54. INCLUDE FILE TSHDSV.EXT
Variables
IDATSV
ITIMSV
IELPSV
ISTPSV
Common block
TSHDSV
Type
Integer*4
Integer* 4
Integer*4
Integer*4
Referenced by Description
POSTAV, RDSTAV, STATE VECTOR ffle time step header block
RUNMGR, WRSTAV
Dimension Description
5-digit JULIAN date (YYDDD) on STATE VECTOR file
6-digit time (HHMMSS) on STATE VECTOR file
Elapsed time (seconds) since start time for model scenario
Step number on STATE VECTOR file
TABLE E-SS. INCLUDE FILE TSTEP&EXT
Variables
MDDATE
MDTIME
MDELAP
MDSTEP
BCDATE
BCTIME
BCELAP
BCSTEP
BMDATE
BMTIME
BMELAP
BMSTEP
BTDATE
BTTIME
BTELAP
BTSTEP
CNDATE
CNTIME
CNELAP
CNSTEP
ICDATE
ICT1ME
ICELAP
ICSTEP
Common block
TSTEPS
Type
Integer*4
lnteger*4
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
lnteger'4
lnteger*4
Integer*4
lnteger*4
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
Integer*4
lnteger*4
lnteger'4
Inleger*4
Integer'4
Integer*4
Integer'4
Referenced by Description
BIGGAM, BCPRCS, BMPRCS, BTPRCS, Time step information for processes
ICPRCS, CNPRCS, INIRUN, RDBCON,
RDBMAT, RDBTRK, RDCONC, RDICON,
RDSTAV, RUNMGR, WRSTAV
Dimension Description
Date of current step in RUNMGR
Tune of current step in RUNMGR
Elapsed time from start of scenario (seconds) in RUNMGR
Current step number in RUNMGR
Date of current step in BCON process
Tune of current step in BCON process
Elapsed time from start of scenario (seconds) in BCON process
Current step number in BCON process
Date of current step in BMAT process
Time of current step in BMAT process
Elapsed time from start of scenario (seconds) in BMAT process
Current step number in BMAT process
Date of current step in BTRK process
Time of current step in BTRK process
Elapsed time from start of scenario (seconds) in BTRK process
Current step number in BTRK process
Date of current step in CONC process
Tune of current step in CONC process
Elapsed time from start of scenario (seconds) in CONC process
Current step number in CONC process
Date of current step in ICON process
Time of current step in ICON process
Elapsed time from start of scenario (seconds) in ICON process
Current step number in ICON process
E-31
-------
TABLE E-56. INCLUDE FILE UNITIO.EXT
Common block Referenced by
Description
UNITIO ADATE, BCPRCS, BIGGAM, BLKMOD,
BMPRCS, BTPRCS, CELLM, CPUTIM,
DATTIM, DUMPHD, FSKffl, ICPRCS,
INIRUN, JFILE2, JFILE5, JFILE6,
JUNIT, LILGAM, NEWICS, OPBCON,
OPBMAT, OPBTRK, OPCONC,
OPICON, OPSTAV, OPWRCN, ORSPBC,
ORSPBM, ORSPIC, POBCON, POBTRK,
POCONC, POICON, POMXBM,
POSTAV, PQ1, PRGSMY, RDBCON,
RDBMAT, RDBT, RDBTRK, RDCONC,
RDHDBM, RDICON, RDMXBM,
RDSTAV, RUNMGR, WRCONC,
WRSTAV
Logical unit numbers of system-defined default FORTRAN
standard input and output files
Variables
Type
Dimension
Description
LUNIN
LUNOUT
Integer*4
Integer*4
Logical unit number of standard input file
Logical unit number of standard output file
TABLE E-57. INCLUDE FILE ZADVSL.EXT
Common block Referenced by
Description
ZERADV
BIGGAM, BLKMOD
Positive "zero" values to replace negative advection solutions
calculated by BIGGAM
Variables
Type
Dimension
Description
ZADVSL(k) Real'4
(NSPECS)
Replacement value for species k
E-32
-------
APPENDIX F
CORE MODEL ERROR CHECKING
F-l
-------
SUBROUTINE ADATE
ERROR CHECK #1:
In subroutine ADATE, the current date (month, day, year) is obtained by calling an internal system routine.
The month, day, and year are then written to an internal formatted buffer (CDATE). The I/O status value of
the formatted write operation is stored in the variable IOST. If IOST is not equal to zero, then a write error
has occurred. A message is written to the log indicating that an error has occurred. The error message con-
tains the subroutine name, the month, the day, the year, and the I/O status value. The program exits by a call
to the system subroutine EXIT.
CODE FOR ERROR CHECK #1:
c
C get date from system
CALL IDATE (IMON, IDAY, IYEAR)
WRITE (CDATE, 1001, IOSTAT=IOST) IMON, IDAY, IYEAR
1001 FORMAT(3I2.2)
IF (IOST .NE. 0> THEN
WRITE (LUNOUT, 2001) IMON, IDAY, IYEAR, IOST
2001 FORMAT(/ 5X, 'XXX ERROR ENCODING DATE IN ADATE1
& / 5X, 'IMON = ', 14, 2X, 'IDAY = ', 14,
& 2X, 'IYEAR = ', 14, 2X, 'I/O STATUS = ', 14)
CALL EXIT
END IF
SUBROUTINE ASORT
NONE
SUBROUTINE BCPRCS
ERROR CHECK #1:
The BCON time step header is read by calling subroutine RDBCON. The parameter IOST is passed to
RDBCON. Upon return from subroutine RDBCON, IOST contains the I/O status of the read of the BCON time
step header. IOST is tested; if IOST is less than zero, an end-of-file marker was reached, and control is passed
back to the calling subroutine BIGGAM. If IOST is not equal to zero, then an error occurred on the read
operation. A message is written to the log indicating that an error has occurred. The error message contains
the subroutine name, the time and date that were read from the BCON time step header, and the I/O status
value. The program exits by a call to the system subroutine EXIT.
F-2
-------
CODE FOR ERROR CHECK #1:
c
C read BC T.S.H.
CALL RDBCON (IOST)
C
IF (IOST .LT. 0) RETURN
IF (IOST .NE. 0) THEN
WRITE (LUNOUT, 2001) BCDATE, BCTIHE, IOST
2001 FORHATC/ 5X, 'XXX READ ERROR ON T.S.H. FROM BCON FILE1
& / 5X, 'BCDATE = ', 16.6, 2X, 'BCTIME = '. 16.6
& / 5X, 'I/O STATUS = ', 18 /)
CALL EXIT
END IF
SUBROUTINE BIGOAM
ERROR CHECK #1:
Calls are made to the following subroutines:
ICPRCS to read the ICON time step header
BCPRCS to read the BCON time step header
BTPRCS to read the backtrack time step header
ICPRCS to read the initial condition concentrations data
LILGAM to write the advection time step header
BCPRCS to read the boundary conditions data
BTPRCS to read the backtrack locations and diffusivities data
A distinctive parameter is passed into each of these subroutines - the I/O status value of a read or write oper-
ation performed in the called subroutine. This parameter is tested. If it is not equal to zero, then either an
end-of-file marker was reached or an error was encountered in the subroutine that was called. If the
parameter is not equal to zero, then the variable IEOFBG in BIGGAM is set to a specific number but not zero.
IEOFBG is set to a number determined by the routine in which the end of file was reached. IEOFBG is tested
in subroutine BIGGAM. If it is not equal to zero, an error message is written to the log indicating that an error
has occurred. The error message contains the subroutine name and the value of IEOFBG. Control is
returned back to the calling program RUNMGR.
F-3
-------
CODE FOR ERROR CHECK #1:
c
C read BGIC T.S.H.
CALL ICPRCS UEOFIC)
IF (IEOFIC .NE. 0) IEOFBG = 1
C
C read BGBC T.S.H.
CALL BCPRCS (IEOFBC)
IF (IEOFBC .NE. 0) IEOFBG = 2
C
C read BGBT T.S.H.
CALL BTPRCS (IEOFBT)
IF (IEOFBT .NE. 0) IEOFBG = 3
C
C read initial condition concentrations
CALL ICPRCS (IEOFIC)
IF (IEOFIC .NE. 0) IEOFBG = 4
C
C write ADVS T.S.H. (done here so that ICON'S can be copied to CONC)
CALL LILGAM (IEOFLG)
IF (IEOFLG .NE. 0) IEOFBG = 7
C
C read boundary conditions
CALL BCPRCS (IEOFBC)
IF (IEOFBC .NE. 0) IEOFBG = 5
C
C read backtrack locations and diffusivities
CALL BTPRCS (IEOFBT)
IF (IEOFBT .NE. 0) IEOFBG = 6
C
IF (IEOFBG .NE. 0) THEN
WRITE (LUNOUT, 2001) IEOFBG
2001 FORMAT(// 5X, 'XXX BIGGAM NOT GETTING DATA1
& / 2X, 'IEOFBG: ', U /)
RETURN
END IF
ERROR CHECK #2:
Calls are made to the following subroutines:
BCPRCS to read the BCON time step header
BTPRCS to read the backtrack time step header
ICPRCS to read the ICON file data
LILGAM to write the advection time step header
BCPRCS to read the boundary conditions data
BTPRCS to read the backtrack locations and diffusivities data
A distinctive parameter is passed into each of these subroutines - the I/O status value of a read or write oper-
ation performed in the called subroutine. This parameter is tested. If it is not equal to zero, then either an
end-of-file marker was reached or an error was encountered in the subroutine that was called. If the
parameter is not equal to zero, then the variable IEOFBG in BIGGAM is set to a specific number - but not to
zero. IEOFBG is set to a number determined by the routine in which the end of file was reached. IEOFBG is
F-4
-------
tested in subroutine BIGGAM. If it is not equal to zero, an error message is written to the log indicating that an
error has occurred. The error message contains the subroutine name and the value of IEOFBG. Control is
returned back to the calling program RUNMGR.
CODE FOR ERROR CHECK #2:
c
C read BGBC T.S.H.
CALL BCPRCS (IEOFBC)
IF (IEOFBC .NE. 0) IEOFBG = 2
C
C read BGBT T.S.H.
CALL BTPRCS (IEOFBT)
IF (IEOFBT .NE. 0) IEOFBG = 3
C
C read BGIC file
CALL ICPRCS (IEOFIC)
IF (IEOFIC .NE. 0) IEOFBG - 1
C
C write ADVS T.S.H.
CALL LILGAM (IEOFLG)
IF (IEOFLG .NE. 0) IEOFBG = 7
C
C read boundary conditions
CALL BCPRCS (IEOFBC)
IF (IEOFBC .NE. 0) IEOFBG = 5
C
C read backtrack locations and diffusivities
CALL BTPRCS (IEOFBT)
IF (IEOFBT .NE. 0) IEOFBG = 6
C
IF (IEOFBG .NE. 0) THEN
WRITE (LUNOUT. 2001) IEOFBG
RETURN
END IF
SUBROUTINE BMPRCS
ERROR CHECK #1:
The BMAT time step header is read by calling subroutine RDBMAT. The parameter IOST is passed to
RDBMAT. Upon return from subroutine RDBMAT, IOST contains the read of the BMAT time step header.
IOST is tested; if IOST is less than zero, then an end-of-file marker was reached and control is passed back to
the calling routine BIGGAM. If IOST is not equal to zero then an error occurred on the read operation. A
message is written to the log indicating that an error has occurred. The error message contains the subroutine
name, the time and date that were read from the BMAT time step header, and the I/O status value. The
program exits by a call to the system subroutine EXIT.
F-5
-------
CODE FOR ERROR CHECK #1:
c
C read BH T.S.H.
CALL RDBMAT (IOST)
C
IF (IOST .LT. 0) RETURN
IF (IOST .NE. 0) THEN
WRITE (LUNOUT, 2001) BCOATE, BCTIME, IOST
2001 FORMAT(/ 5X, 'XXX READ ERROR ON T.S.H. FROM BNAT FILE1
& / 5X, 'BMDATE = ', 16.6, 2X, 'BMTIME = ', 16.6
& / 5X, 'I/O STATUS = ', 18 /)
CALL EXIT
END IF
SUBROUTINE BTPRCS
ERROR CHECK #1:
The BTRK time step header is read by calling subroutine RDBTRK. The parameter IOST is passed to RDBTRK.
Upon return from subroutine RDBTRK, IOST contains the I/O status of the read operation of the BTRK time
step header. IOST is tested; if IOST is less than zero, then an end-of-file marker was reached and control is
passed back to the calling routine BIGGAM. If IOST is not equal to zero, then an error occurred on the read
operation. A message is written to the log indicating that an error has occurred. The error message contains
the subroutine name, the time and date that were read from the BTRK time step header, and the I/O status
value. The program exits by a call to the system subroutine EXIT.
CODE FOR ERROR CHECK #1:
c
C read BT T.S.H.
CALL RDBTRK (IOST)
C
IF (IOST .LT. 0) RETURN.
IF (IOST .NE. 0) THEN
WRITE (LUNOUT, 2001) BTDATE, BTTIME, IOST
2001 FORMAT(/ 5X, 'XXX READ ERROR ON T.S.H. FROM BTRK FILE'
& / 5X, 'BTDATE = ', 16.6, 2X, 'BTTIME = '. 16.6
& / 5X, 'I/O STATUS = ', 18 /)
CALL EXIT
END IF
SUBROUTINE CELLM
ERROR CHECK #1:
The parameter GRDNM is passed to subroutine CELLM. GRDNM is the grid name from the standard input
file (user supplied input). GRDNM is tested against the three defined grid names (NEROS1, SEROS1,
F-6
-------
ROMNET1). If GRDNM is not equal to one of these, then a message is written to the log indicating that an
error has occurred. The error message contains the subroutine name and 'GRDNM'. The program exits by a
call to the system subroutine EXIT.
CODE FOR ERROR CHECK #1:
c
DIMENSION GRDNMS(NREG)
C
DATA GRDNMS / 'NEROS1 ', 'SEROS1 ', 'ROMNET1 ' /
C
C assume grid region is not defined
C
C assume grid region is not defined
DO 101 IRC = 1, NREG
IF (GRDNM .EQ. GRDNMS(IRG)) GO TO 201
101 CONTINUE
C
C grid region is not defined
WRITEUUNOUT, 2001) GRDNM
2001 FORMATC// 5X, '%%% ERROR ABORT IN CELLM JOX'
& 3X, 'GRID REGION ', AS, < IS NOT DEFINED1)
CALL EXIT
SUBROUTINE CLOCKl
NONE
SUBROUTINE CLOCK2
NONE
SUBROUTINE CNPRCS
NONE
SUBROUTINE CPUTIM
NONE
SUBROUTINE DATTJM
F-7
-------
A call to subroutine ADATE is made. The parameter DATE is passed to the subroutine ADATE. ADATE
returns the current date (MMDDYY) from the system in the character string DATE. The parameter TIME is
passed into the subroutine ADATE. ADATE returns the current time (HHMMSS) from the system in the
character string TIME.
ERROR CHECK #1:
From the character string DATE the current month, day, and year are extracted by a formatted read opera-
tion. The I/O status value of the read operation is stored in the variable IOST. IOST is tested, and if not
equal to zero, then a read error has occurred. A message is written to the log indicating that an error has
occurred. The error message contains the subroutine name, the month, day, year, and I/O status value. The
program exits by a call to the system subroutine EXIT.
CODE FOR ERROR CHECK #1:
c
C date
C
CALL ADATE (DATE, TIME. TSTR)
READ (DATE, 1001) IMON, IDAY, IYEAR
1001 FORMAT(3I2)
IF (IOST .NE. 0) THEN
URITE(LUNOUT, 2001) IMON, IDAY, IYEAR, IOST
2001 FORMAT(/ 5X, 'XXX ERROR ENCODING DATE IN DATTIM '/
i / 5X, 'IMON =', 18, 2X, 'IDAY =', 18.
& 2X, 'IYEAR =', 18, 3X, 'I/O STATUS = ', 12 /)
CALL EXIT
END IF
ERROR CHECK #2:
From the character string TIME, the current hour, minute, and second are extracted by a formatted read
operation. The I/O status value of the read operation is stored in the variable IOST. IOST is tested, and if not
equal to zero, then a read error has occurred. A message is written to the log indicating that an error has
occurred. The error message contains the subroutine name, the hour, the minute, the second, and I/O status
value. The program exits by a call to the system subroutine EXIT.
F-8
-------
CODE FOR ERROR CHECK #2:
c
C time
C
READ (TIME, 1001, IOSTAT = IOST) IHR, IMIN, ISEC
IF (IOST .NE. 0) THEN
URITEaUNOUT, 2003) IHR, IMIN, ISEC, IOST
2003 FORMAT(/ 5X, 'XXX ERROR ENCODING TIME IN DATTIM '/
& / 5X, 'IHR =', 18, 2X, 'IMIN =', 18,
& 2X, 'ISEC =', 18, 3X, 'I/O STATUS = ', 12 /)
CALL EXIT
END IF
SUBROUTINE DUMPHD
ERROR CHECK #1:
The BMAT file requires a large amount of disk space and may have been written to several smaller subfiles on
different disk packs (since each pack may not individually have had sufficient space to contain the entire file).
These subfiles would then be assigned separate logical names in the job's run stream. In subroutine RDMXBM,
the subfile number is tested. If the test fails, then the B-matrix subfiles are out of order. A call to subroutine
DUMPHD is made. The subfile number (ISUB) is the parameter passed to DUMPHD. Upon return from
DUMPHD, the program exits by issuing a call to the system subroutine EXIT. In subroutine DUMPHD, a message
is written to the log indicating that an error has occurred. The error message states that an error occurred in
the BMAT sequence, and also states the subfile number, the unit number, the logical name of the file, and the
actual name of the file.
CODE FOR ERROR CHECK #1:
c
From subroutine RDMXBM
C verify subfile
IF (ISUBFL .NE. ISUB .OR. NSUBFL .NE. NSUB) THEN
CALL DUMPHD (ISUB)
CALL EXIT
END IF
In subroutine DUMPHD:
c
c
INQUIRE (FILE = FLNMBM(ISUB). NAME = EQNAME)
WRITEUUNOUT, 2001) ISUB, UNITBM, FLNMBM(ISUB), EQNAME
2001 FORMATC/ 5X, 'XXX SUBFILE INCORRECT IN BMAT SEQUENCE:'
£ / ' SUBFILE NUMBER'. 13, 3X, 'JOB ABORTED1
& / 5X, 'UNITBM = ', 12, 'FNAME = ', A8
& / 3X, 'EONAME = ', A64)
ERROR CHECK #2:
F-9
-------
An iteration over the number of species is made. For each iteration, the following takes place:
. A call to the function INDEXI occurs. SPNAME(ISPC), and SPNMBM are passed into INDEXI.
(SPNMBM is an array containing the species name in the BMAT list, obtained from the BMAT
header. SPNAME is an array containing the list of species names for the model.)
The position of each of the species names of the model is searched for in the list of species names
from the BMAT list of species names. This positional value is assigned to the variable SPCNUM
in the subroutine DUMPHD.
If SPCNUM equals zero, then the species name was not found in the BMAT list of species names.
A message is written to the log indicating that an error has occurred.
The error message contains the subroutine name, the current species name that is being considered (SPNA-
ME(ISPC)), the models species name (SPNMIN(ISPC)), and the value of SPCNUM (which should be zero).
The program exits by a call to the system subroutine EXIT.
CODE FOR ERROR CHECK #2:
c
DO 101 ISPC = 1, NSPCIN
C
C check that model species name is in BHATRIX species list
C and get model list species' name position in BHATRIX list
SPCNUM = INDEX1 (SPNAME(ISPC), NSPCBM, SPNMBM)
IF (SPCNUM -EQ. 0) THEN
WRITEdUNOUT, 2009) SPNAME (ISPC)
2009 FORMAT(/ 5X, 'XXX ERROR IN DUMPHD1
& / 5X, 'species', 2X, A4, 2X,
& 'is not present on BMATRIX file--job aborted')
WRITEdUNOUT. 2011) SPNMIN(ISPC), SPNAME(ISPC), SPCNUM
2011 FORMAT(6X, A4, 8X, A4, 7X, K.2)
CALL EXIT
END IF
101 CONTINUE
SUBROUTINE FSKIP1
This subroutine positions a file by skipping forward or backward.
ERROR CHECK #1:
The amount of records to skip is computed and stored in the variable IFORWD. An iteration of IFORWD
takes place in order to skip the appropriate amount of records. For each iteration the following takes place:
If the file is formatted, then a formatted read takes place.
• If the file is unformatted, then an unformatted read takes place. The record to be skipped is read
into a dummy variable.
F-10
-------
• The I/O status value of the read operation is stored in the variable IOST. If IOST is not equal to
zero, then a read error has occurred. A message is written to the log indicating that an error has
occurred.
The error message contains the subroutine name, the unit number of the file and the I/O status value. The
program exits by a call to the system subroutine EXIT.
CODE FOR ERROR CHECK #1:
DO 111 ISKP = 1, 1FORUD
IF (FMTD) READUDEV, 1001, IOSTAT = IOST) OUM
1001 FORNAT(A1)
IF (.NOT. FNTD) READ(IDEV, IOSTAT = IOST) IDUM
IF (IOST .NE. 0} THEN
URITEaUNOUT, 2001) IOEV, IOST
2001 FORMAT(/ 5X, 'XXX I/O ERROR IN FSKIP1
& / 5X, 'IOEV = ', 13, 2X, 'I/O STATUS = ', 13)
CALL EXIT
END IF
111 CONTINUE
ERROR CHECK #2:
Records are skipped by moving forward through the file until an end-of-file marker is reached. For each
record skipped the following takes place:
If the file is formatted, then a formatted read takes place.
If the file is unformatted, then an unformatted read takes place. The record to be skipped is read
into a dummy variable.
• The I/O status value of the read operation is stored in the variable IOST.
IOST is tested. If IOST is less than zero, then an end-of-file marker is reached. The file is
rewound. Records are skipped and control is returned to the calling subroutine. If IOST is not
equal to zero, then a read error has occurred. A message is written to the log indicating that an
error has occurred.
The error message contains the subroutine name, the unit number of the file, and the I/O status value. The
program exits by a call to the system subroutine EXIT.
F-ll
-------
CODE FOR ERROR CHECK #2:
c
201 CONTINUE
IF (FMTD) READdDEV, 1001, IOSTAT = IOST) DUM
IF (.NOT. FNTD) READdDEV, IOSTAT = IOST) I DUM
IF (IOST .LT. 0) THEN
REWIND IDEV
DO 211 ISKP ^ 1, NSKIPD + IREC
IF (FMTD) READdDEV, 1001, IOSTAT = IOST) DUM
IF (.NOT. FMTD) READdDEV, IOSTAT = IOST)
211 CONTINUE
RETURN
END IF
IF (IOST .NE. 0) THEN
URITEUUNOUT, 2001) IDEV, IOST
CALL EXIT
END IF
NSKIPD = NSKIPD + 1
CO TO 201
ERROR CHECK #3:
Records are skipped by moving forward through the file until an end-of-file marker is reached. For each
record skipped the following takes place:
If the file is formatted, then a formatted read takes place.
If the file is unformatted, then an unformatted read takes place. The record to be skipped is read
into a dummy variable.
The I/O status value of the read operation is stored in the variable IOST.
IOST is tested. If IOST is less than zero, then an end-of-file marker is reached. Control is returned
to the calling subroutine. If IOST is not equal to zero, then a read error has occurred. A message
is written to the log indicating that an error has occurred.
The error message contains the subroutine name, the unit number of the file, and the I/O status value. The
program exits by a call to the system subroutine EXIT.
CODE FOR ERROR CHECK #3:
c
301 CONTINUE
DO 311 ISKP - 1, N2SKIP
IF (FMTD) READdDEV, 1001, IOSTAT = IOST) DUM
IF (.NOT. FMTD} READdDEV, IOSTAT = IOST) I DUM
IF (IOST .LT. 0) RETURN
IF (IOST .NE. 0) THEN
WRITE(LUNOUT, 2001) IDEV, IOST
CALL EXIT
END IF
NSKIPD = NSKIPD + 1
311 CONTINUE
RETURN
NSKIPD = NSKIPD + 1
GO TO 301
F-12
-------
SUBROUTINE GTILDE
NONE
SUBROUTINE HSTEP
NONE
SUBROUTINE ICPRCS
ERROR CHECK #1:
The ICON time step header is read by calling subroutine RDICON. The parameter IOST is passed to RDICON.
Upon return from subroutine RDICON, IOST contains the I/O status of the read operation of the ICON time
step header. IOST is tested; if IOST is less than zero, then an end-of-file marker was reached and control is
passed back to the calling routine BIGGAM. If IOST is not equal to zero, then an error occurred on the read
operation. A message is written to the log indicating that an error has occurred. The error message contains
the subroutine name, the time and date that were read from the ICON time step header, and the I/O status
value. The program exits by a call to the system subroutine EXIT.
CODE FOR ERROR CHECK #1:
c
C initialize file process elapsed time and step
CALL CLOCK1 (ICDATE, ICTIME, SOATIC, STHRIC, TSTPIC,
& ICELAP, ICSTEP)
C
IF (IDATIC .NE. ICOATE .OR. ITIMIC .NE .ICTIME) THEN
WRITE(LUNOUT, 2001) ICOATE, ICTIME, IDATIC, ITIMIC
2001 FORMAT(/ 5X, 'XXX DATES/TIMES DO NOT MATCH IN ICPRCS1
& / 5X, 'EXPECTED DATE AND TIME = ', 16.6, 5X, 16.6
& / 5X, 'DATE AND TIME READ = '. 16.6, 5X, 16.6)
CALL EXIT
EKO IF
ERROR CHECK #2:
A call to subroutine CLOSIC is made to close the ICON file. The parameter IOST contains the I/O status value
of the close operation of the ICON file. IOST is tested. If IOST is not equal to zero, then an error has
occurred on the close operation. A message is written to the log indicating that an error has occurred. The
error message contains the subroutine name and the I/O status. The program exits by a call to the system
subroutine EXIT.
F-13
-------
CODE FOR ERROR CHECK #2:
c
C close the ICON file
CALL CLOSIC (IOST)
WRITE (LUNOUT, 1001) UNITIC
1001 FORMAT(/ 5X, 'ICON file closed on unit ', 12 /)
C
IF (IOST .NE. 0) THEN
WRITE(LUNOUT, 2003) IOST
2003 FORMAT(/ 5X, 'XXX ICON FILE CLOSE ERROR IN ICPRCS OR',
& ' RDICON/CLOSIC: I/O STATUS = ', 110)
CALL EXIT
END IF
FUNCTION INDEX1
NONE
SUBROUTINE INIRUN
ERROR CHECK.#1:
From the standard input file (user supplied input), the grid name is read. This is compared with the accept-'
able grid name obtained from the include file REGION.EXT. If these names do not match, then a message is
written to the log indicating that an error has occurred. The error message contains the expected grid name
and the inputted grid name. The program exits by a call to the system subroutine EXIT.
CODE FOR ERROR CHECK #1:
c
C read grid definition record
C
READ (LUNIN, *) GRONIN
WRITEUUNOUT. 1003} GRDNIN
1003 FORMAT(/ 10X, 'grid name: ', AS)
C
IF (GRDNIN .NE. GRDNAN) THEN
WRITE(LUNOUT, 2001) GRDNAM, GRDNIN
2001 FORMAT(/ 5X, 'XXX EXPECTED REGION NAME, ', AS,
&1X, 'DOES NOT MATCH INPUT REGION NAME, ', AS /)
CALL EXIT
END IF
FUNCTION IOCL
NONE
FUNCTION JFILE2
F-14
-------
ERROR CHECK #1:
This function opens a file and attaches a unit number to it. The I/O status of the open operation is stored in
the variable IOST. If IOST is not equal to zero, then an error has occurred when opening the file. IOST is
then passed as a parameter to function IOCL so that the clause field in the I/O status word can be extracted.
Once extracted, IOST is passed back to JFILEZ A message is written to the log indicating that an error has
occurred. The error message contains the subroutine name, the logical name of the file, the actual name of
the file, the unit number, and the I/O status value. The program exits by a call to the system subroutine EXIT.
CODE FOR ERROR CHECK #1:
IF (RDONLY) THEN
OPEN (UNIT = IDEV,
& IOSTAT = IOST,
& FILE = FNAHE,
& STATUS = 'OLD',
& ACCESS = 'SEQUENTIAL',
& FORM = FORMAT,
& READONLY)
STAT = 'OLD1
RD = 'YES1
ELSE
OPEN (UNIT = IDEV,
& IOSTAT = IOST,
& FILE = FNAHE,
& STATUS = 'UNKNOWN1,
& ACCESS = 'SEQUENTIAL',
& FORM = FORMAT)
STAT = 'UNKNOWN1
RD = 'NO1
END IF
C
INQUIRE (FILE = FNAME, NAME = EQNAME)
C
IF (IOST .NE. 0) THEN
IOST - lOCL(IOST)
WRITE (LUNOUT, 2001) FNAME, EQNAME, IDEV, IOST
2001 FORMAT(/ 5X,'%XX ERROR ABORT IN JFILE2'
& / 5X, 'UNABLE TO OPEN SEQUENTIAL FILE ', A12
£ / 5X, 'EQNAME•= ', A64
& / 5X, 'IDEV = ', 12, 2X, 'I/O STATUS = ', 12)
CALL EXIT
END IF
FUNCTION JFILB5
ERROR CHECK #1:
This function opens a file and attaches a unit number to it. The I/O status of the open operation is stored in
the variable IOST. If IOST is not equal to zero, then an error has occurred when opening the file. IOST is
then passed as a parameter to function IOCL so that the clause field in the I/O status word can be extracted.
F-15
-------
Once extracted IOST is passed back to JFILES. A message is written to the log indicating that an error has
occurred. The error message contains the subroutine name, the logical name of the file, the actual name of
the file, the unit number, and the I/O status value. The program exits by a call to the system subroutine EXIT.
CODE FOR ERROR CHECK #1:
IF
&
&
&
&
&
&
&
&
(RDONLY) THEN
OPEN
STAT
RD
(UNIT =
IOSTAT =
FILE =
STATUS =
ACCESS -
FORM =
RECORDTYPE =
RECL
READONLY)
= 'OLD1
= 'YES'
IDEV,
IOST,
FNAME,
'OLD1,
'SEQUENTIAL',
FORMAT,
'FIXED',
RECLEN,
ELSE
&
&
&
&
&
&
&
OPEN
STAT
RD
(UNIT
IOSTAT =
FILE =
STATUS =
ACCESS =
FORM
RECORDTYPE =
RECL
= 'UNKNOWN'
= 'NO1
IDEV,
IOST,
FNAME,
'UNKNOWN'.
'SEQUENTIAL',
FORMAT,
'FIXED',
RECLEN)
END IF
C
INQUIRE
C
IF
(FILE = FNAME
, NAME = EQNAME)
(IOST .NE. 0) THEN
IOST
= lOCL(IOST)
WRITE (LUNOUT, 2001) FNAME. EQNAME, IDEV, IOST
2001
FORMAT (/ 5X, 'XXX ERROR ABORT IN JFILES'
&
&
&
/ 5X, 'UNABLE TO OPEN SEQUENTIAL FILE ', A12
/ ' EQNAME = ', A64
CALL
/ 5X, 'IDEV
EXIT
= ', 12, 2X, 'I/O STATUS = ', 12)
END IF
FUNCTION JFILE6
ERROR CHECK #1:
This function opens a file and attaches a unit number to it. The I/O status of the open operation is stored in
the variable IOST. If IOST is not equal to zero, then an error has occurred when opening the file. IOST is
then passed as a parameter to function IOCL so that the clause field in the I/O status word can be extracted.
Once extracted IOST is passed back to JFILE6. A message is written to the log indicating that an error has
occurred. The error message contains the subroutine name, the logical name of the file, the actual name of
the file, the unit number, and the I/O status value. The program exits by a call to the system subroutine EXIT.
F-16
-------
CODE FOR ERROR CHECK #1:
IF (RDONLY) THEN
OPEN
&
&
&
&
&
&
&
&
&
STAT
RD
ELSE
OPEN
&
&
&
&
&
&
&
&
STAT
RD
END IF
(UNIT = IDEV,
IOSTAT
FILE = FNAME,
STATUS
ACCESS
FORM = FORMAT,
CARRIAGECONTROL
RECORDTYPE
RECL = RECLEN,
READONLY)
= 'OLD1
B 'YES1
(UNIT = IDEV,
IOSTAT
FILE = FNAME.
STATUS
ACCESS
FORM = FORMAT,
CARRIAGECONTROL
RECORDTYPE
RECL = RECLEN)
= 'UNKNOWN1
= 'NO'
= IOST,
= 'OLD',
= 'SEQUENTIAL',
B 'FORTRAN',
= 'VARIABLE',
= IOST,
= 'UNKNOWN1,
= 'SEQUENTIAL1,
= 'FORTRAN1,
= 'VARIABLE',
INQUIRE (FILE = FNAME, NAME = EQNAME)
C
IF (IOST .NE. 0) THEN
IOST = lOCL(IOST)
WRITE (LUNOUT, 2001) FNAME, EQNAME, IDEV, IOST
2001 FORMAT(/ 5X,'%XX ERROR ABORT IN JFILE61
& / 5X, 'UNABLE TO OPEN SEQUENTIAL FILE ', A12
& / ' EQNAME = ', A64
• & / 5X, 'IDEV = ', 12, 2X, 'I/O STATUS = ', 14)
CALL EXIT
END IF
FUNCTION JULIAN
NONE
FUNCTION JUNIT
ERROR CHECK #1:
This function returns the next available FORTRAN logical unit number. The variable IUN is a counter
variable initially set to 1. Each time it is incremented the following takes place:
IUN is tested to see if the unit number corresponding to it is available. If it is not available, then
IUN is incremented and tested again.
F-17
-------
IUN is also compared with MAXUN. MAXUN is equal to 53 and corresponds to the maximum
allowable unit number available on the system. If IUN is greater than MAXUN, then an error has
occurred.
An error message is written to the log indicating that an error has occurred. The message contains the func-
tion name, the value of IUN, and a table listing the numbers 1 through 53 along with a T or a F, indicating
whether that particular unit number is available or not. The program exits by a call to the system subroutine
EXIT.
CODE FOR ERROR CHECK #1:
IF (IUN .GT. MAXUN) GO TO 301
C
301 CONTINUE
WRITE(LUNOUT, 2001)
2001 FORMAK/// IX, 'XXX ERROR ABORT IN JUNIT XXX' /
& IX, 'NO MORE UNIT NUMBERS AVAILABLE FOR I/O')
WRITECLUNOUT, 2003) (IUN, AVAIL(IUN), IUN = 1, MAXUN)
2003 FORMAT(1X, 'AVAILABLE UNIT NUMBERS ARE: ' /
& 3(1X, 20(12, '-', L1, 2X) /) )
CALL EXIT
SUBROUTINE LILGAM
ERROR CHECK #1:
A call to subroutine PQI is made. In subroutine PQI, IOST is set equal to 1 if values for GPR or ADTVF are
less than or equal to zero. Otherwise IOST is not set. (GPR is the chemistry component of concentration for
a particular species in a specific level; ADTVF is the product of the advection component and the vertical flux
component of concentration for a particular species.) Upon return from subroutine PQI, IOST is tested. If
IOST is not equal to zero, then an error occurred in subroutine PQI. A message is written to the log indicating
that an error occurred. The error message contains the subroutine name, the row, column, layer number,
value for GTTIM, and value for CHETIM. The program exits by a call to the system subroutine EXIT.
CODE FOR ERROR CHECK #1:
c
CALL PQ1 (P1, 01, LEV, IOST)
IF (IOST .NE. 0) THEN
WRITE(LUNOUT, 2001) IROW, ICOL. LEV. GTTIM, CHETIM
2001 FORMAT(/ 5X, 'XXX P,Q PREDICTOR RETURN >'
& / 5X, 'IROW =', 14, 2X, 'ICOL = ', 14,
42X, 'LEV =', 14,
& / 5X, 'GTTIM =', 1PE15.5, 2X, 'CHETIM =', 1PE15.5 /)
CALL EXIT
END IF
ERROR CHECK #2:
F-18
-------
Same as error check #1
CODE FOR ERROR CHECK #2:
c
CALL PQ1 (PP1, 001, LEV, IOST)
IF (IOST .NE. 0) THEN
WRITECLUNOUT, 2003) IROW, ICOL, LEV, GTTIM, CHETIM
2003 FORMAT(/ 5X, 'XXX P,0 CORRECTOR RETURN >'
& / 5X, 'IROW =', 14, 2X, 'ICOL =', 14,
&2X, 'LEV =', 14,
& / 5X, 'GTTIM =', 1PE15.5, 2X, 'CHETIM =', 1PE15.5 /)
CALL EXIT
END IF
SUBROUTINE NEWICS
ERROR CHECK #1:
The NEWICON file is opened unformatted. The first segment is written to an internal buffer using a for-
matted write statement. The I/O status of the write operation is stored in variable IOST. IOST is tested; if
IOST does not equal zero, a write error has occurred. A message is written to the log indicating that an error
has occurred. The error message contains the subroutine name, the segment number and I/O status value.
The program exits by a call to the system subroutine EXIT.
CODE FOR ERROR CHECK #1:
c
C write 1st segment
WRITECSEG1BF, 1001, IOSTAT = IOST)
& COATIC, CTIHIC, SDATIC, STHRIC, TSTPIC, FRSTIC,
& GRDNCN,
& SULNCN, SULTCN, NELNCN, NELTCN,
& OLONCN, DLATCN,
& NCOLCN, NROWCN, NLEVCN, NSPCCN,
& CDBNCN, CTBMCN,
& CDBTCN, CTBTCN,
& CDBCCN, CTBCCN,
& COICCN, CTICCN,
& ICNTCN
1001 FORMAT(6I8.8, A8, 4F8.3, 2F8.5, 414.4, 818.8, 14.4)
IF (IOST .NE. 0) THEN
URITE(LUNOUT, 2001) ' SEGMENT 1 ', IOST
2001 FORMAT(/ 5X, 'XXX INTERNAL WRITE ERROR IN NEWICS: >, A12,
& / 5X, 'I/O STATUS = ', 14)
CALL EXIT
END IF
ERROR CHECK #2:
The unit number, buffer containing the first segment record, and IOST are passed to subroutine WRCHAR,
which writes the first segment record to the NEWICON file. Upon return from subroutine WRCHAR, IOST
contains the I/O status of the first segment write to the NEWICON file. IOST is tested; if IOST does not
F-19
-------
equal zero, a write error has occurred. A message is written to the log indicating that an error has occurred.
The error message contains the subroutine name, the unit number and I/O status. The program exits by a call
to the system subroutine EXIT.
CODE FOR ERROR CHECK #2:
CALL WRCHAR (UMITNI, SEG1BF, IOST)
IF (IOST .NE. 0) THEN
WRITECLUNOUT, 2003) UNITNI, IOST
2003 FORMAT(/ 5X, 'XXX HEADER WRITE ERROR IN NEUICS1,
& / 5X, 'UNIT NUMBER = ', 12, 2X, 'I/O STATUS = ', 14)
CALL EXIT
END IF
ERROR CHECK #3:
The species names record is written to an internal buffer using a formatted write statement. The I/O status of
the write operation is stored in variable IOST. IOST is tested; if IOST does not equal zero, a write error has
occurred. A message is written to the log indicating that an error has occurred. The error message contains
the subroutine name and I/O status value. The program exits by a call to the system subroutine EXJT.
CODE FOR ERROR CHECK #3:
c
C write the species names
WRITE(SPNMBF, 1003, IOSTAT = IOST)
& (SPNMCN(ISPC), ISPC = 1, NSPCCN)
1003 FORMAT( 35(A4) )
IF (IOST .NE. 0) THEN
WRITECLUNOUT, 2001) ' SPEC. NAMES', IOST
CALL EXIT
END IF
ERROR CHECK #4:
The unit number, the buffer containing the species names record, and IOST are passed to subroutine
WRCHAR, which writes the species names record to the NEWICON file. Upon return from the subroutine
WRCHAR, IOST contains the I/O status of the write of the species names record to the NEWICON file. IOST
is tested, and if not equal to zero, then a write error has occurred. A message is written to the log indicating
that an error has occurred. The error message contains the subroutine name and I/O status value. The pro-
gram exits by a call to the system subroutine EXIT.
F-20
-------
CODE FOR ERROR CHECK #4:
CALL WRCHAR (UNITNI, SPNMBF. IOST)
IF (IOST .NE. 0) THEN
WRITE(LUNOUT, 2005) IOST
2005 FORMAT(/ 5X, 'XXX SPECIES NAMES WRITE ERROR IN NEUICS:
&2X, 'I/O STATUS = ', 14)
CALL EXIT
END IF
ERROR CHECK #5:
The layer names record is written to an internal buffer using a formatted write statement. The I/O status of
the write operation is stored in variable IOST. IOST is tested; if IOST does not equal zero, a write error has
occurred. A message is written to the log indicating that an error has occurred. The error message contains
the subroutine name and I/O status value. The program exits by a call to the system subroutine EXIT.
CODE FOR ERROR CHECK #5:
c
C write the level names
WRITEUEVNBF, 1005, IOSTAT = IOST)
& (LVNMCN(ILEV). ILEV = 1. NLEVCN)
1005 FORMAT( 3(A4) )
IF (IOST .NE. 0) THEN
WRITEUUNOUT, 2001) ' LEV. NAMES', IOST
CALL EXIT
END IF
ERROR CHECK #6:
The unit number, the buffer containing the layer names record, and IOST are passed to subroutine WRCHAR,
which writes the level names record to the NEWICON file. Upon return from subroutine WRCHAR, IOST
contains the I/O status of the write of the layer names record to the NEWICON file. IOST is tested, and if not
equal to zero, then a write error has occurred. A message is written to the log indicating that an error has
occurred. The error message contains the subroutine name and I/O status value. The program exits by a call
to the system subroutine EXIT.
CODE FOR ERROR CHECK #6:
CALL WRCHAR (UNITNI, LEVNBF, IOST)
IF (IOST .NE. 0) THEN
WRITE(LUNOUT, 2007) IOST
2007 FORMAT(/ 5X, 'XXX LEVEL NAMES WRITE ERROR IN NEUICS:
&2X, 'I/O STATUS = ', 14)
END IF
ERROR CHECK #7:
F-21
-------
The text records are written by subroutine WRCHAR to the NEWICON file. The I/O status of the write oper-
ation is stored in variable IOST. IOST is tested; if IOST does not equal zero, a write error has occurred. A
message is written to the log indicating that an error has occurred. The error message contains the subroutine
name and I/O status value. The program exits by a call to the system subroutine EXIT.
CODE FOR ERROR CHECK #7:
c
C write file text group
DO 101 ITXT = 1, ICNTCN
CALL WRCHAR (UNITNI. TEXTCN(ITXT), IOST)
IF (IOST .ME. 0) THEM
WRITE(LUNOUT, 2009) IOST
2009 FORMAT(/ 5X, 'XXX TEXT WRITE ERROR IN NEWICS: ',
&2X, 'I/O STATUS = ', 14)
CALL EXIT
END IF
101 CONTINUE
ERROR CHECK #8:
The NEWICON time step header is written by calling subroutine WRFILE. The unit number, number of
words, starting address of the common block RTSHIC, and the variable IOST are passed to subroutine
WRFILE. Upon return from subroutine WRFILE, IOST contains the I/O status value of the write of the
NEWICON time step header. IOST is tested; if IOST does not equal zero, an error occurred while writing the
time step header. A message is written to the log indicating that an error has occurred. The error message
contains the subroutine name, the unit number, the I/O status value and the number of words in the buffer to
be written. The program exits by a call to the system subroutine EXIT.
CODE FOR ERROR CHECK #8:
c
C write NI T.S.H.
CALL WRFILE (UNITNI, IWOLTH, DATIC, IOST}
IF (IOST .NE. 0) THEN
URITEUUNOUT, 2013) UNITNI, IOST, IUDLTH
2013 FORNAT(/ 5X, 'XXX T.S.H. WRITE ERROR IN NEWICS',
& / 5X, 'UNIT NUMBER = ', 12, 2X, 'I/O STATUS = ', 14.
& / 5X, 'NO. OF WORDS = ', 14)
CALL EXIT
END IF
ERROR CHECK #9: . .
Each row of the NEWICON file is written by iterating over the chemical species. For each iteration, a call is
made to subroutine WRFILE. The unit number, number of words, starting address of the common block
ICFILE, and the variable IOST are passed to WRFILE. Upon return from the subroutine call to WRFILE, IOST
contains the I/O status value of a write operation to the NEWICON file. IOST is tested; if IOST is not equal
F-22
-------
to zero, then an error occurred on the write operation. A message is written to the log to indicate an error has
occurred. The error message contains the subroutine name, the unit number, the I/O status value and the
number of words to be written. The program exits by a call to the system subroutine EXIT.
CODE FOR ERROR CHECK #9:
c
C write NEWI CON file
DO 205 ISPC = 1, NSPECS
CALL WRFILECUNITNI, IUDLTH, ICFILE(1,1,ISPC),IOST)
IF (IOST .NE. 0) THEN
URITE(LUNOUT, 2015) UNITNI, IOST, IUDLTH
2015 FORNAT(/ 5X, 'XXX ROW WRITE ERROR IN NEWICS1,
& / 5X, 'UNIT NUMBER = ', 12, 2X, 'I/O STATUS = ', 14,
& / 5X, 'NO. OF WORDS = ', 14)
CALL EXIT
END IF
205 CONTINUE
SUBROUTINE OPBCON
ERROR CHECK #1:
The BCON file is opened. The first segment record is read from the BCON file into a buffer by calling sub-
routine RDCHAR. The unit number, buffer, and IOST are passed to RDCHAR. Upon return from subroutine
RDCHAR, IOST contains the I/O status value of the read of the BCON first segment record. IOST is tested; if
IOST does not equal zero, an error occurred on the read operation. A message is written to the log indicating
that an error has occurred. The error message contains the subroutine name and the I/O status value. The
program exits by a call to the system subroutine EXIT.
CODE FOR ERROR CHECK #1:
c
C read 1st segment
CALL RDCHAR (UNITBC, SEG1BF. IOST)
C
IF (IOST .NE. 0) THEN
WRITE(LUNOUT, 2001) IOST, 'FIRST RECORD '
2001 FORMAT(/ 5X, 'XXX READ ERROR IN OPBCON1
&5X, 'IOSTAT = ', 14, 4X, A16)
CALL EXIT
END IF
ERROR CHECK #2:
Next, a formatted read from the buffer that contains the first segment record is made, and the common block
HEADBC is loaded. The I/O status of the read operation is set to be the variable IOST. IOST is tested; if
F-23
-------
IOST does not equal zero, a read error has occurred. A message is written to the log indicating that an error
has occurred. The error message contains the subroutine in which the error occurred and the I/O status value.
The program exits by a call to the system subroutine EXIT.
CODE FOR ERROR CHECK #2:
c
C convert character header to mixed char & numeric
READ(SEG1BF, 1001, IOSTAT = IOST)
& CDATBC, CTIMBC, SOATBC, STHRBC, TSTPBC, FRSTBC,
& GRDNBC,
& SWLNBC, SULTBC, NELNBC, NELTBC,
& DLONBC, DLATBC,
& NCOLBC, NROUBC, NLEVBC, NSPCBC, ICNTBC
1001 FORMAK6I8.8, A8, 4F8.3, 2F8.5, 514.4)
IF (IOST .NE. 0) THEN
WRITECLUNOUT, 2001) IOST, 'INTERNAL READ #1*
CALL EXIT
END IF
ERROR CHECK #3:
The next record read from the BCON file is the species names record. This record is read into a buffer by
calling subroutine RDCHAR. The unit number, buffer, and IOST are passed to RDCHAR. A formatted read
from the buffer that contains the species names record is made, and the common block HEADBC is loaded.
The I/O status of the read operation is set to be the variable IOST. IOST is tested; if IOST does not equal
zero, a read error has occurred. A message is written to the log indicating that an error has occurred. The
error message contains the subroutine in which the error occurred and the I/O status value. The program
exits by a call to the system subroutine EXIT.
CODE FOR ERROR CHECK #3:
c
C read the species names record
CALL RDCHAR (UNITBC, SPNMBF, IOST)
READ(SPNMBF, 1003, IOSTAT = IOST)
& (SPNMBC(ISPC), ISPC = 1, NSPCBC)
1003 FORMAT(6(10(A4))/)
IF (IOST .NE. 0) THEN
WRITECLUNOUT, 2001) IOST, 'SPECIES NAMES '
CALL EXIT
END IF
ERROR CHECK #4:
The next record read from the BCON file is the layer names record. This record is read into a buffer by calling
subroutine RDCHAR. The unit number, buffer, and IOST are passed to RDCHAR. A formatted read from the
buffer that contains the layer names record is made, and the common block HEADBC is loaded. The I/O
status of the read operation is set to be the variable IOST. IOST is tested; if IOST does not equal zero, a read
F-24
-------
error has occurred. A message is written to the log to indicate that an error has occurred. The error message
contains the subroutine in which the error occurred and the I/O status value. The program exits by a call to
the system subroutine EXIT.
CODE FOR ERROR CHECK #4:
c
C read level names record
CALL RDCHAR (UNITBC, LEVNBF, IOST)
READ(LEVNBF, 1005, IOSTAT = IOST)
& (LVNMBC(ILEV), ILEV = 1, NLEVBC)
1005 FORMAT(2(10(A4))/)
IF (IOST .NE. 0} THEN
WRITE(LUNOUT, 2001) IOST, 'LEVEL NAMES '
CALL EXIT
END IF
ERROR CHECK #5:
The next records read are the text segment records. These records are read into a buffer by iterating over the
number of text records. For each iteration, a call to subroutine RDCHAR is made. The unit number, buffer,
and IOST are passed to RDCHAR. The I/O status of the read operation is set to be the variable IOST. IOST is
tested; if IOST does not equal zero, a read error has occurred. A message is written to the log to indicate an
error has occurred. The error message contains the subroutine in which the error occurred and the I/O status
value. The program exits by a call to the system subroutine EXIT.
CODE FOR ERROR CHECK #5:
c
C read file text group
DO 101 ITXT = 1, ICNTBC
CALL RDCHAR (UNITBC, TEXTBF, IOST)
C
IF (IOST .NE. 0) THEN
URITE(LUNOUT, 2001) IOST, 'TEXT RECORDS '
CALL EXIT
END IF
C
TEXTBC(ITXT) = TEXJBF
101 CONTINUE
SUBROUTINE OPBMAT
NONE
SUBROUTINE OPBTRK
NONE
F-25
-------
SUBROUTINE OPCONC
ERROR CHECK #1:
The CONC file is opened. The first segment record is read from the CONC file into a buffer by calling sub-
routine RDCHAR. The unit number, buffer, and IOST are passed to RDCHAR. Upon return from subroutine
RDCHAR, IOST contains the I/O status value of the read of the CONC first segment record. IOST is tested; if
IOST does not equal zero, an error occurred on the read operation. A message is written to the log indicating
that an error has occurred. The error message contains the subroutine name and the I/O status value. The
program exits by a call to the system subroutine EXIT.
CODE FOR ERROR CHECK #1:
c
C read 1st segment
CALL RDCHAR (UNITCN, SEG1BF, IOST)
C
IF (IOST .NE. 0) THEN
WRITEUUNOUT, 2001) IOST, 'FIRST RECORD '
2001 FORMAT(/ 5X, 'XXX READ ERROR IN OPCONC1
45X. 'IOSTAT = ', 14, 4X, A16)
CALL EXIT
END IF
ERROR CHECK #2:
Next, a formatted read from the buffer that contains the first segment record is made, and the common block
HEADCN is loaded. The I/O status of the read operation is set to be the variable IOST. IOST is tested; if
IOST does not equal zero, a read error has occurred. A message is written to the log indicating that an error
has occurred. The error message contains the subroutine in which the error occurred and the I/O status value.
The program exits by a call to the system subroutine EXIT.
F-26
-------
CODE FOR ERROR CHECK #2:
READ(SEG1BF, 1001, IOSTAT = IOST)
& CDATCN, CTIMCN, SDATCN, STHRCN, TSTPCN, FRSTCN,
& GRDNCN,
& SWLNCN. SULTCN, NELNCN, NELTCN,
& DLONCN, DLATCN,
& MCOLCN, NROUCN, NLEVCN, NSPCCN,
& CDBMCN, CTBMCN,
& CDBTCN, CTBTCN,
& CDBCCN, CTBCCN,
& CDICCN, CTICCN.
& ICNTCN
1001 FORMAT(6I8.8, A8, 4F8.3, 2F8.5, 414.A, 818.8, 14.4)
IF (IOST .NE. 0) THEN
WRITE(LUNOUT, 2001) IOST, 'INTERNAL READ #1'
CALL EXIT
END IF
ERROR CHECK #3:
The next record read from the CONC file is the layer names record. This record is read into a buffer by calling
subroutine RDCHAR. The unit number, buffer, and IOST are passed into RDCHAR. A formatted read from the
buffer that contains the level names record is made, and the common block HEADCN is loaded. The I/O
status of the read operation is set to be the variable IOST. IOST is tested; if IOST does not equal zero, a read
error has occurred. A message is written to the log indicating that an error has occurred. The error message
contains the subroutine name and the I/O status value. The program exits by a call to the system subroutine
EXIT.
CODE FOR ERROR CHECK #3:
CALL RDCHAR (UNITCN, LEVNBF, IOST)
READUEVMBF, 1005, IOSTAT = IOST)
& (LVNHCN(ILEV), ILEV = 1, NLEVCN)
1005 FORMAT( (A4) )
IF (IOST .NE. 0) THEN
WRITE(LUNOUT, 2001) IOST, 'LEVEL NAMES
CALL EXIT
END IF
ERROR CHECK #4:
The next records read are the text segment records. These records are read into a buffer by iterating over the
number of text records. For each iteration, a call to subroutine RDCHAR is made. The unit number, buffer,
and IOST are passed to RDCHAR. The I/O status of the read operation is set to be the variable IOST. IOST is
tested; if IOST does not equal zero, a read error has occurred. A message is written to the log indicating that
an error has occurred. The error message contains the subroutine name and the I/O status value. The pro-
gram exits by a call to the system subroutine EXIT.
F-27
-------
CODE FOR ERROR CHECK #4:
c
C read file text group
DO 101 ITXT = 1, ICNTCN
CALL RDCHAR (UNITCN, TEXTBF, IOST)
IF (IOST .NE. 0) THEN
WRITEaUNOUT, 2001) IOST, 'TEXT RECORDS
CALL EXIT
END IF
101 CONTINUE
SUBROUTINE OPICON
ERROR CHECK #1:
The ICON file is opened. The first segment record is read from the ICON file into a buffer by calling sub-
routine RDCHAR. The unit number, buffer, and IOST are passed to RDCHAR. The I/O status of the read
operation is stored in the variable IOST. In subroutine OPICON, IOST is tested, and if not equal to zero, then
an error occurred on the read operation. A message is written to the log indicating that an error has occurred.
The error message contains the subroutine name and the I/O status value. The program exits by a call to the
system subroutine EXIT.
CODE FOR ERROR CHECK #1:
c
C read 1st segment
CALL RDCHAR (UNITIC, SEG1BF, IOST)
C
IF (IOST .NE. 0) THEN
URITEaUNOUT, 2001) IOST, 'FIRST RECORD <
2001 FORMAT(/ 5X, 'XXX READ ERROR IN OPICON1
S5X, 'IOSTAT s ', 14, «x, A16)
CALL EXIT
END IF
ERROR CHECK #2:
Next, a formatted read from the buffer that contains the first segment record is made, and the common block
HEADIC is loaded. The I/O status of the read operation is set to be the variable IOST. IOST is tested; if
IOST does not equal zero, a read error has occurred. A message is written to the log indicating that an error
has occurred. The error message contains the subroutine in which the error occurred and the I/O status value.
The program exits by a call to the system subroutine EXIT.
F-28
-------
CODE FOR ERROR CHECK #2:
READ(SEG1BF, 1001, IOSTAT - IOST)
& CDATIC, CTIMIC, SDATIC. STHRIC, TSTPIC. FRSTIC,
& GRDNIC,
& SULNIC. SULTIC, NELNIC, NELTIC,
& DLONIC, DLATIC,
& NCOLIC, NROUIC, NLEVIC, NSPCIC,
& (IDUM(ITXT), ITXT = 1, 8),
& ICNTIC
1001 FORMAT(6I8.8, A8, 4F8.3, 2F8.5, 414.4, 818, 14.4)
IF (IOST .NE. 0) THEN
WRITE(LUNOUT, 2001) IOST, 'INTERNAL READ #1'
CALL EXIT
END IF
ERROR CHECK #3:
The next record read from the ICON file is the species names record. This record is read into a buffer by
calling subroutine RDCHAR. The unit number, buffer, and IOST are passed into RDCHAR. A formatted read
from the buffer that contains the species names record is made, and the common block HEADIC is loaded.
The I/O status of the read operation is set to be the variable IOST. IOST is tested; if IOST does not equal
zero, a read error has occurred. A message is written to the log indicating that an error has occurred. The
error message contains the subroutine name and the I/O status value. The program exits by a call to the
system subroutine EXIT.
CODE FOR ERROR CHECK #3:
c
C read the species names record
CALL RDCHAR (UNITIC, SPMMBF, IOST)
READCSPNMBF, 1003, IOSTAT = IOST)
S(SPNMICdSPC), ISPC = 1, NSPCIC)
1003 FORMAT( (A4) )
IF (IOST .NE. 0) THEN
URITE(LUNOUT, 2001) IOST, 'SPECIES NAMES '
CALL EXIT
END IF
ERROR CHECK #4:
The next record read from the ICON file is the layer names record. This record is read into a buffer by calling
subroutine RDCHAR. The unit number, buffer, and IOST are passed into RDCHAR. A formatted read from the
buffer that contains the level names record is made and the common block HEADIC is loaded. The I/O status
of the read operation is set to be the variable IOST. IOST is tested; if IOST does not equal zero, a read error
has occurred. A message is written to the log indicating that an error has occurred. The error message con-
tains the subroutine name and the I/O status value. The program exits by a call to the system subroutine EXIT.
F-29
-------
CODE FOR ERROR CHECK #4:
c
C read the level names record
CALL RDCHAR (UNITIC, LEVNBF, IOST)
READ(LEVNBF, 1005, IOSTAT = IOST)
&(LVNMIC(ILEV), ILEV = 1, NLEVIC)
1005 FORMAT( (A4) )
IF (IOST .NE. 0) THEN
WRITE(LUNOUT, 2001) IOST, 'LEVEL NAMES '
CALL EXIT
END IF
ERROR CHECK #5:
The next records read are the text segment records. These records are read into a buffer by iterating over the
number of text records. For each iteration, a call to subroutine RDCHAR is made. The unit number, buffer,
and IOST are passed to RDCHAR. The I/O status of the read operation is set to be the variable IOST. IOST is
tested; if IOST does not equal zero, a read error has occurred. A message is written to the log indicating that
an error has occurred. The error message contains the subroutine name and the I/O status value. The pro-
gram exits by a call to the system subroutine EXIT.
CODE FOR ERROR CHECK #5:
c
C read file text group
DO 101 ITXT = 1, ICNTIC
CALL RDCHAR (UNITIC. TEXTBF, IOST)
IF (IOST .NE. 0) THEN
WRITEUUNOUT, 2001) IOST, 'TEXT RECORDS '
CALL EXIT
END IF
C
TEXTIC(ITXT) = TEXTBF
101 CONTINUE
SUBROUTINE OPSTAV
ERROR CHECK #1:
The STATE VECTOR file is opened. IOST contains the I/O status of the open operation. IOST is tested; if
IOST is not equal to zero, then an error occurred while opening the file. A message is written to the log
indicating that an error has occurred. The error message contains the subroutine name, the unit number and
the I/O status value. The program exits by a call to the system subroutine EXIT.
F-30
-------
CODE FOR ERROR CHECK #1:
c
C open the STATE VECTOR file
C
UNITSV = JUNITO
OPEN (UNITSV,
& FILE = FLNNSV,
& ACCESS = 'SEQUENTIAL1,
& STATUS = 'UNKNOWN',
& IOSTAT = IOST)
C
IF (IOST .NE. 0) THEN
URITE(LUNOUT, 2001) UNITSV, IOST
2001 FORMAT(/ 5X, 'XXX SV FILE OPEN ERROR IN OPSTAV
& / 5X, 'UNIT NUMBER = ', 12
& / 5X, 'I/O STATUS = ', 14)
CALL EXIT
END IF
ERROR CHECK #2:
A formatted read of the second segment header record of the STATE VECTOR file is made and the common
block HEADSV is loaded. The I/O status of the read operation is stored in the variable IOST. IOST is tested;
if IOST does not equal zero, a read error has occurred. A message is written to the log indicating that an error
has occurred. The error message contains the subroutine name, the unit number, the record number, and the
I/O status value. The program exits by a call to the system subroutine EXIT.
CODE FOR ERROR CHECK #2:
c
C read STATE VECTOR header segment 2
C
READ(UNITSV, FMT = 1003, IOSTAT = IOST)
& NELTSV, DLONSV. DLATSV, NCOLSV,
& NROUSV, NLEVSV, NSPCSV, ICNTSV
1003 FORMATdX. F8.3, 2(1X, F8.5), 5(1X, 14))
IF (ICNTSV .EQ. 0) ICNTSV - 1
IF (IOST .NE. 0) THEN
WRITE(LUNOUT, 2003) UNITSV, RECNSV, IOST
2003 FORMAT(/ 5X, 'XXX SV HEADER READ ERROR IN OPSTAV
& / 5X, 'UNIT NUMBER = ', 12, 5X, 'RECORD = ', 14
& / 5X, 'I/O STATUS = ', 14)
CALL EXIT
END IF
ERROR CHECK #3:
A formatted read of the first segment of the species names record of the STATE VECTOR file is made, and
the common block CHARSV is loaded. The I/O status of the read operation is stored in the variable IOST.
IOST is tested; if IOST does not equal zero, a read error has occurred. A message is written to the log indi-
cating that an error has occurred. The error message contains the subroutine name, the unit number, the
record number, and the I/O status value. The program exits by a call to the system subroutine EXIT.
F-31
-------
CODE FOR ERROR CHECK #3:
c
C read species names records
C
READ(UNITSV, FMT = 1005, IOSTAT s IOST)
& (SPNMSV(ISPC), ISPC = 1, 15)
1005 FORMAT(1X, 15(A4, 1X»
IF (IOST .NE. 0) THEN
WRITE(LUNOUT, 2005) UNITSV, RECNSV, IOST
2005 FORMAT(/ 5X, 'XXX SPECIES NAMES READ ERROR IN OPSTAV
& / 5X, 'UNIT NUMBER = ', 12, 5X, 'RECORD = '. 14
& / 5X, 'I/O STATUS = ', 14)
WRITECLUNOUT, 2005) UNITSV, RECNSV, IOST
CALL EXIT
END IF
ERROR CHECK #4:
A formatted read of the second segment of the species names record of the STATE VECTOR file is made,
and the common block CHARSV is loaded. The I/O status of the read operation is stored in the variable
IOST. IOST is tested; if IOST does not equal zero, a read error has occurred. A message is written to the log
indicating that an error has occurred. The error message contains the subroutine name, the unit number, the
record number, and the I/O status value. The program exits by a call to the system subroutine EXIT.
CODE FOR ERROR CHECK #4:
c
READ(UNITSV, FMT = 1005, IOSTAT = IOST)
& (SPNMSV(ISPC), ISPC - 16, NSPCSV)
IF (IOST .NE. 0) THEN
WRITECLUNOUT, 2005) UNITSV, RECNSV, IOST
CALL EXIT
END IF
ERROR CHECK #5:
A formatted read of the layer names record of the STATE VECTOR file is made, and the common block
CHARSV is loaded. The I/O status of the read operation is stored in the variable IOST. IOST is tested; if
IOST does not equal zero, a read error has occurred. A message is written to the log indicating that an error
has occurred. The error message contains the subroutine name, the unit number, the record number, and the
I/O status value. The program exits by a call to the system subroutine EXIT.
F-32
-------
CODE FOR ERROR CHECK #5:
c
C read level names record
C
RECNSV = RECNSV + 1
READCUNITSV, FMT = 1005, IOSTAT = IOST)
& (LVNMSV(ILEV), ILEV = 1, NLEVSV)
IF (IOST .NE. 0) THEN
WRITEdUNOUT, 2007} UNITSV. RECNSV, IOST
2007 FORMAT(/ 5X, 'XXX LEVEL NAMES READ ERROR IN OPSTAV
& / 5X, 'UNIT NUMBER = >, 12, 5X, 'RECORD = ', 14
& / 5X, 'I/O STATUS = ', K)
CALL EXIT
END IF
ERROR CHECK #6:
The text segment records are read by iterating over the number of text records. For each iteration, a for-
matted read of a text record of the STATE VECTOR file is made, and the common block CHARSV is loaded.
The I/O status of each read operation is stored in the variable IOST. IOST is tested; if IOST does not equal
zero, a read error has occurred. A message is written to the log indicating that an error has occurred. The
error message contains the subroutine name, the unit number, the record number, and the I/O status value.
The program exits by a call to the system subroutine EXIT.
CODE FOR ERROR CHECK #6:
DO 103 ITXT = 1, ICNTSV
RECNSV = RECNSV + 1
READCUNITSV, FMT = 1007, IOSTAT = IOST) TEXTSV(ITXT)
1007 FORMATdX, A80)
IF (IOST .NE. 0) THEN
WRITEUUNOUT. 2003) UNITSV, RECNSV, IOST
CALL EXIT
END IF
103 CONTINUE
ERROR CHECK #7:
A formatted read from the STATE VECTOR file unit number loads the first header segment record into the
common block HEADIN. The I/O status of the read operation is set to be the variable IOST. IOST is tested;
if IOST does not equal zero, a read error has occurred. A message is written to the log indicating that an error
has occurred. The error message contains the subroutine name, the unit number, the record number, and the
I/O status value. The program exits by a call to the system subroutine EXIT.
F-33
-------
CODE FOR ERROR CHECK #7:
c
C read HEADIN header record segment 1
C
READCUNITSV. FMT = 1001, IOSTAT = IOST)
& CDATIN, CTIMIN, SDATIN, STHRIN.
& TSTPIN, FRSTIN, GRONIN, SWLNIN,
& SULTIN, NELNIN
IF (IOST .NE. 0) THEN
URITE(LUNOUT, 2009} UNITSV. RECNSV, IOST
2009 FORMAT(/ 5X, 'XXX HE AD IN READ ERROR IN OPSTAV
& / 5X, 'UNIT NUMBER - ', 12, 5X, 'RECORD = ', 14
& / 5X, 'I/O STATUS = ', 14)
CALL EXIT
END IF
ERROR CHECK #8:
A formatted read from the STATE VECTOR file unit number loads the second header segment record
(species index records) into the common block HEADIN. The I/O status of the read operation is set to be the
variable IOST. IOST is tested; if IOST does not equal zero, a read error has occurred. A message is written
to the log indicating that an error has occurred. The error message contains the subroutine name, the unit
number, the record number, and the I/O status value. The program exits by a call to the system subroutine
EXIT.
CODE FOR ERROR CHECK #8:
c
C read HEADIN header record segment 2
C
READCUNITSV, FMT = 1003, IOSTAT = IOST)
& NELTIN, DLONIN, DLATIN, NCOLIN,
& NROWIN, NLEVIN, NSPCIN, ICNTIN
IF (IOST .NE. 0) THEN
URITECLUNOUT, 2011) UNITSV, RECNSV, IOST
2011 FORMAT(/ 5X, 'XXX HEADIN INTERNAL READ ERROR IN OPSTAV '
& / 5X, 'UNIT NUMBER = ', 12, 5X, 'RECORD = ', 14
& / 5X, 'I/O STATUS = ', 14)
CALL EXIT
END IF
ERROR CHECK #9:
A formatted read from the STATE VECTOR file unit number loads the chemistry control records into the
common block CHEMIN. The I/O status of the read operation is set to be the variable IOST. IOST is tested;
if IOST does not equal zero, a read error has occurred. A message is written to the log indicating that an error
has occurred. The error message contains the subroutine name, the unit number, the record number, and the
I/O status value. The program exits by a call to the system subroutine EXIT.
F-34
-------
CODE FOR ERROR CHECK #9:
c
RECNSV = RECNSV + 1
READ(UNJTSV, FMT = 1015, IOSTAT = IOST)
& (ISPEC(ISPC), ISPC = 1, NCOUT), ULIM, BUM, FNOLIH
1015 FORMATdX, (I4.3, 1X>, 3(E10.3, 1X) )
IF (IOST .NE. 0) THEN
WRITE(LUNOUT, 2013) UNITSV, RECNSV, IOST
2013 FORMAT(/ 5X, 'XXX CHEHIN READ ERROR IN OPSTAV'
& / 5X, 'UNIT NUMBER = ', 12, 5X, 'RECORD = ', 14
& / 5X, 'I/O STATUS = ', 14)
CALL EXIT
END IF
ERROR CHECK #10:
A formatted read from the STATE VECTOR file unit number loads the ICON species index records into the
common block NDXSPC. The I/O status of the read operation is set to be the variable IOST. IOST is tested;
if IOST does not equal zero, a read error has occurred. A message is written to the log indicating that an error
has occurred. The error message contains the subroutine name, the unit number, the record number, and the
I/O status value. The program exits by a call to the system subroutine EXIT.
CODE FOR ERROR CHECK #10:
READ(UNITSV. FMT = 1019, IOSTAT = IOST)
& (NXSPIC(ISPC), ISPC = 16, NSPECS)
C
IF (IOST .NE. 0) THEN
URITEdUNOUT, 2015} UNITSV. RECNSV, IOST
2015 FORMAT(/ 5X, 'XXX SPECIES ORDER LIST READ ERROR IN OPSTAV
& / 5X, 'UNIT NUMBER = ', 12, 5X, 'RECORD = ', 14
& / 5X, 'I/O STATUS = ', 14)
CALL EXIT
END IF
SUBROUTINE OPWRCN
ERROR CHECK #1:
A formatted write of the common blocks CHARCN and HEADCN is made to a buffer. CHARCN and
HEADCN contain the first segment record of the CONC file. The I/O status of the write operation is stored
in the variable IOST. IOST is tested; if IOST does not equal zero, a write error has occurred. A message is
written to the log indicating that an error has occurred. The error message contains the subroutine in which
the error occurred and the I/O status value. The program exits by a call to the system subroutine EXIT.
F-35
-------
CODE FOR ERROR CHECK #1:
c
C write 1st segment
WRITECSEG1BF, 1001, IOSTAT = IOST)
& COATCN, CTIMCN, SOATCN, STHRCN, TSTPCN, FRSTCN,
& GRDNCN,
& SWLNCN, SULTCN, NELNCN, NELTCN,
& DLONCN, DLATCN,
& NCOLCN, NROWCN, NLEVCN, NSPCCN,
& COBMCN, CTBMCN,
& COBTCN, CTBTCN,
& COBCCN, CTBCCN,
& CDICCN, CTICCN,
& ICNTCN
1001 FORHAK6I8.8, AS, 4F8.3, 2F8.5, 414.4, 818.8, 14.4)
IF (IOST .NE. 0) THEN
WRITE(LUNOUT, 2001) ' SEGMENT 1 ', IOST
2001 FORMAT(/ 5X, 'XXX INTERNAL WRITE ERROR IN OPWRCN: ', A12,
& / 5X, '1/0 STATUS = ', 14)
CALL EXIT
END IF
ERROR CHECK #2:
The buffer containing the first segment record is written to the CONC file by calling subroutine WRCHAR.
The unit number, buffer, and IOST are passed to WRCHAR. Upon return from subroutine WRCHAR, IOST
contains the I/O status value of the write of the CONC first segment header record. IOST is tested; if IOST
does not equal zero, then an error occurred on the write operation. A message is written to the log to indicate
an error has occurred. The error message contains the subroutine name and the I/O status value. The pro-
gram exits by a call to the system subroutine EXIT.
CODE FOR ERROR CHECK #2:
CALL WRCHAR (UNITCN, SEG1BF, IOST)
IF (IOST .NE. 0) THEN
WRITE(LUNOUT, 2003) UNITCN, IOST
2003 FORMAT(/ 5X, 'XXX HEADER WRITE ERROR IN OPWRCN1
& / 5X, 'UNIT NUMBER = ', 12, 2X, 'I/O STATUS * ', 14)
CALL EXIT
END IF
ERROR CHECK #3:
A formatted write of the common block CHARCN is made to a buffer. CHARCN contains the species names
record of the CONC file. The I/O status of the write operation is stored in the variable IOST. IOST is tested;
if IOST does not equal zero, a write error has occurred. A message is written to the log indicating that an
error has occurred. The error message contains the subroutine in which the error occurred and the I/O status
value. The program exits by a call to the system subroutine EXIT.
F-36
-------
CODE FOR ERROR CHECK #3:
c
C write the species names
URITE(SPNMBF, 1003, IOSTAT = IOST)
& (SPNHCN(ISPC), ISPC = 1, NSPCCN)
1003 FORMAT( (A4) )
IF (IOST .ME. 0) THEN
URITECLUNOUT, 2001) ' SPEC. NAMES', IOST
CALL EXIT
END IF
C
ERROR CHECK #4:
The buffer containing the species names record is written to the CONC file by calling subroutine WRCHAR.
The unit number, buffer, and IOST are passed to WRCHAR. Upon return from subroutine WRCHAR, IOST
contains the I/O status value of the write of the CONC species names record. IOST is tested; if IOST does not
equal zero, then an error occurred on the write operation. A message is written to the log indicating that an
error has occurred. The error message contains the subroutine in which the error occurred and the I/O status
value. The program exits by a call to the system subroutine EXIT.
CODE FOR ERROR CHECK #4:
CALL WRCHAR (UNITCN, SPNMBF, TOST)
- IF (IOST .NE. 0) THEN
WRITE(LUNOUT. 2005) IOST
2005 FORMAT(/ 5X, 'XXX SPECIES NAMES WRITE ERROR IN OPWRCN:
&2X, 'I/O STATUS = ', U)
CALL EXIT
END IF
ERROR CHECK #5:
A formatted write of the common block CHARCN is made to a buffer. CHARCN contains the level names
record of the CONC file. The I/O status of the write operation is stored in the variable IOST. IOST is tested;
if IOST does not equal zero, a write error has occurred. A message is written to the log to indicate that an
error has occurred. The error message contains the subroutine in which the error occurred and the I/O status
value. The program exits by a call to the system subroutine EXIT.
F-37
-------
CODE FOR ERROR CHECK #5:
c
C write the level names
URITE(LEVNBF, 1005, IOSTAT = IOST)
& (LVNMCN(ILEV), ILEV = 1, NLEVCN)
1005 FORMAT( (A4) )
IF (IOST .NE. 0) THEN
URITECLUNOUT, 2001) ' LEV. NAMES', IOST
CALL EXIT
END IF
ERROR CHECK #6:
The buffer containing the level names record is written to the CONC file by calling subroutine WRCHAR. The
unit number, buffer, and IOST are passed to WRCHAR. Upon return from subroutine WRCHAR, IOST con-
tains the I/O status value of the write of the CONC level names record. IOST is tested; if IOST does not equal
zero, then an error occurred on the write operation. A message is written to the log indicating that an error
has occurred. The error message contains the subroutine in which the error occurred and the I/O status value.
The program exits by a call to the system subroutine EXIT.
CODE FOR ERROR CHECK #6:
c
CALL URCHAR (UNITCN, LEVNBF, IOST)
IF (IOST .NE. 0) THEN
WRITE(LUNOUT, 2007) IOST
2007 FORMAT(/ 5X, 'XXX LEVEL NAMES WRITE ERROR IN OPWRCN: ',
&2X, 'I/O STATUS = ', 14)
END IF
ERROR CHECK #7:
The first text record is written to the CONC file by calling subroutine WRCHAR. The unit number, buffer, and
IOST are passed to WRCHAR. Upon return from subroutine WRCHAR, IOST contains the I/O status value of
the write of the first CONC text record. IOST is tested; if IOST does not equal zero, then an error occurred
on the write operation. The I/O status of the write operation is stored in the variable IOST. IOST is tested; if
IOST does not equal zero, a write error has occurred. A message is written to the log to indicate that an error
has occurred. The error message contains the subroutine in which the error occurred and the I/O status value.
The program exits by a call to the system subroutine EXIT.
F-38
-------
CODE FOR ERROR CHECK #7:
c
C copy input text records to CONC file
TEXTCN(I) = 'ROM2.1 '
CALL WRCHAR (UNITCN, TEXTCN(1), IOST)
IF (IOST .NE. 0) THEN
WRITE(LUNOUT, 2009) IOST
2009 FORMAK/ 5X, 'XXX TEXT WRITE ERROR IN OPWRCN:
&2X, 'I/O STATUS = '. 14)
CALL EXIT
END IF
ERROR CHECK #8:
The remaining text records are written to the CONC file by iterating over the number of text records. For
each iteration a call to subroutine WRCHAR is made. The unit number, buffer, and IOST are passed to
WRCHAR. Upon return from subroutine WRCHAR, IOST contains the I/O status value of the write of the
CONC text records. IOST is tested; if IOST does not equal zero, then an error occurred on the write opera-
tion. A message is written to the log indicating that an error has occurred. The error message contains the
subroutine in which the error occurred and the I/O status value. The program exits by a call to the system
subroutine EXIT.
CODE FOR ERROR CHECK #8:
. DO 103 ITXT = 2, ICNTCN
TEXTCN(ITXT) * TEXTINdTXT - 1)
C
CALL WRCHAR (UNITCN, TEXTCN(ITXT). IOST)
IF (IOST .NE. 0) THEN
WR1TEUUNOUT, 2009) IOST
CALL EXIT
END IF
103 CONTINUE
SUBROUTINE ORSPBC
ERROR CHECK #1:
An iteration over the number of species is made. For each iteration, the following takes place:
• A call to the function INDEXI occurs. SPNAME(ISPC) and SPNMBC are passed into INDEX1.
(SPNMBC is an array containing the species name in the BCON list, obtained from the 'BCON
header. SPNAME is an array containing the list of species names for the model.)
The position of each of the species names of the model is searched for in the list of species names
from the BCON list of species names. This positional value is assigned to the variable SPCNUM.
If SPCNUM equals zero, then the species name was not found in the BCON list of species names.
A message is written to the log indicating that an error has occurred.
F-39
-------
The error message contains the subroutine name, the current species name that is being considered (SPNA-
ME(ISPC)), the models species name (SPNMIN(ISPC)), and the value of SPCNUM (which should be zero).
The program exits by a call to the system subroutine EXIT.
CODE FOR ERROR CHECK #1:
c
00 101 ISPC = 1, NSPCIN
C
C check that model species name is in BOON species list
C and get model list species' name position in BCON list
C
SPCNUM = INDEX1 (SPNANE(ISPC), NSPCBC, SPNNBC)
If (SPCNUM .EO. 0) THEN
WRITE(LUNOUT, 2001) SPNAME(ISPC)
2001 FORMAT(/ 5X, 'XXX ERROR IN ORSPBC' /
& / 5X, 'SPECIES', 2X, A4, 2X,
& 'IS NOT PRESENT ON BCON FILE --JOB ABORTED')
WRITECLUNOUT, 1003) SPNMIN(ISPC), SPNAME(ISPC), SPCNUM
1003 FORMAT(6X, A4, 8X, A4, 7X, 14.2)
CALL EXIT
END IF
101 CONTINUE
SUBROUTINE ORSPBM
ERROR CHECK #1:
An iteration over the number of species is made. For each iteration the following takes place:
• A call to the function INDEXI occurs. SPNAME(ISPC) and SPNMBM are passed into INDEX1.
(SPNMBM is an array containing the species name in the BMAT list, obtained from the BMAT
header. SPNAME is an array containing the list of species names, for the model.)
• The position of each of the species names of the model is searched for in the list of species names
from the BMAT list of species names. This positional value is assigned to the variable SPCNUM.
• If SPCNUM equals zero, then the species name was not found in the BMAT list of species names.
A message is written to the log to indicate that an error has occurred.
The error message contains the subroutine name, the current species name that is being considered (SPNA-
ME(ISPC)), the models species name (SPNMIN(ISPC)), and the value of SPCNUM (which should be zero).
The program exits by a call to the system subroutine EXIT.
F-40
-------
CODE FOR ERROR CHECK #1:
c
DO 101 ISPC = 1, NSPCIN
C
C check that model species name is in BMATRIX species list
C and get model list species' name position in BMATRIX list
C
SPCNUM = INOEX1 (SPNAME(ISPC), NSPCBM, SPNMBM)
IF (SPCNUM .EQ. 0) THEN
URITECLUNOUT, 2001} SPNAME(ISPC)
2001 FORMAT(/ 5X, 'XXX ERROR IN ORSPBM1
& / 5X, 'SPECIES'. 2X, A4. 2X,
& 'IS NOT PRESENT ON BMATRIX FILE-JOB ABORTED')
WRITECLUNOUT, 1003) SPNMIN(ISPC), SPNAME(ISPC), SPCNUM
1003 FORMAT(6X, A4, 8X, A4, 7X, 14.2)
CALL EXIT
END IF
C
101 CONTINUE
SUBROUTINE ORSPIC
ERROR CHECK #1:
An iteration over the number of species is made. For each iteration the following takes place:
. A call to the function INDEXI occurs. SPNAME(ISPC) and SPNMIC are passed into INDEX1.
(SPNMIC is an array containing the species name in the ICON list, obtained from the ICON
header. SPNAME is an array containing the list of species names for the model.)
The position of each of the species names of the model is searched for in the list of species names
from the ICON list of species names. This positional value is assigned to the variable SPCNUM.
If SPCNUM equals zero, then the species name was not found in the ICON list of species names.
A message is written to the log to indicate that an error has occurred.
The error message contains the subroutine name, the current species name that is being considered (SPNA-
ME(ISPC)), the models species name (SPNMIN(ISPC)), and the value of SPCNUM (which should be zero).
The program exits by a call to the system subroutine EXIT.
F-41
-------
CODE FOR ERROR CHECK #1:
c
DO 101 ISPC = 1, NSPCIN
C check that model species name is in ICON species list
C and get model list species' name position in ICON list
C
SPCNUM = INDEX1 (SPNAME(ISPC), NSPCIC, SPNMIC)
If (SPCNUM .EQ. 0) THEN
URITEaUNOUT, 2001) SPNANE(ISPC)
2001 FORMAT(/ 5X, 'XXX ERROR IN ORSPIC1 /
& / 5X, 'SPECIES', 2X, A4, 2X,
& 'IS NOT PRESENT ON ICON FILE-JOB ABORTED1}
URITE(LUNOUT, 1003) SPNNIN(ISPC), SPNAME(ISPC), SPCNUM
1003 FORMAT(6X, A4, 8X. A4, 7X, K.2)
CALL EXIT
END IF
SUBROUTINE POBCON
ERROR CHECK #1:
A call to subroutine CLOCKI is made to obtain the file process elapsed time step. The first time step
(FRSTBC) is compared with the elapsed time step (IELPBC) obtained from the call to CLOCKI. If the times
do not match, then an error has occurred. The error message contains the subroutine in which the error
occurred, the expected time, the time read from the time step header, the scenario start time, the elapsed time,
the step number, and the time step size. The program exits by a call to the system subroutine EXIT.
CODE FOR ERROR CHECK #1:
. c
C compute intervals and records to be skipped
C
CALL CLOCK1 (IDATE, ITINE, SDATBC, STHRBC, TSTPBC,
& IELPBC, ISTPBC)
C
IF (IELPBC .LT. FRSTBC) THEN
URITEUUNOUT, 2001) IDATE, I TIME, SDATBC, STHRBC,
& IEUPBC, ISTPBC, TSTPBC
2001 FORMAT(/ 5X, 'XXX DATE/TINE PRECEDES FIRST DATE/TINE ON ',
i'BCON FILE IN POBCON1
& / 5X, 'REQUESTED DATE/TIME: •, 15, 5X, 16
& / 5X, 'SCENARIO START: ', 15. 3X, 14
& / 5X, 'ELAPSED TIME: ', 17, 2X, 'STEP NUMBER: ', 13
& / 5X, 'TIME STEP SIZE: ', 14 /)
CALL EXIT
END IF
ERROR CHECK #2:
A call to FSKIPI is made in order to skip the appropriate amount of records. The parameter IOST is passed to
FSKIPI. Upon return from the subroutine FSKIPI, IOST contains the I/O status value of the read operation.
IOST is tested; if IOST is equal to zero, then an end-of-file marker was reached while reading records on the
F-42
-------
BCON file. The error message contains the subroutine in which the error occurred, the requested time, the
scenario start time, the elapsed time, the step number, the I/O status value and the time step size. The pro-
gram exits by a call to the system subroutine EXIT.
CODE FOR ERROR CHECK #2:
c
C and skip forward -
C
CALL FSKIP1 (UNITBC. FMTD, SKIPDR, NSKIP, RECTOS, SKIPNO,
& IOST)
IF (IOST .HE. 0) THEN
URITEUUNOUT, 2003} I DATE, ITIME, SOATBC, STHRSC,
& IELPBC, ISTPBC, IOST, TSTPBC
2003 FORMAT(/ 5X, 'XXX ERROR ENCOUNTERED ON BCON FILE1,
& ' BEFORE DATE/TIME REACHED IN POBCON1
& / 5X, 'REQUESTED DATE/TIME: ', 15, 5X, 16
& / 5X, 'SCENARIO START: ', 15, 3X, 14
& / 5X, 'ELAPSED TIME: ', 17, 2X, 'STEP NUMBER: ', 13
& / 5X, 'IOSTAT: ', 18, 4X, 'TIME STEP SIZE: ', 14 /)
CALL EXIT
END IF
SUBROUTINE POBTRK
ERROR CHECK #1:
A call to subroutine CLOCKI is made to obtain the file process elapsed time step. The first time step
(FRSTBT) is compared with the elapsed time step (IELPBT) obtained from the call to CLOCKI. If the times
do not match, then an error has occurred. A message is written to the log to indicate that an error has
occurred. The error message contains the subroutine in which the error occurred, the expected time, the time
read from the time step header, the scenario start time, the elapsed time, the step number, and the time step
size. The program exits by a call to the system subroutine EXIT.
CODE FOR ERROR CHECK #1:
c
C determine and validate step nunber
c
CALL CLOCK1 (IDATE, ITIME, SOATBT, STHRBT, TSTPBT,
& IELPBT, ISTPBT)
C
IF (IELPBT .LT. FRSTBT) THEN
URITEaUNOUT, 2001) IDATE, ITIME, SDATBT, STHRBT,
& IELPBT, ISTPBT, TSTPBT
2001 FORMAT(/ 5X. 'XXX DATE/TIME PRECEDES FIRST DATE/TIME ON ',
&'BTRK FILE IN POBTRK'
& / 5X, 'REQUESTED DATE/TIME: ', IS, 5X, 16
& / 5X, 'SCENARIO START: ', 15, 3X, 14
& / 5X, 'ELAPSED TIME: ', 17, 2X, 'STEP NUMBER: ', 13
& / 5X, 'TIME STEP SIZE: ', 14 /)
CALL EXIT
END IF
F-43
-------
ERROR CHECK #2:
To skip the appropriate amount of records an iteration over the number of records is done. For each iteration
a call to RDBT is made. The parameter IOST is passed to RDBT. Upon return from subroutine RDBT, IOST
contains the I/O status value of a read operation. If IOST is not equal to zero, then an error occurred while
reading records on the BTRK file. A message is written to the log to indicate an error has occurred. The error
message contains the subroutine in which the error occurred, the expected time, the time read from the time
step header, the scenario start time, the elapsed time, the step number, and the time step size. The program
exits by a call to the system subroutine EXIT.
CODE FOR ERROR CHECK #2:
c
DO 101 ISKIP = 1, NSKIP
CALL RDBT (1, ADUM, IOST)
IF (IOST .NE. 0) THEN
WRITE(LUNOUT. 2003) IDATE, ITIME, SDATBT, STHRBT,
& IELPBT, ISTPBT, IOST, TSTPBT
2003 FORMAT(/ 5X, 'XXX ERROR ENCOUNTERED ON BTRK FILE1,
& ' BEFORE DATE/TIME REACHED IN POBTRK1
& / 5X, 'REQUESTED DATE/TIME: ', 15, 5X, 16
& / 5X, 'SCENARIO START: ', IS, 3X, K
& / 5X, 'ELAPSED TIME: ', 17,
& 2X, 'STEP NUMBER: ', 13
& / 5X, 'IOSTAT: ', 18,
& 4X, 'TIME STEP SIZE: ', 14 /)
CALL EXIT
END IF
101 CONTINUE
SUBROUTINE POCONC
ERROR CHECK #1:
A call to subroutine CLOCKI is made to obtain the file process elapsed time step. The first time step
(FRSTCN) is compared with the elapsed time step (IELPCN) obtained from the call to CLOCKI. If the times
do not match, then an error has occurred. A message is written to the log to indicate that an error has
occurred. The error message contains the subroutine in which the error occurred, the expected time, the time
read from the time step header, the scenario start time, the elapsed time, the step number, and the time step
size. The program exits by a call to the system subroutine EXIT.
F-44
-------
CODE FOR ERROR CHECK #1:
c
C compute step number
C
CALL CLOCK1 (IDATE, ITINE, SDATCN, STHRCN, TSTPCN,
& IELPCN, ISTPCN)
C
IF (IELPCN .LT. FRSTCN) THEN
WRITEUUNOUT, 2001) I DATE, ITIME, SDATCN, STHRCN,
& IELPCN, ISTPCN, TSTPCN
2001 FORMAT(/ 5X, 'XXX DATE/TIME PRECEDES FIRST DATE/TIME ON ',
i'CONC FILE IN POCONC'
& / 5X, 'REQUESTED DATE/TIME: ', 15, 5X, 16
& / 5X, 'SCENARIO START: ', 15, 3X, 14
& / 5X, 'ELAPSED TIME: ', 17, 2X, 'STEP NUMBER: ', 13
& / 5X, 'TIME STEP SIZE: ', 14 /)
CALL EXIT
END IF
ERROR CHECK #2:
A call to FSKIPI is made in order to skip the appropriate amount of records. The parameter IOST is passed to
FSKIPI. Upon return from the subroutine FSKIPI, IOST contains the I/O status value of the read operation.
IOST is tested; if IOST is equal to zero, then an end-of-file marker was reached while reading records on the
CONC file. The error message contains the subroutine in which the error occurred, the requested time, the
scenario start time, the elapsed time, the step number, the I/O status value and the time step size. The pro-
gram exits by a call to the system subroutine EXIT.
CODE FOR ERROR CHECK #2:
. c
C and skip forward -
C
CALL FSKIP1 (UNITCN, FMTD, SKIPDR, NSKIP, RECPOS, SKIPNO,
& IOST)
IF (IOST .NE. 0) THEN
WRITE(LUNOU7, 2003) IDATE, ITIME, SDATCN, STHRCN,
& IELPCN, ISTPCN, IOST, TSTPCN
2003 FORMAT(/ 5X, 'XXX ERROR ENCOUNTERED ON CONC FILE',
& ' BEFORE DATE/TIME REACHED IN POCONC'
& / 5X, 'REQUESTED DATE/TIME: ', 15, 5X, 16
& / 5X, 'SCENARIO START: ', 15, 3X, 14
& / 5X, 'ELAPSED TIME: ', 17, 2X, 'STEP NUMBER: ',13
& / 5X, 'IOSTAT: ', 18, 4X, 'TIME STEP SIZE: ', 14 /)
CALL EXIT
END IF
SUBROUTINE POICON
ERROR CHECK #1:
F-45
-------
A call to subroutine CLOCKI is made to obtain the file process elapsed time step. The first time step
(FRSTIC) is compared with the elapsed time step (IELPIC) obtained from the call to CLOCKI. If the times do
not match, then an error has occurred. A message is written to the log to indicate that an error has occurred.
The error message contains the subroutine in which the error occurred, the expected time, the time read from
the time step header, the scenario start time, the elapsed time, the step number, and the time step size. The
program exits by a call to the system subroutine EXIT.
CODE FOR ERROR CHECK #1:
c
C compute step number
C
CALL CLOCK1 (IDATE, ITIME, SOAT1C, STHRIC, TSTPIC,
& IELPIC, ISTPIC)
C
IF (IELPIC .LT. FRSTIC) THEN
WRITECLUNOUT, 2001) IDATE, ITIME, SDATIC, STHRIC.
& IELPIC, ISTPIC, TSTPIC
2001 FORMAT(/ 5X, 'XXX DATE/TIME PRECEDES FIRST DATE/TIME ON ',
S'lCON FILE IN POICON'
& / 5X, 'REQUESTED DATE/TINE: ', IS, 5X, 16
& / 5X. 'SCENARIO START: ', 15, 3X, 14
& / 5X, 'ELAPSED TIME: ', 17, 2X, 'STEP NUMBER: ', 13
& / 5X, 'TIME STEP SIZE: ', 14 /)
CALL EXIT
END IF
ERROR CHECK #2:
A call to FSKIPI is made in order to skip the appropriate amount of records. The parameter IOST is passed to
FSKIPI. Upon return from the subroutine FSKIPI, IOST contains the I/O status value of the read operation.
IOST is tested; if IOST is equal to zero, then an end-of-file marker was reached while reading records on the
ICON file. The error message contains the subroutine in which the error occurred, the requested time, the
scenario start time, the elapsed time, the step number, the I/O status value and the time step size. The pro-
gram exits by a call to the system subroutine EXIT.
F-46
-------
CODE FOR ERROR CHECK #2:
c
C and skip forward -
C
CALL FSKIP1 (UNITIC, FHTD, SKIPDR, NSKIP, RECTOS, SKIPNO,
& IOST)
IF (IOST .WE. 0) THEN
VRITE(LUNOUT, 2003) IDATE, ITIME, SDATIC, STHRIC,
& IELPIC, ISTPIC, IOST, TSTPIC
2003 FORMAT(/ 5X, 'XXX ERROR ENCOUNTERED ON ICON FILE1,
& ' BEFORE DATE/TIME REACHED IN POICON'
& / 5X, 'REQUESTED DATE/TIME: ', IS. 5X, 16
& / 5X, 'SCENARIO START: ', IS, 3X, 14
& / 5X, 'ELAPSED TIME: ', 17, 2X, 'STEP NUMBER: ', 13
& / 5X, MOSTAT: ', 18, 4X, 'TIME STEP SIZE: ', 14 /)
CALL EXIT
END IF
SUBROUTINE POMXBM
ERROR CHECK #1:
A call to subroutine CLOCKI is made. The first time step (FRSTBM) is compared with the elapsed time step
(IELPBM) obtained from the call to CLOCKI. If the times do not match, then an error has occurred. A mes-
sage is written to the log to indicate that an error has occurred. The error message contains the subroutine in
which the error occurred, the expected time, the time read from the time step header, the scenario start time,
the elapsed time, the step number, and the time step size. The program exits by a call to the system subroutine
EXIT.
CODE FOR ERROR CHECK #1:
c
C get position step muter
C
CALL CLOCK1 (IDATE, ITIME, SDATBM, STHRBM, TSTPBM,
& IELPBM, ISTPBM)
C
IF (IELPBM .LT. FRSTBM) THEN
WRITE(LUNOUT, 2001) IDATE, ITIME, SDATBM, STHRBM,
& IELPBM, ISTPBM. TSTPBM
2001 FORMAT(/ 5X, 'XXX DATE/TIME PRECEDES FIRST DATE/TIME ON '.
&'BMAT FILE IN POMXBM'
& / 5X, 'REQUESTED DATE/TIME: ', 15, 5X, 16
& / 5X, 'SCENARIO START: ', IS, 3X, 14
& / 5X, 'ELAPSED TIME: ', 17, 2X, 'STEP NUMBER: ', 13
& / 5X, 'TIME STEP SIZE: ', 14 /)
CALL EXIT
END IF
ERROR CHECK #2:
F-47
-------
The appropriate amount of records are skipped by performing an IF test. For each test, a call to subroutine
RDMXBM is made. The parameter IOST is passed to RDMXBM. Upon return from subroutine RDMXBM, IOST
contains the I/O status of a read of the BMAT file. A test of the correct subfile is made. If this test fails, then
IOST is tested; if IOST is less than zero, then the end-of-file marker is reached on the BMAT file. If this test
passes, then IOST is tested, and if not equal to zero, then an error occurred while reading the BMAT file. An
error message is written to the log indicating that an error has occurred. The error message contains the
subroutine name, the requested time, the scenario start time, the elapsed time, the step number, the I/O status
value, and the time step size. The program exits by a call to the system subroutine EXIT.
CODE FOR ERROR CHECK #2:
c
CALL RDMXBM (1, ADUM, IOST)
IF ( ISUBFL .NE. JSUBFL) THEN
IF (IOST .LT. 0) THEN
WRITE(LUNOUT, 2003} ' EOF1, IDATE, ITIME, SOATBM,
& STHRBH, IELPBM, ISTPBM, IOST, TSTPBM
2003 FORMAT(/ 5X, 'XXX1, A6, ' ENCOUNTERED ON BMAT FILE1
& ' BEFORE DATE/TIME REACHED IN POMXBM'
& / 5X, 'REQUESTED DATE/TIME: '. 15, 5X, 16
& / 5X, 'SCENARIO START: ', 15, 3X, 14
& / 5X, 'ELAPSED TIME: ', 17. 2X, 'STEP NUMBER: ', 13
& / 5X, 'IOSTAT: ', 18, 4X, 'TIME STEP SIZE: ', 14 /)
CALL EXIT
ELSE
ELSE IF (IOST .NE. 0) THEN
URITE(LUNOUT, 2003) ' ERROR', IDATE. ITIME, SDATBM, STHRBM,
& IELPBM, ISTPBM, IOST, TSTPBM
CALL EXIT
END IF
SUBROUTINE POSTAV
ERROR CHECK #1:
A call to subroutine CLOCKI is made to obtain the elapsed time (IELPSV). This time is tested against the first
time step (FRSTSV). If the elapsed time step is less than the first time step, then an error has occurred. An
error message is written to the log indicating that an error has occurred. The error message contains the
subroutine in which the error occurred. The program exits by a call to the system subroutine EXIT.
F-48
-------
CODE FOR ERROR CHECK #1:
c
C compute step
C
CALL CLOCK1 (IDATE, ITIHE, SDATSV, STHRSV, TSTPSV,
& IELPSV, ISTPSV)
C
IF (IELPSV .LT. FRSTSV) THEN
WRITE(LUNOUT, 2201)
2201 FORMAT(/ 5X, 'XXX DATE/TIME PRECEDES FIRST DATE/TIME ON
S'STATE VECTOR IN POSTAV)
CALL EXIT
END IF
ERROR CHECK #2:
To skip records on the STATE VECTOR file, an iteration is made. For each iteration, a formatted read from
the STATE VECTOR file is done, and the I/O status is saved in the variable IOST. IOST is tested after each
read; if IOST does not equal zero, then an error has occurred. A message is written to the log indicating that
an error has occurred. The message contains the subroutine name, the unit number, the current record being
read, and the I/O status value. The program exits by a call to the system subroutine EXIT.
CODE FOR ERROR CHECK #2:
DO 101 IREC = 1, RECNSV
READ(UNIT = UNITSV, FMT = 1001, IOSTAT = IOST) DUMBUF
1001 FORMAT(A80)
IF (IOST .NE. 0) THEN
WRITE(LUNOUT, 2003) UNITSV. IREC, DUMBUF, IOST
2003 FORMAT(/ 5X, 'XXX READ ERROR IN POSTAV
& / 5X, 'UNIT NUMBER = ', 12, 5X, 'RECORD = ', 14
& / 5X, A80
& / 5X, 'I/O STATUS = ', 14)
CALL EXIT
END IF
101 CONTINUE
SUBROUTINE PQ1
ERROR CHECK #1:
For each iteration over each species, each chemistry component (GPR) is tested to see if it is less than or
equal to zero. At the same time, each ADTVF component is also tested to see if it is less than or equal to
zero. If either test fails, then an error has occurred. A message is written to the log. The message contains the
subroutine name, as well as some chemistry reaction and rate constant values. IOST is then set to have a value
of 1, and control is returned to the calling subroutine.
F-49
-------
CODE FOR ERROR CHECK #1:
DO 201 ISPC = 1, NSPCIN
IF (GPR(ISPC,LEV) .LE. 0.0 .OR. ADTVFUSPC) .LE. 0.0) THEN
WRITE(LUNOUT, 2001) LEV
2001 FORMAT(/ 5X, 'XXX ERROR REPORTED FROM PQ1 FOR LEVEL1, 12,
& 2X, '- ZERO OR NEGATIVE CONG COMPONENT VALUES:',
& / 5X, 'SPEC --- CPR(SPEC) ---
& GTI(SPEC) --- CGA(SPEC)1)
DO 101 JSPC = 1, NSPCIN
WRITE(LUNOUT, 2003) SPNAME(JSPC), GPR(JSPC,LEV),
& GTI(JSPC,LEV), CGA(JSPC,LEV)
2003 FORMATC/ 5X. A4. 2X, 3(2X, 1PE12.5))
101 CONTINUE
IOST = 1
RETURN
END IF
201 CONTINUE
SUBROUTINE PQCOEF
NONE
SUBROUTINE PRGSMY
ERROR CHECK #1:
A call to subroutine ADATE is made to obtain the run date and run time. A formatted read from the buffer
extracts the date and time. IOST contains the I/O status of this read operation. If IOST is not equal to zero,
then an error has occurred. A message is written to the log indicating that an error has occurred. The error
message contains the subroutine name, the run date and run time extracted from the call to ADATE, the entire
character string from ADATE, and the I/O status value. The program exits by a call to the system subroutine
EXIT.
CODE FOR ERROR CHECK #1:
c
C get run date, time
C
CALL ADATE (CDT(1), CDT(2), DUMSTR)
READ(STRING, 1001, IOSTAT - IOST) RUNDTE, RUNTIM
1001 FORMAT(I6, 2X, 16)
IF (IOST .NE. 0) THEN
IOST = IOCL (IOST)
URITE(LUNOUT, 2001) RUNDTE, RUNTIM, STRING, IOST
2001 FORMAT(/ 5X, 'XXX ERROR ABORT IN PRGSMY -'
& ' UNABLE TO DECODE DATE/TIME1
& / SX, 'RUNDTE = ', 18, 2X, 'RUNTIM = ', 18
& / 5X, 'STRING = ', A16, 2X, MOST = ', 16)
CALL EXIT
END IF
F-50
-------
SUBROUTINE RATED
NONE
SUBROUTINE RDBCON
ERROR CHECK #1:
The BCON file is opened by a call to OPBCON, and the common block HEADBC is loaded. HEADBC con-
tains the header information for the BCON file. The header information is then tested in the subroutine
RDBCON. If any of the parameters fail the test, then an error has occurred. A message is written to the log to
indicate that an error has occurred. The message contains the subroutine name and the BCON header infor-
mation. The program exits by a call to the system subroutine EXIT.
CODE FOR ERROR CHECK #1:
C open BCON file
CALL OPBCON
C
C check header parameters
IF ((TSTPBC .NE. TSTPIN) .OR. (GRDNBC .NE. GRDNIN) .OR.
& (ABSCSWLNBC - SWLMIN) .GT. 0.001) .OR.
2001
&
&
&
&
&
&
&
&
&
&
&
&
&
&
&
£
&
WRITEaUNOUT, 2001) TSTPBC, TSTPIN, GRDNBC.
SWLNBC, SWLNIN, SWLTBC, SULTIN,
NELNBC, NELNIN, NELTBC, NELTIN,
DLONBC, DLONIN, DLATBC, DLATIN,
NCOLBC, NCOLIN, NROUBC, NROWIN,
NLEVBC, NLEVIN, NSPCBC, NSPCIN
FORMAT (/ 5X, 'XXX PARAMETER CHECK FAILURE IN
/ 5X, 'TSTPBC
/ 5X, 'GRDNBC
/ 5X, 'NELNBC
/ SX, 'NELTBC
/ SX, 'SWLNBC
/ 5X, 'SULTBC
/ 5X, 'DLONBC
/ 5X, 'DLATBC
/ 5X, 'NCOLBC
/ SX, 'NROUBC
/ SX. 'NLEVBC
/ SX, 'NSPCBC
IS, 5X, 'TSTPIN =
A8, SX, 'GRDNIN =
F10.5, SX, 'NELNIN
F10.S, SX, 'NELTIN
F10.5, SX, 'SULNIN
F10.5, SX, 'SWLTIN
F10.5, SX, 'DLONIN
F10.S, SX, 'DLATIN
13, SX, 'NCOLIN =
13, SX, 'NROWIN =
13, SX, 'NLEVIN =
13, SX, 'NSPCIN =
CALL EXIT
GRDNIN,
RDBCON'
', 15
', AS
F10.5
F10.5
F10.5
F10.5
F10.5
F10.5
', 13
'. 13
'. 13
'. 13 )
END IF
ERROR CHECK #2:
The BCON time step header is read by calling the subroutine RDFILE. The unit number, number of words to
be read, starting address of the common block RTSHBC, and IOST are passed to RDFILE. Upon return from
the call to RDFILE, IOST contains the I/O status of the read operation of the BCON time step header. IOST
F-51
-------
is tested; if IOST is less than zero, then an end-of-file marker was reached. Control is returned to the calling
subroutine. If IOST is not equal to zero, then an error occurred on the read operation. A message is written
to the log to indicate that an error occurred. The message contains the subroutine name, the number of words
to read, the value of the I/O status, the time step date and time. The program exits by a call to the system
subroutine EXIT.
CODE FOR ERROR CHECK #2:
c
C read BCON T.S.H.
C
CALL RDFILE (UNITBC, NUDTSH, DATBC, IOST)
C
IF (IOST .LT. 0) RETURN
IF (IOST .NE. 0) THEN
URITE(LUNOUT, 2003) NUDTSH, IOST, BCOATE, BCTIME
2003 FORMAT(/ 5X, 'XXX T.S.H. READ ERROR IN RDBCON1
& / 5X, 'NO. OF WORDS = ', 12, 2X, 'I/O STATUS = ', 14,
&2X, 'BCDATE = ', 16.6, 2X, 'BCTIME = ', 16.6)
CALL EXIT
END IF
ERROR CHECK #3:
The current model time step time and date are compared with the time step time and date obtained from
reading the BCON time step header. If either the time or date do not match, then an error has occurred. A
message is written to the log to indicate that an error has occurred. The message contains the subroutine
name, date and time from the time step header, and the model date and time. The program exits by a call to
the system subroutine EXIT.
CODE FOR ERROR CHECK #3:
c
IF (BCOATE .NE. IDATBC .OR. BCTIME .NE. ITIMBC) THEN
WRITE(LUNOUT, 2005) IDATBC, ITIMBC, BCDATE, BCTIME
2005 FORMAT(/ SX. 'XXX DATES/TINES DO NOT MATCH IN RDBCON'
& / SX, 'IDATBC - ', 16.6, 2X, •ITIMBC = ', 16.6
& / 5X, 'BCDATE = ', 16.6, 2X, 'BCTIME = ', 16.6)
CALL EXIT
END IF
ERROR CHECK #4:
The western boundary conditions are read by iterating over the number of species. For each iteration, a call
to RDFILE is made. The unit number, the number of words to read, the starting address of the common block
BCFILE, and IOST are passed to RDFILE. Upon return from RDFILE, IOST contains the I/O status of the
F-52
-------
read operation. If IOST is not equal to zero, then an error has occurred. A message is written to the log to
indicate that an error has occurred. The message contains the subroutine name, the unit number, the I/O
status value, and the number of words. The program exits by a call to the system subroutine EXIT.
CODE FOR ERROR CHECK #4:
c
C read Western boundary conditions species*
C
DO 201 ISPC = 1, NSPECS
CALL RDFILE (UNITBC, ROUURO, WEST(1,1,ISPC), IOST)
IF (IOST -ME. 0) THEN
URITE(LUNOUT, 2007) UNITBC, IOST, ROUWRD
2007 FORMAT(/ 5X, 'XXX READ ERROR IN RDBCON1
& / 5X, 'UNIT = ', 13, 2X, 'I/O STATUS = ', 13,
& 2X, 'NO. OF WORDS = ', 14)
CALL EXIT
END IF
201 CONTINUE
ERROR CHECK #5:
The eastern boundary conditions are read by iterating over the number of species. For each iteration, a call to
RDFILE is made. The unit number, the number of words to read, the starting address of the common block
BCFILE, and IOST are passed to RDFILE. Upon return from RDFILE, IOST contains the I/O status of the
read operation. If IOST is not equal to zero, then an error has occurred. A message is written to the log to
indicate that an error has occurred. The message contains the subroutine name, the unit number, the I/O
status value, and the number of words. The program exits by a call to the system subroutine EXIT.
CODE FOR ERROR CHECK #5:
c
C read Eastern boundary conditions species*
C
DO 301 ISPC = 1, NSPECS
CALL RDFILE (UNITBC, ROWWRD, EAST(1,1,ISPC), IOST)
IF (IOST .NE. 0) THEN
WRITE(LUNOUT, 2007) UNITBC, IOST, ROWWRD
CALL EXIT
END IF
301 CONTINUE
ERROR CHECK #6:
The northern boundary conditions are read by iterating over the number of species. For each iteration, a call
to RDFILE is made. The unit number, the number of words to read, the starting address of the common block
BCFILE, and IOST are passed to RDFILE. Upon return from RDFILE, IOST contains the I/O status of the
F-53
-------
read operation. If IOST is not equal to zero, then an error has occurred. A message is written to the log to
indicate that an error has occurred. The message contains the subroutine name, the unit number, the I/O
status value, and the number of words. The program exits by a call to the system subroutine EXIT.
CODE FOR ERROR CHECK #6:
c
C read Northern boundary conditions species*
C
DO 401 ISPC = 1, NSPECS
CALL RDFILE (UNITBC, COLURD, NORTH(1.1,ISPC), IOST)
IF (IOST .NE. 0) THEN
WRITECLUNOUT. 2007) UNITBC, IOST, COLWRD
CALL EXIT
END IF
401 CONTINUE
ERROR CHECK #7:
The southern boundary conditions are read by iterating over the number of species. For each iteration, a call
to RDFILE is made. The unit number, the number of words to read, the starting address of the common block
BCFILE, and IOST are passed to RDFILE. Upon return from RDFILE, IOST contains the I/O status of the
read operation. If IOST is not equal to zero, then an error has occurred. A message is written to the log to
indicate that an error has occurred. The message contains the subroutine name, the unit number, the I/O
status value, and the number of words. The program exits by a call to the system subroutine EXIT.
CODE FOR ERROR CHECK #7:
c
C read Southern boundary conditions species* •
C
DO 501 ISPC = 1, NSPECS
CALL RDFILE (UNITBC, COLWRD, SOUTH(1,1,ISPC), IOST}
IF (IOST .NE. 0) THEN
URITE(LUNOUT, 2007) UNITBC, IOST, COLURD
CALL EXIT
END IF
501 CONTINUE
SUBROUTINE RDBMAT
ERROR CHECK #1:
The BMAT file is opened by a call to OPBMAT, and the common block HEADBM is loaded. HEADBM
contains the header information for the BMAT file. The header information is then tested in the subroutine
RDBMAT. If any of the parameters fail the test, then an error has occurred. A message is written to the log to
indicate that an error has occurred. The message contains the subroutine name and the BMAT header
information. The program exits by a call to the system subroutine EXIT.
F-54
-------
CODE FOR ERROR CHECK #1:
C open BMAT file
CALL OPBMAT
C
C check parameters in headers
IF ((TSTPBM .NE. TSTPIN) .OR. (GRDNBM .NE. GRDNIN) .OR.
& (ABS(SWLNBM - SWLNIN) .GT. 0.001) .OR.
& (ABS(SWLTBM - SWLTIN) .GT. 0.001) .OR.
2001 FORMAT(/ 5X, 'XXX PARAMETER CHECK FAILURE IN RDBMAT1
15, 5X, 'TSTPIN = ', 15
AS, 5X, 'GRDNIN = ', A8
F10.5, 5X, 'NELNIN = F10.5
F10.5, 5X, 'NELTIN = F10.5
F10.5, 5X, 'SWLNIN = F10.5
F10.5, 5X, 'SWLTIN = F10.5
F10.5, 5X, 'DLONIN = F10.5
F10.5, 5X, 'DLATIN = F10.5
13, 5X, 'NCOLIN = ', 13
13, 5X, 'NROWIN * ', 13
13, 5X, 'NLEVIN = ', 13
13, 5X, 'NSPCIN « ', 13 )
&
&
&
&
&
&
&
&
&
&
&
&
CALL
END IF
/ 5X,
/ 5X,
/ 5X,
/ 5X,
/ 5X,
/5X,
/5X,
/ 5X.
/ 5X,
/ 5X,
/ 5X,
/ 5X.
EXIT
'TSTPBM =
'GRDNBM =
'NELNBM =
'NELTBM =
'SWLNBM =
'SWLTBM =
'DLONBM =
'DLATBM =
'NCOLBM =
'NROWBM =
'NLEVBM &
'NSPCBM s
ERROR CHECK #2:
The BMAT time step header is read by calling the subroutine RDMXBM. The number of words to be read,
starting address of the common block RTSHBM, and IOST are passed to RDMXBM. Upon return from the
call to RDMXBM, IOST contains the I/O status of the read operation of the time step header. IOST is tested;
If IOST is less than zero, then an end-of-file marker was reached. Control is returned to the calling subrou-
tine. If IOST is not equal to zero, then an error occurred on the read of the BMAT time step header. A
message is written to the log indicating that an error occurred. The message contains the subroutine name,
the number of words to read, the value of the I/O status, the time step date and time. The program exits by a
call to the system subroutine EXIT.
CODE FOR ERROR CHECK #2:
C read BMAT T.S.N.
CALL RDMXBM (NUDTSH, DATBM, IOST)
C
IF (IOST .LT. 0) RETURN
IF (IOST .NE. 0) THEN
URITEUUNOUT, 2003) NUDTSH, IOST, BMDATE, BMTIME
2003 FORMAT(/ 5X, 'XXX T.S.H. READ ERROR IN RDBMAT'
& / 5X, 'NO. OF WORDS = ', 12, 2X, 'I/O STATUS = ', 14,
&2X, 'BMDATE = ', 16.6, 2X, 'BMTIME = ', 16.6)
CALL EXIT
END IF
ERROR CHECK #3:
F-55
-------
The current model time step time and date are compared with the time step time and date obtained from
reading the BMAT time step header. If the time or date do not match, then an error has occurred. A message
is written to the log to indicate that an error has occurred. The message contains the subroutine name, date
and time from the time step header, and the model date and time. The program exits by a call to the system
subroutine EXIT.
CODE FOR ERROR CHECK #3:
c
IF (IDATBM .ME. BMOATE .OR. ITIMBM .NE. BMTIME) THEN
WRITE(LUNOUT, 2005) IDATBM, ITIMBM, BMDATE, BMTIME
2005 FORMAT(/ 5X, 'XXX DATES/TIMES DO NOT MATCH IN RDBMAT1
& / 5X, 'IDATBM = ', 16.6, 2X, 'ITIMBM = ', 16.6
& / 5X, 'BMDATE = ', 16.6, 2X, 'BMTIME = ', 16.6)
CALL EXIT
END IF
ERROR CHECK #4:
Part 1 of the BMAT file is read by calling subroutine RDMXBM. The number of words to read, the starting
address of the common block BMFILE, and IOST are passed to RDMXBM. Upon return from RDMXBM, IOST
contains the I/O status of the read operation. If IOST is not equal to zero, then an error has occurred. A
message is written to the log to indicate that an error has occurred. The message contains the subroutine
name, the I/O status value, the number of words, the model step date and time, and the row. The program
exits by a call to the system subroutine EXIT.
CODE FOR ERROR CHECK #4:
c
C read BMAT part 1
CALL RDMXBM (NUDBM1, XB12, IOST)
IF (IOST .NE. 0) THEN
WRITE(LUNOUT, 2007) IOST, NUDBM1, BMOATE. BMTIME, RDBMRU
2007 FORMAT(/ 5X, 'XXX FILE READ ERROR IN PART 1 IN RDBMAT1
& / 5X, 'I/O STATUS - ', K, 2X, 'NO. OF WORDS - ', 16
& / 5X, 'BMDATE = ', 16.6, 2X, 'BMTIME = ', 16.6,
&2X, 'ROW = '. 13)
CALL EXIT
END IF
ERROR CHECK #5:
Part 2 of the BMAT file is read by iterating over the number of reduced species. For each iteration, a call to
RDMXBM is made. The number of words to read, the starting address of the common block BMFILE, and
IOST are passed to RDMXBM. Upon return from RDMXBM, IOST contains the I/O status of the read opera-
tion. If IOST is not equal to zero, then an error has occurred. A message is written to the log to indicate that
F-56
-------
an error has occurred. The message contains the subroutine name, the I/O status value, the number of words,
the time step date and time, the row, and the species number. The program exits by a call to the system
subroutine EXIT.
CODE FOR ERROR CHECK #5:
c
C read BMAT part 2 (reduced species list)
DO 301 ISPC = 1, BMSPRD
CALL RDHXBM (NUDBM2, AA1(1,1,ISPC), IOST)
IF (IOST .NE. 0) THEN
URITE(LUNOUT, 2009) IOST, NUDBM2, BMDATE, BMTIME,
& RDBMRU, ISPC
2009 FORMAT(/ 5X, 'XXX READ ERROR ON PART 2 IN RDBMAT1
& / 5X, 'I/O STATUS = ', 14
& / 5X, 'NO. OF WORDS = ', 13
& / 5X, 'BMDATE = ', 16.6, 5X, 'BMTIME = ', 16.6
& / 5X, 'ROW = ', 12, 5X, 'SPECIES NUMBER = ', 12)
CALL EXIT
END IF
301 CONTINUE
SUBROUTINE RDBT
ERROR CHECK #1:
RDBT is called by RDBTRK to read the header record. In RDBT, the header record is read from the BTRK file
into the character buffer RECONR The I/O status value for the read operation is stored in the variable
IOST. IOST is tested; if IOST does not equal zero, an error has occurred on the read. A message is written to
the log to indicate that an error has occurred. The error message contains the subroutine name, the unit
number, the logical name of the file, the actual name of the file, the I/O status value, and the record length.
The program exits by a call to the system subroutine EXIT.
CODE FOR ERROR CHECK #1:
c
C read header record
C
READ(UNITBT, IOSTAT = IOST) RECONE
IF (IOST .NE. 0) THEN
INQUIRE (FILE = FLNMBT. NAME = EQNAME)
WRITEUUNOUT, 2001) UNITBT, FLNMBT, EQNAME. IOST, RECLEN,
& 'FIRST RECORD '
2001 FORMAT(/ 5X, 'XXX ERROR READING HEADER RECORD IN RDBT',
& / 5X, 'UNITBT = ', 12, 2X, 'FNAME = ', AS
& / 5X, 'EQNAME = ', A64
& / 5X, 'IOST = ', 14, 2X, 'RECLEN = ', 14, 2X, A16)
CALL EXIT
END IF
ERROR CHECK #2:
F-57
-------
A formatted read of the buffer containing the header information (RECONE) is made to convert the header
information into character or numeric type, and the common block HEADBT is loaded. IOST is tested; if
IOST does not equal zero, an error has occurred on the read. A message is written to the log to indicate that
an error has occurred. The error message contains the subroutine name, the unit number, the logical name of
the file, the actual name of the file, the I/O status value, and the record length. The program exits by a call to
the system subroutine EXIT.
CODE FOR ERROR CHECK #2:
c
C convert character to mixed character & numeric
C
READ(RECONE, 1001, IOSTAT = IOST)
& COATBT, CTIMBT, SOATBT, STHRBT, TSTPBT, FRSTBT,
& GRDNBT, SULNBT, SULTBT, NELNBT, NELTBT, DLONBT.
& DLATBT, NCOLBT, NROWBT, NMIFBT, ICNTBT
1001 FORMAT(6I8.8, AS. 4F8.3, 2F8.5, 414.4)
IF (IOST .HE. 0) THEN
INQUIRE (FILE = FLNMBT, NAME = EQNAME)
WRITE(LUNOUT, 2001) UNITBT, FLNMBT, EQNAME, IOST, RECLEN,
& 'INTERNAL READ #1'
CALL EXIT
END IF
ERROR CHECK #3:
The MIF data records are next read by iterating over the number of MIF data records. For each iteration, an
MIF record is read into the buffer RECMIF. The I/O status value from the read operation is stored in the
variable IOST. IOST is tested; if IOST does not equal zero, an error has occurred on the read. A message is
written to the log to indicate that an error has occurred. The error message contains the subroutine name, the
unit number, the logical name of the file, the actual name of the file, the I/O status value, and the record
length. The program exits by a call to the system subroutine EXIT.
CODE FOR ERROR CHECK #3:
c
C read the MIF data records
C
DO 101 IMIF = 1, NMIFBT
READ(UNITBT, IOSTAT = IOST) RECMIF
IF (IOST .NE. 0) THEN
INQUIRE (FILE = FLNMBT, NAME = EQNAME)
URITE(CHBUF, 1003) ' MIF - NO. ', IMIF
1003 FORMAT(A12, 14)
URITEUUNOUT, 2001) UNITBT, FLNMBT, EQNAME, IOST,
& RECLEN, CHBUF
CALL EXIT
END IF
ERROR CHECK #4:
F-58
-------
A formatted read of the buffer containing the MIF information (RECMIF) is made to convert the MIF
records to character and numeric data, and the common blocks HEADBT and CHARBT are loaded. IOST is
tested; if IOST does not equal zero, an error has occurred on the read. A message is written to the log to
indicate that an error has occurred. The error message contains the subroutine name, the unit number, the
logical name of the file, the actual name of the file, the I/O status value, and the record length. The program
exits by a call to the system subroutine EXIT.
CODE FOR ERROR CHECK #4:
READ(RECMIF, 1005, IOSTAT = IOST)
& MFNMBT(IMIF), CDMFBT(IMIF), CTMFBT(IMIF),
& UOMFBT(IMIF), UTMFBT(IHIF)
1005 FORMAKA12, 418.8)
IF (IOST .NE. 0) THEN
INQUIRE (FILE = FLNMBT, NAME = EQNAME)
URITE(CHBUF, 1003) 'INTERNAL READ #2>. IMIF
WRITEUUNOUT, 2001) UNITBT, FLNMBT, EQNAME, IOST,
& RECLEN, CHBUF
CALL EXIT
END IF
101 CONTINUE
ERROR CHECK #5:
The text records are next read by iterating over the number of text records. For each iteration, a text record is
read into the buffer RECTXT. The I/O status value from the read operation is stored in the variable IOST.
IOST is tested; if IOST does not equal zero, an error has occurred on the read. A message is written to the log
to indicate that an error has occurred. The error message contains the subroutine name, the unit number, the
logical name of the file, the actual name of the file, the I/O status value, and the record length. The program
exits by a call to the system subroutine EXIT.
CODE FOR ERROR CHECK #5:
c
C read header text records
C
DO 201 ITXT = 1. ICNTBT
READ (UNITBT, IOSTAT = IOST) RECTXT
IF (IOST .NE. 0) THEN
INQUIRE (FILE - FLNMBT, NAME = EQNAME)
URITE(LUNOUT, 2001) UNITBT, FLNMBT, EQNAME, IOST, RECLEN.
& 'TEXT RECORDS '
CALL EXIT
END IF
ERROR CHECK #6:
F-59
-------
A formatted read of the buffer containing the text records (RECTXT) is made to convert the text records to
character data, and the common block CHARBT is loaded. IOST is tested; if IOST does not equal zero, an
error has occurred on the read. A message is written to the log to indicate that an error has occurred. The
error message contains the subroutine name, the unit number, the logical name of the file, the actual name of
the file, the I/O status value, and the record length. The program exits by a call to the system subroutine EXIT.
CODE FOR ERROR CHECK #6:
READ(RECTXT, 1007, IOSTAT = IOST) TEXTBT(ITXT)
1007 FORMAT(ABO)
IF (IOST .NE. 0) THEN
INQUIRE (FILE = FLNHBT, NAME = EQNAME)
WRITE(LUNOUT, 2001) UNITBT, FLNNBT, EQNAME, IOST, RECLEN,
& 'INTERNAL READ <
URITE(LUNOUT, 1009) RECTXT
1009 FORMAT(5X, ABO)
CALL EXIT
END IF
ERROR CHECK #6:
Each row of the backtrack file is read. The I/O status value of the read is stored in the variable IOST. IOST
is tested; if IOST is less than zero, then the end-of-file marker is reached. An informational message is written
to the log. The backtrack file is closed and control returns to the calling subroutine. If IOST is not equal to
zero, then .an error has occurred on the read. A message is written to the log to indicate that an error has
occurred. The error message contains the subroutine name, the unit number, the file name, the logical file
name, the I/O status value, and the record length. The program exits by a call to the system subroutine EXIT.
F-60
-------
CODE FOR ERROR CHECK #6:
c
C read data record
C
READ (UNITBT, IOSTAT = IOST) DATA
RECNO = 1
C
C read body record*
C
IF (IOST .LT. 0) GO TO 401
IF (IOST .NE. 0) THEN
INQUIRE (FILE = FLNMBT, NAME = EQNAHE)
WRITE(LUNOUT, 2003) UNITBT, FLNMBT, EQNAME, IOST, RECNO
2003 FORMAT(/ 5X, 'XXX ERROR READING DATA RECORD IN RDBT1,
S5X, 'CONTROL RETURNED TO RDBTRK'
& / 5X. 'UNITBT = ', 12, 2X, 'FNAME = ', A8
& / 5X, 'EQNAME = ', A64
& / 5X, MOST = ', 12, 2X, 'RECNO = ',14)
RETURN
END IF
401 CONTINUE
WRITE(LUNOUT, 1011) RECNO, FLNMBT, UNITBT
1011 FORMAT(5X, 16, ' records read on ', A12, ' from unit ', 12 /)
C
CLOSE (UNIT = UNITBT)
SUBROUTINE RDBTRK
ERROR CHECK #1:
The BTRK file is opened by a call to OPBTRK, and the common block HEADBT is loaded. HEADBT con-
tains the header information for the BTRK file. The header information is then tested in the subroutine
RDBTRK. If any of the parameters fail the test, then an error has occurred. A message is written to the log to
indicate that an error has occurred. The message contains the subroutine name and the BTRK header infor-
mation. The program exits by a call to the system subroutine EXIT.
F-61
-------
CODE FOR ERROR CHECK #1:
C open BTRK file
CALL OPBTRK
C
C check parameters in headers
IF ((TSTPBT .HE. TSTPIN) .OR. (GRDNBT .NE. GRDN1N) .OR.
& (ABS(SULNBT - SWLNIN) .GT. 0.001) .OR.
& (ABS(SULTBT - SULTIN) .GT. 0.001) .OR.
URITE(LUNOUT, 2001) TSTPBT, TSTPIN, GRDNBT, GRDNIN,
2001
& SULNBT, SULNIN, SULTBT, SULTIN,
& NELNBT, NELNIN, NELTBT, NELTIN,
& DLONBT, DLONIN, DLATBT, DLATIN,
& NCOLBT, NCOLIN, NROUBT, NROWIN
FORMAT (/ 5X, 'XXX PARAMETER CHECK FAILURE IN RDBTRK1
& / 5X, 'TSTPBT =
& / 5X, 'GRDNBT =
& / 5X, 'NELNBT -
& / 5X, 'NELTBT =
& / 5X, 'SWLNBT =
& / 5X, 'SULTBT =
& / 5X, 'DLONBT =
& / 5X, 'DLATBT =
& / 5X, 'NCOLBT =
& / 5X, 'NROUBT -
15, 5X, 'TSTPIN = ', 15
A8, 5X, 'GRDNIN = ', AS
F10.5. 5X, 'NELNIN = F10.5
F10.5, 5X, 'NELTIN = F10.5
F10.5, 5X, 'SULNIN = F10.5
F10.5, 5X, 'SULTIN = F10.5
F10.5, 5X, 'DLONIN = F10.5
F10.5, 5X, 'DLATIN = F10.5
13, 5X, 'NCOLIN = ', 13
13, 5X, 'NROWIN = ', 13 /)
CALL EXIT
END IF
ERROR CHECK #2:
The BTRK time step header is read by calling the subroutine RDBT. The number of words to be read, starting
address of the common block RTSHBT, and IOST are passed to RDBT. Upon return from the call to RDBT,
IOST contains the I/O status of the read operation of the BTRK time step header. IOST is tested; if IOST is
less than zero, then an end-of-file marker was reached. Control is returned to the calling subroutine. If IOST
is not equal to zero, then an error occurred on the read operation. A message is written to the log to indicate
that an error occurred. The message contains the subroutine name, the number of words to read, the value of
the I/O status, the time step date and time. The program exits by a call to the system subroutine EXIT.
CODE FOR ERROR CHECK #2:
C read BTRK T.S.H.
CALL RDBT (IUDLN1, DATBT, IOST}
C
IF (IOST .LT. 0) RETURN
IF (IOST .NE. 0) THEN
URITE(LUNOUT, 2003) IUDLN1, IOST, BTDATE, BTTIME
2003 FORMAT(/ 5X, 'XXX T.S.H. READ ERROR IN RDBTRK'
& / 5X, 'UDLN1 = ', 12, 2X, 'I/O STATUS = ',
&2X, 'BTDATE = ', 16.6, 2X, 'BTTIME = ', 16.6)
CALL EXIT
END IF
I*.
F-62
-------
ERROR CHECK #3:
The current model time step time and date are compared with the time step time and date obtained from
reading the BTRK time step header. If the time or date do not match, then an error has occurred. A message
is written to the log to indicate that an error has occurred. The message contains the subroutine name, date
and time from the time step header, and the model date and time. The program exits by a call to the system
subroutine EXIT.
CODE FOR ERROR CHECK #3:
c
IF (IDATBT .NE. BTDATE .OR. ITIMBT .NE. BTTIME) THEN
WRITE(LUNOUT, 2005) IDATBT, ITIMBT, BTDATE, BTTIHE
2005 FORMAT(/ 5X, 'XXX DATES/TIMES DO NOT MATCH IN RDBTRK1
& / 5X, 'IDATBT = ', 16.6, 2X, 'ITIMBT = ', 16.6
i / 5X, 'BTDATE = ', 16.6, 2X, 'BTTIME = ', 16.6)
CALL EXIT
END IF
ERROR CHECK #4:
The backtrack file records are read by calling subroutine RDBT. The number of words to read, the starting
address of the common block BTFILE, and IOST are passed to RDBT. Upon return from RDBT, IOST con-
tains the I/O status of the read operation. If IOST is not equal to zero, then an error has occurred while
reading a record from the BTRK file. A message is written to the log to indicate that an error has occurred.
The message contains the subroutine name, the I/O status value, the number of words, the time step date and
time, and the row. The program exits by a call to the system subroutine EXIT.
CODE FOR ERROR CHECK #4:
c
C read BTRK
CALL RDBT (IUDLN2, XRU, IOST)
IF (IOST .NE. 0) THEN
URITE(LUNOUT, 2007) IOST, IUDLN2, BTDATE, BTTIME, RDBTRU
2007 FORMAT(/ 5X, 'XXX FILE READ ERROR IN RDBTRK1
& / 5X, 'I/O STATUS = ', 14, 2X,
& 'NUMBER OF WORDS TO BE READ - ', 13
& / 5X, 'BTDATE = ', 16.6, 2X, 'BTTIME « ', 16.6,
&2X, 'ROW = ', 13)
CALL EXIT
END IF
SUBROUTINE RDCHAR
NONE
F-63
-------
SUBROUTINE RDCONC
ERROR CHECK #1:
The CONC file is opened by a call to OPCONC, and the common block HEADCN is loaded. HEADCN con-
tains the header information for the CONC file. The header information is then checked in the subroutine
RDCONC If any of the parameters fail the test, then an error has occurred. A message is written to the log to
indicate that an error has occurred. The message contains the subroutine name and the CONC header infor-
mation. The program exits by a call to the system subroutine EXIT.
CODE FOR ERROR CHECK #1:
C open the CONC file
CALL OPCONC
C
C check parameters in header
IF ((TSTPCN .NE. TSTPIN) .OR. (GRDNCN .NE. GRDNIN) .OR.
& (ABS(SULNCN - SWLNIN) .GT. 0.001) .OR.
& (ABSCSWLTCN - SWLTIN) .GT. 0.001) .OR.
WRITEUUNOUT, 2001) TSTPCN, TSTPIN, GRDNCN, GRDNIN.
& SULNCN, SULNIN, SULTCN, SWLTIN,
& NELNCN, NELNIN, NELTCN, NELTIN,
& DLONCN, DLONIN, DLATCN, DLATIN,
& NCOLCN, NCOLIN, NROWCN, NROUIN,
& NLEVCN, NLEVIN, NSPCCN, NSPCIN
2001 FORMAT(/ 5X, '«X PARAMETER CHECK FAILURE IN RDCONC1
&
I
CALL
END IF
/ 5X,
/ 5X.
/ 5X,
/ SX.
/ SX.
/ 5X,
/ SX.
/ 5X.
/ 5X.
/ SX,
/ 5X,
/ 5X,
EXIT
'TSTPCN ••
'GRDNCN
'NELNCN
'NELTCN
'SULNCN
'SULTCN
'DLONCN
'DLATCN
'NCOLCN «
'NROWCN :
'NLEVCN
'NSPCCN
E
e
s
E
15,
AS,
F10
F10
F10
F10
F10
F10
13,
13,
13,
13,
5X,
5X,
•5,
•5.
•5,
• 5,
.5.
•5,
5X,
5X,
5X.
5X.
'TSTPIN =
'GRDNIN =
SX, 'NELNIN
SX, 'NELTIN
SX, 'SULNIN
SX, 'SWLTIN
SX, 'DLONIN
SX, 'DLATIN
'NCOLIN =
'NROUIN =
'NLEVIN =
•NSPCIN =
'. I
'' *
S
S
s
s
s
', I
', I
', I
', I
5
8
F10.5
F10.5
F10.5
F10.5
F10.5
F10.5
3
3
3
3)
ERROR CHECK #2:
The CONC time step header is read by calling the subroutine RDFILE. The unit number, the number of words
to be read, starting address of the common block RTSHCN, and IOST are passed to RDFILE Upon return
from the call to RDFELE, IOST contains the I/O status of the read operation of the CONC time step header.
IOST is tested; if IOST is less than zero, then an end-of-file marker was reached. Control is returned to the
calling subroutine. If IOST is not equal to zero, then an error occurred on the read operation. A message is
F-64
-------
written to the log to indicate that an error occurred. The message contains the subroutine name, the number
of words to read, the value of the I/O status, the time step date and time. The program exits by a call to the
system subroutine EXIT.
CODE FOR ERROR CHECK #2:
c
C read CONC T.S.H.
CALL RDFILE (UNITCN, NUDTSH, DATCN, IOST)
C
IF (IOST -LT. 0} RETURN
IF (IOST .NE. 0) THEN
WRITEdUNOUT, 2003) NUDTSH, IOST, CNDATE, CNTINE
2003 FORMAT(/ 5X, 'XXX T.S.H. READ ERROR IN RDCONC1
& / 5X, 'NO. OF WORDS = ', 12, 2X, 'I/O STATUS = ', 14,
S2X, 'CNDATE = ', 16.6, 2X, 'CNTIME = ', 16.6)
CALL EXIT
END IF
ERROR CHECK #3:
The current model time step time and date are compared with the time step time and date obtained from
reading the CONC time step header. If either the time or date do not match, then an error has occurred. A
message is written to the log to indicate that an error has occurred. The message contains the subroutine
name, date and time from the time step header, and the model date and time. The program exits by a call to
the system subroutine EXIT.
CODE FOR ERROR CHECK #3:
c
IF (IDATCN .NE. CNDATE .OR. ITIMCN .NE. CNTIME) THEN
URITE(LUNOUT, 2005) IDATCN, ITIMCN, CNDATE, CNTIME
2005 FORMAT(/ 5X, 'XXX DATES/TIMES DO NOT MATCH IN RDCONC'
& / 5X, 'IDATCN = ', 16.6, 2X, 'ITIMCN = ', 16.6
& / 5X, 'CNDATE = ', 16.6, 2X, 'CNTIME = ', 16.6)
CALL EXIT
END IF
ERROR CHECK #4:
Each row of the CONC file is read by iterating over the number of species. For each iteration, a call to RDFILE
is made. The unit number, the number of words to read, the starting address of the common block CNFILE,
and IOST are passed to RDFILE. Upon return from RDFILE, IOST contains the I/O status of the read opera-
tion. If IOST is not equal to zero, then an error has occurred. A message is written to the log indicating that
an error has occurred. The message contains the subroutine name, the I/O status value, the number of words,
the time step date and time, and the row. The program exits by a call to the system subroutine EXIT.
F-65
-------
CODE FOR ERROR CHECK #4:
c
C read CONC row
DO 211 ISPC = 1, NSPECS
CALL RDFILE (UNITCN, NUDSCN, CNFILE(1,1,ISPC), IOST)
IF (IOST .NE. 0) THEN
WRITECLUNOUT, 2007) IOST, NUDSCN, CNOATE, CNTIME, RDCNRU
2007 FORMATC/ 5X,
-------
CODE FOR ERROR CHECK #4:
c
C read the index group record
C
READ(UNIIBM, IOSTAT = IOST) RECNDX
IF (IOST .NE. 0) THEN
URITE(LUNOUT, 2001) ISUB, IOST, 'XXX INDEX GROUP
CALL EXIT
END IF
ERROR CHECK #5:
A formatted read of RECNDX, the buffer containing the index group record, converts the index group record
to character and numeric data, and loads the record into HEADBM. The I/O status of the read operation is
stored in the variable IOST. IOST is tested; if IOST does not equal zero, an error has occurred on the read.
A message is written to the log to indicate that an error has occurred. The error message contains the sub-
routine name, the subfile number, and the I/O status value. The program exits by a call to the system sub-
routine EXIT.
CODE FOR ERROR CHECK #5:
c
C convert character to numeric and load HEADBM
C
READ(RECNDX, 1003, IOSTAT = IOST)
& (BMINDX(ISPC), ISPC = 1, NSPCBM)
1003 FORMAT(I4)
IF (IOST .NE. 0) THEN
URITE(LUNOUT, 2001) ISUB, IOST. 'INTERNAL READ, INDEX REC'
CALL EXIT
END IF
ERROR CHECK #6:
The layer names record is read from the BMAT file and loaded into CHARBM. The I/O status of the read
operation is stored in the variable IOST. IOST is tested; if IOST does not equal zero, an error has occurred
on the read. A message is written to the log to indicate that an error has occurred. The error message con-
tains the subroutine name, the subfile number, and the I/O status value. The program exits by a call to the
system subroutine EXIT.
F-68
-------
CODE FOR ERROR CHECK #6:
c
C read the level names record
C
READCUNITBM, IOSTAT = IOST) (LVNMBM(LEV), LEV = 1, NLEVBM)
IF (IOST .NE. 0) THEN
WRITEUUNOUT, 2001) I SUB, IOST, 'XXX LEVEL NAMES
CALL EXIT
END IF
ERROR CHECK #7:
The MIF data records are next read by iterating over the number of MIF data records. For each iteration, an
MIF record is read into the buffer RECMIF. The I/O status value from the read operation is stored in the
variable IOST. IOST is tested; if IOST does not equal zero, an error has occurred on the read. A message is
written to the log to indicate that an error has occurred. The error message contains the subroutine name, the
subfile number, and the I/O status value. The program exits by a call to the system subroutine EXIT.
CODE FOR ERROR CHECK #7:
c
C read the MIF data records
C
DO 101 IMIF = 1, NMIFBM
READ(UNITBM, IOSTAT = IOST) RECMIF
IF (IOST .NE. 0) THEN
WRITE(CHBUF, 1005) 'XXX MIF - NUMBER ', IMIF
1005 FORMAT(A20, 14)
URITECLUNOUT, 2001) I SUB, IOST, CHBUF
CALL EXIT
END IF
ERROR CHECK #8:
A formatted read of RECMIF, the buffer containing the MLF record, converts the MIF record to character
and numeric data, and loads the record into HEADBM. The I/O status of the read operation is stored in the
variable IOST. IOST is tested; if IOST does not equal zero, an error has occurred on the read. A message is
written to the log to indicate that an error has occurred. The error message contains the subroutine name, the
subfile number, and the I/O status value. The program exits by a call to the system subroutine EXIT.
F-69
-------
CODE FOR ERROR CHECK #8:
READCRECMIF, 1007, IOSTAT = IOST)
& MFNMBM(IMIF), CDNFBM(IHIF), CTHFBM(IMIF),
& UOMFBH(IMIF), UTMFBM(INIF)
1007 FORMAT(A12, 418.8)
IF (IOST .NE. 0) THEN
URITE(CHBUF, 1005) 'INTERNAL READ, MIF DATA ', INIF
WRITECLUNOOT, 2001) ISUB, IOST, CHBUF
CALL EXIT
END IF
101 CONTINUE
ERROR CHECK #9:
The text records are next read by iterating over the number of text records. For each iteration, a text record is
read into the buffer RECTXT. The I/O status value from the read operation is stored in the variable IOST.
IOST is tested; if IOST does not equal zero, an error has occurred on the read. A message is written to the log
to indicate that an error has occurred. The error message contains the subroutine name, the subfile number,
and the I/O status value. The program exits by a call to the system subroutine EXIT.
CODE FOR ERROR CHECK #9:
c
C read header text records
C
DO 201 ITXT = 1, ICNTBM
READ(UNITBM, IOSTAT = IOST) RECTXT
IF (IOST .NE. 0) THEN
WRITE(LUNOUT, 2001) I SUB, IOST,
& 'XXX TEXT RECORDS '
• CALL EXIT
END IF
ERROR CHECK #10:
A formatted read of RECTXT, the buffer containing the text records, converts the text records to character
data, and loads them into CHARBM. The I/O status of the read operation is stored in the variable IOST.
IOST is tested; if IOST does not equal zero, an error has occurred on the read. A message is written to the log
to indicate that an error has occurred. The error message contains the subroutine name, the subfile number,
and the I/O status value. The program exits by a call to the system subroutine EXIT.
F-70
-------
CODE FOR ERROR CHECK #10:
c
READ(RECTXT, 1009, IOSTAT = IOST) TEXTBM(ITXT)
1009 FORMAT(A80)
IF (IOST .NE. 0) THEN
URITEaUNOUT, 2001) I SUB, IOST,
& 'INTERNAL READ, TEXT RECS1
URITE(LUNOUT, 1011} RECTXT
1011 FORMAT(5X, A80)
CALL EXIT
END IF
201 CONTINUE
ERROR CHECK #11:
The subfile header record is read from the BMAT file into the buffer FLCREC. The I/O status of the read
operation is stored in the variable IOST. IOST is tested; if IOST does not equal zero, an error has occurred
on the read. A message is written to the log to indicate that an error has occurred. The error message con-
tains the subroutine name, the subfile number, and the I/O status value. The program exits by a call to the
system subroutine EXIT.
CODE FOR ERROR CHECK #11: .
c
C read subfile header record
C
READ(UNITBM, IOSTAT = IOST) FLCREC
IF (IOST .NE. 0) THEN
WRITEUUNOUT. 2001) I SUB, IOST, >X%X SUBFILE RECORD '
CALL EXIT
END IF
ERROR CHECK #12:
A formatted read of FLCREC, the buffer containing the subfile header record, converts the record to integer
data, and loads the record into HEADBM. The I/O status of the read operation is stored in the variable
IOST. IOST is tested; if IOST does not equal zero, an error has occurred on the read. A message is written to
the log to indicate that an error has occurred. The error message contains the subroutine name, the subfile
number, and the I/O status value. The program erits by a call to the system subroutine EXIT.
F-71
-------
CODE FOR ERROR CHECK #12:
C convert character to numeric and load HEADBM
C
READ(FLCREC, 1013, IOSTAT = IOST) ISUBFL, NSUBFL,
& FRSTSF, LSSTSF
1013 FORMAT(4I4)
IF (IOST .NE. 0) THEN
WRITE(LUNOUT, 2001) I SUB, IOST, 'XXX SUBFILE COUNT
CALL EXIT
END IF
SUBROUTINE RDICON
ERROR CHECK #1:
The ICON file is opened by a call to OPICON, and the common block HEADIC is loaded. HEADIC contains
the header information for the ICON file. The header information is then tested in the subroutine RDICON.
If any of the parameters fail the test, then an error has occurred. A message is written to the log to indicate
that an error has occurred. The message contains the subroutine name and the ICON header information.
The program exits by a call to the system subroutine EXIT.
CODE FOR ERROR CHECK #1:
C open the ICON file
CALL OPICON
C
C check parameters in header
2001
&
&
&
&
&
&
&
&
&
&
&
&
&
&
&
&
&
&
&
IF ((TSTPIC .NE. TSTPIN) .OR. (GRDNIC .NE. GRDNIN) .OR.
(ABS(SULNIC - SULNIN) .GT. 0.001) .OR.
(ABSCSWLTIC - SULTIN) .GT. 0.001) .OR.
WRITECLUNOUT, 2001) TSTPIC, TSTPIN, GRDNIC, GRDNIN,
SWLNIC, SULNIN, SULTIC, SULTIN,
NELNIC, NELNIN, NELTIC, NELTIN,
DLONIC, DLONIN, DLATIC, OLATIN,
NCOLIC, NCOLIN, NROUIC, NROWIN,
NLEVIC, NLEVIN, NSPCIC, NSPCIN
FORMATC/ 5X, 'XXX PARAMETER CHECK FAILURE IN RDICON1
/ 5X, 'TSTPIC =
/ 5X, 'GRDNIC =
/ 5X, 'NELNIC =
/ 5X, 'NELTIC =
/ 5X, 'SWLNIC =
/ 5X, 'SWLTIC =
/ 5X, 'DLONIC *
/ 5X, 'DLATIC =
/ 5X, 'NCOLIC =
/ 5X. 'NROUIC =
/ 5X. 'NLEVIC =
/ 5X, 'NSPCIC =
15, 5X, 'TSTPIN = ', 15
AS, 5X, 'GRDNIN = ', A8
F10.5, 5X, 'NELNIN = F10.5
F10.5, 5X, 'NELTIN = F10.5
F10.5, 5X, 'SULNIN = F10.5
F10.5, 5X, 'SULTIN = F10.5
F10.5, 5X, 'DLONIN = F10.5
F10.5, 5X, 'DLATIN = F10.5
13, 5X, 'NCOLIN = ', 13
13, 5X, 'NROUIN = ', 13
13, 5X, 'NLEVIN = ', 13
13, 5X, 'NSPCIN = ', 13)
CALL EXIT
END IF
ERROR CHECK #2:
F-72
-------
The ICON time step header is read by calling the subroutine RDFILE. The unit number, the number of words
to be read, starting address of the common block RTSHIC, and IOST are passed to RDFILE. Upon return
from the call to RDFILE, IOST contains the I/O status of the read operation of the ICON time step header.
IOST is tested; if IOST is less than zero, then an end of file was reached. Control is returned to the calling
subroutine. If IOST is not equal to zero, then an error occurred on the read operation. A message is written
to the log to indicate that an error occurred. The message contains the subroutine name, the number of words
to read, the value of the I/O status, the time step date and time. The program exits by a call to the system
subroutine EXIT.
CODE FOR ERROR CHECK #2:
c
C read ICON T.S.H.
CALL RDFILE (UNITIC, NUOTSH, DATIC, IOST)
C
IF (IOST .LT. 0) RETURN
IF (IOST .NE. 0) THEN
WRITEUUNOUT, 2003} NUDTSH, IOST, ICDATE, ICTIME
2003 FORMAT(/ 5X, -XXX T.S.H. READ ERROR IN RDICON1
& / 5X. 'NO. OF WORDS = ', 12, 2X, 'I/O STATUS = ', 14,
&2X, 'ICDATE = ', 16.6, 2X, 'ICTIME = ', 16.6)
CALL EXIT
END IF
ERROR CHECK #3:
The current model time step time and date are compared with the time step time and date obtained from
reading the ICON time step header. If either the time or date do not match, then an error has occurred. A
message is written to the log to indicate that an error has occurred. The message contains the subroutine
name, date and time from the time step header, and the model date and time. The program exits by a call to
the system subroutine EXIT.
CODE FOR ERROR CHECK #3:
c
IF (IDATIC .NE. ICDATE .OR. ITINIC .HE. ICTIME) THEN
URITEUUNOUT. 2005) IDATIC, ITIMIC, ICDATE, ICTIME
2005 FORMAT(/ 5X, 'XXX DATES/TIMES DO NOT MATCH IN RDICON'
& / 5X, 'IDATIC = ', 16.6, 2X, 'ITIMIC = ', 16.6
& / 5X, 'ICDATE = ', 16.6, 2X, 'ICTIME = ', 16.6)
CALL EXIT
END IF
ERROR CHECK #4:
Each row of the ICON file is read by iterating over the number of species. For each iteration, a call to RDFILE
is made. The unit number, the number of words to read, the starting address of the common block ICFILE,
and IOST are passed to RDFILE. Upon return from RDFILE, IOST contains the I/O status of the read opera-
F-73
-------
tion. If IOST is not equal to zero, then an error has occurred. A message is written to the log to indicate that
an error has occurred. The message contains the subroutine name, the I/O status value, the number of words,
the time step date and time, and the row. The program exits by a call to the system subroutine EXIT.
CODE FOR ERROR CHECK #4:
c
C read ICON row
DO 211 ISPC = 1, NSPECS
CALL RDFILE (UNITIC, NUDSIC, ICFILE(1,1,ISPC), IOST)
IF (IOST .NE. 0) THEN
WRITE(LUNOUT, 2007) IOST. NUDSIC, ICDATE, ICTIME, IROW
2007 FORMAT(/ 5X, 'XXX FILE READ ERROR ON ROW IN RDICON1
&/ 5X, 'I/O STATUS = ', 14, 2X, 'NO. OF WORDS = ', 16
&/ 5X, MCDATE = ', 16.6, 2X, 'ICTIME * ', 16.6,
& 2X, 'ROW = ', 13)
CALL EXIT
END IF
211 CONTINUE
SUBROUTINE RDMXBM
ERROR CHECK #1:
The BMAT file requires a large amount of disk space and may have been written to several smaller subfiles on
different disk packs (since each pack may not individually have had sufficient space to contain the entire file).
These subfiles would then be assigned separate logical names in the job's run stream. In subroutine RDMXBM,
the subfile number is tested. If the test fails, then the BMAT subfiles are out of order. A call to subroutine
DUMPHD is made. The subfile number is the parameter passed to DUMPHD. Upon return from DUMPHD, the
program exits by issuing a call to the system subroutine EXIT.
CODE FOR ERROR CHECK #1:
c
C verify subfile
IF (ISUBFL .NE. ISUB .OR. NSUBFL .NE. NSUB) THEN
CALL DUMPHD (ISUB)
CALL EXIT
END IF
ERROR CHECK #2:
A record is read from the BMAT subfile. The I/O status value for the read operation is stored in the variable
IOST. IOST is tested; if IOST is less than zero, then the last record was read. An informational message is
printed to the log. The BMAT subfile is closed and control is returned to the calling subroutine. If IOST is
not equal to zero, then an error has occurred. A message is written to the log to indicate that an error has
F-74
-------
occurred. The message contains the subroutine name, the subfile number, the unit number, the logical name
of the file, the actual name of the file, the I/O status value, and the number of records. Control is returned to
the calling subroutine.
CODE FOR ERROR CHECK #2:
c
C read record
C
READ (UNITBM, IOSTAT = IOST) BUFF
RECNO = 1
C
C read subfile body
C
IF (IOST .LT. 0) GO TO 301
IF (IOST .NE. 0) THEN
INQUIRE (FILE = FLNMBM(ISUB), NAME = EQNAME)
URITECLUNOUT, 2001) I SUB, UNITBM, FLMMBM(ISUB), EONAME,
& IOST, RECNO
2001 FORMAT(/ 5X, 'XXX ERROR READING RECORD IN RDHXBM',
& ' FOR SUBFILE NUMBER', 13,
&5X, 'CONTROL RETURNED TO RDBMAT1
& / 5X, 'UNITBM = ', 12, 2X, 'FNAME = ', A8
& / 5X, 'EQNAME = ', A64
& / 5X, 'IOST = ', 12, 2X, 'RECNO = ',14)
RETURN
END IF
301 CONTINUE
WRITEUUNOUT, 1001) RECNO, FLNMBM(ISUB), UNITBM
1001 FORMAU// 5X, 16, ' records read on ', A12,
& ' from unit ', 12 /)
TOTREC = TOTREC + RECNO
C
C close subfile
INQUIRE (UNIT = UNITBM, NAME = EQNAME)
CLOSE (UNIT = UNITBM)
WRITEC6, 1003) ISUB, EQNAME, UNITBM
1003 FORMAT(/ 3X, 'Subfile ', 12, ', ' A64
& / 3X, 'closed on unit ', 12)
ERROR CHECK #3:
Each record of the last subfile of the BMAT file is read. The I/O status value for the read operation is stored
in the variable IOST. IOST is tested; If IOST is less than zero, then the last record was read. An informa-
tional message is printed to the log. The BMAT subfile is closed and control is returned to the calling sub-
routine. If IOST is not equal to zero, then an error has occurred. A message is written to the log to indicate
that an error has occurred. The message contains the subroutine name, the subfile number, the unit number,
the logical name of the file, the actual name of the file, the I/O status value, and the number of records.
Control is returned to the calling subroutine.
F-75
-------
CODE FOR ERROR CHECK #3:
c
C read record
C
READ (UNITBM, IOSTAT - IOST) BUFF
RECNO = RECNO + 1
C
C last subfl body record*
C
IF (IOST .LT. 0) GO TO 601
IF (IOST .NE. 0) THEN
INQUIRE (FILE = FLNMBM(ISUB), NAME = EONANE)
WRITEdUNOUT, 2001) I SUB, UNITBM, FLNMBM(ISUB), EQNAME,
& IOST, RECNO
RETURN
END IF
C last subfl body end
C
601 CONTINUE
URITEUUNOUT, 1001) RECNO, FLNNBM(ISUB), UNITBM
TOTREC = TOTREC + RECNO
WRITEdUNOUT, 1005) NSUB, TOTREC
1005 FORMAT(/ 5X, 'Total number of records read from1, 13,
& ' subfiles - ', 18)
C
C close subfile
INQUIRE (UNIT = UNITBM, NAME = EQNAME)
CLOSE (UNIT = UNITBM)
URITE(6, 1003) I SUB, EQNAME, UNITBM
SUBROUTINE RDSTAV
ERROR CHECK #1:
A formatted read from the STATE VECTOR file (RESTRT) is made to obtain the STATE VECTOR time
step header. This formatted read converts the data into integer data, and loads them into the common block
TSHDSV. The I/O status value for the read operation is stored in variable IOST. IOST is tested; if IOST
does not equal zero, an error has occurred on the read. A message is written to the log to indicate that an
error has occurred. The error message contains the subroutine name, the unit number, the record number,
and the I/O status value. The program exits by a call to the system subroutine EXIT.
F-76
-------
CODE FOR ERROR CHECK #1:
c
C read STATE VECTOR T.S.H.
RECNSV = RECNSV + 1
READCUNITSV, FMT - 1001, IOSTAT = IOST)
& IDATSV, ITIMSV, IELPSV, ISTPSV
1001 FORMAT(1X, 15, 1X, 16, 1X, 18, 1X, 14)
C
IF (IOST .NE. 0) THEN
WRITE(LUNOUT, 2001) UNITSV, RECNSV, IOST
2001 FORMAT(/ 5X, 'XXX T.S.H READ ERROR IN RDSTAV
& / 5X, 'UNIT NUMBER = ', 12, 5X, 'RECORD = ', 14
& / 5X, 'I/O STATUS = ', 14)
CALL EXIT
END IF
ERROR CHECK #2:
The current model time step time and date are compared with the time step time and date obtained from
reading the STATE VECTOR time step header. If either the time or date do not match, then an error has
occurred. A message is written to the log to indicate that an error has occurred. The message contains the
subroutine name, date and time from the time step header, and the model date and time. The program exits
by a call to the system subroutine EXIT.
CODE FOR ERROR CHECK #2:
c
IF (IDATSV .NE. IDATE .OR. ITIMSV .NE. ITIME) THEN
WRITEUUNOUT, 2003) IDATE, ITIME, IDATSV, ITIMSV
2003 FORMAT(/ SX, 'XXX DATES/TIMES DO NOT MATCH IN RDSTAV
& / 5X, 'IDATE = ', 16.6, 2X, 'ITIME = ', 16.6
& / 5X, 'IDATSV = ', 16.6, 2X, 'ITIMSV = ', 16.6)
CALL EXIT
END IF
ERROR CHECK #3:
A formatted read from the STATE VECTOR file is made to obtain the text pointers record. .This formatted
read converts the data into integer .data, and loads them into the common block TEXTPT. The I/O status
value for the read operation is stored in variable IOST. IOST is tested; if IOST does not equal zero, an error
has occurred on the read. A message is written to the log to indicate that an error has occurred. The error
message contains the subroutine name, the unit number, the record number, and the I/O status value. The
program exits by a call to the system subroutine EXIT.
F-77
-------
CODE FOR ERROR CHECK #3:
c
C read text pointers
C
RECNSV = RECNSV + 1
REAOCUNITSV, FMT = 1003, IOSTAT = IOST)
& BIGMPT, LILGPT,
& BCPSPT, BMPSPT, BTPSPT, CNPSPT, ICPSPT,
& RDBCPT, RDBNPT, RDBTPT, ROCNPT, RDICPT,
& WRCNPT, WRSVPT
1003 FORMAT(1X, U(I3, 1X))
IF (IOST .NE. 0) THEN
WRITECLUNOUT, 2005) UNITSV, RECNSV, IOST
2005 FORMAT(/ 5X, 'XXX TEXTPT READ ERROR IN RDSTAV
& / 5X, 'UNIT NUMBER = ', 12, 5X, 'RECORD = ', 14
& / 5X, 'I/O STATUS = ', 14)
CALL EXIT
END IF
ERROR CHECK #4:
A formatted read from the STATE VECTOR file is made to obtain the row counters record. This formatted
read converts the data into integer data, and loads them into the common block ROWSCT. The I/O status
value for the read operation is stored in variable IOST. IOST is tested; if IOST does not equal zero, an error
has occurred on the read. A message is written to the log to indicate that an error has occurred. The error
message contains the subroutine name, the unit number, the record number, and the I/O status value. The
program exits by a call to the system subroutine EXIT.
CODE FOR ERROR CHECK #4:
c
C read row counters
C
RECNSV = RECNSV + 1
READ(UNITSV, FMT = 1005, IOSTAT = IOST)
& BMPSRU, RDBMRU, RDBTRU, RDCNRU
1005 FORMAT<1X, 4(13, 1X))
IF (IOST .NE. 0) THEN
WRITECLUNOUT, 2023) UNITSV. RECNSV, IOST
2023 FORMAT(/ 5X, 'XXX ROW COUNTERS READ ERROR IN RDSTAV
& / 5X, 'UNIT NUMBER = ', 12, 5X, 'RECORD = '. 14
& / 5X, 'I/O STATUS = ', 14)
CALL EXIT
END IF
ERROR CHECK #5:
A formatted read from the STATE VECTOR file is made to obtain the scenario time record. This formatted
read converts the data into integer data, and loads them into the common block TSTEPS. The I/O status
value for the read operation is stored in variable IOST. IOST is tested; if IOST does not equal zero, an error
F-78
-------
has occurred on the read. A message is written to the log to indicate that an error has occurred. The error
message contains the subroutine name, the unit number, the record number, and the I/O status value. The
program exits by a call to the system subroutine EXIT.
CODE FOR ERROR CHECK #5:
c
C read scenario time
C
RECNSV = RECNSV + 1
READ(UNITSV, FMT = 1007, IOSTAT = IOST)
& HDDATE, MDTIME, HOELAP, MDSTEP,
& BCDATE. BCTIME, BCELAP, 8CSTEP,
& BMDATE, BHTIME, BMELAP, BMSTEP
1007 FORMATdX, 3(15, 1X, 16, 1X, 18, 14, 1X)>
C
RECNSV = RECNSV + 1
READ(UNITSV, FMT = 1007, IOSTAT = IOST)
& BTDATE, BTTIME, BTELAP, BTSTEP,
& CNDATE, CNTIME, CNELAP, CNSTEP,
& ICDATE, ICTIME, ICELAP, ICSTEP
IF (IOST .NE. 0) THEN
WRITE(LUNOUT, 2007) UNITSV, RECNSV, IOST
2007 FORMAT(/ 5X, 'XXX SCENARIO TIME READ ERROR IN RDSTAV
& / 5X, 'UNIT NUMBER = ', 12, 5X, 'RECORD = ', 14
& / 5X, 'I/O STATUS = ', 14)
CALL EXIT
END IF
C
C file body end
SUBROUTINE RTPHO
NONE
SUBROUTINE RTSET
NONE
SUBROUTINE RUNMGR
ERROR CHECK #1:
The CONC file is opened, and the file header time and date information is tested against the STATE VEC-
TOR file (RESTRT) header time and date. If either the time or date do not match, then an error has
occurred. A message is written to the log to indicate that an error has occurred. The message contains the
subroutine name, date and time from the CONC file time step header, and the date and time from the STATE
VECTOR file time step header.
F-79
-------
CODE FOR ERROR CHECK #1:
c
C open CONG file and check file header with STATE VECTOR file header
CALL OPCONC
IF (CDATCN .NE. CDATSV .Oft. CTINCN .NE. CTIMSV) THEN
WRITE(LUNOUT, 2001) CDATCN, CTIMCN, CDATSV, CTIMSV
2001 FORMAT(// 22X, '!!! WARNING IN RUN MANAGER I!!1
& / 36X, 'CONC FILE CREATION DATE/TIME: ', 16, 2X, 16
& / 3X, 'DOES NOT MATCH CREATION DATE/1,
& 'TIME ON STATE VECTOR FILE HEADER: ', 16, 2X, 16 //)
END IF
ERROR CHECK #2:
The BCON file is opened. The CONC file header time and date information is tested against the BCON file
header time and date. If either the time or date do not match, then an error has occurred. A message is
written to the log to indicate that an error has occurred. The message contains the subroutine name, date and
time from the CONC file time step header, and the date and time from the BCON file time step header.
CODE FOR ERROR CHECK #2:
c
C open BCON file and check file header
CALL OPBCON
IF (CDATBC .NE. CDBCCN .OR. CTIMBC .NE. CTBCCN) THEN
WRITE(LUNOUT, 2003) 'BCON1, CDATBC, CTIMBC, CDBCCN, CTBCCN
2003 FORMAT(// 22X. Mil WARNING IN RUN MANAGER IN1
& / 29X, A4, ' FILE CREATION DATE/TIME: ', 16, 2X, 16
& / 4X, 'DOES NOT MATCH CREATION DATE/',
& 'TIME ON CONC FILE HEADER: ', 16, 2X, 16 //)
END IF
ERROR CHECK #3:
The BTRK file is opened. The CONC file header time and date information is tested against the BTRK file
header time and date. If either the time or date do not match, then an error has occurred. A message is
written to the log to indicate that an error has occurred. The message contains the subroutine name, date and
time from the CONC file time step header, and the date and time from the BTRK file time step header.
CODE FOR ERROR CHECK #3:
c
C open BTRK file and check file header
CALL OPBTRK
IF (CDATBT .NE. CDBTCN .OR. CTIMBT .NE. CTBTCN) THEN
WRITE(LUNOUT, 2003) 'BTRK1, CDATBT, CTIMBT, CDBTCN, CTBTCN
END IF
ERROR CHECK #4:
F-80
-------
The BMAT file is opened. The CONC file header time and date information is tested against the BMAT file
header time and date. If either the time or date do not match, then an error has occurred. A message is
written to the log to indicate that an error has occurred. The message contains the subroutine name, date and
time from the CONC file time step header, and the date and time from the BMAT file time step header.
CODE FOR ERROR CHECK #4:
c
C open BMAT file and check file header
CALL OPBMAT
IF (CDATBM .NE. CDBMCN .OR. CTIHBM .NE. CTBMCN) THEN
URITE
-------
zero, then an end-of-file marker or error was encountered while reading the input files. A message is written
to the log indicating in which file the error or end-of-file marker was reached, along with the model time step
and date. The program exits by a call to the system subroutine EXIT.
CODE FOR ERROR CHECK #8:
c
C run body time step*
DO 301 JSTEP = 1, NSTEPS
C
C read MODEL T.S.H.
CALL BIGGAM (IEOFBG)
C
IF (IEOFBG .EQ. 1) THEN
WRITEUUNOUT, 2009) 'ICON1, IDATMD, ITIMMD
2009 FORHAT(/ 5X, 'XXX EOF ON ', A4, ' FILE AT MODEL DATE:*,
& 16, 2X, 16)
CALL EXIT
ELSE IF (IEOFBG .EQ. 2) THEN
WRITEUUNOUT, 2009) 'BCON', IDATMD, ITIMMD
CALL EXIT
ELSE IF (IEOFBG .EQ. 3) THEN
WRITE(LUNOUT, 2009) 'BTRK1, IDATMD, ITIMMD
CALL EXIT
ELSE IF (IEOFBG .EQ. 7) THEN
WRITEUUNOUT, 2009) 'BMAT', IDATMD, ITIMMD
CALL EXIT
ELSE IF (IEOFBG .EQ. 4) THEN
WRITEUUNOUT, 2011) 'ICON1, IDATMD, ITIMMD
2011 FORHATC/ 5X, 'XXX ERROR READING ', A4,
& ' FILE AT MODEL DATE:',
& 16, 2X, 16)
CALL EXIT
ELSE IF (IEOFBG .EQ. 5) THEN
WRITEUUNOUT, 2011) 'BCON', IDATMD, ITIMMD
ELSE IF (IEOFBG .EQ. 6) THEN
WRITEUUNOUT, 2011) 'BTRK', IDATMD, ITIMMD
END IF
SUBROUTINE TIMER
NONE
SUBROUTINE WRCHAR
NONE
SUBROUTINE WRCONC
ERROR CHECK #1:
F-83
-------
The CONC time step header is written by calling the subroutine WRFILE. The unit number, the number of
words to be written, starting address of the common block RTSHCN, and IOST are passed to WRFILE. Upon
return from the call to WRFILE, IOST contains the I/O status of the write operation of the CONC time step
header. IOST is tested. If IOST is less than zero, then an end-of-file marker was reached. Control is returned
to the calling subroutine. If IOST is not equal to zero, then an error occurred on the write operation. A
message is written to the log to indicate that an error occurred. The message contains the subroutine name,
the unit number, the I/O status value, and the number of words to write. The program exits by a call to the
system subroutine EXIT.
CODE FOR ERROR CHECK #1:
c
C write CONC T.S.H.
C
CALL WRFILE (UNITCH, IWLTSH, DATCN, IOST)
C
IF (IOST .NE. 0) THEN
WRITE(LUNOUT, 2001) UNITCN, IOST, IWLTSH
2001 FORMAT(// 5X, 'XXX ERROR WRITING TSH IN WRCONC1
& / 5X, 'UNIT NUMBER = ', 12, 2X, 'I/O STATUS = ', 14,
& 2X, 'NO. OF WORDS = ', 14 /)
CALL EXIT
END IF
ERROR CHECK #2:
Each row of the CONC file is written by iterating over the number of species. For each iteration, a call to
WRFILE is made. The unit number, the number of words to read, the starting address of the common block
CNFILE, and IOST are passed to WRFTLE. Upon return from WRFILE, IOST contains the I/O status of the
write operation. If IOST is not equal to zero, then an error has occurred. A message is written to the log to
indicate that an error has occurred. The message contains the subroutine name, the I/O status value, the
number of words, the time step date and time, and the row. The program exits by a call to the system sub-
routine EXIT.
CODE FOR ERROR CHECK #2:
c
C write CONC row
DO 301 ISPC = 1, NSPECS
CALL WRFILE (UNITCN. IWDLN2, CNFILE(1,1,ISPC), IOST)
IF (IOST .NE. 0) THEN
WRITEUUNOUT, 2003) UNITCN, IOST, IWDLN2
2003 FORMAT(// 5X, 'XXX ERROR WRITING ROW IN WRCONC'
& / 5X, 'UNIT NUMBER = ', 12,
& 2X, M/0 STATUS = ', 14,
& 2X, 'NO. OF WORDS = ', 14 /)
CALL EXIT
END IF
301 CONTINUE
F-84
-------
SUBROUTINE WRFILE
NONE
SUBROUTINE WRSTAV
ERROR CHECK #1:
The STATE VECTOR file is opened. The I/O status value of the open statement is stored in variable IOST.
IOST is tested; if IOST does not equal zero, an error occurred on the open statement. A message is written to
the log to indicate that an error has occurred. The message contains the subroutine name, the unit number
and the I/O status value. The program exits by a call to the system subroutine EXIT.
CODE FOR ERROR CHECK #1:
c
C open STATE VECTOR file, formatted, read/write access
C
UNITSV = JUNITO
OPEN (UNITSV,
& FILE = FLNMSV,
& ACCESS = 'SEQUENTIAL1,
& STATUS = 'UNKNOWN1,
& IOSTAT = IOST)
C
IF (IOST .NE. 0) THEN
URITEUUNOUT, 2001) UNITSV, IOST
2001 FORMAT(/ 5X, 'XXX SV FILE OPEN ERROR IN WRSTAV
& / 5X, 'UNIT NUMBER = ', 12
& / 5X, 'I/O STATUS = ', 14)
CALL EXIT
END IF
ERROR CHECK #2:
A formatted write of the common blocks CHARSV and HEADSV writes the first header segment record to
the STATE VECTOR file. The I/O status of the formatted write statement is stored in the variable IOST.
IOST is tested; if IOST does not equal zero, an error has occurred on the write operation. A message is
written to the log to indicate that an error has occurred. The message contains the subroutine name, the unit
number, the record number, and the I/O status value. The program exits by a call to the system subroutine
EXIT.
F-85
-------
CODE FOR ERROR CHECK #2:
c
C write STATE VECTOR header segment 1
C
RECNSV = RECNSV + 1
URITECUNITSV, FMT = 1001, IOSTAT = IOST)
& CDATSV, CTIMSV, SOATSV, STHRSV. TSTPSV,
& FRSTSV, GRDNSV, SWLNSV, SULTSV, NELNSV
1001 FORMATdX, 2(16, 1X), 15, IX, 12, 1X, 2(18, IX),
& AS, 1X, 3CF8.3, 1X))
IF (IOST .ME. 0) THEN
WRITE(LUNOUT, 2003) UNITSV, RECNSV, IOST
2003 FORMAT(/ 5X, 'XXX HEADER WRITE ERROR IN WRSTAV
& / 5X, 'UNIT NUMBER = ', 12, 5X, 'RECORD = ', 14
& / 5X, 'I/O STATUS = ', 14)
CALL EXIT
END IF
ERROR CHECK #3:
A formatted write of the common block HEADSV writes the second header segment record to the STATE
VECTOR file. The I/O status of the formatted write statement is stored in the variable IOST. IOST is tested;
if IOST does not equal zero, an error has occurred on the write operation. A message is written to the log to
indicate that an error has occurred. The message contains the subroutine name, the unit number, the record
number, and the I/O status value. The program exits by a call to the system subroutine EXIT.
CODE FOR ERROR CHECK #3:
c
C write STATE VECTOR header segment 2
C
RECNSV = RECNSV + 1
WRITE(UNITSV, FMT = 1003, IOSTAT = IOST)
& NELTSV, DLONSV, DLATSV, NCOLSV.
& NROUSV, NLEVSV, NSPCSV, ICNTSV
1003 FORMATdX, F8.3, 2(1X, F8.5), 5(1X, 14))
IF (IOST.NE.O) THEN
WRITE(LUNOUT, 2003) UNITSV, RECNSV, IOST
CALL EXIT
END IF
ERROR CHECK #4:
A formatted write of the common block CHARSV writes the first segment of the species names record to the
STATE VECTOR file. The I/O status of the formatted write statement is stored in the variable IOST. IOST
is tested; if IOST does not equal zero, an error has occurred on the write operation. A message is written to
the log to indicate that an error has occurred. The message contains the subroutine name, the unit number,
the record number, and the I/O status value. The program exits by a call to the system subroutine EXIT.
F-86
-------
CODE FOR ERROR CHECK #4:
c
C write the species names records
C
RECNSV = RECNSV + 1
URITE(UNITSV, FMT = 1005, IOSTAT = IOST)
& (SPNMSV(ISPC), ISPC = 1, 15)
1005 FORMATdX, 15(A4, 1X»
IF (IOST .NE. 0) THEN
WRITE(LUNOUT, 2005) UNITSV, RECNSV, IOST
2005 FORMAT(/ 5X, 'XXX SPECIES NAME WRITE ERROR IN URSTAW
& / 5X, 'UNIT NUMBER = ', 12, 5X, 'RECORD = ', K
& / 5X, 'I/O STATUS = ', K)
CALL EXIT
END IF
ERROR CHECK #5:
A formatted write of the common block CHARSV writes the second segment of the species names record to
the STATE VECTOR file. The I/O status of the formatted write statement is stored in the variable IOST.
The second species names record is written. The I/O status of the formatted write statement is stored in the
variable IOST. IOST is tested; if IOST does not equal zero, an error has occurred on the write operation. A
message is written to the log to indicate that an error has occurred. The message contains the subroutine
name, the unit number, the record number, and the I/O status value. The program exits by a call to the system
subroutine EXIT.
CODE FOR ERROR CHECK #5:
c
RECNSV = RECNSV + 1
WRITE(UNITSV, FMT = 1005, IOSTAT = IOST)
& (SPNMSV(ISPC), ISPC = 16, NSPCSV)
IF (IOST .NE. 0) THEN
URITEaUNOUT, 2005) UNITSV, RECNSV, IOST
CALL EXIT
END IF
ERROR CHECK #6:
A formatted write of the common block CHARSV writes the levels names record to the STATE VECTOR
file. The I/O status of the formatted write statement is stored in the variable IOST. IOST is tested; if IOST
does not equal zero, an error has occurred on the write operation. A message is written to the log to indicate
that an error has occurred. The message contains the subroutine name, the unit number, the record number,
and the I/O status value. The program exits by a call to the system subroutine EXIT.
F-87
-------
CODE FOR ERROR CHECK #6:
c
C write the level names record
C
RECNSV = RECNSV + 1
URITE(UNITSV, FNT = 1005, IOSTAT = IOST)
& (LVNMSV(ILEV), I LEV = 1. NLEVSV)
IF (IOST .NE. 0) THEN
WRITEUUNOUT, 2007} UNITSV, RECNSV, IOST
2007 FORMAT(/ 5X, 'XXX LEVEL NAMES WRITE ERROR IN URSTAV
& / 5X, 'UNIT NUMBER = ', 12, 5X, 'RECORD - ', 14
& / 5X, 'I/O STATUS = ', 14}
CALL EXIT
END IF
ERROR CHECK #7:
The header text records are written by iterating over the number of text records. For each iteration, a for-
matted write of the common block CHARSV writes a text record to the STATE VECTOR file. The I/O
status of the formatted write statement is stored in the variable IOST. IOST is tested; if IOST does not equal
zero, an error has occurred on the write operation. A message is written to the log to indicate that an error
has occurred. The message contains the subroutine name, the unit number, the record number, and the I/O
status value. The program exits by a call to the system subroutine EXIT.
CODE FOR ERROR CHECK #7:
c
C write header text records
C
DO 109 ITXT = 1, ICNTSV
RECNSV = RECNSV + 1
WRITE(UNITSV, FMT = 1007, IOSTAT = IOST) TEXTSV(ITXT)
1007 FORMATdX, A80)
IF (IOST .NE. 0) THEN
WRITE(LUNOUT, 2009) UNITSV, RECNSV, IOST
2009 FORMAT(/ 5X, 'XXX SV HEADER WRITE ERROR IN WRSTAV
& / 5X, 'UNIT NUMBER - ', 12, 5X, 'RECORD - ', 14
& / 5X, 'I/O STATUS = ', 14)
CALL EXIT
END IF
109 CONTINUE
ERROR CHECK #8:
A formatted write of the common block HEADIN writes the first header segment record to the STATE
VECTOR file. The I/O status of the formatted write statement is stored in the variable IOST. IOST is tested;
if IOST does not equal zero, an error has occurred on the write operation. A message is written to the log to
indicate that an error has occurred. The message contains the subroutine name, the unit number, the record
number, and the I/O status value. The program exits by a call to the system subroutine EXIT.
F-88
-------
CODE FOR ERROR CHECK #8:
c
C write HEADIN header record segment 1
C
RECNSV = RECNSV + 1
WRITE(UNITSV, FMT = 1001, IOSTAT = IOST)
& CDATIN, CTIMIN, SDATIN, STHRIN, TSTPIN, FRSTIN,
& GRDNIN, SULNIN, SWLTIN, NELNIN
IF (IOST .NE. 0) THEN
URITE(LUNOUT, 2011) UNITSV, RECNSV, IOST
2011 FORMAT(/ 5X, 'XXX FILE WRITE ERROR IN URSTAV
& / 5X, 'UNIT NUMBER = ', 12, 5X, 'RECORD - ', 14
& / 5X, 'I/O STATUS = ', 14)
CALL EXIT
END IF
ERROR CHECK #9:
A formatted write of the common block HEADIN writes the second header segment record to the STATE
VECTOR file. The I/O status of the formatted write statement is stored in the variable IOST. IOST is tested;
if IOST does not equal zero, an error has occurred on the write operation. A message is written to the log to
indicate that an error has occurred. The message contains the subroutine name, the unit number, the record
number, and the I/O status value. The program exits by a call to the system subroutine EXIT.
CODE FOR ERROR CHECK #9:
c
C write HEADIN header record segment 2
C
RECNSV = RECNSV + 1
WRITECUNITSV, FMT = 1003, IOSTAT = IOST)
& NELTIN, DLONIN, DLATIN, NCOLIN, NROWIN,
& NLEVIN, NSPCIN, ICNTIN
IF (IOST .NE. 0) THEN
WRITEUUNOUT, 2011) UNITSV, RECNSV, IOST
CALL EXIT
END IF
ERROR CHECK #10:
A formatted write of the common block CHEMIN writes the header record to the STATE VECTOR file.
The I/O status of the formatted write statement is stored in the variable IOST. IOST is tested; if IOST does
not equal zero, an error has occurred on the write operation. A message is written to the log to indicate that
an error has occurred. The message contains the subroutine name, the unit number, the record number, and
the I/O status value. The program exits by a call to the system subroutine EXIT.
F-89
-------
CODE FOR ERROR CHECK #10:
c
C write CHEMIN header record
C
RECNSV = RECNSV + 1
WRITECUNITSV, FMT = 1013, IOSTAT = IOST)
& ATS, GTS, UFRAX, BFRAX, FACTOR,
& DIVP, DIVQ, NCOUT
1013 FORMATdX, 2(F8.2, 1X), 5CF8.5, IX), 15)
IF (IOST .NE. 0) THEN
WR1TE(LUNOUT, 2013) UNITSV, RECNSV, IOST
2013 FORMAT(/ 5X, 'XXX CHEMIN INTERNAL WRITE ERROR IN WRSTAV
& / 5X. 'UNIT NUMBER = ', 12, 5X, 'RECORD = ', 14
& / 5X, 'I/O STATUS = ', 14)
CALL EXIT
END IF
ERROR CHECK #11:
A formatted write of the common block CHEMIN writes the species record to the STATE VECTOR file.
The I/O status of the formatted write statement is stored in the variable IOST. IOST is tested; if IOST does
not equal zero, an error has occurred on the write operation. A message is written to the log to indicate that
an error has occurred. The message contains the subroutine name, the unit number, the record number, and
the I/O status value. The program exits by a call to the system subroutine EXIT.
CODE FOR ERROR CHECK #11:
c
RECNSV = RECNSV + 1
WRITECUNITSV, FMT = 1015, IOSTAT = IOST)
& (ISPEC(ISPC), ISPC = 1, NCOUT), ULIM, BLIM, FNOLIM
1015 FORMATdX, (I4.3, 1X), 3(E10.3. IX) )
IF (IOST .NE. 0) THEN
URITEUUNOUT, 2013) UNITSV, RECNSV, IOST
CALL EXIT
END IF
ERROR CHECK #12:
A formatted write of the common block NDXSPC writes the first segment record of the species ordering
header record (NXSPBM) to the STATE VECTOR file. The I/O status of the formatted write statement is
stored in the variable IOST. IOST is tested; if IOST does not equal zero, an error has occurred on the write
operation. A message is written to the log to indicate that an error has occurred. The message contains the
subroutine name, the unit number, the record number, and the I/O status value. The program exits by a call
to the system subroutine EXIT.
F-90
-------
CODE FOR ERROR CHECK #12:
c
C write species ordering header record
C
RECNSV = RECNSV + 1
WRITE(UNITSV, FMT = 1019, IOSTAT = IOST)
& (NXSPBM(ISPC), ISPC = 1, 15)
1019 FORMAT <1X, 15(13, 1X)>
IF (IOST .NE. 0) THEN
WRITE(LUNCUT, 2015} 'BH-1', UNITSV, RECNSV, IOST
2015 FORMAT(/ 5X, 'XXX FILE WRITE ERROR IN WRSTAV: NXSP', A4
& / 5X, 'UNIT NUMBER = ', 12, 5X, 'RECORD = ', 14
& / 5X, 'I/O STATUS = ', 14)
CALL EXIT
END IF
ERROR CHECK #13:
A formatted write of the common block NDXSPC writes the second segment record of the species ordering
header record (NXSPBM) to the STATE VECTOR file. The I/O status of the formatted write statement is
stored in the variable IOST. IOST is tested; if IOST does not equal zero, an error has occurred on the write
operation. A message is written to the log to indicate that an error has occurred. The message contains the
subroutine name, the unit number, the record number, and the I/O status value. The program exits by a call
to the system subroutine EXIT.
CODE FOR ERROR CHECK #13:
c
RECNSV = RECNSV + 1
URITE(UNITSV, FMT = 1019, IOSTAT = IOST)
& (NXSPBM(ISPC), ISPC = 16, NSPECS)
IF (IOST .NE. 0) THEN
WRITE(LUNOUT, 2015) 'BM-21, UNITSV, RECNSV, IOST
CALL EXIT
END IF
ERROR CHECK #14:
A formatted write of the common block NDXSPC writes the first segment record of the species ordering
header record (NXSPBC) to the STATE VECTOR file. The I/O status of the formatted write statement is
stored in the variable IOST. IOST is tested; if IOST does not equal zero, an error has occurred on the write
operation. A message is written to the log to indicate that an error has occurred. The message contains the
subroutine name, the unit number, the record number, and the I/O status value. The program exits by a call
to the system subroutine EXIT.
F-91
-------
CODE FOR ERROR CHECK #14:
c
RECNSV = RECNSV + 1
WRITE(UNITSV, FMT = 1019, IOSTAT = IOST)
& (MXSPBC(ISPC), ISPC = 1. 15)
IF (IOST -ME. 0) THEN
URITEaUNOUT, 2015) 'BC-V, UNITSV, RECNSV, IOST
CALL EXIT
END IF
ERROR CHECK #15:
A formatted write of the common block NDXSPC writes the second segment record of the species ordering
header record (NXSPBC) to the STATE VECTOR file. The I/O status of the formatted write statement is
stored in the variable IOST. IOST is tested; if IOST does not equal zero, an error has occurred on the write
operation. A message is written to the log to indicate that an error has occurred. The message contains the
subroutine name, the unit number, the record number, and the I/O status value. The program exits by a call
to the system subroutine EXIT.
CODE FOR ERROR CHECK #15:
c
RECNSV = RECNSV + 1
WRITECUNITSV, FMT = 1019, IOSTAT = IOST)
& (NXSPBC(ISPC), ISPC = 16, NSPECS)
IF (IOST .NE. 0) THEN
WRITEUUNOUT, 2015) 'BC-2', UNITSV, RECNSV, IOST
CALL EXIT
END IF
ERROR CHECK #16:
A formatted write of the common block NDXSPC writes the first segment record of the species ordering
header record (NXSPIC) to the STATE VECTOR file. The I/O status of the formatted write statement is
stored in the variable IOST. IOST is tested; if IOST does not equal zero, an error has occurred on the write
operation. A message is written to the log to indicate that an error has occurred. The message contains the
subroutine name, the unit number, the record number, and the I/O status value. The program exits by a call
to the system subroutine EXIT.
F-92
-------
CODE FOR ERROR CHECK #16:
c
RECNSV = RECNSV + 1
WRITE(UNITSV, FMT = 1019, IOSTAT = IOST)
& (NXSPIC(ISPC), ISPC = 1, 15)
IF (IOST .NE. 0) THEM
VRITE(LUNOUT, 2015) 'IC-11, UNITSV, RECNSV, IOST
CALL EXIT
END IF
ERROR CHECK #17:
A formatted write of the common block NDXSPC writes the second segment record of the species ordering
header record (NXSPIC) to the STATE VECTOR file. The I/O status of the formatted write statement is
stored in the variable IOST. IOST is tested; if IOST does not equal zero, an error has occurred on the write
operation. A message is written to the log to indicate that an error has occurred. The message contains the
subroutine name, the unit number, the record number, and the I/O status value. The program exits by a call
to the system subroutine EXIT.
CODE FOR ERROR CHECK #17:
c
RECNSV = RECNSV + 1
WRITE(UNITSV, FMT = 1019, IOSTAT = IOST)
& (NXSPIC(ISPC), ISPC = 16, NSPECS)
. IF (IOST .NE. 0) THEN
WRITEUUNOUT, 2015) 'IC-21, UNITSV, RECNSV, IOST
CALL EXIT
END IF
ERROR CHECK #18:
A formatted write of the common block TSHDSV writes the time step header record of the STATE VEC-
TOR file. The I/O status of the formatted write statement is stored in the variable IOST. IOST is tested; if
IOST does not equal zero, an error has occurred on the write operation. A message is written to the log to
indicate that an error has occurred. The message contains the subroutine name, the unit number, the record
number, and the I/O status value. The program exits by a call to the system subroutine EXIT.
F-93
-------
CODE FOR ERROR CHECK #18:
c
RECNSV = RECNSV + 1
URITE(UNITSV, FMT = 1031, IOSTAT = IOST)
& IDATSV, ITIMSV, IELPSV, ISTPSV
1031 FORMAK1X, 15, 1X, 16, 1X, 18, 1X, 14)
C
IF (IOST .NE. 0) THEN
WRITE(LUNOUT, 2017) UNITSV, RECNSV, IOST
2017 FORMAT(/ 5X, 'XXX T.S.H WRITE ERROR IN WRSTAV
& / 5X, 'UNIT NUMBER = ', 12. 5X, 'RECORD = ', 14
& / 5X, 'I/O STATUS = ', 14)
CALL EXIT
END IF
ERROR CHECK #19:
A formatted write of the common block TEXTPT writes the text pointers record to the STATE VECTOR
file. The I/O status of the formatted write statement is stored in the variable IOST. IOST is tested; if IOST
does not equal zero, an error has occurred on the write operation. A message is written to the log to indicate
that an error has occurred. The message contains the subroutine name, the unit number, the record number,
and the I/O status value. The program exits by a call to the system subroutine EXIT.
CODE FOR ERROR CHECK #19:
c
C write text pointers
C
RECNSV = RECNSV + 1
WRITE(UNITSV, FMT = 1033, IOSTAT = IOST)
& BIGMPT, LILGPT,
& BCPSPT, BMPSPT, BTPSPT, CNPSPT, ICPSPT,
& RDBCPT, RDBMPT, RDBTPT, RDCNPT, RDICPT,
& URCNPT, WRSVPT
1033 FORMATOX, 14(13, 1X))
IF (IOST .NE. 0) THEN
WRITECLUNOUT, 2019} UNITSV, RECNSV, IOST
2019 FORMAT(/ 5X, 'XXX TEXTPT WRITE ERROR IN WRSTAV
& / 5X, 'UNIT NUMBER = •, 12, 5X, 'RECORD = ', 14
& / 5X, 'I/O STATUS = ', 14)
CALL EXIT
END IF
ERROR CHECK #20:
A formatted write of the common block ROWSCT writes the row counters record to the STATE VECTOR
file. The I/O status of the formatted write statement is stored in the variable IOST. IOST is tested; if IOST
does not equal zero, an error has occurred on the write operation. A message is written to the log to indicate
that an error has occurred. The message contains the subroutine name, the unit number, the record number,
and the I/O status value. The program exits by a call to the system subroutine EXIT.
F-94
-------
CODE FOR ERROR CHECK #20:
c
C write row counters
C
RECNSV = RECNSV + 1
WRITE(UNITSV, FMT = 1035, IOSTAT = IOST)
& BMPSRW, RDBMRU, RDBTRU, RDCNRW
1035 FORMAK1X, 4(13, 1X))
IF (IOST .NE. 0) THEN
WRITE(LUNOUT, 2021) UNITSV, RECNSV, IOST
2021 FORMAT(/ 5X, 'XXX ROW COUNTERS WRITE ERROR IN WRSTAV'
& / 5X, 'UNIT NUMBER = ', 12, 5X, 'RECORD = ', 14
& / 5X, 'I/O STATUS = ', 14)
CALL EXIT
END IF
ERROR CHECK #21:
A formatted write of the common block TSTEPS writes the scenario time record to the STATE VECTOR
file. The I/O status of the formatted write statement is stored in the variable IOST. IOST is tested; if IOST
does not equal zero, an error has occurred on the write operation. A message is written to the log to indicate
that an error has occurred. The message contains the subroutine name, the unit number, the record number,
and the I/O status value. The program exits by a call to the system subroutine EXIT.
CODE FOR ERROR CHECK #21:
c
C write scenario time
C
RECNSV = RECNSV + 1
WRITE(UNITSV, FMT = 1037, IOSTAT = IOST)
& MDDATE, MDTIME, MDELAP, MDSTEP,
& BCDATE, BCTIME, BCELAP, BCSTEP,
& BMOATE, BMTIME, BMELAP, BMSTEP
1037 FORMAU1X, 3(15, 1X, 16, 1X, IB, 14, 1X)>
C
RECNSV = RECNSV + 1
WRITE(UNITSV, FMT = 1037, IOSTAT - IOST)
& BTDATE, BTTIME, BTELAP, BTSTEP,
& CNDATE, CNTIME, CNELAP, CNSTEP,
I ICDATE, ICTIME, ICELAP, ICSTEP
IF (IOST .NE. 0) THEN
WRITEaUNOUT, 2023) UNITSV, RECNSV, IOST
2023 FORMAT(/ 5X, 'XXX SCENARIO TIME WRITE ERROR IN WRSTAV
& / 5X, 'UNIT NUMBER = ', 12, 5X, 'RECORD = ', 14
& / 5X. 'I/O STATUS = ', 14)
CALL EXIT
END IF
F-95
-------
This page is intentionally left blank.
-------
INDEX
Chemical species
mapping, 39
Error checking in
ADATE, F-2
ASORT, F-2
BCPRCS, F-2
BIGGAM, F-3
BMPRCS, F-5
BTPRCS, F-6
CELLM, F-6
CLOCK1, F-7
CLOCK2, F-7
CNPRCS, F-7
CPUTIM, F-7
DATTIM, F-7
DUMPHD, F-9
FSKIP1, F-10
GTILDE, F-13
HSTEP, F-13
ICPRCS, F-13
INDEX!, F-14
INIRUN, F-14
IOCL, F-14
JFILE2, F-14
JFILE5, F-15
JFILE6, F-16
JULIAN, F-17
JUNIT, F-17
LILGAM, F-18
NEWICS, F-19
OPBCON, F-23
OPBMAT, F-25
OPBTRK, F-25
OPCONC, F-26
OPICON, F-28
OPSTAV, F-30
OPWRCN, F-35
ORSPBC, F-39
ORSPBM, F-40
ORSPIC, F-41
POBCON, F-42
POBTRK, F-43
POCONC, F-44
POICON, F-45
POMXBM, F-47
POSTAV, F-48
PQ1, F-49
PQCOEF, F-50
PRGSMY, F-50
RATED, F-51
RDBCON.F-51
RDBMAT,F-54
RDBT, F-57
RDBTRK, F-61
RDCHAR, F-63
RDCONC, F-64
RDFILE,F-66
RDHDBM, F-66
RDICON, F-72
RDMXBM, F-74
RDSTAV, F-76
RTPHO, F-79
RTSET.F-79
RUNMGR, F-79
TIMER, F-83
WRCHAR, F-83
WRCONC, F-83
WRFILE, F-85
WRSTAV, F-85
Include files (.EXT)
ADVSFL, E-3
BCFILE, E-3
BGBCFL, E-3
BGBTFL, E-4
BGICCN, E-4
BMCOEF.E-5
BMFILE,E-6
BTFILE,E-7
CHEMIN, E-7
CHEMSW, E-8
CNFILE,E-8
COMPAQ E-8
DIMENS, E-2
ERRG, E-9
FLNAMS, E-9
GTCOEF.E-10
HDFMTS, E-10
HDSTAV,E-11
HEADBQE-12
HEADBM, E-13
HEADBT, E-14
HEADCN, E-15
HEADIC, E-16
HEADIN, E-17
HSTEPS, E-18
ICFILE, E-18
LGBMFL, E-19
LILGSP, E-20
LUNITS, E-20
LVNAME, E-21
NDXPC, E-21
NROOTS, E-21
REGION, E-2
RKLEVS, E-22
ROWSCT, E-22
RTCONS, E-23
RTSHBC, E-23
RTSHBM, E-24
RTSHBT, E-24
RTSHCN, E-25
RTSHIC, E-25
RUNTMS, E-26
SPNAME, E-26
STOPFL, E-26
SUBID, E-27
TEXTPT, E-28
TILDE, E-28
TSHDBC, E-29
TSHDBM, E-29
TSHDBT, E-29
TSHDCN, E-30
TSHDIC, E-30
TSHDMD.E-30
TSHDSV, E-31
TSTEPS, E-31
UNITIO, E-32
ZADVSL, E-32
Input files
BCON.27
BMAT.34
BTRK,48
ICON, 54
NEWICON, 60
PROG, 68
RESTRT, 69
Index-1
-------
Output file
CONC, 90
Procedures, 3,69
Processes, 3,69
Program inversion, A-4
Restarting the core model
RESTRT, 70
ROM
chemical species in, 8
CPU time on a VAX 8650,122
episode, 3
horizontal grid resolution, 3
IBM 3090 clock time, 2
IBM 3090 CPU time, 14
modeling domains, 4
principal components, 10
start time of the Core Model,
13
time zone, 3
vertical layers, 5
Run manager, 16
Start time, 13
Starting the core model
RESTRT, 70
State vectors, A-4
Subprograms
ADATE,23
ASORT, 23
BCPRCS, 18
BLKMOD, 24
BMPRCS, 20
BTPRCS, 19
CELLM, 23
CLOCK1,23
CLOCK2,23
CNPRCS, 21
CPUTIM, 24
DATTIM, 23
DUMPHD, 21
FSKIP1,23
GTILDE, 21
HSTEP, 21
ICPRCS, 17
INDEX1, 23
INIRUN, 16
IOCL, 23
JFILE2,23
JFILE5,23
JFILE6.23
JULIAN, 23
JUNIT.23
NEWICS, 24
OPBCON, 18
OPBMAT, 20
OPBTRK, 19
OPCONC, 22
OPICON, 17
OPSTAV, 22
OPWRCN, 21
ORSPBC, 19
ORSPBM, 21
ORSPIC, 18
POBCON, 19
POBTRK, 19
POCONC, 22
POICON, 18
POMXBM, 21
POSTAV, 22
PQ1,21
PQCOEF, 21
PRGSMY, 24
RATED, 21
RDBCON, 18
RDBMAT, 20
RDBT, 19
RDCHAR.23
RDCONC, 22
RDFELE,24
RDHDBM, 20
RDICON, 17
RDMXBM, 20
RDSTAV, 22
RTBTRK.19
RTPHO, 21
RTSET, 21
TIMER, 24
WRCHAR.24
WRCONC, 21
WRFILE.24
WRSTAV, 22
Variable-state subroutines, 3,69,
A-4
VAX to IBM translation, 14
Index-2
------- |