COMPUTER MAPPING OF COAL RESERVES
BY SULFUR LEVEL:
STUDY AREA REPORT
LABORATORY FOR COMPUTER GRAPHICS
AND SPATIAL ANALYSIS
Contract No. CPA 70-16 Harvard Graduate School of Desig
Air Pollution Control Office Cambridge, Massachusetts
-------
COMPUTER MAPPING OF COAL RESERVES
BY SULFUR LEVEL
' Principal Investigators
John C. Goodrich
Howard T. Fisher*
Study Area Investigators
David Sheehan - Phases I and II (Appalachian Coal Region)
Timothy Murray - Phase III (Allegany and Garrett Counties)
Technical Guidance on Geology
Michael Woldenberg
Research Assistants
Whitmore John
Gail Howrigan
' Computer Programming
Nancy Peyton
Se cretarial and Administration
Lois Kramer
Tina McGeary
Report Preparation
Albert Davis
Laurence Yont
-------
PREFACE
The Air Pollution Control Office (APCO) of the
Environmental Protection Agency is interested in ways
of displaying the quantity of coal available at various
sulfur levels. It was felt that computer mapping of the
coal data might provide a relatively quick and inexpensive
tool.
Data was provided by the Bureau of Mines for
study areas in the Appalachian Coal Region and, in
particular, in a two county area in Maryland. Values
for sulfur content, bed thickness, and quantity of coal
were manipulated and mapped using several computer
mapping programs available at the Laboratory for Computer
Graphics and Spatial Analysis at the Harvard Graduate
School of Design. Although the work was carried out for
these study areas, the main emphasis of the project was
on providing a tool with general applicability to coal data
anywhere in the United States.
The problems of obtaining sufficient reliable data
for mapping cannot be emphasized strongly enough. At
first glance it would appear that there is a great deal of
coal data available. However, when the data are broken
down by type, coal seam, and location, there is actually
very little useable information available at the present
time for all the types of data needed for a study such as
this.
It is felt, and has been stressed throughout, that
more meaningful results could have been attained if there
had been a more equal distribution of data points and a
-------
more equal number of analyses taken at these points.
All of us were aware that this might be a problem at
the outset of the project and our fears have been
substantiated. We hope that pointing this out is the kind
of constructive criticism that may prove most helpful in
overcoming deficiencies in data acquisition so that more
meaningful results may be attained in future work.
Of significant importance, however; is the potential
applicability of each of the types of maps produced and
their usefulness to APCO as adequate data are made available.
The Study Area Report is divided into three main
sections, entitled:
I BACKGROUND
II STUDY AREA FINDINGS
III APPENDICES
The first section covers the problem definition, computer
program used, study area selection and data acquisition,
and the graphic techniques used. The second section
discusses the study areas and data used, and the computer
maps produced in some detail. The final section contains
the step by step description of the mapping procedures used
for the study areas, the references, and the technical
aspects of program development.
The Study Area Report does not contain any
speculative information as to possible future uses of the
techniques nor does it include all the computer program
manuals and specifications necessary for application of
-------
the techniques to other coal data and study areas. These
manuals and related information are contained in a companion
report on General Documentation.
As stated in the problem definition, the project was
designed to demonstrate the capabilities of the SYMAP Computer
Mapping Program to display the sulfur levels of coal and
the quantities of coal available at various sulfur levels. The
types of maps desired and the criteria for selecting study
areas for demonstration are discussed.
The requirements and specifications for each of the
three computer mapping programs are explained in the next part
as well as the necessary output devices i and the significant
aspects of program development which took place during the
project. This technical section serves as an introduction to
the use of the program found in Section III (Appendices) and
Volume Two.
Study area selections and data acquisition are presented
in terms of the preliminary examination of areas and a discussion
of the actual study areas chosen: the four-state Appalachian
Coal Region, and Allegany and Garrett Counties in Maryland.
The final part of the first section is concerned with
graphic techniques, particularly the contour intervals and
symbolism used for mapping. The problems of map size,
particularly for final reproduction and display, are briefly
discus sed.
The second section is concerned with the actual study
area procedures and findings. The data used for the two study
areas is discussed in the first part: (i) the sulfur content data
-------
aggregated by counties for the Appalachian Coal Region, (ii)
sulfur data aggregated by towns for this area, (iii) sulfur and
bed thickness data by mine location for Allegany and Garrett
counties in Maryland, and (iv) washability data for Maryland
by mine location. _ __.
j
The computer maps produced for each study area and
set of data are discussed in some detail, followed by the
conclusions and recommendations resulting from the case study
areas. The material provides the background for the more
detailed mapping technique discussions of Section III (Appendices)
and Volume Two: General Documentation.
Section III (Appendices) contains the step by step
computer mapping procedures for each study area. It also
contains the references used in the project and a discussion
of the significant programming developments, including the
use of statistical routines, special data handling procedures,
and the changes made to the mapping programs for this project.
We would like to acknowledge the invaluable support and
advice of Russell Flegal, Project Officer, Office of Program
Development, Air Pollution Control Office, Environmental
Protection Agency, and S. J. Aresco, Coordinator, Energy
Data Bank, Bureau of Mines, U. S. Department of the Interior.
We would also like to thank Jerrold G. Thompson, Branch
of Computer Sciences and Engineering, Bureau of Mines,
U. S. Department of the Interior, Harry Buckley, Director,
Maryland Geological Survey, and William E. Edmunds, Head
Coal Geologist, Pennsylvania Bureau of Topographic and
Geologic Survey for their assistance.
-------
All unpublished data used in the project were supplied
by the Air Pollution Control Office and the Bureau of Mines.
Cambridge, Massachusetts John C. Goodrich
April 1971 Project Director and Editor
-------
STUDY AREA REPORT
Table of Contents
PREFACE
TABLE OF CONTENTS
LIST OF ILLUSTRATIONS
I. BACKGROUND
1. Problem Definition
2. SYMAP and Related Programs
2.1. The SYMAP Program
2. 2 The GRID Program
2.3. The SYMVU Program
2.4. Output Devices
2. 5. Development of Programs
3. Study Area Selection and Data Acquisition
3.1. Preliminary Examination of Areas
3.2. Appalachian Coal Region
3.3. Allegany and Garrett Counties
4. Graphic Techniques
4.1. General Symbolism Considerations
4. 2. Symbolism Used in the Project
4. 3. Reproduction Considerations
H. STUDY AREA FINDINGS
1. Discussion of Study Areas and Data Used
1.1. Appalachian Coal Region - Phase I
1.2. Appalachian Coal Region - Phase II
1.3. Allegany and Garrett Counties - Phase III
Stage One
1.4. Allegany and Garrett Counties - Phase
III, Stage Two
2. Discussion of Graphics
2. 1. Summary of Phases I and II
2. 2. Map Series A
2. 3. Map Series B
2.4. Map Series C
2. 5. Map Series D
2.6. Summary of Phase III
2. 7. Map Series E
2. 8. Map Series F
2. 9. Map Series G
2. 10. Map Series H
-------
3. Conclusions and Recommendations
3.1. Study Area and Data Considerations
3.2. Mapping Considerations
3.3. Applications
III. APPENDICES
1. Phases I and II: Appalachian Coal Region
1.1. Input to SYMAP
1.2. Base Map Preparation
1.3. Phase I Data Bank
1.4. Phase I Data Point Placement
1.5. Phase II Data Bank
1.6. Phase II Data Point Placement
1.7. Subroutine FLEXIN
1.8. Map Execution
2. Phase III: Allegany and Garrett Counties
2. 1. Input to GRID
2. 2. Base Map Preparation
2.3. Phase III Data Banks
2.4. Phase III Data Point Placement
2. 5. Subroutine FLEXIN
2.6. Map Execution
2. 7. Input to SYMVU
3. General References
3.1. Appalachian Coal Region References
3. 2. Allegany and Garrett County References
3. 3. Other References
4. Program Development
4.1. SYMAP Printout Statements
4.2. SYMAP Subroutine MANIP
4.3. Legends with GRID
4. 4. Use of Statistical Routines
4. 5. Data Handling Procedures
4. 6. Displaying Number of Analyses on
Base Maps
-------
List of Illustrations
Figure 1: Map of Study Areas
Tables:
1-1
1-2
1-3
Study Areas and Data
Levels Used for Each Map Series
Symbolism Used for Mapping
Figure 2: Sample Plot of Sulfur Content Versus Yield
Tables:
II-1
n-z
II-3
II-4
Data Banks Used
Summary of Data Available for Phase III
Phase I and II Maps
Phase III Maps
Figures:
A-l thru A-8 Phase
B-l thru B-21 Phase
C-l thru C-7 Phase
D-l thru D-7 Phase
E-l thru E-ll Phase
F-l thru F-V Phase
G-l thru G-10 Phase
H-l thru H-7 Phase
I - Pittsburgh Bed
II - Pittsburgh Bed
II - Middle Kittanning Bed
II - Detail Areas and Difference Maps
III (Stage One) - Sulfur Content
III (.Stage One) - Thickness and Quantity
III (Stage Two) - Sulfur Content
III (Stage Two) - Quantity
Tables:
III-l Phase I Pittsburgh Bed - Raw Data Points
IH-2 Phase I Pittsburgh Bed - Raw Values
III-3 Phase I Pittsburgh Bed - Washed Data Points
III-4 Phase I Pittsburgh Bed - Washed Values
III-5 Phase II Pittsburgh Bed - Raw Data Points
III-6 Phase II Pittsburgh Bed - Raw Values
III-7 Phase II Pittsburgh Bed - Washed Data Points
III-8 Phase II Pittsburgh Bed - Washed Values
III-9 Phase II Middle Kittanning Bed - Raw Data Points
III-10 Phase II Middle Kittanning Bed - Raw Values
III-11 Subroutine FLEXIN - Phases I & II
in-12 F-MAP package
III-13 FLEXIN for Subroutine MANIP
111-14 Upper. Freeport Bed - Sulfur Content
111-15 Upper Freeport Bed - Thickness
III-16 Upper Bakerstown Bed - Sulfur Content
III-l7 Upper Bakerstown Bed - Thickness
111-18 Upper Freeport Bed - Washability
-------
Tables:
III-19 Superimposed Data Points
III-ZO Subroutine FLEXIN - SYMAP
111-21 Subroutine FLEXIN - GRID
111-22 Control Cards for SYMVU
-------
L BACKGROUND
1. Problem Definition
2. SYMAP and Related Programs
3. Study Area Selection and Data Acquisition
4. Graphic Techniques
-------
1. PROBLEM DEFINITION
The Air Pollution Control Office (APCO) of the Environ-
mental Protection Agency (EPA) is interested in ways of dis-
playing the quantity of coal available at various sulfur levels.
This is in conjunction with efforts to find ways for supplying
low-polluting fuels to areas of the country adopting emission
control regulations. It was felt that a relatively quick and
inexpensive tool such as computer mapping could prove to be
useful in the process.
The SYMAP Computer Mapping program of the Laboratory
for Computer Graphics and Spatial Analysis enables one to express
clearly and visually information that has generally been available
only in extremely long and cumbersome lists of tabular data.
The Laboratory has demonstrated the unique and valuable appli
cations of the SYMAP and related programs in numerous projects
including a demonstration grant running from 1968 to 1970
from APCO (68A-2405D) undertaken to show the application of
computer graphics to air pollution analysis and control. It was
demonstrated that as more and more air quality data becomes
available in computer usable form, a tool such as computer map-
ping can be very useful as an aid in air pollution studies, par-
ticularly for displaying changes in air quality over time
The scope of work for investigating the computer mapping
of coal reserves by sulfur level emphasized several major points.
It was hoped that the proposed work would demonstrate the capa-
bilities of SYMAP to display the sulfur levels of coal and the
quantities of coal available at various sulfur levels. As a result
of the work a new and valuable tool would be provided for the
-------
decision-making process which is concerned with locating low
sulfur coal for the potential control of air pollution.
Recognizing the need for a tool with general applicability,
the program developed was to be demonstrated for a specific
area of the country with applicability to coal data anywhere in
the United States. The necessary data was to be provided by
APCO in cooperation with other Federal agencies, including the
US Bureau of Mines, and state and local agencies, including
state mining and geological survey departments. Whenever
possible this data was to be provided in a computer form
compatible with the needs of the SYMAP program.
The study area was to be selected by reviewing available
analytical, washability, and geological data for at least three
possible areas. The criteria for selecting a study area included
(i) a reasonable and manageable number of
seams, and,
(ii) adaquate availability of data for each seam
for sulfur content, seam thickness and extent, and
washability.
The area selected was to be as extensive as possible,
subject to the above criteria. It was suggested that the output
maps be approximately three feet by four feet.
Computer maps were to be prepared for the selected
study area showing the following:
--Isolines of sulfur level in percent sulfur by seam, in class
intervals suggested to be 0.0 to 0.5 percent; 0. 5 to 1.0, 1.0
to 1.5, 1.5 to 2.0, 2.0 to 2.5, 2.5 to 3.0, 3.0 to 5.0, and
greater than 5%;
-------
--Quantities in tons per acre of mineable reserves of raw
coal by seam with sulfur contents less than 1%, assuming
current mining practice. APCO was to provide the data in
a form that would explain the assumed current mining practice
to be used. When more than one seam existed in an area,
maps were to be made for each seam or several selected seams
and a composite summing all of the selected seams 20" or more
in thickness was to be made;
--Isolines, as the coal lies in the ground, showing the sulfur
level the coal could be washed to at 65% yield by seam. The
class intervals of sulfur level were suggested to be 0 to 1%,
1.0 to 1.5, 1.5 to 2.0, 2.0 to 3.0, 3.5 to 5.0 and greater
than 5%;
--Quantities in the ground in tons per acre of washed marketable
coal of less than 1% sulfur content, assuming, first, the
economics of present coal preparation and washing techniques in
mining practice, and, second, that the market will take incre-
mental price increases. APCO was to provide the data in a
form that would explain the various preparation, washing, and
mining techniques to be considered for both the current practice and
the incremental price increase cases.
The project was to involve any necessary adapting of the
SYMAP techniques and development of the necessary computer
programs and analysis to produce these low sulfur coal availability
maps. In addition, the feasibility and applicability to APCO's
needs of' utilizing various three-dimensional graphical techniques in
conjunction with the two-dimensional SYMAP portrayal was to
be investigated. Finally, the technical report was to include a
general discussion of the program, permitting persons to apply
the general program to other coal areas, and a discussion of the
limitations on the use of the program.
-------
This emphasis on documentation was thought to be very
important, particularly because of the lack of data for adequately
demonstrating the techniques for the study areas to be examined.
Even if no definitive conclusions can be drawn for the study
areas, the general applicability of the program to any areas of
the country where reasonable data can be made available is a
significant contribution in itself.
Data on sulfur content and seam thickness are available
for many coal beds in the country. From the data one can
determine the quantity of coal presently available at various
sulfur levels. Wherever information on washability is obtained,
a series of maps can be made which show the quantities of coal
that may be obtained at various yields. Decisions can then be
made as to what cost should be incurred to supply low sulfur
coal to areas needing this type of fuel.
Such a computer mapping approach is important for several
reasons:
(i) it can be automatically tied to large scale computer-based
data banks, such as exist for coal and related data;
(ii) alternative situations can be postulated and mapped, with
comparisons made quickly and inexpensively between the alter-
natives; and,
(iii) as more and more data on coal reserves, mining practices,
and demand requirements are made available, the tool can be
immediately refined to give more meaningful and accurate
results.
-------
Not all of the objectives of the project could be fully
met. We were able to display the sulfur levels of coal for
a number of seams at different scales depending upon the
study area size. We were also able to display the quantity
of coal available for less than 1% sulfur content, but the lack
of data prevented us from calculating quantity for the study
area with the accuracy hoped for.
The program was demonstrated for specific areas with
applicability to coal data anywhere in the United States. How-
ever, it is likely that many of the problems of map accuracy
and data availability encountered with the study area will be
found with other parts of the country in the near future.
The study area selection criteria were somewhat
academic in that adequate data were not available for any
study area--for all variables concurrently and more than
one coal seam. The four-state study area chosen was quite
extensive but the scale of the mapping--26" wide maps--
resulted in a reduction to 25% of original size for printing
in this report and a great deal of lost detail.
The isolines for sulfur content used were almost the
same as those initially suggested. However, only a few of
these isolines were actually relevant, in many cases.
The washability and market price information had to
be re-stated in terms of sulfur content versus the expected
yield of coal. This did not significantly affect the results,
although the relationship of quantity to cost was thereby
eliminated from the analysis.
-------
In general, the computer mapping techniques were
demonstrated for the study area, these study areas are
representative of data and problems encountered elsewhere,
and the techniques can now be applied to other areas of the
U.S. where coal data are available, subject to the constraints
and accuracy of the required data discussed at length in this
report.
-------
2. SYMAP AND RELATEt) PROGRAMS
Three computer mapping programs were used in this
project. The SYMAP program was used for the majority of
the maps produced, particularly for maps of sulfur content and
bed thickness. The GRID program--which was adapted from
SYMAP to specifically handle data input in terms of a regular
grid--was used in all cases where maps or data sets were compared,
as in determining the quantity of coal at various levels of sulfur
content. Finally, in a few cases the SYMVU program was used
to display the three-dimensional representation of the same surface
shown in two-dimensions with the SYMAP or GRID programs.
All three programs were developed principally at the Laboratory
for Computer Graphics and Spatial Analysis.
2- *• THE SYMAP PROGRAM
SYMAP generates map displays showing the values of
spatially distributed data, according,to their actual geographic
location on a base map. Combinations of standard computer
printout symbols are used to print a scale of tone from black
to white which can correspond to a data value range from large
to small. Mathematical computations can be performed on the
data if necessary before mapping. Options are provided for the
rescaiing of raw data and for variations in symbolism.
For this project values for sulfur content, bed thickness,
and quantity of coals in tons per acre were displayed for each
of several coal seams. Uniform scales of symbolism were used
so that maps in a series could easily be compared. In many cases
mathematical computations were performed prior to mapping,
particularly to determine the quantity oi coal available for a
given sulfur content, given various yields from washing.
-------
The SYMAP program is written in Fortran IV. It can
be used on a computer system equivalent to the IBM 360-40, or
CDC3600. Computer storage requirements are at least 32, 000
words, or 128,000 bytes. Such systems are available in most
commercial, government, and university computation centers.
All work on this project was done using Fortran IV on
an IBM 360-65 at the Harvard Computing Center. Version 5. 15 of
the SYMAP program was used which requires 15OK bytes of
storage. Certain operations--comparing surfaces using a newly-
written subroutine for SYMAP--required 200K bytes.
The program can produce conformant, contour, proximal,
and dot maps. In conformant maps, predefined areas or zones
are filled with the symbol tone assigned to their data values.
Contour or isoline maps show the variation of the data over the
whole study area; the display is analogous to the contour bands
used to show changes of elevation on topographic maps. Proximal
maps use the nearest-neighbor principle, with any non-data position
being given the value of its nearest data point. They are similar
in appearance to conformant maps. The program will also produce
dot maps, using a variety of symbols. The maps produced in this
study were ail of the isoiine or contour type. This was because
it was felt that the data being mapped--sulfur content, thickness,
and quantity--all tend to vary continttausly over a coal bed rather
than to exhibit fixed values for data zones.
To produce computer maps, a deck of punched cards
must be prepared as input to the computer. This deck consists
of a number of "packages," each covering a specific category
of information. The most basic of these packages, with a
brief explanation of their general purpose, are listed below.
-------
The MAP package is always required; use of the other pack-
ages is determined by the type of map desired.
The MAP package specifies, below the map, an appropriate
title for the identification of each separate map that is to be
run. It then instructs the computer to make each map by using
certain electives that may be useful. It is not necessary to
specify any of these electives since each has a standard default.
The availability of these electives is a major feature of the
SYMAP program and adds greatly to its flexibility. Some of
the electives are:
(i) Size: used to specify the overall dimensions of the
rectangular border surrounding the study area on the output
map;
(ii) Content: used to specify the portion or extent of the
study area (as shown on the source map) to be mapped by
the computer; the scale of the map to be produced can be
changed, or a portion of the study area may be shown, at
any scale.
(iii) Number of Levels: used to specify the number of levels,
or class intervals, (up to twelve) into which the total range of
the data is to be subdivided for mapping purposes; another
elective can be used to specify the value range intervals for
these levels;
(iv) Symbolism: used to specify the black-to-white tone
symbolism for the levels as well as symbolism for the data
point locations, contour lines, and areas outside the study area;
and,
(v) Text: used to provide supplementary textual information
below the map.
The OUTLINE package is used to describe the outline of
the study area by specifying the x-y coordinate locations of
the outline vertices. For this project the outline
-------
was generally the seam outcrop line or the line of 28" bed
thickness.
The DATA POINTS package is used to give the positions
of the data points to which data are to be related, by specifying
their coordinate locations. Data, points may be either points for
which data are available or centers of areas ("data zones") for
which data are available. When warranted by the nature of the
study, other "centers" may be used. For this project the data
points were the centers of counties or towns for which data on
sulfur content were known or the actual or assumed location of
mines, mine openings, or core samples.
The LEGENDS or OTOLEGENDS package causes supple-
mentary information to appear on the face of the map by specifying
coordinate locations and content. Legends include such things as
political boundaries, location nameSja scale and north arrow and
the title block. The principal legends used in this project were
state and county boundaries, relevant place names, and the scale.
These three packages--OUTLINE, DATA POINTS , and
OTOLEGENDS are used (together with a MAP package which is
always required) to produce a 'base map1 for testing purposes.
The base map and consequently these packages generally remain
the same for a whole series of maps. Once the base map has
been determined and tested it is necessary to relate the appropriate
Values to the data points--by use of the VALUES package--and
select any options desired together with the title with the MAP
package.
-------
The VALUES package is used to assign the appropriate
vaiues to the data points. In this project this might be the
assigning of values for sulfur content averaged for all samples
in a county for the seam in question to the centroid of the
counties which would be the data point locations. Very often a
series of maps is to be run from a large data bank of variables.
In this case an extremely flexible user-subroutine is employed in
conjunction with the VALUES package to select the values for
the particular variable to be mapped. Raw data given to the
computer may be related, manipulated, weighted, and aggregated
in nearly any manner desired with this subroutine to produce the
values for mapping. The subroutine may also be used in con-
junction with the other packages, often to transform coordinate
locations.
Immediately ahead of the packages of data cards described
above, certain introductory cards (as required by the particular
computing center being used) will need to be provided--together
with a copy of the SYMAP program on cards, tape, or disk.
Producing isopleth maps with a computer program such
as SYMAP has numerous advantages over producing maps by hand.
First, if one is producing a considerable number of maps computer
mapping is a much faster, less expensive, and more accurate
means of displaying spatially variable data. For this project and
the general application of the techniques used, a considerable
number of maps would be produced in a series. The very nature
of computer mapping eliminates the drudgery of making a great
number of maps by hand.
Second, a wide variety of options are available to the
computer map maker that are not possible with maps made by
-------
hand. The user of SYMAP can experiment with different
symbolism and shading to achieve the desired tones; he can use
various contour level intervals to portray different meanings about
the data, and he can add or delete political boundaries, rivers
and legends by changing only a few cards from map to map.
When mapping by hand one cannot afford to make several maps
and select only one of them. Because of the time and financial
constraints involved, one must often settle for a finished product
that is less than satisfactory and may not show the information
desired.
Third, the human error often encountered in the associating
of values with the appropriate data points is virtually eliminated
once the DATA POINTS and VALUES packages have been estab-
lished and verified to have a 1:1 correspondence. There is no
chance of transfer ing data from a table to the wrong data point
on the map as is often the case with mapping by hand. In
addition if a data bank is used which contains the data points and
all their associated values, errors for the whole series of maps
can be eliminated at the outset quickly and efficiently.
Finally, the contouring algorithm in SYMAP avoids the
subjective interpolation error so inherent in the decisions of
individual hand mappers or cartographers. SYMAP not only
standardizes this contouring process but also greatly reduces the
painstaking time and energy normally involved.
2- 2- THE GRID PROGRAM
The GRID program is a modified version of SYMAP
specifically designed for displaying data collected on the basis
of geographic grid cells. Each grid cell may be shown on the
computer-produced map as a single character or group of
-------
characters. Whereas SYMAP accepted information at irregularly
spaced data points such as mine locations, and then interpolated
values at all points on an assumed surface, the GRID program
requires that a value be read in for every grid cell. These values
are then displayed directly without interpolation. The GRID
program was used in this project to determine quantity of coal
at each grid cell mapped (where the sulfur content was less than
1%) based on the previously mapped values for bed thickness
interpolated from data points with the SYMAP program. The
program was also used to produce composite maps of quantity
of coal for more than one bed; here the quantity for each of the
beds was summed at each of the grid cell locations.
The program displays its data using a line printer but
operates more efficiently and economically because of its special
purpose design, which eliminates the need for the relatively
expensive interpolation algorithm. For a user who does not need
the added flexibility of the SYMAP program, the GRID program
can provide substantial savings. The GRID program presently
operates on an IBM 360/40.
Version 3 of the program was used for this project. It
was run on an IBM 360-65 at the Harvard Computing Center
and required 150K bytes of storage.
To obtain a map, the user must provide three sets of
instructions and has the option of providing a fourth set. The
instructions are prepared in the following packages: Data Package
(usually a separate tape), Map Package, Irregular Outlines
Package (Optional), and Subroutine Flexin.
-------
The DATA package contains the data which generate the
graphic display. It is similar to the VALUES package in
SYMAP. The program is written for a maximum of 10,000 grid
ceils, but the Multiple Data Set Option permits the user to handle
unlimited numbers of data cells. For this project approximately
12, 000 grid cells were used. These were displayed as single
character cells allowing for direct comparability between the
SYMAP produced maps and the GRID produced maps derived from
them.
The MAP package permits the user to specify the precise
form of the map output in terms of various eiectives. It is
similar to the MAP package in SYMAP; many of the eiectives are
specified with exactly the same format.
The IRREGULAR OUTLINE package allows the user to
specify the boundaries of the study area when he is not dealing
with a grid which is rectangularly bounded. For this project
the same outline was specified as was used with the SYMAP
program.
No additional package having to do with legends was
necessary for this project. A special edited version of the SYMAP
program was used to create a data file containing the legends, and
an edited version of the GRID program was then used to read the
data file. Therefore, the same legends were used with both
GRID and SYMAP.
SUBROUTINE FLEXIN is the Fortran subroutine which
allows the user to specify the format of the data. It is identical
in use to the user subroutine of SYMAP; however, with the GRID
program the subroutine must always be used to specify the format
for reading in the values. For this project values created by
the SYMAP program were read in, compared or manipulated, and
resultant values mapped on a cell by cell basis.
-------
2. 3. THE SYMVU PROGRAM
The SYMVU program provides the user with oblique views
of a given surface.
SYMVU is especially well suited to be run in conjunction
with SYMAP or GRID which will generate the necessary matrix
values.
Version 1. U of SYMVU was used for this project. It was
run on an IBM 360-65 at the Harvard Computing Center and
required 200K bytes of storage. Only a few SYMVUs were run
of surfaces mapped with SYMAP and GRID as a demonstration
of the applicability of the program.
The user of SYMVU needs to supply a matrix of surface
values, and a set of parameters specifying the type of oblique
view desired. The basic options available to the user are:
(i) the oblique view may be orthographic or in perspective;
(ii) the viewing angle may be varied in both the vertical (z)
direction and horizontal (x, y) plane;
(iii) the lines that SYMVU generates to define the surface of the
three-dimensional view may follow down the columns, across the
rows, or through the diagonals of the input matrix;
(iv) the size is specified by width and height;
(v) treatment of the non-study area may consist of deleting
it, darkening it, or setting it to a value of zero.
-------
2.4. OUTPUT DEVICES
The maps produced by either the SYMAP or GRID
program are generated by a standard chain line printer.
The printer used for this project was an IBM 1403 at the
Harvard Computing Center. The printer is capable of pro-
ducing 132 characters on a line; at 10 characters per inch,
this means a map 13 inches wide, with two characters for
the vertical borders. The printer produces eight lines per
inch with these mapping programs.
If a map is too large to fit on one sheet of paper,
the SYMAP program will subdivide a map into 13 inch sections.
Since this is frequently the case, a simple splicing job is
required. The GRID program has a similar option for pro-
ducing as many map sheets as are required; generally each
map sheet is associated with a separate data set. With these
programs a special carriage control tape is used to assure
the proper line spacing for each page; in this way the normal
spacing at the top of a page does not occur in the middle of
a map. The programs allow one to print up to four over-
printed characters to achieve the desired gray scale for
mapping.
The three-dimensional views produced by the SYMVU
program are generated by a pen plotter. The plotter used
for this project was an 11 inch CALCOMP plotter at the
Harvard Computing Center. The vertical size of the plot is
governed by the height of the paper which is 11 inches; there
is no constraint on the length of the plot because the paper is
on a continuous spool.
-------
2.5. DEVELOPMENT OF PROGRAMS
The concept, overall design, and mathematical
model for the SYMAP program were developed in the autumn
of 1963 by Howard T. Fisher, working at Northwestern
Technological Institute. Programming was carried out by
Mrs. O. G. Benson of the Northwestern University Computing
Center. Since that time many others have contributed ideas
and cooperated in bringing the program to its present state.
Major improvements in the program were made by two
Harvard students, Robert A. Russell and Donald S. Shepard,
who was responsible for the algorithms on spatial interpolation.
The GRID program was written at the Laboratory for
Computer Graphics and Spatial Analysis in 1968 by David
Sinton and Carl Steinitz. The SYMVU program was written
at the Laboratory in 1967 and 1968 by Frank J. Rens under
the direction of Howard T. Fisher. Since then it has under-
gone major changes and improvements.
Significant programming developments resulting from
this project included:
(i) capability to compare surfaces directly using
the SYMAP program;
(ii) incorporation of SYMAP-type legend options into
the current version of the GRID program; and,
(iii) development of data handling procedures to
quickly and inexpensively interface these three programs
-------
Program development is discussed in detail in Section
III (Appendices). In the SYMAP program changes were made
in the printout of the textual information below the map.
In addition, a new subroutine was written for comparing surfaces
having the same study area outline. Since the GRID program
can generally be used more efficiently for comparison of
surfaces, this SYMAP subroutine is only useful for the user
who wishes to perform the comparison with one program and
one submission.
The major changes to the GRID program involved the
rewriting of the legends routine that had been included in
previous versions of the program. This enables the same
kind of legends used with the SYMAP program to be displayed
on a map produced by GRID. With the inclusion of these
changes, the two programs are now completely compatible in
terms of graphic output.
A number of other important changes in the mapping
procedure were undertaken as part of the program development.
These include:
(i) The use of statistical routines to aid in the
choosing of the levels or contour intervals for mapping;
(ii) data handling procedures used to facilitate the
mapping, particularly when a large series of maps is
involved; and,
(iii) the use of a simple program to create legends
showing the number of analyses associated with each
data point.
-------
3. STUDY AREA SELECTION AND DATA ACQUISITION
The study area for which the computer mapping techniques
were to be demonstrated was to be selected after reviewing
the available analytical, washability and geological data for at
least three possible areas. The criteria for selecting an area
for detailed study included: (1) a reasonable and manageable
number of seams, and (2) adequate availability of data for
each seam for sulfur content, seam thickness and extent, and
washability. Subject to these criteria we wished to pick as
extensive an area as practicable so as to be able to develop
and test large-scale mapping techniques useful to APCO.
The review of possible study areas began with a literature
search and was followed by trips to the data bank and computer
facilities of the Bureau of Mines, since nearly all of the useful
available data is contained in their computerized data banks.
The initial review pointed to areas in Clearfield County,
Pennsylvania, and Garrett and Allegany Counties in Maryland
as providing the best data.
To avoid the use of detailed geological maps at the outset,
the computer mapping techniques were first tested for an 84
county four-state area in the Appalachian Coal Region. At this
scale readily available data aggregated by towns and counties
and uncomplicated estimations of bed outcrop lines can be used
to give an overview of the sulfur content of coals in a region.
Figure 1 shows this Appalachian Coal Region study area.
Phase I of the project involved mapping the sulfur content of
reserves by counties; Phase II involved mapping reserves by
towns.
-------
The maps produced show that Garrett and Allegany
Counties in Maryland have large deposits of low sulfur coal
and are worthy of closer examination. This area appeared to
have the best data and was chosen as the principal study area
for Phase III. Stage One of Phase III involved the mapping of
reserves by mines, while Stage Two was concerned with the
mapping of the washability data.
The Phase III study area is shown by a square box in
Figure 1. A summary of the study areas, coal beds, and
data for all these phases of the project is shown in Table I—I.
As expected, the greatest difficulty encountered in the
project was the obtaining of adequate data to work with. Sufficient
data to produce meaningful results never were available for the
detailed study area of Phase III, although this did not significantly
detract from the demonstrations of the applicable techniques. The
table below shows the relative amounts of time spent on data
acquisition, the data handling necessary prior to mapping, the
actual testing and perfecting of the mapping techniques, and the
analysis of the maps.
The table also shows the percent of time we would expect
to spend on each phase if the data were readily available.
acquisition handling mapping analysis
actual 30 25 30 15
expected 15 15 40 30
In addition the computer base map development and testing
takes place at the same time as data acquisition and handling;
in this way, the base maps and procedures are all prepared in •
time for the actual mapping phase.
-------
E t K
MERCER
BU T'LER J J- '
IBEAVER i \ ^ /
c LE ARFIELD
^CARROLL'
HOLMES I ^r|
,-
«.' »,,_•£ Afj
* j>°I
.GUERNSEY JSEtMONI
j>°IWASHINGTON
fMliSKINGUM"' I / CL'" ' ~' ~~ C
, , l-r-^.-i //^GREENE (
LPERR'Y! : I^OBIE I
. . .MONRO E %-
I I. ^ / ' -wununuHiiM •
;MORGA_N. Lr1"'-1! x-WE'7 r L ' V " ~". •
PI FACAkJTt" "^^ ^ t /
)A
! BE OFQRD
/ i
/SOMERSET '
EGANY
§!.., ["ORGAN, L^-H ^/WET.ri/^--N. / / /^
,..,,. "f"T J PLEASANTS X^ ^- \ '•, / |!&ARREtT< f
"",-' --.,-««».NC,ON v-v"vV^""N>;'"»'°l. /-^-'
-;..«.,k ,'XrJ-"'' //f\ ,wJ, iL^x.-X'
ON! 1~rJ S ./ ' Vvx /.'UCKI«COI.ANT;
'--• _ V '.' «-^^ -X I • -. , <-. / .HARDY
GIIMER
-J / y-V ^ J"'IMJ8~^ C-UPSHU^ /"
GALIIA/ \ '0 JROANE-IS^' />> I'-* ,- /f
~~, \ --•N'1' V ( -BRAXTON/ A /"""y
>c '^' i */\A -/X x' \ / X
. /PENDlETONj
Ll O G A
. •>* '
r'V.KANAWHAi,' s '
) V N I C H O L A S .^-IPOCAHONT^
> '-. /'-^ «~ > ' A
UOONE^/FAYETTE^, ^"
%— - f '^~'^* f'GREENBRIER
\-L
1 A L E I G Hi
40 20 0
40
Scale, miles.
80
N
Figure 1
Map of Study Areas
-------
APPALACHIAN COAL REGION
DATA AND BEDS MAPPED
PROJECT NAME
Reserves mapped by
counties.
Sulfur Content Quantity
Pittsburgh
Phase I
Reserves mapped by
towns
Pittsburgh
Middle Kittanning
Phase II
ALLEGANY AND GARRETT COUNTIES
Reserves mapped by
nines
Upper Freeport Upper Preeport
Upper Bakerstown Upper Bakerstown
Phase III
Stage One
Washability data mapped
by mines
Upper Freeport Upper Freeport
Phase III
Stage Two
Table 1-1
-------
The apparent delays in data acquisition were a result of
(1) data aggregation procedures and technical computer difficulties
with the necessary data banks for Phase II and (2) data
aggregation procedures and unavailability of data expected from
private sources for Phase III. The delays in the data handling
aspects in Phase I were due to expected trial and error attempts
to mate the data banks with the mapping programs, and in
Phase II to the need to determine all data point locations by eye
from detailed maps of the region.
In conclusion, much of the time that was devoted to data
acquisition, data handling, and the actual mapping for the study
areas chosen can be accounted for by the demonstration aspects
of the project; however, some of the time spent is related to
the nature and availability of data and, thus, will be a factor
in applying the techniques to other study areas in the near future.
This is particularly true of (1) the washability data, which are
available in sufficient quantity for only a few locations and were
not yet incorporated into the Bureau of Mines data banks, at the
time, (2) data point locations which had to be taken from maps by
eye, and (3) geological maps showing bed outcrops which must be
pieced together for large areas and are often incomplete for parts
of a region.
To successfully use computer mapping as a reliable tool,
the problems of data availability and form need to be overcome.
The mapping programs will require:
(1) information to easily and inexpensively develop base
maps, including bed outcrop maps;
(2) computerized data locations, in addition to the coal
data values themselves; and,
-------
(3) sufficient data, by seam, with adequate spacing of
data point locations; this is particularly true for the
washability data.
The following sections discuss briefly the preliminary
examination of areas, the Phase I and II Appalachian Coal
Region study area, and the Phase III Allegany and Garrett
Counties study area. The actual references for the literature
search and data sources, will be found under General References
in the Appendices (Section III).
-------
3. 1. PRELIMINARY EXAMINATION OF AREAS
At the beginning of the project our literature search
concentrated on Clearfield County, Pennsylvania, at the suggestion
of APCO. This is a low sulfur coal region worthy of detailed
study. ¥ery thorough geological maps and descriptions of the
area had been prepared (Edmunds; Crentz et al; Blaylock et al.« )>
however, our investigation revealed that there were not sufficient
coal analyses or washability data available for mapping.
The following quotations from a letter from William E.
Edmunds, Head Coal Geologist for the Pennsylvania Bureau
of Topographic and Geologic Survey, summarize aptly the problems
encountered with the Clearfield County data and with data for
most areas one might wish to study:
". . .all of the readily available coal analyses for the
northern Houtzdale area are included in (the) report. There are,
doubtless, many more analyses from that area (as well as
almost any other area), but they are widely scattered among
private and public sources. At any rate, assembling, verifying
and identifying them as to coal seam and geographic location
3
t is a very time consuming operation.
"My experience in this sort of study has been that the
greatest problem is the acquisition and verification of data. The
ratio of time spent to amount of raw data acquired is very high.
"I have no doubt that coal reserves and sulfur content can
be mapped out by computer methods, but the stratigraphic relations
of the various coals must be well understood and considerable
data on thickness and sulfur content available. The stratigraphic
-------
relations of the various coals in the northern Houtzdale quadrangle
are well in hand, but they happen to display complex splitting of
some coal seams into separate coals as well as extensive structural
faulting. In addition, I doubt that a sufficient number of sulfur
analyses are available.
"We have finished, but not published, the coal geology
from a number of adjacent quadrangles to northern Houtzdale,
but the same problems of complex geology and little sulfur data
exist there as well. In the southeastern quarter of the Houtzdale
quadrangle, we have had a number of recent coal analyses made,
but we have not completed the geologic map and the stratigraphic
and structural complexities are extensive.
In Washington County (south of Pittsburgh) we have had a
number of cooperative studies made by the U. S. Geological Survey,
which provide good maps and stratigraphic control in that area.
Here again I don't think there are enough sulfur analyses available
publicly. However, in this area there are a relatively few
large coal-producing companies who, I'm sure have copious records,
if they were willing to cooperate.
".... Your greatest difficulty is going to be finding an area
where the geology is relatively uncomplicated and sufficient
analyses available. "
The other area examined in detail through a literature search
was Garrett and Allegany Counties in Maryland (Boyd; Crentz
and Graser; Snyder and Aresco; Toenges et al). This was also
a low sulfur coal region worthy of study. Our investigation
revealed the following characteristics that made this region superior
to Clearfield County for the purposes of this study:
-------
(1) adequate source maps for geological conditions and
base map development;
(2) uncomplicated stratigraphic relationships and a
reasonable number of seams for mapping; and,
(3) satisfactory availability of data for sulfur content,
bed thickness, and washability, with the prospects of
additional washability data from private sources.
Accordingly, this two county area in Maryland was selected for
the detailed study of Phase III.
Instead of examining a third region before selecting the
study area, the large-scale four state coal region was selected
for preliminary investigation in Phases I and II. This was done
because: (1) our initial investigations did not reveal a third area
with the possibility of providing adequate data at this time,
although washability data are presently being gathered and made
available for numerous areas, and (2) we wanted to begin testing
the techniques for a study area without waiting for the availability
of detailed geological information for base maps and coal data
for mapping. The four state Appalachian Coal Region afforded
us such an opportunity.
We investigated the form and extent of the data as well
as its general availability. Since it was apparent at the outset
that nearly all of the relevant data were available through the
Bureau of Mines, our initial review included discussions with
people in the Bureau in College Park, Maryland, and Denver,
Colorado, involved directly with the coal data banks. We found
that for whatever region we selected we could obtain data on
sulfur content for raw or clean (washed) samples by county,
town, or mine for each coal seam. We could also obtain
-------
data for the number of analyses taken at a mine, or for a town
or county as a whole. This form determined the data banks
for sulfur content used in the project:
Phase I: low, mode, high values by counties;
Phase II: low, mode, high values by towns;
Phase III; low, mode, high values by mines.
The Bureau of Mines data banks also contained information
on reserves and bed thickness by counties and in some cases
by mine; however, these data were incomplete, particularly
on the sub-county level. The data banks did not contain any
information on location according to a coordinate system nor
on washability data, although both are presently being incor-
porated into the data banks. It was, therefore, apparent that
locations would have to be taken directly from maps by eye and
that we would be dependent upon published and unpublished reports
for the washability data.
We had originally hoped that all data could be provided
in a computer-us eable form. Therefore, it was necessary to
(1) directly acquire and assemble as much information as
possible on data locations and washability, and (2) translate
this information into a computer-useable form. In some cases
this was extremely time consuming; in other cases, sufficient
data could never be acquired. In all cases the relevant information
is being incorporated into the data banks in computer-useable form.
-------
3.2. APPALACHIAN COAL REGION
The Phase I and II study area was defined by the boundaries
of counties in the Bureau of Mines data banks which (1) had
data on the sulfur content of coal and (2) were in the area
referred to as the Appalachian Region. This study area includes
parts of Ohio, Pennsylvania, Maryland, and West Virginia and
is shown in Figure 1. Two coal seams were chosen for mapping,
based on their extent, number of analyses, and variation in
sulfur content: the Pittsburgh and the Middle Kittanning.
A base map was developed showing the seam outcrop lines
and legends such as state and county boundaries. The Report
of Investigation Series published by the Bureau of Mines shows
seam outcrop lines and mined out areas for each coal bed by
county. We had anticipated using this series to develop the
base maps for the Pittsburgh and Middle Kittanning beds. Un-
fortunately, the reports had been published for only some of
the counties (13 out of 33 for the Pittsburgh Bed) and, therefore,
this approach could not be used to obtain a complete base map.
A map of the Pittsburgh Bed outcrop as of 1938 was
available from the Bureau of Mines. This map was used in
conjunction with the Report of Investigations series to develop
the base map for the Pittsburgh Bed; this same base map was
used for both Phase I (mapping by counties) and Phase II (mapping
by towns).
Similar source maps are not available for the Middle
Kittanning bed or for most other coal seams; this 1938 Pittsburgh
map had been produced for a specific application. To map a
large scale area such as the four state region, one presently
has recourse to two approaches:
-------
(1) Determining the outcrop lines from a series
of smaller scale U. S. or state geological survey maps;
this is very time consuming, a good knowledge of the
geology is necessary, and the maps are not always
available for every portion of the region. This method
was not used.
(2) Defining the study area by the location of data
points rather than by the outcrop lines; for some of
the mapping in Phase II, the study area was defined
by those locations within five miles (or ten miles) of
each data point or town. Unfortunately, it is not
possible with such an approach to know the true extent
of a coal seam, and, therefore, of a low sulfur coal
deposit. The base map preparation is easier and
large scale mapping is possible Without the tedious
coding of an outline from the geological maps. In
Phase II both the Pittsburgh and Middle Kittanning
Beds were mapped by this method.
The Phase I data was obtained for the Pittsburgh Bed
from The Bureau of Mines in a form compatible with our
computer programs. This included a computer card deck of
the values for the number of analyses, low, mode, and high
values of sulfur content for each county for both raw and
washed samples. It did not include the data point locations
which were determined by a manual procedure described
in Section III (Appendices).
For both the Pittsburgh and the Middle Kittanning
Beds data similar to that of Phase I--the number of analyses,
low, mode, and, high values of sulfur content for raw and
washed samples--were available by mine for Phase II. The
-------
data could not be mapped mine by mine as originally received
for one reason: no data was readily available as to the location
of the mines in terms of a geographic coordinate system.
For the actual mapping, it was necessary to determine the
location of each town on a map by eye. The data point
locations are presently being made a part of the computer
data banks.
These data can be aggregated from mines to towns
by a Bureau of Mines computer program. Due to technical
difficulties that would have resulted in delays, the data were
aggregated by hand; for other study areas, however, it is
important to note that the data can be automatically aggregated
using a relatively simple computer program.
Portions of the four state region were selected for
mapping at a more detailed scale. These included the Phase
III study area of Allegany and Garrett Counties, and the
northern and northeastern portions of the Appalachian Coal
Region, abundant in both data point locations and low sulfur
coal. The accuracy obtained by using town data versus county
data was also compared. No additional data were necessary
for these sub-area studies; they were conducted to demonstrate
the flexibility of the mapping program to select and map
areas of particular interest at any scale.
-------
3. 3. ALLEGANY AND GARRETT COUNTIES
Allegany and Garrett Counties were selected as the Phase
III study area, based on the availability of data. Stage One of
this work was similar to Phases I and II: mapping of sulfur
content for low, mode, and high values by mines. The Upper
Freeport and Upper Bakerstown Coal beds were chosen for
mapping based on the availability of data and source maps
showing the seam outcrop lines.
The data values were provided in the same form as the
Phase I and II data. The data points had to be located on the
source maps and extracted from field reports; this was done in
conjunction with Bureau of Mines personnel and would have been
very time consuming if more data points had been involved.
We had expected the washability data to be limited, but the
sulfur data was also sparse. This is due to the number of
seams in the area: when the data is assigned to the appropriate
seams, individual seams often do not have enough data points
for useful mapping.
Information published by the State of Maryland (1967 annual
report) was used to determine the bed thickness for these two
coal beds; the Bureau of Mines provided the location for each
of the points where thickness had been measured. The availability
of similar data and its extent for other areas would significantly
affect the use of the mapping programs to determine the quantity
of coal (reserves) for a region.
The base maps showing outcrop lines and areas over 28"
thick were developed using two sources (Boyd; 1953 Garrett
County map; 1965 Allegany County map) as described in Section
III (Appendices). Together, these sources contained sufficient
-------
information to adequately determine outlines, but similar reports
may be available for only a few regions.
Stage Two of this work was concerned with the washability
data and the quantities of coal that might be made available
under various mining and economic assumptions. Our greatest
difficulties with the data were encountered here, as had been
expected. Data were incomplete with regard to float and sink
analysis for coals of various top sizes. Of some 15 possible
seams with data only the Upper Freeport Bed had sufficient
washability data and adequate sulfur content, thickness, and
base map information for mapping. The results are, therefore,
purely demonstrable and cannot be^taken literally in terms of
quantities of coal available under differing assumptions.
The data from government sources was to have been
supplemented with information from private sources, but this
was not possible within the time-frame of the project. Private
sources will need to provide much of the necessary data in
the futur e.
The washability data was provided in a form such that a
minimal amount of data handling was necessary prior to mapping.
This data included the associated x and y coordinates which
greatly facilitated the coding procedure. The location of the data
points for the washability data (and the Stage One data points)
determined the study area for Stage Two. This study area is the
central portion of Allegany and Garrett Counties as shown in
Figure G-l.
Data handling prior to mapping generated values for both
total sulfur content and pyritic sulfur content, given different
-------
yields from washing with several screen sizes. The sulfur
content that could be attained through washing is the total
sulfur content, less all or part of the pyritic sulfur, since
the pyritic sulfur is easier to remove than organic. We do not know
what percent of the pyritic sulfur can be removed; therefore,
both the total sulfur content and the total less all pyritic sulfur
were mapped to represent the two extreme assumptions. The
quantities of coal that could be made available at different yields
were determined accordingly.
It is difficult to draw any statistically sound conclusions or
inferences on account of the insufficient washability data. However-
this lack of data did not prevent the demonstration of the relevant
computer mapping techniques. Likewise, the particular study areas
may not be representative of the country as a whole but they
can adequately demonstrate the scales of analysis useful to
APCO.
-------
4. GRAPHIC TECHNIQUES
With the exception of class levels, or value range
intervals, the map symbolism is the only and, therefore the
most important graphic way to portray the meaning of the
data. The value of a map may be easily lost or destroyed
by a poor choice of symbolism.
In SYMAP this choice of symbolism is usually made
by the program; there is a prestored set of symbolism for
up to 12 levels. For this project it was felt that this set of
symbolism was inadequate because (1) there was not enough
differentiation between some of the levels, and (2) the darkness
of adjacent levels seemed to be reversed in some cases.
Therefore, a unique set of symbolism was devised.
It is fairly easy to create distinctive symbolism for
up to 5 levels. Numerous studies into human tone perception
conclude that the eye can discern no more than 7 different
shades of gray. Above 7 shades differentiation may be
difficult. Since SYMAP's symbolism is derived from over-
printed characters, each tone is unique in make-up; therefore,
if it is difficult to discern the relative darkness of two
adjacent tones, the makeup of the symbolism can show which
was intended to be the darker. This is not possible in a
hand-rendered map which uses standard dot screen patterns.
The final choice of symbolism was arrived at by
considering what had been used before and by experimenting
with a number of test maps.
-------
4. 1. GENERAL SYMBOLISM CONSIDERATIONS
Given any symbol, its appearance would seem to
be a function of several factors: the amount or percent of
the ceil area darkened or blackened, the amount of white
space between characters (external white space), the amount
of white space within the Characters (internal white space),
the intensity or blackness of a character, and the texture
and grain of the symbol.
In most cases, the symbols used on a map should
be non-directional; highly directional symbols like I and/should
be avoided except when emphasis is desired. There are also
some symbols which are ambiguous; they may be intensely
black, yet have a high percentage of external white space.
As a consequence they are both dark and light and do not
occupy any position on the grey scale. Examples are symbols
like: { = +01. The circle or "U" is also troublesome since
the external white space is minimal, but there is a great deal
of internal white space. For this reason, it may be wise to
use "0" only when overprinted with some symbol which reduces
the internal white space.
The approach to symbolism which seems most
successful is to begin with a basic symbol and vary it. As
a consequence there is a logical progression created from
symbol to symbol. The symbol not only becomes darker,
but more complex as well. For example,
+ X X 8 B , or,
o e a 8 B.
-------
There is often a temptation to dismiss symbols for
adjacent levels which are not readily distinguishable, but it
should be remembered that in a 10 level map, level 7 is
similar to level 6, and some similarity between the symbols
is desirable. For most maps the levels represent quantitative
variations of a single variable--and this should be apparent
visually through the symbolism. A different approach to
symbolism is required when, for example, level 1 represents
dairy farming and level 2 represents vegetable raising. In
this case, the symbols should be descriptive rather than
evaluative. This is often the case with conformant or
proximal maps.
It would seem that one of the most important consider-
ations should be that the units on the visual scale correspond
to the quantitative units on the data distribution. For example,
if the value range is divided into five equal levels, the symbols
should be chosen so that the grey scale is divided by 5 equal
intervals. Carrying this to its logical conclusion means that
when the data is skewed and it is necessary to manipulate the
levels so that all the observations will not fall in one level, the
symbolism should be adjusted appropriately. Eliminating the
skewed distribution by assigning an equal number of observations
to each level is often unwise, since the skew of the distribution
is a fact which should be represented. The unevenness of the
distribution may be the most important thing about the map.
There are limits to how precisely this can be done,
but there is no doubt that it can and should be done in some
cases, where use of equal interval symbolism would be grossly
misleading. There are enough easily distinguishable symbols
and textures to maintain distinctions between levels and still skew
the symbolism.
-------
It may also be desirable to manipulate the symbolism
in other ways. For instance, suppose one is concerned with
distinguishing between 3 of 10 levels. It might be desirable
to use highly distinctive symbolism for these three areas in
order to emphasize their locations. The manipulation o± the
symbolism must be left to the judgement of the individual,
but a fair amount of effort should be expended to relate the
data to the symbolism. Because of the limited number of
levels, it is often necessary to assign a wide value range to
one value interval which may reduce the precision of the map.
To avoid this, great care should be exercised when dealing
with variables which can be controlled. The map will not
be effective if it does not represent visually what the histo-
gram bar and the data say it represents .
4.2. SYMBOLISM USED IN THE PROJECT
Many varieties of symbolism are available to the
user; the principal objective in selecting symbolism is the
readability and graphic tone. Experiments and tests were
made on the mapping for all phases and a consistent symbolism
developed for use throughout the whole project.
As discussed in the Appendices (Section III) under use of
Statistical Routines, an equal interval value range was used
for most of the maps produced. it had nine graphic levels as
well as a low and a high value symbolism for data exceeding
the specified minimum and maximum values. Therefore, the
task became one of determining the appropriate level break-
down for each of the types of computer maps and applying the
standard symbolism to this breakdown.
-------
The range and distribution oi values from the
experimental runs for Phases I and II were reviewed and the
decision made to map the sulfur content to the nearest 1/2%,
with a maximum of 5% and a minimum of U. 5%. Symbolism
was also assigned to the few values that might exceed the
maximum and likewise to those areas that were below the
minimum. The same level breakdown was used for the sulfur
content data in Phase III. The symbolism used as coded in
the F-MAP package is shown in Table 111-12 of the Appendices
(Section III).
For Phase II 'critical level1 breakdowns were also
used to emphasize the isolines of 0. 5% and 1% sulfur content.
Special symbolism was used for Base Maps in ail three phases
and for the surface comparison in Phase II. Table 1-2
summarizes the use of symbolism for the different map series
and Table 1-3 shows the various symbolism used.
In Phase III maps were produced for bed thickness
and quantity of coal in addition to sulfur content. The level
breakdown for bed thickness was 4" per level. After reviewing
the minimum, maximum, and distribution o± thickness values
it was apparent that areas greater than or less than 28" thick
were of particular interest. Therefore, level adjustments
were made that reflected this 28" value; the minimum value
set was 16" and the maximum 52".
Preliminary review of the values for quantity of coal
indicated that a value range of 500 tons per acre seemed
appropriate to use as the level breakdown. The minimum
was set at 5000 tons per acre and the maximum at 9500
tons per acre, with special symbolism assigned for values
exceeding the maximum or minimum values.
-------
MAP SERIES Standard Critical Base
Levels Levels Maps
Appalachian Coal Region
A. Phase I - Pittsburgh X X
B. Phase II - Pittsburgh X XX
C. Phase II - Middle Kittanning X X
D. Phase II - Detail X
Allegany and Garrett Counties
E. Stage One - Reserves X X
F. Stage One - Thickness/Quantity X
G. Stage Two - Washability X X
H. Stage Two - Quantity X
NOTE: The Special Symbolism was used for the surface comparison of
maps D-l and D-2.
Table 1-2
Leveld Usied for Each Map Series
-------
PERCENTAGE CF TOTAL ABSOLUTE VALUE RANGE APPLYING TO EACH LeVEL
11.11 11.11 11.11 11.11 11.11 11.11
11.11
11.11
11.11
FREQUENCY DISTRIBUTION OF DATA POINT VALUES IN EACH LEVEL
LEVEL 12345
s as Es:es3BS3S33 a as 3s sssessssss 3333 333 ssescsa S3 33 333 s 333 33=3333:33 333:333333333 333 s ass-sse*rs«33S3sss
SYMBOLS
A.
....
!.... --- 2 ----
......... — ————
......... ———-•—
S333S3SSS333SS33:
SS3S33B3 +++++++++ ********* ********* ********* MMMMI MMMAM
33333333 +++++++++ ********* ********* ********* •MMMM MMMAM
****5**** ****6**** ****?**** eiitenii MM9B8M
•& "&. •& &•& & &: + & fc****-g*^wwwwv •MM MMMMMM4f • AflAABBfldfl AJfeJ
^.^^^^^^^ ^ 9C7f7%n7in7w%n Wftw«w**flfWI ••••(••••IV OV1
********* ********* ********* eiiieeeai ~
Standard Symbolism: nine levels, plus high and low levels.
(the low is the same as level 1; the high is shown below)
PERCENTAGE OF TOTAL ABSOLUTE VALUE RANGE APPLYING TO EACH LEVEL
100.00
FREQUENCY DISTRIBUTION OF DATA POINT VALUES IN EACH LEVEL
LEVEL 1 H
3S38333S33S3SB*SSS*«S
XXXXXXXXX IMMMM
XXXXXXXXX MftMMM
SYMBOLS XXXX1XXXX •IIIHIIII
XXXXXXXXX •••••Mil
XXXXXXXXX MBMMH
B.
Critical Levels Symbolism: one level, plus high and low levels.
(low shows less than U. 5% sulfur coal; high shows 1% or higher)
Table 1-3
-------
PERCENTAGE OF TOTAL ABSOLUTE VALUfc PANGF APPLYING TO EACH LEVcL
ICC.CO
FKfcCUENCY DISTRIBUTION OF DATA POINT VALUES IK iEACH LfcVEL
LgViEL i
SYMBOLS
!S = = = =
C. Base Map Symbolism: one level.
(two of the three symbols used on base maps; the third substitutes
'pluses' for the 'dots'. )
PERCENTAGE CF TCTAL A8SCLITE VALUE RAKGE APPLYING TO EACH LEVEL
25-OC 2C.CC 1C. 00 20.00 25.00
FREQUENCY CISTfilBUTICN CF CATA PCINT VALUES I* EACH LEVEL
LEVEL L 1 2 3 4 5 H
Iflllllll
... .....
5YP3CL5 IflBHlIII KXNX3KMMX -»i-»-»2-»-»-»-» ....1.... +++-»2++-»* *#**3#*#* IfiflHECM
•fiffitaaa
• Illllill
S = ZS3SS3SS33SSSS.CSBSaS 33=
D. Special Symbolism used for Surface Comparison: five levels, phis low and high levels.
(the greater the difference in the two surfaces, the darker the symbolism)
Table 1-3
-------
Significantly different level breakdowns were used
for Stage Two--washability data--of Phase III for, sulfur
content, quantity, and surface comparison. In all cases,
however, the standarcj nine level symbolism was used.
All of the level breakdowns used for the project are sum-
marized below:
Map
Series
B
A
B
C
D
D
E
F
F
G
G
G
H
H
No. of
Levels
1
9
5
9
9
9
9
9
9
9
9
Value
Range
0.5%
0.5%
varies
0.5%
4"
500 tons/
acre
0. 10%
0. 05%
(eq-ual)
(equal)
(equal)
Minimum
0.5%
0.5%
-1.0%
0.5%
16"
5000
0. 60%
0. 45%
0. 00%
900
(966.43)
Maximum
1.0%
5.0%
+ 1.0%
5. 0%
52"
9500
1.50%
0. 90%
0. 20%
3300
(1340. 80)
Data
Sulfur content
Sulfur content
Surface comparison
Sulfur content
Thickness
Quantity of reserves
Total sulfur content
Total less pyritic
Comparison of mesh
size
Quantity of washed
coal
Difference in quantity
The sulfur content maps in Stage Two used different
level breakdowns for total sulfur content, total less pyritic
sulfur, and the comparison of sulfur content for different screen
sizes; all of these used equal interval value ranges. The quantity
of washed coal was mapped with a value range of approximately
266. 67 tons per acre in each level. For the comparison of
quantity at 60% and 100% yield the actual extremes of the
data were used as the minimum and maximum.
-------
4. 3. REPRODUCTION CONSIDERATIONS
All but two of the final maps produced in Phases
I and II were 26" wide, and these two plus all of the maps
in Phase III were 13" wide. The maps were photographically
reduced for publication in this report. The 26" maps were
reduced to 25% of their original size and printed one to a
page; the 13" maps were reduced to 30% of their original
size and printed two to a page. Previous studies had
demonstrated that such size reductions gave visually good
results.
However, photographic reduction of the maps has
certain difficulties . In particular the legends may become
illegible and the textual and statistical material printed below
the map may lose much of its usefulness. Moreover, the
tone dis cernability of the symbolism is somewhat reduced in
almost direct proportion to the amount the maps were photo-
graphically reduced. But, it is still felt that a relatively
good quality can be attributed to the symbolism.
Some of the problems can be overcome by photographing
the legends and explanatory information at another scale. In fact,
this was done with the titles shown at the bottom of each page;
these were separately photographed and stripped into the
negatives before printing.
The best solution is to experiment at the beginning
of a project to determine the optimum combination of study area
size, reduction scale, and legend and explanatory information
detail to use in the final presentation or publication of the
maps.
-------
II. STUDY AREA FINDINGS
1. Discussion of Study Areas and Data Used
2. Discussion of Graphics
3. Conclusions and Recommendations
-------
1. DISCUSSION OF STUDY AREAS AND DATA USED
The Appendices (Section III) describe step by step the
mapping procedures for both the four state Appalachian Coal
Region and Allegany and Garrett Counties in Maryland. This
section summarizes the information on study areas, data, and
procedures, provides an introduction to the discussion of
the graphics, and sets forth conclusions and recommendations.
The studies were made at two scales: the Pittsburgh coal
was mapped for a large four-state region and, the Freeport
and Bakerstown coal were more intensively mapped for a small
region in Maryland. Both the regions studied lie within the
Appalachian Mountains where Paleozoic beds have been folded
into regular parallel anticlines and synclines, tending northeast-
southwest. The width of the folds is from four to five miles.
The coal seams are concentrated in rocks of Pennsylvanian
age; the Pennsylvanian in Garrett County is about 1,700 feet
thick (Amsden, 1953). The coals are concentrated in the
synclines in Garrett and Allegany Counties since the Pennsylvanian
rocks overlying the anticlines have been stripped off by erosion.
-------
1.1. APPALACHIAN COAL REGION - PHASE I
A four-state area in the heart of the Appalachian coal
region was delineated on the basis of available county data.
This area is worthy of study because of:
1. the size and extent of the coal seams,
2. the existence of maps showing the extent of some of
the seams,
3. the amount of data already collected and in computer -
useable form,
4. and the high potential this coal region might have in
solving some of the problems of air pollution.
This area is abundant in low sulfur bituminous coal. The
total area of the region comprises 293,439 square miles, of which
33,369 square miles is laden with Pittsburgh coal (11.4% of
the total study area). Almost 82% of the coal (27,279 square
miles) lies in a single pie-wedge shaped seam which is centered
near the S.W. corner of Pennsylvania. The remainder of the
coal is located in 65 outlying seams over the four state area.
Only maps for the Pittsburgh seam were created, since
insufficient information was available to produce a study area
outline for the Middle Kittanning seam outcrop. The outcrop
used for the Pittsburgh seam was the 1938 strippable reserves,
and not the deep reserves. The Phase I base map, including
the seam outcrop, was made with a great deal of detail so that
particular sections could be 'blown up' for closer examination
in Phase II. It was felt that this feature of the technique would
have special usefulness to APGO and the Bureau of Mines. Each
seam mapped requires a detailed outline of 500-1000 points to
achieve this accuracy at the enlarged scale.
-------
The data points representing counties had to be located
on the source maps by eye. The rationale behind the placement
of the county data points was as follows: (1) If a map from the
Report of Investigation series was available, the data point was
located in the center of a coal deposit 28" or greater; (2) If
no map was found, the data point was located in the "center
of gravity" of the seam for that county, based on the 1938
Pittsburgh seam map.
The data bank of values was provided in a computer-useable
form by the Bureau of Mines. Each card of the data bank
represented a county's information; on this card were the following
variables: state, county, and bed codes, the mode, low, and high
values and the number of analyses for the sulfur content of raw
samples, and, finally, on 19 of the 33 Pittsburgh Bed cards,
the mode, low, and high values and the number of analyses for
the sulfur content of cleaned (or washed) samples.
The original data bank was manipulated so as to (1) separate
the Pittsburgh and Middle Kittanning data, (2) create separate
decks for the raw and washed values, and (3) store the variables
in the order useful for analysis and mapping, as shown in Tables
III-2 and III-4 of Section III (Appendices). This order was:
the number of analyses, low, mode, and high values, bed code,
state and county code, and state and county name. This exemplifies
the type of straight-forward data manipulation that can be performed -
either by hand or computer, depending upon the amount of data--
to set up the data banks in a convenient form for mapping.
-------
1.2. APPALACHIAN COAL REGION - PHASE II
Phase II dealt with the same four-state coal region, but
on a more detailed level of analysis. Whereas Phase I
measured the variation in sulfur levels from one county to
another. Phase II dealt primarily with variations between
towns within these counties.
Much of the base map and data handling work for Phase II
was done in conjunction with Phase I; the main additional base
map work of Phase II involved determining the x and y coordinates
for the towns for both the Pittsburgh and Middle Kittanning Beds,
as described in Section III (Appendices). The Bureau of Mines
data by mines were aggregated to towns in Phase II because: (1)
many of the mine locations were not yet referenced to the grid co-
ordinates, and (2) such small-scale mapping would create a great
number of superimposed data points that would cancel out the
accuracy obtained at the mine level. At the town level there
were 8 superimposed points at 4 locations for the Pittsburgh Bed
and 2 at 1 location for the Middle Kittanning Bed for the raw
samples; the effective number of data points (considering
superimposition) is shown in Table H-l.
It was also necessary to aggregate the data from mines
to towns according to the procedures discussed in section III
(Appendices). Three data banks of values and associated data
point locations were created? (1) Pittsburgh number of analyses,
low, mode, and high sulfur content for raw samples, (2) Pittsburgh
data for washed samples, and (3) Middle Kittanning data for raw
samples; with only 15 data points the washed Middle Kittanning
data was not mapped.
-------
APPALACHIAN COAL REGION
Phase I - Counties
Pittsburgh Bed
Raw Sulfur Content
(points mapped)(effective
locations)
33
33
Washed Sulfur Content
(points mapped)(effective
locations)
19
19
Phase II - Towns
Pittsburgh Bed
Middle Kittanning Bed
203
90
199
89
56
56
ALLEGANY AND GARRETT COUNTIES
Phase III - Mines
Upper Freeport Bed
Upper Bakerstown Bed
Raw Sulfur Content
(points)(effective)
20 19
7 6
Bed Thickness
(points)(effective!"
7 7
16 16
Washability
(points) (effective)
5 5
Table II-1
-------
The Pittsburgh data was mapped using the same equal
value range intervals as Phase I; maps were also produced
using the critical isolines and symbolism discussed in the
section on Graphic Techniques. A set of maps for the Pittsburgh
raw data was produced using five mile and ten mile search radii
(or circles of symbolism) about each data point to define the
study area, rather than the outcrop lines. This same technique
was then used with the Middle Kittanning data since no outcrop
line was readily available for this bed. Although no reliable
information as to extent of the coal seam can be obtained from
such maps, the accuracy around each data point is relatively
equal.
The final set of maps run in Phase II utilized previously
mentioned data. This included surface comparisons between
the mode values of sulfur content mapped by towns versus that
mapped by counties. The mode values for raw samples were
mapped at several scales for different portions of the study area,
including the Allegany and Garrett Counties region of Phase in.
-------
1.3. ALLEGANY AND GARRETT COUNTIES - PHASE HI, STAGE ONE
The general purpose of Phase III was to develop and test
techniques for determining the quantity of coal in the ground
(depending upon sulfur content) and the quantity of coal that
could be made available through washing, assuming various mining
practices and market economics. The Stage One work was
concerned only with calculating the reserves; Stage Two involved
the data on washability.
The study area for Stage One was approximately 1000
square miles and included all of Garrett County and the Western
one-fourth of Allegany County. Preliminary investigation of
available data pertaining to the three categories of sulfur content,
thickness, and washability for each of the coal seams that exists
within the two county study region led us to the conclusion that
the Upper Freeport and Upper Bakerstown seams were the most
feasible for our studies. The other seams in the study area
presented difficulties due to either an insufficient number of
data points, super imposition* of data points, or a lack of data
in one or more of the three above catagories. Table II-l shows
the data banks actually used, while Table II-2 summarizes the
available information for 15 seams in the study region.
In the case of the sulfur content data mapped in Stage One,
superimposition of data points reduced the effective number of
locations for 4 beds, particularly for the Pittsburgh and Sewickley
Beds. Superimposition also affected the number of data points
for the washability information of Stage Two, but it did not
affect the seam thickness data for the Upper Bakerstown and
Upper Freeport Beds.
-------
I
U)
Bed Name
Waynesburg
Bewick ley (Tyson)
Pittsburgh
Little Pittsburgh
Barton (Elk Lick)
UppeerBakerstown
Lower Bakerstown
Upper Freeport
Lower Freeport
Upper Kit tanning
Lower Kit tanning
Clarion
Harlem
Franklin
Redstone
Code
023
029
036
038
055
062
063
071
074
076
084
087
Counties
Alle- Garrett
Sany
X
X
x x
X
X
X X
X
X
X X
X
X
X
Base Map
Hary- Boyd
land
X
X
X
X
X X
X
X
} x x
J x
1 X
)x x
X
Sulfur
Origin
nal
2
15
16
2
4
7
2
18
2
1
3
2
Content
Effective C
2
9
10
2
4
6
2
17
2
1
3
2
Wash a
)riginal
2
2
1
1
2
-
2
6
1
1
3
—
1
6
2
bility
Effective
2
2
1
1
2
-
1
5
1
1
2
—
1
3
2
NOTE: Bed Thickness Data consisted of 16 data points (original and effective)£€or
the Upper Bakerstown Bed and 7 for the Upper Freeport.
Table II-2
-------
The base maps for Stage One and Two were the same
with the exception of data point locations. Geological maps for
Allegany and Garrett Counties were used, as well as a less
detailed set of maps from a consultant's report (Boyd, 1964). The
map for Allegany county showed few outcrops. The geological
map for Garrett County was used as a base since it had out-
crop lines for the Brush Creek, Barton, Harlem and Freeport
beds. The maps were developed with a great deal of detail so as to
extract small areas for closer analysis. Unfortunately, the data
contained almost no points for the Brush Creek, Barton, or
Harlem beds for either the sulfur or washability data. Consequently,
the Upper Freeport base map was the only one which used this
source map.
The maps from the Boyd report had to be enlarged to
determine the seam outcrop for the Allegany County portion
of the study area, and for all of the Bakerstown base map. The
Freeport, Bakerstown, Sewickly (Tyson), and Pittsburgh seams
were the only ones with sufficient data points for mapping of the
sulfur level data. Sufficient data points for washability and
thickness were available for only the Freeport bed. The Pittsburgh
and Sewickley base maps were not developed because it was felt
that no additional information would be obtained from mapping
these two beds; all of the procedures and techniques could be
tested with the Upper Freeport and Upper Bakerstown coal seams.
Base maps for areas over 28" thick were also developed from
the Boyd reports.
The sulfur content data for Phase III were made available in
two parts':' (1) data bank of values, and (Z) data point locations.
The values, as usual, were available in a computer useable
form, similar to Phases I and II. Due to the timing, the data
-------
point locations were derived from field notations in Bureau
of Mines' reports-- "One mile west of-town X"--or by assigning
the mine location to the nearest town when more precise locations
were unavailable. Twelve beds had data for raw samples; 8 of
these had 4 or less data point locations each. The greatest
number of locations for washed samples was 6 for the Upper
Freeport Bed, and all othe rs had 1 to 2 locations; consequently,
the washed samples were not used.
The bed thickness data were also made available in two
parts: (1) data bank of values from the state of Maryland 1967
annual report, and (2) the data point locations from the Bureau
of Mines. These data were obtained specifically for the Upper
Bakerstown and Upper Freeport beds after it had been determined
that only these two would be mapped; consequently, the relative
availability of data for other .seams is not known.
In addition to sulfur content and bed thickness, the
quantity of less than 1% sulfur coal was determined and mapped
for the Upper Bakerstown and Upper Freeport Coal Beds.
Quantity was determined as follows: if the sulfur content was less than
1% for any character location on the map, then the quantity would
be calculated and mapped as a function of the bed thickness at that
location. The algorithm for determining quantity from thickness
was obtained from the Bureau of Mines and is discussed in
Section III (Appendices); it assumes that there are 1676 tons of
coal per acre-foot.
Composite sulfur content and quantity maps were also made
for the Upper Freeport and Upper Bakerstown Beds.
-------
1.4. ALLEGANY AND GARRETT COUNTIES - PHASE III, STAGE TWO
The central portion of the two-county area was selected
for further mapping and analysis because of the availability of
washability data for the Upper Freeport Bed. The area is
approximately 500 square miles. The maps for both Stage One
and Stage Two were made at a smaller size then those in Phase
I and Phase II. After reviewing several test maps showing the
Upper Freeport total outcrop at the Phase I size, it was determined
that the information could better be provided at a much reduced
map size. This decision was based on the following:
(1) The number of data points was relatively small for
sulfur content, thickness, and washability information; it
became quite apparent that, because of the small number of
data points, the larger maps did not show any more infor-
mation than did maps made at the reduced scale.
(2) Because of the large amount of testing and experimental
work that had to be done, it is easier to quickly evaluate the
results from the smaller maps; this avoids the extra time
and effort to splice together the several panels or sections
that would result if the larger maps were made.
(3) The larger maps in general incur higher computing and
printing costs, nearly proportional to the area of the map.
The final Stage One maps were printed at a scale of
approximately one inch = three miles, and the Stage Two maps
at a scale of approximately one inch = two miles.
All of the data used in Stage Two were contained in one
small data bank for the Upper Freeport Bed. This included
the x and y locations for the 5 washability sampling points used
-------
(1 of the original 6 was in the top, rather than the bottom,
bench of the bed and was therefore removed). The washability
data were composed of values for total sulfur content and pyritic
sulfur content for different yields and top sizes (3/8", 1 1/2", etc)
The first task was to determine the yields to be mapped,
in light of current mining practices and incremental price
increases at the market. The criteria were translated into
50%, 60%, 70%, and 80% yields as being representative of the
problem, with the 60% yield being the one of primary interest.
The second task was to determine the appropriate sulfur
content for these yields, given the sampled and derived relation-
ships from the Bureau of Mines. Figure 2 shows the sample
plot for the fifth mine listed in Table III-18, (Appendices)
for the 3/8" mesh data shown at the top of the table. The values
for 50%, 60%, and 70% yield had been statistically derived by
the Bureau of Mines, but the values for 80% were interpolated
from the plot. Similar plots were done for the other 4 mines
for the 3/8" top size, and for all 5 mines for the 1 1/2" top
size.
From this set of washability data and the sulfur content and
thickness data of Stage One, four types of maps were run in Stage
Two:
(1) sulfur content of washed data, by yield and top size;
(2) comparison of sulfur content from 3/8" versus 1 1/2"
top size;
/ 0\
quantity of less than 1% sulfur coal assuming washing,
by yield and top size; and,
-------
(O
Figure 2
Sample Plot of Sulfur Content Versus Yield
-------
(4) comparison of the quantity of mineable reserves and
washed marketable coal at 60% yield.
In most cases, both total sulfur and total less pyritic sulfur
were mapped, for the total outcrop and for the areas over 28"
thick for the Upper Freeport Bed.
-------
2. DISCUSSION OF GRAPHICS
The following sections contain a detailed discussion
of all the graphics produced; the maps themselves are at
the end of each section and Tables II-3 and II-4 summarize
all of the maps produced in each phase or stage of the
project.
We strongly caution the reader about the lack
of reliable data and the commensurate unreliability of many
of the maps. Perhaps the best safeguard against making
any false inferences is to look at a map in one hand and at
the base map which shows the data points and the number
of analyses in the other hand. In this way, one is made
aware of the number of analyses attributed to a data point
and the validity one can infer about any given area of
the map.
It is felt, and has been stressed throughout, that
more meaningful results could have been attained if there
had been a more equal distribution of data points and a more
equal number of analyses taken at these points. All of us
were aware that this might be a problem at the outset of
the project and our fears have been substantiated. We
hope that pointing this out is the kind of constructive
criticism that may prove most helpful in overcoming de-
ficiencies in data acquisition so that more meaningful
results may be attained in future work.
Of significant importance, however, is the potential
applicability of each of the types of maps produced and
their usefulness as adequate data are made available.
-------
2.1. SUMMARY OF PHASES I AND II
Examination of the graphics produced in Phases I
and II has shown that there is a definite need for: (1)
a more equal number of analyses at each sampling
location and (2) a more evenly distributed pattern of
sampling locations. The latter problem is especially
acute in Phase II while the former problem pertains to
both phases.
In Phase I, values were assigned to data points
which occur in the center of gravity of seams of 28"
thickness within a county. These points represent an
area, and, therefore, the data points themselves have no
real locational meaning. As a result, any contour maps
made from such data point locations can at best give only
an impression of the sulfur concentrations. They cannot
be interpreted literally, even at the data points themselves.
This is in contrast to the maps made in Phase III where
the data points are positioned as accurately as possible
at the actual sampling locations on the map. Contour maps
made in Phase III are at least accurate at the data points
themselves.
The maps with county data points (Phase I) had a
fairly even spread of data, while the town data point maps
(Phase II) had a very uneven spread of data. The county
maps could have used quite a few more data points; the
town maps could have used many more data points,
especially in the center of the main seam. Therefore,
of the two sets of maps, Phase I is of reasonable accuracy
throughout, while Phase II is of fairly good accuracy along
the northern and eastern edges of the Pittsburgh Bed out
crop line but of very poor accuracy in the center.
-------
The use of a limiting search radius around the
data points for both the Pittsburgh and Middle Kittanning
Beds increased the accuracy of the areas mapped; however,
comprehension of the extent of the coal seams for varying
sulfur levels was lost. This can be overcome by using
outlines and legends in combination--one to show the out-
crop line, or extent, and the other to show the areas of
nearly equal accuracy, as defined by the search radius.
-------
FOUR STATE APPALACHIAN COAL REGION
^,— Bed Outcrop
Washed
County [Levels] [Critical]
«
£
§
CO
EH
EH
H
04
O
1 S5
QN LJ
00 g;
' pi] JZ
1-5 <5
Q EH
Q EH
H H
S «
# analyses
low
mode
high
Towns
f analyses
low
mode
high
Towns
# analyses
low
mode
high
Difference
A-l
A- 2
A- 3
A- 4
B-l
B-2 B-5
B-3 B-6
B-4 B-7
Raw
[Levels] [Critical]
A- 5
A- 6
A- 7
A- 8
B-8
B-9 B-12
B-10 B-13
B-ll B-14
+ S(
5 mi.
[Level
B-15
B-16
B-17
B-18
C-l
C-2
C-3
C-4
ffi
moe
D- 1
§ Detail of Towns (mode)
co Allegany and Garrett
^ Northern
^ Northeastern Blow up
D-2
D-3-
D-4
D-5
-D-3
Search Radius +
5 mi. radius-Raw-10 mi. radius
[Levels] T
CO
D-6
D-7
B-19
B*20
B-21
C-5
C-6
C-7
CO
Table II-3
-------
2. 2. MAP SERIES A
Map series A contains all of the maps produced in Phase I.
Figures A-l and A-5 show the base maps for the Pittsburgh Bed,
including the data point locations (washed and raw respectively)
and the number of analyses assigned to each county. The spread
of data points is quite even throughout for both sets of data, but
there are significant disparities in the number of analyses used
for calculating the values at each data point. This is particularly
relevant when examining the maps for the low, mode, and high
values for each county (Figures A-2 thru A-4 for washed and
A-6 thru A-8 for raw samples). All six surfaces show a definite
trend from high sulfur coal in the southwest (up to 5. 0%) to low
sulfur coal (less than 1%) in the northeast and in Allegany and
Garrett Counties in Maryland.
2. 3. MAP SERIES B
Map series B contains the maps produced in Phase II
for the Pittsburgh bed. Figures B-l and B-8 show the base maps,
including the data point locations (washed and raw respectively)
and the number of analyses assigned to each town. The data
points are not evenly spread out for either set of data. There
is a large central area in south-western Pennsylvania that has
few data points; this reduces the accuracy in this area. When the
base maps are run at this scale, some of the data points and
legends showing the number of analyses are lost under the state
boundary legends or under other data points. Eight data points are
actually superimposed and are shown by a "/" rather than a black
square. Figure B-8 contains a box showing the study area for
the Phase III analysis of coal in Allegany and Garrett Counties.
-------
irz;; iz^nmnmz i~
;j$8;jf*!r•«-™- ,——" •••&
—— ——-— — i — f-~-— - 0 C 0 001—
WOR S:S
FUCtNUCI Of TOTAL M}O.VTt VALUE *MM MPIV|M Tp (»U L|VIL
Figure A-l
Pittsburgh Coal Bed
Sulfur Content Mapped by Counties
Number of Analyses for Washed Samples
-------
APPALACHIAN
COAL REGION
SKISS !:!! 1:!!
"ffllp"
KIIUTIOH Of MTi POlin VU.UE1 IN C*tM 16«L
Figure A-2
Pittsburgh Coal Bed
Sulfur Content Mapped by Counties
Low Values for Washed Samples
-------
APPALACHIAN
COAL REGION
MOMC03
,"T5
«. 3
I S
-
— J
I
.,-•£
SUflSS 8:»
!:!8 i:i! !:SS !:SI t:S
::»!"" -—2 ™~™ £?*!*£• IUMH
—— :::":" Tn^nTT :^s?
Figure A-3
Pittsburgh Coal Bed
Sulfur Content Mapped by Counties
Mode for Washed Samples
-------
APPALACHIAN
COAL REGION
, i:ii ,.i«r>
tl.ll 11.11 11.11 11.11
Figure A-4
Pittsburgh Coal Bed
Sulfur Content Mapped by Counties
High Values for Washed Samples
-------
sins! u
Figure A-5
Pittsburgh Coal Bed
Sulfur Content Mapped by Counties
Number of Analyses for Raw Samples
-------
APPALACHIAN
COAL REGION
J D OOOOOOu 0 0 C C 00 C 0000 DO
J-J
»og «»
4
8 {N
; •«*
sFA.
I
A. .!
r »
MSIBS 8:1! i:S* !:!« i:« i:ii !:S* !:i( S:St 9:1!
Figure A-6
Pittsburgh Coal Bed
Sulfur Content Mapped by Counties
Low Values for Raw Samples
-------
APPALACHIAN
COAL REGION
•-. t™ \
'¥ J
Figure A-7
Pittsburgh Coal Bed
Sulfur Content Mapped by Counties
Mode Values for Raw Samples
-------
APPALACHIAN
COAL REGION
at ta
Figure A-8
Pittsburgh Coal Bed
Sulfur Content Mapped by Counties
High Values for Raw Samples
-------
Figures B-2 thru B-4 show the low, mode, and high
values for each town; they have the same slope from low sulfur
in the northeast to high sulfur in the southwest, as in the county
maps but the surfaces are not as smooth, due in part to the greater
number of data points and the uneven spread of points. Figures
B-5 thru B-7 are identical to B-2 thru B-4 except for the value
range intervals selected; the critical isolines of 0.5% and 1%
sulfur content are shown. These maps clearly show at a glance
that there are no areas with 0. 5% or less sulfur coal and very
few areas with 1. 0% or less sulfur coal.
Figures B-9 thru B-ll show the low, mode, and high values
for the raw samples and Figures B-12 thru B-14 the same surfaces
for the critical value range intervals. We would assume that
fewer areas would be shown as low sulfur (less than 1%) than with
the washed samples; however, the opposite is true (compare
Figures B-6 and B-13) due to the larger number of data points
for raw samples and particularly the much larger number in the
low sulfur northeastern part of the study area (compare Figures
B-l and B-8). It is very important to refer to the base maps
when comparing surfaces with different data bases.
Another set of maps was run in Phase II using a limiting
search radius of five or ten miles around each data point as a
study area outline. This serves several purposes:
1) it is not necessary to take the time to code the
bed outcrop as an outline, and
2) equal accuracy is maintained for all areas mapped;
no symbolism- ajapears for areas without data points such as the
southwestern part of Pennsylvania. Figure B-15 shows the base
map for this set of maps; the data point locations are exactly
-------
Figure B-l
Pittsburgh Coal Bed
Sulfur Content Mapped by Towns
Number of Analyses for Washed Samples
-------
APPALACHIAN
COAL REGION
3:.
••if:
IS,
**««!•!•
r~
\
~t
..™..™
- " "
Figure B-2
Pittsburgh Coal Bed
Sulfur Content Mapped by Towns
Low Values for Washed Samples
-------
APPALACHIAN
COAL REGION
,tu;i
fr<
*>"MH« R I "*!!••
"
S181S! i:!S l:ii hii i:ii
..™..
Figure B-3
Pittsburgh Coal Bed
Sulfur Content Mapped by Towns
Mode Values for Washed Samples
-------
APPALACHIAN
COAL REGION
ouuy c ji
•as-
Figure B-4
Pittsburgh Coal Bed
Sulfur Content Mapped by Towns
High Value for Washed Sample
-------
APPALACHIAN
COAL REGION
Figure B-5
Pittsburgh Coal Bed
Sulfur Content Mapped by Towns
Low Values for Washed Samples
-------
APPALACHIAN
COAL REGION
ims
Figure B-6
Pittsburgh Coal Bed
Sulfur Content Mapped by Towns
Mode Values for Washed Samples
-------
APPALACHIAN
COAL REGION
titltti r.K ..si!'
Figure B-7
Pittsburgh Coal Bed
Sulfur Content Mapped by Towns
High Values for Washed Samples
-------
Figure B-8
Pittsburgh Coal Bed
Sulfur Content Mapped by Towns
Number of Analyses for Raw Samples
-------
APPALACHIAN
COAL REGION
_YO(.Oa!
i t^Hti ••
i* .w?^
1 i
aSISS 3:f, t:», i:ii 1:1, i:Si i:s< !:i< MB
Figure B-9
Pittsburgh Coal Bed
Sulfur Content Mapped by Towns
Low Values for Raw Samples
-------
APPALACHIAN
COAL REGION
"MISS ;:!<
u!
-------
APPALACHIAN
COAL REGION
Figure B-ll
Pittsburgh Coal Bed
Sulfur Content Mapped by Towns
High Values for Raw Samples
-------
APPALACHIAN
COAL REGION
K1KK M'. ..!«'
Figure B-12
Pittsburgh Coal Bed
Sulfur Content Mapped by Towns
Low Values for Raw Samples
-------
APPALACHIAN
COAL REGION
™"j
_,:
'~t
KSIBS
Figure B-13
Pittsburgh Coal Bed
Sulfur Content Mapped by Towns
Mode Values for Raw Samples
-------
APPALACHIAN
COAL REGION
"•'sf
iiiiB
Figure B-14
Pittsburgh Coal Bed
Sulfur Content Mapped by Towns
High Values for Raw Samples
-------
I APPALACHIAN
! COAl REGION
"!
!'•
i«fl • ««*"
Figure B-15
Pittsburgh Coal Bed
Sulfur Content Mapped by Towns
Number of Analyses for Raw Samples
-------
APPALACHIAN
COAL REGION
r J
'<"«!«„!
...I...
liiin;
«™_.-:::::
a
ISISI lit hi*
!:S1
!
Figure B-16
Pittsburgh Coal Bed
Sulfur Content Mapped by Towns
Low Values for Raw Samples
-------
APPALACHIAN
COAL REGION
I,,,
«
"fc. "
.<..„.
ii:iiiiiu!uijiil:i:fi
•••"iiiiluliliiiii! ji;;i
ifiT
i
F
"r
T,, «•
•r
!•!
-Mi-
—i—. ---^™ JHHHH HHHH; :K!SM» yS&I
Figure B-17
Pittsburgh Coal Bed
Sulfur Conten Mapped by Towns
Mode Values for Raw Samples
-------
APPALACHIAN
COAL REGION
MlCOGCC'-oct \
IIMIIl'tl* • •l!tt>MI|£*i"
liiiiii:
•••iiilifeiiiir
lf"!iii
ill iiiiij
iiilll"
Figure B-18
Pittsburgh Coal Bed
Sulfur Content Mapped by Towns
High Values for Raw Samples
-------
APPALACHIAN
COAL REGION
•••Hii...
I:;::::::;::::::::
!
"-I
:ltllllllll
ill
1:4*
-WS i-5* 4:-S( 1:4!
Figure B-19
Pittsburgh Coal Bed
Sulfur Content Mapped by Towns
Low Values for Raw Samples
-------
APPALACHIAN
COAL REGION
'"' ™,f
I'^S"'
Z ""
UiPi
»».*••• IHHHM HliilM*
• •i*]l**« EMIMAMIH! KH«i7»M
Figure B-20
Pittsburgh Coal Bed
Sulfur Content Mapped by Towns
Mode Values for Raw Samples
-------
APPALACHIAN
COAL REGION
UJKS!
«* 4;!'
"sjjit" |;;;;r;;;
~^E= K::H::: SSHH::::]:::: {BOESE is::;::::
Figure B-21
Pittsburgh Coal Bed
Sulfur Content Mapped by Towns
High Values for Raw Samples
-------
the same as those shown in Figure B-8. The portions of the
study area enlarged in map series D are shown by rectangular
borders. Figures B-16 thru B-18 show the low, mode, and high
values for the raw samples with a five mile search radius.
Figures B-19 thru B-21 depict the identical information with a
ten mile search radius. In both cases the same values are
mapped as were shown in Figures B-9 thru B-ll; in fact, the
only differences are in the effective study area outlines.
2.4. MAP SERIES C
For the Middle Kittanning Coal Bed sufficient information
was not available to use the bed outcrop as a study area outline.
Accordingly, the data for the raw samples were mapped using a
five and ten mile search radius as was employed in one set of
map series B. Figure C-l shows the base map for the Middle
Kittanning Bed; the data points are very poorly spread out over
the study area and are clustered. Figures C-2 thru C-4 show
the low, mode, and high values for the raw samples with a five
mile search radius; Figures C-5 thru C-l depict the same infor-
mation with a ten mile search radius.
These sets of maps do not give us very much information
about the extent of the coal seam, or the extent at various sulfur
levels. However, they give us as much and as accurate infor-
mation in the vicinity of the towns (or mines) as if we had known
the outcrop area and mapped it. We can choose a search radius based
on the reliability we attach to the data and the number and spacing
of data point locations. Five miles may be conservative in that
the information conveyed graphically is scanty; on the other hand
ten miles may be misleading as to the extent of the coal seam in
some areas. Finally, it is important to keep in mind that we are
-------
dealing with samples aggregated by towns and that the data
point locations only represent the nearest town and not the
actual mine or sample location.
In examining Figures C-3 and C-6 (the mode values for
five and ten mile search radius, respectively) we find that 9
out of 90 or 10% of the towns have mode values of less than 1%
sulfur content. This is in contrast to the 26 out of 203 or
13% of the towns with mode values for raw samples less than
1% sulfur content for the Pittsburgh Bed (see Figure B-10).
However; the low sulfur data points are more spread out (less
clustered) in the Middle Kittanning Bed and therefore give the
illusion of greater areas of low sulfur coal.
It is of particular interest to compare the surfaces for the
low (Figure C-5);, mode (Figure C-6), and high (Figure C-7) values.
The surface of mode values is what we would be most apt to
find; the surface of low values is hypothetical and shows the lowest
sample found for each town, and, likewise, the surface of high
values shows the highest values found. Out of the 90 data points
13 are under 1% sulfur content for the low values, 9 for the mode
values and only 3 for the high values. In all three cases low
sulfur data points are spread throughout the eastern half of the
coal region and three points occur in the western half for the
low values. The mode values delineate the areas of interest for
possible small scale analysis and mapping. The high values tell us
that 3 of the 9 areas shown on the surface will have 'all1 low
sulfur coal, as determined by the samples taken. (Unfortunately,
when we refer to the base map—Figure C-l--we find that these
three points have only 1, 1, and 7 samples each and, therefore,
do not represent a very good statistical sample; however, the use
of the map of high values with the one of mode values is still
informative). On the other hand the map of low values tells us
-------
APPALACHIAN
COAL REGION
«iDcf c;i 655 coectU o
/ "V »•
I,
r8
!"*
Figure C-l
Middle Kittanning Coal Bed
Sulfur Content Mapped by Towns
Number of Analyses for Raw Samples
-------
APPALACHIAN
COAL REGION
•*•«»: "
•»!««• MB
*•;•;*;::;••!:::
***n;^^r;!ffl
3 ....
m "
A
1
!'«
i
{JIMS MS fcJ! fci* i« 4:4« 4:«
........
Figure C-2
Middle Kittanning Coal Bed
Sulfur Content Mapped by Towns
Low Values for Raw Samples
-------
APPALACHIAN
COAL REGION
..'"1 „""*'
.:::::::
::::
111
.
L m*Te«
! •"•
C.M
i "l
U!W 8rf! ISI 4:4! IS!
SrfS ):K ..!;('
Figure C-3
Middle Kittanning Coal Bed
Sulfur Content Mapped by Towns
Mode Values for Raw Samples
-------
APPALACHIAN
COAL REGION
I !1*S!!K ijjSEEUjjij E»»iES*
Figure C-4
Middle Kittanning Coal Bed
Sulfur Content Mapped by Towns
High Values for Raw Samples
-------
APPALACHIAN
COAL REGION
—,—"*tSK!
f V
«
Figure C-5
Middle Kittanning Coal Bed
Sulfur Content Mapped by Towns
Low Values for Raw Samples
-------
APPALACHIAN
COAL REGION
WSISS -i;4S 4:!S 4;«-
-4*
i:is
iiil-iiiiiiii-iiii
Figure C-6
Middle Kittanning Coal Bed
Sulfur Content Mapped by Towns
Mode Values for Raw Samples
-------
APPALACHIAN
COAL REGION
(!!!«« Ml 1:8!
S:4S
4:4S
Figure C-7
Middle Kittanning Coal Bed
Sulfur Content Mapped by Towns
High Values for Raw Samples
-------
that all but 13 locations have virtually no low sulfur coal as
determined by the samples taken.
2.5. MAP SERIES D
Map series D includes: (1) surfaces comparing the mode
values for the washed and raw samples for the Pittsburgh Bed
as mapped by towns and mapped by counties (Figures D-l and
D-2), and, (2) enlargements of different parts of the study area
for the surface of the mode values for raw samples shown in
Figures B-10, B-13, and B-17. (Figures D-3 thru D-7).
The values for the interpolated sample surfaces generated
from town data points were subtracted from the interpolated
surfaces generated from county data points. These surfaces were
subtracted at a regular grid interval of about 9/10" or 9 miles
on the source map. The new surfaces generated (Figures D-l
and D-2) are subject to considerable error. When we subtract the
two surfaces we get a good deal of detail which cannot be trusted
except along the eastern, western, and northern edges of the maps
where there are more data points on the original surfaces.
In. the ideal case, the detail shown on the surface should
not be greater than that of the lesser detailed surface from which
the map was generated. In this case the detail on the difference
surfaces should not be greater than the detail on the map generated
from county data points. The town map should be used as a
reference surface since it has the best accuracy in the outlying areas
and better accuracy than any place on the county map.
Figures D-l and D-2 show absolute differences in sulfur
content from -1:35 to +1:02% for the washed samples and -1.67
to +1.39%-for the raw samples. It would have been more
-------
informative, in retrospect, to have shown the relative differences
by positive and negative ratios. In other words, if at a particular
location the town map shows 2% sulfur and the county map shows
1% sulfur, then the value to be mapped on the difference surface
would be +0.5, using the town map as the reference surface:
Vd = (Vt - Vc)
Vt
where^: Vd = value on difference surface
Vt = value on town surface
Vc = value on county surface.
The actual values mapped were derived from the equation:
Vd = Vt - Vc
which would give a value of +1. 0% sulfur in this case.
Since we are really concerned with determining what areas
have less than 1% sulfur coal and which level of detail we need
to adequately show this, an alternative approach could be used:
1) use the town map as the control or reference
surface, and,
2) map only those areas of less than 1% sulfur coal
shown on the town surface and not on the county surface.
As a refinement the portions of the study area that lack accuracy can
be blanked out and only the 'horseshoe1 of data points on the town
map used as a boundary. This could be accomplished by using
the five mile or ten mile search radius areas of Figures B-17
or B-20 as the study area outline for the difference surfaces.
There is no easy way to create this new outline with the SYMAP
program; a new base map outline package would have to be coded
or else the difference maps would have to be generated with the
GRID program, as in Phase III.
-------
The changing accuracy of the maps can best be shown
by examining the series of maps for the raw Pittsburgh samples
by towns in the following order:
1) the base map with data points (Figure B-8)
2) the map with a 5 mile search radius (B-17)
3) the map with a 10 mile search radius (B-20)
4) the map with an unlimited search radius where
the entire outcrop area is filled with symbolism
(Figure B-10).
As we go through this series of maps we are adding areas with
decreasing accuracy. The same approach would be applied to
the difference surfaces using the search radius areas from the
town data points.
Figure D-3 shows a portion of the surface of mode values
for raw samples mapped by towns (Figures B-10 and B-13) en-
larged to the scale of the Phase III maps. The same areas of
Allegany and Garrett Counties as mapped in Phase III are shown.
Figure D-3 can be compared with Figure E-l, one of the base
maps for Phase III. In Figure D-3 the line legends for the state
and county borders were blown up from the Phase I and II base
map while in Figure E-l these legends were taken directly from
the Phase TIT source map; the accuracy lost by enlarging from the
small scale to the large scale is not significant.
When a map is blown up from a small scale to a larger
scale, the data points used for all of the smaller scale maps
should be shown. Then, the true area used for the interpolation
of the larger scale map will be evident and the reliability of the
map better understood. Note in Figure D-3 that the bar chart
below the map shows the frequency distribution for all data points
-------
in the study area, not just those in the portion mapped. The
frequency distribution of the ones within the map boundary
must be determined by eye and by refer ing to;the base map,
Figure B-8:
level 1 = 6 points
level 2 = 2 points
Total = 8 points
It is immediately evident from both maps (the one showing
1/2% isolines of sulfur content, and the one showing the critical
isolines of 0.5% and 1.0% sulfur) that this is a low sulfur coal
area. By specifying a detailed base map originally, it was
possible to blow up this portion of the study area and demonstrate
that it was worthy of closer examination, based on the sulfur
content of the Pittsburgh bed.
Perhaps the most meaningful maps produced in Phases
I and II are Figures D-4 thru D-7. These maps show detailed
portions of the study area outlined on the base map, B-15. The
"cut-out" region of maps D-4 and D-6 displays the greatest
extremes in the value range and in the spacing of data points of
the entire study area. Both of these maps appear at the original
base map scale of 1:250, 000 and thereby show the amount of
detail that went into the preparation of the bed outline. One map
was created using the regular SYMAP interpolation with an un-
limited search radius (D-4) while the other was restricted to a
5 mile search radius (D-6); at this scale the search radius option
produces circular shapes, rather than the 'squares' depicted
previously.
The upper-right quadrant of each of these maps was blown -
up to twice the original size (or, to a 1:|25,000 scale) to produce
maps D-5 and D-7; this is the only case where maps were produced
-------
at a larger scale than the source maps. This particular quadrant
was enlarged because (1) it is a large area with the lowest sulfur
coal of the entire region and, (2) there is a significant number
of data points that are fairly equally spaced. For these reasons
the maps are probably the best series of either Phase I or II and
convey the most accurate and detailed information, particularly
where the five mile search radius was used. Note that in
Figures D-6 and D-7 where the search radius has been used as
a study area outline, the bed outline has been shown as a line
legend for reference. We can now visually compare the extent
of the coal bed--based on the bed outline--and the relatively
accurate values for sulfur content--based on the search radius
from each data point. When our bed outline and data points
can be accurately located, we can use the bed outline as a study
area boundary but limit interpolation (and symbolism) with a five
mile search radius to areas inside the bed outline. For this
study area the information on locations did not warrant such a
detailed set of criteria.
Even the results attained in this series are somewhat
questionable, considering the extreme range in the number of
analyses attributed to the data points. Furthermore, this
problem is not as severe in this series as it is in the rest of
Phases I and II. In maps D-5 and D-7 there is a range of from
1 to 126 analyses per data point with most of the values being
near these extremes. The greatest range is found for the complete
set of data points in Phase I where the number varies from 1 to
2448. SYMAP does not weight the 1 analysis any differently from
the 2448 in the interpolation process; their associated values
are treated equally. With such an extreme range in the number
of analyses, the resultant maps are somewhat questionable and
should be treated with an even greater degree of caution.
-------
APPALACHIAN
COAL REGION
I
Figure D-l
Pittsburgh Coal Bed
Sulfur Content - Surface Comparison
Mode Values for Washed Samples
-------
APPALACHIAN
COAL REGION
«,•"''! «"
i ::
.«
,:
*"s":™ ">!!!» KS»S"!»"H!"l ;SJI
:;;;;:;:::-; ••::
:iF«?l ,/'
Figure D-2
Pittsburgh Coal Bed
Sulfur Content - Surface Comparison
Mode Values for Raw Samples
-------
| -=,;-
i i j
j ! sir" s
| ee
1 '•" |*"
j | Tj^,.,'! S
I f'S: -"XI
1 I-/""
3 5 «e ...
I s*T!
} S «j' '::::.
i i ««M .....
I
I -r*
| tMBMQ
e a
| ^«* LEGEND
1 / V0« COUNTY OUTL,He
a «e «j ee«e STATE OUTLINE
is *
| «° SCALEi IH MILES
1
SYMAP
GARRETT AMD ALLECAMY COUNTIES, ftARVLANO
PITTSBURGH BED - RAH SULFUR CGMTEhT IHCDEI
EXPRESSED AS A PERCENT BY TOM
DATA VALUE EXTREMES ARE O.dO 5.30
MMlMm 0.50 l.OO 1.50 2.00 Z.SO 3.00 3. 50 4.00 4.50 5.00
HAX1NJH 0.99 1.49 1.99 2.49 2.99 3.49 3.99 4.49 5.00 ABOVE
PERCENTAGE OF TOTAL ABSOLUTE VALUE RANGE APPLYING TO EACH LEVEL
11.11 11.11 11.11 11.11 11.11 11.11 11.11 11.11 11.11
11
! I
I
i
1
e
1 !
i ''
i
i ,—
i r--1' "
! .**"
i „„.,..„.„'
! /
1
« ee
1 /'
""I*'
SYKiP
PITTSBURGH 8EL - HAti SULFUR CGhTEMT IMDCE)
MNIKUM 0.50 1.00
PERCENTAGE OF TOTAL ABSOLUTE VALUE RAhGE APPLYING TO EACH LEV
100.00
FflEOU|NCt OlSTRjeUHON OF;CATA OCIKT VALUES IN EACH LEVEL 6 7 e , H "CKX" —"i" «" ' ' "LU" '" "" "'"
j,..a.s ::::i:::: —2— ;.„»—. IXXiiii BiSJS Sffl|^ ::"KU Uj;^!: UIK^ •«•**£!
"— nurt
sass
'
I
0 0000
« «me
COUhTY OUTLINE
STATE OUTLINE
Figure D-3
Pittsburgh Coal Bed
Sulfur Content Mapped by Towns
Mode Values for Raw Samples
-------
M«I«lIS*iI * 1 • Ul lUHuil* MHMU^IWJUM*
»»*..».»»
m!K«!m»»u"*H!u'''" i»
SiSlffi! l:a !:!! 1:18 i:8
I'M M*
lhi v"-u" 'K "CH lfv!L
Figure D-4
Pittsburgh Coal Bed
Sulfur Content Mapped by Towns
Mode Values for Raw Samples
-------
tiiiffi
!:J8 !:S! 1:1!
Figure D -5
Pittsburgh Coal Bed
Sulfur Content Mapped by Towns
Mode Values for Raw Samples
-------
HaSsssHSHS-Ec-H™
GOOD-
IE A. " i<^=-
^5.4r_-=-«:-E-^^-S'
Figure D-6
Pittsburgh Coal Bed
Sulfur Content Mapped by Towns
Mode Values for Raw Samples
Detail of Northern Sector
-------
,:. :.:,.:•:...is -.::.„
MSS
Figure D -7
Pittsburgh Coal Bed
Sulfur Content Mapped by Towns
Mode Values for Raw Samples
(Blow up of upper right quadrant of D-4)
-------
2.6. SUMMARY OF PHASE III
The data point locations in Phase III represent actual
sampling locations, or best estimates of these locations. For the
most part they are no more accurate than the known location of
a mine but this is sufficient at the scale of the output maps; if we
were to map a smaller scale area--such as one mine property,
then we would need to know the core sampling locations and the
general origin location for run-of-mine or tipple samples.
Because the data points are actual sample locations unlike
Phases I and II, if we assume no error in the data collection or
in the positioning of the data points, the contouring can then be
interpreted as being more or less accurate, depending on how
many data points there are and how far apart they are spread.
It is imperative, therefore, to display the data point locations
so that the level of accuracy can be clearly ascertained. In no
case in Phase III was the number or spacing of data points
sufficient to ensure reasonable accuracy, except in the immediate
vicinity of data points. In many cases the data points were very
clustered or were poorly spaced in relation to the outline of the
study area being mapped.
Furthermore, sufficient information was not known about such
parameters as the vertical location of samples in a bed to ensure
representative data for mapping--variations in sulfur content may
be due more to vertical sample location than the horizontal location
differences implied by the maps (Gluskoter and Simon, 1968, p. 15).
Until sufficient data is available to standardize the samples and
remove the effects of parameters other than spatial location, sub-
stantive results cannot be obtained.
-------
ALLEGANY AND GARRETT COUNTIES
STAGE ONE
Reserves-Mines
Traw-levelsJ
Base
Low
Mode
High
[thickness]
[quantity]
Mode
Three Dimensional
Reserves (Mode)
Thickness
Upper
Freeport
Outcrop >28"
E-l
E-2
E-3
E-4
F-l
F-3
E-l
E-2
E-3
E-4
F-l
F-3
F-6
F-8
F-7
F-9
Upper
Bakerstown
Outcrop >28"
E-5
E-6"
E-7"
E-8"
F-2"
F-4"
E-5
~E-6
"E-7
"E-8
"F-2
"F-4
Composite
Outcrop >28"
E-9
E-1(T
E-ll"
F-5
E-9
"E-10
F-5
UPPER FREEPORT
-si
I
STAGE TWO
[Yield-Levels]
Base
60%
50%
70%
80%
[Quantity]
60%
50%
70%
80%
100%
100-60%
[3/8" Mesh]
Outcrop
Tot. Tot.-
Pyritic
G-l
G-2 G=l
H-l H-l
H-6
H-7
H-2
H-3
H-4
H-5
H-6
H-7
>28"
~fot. Tot-
Pyritic
G-l
G-3 G-3
G-8 G-8
G-S G-9
G-10G-10
_H-2
"H-4
~H-5
[X&1/2" Mesh]
Outcrop >28"
Tot. Tot- Tot. TotT
Pyr. Pyr,
[Comparison]
Outcrop >28"
Tot. Tot- Tot. Tot-
Pyr. Pyr.
G-4 G-4 G-5 G-5 G-6 G-6 G-7 G-7
Table II-4
-------
2. 7. MAP SERIES E
Map series E contains the maps produced in
Phase III, Stage One, showing sulfur content for the
Upper Freeport and Upper Bakerstown Beds. Maps were
produced for both the total outcrop and for the areas over
28" in thickness, using precisely the same data point
locations and values in each case. Base maps are
shown in Figure E-l for the Upper Freeport Bed and
Figure E-5 for the Upper Bakerstown Bed. In Figure
E-l there are 20 data point locations; only 14 show as
black squares on the base map. Of the remainder 2
are superimposed at one location (in the cluster in the
upper center) and four occur under the state and county
boundary legends. The full set of data points and
accompanying information can be shown by running an
additional base map with (1) no legends, (2) distinctive
symbolism for the superimposed data points, and (3)
the number of analyses shown as legends above each point
as in Phase I and II. The number of analyses was not
shown since out of 20 data points only 7 had more than
one analysis and only 4 of these more than 3 analyses.
Since the number of data points is also very small we
are clearly working with a poor statistical sample.
Of the 20 data points 7 (including the two that are
superimposed) are clustered together in one small area;
the remaining ones are not well spread out over the study
area. The problem is particularly acute when the areas
over 28" thick are shown. Of the 16 points not hidden
by legends, only 5 fall within the study area boundaries
and the major portions of the eastern syncline contain
no data points.
-------
For the Upper Freeport bed the original base map
contained several other outcrop islands, mainly in the
northwestern part of Garrett County. These were
eliminated because there were no data points associated
with them. This problem did not exist with the Upper
Bakerstown base map. The question arose as to the
validity of interpolating across the area between the smaller
western syncline and larger eastern syncline (see Figure
E-l). It is possible with SYMAP to place a barrier to
prohibit or inhibit the interpolation when appropriate.
Since this area does not have complex geology, we can
assume that geologic conditions when the coal beds were
laid down were the same for both synclines and that the
history since then has been relatively the same. Without
any information to the contrary we assumed that there was
a single formation which was eroded in the middle to
create the gap between the synclines. Therefore, we
treated them as one formation and interpolated between
synclines without the use of a barrier.
Figure E-5 shows the base maps for the Upper
Bakerstown Beds. Of the 7 data points four are shown
as black squares (2 of these are clustered in the western
syncline), 2 are superimposed (in the upper right corner
of the eastern syncline) and 1 is under the state boundary
legend in the center of the study area. Only the super-
imposed points are inside the outline of the area over 28"
thick; this reduces the accuracy for this series of maps.
Figures E-2 thru E-4 and E-6 thru E-8 show the
low, mode, and high values for sulfur content mapped by
mines for the Upper Freeport and Upper Bakerstown Beds,
-------
respectively. The data points were located for these
particular maps on the basis of geologist's reports at
the Bureau of Mines. Locations of a mine are generally
no more detailed than the number of miles from a town
in a certain compass direction. The location of a sample
in the area of the mine is almost never known. A further
shortcoming of the data is the lack of information as to
where the samples were taken in the bed. Very often the
concentration of sulfur is quite high at the top and bottom
of the bed but not in the middle. If the samples are
taken near the top or the bottom they may give a high
sulfur reading when the average value of the coal may
be low. Ideally, the data should tell us sample location
with a great deal of detail in the x, y, and z directions,
where x and y are the spatial location of the sample and z
is the sectional location in the bed.
With these shortcomings of the data in mind we
can examine the maps of sulfur content. These maps
show, in general, that the higher sulfur content coals
(by percent) are located near the center of the study area.
in the Upper Freeport Bed; however; in the Upper Bakers -
town Bed the higher concentration tends to be in the
southwest section of the study area and the sulfur content
decreases towards the center and the northeast section
of the bed.
As an example we can examine Figure E-3, the
map of mode values for the Upper Freeport Bed. As
shown on the base map the eastern syncline has one data
point in the very north and several in the middle portion.
The interpolation between these displays a gradation from
a low concentration of sulfur in the north to a moderate
-------
amount in the middle portion of the syncline. The
interpolation is affected by the data points in the western
syncline. One cannot be certain at all about where the
isolines of sulfur content should be. The best one can
say is that there seems to be a trend from low con-
centrations at the study area boundaries to intermediate
concentrations toward the middle of the syncline.
In general, one should be very cautious of the
extrapolation beyond the data points at the extremes of
the region, and of the interpolation between data points
which are fairly far apart. The misleading extrapolation
is shown in the bottom left hand corner of Figure E-6;
the potentially misleading interpolation is evident in the
center of the same illustration.
The final set of maps in Series E were composite
maps of the sulfur content values for both the Upper
Freeport and Upper Bakerstown Beds. These maps
combined the A-OUTLINE, B-DATA POINTS nad E-VALUES
packages for both the Upper Freeport outcrop and Upper
Bakerstown outcrop. Maps were then made which assumed
that both beds were part of the same formation. Similar
maps were made for the areas over 28" thick, for the low,
mode, and high values, (see Figures E-9 thru E-ll).
This is really only a mapping exercise: because of
the difference in elevation separating the two seams, the
composite maps produced are of very little value. They
assume that the two beds are fairly close together for
mining purposes, which, in fact they are not. They also
assume that the two beds are similar enough in geologic
-------
II
i i
i s
MMs?s?!Sfss?;;?s?s???8Meeo8ef
-?•**«***« »***»**
If i r
r
»§»
KB:
"»"• S
I | GARRETT CO
i i
!i
I 9
1 1
u
'I
I ! ... »:
I | ++JJtJtt+
i i :::5HH2
i S »*• t**++*t»e*
i 1 .HlHj?"
! • tltit tt^ta
} e+nn> **ee
j 'ir5""
i
*:::::::::t::r "...
•::::::::::::::• •
.:::!::::::::r
...j:::::::::::::j ivs*
.::::::::::::;::;::%98
**«*»**»**+**»;9
.:s::ts:;::j;::ss*
^5:5" -a:-
mea
0
SfMiP ' *
TINE - 0.0
1 3UIFIRI CONTENT - B13E HAP
G«RRFTT »HO ALLECtKT COUNTIES
t*
.**•*. CENrEH PORTION
0 0000 COUHTT OUTLINE
8 9BW ST»TE OUTLINE
SCALE! IN HFLES
S
]
j
|
!
I
I
Ii i
| | „«,,„
I I
II
Is
II
1 s .«H:
i ! ::::::;!
i! .s::::?
?! ..HH U"
i I :::»: •
j !%rK"
. /
.„: •' .„..-'»
***•• ***** e e
** ***;«
ew
€
• as9
:. »'"'%»'S
0
TIME " 0.0
UPPER FREEPORT BED AREA OVER IB- THICK
ew
WEST VI
0 0000 COUNTY OUTLINE
0
e ewe STATE OUTLINE
S!!BS
I POINTS IS 2. THESE OCCUR IN I LOCAT
S j?^? tSSIt ,
SIfiSi!
=BEO. 20
Figure E-l
Allegany and Garrett Counties
Upper Freeport Coal Bed
Sulfur Content Mapped by Mines
Base Map for Raw Samples
-------
ti
I i .iiii
i | ::::::::i't;
IMJKSSS5*
SijjjjsjjHjHSr* e f
«»
•«
'
1 |
1 i
Dl**»l**«**
i5s:::s::'
8:38 i:SS
~ SSSS !»!!!!!
Figure E-2
Allegany and Garrett Counties
Upper Freeport Coal Bed
Sulfur Content Mapped by Mines
Low Values for Raw Samples
-------
i
i
j
j
i
f i
i ;
j
I i
i i
II
i 1
II
• :
i 1
! '
TI*E
» SUL
»ESOL
F.tUU
SVC
'"Si
* 4 1* *•!•«• 4
••••»!•**•!•••
,;**•••*••••»»•*
illF
,
•
*
^^":::;iM*i::l«
M* •«
— — """.*•
— ioiii*
•5^*"*"
i
0.0
. , tll[..v.,, „,.,„.„, ,. ,.„ L.v
«.i.a.......5....!....!....S....!....»....^.......:..;.J....S....
:!!!!!==:"=r .:!===in=:. 1 i
:"" ..:r;£::::i=;:E::::::"(lllc..,
•nliilsllliliililiiiF .~'v
.JK::.;;;.::.:!.-!::::::"' j ! i
NM """S".!!!**'.^.!8^" ** S**** ^
"!«••«**«" .1 i
I
1
i
i
i .,.,.< |
1 B **M STATE CL1LIN6 1 j T
1 I i
1 [
SCALEi IN MIES j
1 * ^ 1
!
+ J fl S . 1 s
»
UK
l:« ,-.« ::« .S
„
Ililiiiiilil
.................M««^.'"'™u""""""a<'^'^mte'"!'^i""""a"t"^1""i""Sii">,
{ • - T =: -"',»
,'' -="• J1
srk = s!
»8*I!*!IIJ** M* *
,.^ J:^_::> 3
.==•••••• .•' "' j
..--r-tU""" C J
•Fi— --~-i a ««« ST»Tt CtTUhE 1
ULFUR c^:^Tl^T - OCCE KREJ cvfo ;c n-icot
Et fSfctFCSI EEE
oitTf.aas.!.'Hc(sst.'.s H.KS. unit «».
ssir .„,.,.«,„ Cfi,,,, ,«,„ v,luss ,, E,(t t!,|t t
YKECIS i. . — i — :"••_•_:: SJK**** :u.i.:u IfHlju" SSfmH jsjlpii IlilliiP
; i ;
Figure E-3
Allegany and Garrett Counties
Upper Freeport Coal Bed
Sulfur Content Mapped by Mines
Mode Values for Raw Samples
-------
J. ....... ^^^——^^^^
1 , . JiffP" \
1 I JiiaaiiUiii'' .:::8:=:==:- .1;*,, ."j
I .alasaaaT .~S!:===E»EHf | j
| - U!N&MMS^ ..4SJ3JIJ5+JS3S""" *• !
a ""!!!:' ,::::::::;s::::;::::::::r* «e 8B \
8 .jp:J!:p::::jp:" ...» \
\ I ,!!:K»!!;i:ass'i::s:r" j i
1 1 - •":• "iil-HJiif " 1
i 1 ~IUll:_::2%sas? 1
S ...::=—:::": r.V i
f | ...:::::=-=::::-.i " £ s ' " ' " ' ' "' ' i
.1 ::ii;;ill!i!i::r- j
I .:::::a:::::::^ 1
-^ — =.*"'jjrMi"?4 j
• ! — ^=;:::::« •«. i r >
.! "Z^ns""4" ' °°°°" ""r""Lll"'s j
I • ..::::::::se • !
St-*"'1" c ! 10 j
I * *-=^= * ' 1
„„- 0.0
1 SULFLft CONTENT 1G V LUE ITCT L tU C C 1
UPPER FBEEPCBT BfC 1
DATA VALUc EXTREMES i»t O.SO S.tC
KIN1HJH 0. 5C 1.00 1.5C Z.OO 2.5C 3. CO J.50 'r.tC 4.3C I.CO
f 4X1 HUH 1.00 1.5C 2.00 2.50 J.CO J.50 '..GO 4 .SO S.OO
* ...«>«= ••**>*++« *•**•***• sjjssjjjsji SJ15S55S55 !••!!!••• ••••••••• •••SiiSSS
stMotLs .""^rr"^ im^^n »l«^"»^ TtTittTT< ISSIsIII* SuSaliMS KEB'SSS •u!"!!! ••••"••IS ••••!<••••
••
-T
!
"f
H«.i.«.<..<~«.««««..M""°"''""e" i;""""" "°™"™ •"' | , j
jf "' -^
• ' 1 = \ — l"
* S -^"^— B
4n. - «s j
I S:!»*» « {
g!!!*!!!!!' 8ea«€ :
;=: '. 'Xf"^'
s=-. ::::.: • j*S*
•M***
a
I'8* :
_-i- ^f-. -t i
,=:g:ai.' v^ ;™;;'^^; !
_.™:::::r | ' |
.«"**' C I 10
. ...
.B FBEEFCFT 6ft
V.1.1 «»E>,< ... ..5.
H«Mj":M«""'L:«r"ic;* i=. ^ ™ ™ ™ -
S|,G, t,S,,,ft,lC« GF.t,,. »»J V.L.ES ,} E,CK LS.|L ,
Figure E -4
Allegany and Garrett Counties
Upper Freeport Coal Bed
Sulfur Content Mapped by Mines
High Values for Raw Samples
-------
:J
8°:
1 !
Figure E -5
Allegany and Garrett Counties
Upper Bakersiown Coal Bed
Sulfur Content Mapped by Mines
Base Map for Raw Samples
-------
i I
i i
i i
i!
***^*OB"*'a8eeB8^B"S8° * eseeeieeeoetoe
~ipy
::«!'"«"
•:::^s. •:••".
^-.=. e« fle%8 % e
..SKSBST
*»»«»iix«»»aa
s::i
ii«B«8aeeeelaflO68flB«o
UPPEH E.KE.SIthN EtD
ISTRIBUTIGfc OF C6TJ
>**t*«l* KMUtKKKM*
IH, tttt *********
K • tan ui e IIHHIH
! ISKSK HSSSS
:::::::... xisas
i Tt-tSt CCCUB IN 1 LOCATIONS.
:^ ;;;;:•:;;• iiii iii iiiiiilii 1111.1111
Figure E -6
Allegany and Garrett Counties
Upper Bakerstown Coal Bed
Sulfur Content Mapped by Mines
Low Values for Raw Samples
-------
i :
GARRET! COUNTY
.*«**«****
..3:1
*jgj{5..««;«*" *<
..—.
IMI
IT
JJaF'
•r
> 13! » * 0+**+** - " mm " '
™!S.U£
~_r—
--- 1 -- a
---- A £
-
"
, -s
:
UPPER EiHteSTCliN E
HESE CCCUR Ik 1 i
SHI! JMSSKJJ "HMMMKM. i
*S**** VMMII6MM* MNHMTHKHM I
tsss sssssiw* ;i:"!!"s i
Figure E-7
Allegany and Garrett Covinties
Upper Bakerstown Coal Bed
Sulfur Content Mapped by Mines
Mode Values for Raw Samples
-------
i
•
i
If
i
•
1 i
i '
:
• w
sii
%K
*
iff •
™
UILLI.
H.JVUE
„»
MMflimcHMMa0M«m«^M^M^MeB^e<^M^ea*SS*"sH"^3S.^^^*"*^^^^^^^i""^""""''"^B"^i'
»*SS)i«»I"""*" o ;
•JHSKSS" -2= "
,J1IP:: _:]!:==:
M!,!8 "as" :::;&^=~- " ALL,.,*,
..!!»« IS" .:::;?:::::=::=:— "L""' ;
«****44**o?**».. ...... «aa%o*
.."I:::::;;:::::::::::" ,^
'^?k^'~- ."••''
.::::. ..."•"• 8 I
."ri"."g ««•» ea
.--:™:»2" ""«. S
— •••• ;V
=• $ x- „., .,..,.,.
ni««««*M 6
"".nil SOLE! IN K]i:v.
sas"
H0« 0 S 10
c.o
EAKERSIChN EEC
r:r:;:;":fiht;;;;
•» •!:«:::• ~ =S= :^~: J::P:::: ||1 :i|| |||JiBJj jJiHSS
so] i * ' * '"" * J ieU6Mee6oB«aoe«M6«eeBoi^M«l«*efl«eaBfl8eae9eee5«awtiM«MMewaBMB«>*«4i«e«8ieieeei««
| S o= /
i ! r^ —
j | -g=m^n-
! ! , sis-:: s|
| j,-^,, .*'
i" 8:™?SK . .'
I /*r? S
i 1 ...A r'
/ '— . ,-'
! jj ;• xj
'i i J
! ! ••
! ! /••'• .,„ „..,.,.
! 1 x^
i ! '/
i i! .... f
1 1 ..- - ,£56.0
j j . js. 03..00 CC»hI< OUTLINE
{ 1 e fl a «««n STATE OUTLINE
: !
: • .»
i «..«.--' . s"iii r"uss
! i • j ^-"
T,,. - ...
TSaH^™:BH'T!t«» .:« 1^1 <:» ^ i:ii
LtVcl. 1234 Lt"|L 6 T a 9
— ;;;;;;;;; -— ;;?==: ^=H spn ||pj slips !•!•:•.!• »i|
2 1
Figure E-8
Allegany and Garrett Counties
Upper Bakerstown Coal Bed
Sulfur Content Mapped by Mines
High Values for Raw Samples
-------
I .sSiBF _ 88 -_r '"""" I"!
1 -SSTssa" :=~gr- ~~ J \
1 I .n-::::::Hr HH:8: :::•— — ~ :' {
1 ::|!!!;!jj™"!"'8' 7^ J i
i « GARRLTT CCUKTY ******»««*e 66 5
j i *tTt *t< !«S»» * B eeeae a i
3 ~~~~""-' ."^p89866 }
| _-•-• "Ofloe e f
/ \
i 1>^ B56 i *
!l .::::::::3i1' ' i
"v"*" ° __; l° j
.. ..^ L J
TIME - C.O
I SULFUR CONTENT - LOU VALUE (TOT*L CUTCRCPSl
CdRRETT ANC ALLEGINY CCUNTIES
-",.««„,. «««„,
::::::::: ==: —~" S5SSS SBSESS! ISlKiR KKU!!! SSiSSi
, S .-_ - COUkT.
S . / =^=:
1 8° ' -^ f
\ 11!" / "
«' * sSHH? ..s
1 .;a:.:!:!:::' |"
§ oe •***» oeee
i g"..-6-8 v-;. j
I , ' 8i8' '"'
• . j,j ,
i .•""*
i _ • y
i S - «.'. S .
| :: : :::i"
•S ..::::»•
1 i;;U«'
bAKkETT ^t 6LLE6/rr CCUMIES
^^,,»^^,^^^^
— :::H:»: S — HiiHiH -HH:: ijl||jj |JB |S jjjjii jjjijjj;
Figure E-9
Allegany and Garrett Counties
Composite of Upper Freeport and Upper Bakerstown Coal Beds
Sulfur Content Mapped by Mines
Low Values for Raw Samples
-------
UTC'H*!(l«uB5hIHC*WED1IN HIGHEST LEvR OHIYI
11.11 11.11 11.11 11.11 11.11
'! "" LiT
SYMBOLS 1.
•••aSi'MSiJilfBii iiii'iiii
:;:::::;;: aasss :::::&:: usaua
TIKE • C.O
Figure E-10
Allegany and Garrett Counties
Composite of Upper Freeport and Upper Bakerstown Coal Beds
Sulfur Content Mapped by Mines
Mode Values for Raw Samples
-------
_-§i=::::i-:-
If
! S
! 1
!!
1!
1 i
NNNM*O! S8
!l!!!,!!!
••
•:::::i::::i'
-.5™ 'iii'
;! „ „„
i I.THrrHH!'
*N3 ALLEGiKV CCUNTIES
i::;:::;:;:::::!»:::: sss&ir. ISSSK
EiilEHKEifiifflifliiiEii
-— : :::::::= :s«H:; s:?:::: iiiiiig Jjii
:::: ;::; i_;_^ ....;,= £££. ....J..., !!S!s!S! KSSiKS i:;:
Figure E-ll
Allegany and Garrett Counties
Composite of Upper Freeport and Upper Bakerstown Coal Beds
Sulfur Content Mapped by Mines
High Values for Raw Samples
-------
formation and sulfur content that all points can be used
together for interpolation; comparing Figures E-3 and E-7
tends to show that this assumption is also probably in
error.
If we confine our interpretation solely to the areas
surrounding data points, then the composite map can give
us a quick and useful overview of any and all low sulfur
coal areas for the beds we are interested in. The
limiting search radius in conjunction with the outline would
be very useful here, particularly for the areas over 28" thick.
2.8. MAP SERIES F
Map series F contains the maps produced in Phase III,
Stage One, showing bed thickness and quantity of less than
1% sulfur coal in tons per acre for the Upper Freeport and
Upper Bakerstown Beds. Figures F-l and F-2 show the
bed thickness in inches for the Upper Freeport and Upper
Bakerstown Beds, respectively; these figures also serve
as the base maps for the thickness data. No base map
changes are made other than in the location of data points.
For the Upper Freeport Bed there are 7 data points, only
4 of which are inside the bed outline. Since the low value
of 30" is in the upper center and the high value is in the
lower left hand corner, we get a gradual slope from
northeast to southwest. This is based on too few data points
and too little change in absolute values to be very conclusive.
Major discrepancies are evident when we examine
the map of areas over 28" thick in Figure F-l. Although all
of our thickness values were over 28", the source maps
used indicate only a small portion of the outcrop is actually
-------
over 28" thick. It would appear that the thickness data
are not representative or complete enough to be meaning-
ful. It may well be that thickness varies considerably
on a small scale, particularly in relation to the absolute
value ranges found in this set of data. The interpretation
of the map in the southwestern portion of the eastern
syncline is very much suspect because the data has been
extrapolated 15 miles. Likewise in the northern part of
this syncline there are no data points; it is the data point
in the northern portion of the syncline to the west which
greatly determines the values shown in this portion of
the eastern syncline.
In summary, only the central portion of the eastern
syncline where 3 data points are located can be considered
representative and meaningful. This should be kept in
mind in examining all of the quantity maps for the Upper
Freeport Bed in both Stages One and Two; the quantity in
tons per acre is directly proportional to the bed thickness
shown in Figure F-l.
The data point locations and their relationship to
the Upper Bakerstown outcrop are more reasonable (see
Figure F-2) There are considerably more data points
describing the thickness of the Upper Bakerstown bed
than there were for the Upper Freeport bed; therefore this
surface is more representative and meaningful. Only one
data point has a value less than 28" thick and this is out-
side both the outcrop and area over 28" thick outline. Of
the 14 locations (16 data points with 3 superimposed at the
same location) all but 4 are within the outcrop outline but
only 1 is in the area over 28" thick.
-------
This results in a very serious discrepancy between
the area defined by the source maps as over 28" thick and
the portions of the outcrop found to be over 28" thick
according to the data points and values used. This can
be explained by one or more of three assumptions:
1) the data base or criteria for "thickness" are
different for the Bureau of Mines thickness data and
the Boyd report isolines of bed thickness.
2) there are significant errors in one or both of
the data bases.
3) thickness varies a great deal locally; therefore,
we cannot interpolate a surface from so few points,
nor can definitive isolines be drawn as in the Boyd
report.
The "coal seam thickness lines" from the Boyd report and
accompanying maps and the bed thickness data from the
Bureau of Mines do not appear to differ in their criteria;
likewise, there is no reason to assume significant errors
in either set of data. However, according to the Bureau of
Mines the manner in which thickness is measured may vary
significantly from location to location and thus introduce
errors as well as bias between the two data bases. The
third assumption may well explain some of these discrepancies.
but there is no way to standardize for the other two.
Figures F-3 and F-4 show the quantity of coal in the
ground (in tons per acre) for those parts of the Upper Free-
port and Upper Bakerstown Beds, respectively, with less
than 1% sulfur coal—based on the mode values for the raw
samples. The 'accuracy' of these maps is a function of
(1) the maps of sulfur content (Figures E-3 and E-7) and
-------
(2) the maps of bed thickness (Figures F-l and F-2).
As discussed we know that the accuracy of these surfaces
from which the quantity is derived is very poor. However,
the areas shown in Figures F-3 and F-4 are ones where
data point location--and therefore reliability--were much
better than average in each of the four original maps
(Figures E-3, E-7, F-l and F-2). These are the first
of the maps produced with the GRID program; a separate
value is mapped for each cell or computer print position
based on the original maps. No interpolation takes place.
Figure F-5 shows the total quantity in tons per acre
for the Upper Freeport and Upper Bakerstown Beds combined.
It is not a summation of Figures F-3 and F-4; it is derived
from the composite sulfur content map (Figure E-10) and the
combined thickness (the summation of Figures F-l and
F-2). The significant reservations about all the original
surfaces in this case make these maps quantifiably mean-
ingless; however, the approach is very useful if accurate
data are available from the original maps and the seams
being combined are fairly close together in elevation. Such
a map can help locate areas of significant low sulfur coal
that would yield a greater total quantity of coal when all
mineable seams are considered. Ideally, we would include
another variable: overburden, so that the economics of
mining could be included.
Figures F-6 thru F-8 are three-dimensional views
of surfaces produced with SYMVU for sulfur content and
bed thickness for the Upper Freeport Bed. Figures 6a
and 6b show two views of the total outcrop for the mode
values of sulfur; Figures 7a and 71-, show the same values
-------
i c««e««aM«6e
: i
i!
!ss
II
i !
+j°j**~«*:****''i
.:iM::HH:::::::
MNM!!!!**!!"*
«««*«•!£»?
.::::::!«••
tSSSSSf
•«:•
**l****t**0>****«********
• •••»*K«**p*******4***»*
********«**0l««l**«**«***
HnS»Mium**«»o •»•••**••*•
Jlljjji8 ""'° TL. J
Eai- "... I
11 ,i
I I
» !
«r
rKICKNESS OF BED IICT4L CUTCPCP]
«METT AKC ALL£G*KY CCUMTIES
KIH1HUH It.OQ 20.00 J-..CC 28.00 32.00 3f.CC
f^AXI^UH 2O.OO 24.OC 26.00 32. OC 36.OC ".O.CC
>£RCENT«GE CF TOTAL ABSOLUTt VALUE R1AGE /PPLUNG 10 £JCH LEVEL
;KE" "'""!"
1 PCIhT VALUES IN E«CH LEVEL
i'HHS—'HJHHHlrss^rsJigssysls;!!
lolcc iSloc telcc !!lcc aeTcc
11.11 11.11 11.11 ii.!
Figure F-l
Allegany and Garrett Counties
Upper Freeport Coal Bed
Thickness of Bed
-------
II
If
II
[I
.
..sllssr'
„
'
'* .**** e
* / , * **+**ea
,,!,.K;-::. ..»<
!"!|s:.",!.8.'"
,-FlP*
,,,.S,*,,.»*KfiSSSS 5
,;:$::*:":::::::&* J
J>LLE6*ST CCUHTISS
"°LUIf.l!jH6ulH'5fetSEaIS SMffV tISIt ONLY.
EHitS JS:88 IS:i8 18:88 IS:88 si:8S IS:
IN i.e. L»|L
" ***""*"*•
.1J 11.11
. ...>.•.„ 4*-t-+tH-t+ •*•*••»•* MKHNIMII* •*«•»**«• IUIIU
:::::;:— K:SS:: ;»;J»;: ;«;]•!» !!K"?a (iBIS
Figure F-2
Allegany and Garrett Counties
Upper Bakerstown Coal Bed
Thickness of Bed
-------
J
j
I / =r
6 B •
f / :|
| « A R Y I . N 0 55 .•"„•
/ ./ X
i .i. .— •'
i «„«„„».,, /"'""" '"'". -
f 4= «j'
•1 ~ .•/•''
f
8 BS ft een» STATE OUTLINE
fl J| ^^
„•••= ==i;= :;s:::::™n^^.=SS:^:-n^^:-5SSS =!KS::!!!L!;5!;!S!?L!!fl^.™ffi?.5!H!SK=
. _ _
i / _
i * «"•" r
i ! 8° ..«
i 1
S „ . . , 8 ._ '.
1 / /"
: s «° ,'*
I
! / j
i , K ."••
i I „"«." • ! !
' * 1 «•« ««
i! ~ / X!
1
1 «»""*
i { rs
:: r
i! . ... is
1 I «*eaaBae 0 OCOO COUNTY OUTLINE
i a ae
j | 6 ft «eM ST4TE OUTLINE
: i
1 2 9a
i i .».- o r
» "/,.,„"' J£..:a""tisS! "MSHIP «S,g.!!f s-8;jc- ™» «.!'"•'-
KiK »fc:, iffikJI 8fe:a «J?:a Hto, Bttb W:^ «:88 2IS8:88
ii-i 11-11 u.it 11.11 ii.ii 11.11 11.11 11.11 n.u
.::^^s^.:::::i- i-^:nn^n"^:-:Jb™.^*SH!L;«sSML^^S5.'^™J!~H^^^,
:UUENC¥ 2
Figure F-3
Allegany and Garrett Counties
Upper Freeport Coal Bed
Quantity (in the ground) of Coal with less than 1% Sulfur
Based on Mode Values for Raw Samples
-------
I / '" ™™
to ee
0 COUNTY B
i / f
| « « « V L « « 0 ^0 .'"„•
I1 f •/
I .. J. ,""'""
| ' "" _/ ""'"--1"
| ""eea0
f
I o*** 9o«*
1 aaaea8 OOOOOD COUNTY OUTLINE
| 98 s e&ee STATE OUTLINE
8
V , + ,
!
i
t
i
I
2"
I
!
I "™STI cou""
1
1 J
1 . f
1 f ""
1 »••"•
I .•«'
1 68B
a ewe
..,
"-J
SCALEI 1H HILES
—I
iii:;::::
:;!,%:::;
i»E:i!i
:,•:;:«: SB
.,G.,JL
.I::,-::::!;;:;:::;';::':;;;:
**>0>n* KHHHUNHNX IKIIM7MWCM
::::::::;;;;;;;;; ustiS",
1O
Figure F-4
Allegany and Garrett Counties
Upper Bakerstown Coal Bed
Quantity (in the ground) of Coal with less than 1% Sulfur
Based on Mode Values for Raw Samples
-------
H
« J
V,, |
8
~l«f-gj""j=>'"l" ««g.« 5Sg;-Sl- J?,r,'f S^r-s
Figure F-5
Allegany and Garrett Counties
Composite of Upper Freeport and Upper Bakerstown Coal Beds
Quantity (in the ground) of Coal with less than 1% Sulfur
Based on Mode Values for Raw Samples
-------
tNJ
1.50
1.00
0.50
0.00
, 3.67
2.UE
1.22
0.00
Figure F-6a
Allegany and Garrett Counties
Upper Freeport Coal Bed Three Dimensional View
Sulfur Content Mapped by Mines
Mode Values for Raw Samples
for total outcrop; azimuth 0
flZIMUTH = 0
WIDTH = 6.00
014/19/71
flLTITUDE = 60
-------
i.oo
o.so
o.oo
3.87
2.80
1.30
1 0.00
Figure F-6b
Allegany and Garrett Counties
Upper Freeport Coal Bed Three Dimensional View
Sulfur Content Mapped by Mines
Mode Values for Raw Samples
for total outcrop; azimuth 309
U. FREEPORT BED (TOTflL OUTCROP). X SULFUR CONTENT (MODE).G4fl CO.MflRYLflND
flZIMUTH = 309
WIDTH = 6.00
flLTITUDE = 45
-------
1.50
1.00
, 2.87
1 1.91
0.50 4- 0.96
0.00 1 0.00
Figure F-7a
Allegany and Garrett Counties
Upper Freeport Coal Bed Three Dimensional View
Sulfur Content Mapped by Mines
Mode Values for Raw Samples Q
for area over 28"thick; azimuth 0
RZIMUTH = 0
WIDTH = 6.00
04/16/71
RLTITUDE = 60
-------
, 2.87
1.00 1 2.03
O.SO 1 1.01
0.00 1 0.00
Figure F-7b
Allegany and Garrett Counties
Upper Freeport Coal Bed Three Dimensional View
Sulfur Content Mapped by Mines
Mode Values for Raw Samples
for area over 28" thick; azimuth 309
U. FREEPORT BED (RREfl >28"THICK) .7. SULFUR CONTENT (MODE).G4fl CO.MflRYLRND
flZIMUTH = 309
WIDTH = 6.00
flLTITUDE = US
-------
1.50 _ 48.00
1.00
0.50
0.00
32. S.O
16. CM)
0.00
Figure F-8a
Allegany and Garrett Counties
Upper Freeport Coal Bed Three Dimensional View
Thickness of Bed
for total outcrop; azimuth 0
flZIMUTH = 0
WIDTH = 6.00
01/16/71
flLTITUDE = 60
-------
1.00
O.SO
0.00
, 18.00
•J2.11
35.07
23.00
Figure F-8b
Allegany and Garrett Counties
Upper Freeport Coal Bed Three Dimensional View
Thickness of Bed Q
for total outcrop; azimuth 309
U. FREEPORT BED.THICKNESS OF TOTfll OUTCROP. GRRRETUflLLEGflNY CO.MRRYLflND
flZIWJTH = 309
WIDTH = 6.00
04/16/71
flLTITUDE = 45
-------
00
i.so , te.oo
1.00 .. 32.00
O.SO .. 16.00
0.00 1 0.00
Figure F-9a
Allegany and Garrett Counties
Upper Freeport Coal Bed Three Dimensional View
Thickness of Bed o
for area over 28" thick; azimuth 0
flZIMUTH = 0
WIDTH = 6.00
04/16/71
ALTITUDE = 60
-------
1.41 - 48.00
1.00 .. 42.14
O.SO .. 3S.07
0.00 1 28.00
Figure F-9b
Allegany and Garrett Counties
Upper Freeport Coal Bed Three Dimensional View
Thickness of Bed
for area over 28" thick; azimuth 309
U. FREEPORT BED.THICKNESS OF flRERS >28",GRRRETTAflLLEGflNY COUNTY.MRRYLflND
flZIMUTH = 309
WIDTH = 6.00
flLTITUDE = 45
HEIGHT = 2.00
-------
for the area over 28" thick. Figures 8a and 8b show
the bed thickness for the total outcrop and Figures 9a and 9b
for the area over 28" thick. These views were produced
to demonstrate this particular graphical portrayal. A
set of views of a surface from several different angles
and altitudes can give a great deal more information about
the whole surface than the arbitrary isolines of a SYMAP
or GRID two-dimensional representation. With special
versions of the program legends and contours between
levels--such as the isoline of 1% sulfur content--can be
shown on the surface.
The views chosen here are (1) "head-on", or from
the south, and (2) from the southwest.
2.9. MAP SERIES G
Map series G and map series H comprise Stage
Two of the mapping. This stage involves the mapping of
washability data for the Upper Freeport Bed for the center
portion of Allegany and Garrett Counties; this portion is
shown in Figure F-l. Insufficient washability data was
available for the Upper Bakerstown Bed.
Map series G is concerned only with the sulfur
content resulting from the washed samples. For the
Upper Freeport Bed there were useable samples at 5
locations. These had been washed to various yields
through screen sizes of 3/8" mesh arid 1 1/2" mesh
among others. The mapping involved the sulfur content
that coal could be washed to at various yields using these
two screen sizes. Figure G-l shows the base map for
-------
this series. Four of the five data points fall within
the center portion to be mapped and the fifth is to the
northeast; the spacing of the points is quite good, although
extrapolation occurs in the southwest portion, of the total
outcrop. The spacing is also good for the area over
28" thick, with the exception of the southwest corner.
The total sulfur content and the sulfur content
resulting if all pyritic sulfur were to be removed were
mapped for purposes of comparison. Figures G-2 and G-3
show the total sulfur and total less pyritic sulfur for the
3/8" mesh screen at 60% yield for the total outcrop and
the areas over 28" thick. Figures G-4 and G-5 show
the same type of information for the 1 1/2" mesh screen.
Not surprisingly, the greatest variability occurs in the
center area where 3 of the 5 data points are located.
For the total less pyritic sulfur all areas are less than
1% sulfur for either the 3/8" or 1 1/2" mesh; the north-
eastern and central areas are less than 1% in both cases
for total sulfur.
Figures G-6 and G-7 show the differences in sulfur
content between washing with a 3/8" and a 1 1/2" mesh
screen for the same yield of 60%. The difference in the
total sulfur between the 3/8" and 1 1/2" mesh screens is
small except in the far northeastern corner of the study
area. In all cases the sulfur content is greater with the 11/2"
mesh screen. The patterns reflect the location and number
of data points and are particularly evident in the very sharp
contour intervals. For the total less the pyritic sulfur,
the difference is larger than for the total sulfur data alone.
-------
1 i E H HF"
•:;:::::::.::-;::::::s:: •:::•::::: .... s
.:: ::" '"svHtrLTi'V:" «UEO,«, s* "
«»E.. C..NTY ::H :::H::Y •••". """"* ."'
Ulintlt "I"!:. I "V "e
::::::::::::K::::::::f "'.J
I \
::::::::;::!::::::::::: si s :s
..::::::::;::::•:::::::::::•„•
;H:::::::::::H::::::..»"''
;::::::::•::::::::::::::: !
*"t t ; ii++* t *t*M
:::•: : :JIJ;!" s s* '
>*•••!•<• <• eg
I::::::.1 «.ie, ,„„,«,
•"' 0 !
.. ,_,._.....,._„_.._..-. --^^ ' 1
, . . , L . . » I^IIIHiH «s' i
8 •:::::::: ,»",^ ;
8e%«..r".s %;;* «^r |.
.• *...« e™-"
~ s> --«, :•
• «.«,•
e ee
S"
H:!- -Af
:::::!. ::: ::":' , .;:! « « * ' • > • « > « ' • i
"" :-l -us,8" 1
.""•* ^
$ « I i C F » 0
eee 0 OOM CQUNTY DUTllNE
| e ee»e STSIF OUTLINE
— Mt-«8
+; g
^?!> „ ,!c'tEi r"1" .
! -
Figure G-l
Center Portion of Allegany and Garrett Counties
Upper Freeport Coal Bed
Sulfur Content of Washed Coal
Base Map of Samples
-------
A
Tr=i.'w ••—
::'."—*• V
;s;
J
StJSlSK
F"«" '
K!~J'
n
L E G E N
0 DCOO CCthlV
6 «««« STAT£
a
OUTLIKE
OUTLINE
8:12 S:JS §:!.' S:88 CSS !:JS i:18 f:J8
IfieuIICK OF CAT* PCI 1,1 H*LUES IN EJCh LCVEL
r^-zqH:'
i^Sa^T^a1
[•"fcX *»"^J ALteswiv J
S CCUHTY •
8:18 l:\l l\ll
SCALEI IK MILES
!:» S:ii 8:SS 8:13
FKEObthi.* 01S1RI6L1ICN LF_LAT* FC1M V*LUiI Ih £*th LEVEL
::::::::: ^= ==== ;TT?H:H ^S ^^
i i
Figure G-2
Center Portion of Allegany and Garrett Coimties
Upper Freeport Coal Bed
Sulfur Content of Washed Coal
At 60% yield
3/8" mesh; for total outcrop
-------
TIM • C-C
•*««• *+••— -'
'BBS S:::?-
"!:B: SHr"
';!!»
,
J.
I
!'. S'V
"
::•;:::; HHns s!.H:::: aiisiB KSMSK iiiiilii: iiii'ii::
o 555SJJ5SSS
"::;::««
SULll I» MIES
n::::::::::: tstss*. SRSKX istissii ins
"" ;::::::•: ilBl::::::::!: iiiKai: iiis
Figure G-3
Center Portion of Allegany and Garrett Counties
Upper Freeport Coal Bed
Sulfur Content of Washed Coal
At 60% yield
3/8" mesh; for area over 28" thick
-------
«:n::::j;;i;;:::.^t
(SMMXMM!«»U?-^" i
ieS«MM«*»++«->-.— «
"iBEKSKSKS^"" 2
.. _jH*S"*5l'KS3K£V8
UliJtJtHH^...-
illsV I
.HHJSsst;']''^^""'5* "V* M™"" '••
«m*tt««4t* S eeww ae
-";;*;*+•+ o e"« «
L
OOODCD
e eeea
E G E (i C
COUNT* OUTLINE
S1*TE OUTLINE
ttUUtt *****»*t KMX&SKMN xiii?*!!! ••••'!»• •••"•••a
—:::: :::^:::: jy;::;;;;;:;;:::: ::s::a!! :::8i:B
Figure G-4
Center Portion of Allegany and Garrett Counties
Upper Freeport Coal Bed
Suliur Content of Washed Coal
At 60% yield
1 1/2" mesh; for total outcrop
-------
1 .. «f.;:pr ,•""•''
j ""•" """" 1, ""*""•„ •'"
\ ^ -"-J
! *.....!
i ...::• i .... II *"*
1 . I" '
i i ;.-„ s,,, „,,„„
i. ...".^«' *
i'"" 2 n • ;
— f.aHSj.'HciEa'tsnafflksBtc.u,
"""•" '•" ''" '•" ''°° ''" "" '•" U8
- Hji!;!!; - -y iliii! = iilili! illli illl 11111011 iiilillll iiiiii
t
* *S» 1 *555*J»« " ***
'.r/""
1 " L i , t , C
«««•
&&CLL1C- iiUJt h<^C-t JlpPLYISG Tu t*Ch LtVcL
I'ttilH.^' 1M.LUCEC IK MCI- 1ST LiVtL CHY 1
WSK s:18 S:^? S:ie 8:S S:SS S:!S S:iS S:!§ 8:1!
- iliii!!!: ^== Illini ^ Ilililii I«l ill IBJISJ! iiiiHii
Figure G-5
Center Portion of Allegany and Garrett Counties
Upper Freeport Coal Bed
Sulfur Content of Washed Coal
At 60% yield
1 1/2" mesh; for area over 28" thick
-------
... .~
.:::::::.._ :::::::::= — f •
r
F 3/d IMtH AHD 1-1/1 INC
zj^K^^f''^ ^r \<
Sflfi"***- _• ^aan «*
'-. !
LFUA (TOTAL OUTCROP)
-*T "2" *" S
IdlOUUTt VALUE HAUGfi APPLYING TO EACH LEVEL
MINI HUH g-g2 0.02 0.0\ O.U7 0.09 0.11 i,.l3 0.16 0.18
TUTAL ABajLUTE^ALUe^ftNGE A^LJISC TO ejC^LEVEL ^^ ^^ ^^^ ^^^ ^^^
V4LUES IN dftCH LEVEL
i sssss :
25 19
0.0 ^ O.OJ 0.04 O.QT
C.02 O.IH 0.07 0.09
.UES IN EACH LEVEL
SB 25
Figure G-6
Center Portion of Allegany and Garrett Counties
Upper Freeport Coal Bed
Sulfur Content of Washed Coal
At 60% yield
Comparison of 3/8" and 1 1/2" mesh; for total outcrop
-------
= 1 1lM\!«v
.t r% •
..
..•i
SCW.E1 IN MILES
I J.L^ 0.0. " X*07 0*0T °"C? "*11 °"T3
Figure G-7
Center Portion of Allegany and Garrett Counties
Upper Freeport Coal Bed
Sulfur Content of Washed Coal
At 60% yield
Comparison oi 3/8" and 1 1/2" mesh; for area over 2b" thick
-------
ssr..:
'»sS:
i _ j.
€ eew suit OUTLIN
K;;::, isik?:
"KiiSilr1* ..J
ftf**
.8
eoo
'v y
1 1
Figure G-8
Center Portion of Allegany and Garrett Counties
Upper Freeport Coal Bed
Sulfur Content of Washed Coal
At 50% yield
3/8" mesh; for area over Z8" thick
-------
..JJillP' y
— (K-"-»>--— O*
«'•.««/'.' V" ;^;m j.
._=:", .„. fjl*%*
' ^
J" 1 """ I
!
^•^•^jf \ " 1
!!« «S
as? ""-" "•"II"
we c z i o u
a «
...::::", .„. I's "r"
:H:: -:.. S3** *••"
8
. S
S™
f
.... .«
leee88
, a
SCALEf Ih flLES
*L AescLi/re V*LUE HAi>Gfc *PM.TING ID E*CH LE
TICN CF.CAT* PClhl
8:11
Figure G-9
Center Portion of Allegheny and Garrett Counties
Upper Freeport Coal Bed
Sulfur Content of Washed Coal
At 70% yield
3/8" mesh; for area over 28" thick
-------
/•
f^jr-.J* xr f'°8
S "'".
8 •".. •.
-S-i:iSHr"" •""
s>..rt-'Ve" -— >
IB!"
• IUU •**" " *••§
SSu»n !in«--" e sSI
%•••••• u****-***- « «*ea «
•:ffl: Kjsrr 8
. «n
SLEi IN PILES
..a,;;;;;;;;; ™s~ H-SH sj|H:j ||K| ]|jj||j]| jjjj|{ji| ij||;jjjj
!!!!!!!" -=.™.= *^^tttX+ »*•*»•*** MMKMMM
sypcLLi ;;;;;";; -• -^ •- ;*°;j^f^ "ilii^Ji I******** |
Figure G-10
Center Portion of Allegheny and Garrett Counties
Upper Freeport Coal Bed
Sulfur Content of Washed Coal
At 80% yield
3/8" mesh; for area over 28" thick
-------
This area of greatest difference is in the center and is
undoubtedly related to the location of the washability data
points. We cannot make any conclusions as to the reduction
in sulfur content due to using a 3/8" versus a 1 1/2"
mesh screen since the data are too scanty for this case;
the graphic technique should, however; prove useful.
The final set of maps in Series G (Figures G-8 thru
G-10) show the same type of sulfur information as Figure
G-3 for the area over 28" thick but with different yields
using a 3/8" mesh screen. For these particular washability
analyses mapped for the Upper Freeport Bed, yields of
50%, 60%, 70%, or 80% do not significantly affect the
sulfur content. This would be intuitively apparent from the
sample plot of sulfur content versus yield shown in Figure
2. For the total less the pyritic sulfur all values are less
than 1. 0% sulfur regardless of yield; non are less than
0. 5% sulfur. For total sulfur in all cases 2 of the 5
values were less than 1. 0% and none were less than 0. 5%.
2. 10. MAP SERIES H
The final map series shows the quantity of coal that
would be expected to be available assuming the washability
mapped in Series G and the bed thickness mapped in Series
F. All of the maps in this series were produced with
the GRID program. As previously discussed, the reliability
of the thickness data is very suspect and the spatial
interpretation of the washability data limited due to the
number of data points; consequently, it is not possible to
draw any conclusions as to the actual quantity of coal that
might be available for this particular study area. In fact,
-------
it would be grossly misleading to attempt to draw any
such conclusions. All of our mapping and analysis efforts
do, however, reinforce the initial hypothesis that the
techniques are a powerful tool for this application as long
as we can provide sufficient data.
Figure H-l shows the quantity of coal washed to
less than 1% sulfur (at 50% recovery, assuming deep
mining so as to give conservative estimates) in tons per
acre for a 60% yield with a 3/8" mesh screen. This data
was mapped for the total outcrop for both total sulfur and total
less pyritic sulfur. The map of total less pyritic sulfur is
directly proportional to the thickness map shown in Figure
F-l since: (1) all of the outcrop is less than 1% sulfur for
total less pyritic, and, (2) quantity is directly proportional
to thickness for those areas where sulfur content is less
than 1%.
Figures H-2 thru H-5 show the same information
as Figure H-l for the areas over 28" thick for 60%,
50%, 70%, and 80% yield, respectively. These four pairs
of maps differ slightly in terms of (1) the quantity at any
character location mapped in direct proportion to the percent
yield, and (2) the cells mapped, depending upon the small
variations in. sulfur content found in Figures G-3, G-8, G-9,
and G-10 respectively. If sufficient reliable data had been
available these subtle differences and distinctions might
have proven significant and could easily have been quantified,
as in Figure H-7.
The final maps in this series attempted to show the
differences in quantity of coal obtained by washing to a
specified yield of 60%. First of all the quantity that could
be recovered with 100% yield was determined and mapped
in Figure H-6. Figure H-6 is merely a map of the central
portion of Figure F-3 at a larger scale; the quantities
-------
at each location are multiplied by 50% to estimate the
quantity of the reserves that would be recovered. It
is interesting to note that the scale change from Figure
F-3 to Figure H-6 reveals a small pocket of coal not
shown in Figure F-3 for the areas over 28" thick. Figure
H-7 shows the additional quantities of coal that would be
recoverable by washing to a 60% yield.
When we examine the difference maps for the
100% yield surface (the quantities determined from the
raw sulfur data points) a1"* the 60% yield surface (determined
from the washability data) we can only make a comparison
for a small center portion of the study area. This is the
only area where the spread of data points from the two
sets of data is in any way compatible. For all other
portions of these maps there are data points for only one
of the two sets of data. In addition the thickness data
are best for this portion of the study area.
Mapping quantity of coal in tons per acre by grid
cell may not be the best procedure: quantity is a variable
that is better aggregated for areas than displayed spatially
as a surface. On the other hand sulfur content and bed
thickness are variables that lend themselves well to spatial
presentation as surfaces. To show useful information for
both sulfur content and quantity at the same time the following
procedure might be appropriate:
1) map sulfur content as done in this study.
2) map bed thickness as done in this study.
3) critical isolines such as 0.5% or 1.0% could
be determined as in this study and a map made
showing these for each of the yields.
-------
4) for each area between isolines, such as the
area greater than 0.5% but less than 1.0%, the
quantity of coal could be calculated based on the
varying bed thickness for those areas.
5) these quantities could be printed out below
the map or as legends on the map in the appropriate
shaded level.
6) the reader can then chose the sulfur content of
interest to him and visually see the extent of the
area involved and the quantities of coal in all those
areas of lesser sulfur content.
7) comparisons can be made between various yieids,
and screen sizes, with quantities calculated and dis-
played in a similar manner.
-------
SCALE I IN H1LCS
:!;:::::".„..«*• ~"Vi"t *•"" ALH.GAN* 19
:™:::::^::::|« •***" * S ™w •
fiF
•*«•*» I
.^u:u»
SKIIIS!*
ii
4
OUANITT OF IQAL AT SOS t.ECO»tftT / bO* TItLD 11/6
: S T V I R G 1
4 «•*<
STATE OUTLINE
SCU.EI IN HILtS
EEPOHI BED ALLEGAhT *WJ S4RRETT COUNTIES
' 'ppiI!SI0 ^SU"'1
F.ioUSNC. »IST«jgjTj» JF OAIA FOINI ,.LUSS ,» EAIK unL
L=VcLS 0 1 i 3 , 5 f,
10.00 iTt.6.67 3033.33
S.6.67 3033.33 3300.03
LbSS.CO £011.20 l-V«,-t.Si E 17*1.54 2501.90 0.0 O.O
P.UE.T.OE OF TOTAL A.SOLOT.^.LOE^.fG. APPL.lfG TO EAf^LEVEL ll-U ^ ^ u_u UJ1
Figure H-l
Center Portion of Allegheny and Garrett Counties
Upper Freeport Coal Bed
Quantity (at 50% recovery) of Coal Washed to less than i% Sulfur
At 60% yield
3/8" mesh; for total outcrop
-------
.=§1=™:"-:!"" .-'
. f " "'""• J'
„•«,•*""
J
ft« e0eee« STATE OUTLINE
ae
M» SCALEl lit H1LES
T __4- -: I—I
^
V"™°™° * *i JoUNT. — •"
.%!•' "''"'
* •*•*• •••• ***a ee
**•*•• «•*••*• e*eaaeB%
MM
t66 0 0000 COUNTY OUTLINE
0
e e eeee STATE OUTLINE
SO.'' S»LE, LMItH
.»«:«;«£ OF TOT.L ..SCJJTT^.^^.Jt .WJJ.Jj ,0 EACJje.EL ^
^!,rii,.u,«i..jG6 .PPLJJJO ,0 i{c»Ae,tL
t *n*s*»«
*
Figure H-2
Center Portion of Allegheny and Garrett Counties
Upper Freeport Coal Bed
Quantity (at 50% recovery) of Coal Washed to less than 1% Sulfur
At 60% yield
3/8" mesh; for area over 28" thick
-------
TOTAL SULFUR (AREA OVcH 29 INCHEi THICK)
OPPtH FHtcPURT BCD ALLcGANV AM) GARRtTT COUNTIES
QAT« NAPPcO IN 9 L£tfcLS BET«tN cXTrte"£ «?WiS OF 900.00 AND 330O.QO
319.00 1476.00 1204.55 1*06.71 1598.98 0.0 0.0
4BSCCUTE VALUe RAH&E APPLTIW TO EACH LEVEL
--
0
:S! SISI:S
888:8? iSII:
• *
r SO* K.-tOVi.HY / SOI YIELD (3/6 INCH HESH1
UPPER FHEEPORT BLO ALLEGANY AND GAHRbTT COUNTIES
"TS,;:5!'° i!,.?.b'VEH1.8!8EE".8E5!lE "SfeSS' '8V
"E ""°s "SS-S T°nl:!fvefe:33 »88:8S
j r
>a,i!
0 0 192 127 0
. VJLUi
Figure H-3
Center Portion of Allegheny and Garrett Counties
Upper Freeport Coal Bed
Quantity (at 50% recovery) of Coal Washed to less than 1% Sulfur
At 50% yield
3/8" mesh; for area over 28" thick
-------
UPPi-R FRS-POKT
.LCCANV S
»„„ ^.'
> %
". f
E S T V I R G I ft
IALEI IN MILcS
TMtcN ElHUHE VALULS OF 900.00 AW 3300.OO
5S 19*4.93 276S.T7 0.0 o.fi
i VALUE RAhG£ APPLV1M, TJ EACH LEVEL
i- ii.li 11.11 11.11 li.ll 11.11 11.11 11.11 11.li
s; ssaas BBSS!
ftL LtSi PYRIT1C SULFUS IARLA OVdR 28 INCHES THICK)
P-K FRE^PORT BL-C ALLEGANV AM> liARRLTT COUNTIES
!5lK""" """" "IS'sf ™{«S:Sf VEt53S:33 1388:SS SSI:SI IB8:S !5!J:8S i@i:!I !S38:il
Figure H-4
Center Portion of Allegheny and Garrett Counties
Upper Freeport Coal Bed
Quantity (at 50% recovery) of Coal Washed to less than 1% Sulfur
At 70% yield
3/8" mesh; for area over 28" thick
-------
1 . jUt*
,-•' ''"•-... i:
e a 8
MM | 69
^g^sEjB a |e 8"°
.r1'"
eeS LEGEND
| 0 0000 COUNTY OUTLINt
e8 e flflfl* STATE OUTLINt
.„•<"•«»,<
02 + 48
/
%m«,/sF'<*^ ;-— j.
I- ""'"-« :'"
t" "•«.*
.uSii tMS!.BB" "II
ttHKNMHJI MNHMIMIM H0 F^TVIRGIKIA
:;;:!i J;;K!:!" «•«.«,".
9.".'""
0 e BB66 STATE OUTLINE
BMMH e88
MMHKJI 6
SKSSUe0 SCALE i IN HILES
0686 -0 2 T 6 -B
C3/8 INCH MESH)
Doloo TLJL166^07CVCU33.33 1700.00 194.6.67 2233.33 2SOO.OQ 276o.fi7 3033.33
06.o7 143i.3J 1700.00 4966.67 2233.33 2500.00 2766.67 3033.33 3300.00
iiIi.v"LU'."uL A"ii!u ™ "KuEV£L 11.11 11.11 11.11 ii.il 11.11
F DATA PJIIIT VALUES IN ^.CH LEVt
Figure H-5
Center Portion of Allegheny and Garrett Counties
Upper Freeport Coal Bed
Quantity (at 50% recovery) of Coal Washed to less than 1% Sulfur
At 80% yield
3/8" mesh; for area over 28" thick
-------
S
%
."".'
•
-
SCALEl ltt H1L£S
IHiUTT COUNTIES
N OF DATA POINT VALOES IN E
.11 1L.11 11.11 11.11
°ii!l"!rs!,..:.3 1?2S:SS ill!:!! 1S:II !5S!:SS I5!S:!I 1111:1.
.SOLU j.V.U..^.^ .PPJJ1NC T, E.C^LE.EL ll-u u-u ^^
K.ALOi
Figure H-b
Center Portion of Allegheny and Garrett Counties
Upper Freeport Coal Bed
Quantity (at 50% recovery) of Coal with less than 1% Sulfur
(100% yield, based on Figure F-3)
-------
•
„. s
-. I
efle
•IS!
SCALEi IN MILES
-i rt^PPLO IN 5 Li.Vi.Lj Bf.TKL-U cXTScME VALUES OF 966.43 flNQ 13^0.80
iti.CiO i J,O.QO '-i.o.-i-- i: jT.jt 1616.20 0.0 O.O
iiffl! "L!!tS:S' "SI-s '"ilS-sr'iKs-Jf tiS:Si ffiKSt i&
-------
3. CONCLUSIONS AND RECOMMENDATIONS
3.1. STUDY AREA AND DATA CONSIDERATIONS
As expected, the greatest difficulty encountered in the
project was the obtaining of adequate data to work with.
Sufficient data to produce meaningful results never were
available for the detailed study area of Phase III, although
this did not significantly detract from the demonstrations of
the applicable techniques.
Examination of the graphics produced in Phases I
and II has shown that there is a definite need for: (1) a more
equal number of analyses at each sampling location and (2)
a more evenly distributed pattern of sampling locations. The
latter problem is especially acute in Phase II while the former
problem pertains to both phases.
In Phase III it was difficult to draw any statistically
sound conclusions. The reliability of the thickness data is
very suspect and the spatial interpretation of the washability
data limited due to the number of data points; consequently, it is
not possible to determine with any accuracy the actual quantity
of coal that might be available for this particular study area.
Furthermore, not enough information was known about
such parameters as the vertical location of samples in a bed
to ensure representative data for mapping. Until sufficient
data are available to standardize the samples and remove the
effects of parameters other than spatial location, substantive
results cannot be obtained, except on a larger-scale basis
such as Phases I and II.
-------
The study areas for which the computer mapping
techniques were demonstrated were selected after reviewing
the available analytical, washability and geological data for
several possible areas. The criteria for selecting an area
for detailed study included: (1) a "reasonable" and "manageable"
number of seams, and (2) "adequate" availability of data for
each seam for sulfur content, seam thickness and extent, and
washability. Subject to these criteria we wished to pick as
extensive an area as practicable so as to be able to develop
and test large-scale mapping techniques useful to APCO.
The criterion of a "reasonable" and "manageable" number
of seams was merely operational: no additional techniques could
be demonstrated by working with 10 or 100 seams than with
2 to 5. We were interested in summing quantity of coal for
2 or more seams and were able to demonstrate the technique
with 2 seams for Allegany and Garrett Counties in Phase III,
Stage One.
In practice any number of seams could be analyzed,
mapped, and summed to form a composite. The constraints
are: (i) preparing base maps and data banks for each seam,
(ii) having data points 'adequately' spaced for each seam,
(iii) having a separate computer storage file for each seam,
and (iv) having the additional computer budget to prepare maps
for each seam. None of these constraints are significant.
The criterion of "adequate" availability of data for each
seam for sulfur content, seam thickness and extent, and wash-
ability is harder to define. There is no one magic number--
or even range of numbers --that can define the adequacy of
data points.
-------
The number of datk points necessary for accurate
mapping is a function of:
(i) the scale of the map
(ii) the level of detail desired at that scale
(iii) the spacing and clustering of data points, and
(iv) the variability in the surface for the particular
variable or phenomenon.
As a minimum, it is difficult to imagine a surface that
would be well represented at any scale with fewer than 10 data
points-!' and these 10 should be
a) fairly uniformly spaced
b) non-clustered
c) near the periphery as well as the center of the
study area.
If data points are clustered or missing for portions of the
study area, additional points should be incorporated to determine
a minimum.
The two criteria that are the most changeable are the
relationship of the level of detail desired to the scale of the
map^ and the variability in the surface. If we have a map
10" by 10" and we desire resolution to the nearest inch, then
a data point spacing of about 2" should be adequate. The scale
of the map is immaterial as long as this ratio of 1" to 10"
(or 1:3.0) is maintained. (Compare Figures B-13 and D-3.)
for a procedure to determine the adequacy of data points, see the section
on Map Accuracy and Sampling Procedures in Volume Two: General
Documentation.
-------
The more variable the surface is for a phenomenon,
the greater the number of data points that will be necessary
to explain that surface, all else being constant. This can
only be determined by statistical measures and sampling for
each phenomenon, as discussed in Volume Two: General
Documentation.
If we examine our base maps again, the following general
conclusions can be made about number of data points and spacing:
1. Phase I (Figures A-l and A-5) -we cannot infer
accuracy below the county level and have large areas of
the study area without data points for sulfur content.
2. Phase II (Figures B-l, B-8, B-15, and C-l)--
we cannot infer accuracy below the town level, and--
relative to the scale and detail--have even larger portions
of the study area without data points; Map series C,
however, has equal accuracy for all areas mapped
because of the use of a maximum search radius about
each data point.
3. Phase III - sulfur content (Figures E-l and E-5)--
the data points are not evenly spaced and are clustered;
there are virtually no portions of the study areas with
adequate data points.
4. Phase III - bed thickness (Figures F-l and F-2)--the
same is true as with the sulfur content data points. Note
also that the inadequacy of data points--both the number and
the spacing--is compounded when sulfur content and thickness
are compared to determine quantity.
5. Phase III - washability (Figure G-l)--there are
only five data points; it is not possible to develop a
meaningful surface from so few points.
-------
Although the particular study areas may not be re-
presentative of the country as a whole, they can adequately
demonstrate the scales of analysis useful to APCO. The larger
the scale — given the same ratio of detail to scale — the less
constraining are the data point requirements, as shown by the
Phase I and Phase II maps. But at the scale of one or more
counties (Phase III) it is necessary to have good data point
coverage by mine location.
As with any contour map, values are exact only at
control points. As one moves farther from control points values
become less trustworthy. A region with many control points is
more likely to have accurate contours than is a region with
few control points, all other things being equal. Values at the
margin of the map, which are outside the control points are
of dubious accuracy. The SYMAP program places all control
points on the map so that the reader can have some idea of
the likelihood of accuracy.
A final note of caution should be raised concerning the
variation in the number of analyses averaged at each data point.
Perhaps the best safeguard against making any false inferences
is to look at a map in one hand and at the base map which shows
the data points and the number of analyses in the other hand.
In this way, one is made aware of the number of analyses
attributed to a data point and the validity one can infer about
any given area of the map.
-------
3.2. MAPPING CONSIDERATIONS
There are several important mapping considerations
that have come out of this study, separate from the study area
and data considerations. These include:
(i) photographic reproduction,
(ii) graphic symbolism,
(iii) legends for number of analyses, and
(iv) use of composite maps.
Photographic reduction of the maps has certain
difficulties. In particular the legends may become illegible
and the textual and statistical material printed below the map
may lose much of its usefulness. Moreover; the tone dis-
cernability of the symbolism is somewhat reduced in almost
direct proportion to the amount the maps were photographically
reduced. But, it is still felt that a relatively good quality
can be attributed to the symbolism.
Some of the problems can be overcome by photographing
the legends and explanatory information at another scale.
In fact, this was done with the titles shown at the bottom of
each page; these were separately photographed and stripped into
the negatives before printing. Overlays for certain of the
legends can also be used.
The best solution is to experiment at the beginning of
a project to determine the optimum combination of study area
size, reduction scale, and legend and explanatory information
detail to use in the final presentation or publication of the
-------
3.2. MAPPING CONSIDERATIONS
There are several important mapping considerations
that have come out of this study, separate from the study area
and data considerations. These include:
(i) photographic reproduction,
(ii) graphic symbolism,
(iii) legends for number of analyses, and
(iv) use of composite maps.
Photographic reduction of the maps has certain
difficulties. In particular the legends may become illegible
and the textual and statistical material printed below the map
may lose much of its usefulness. Moreover, the tone dis-
cernability of the symbolism is somewhat reduced in almost
direct proportion to the amount the maps were photographically
reduced. But, it is still felt that a relatively good quality
can be attributed to the symbolism.
Some of the problems can be overcome by photographing
the legends and explanatory information at another scale.
In fact, this was done with the titles shown at the bottom of
each page; these were separately photographed and stripped into
the negatives before printing. Overlays for certain of the
legends can also be used.
The best solution is to experiment at the beginning of
a project to determine the optimum combination of study area
size, reduction scale, and legend and explanatory information
detail to use in the final presentation or publication of the
-------
maps. For this project the decisions to map Phase I and
Phase II at the scale chosen and to show all legends at the
same scale (without overlays) were ultimately responsible for
the difficulties in reading some of the maps.
With the exception of class levels, or value range
intervals, the map symbolism is the only and, therefore the
most important graphic way to portray the meaning of the
data. The value of a map may be easily lost or destroyed
by a poor choice of symbolism.
In SYMAP this choice of symbolism is usually made
by the program; there is a prestored set of symbolism for
up to 12 levels. For this project it was felt that this set of
symbolism was inadequate because (1) there was not enough
differentiation between some of the levels, and (2) the darkness
of adjacent levels seemed to be reversed in some cases.
Therefore, a unique set of symbolism was devised. Whenever
possible this set of symbolism and value range intervals was
used.
The numbers of analyses associated with the data points
were shown on the base maps as legends to aid the reader.
The program which produced these legends can easily be
generalized to produce other legends such as mine codes, town
names, year samples taken, etc. to be printed on the base
maps at or adjacent to the data points. If large scale maps
are used or there are few data points, these legends can be
used on all maps in a series; because they often obscure the
symbolism they are generally used only with base maps.
In Phase III a composite map was made showing the
quantity of coal from two different coal seams. This is really
-------
only a mapping exercise: because of the difference in elevation
separating the two seams, the composite maps produced are
of very little value. They assume that the two beds are
fairly close together for mining purposes, which, in fact they
are not. They also assume that the two beds are similar
enough in geologic attributes to be shown together. In producing
such composite maps caution must be taken to use surfaces that
can meaningfully be combined.
The mapping characteristics of the SYMAP and GRID
programs do seem well suited to the displays shown. Overlays
of certain legends would probably improve the readability, as
would a different scale of photographic reduction. The SYMVU
program is useful in conjunction with SYMAP and mainly to
visualize the variability in the surfaces.
-------
3.3. APPLICATIONS
The most important application is the mapping of sulfur
content and quantity on the same map. Mapping quantity of
coal in tons per acre by grid cell may not be the best procedure:
quantity is a variable that is better aggregated for areas than
displayed spatially as a surface. On the other hand sulfur.
content and bed thickness are variables that lend themselves
well to spatial presentation as surfaces. To show useful infor-
mation for both sulfur content and quantity at the same time
the following procedure might be appropriate:
1) map sulfur content as done in this study
2) map bed thickness as done in this study.
3) critical isolines such as 0.5% or 1.0% could
be determined as in this study and a map made showing
these for each of the yields.
4) for each area between isolines, such as the area
greater than 0.5% but less than 1.0%, the quantity of
coal could be calculated based on the varying bed thickness
for those areas.
5) these quantities could be printed out below the
map or as legends on the map in the appropriate shaded
level.
6) the reader can then choose the sulfur content of
interest to him and visually see the extent of the area
involved and the quantities of coal in all those areas of
lesser sulfur content.
7) comparisons can be made between various yields,
and screen sizes, with quantities calculated and dis-
played in a similar manner.
-------
In general for analytical or working purposes, a series
of maps could be made with sulfur content or quantity varying
for each map and the other variable varying through the series
of maps. The recommended procedure would be to produce
maps in the series, varying by seam thickness or overburden
in intervals. For each interval of seam thickness or over-
burden sulfur content would be mapped and a number or bar
would be printed within each isoline of sulfur level showing
the quantity of coal contained therein.
Once the base map is created maps of any number of
possibilities and variables may be made inexpensively. The
most suggestive may be used. In addition, historical maps of
production and reserves may be made, as data over time are
accumulated. Some possible maps are:
1) Isopach maps (thickness) by bed and for the
total amount of coal underground;
2) Depth of overburden to top of bed(structure contour
map)'} among other things, this will tell us whether
strip mining is feasible;
3) Total volume of coal in the ground per acre;
4) Amount of coal recoverable by present methods
before treatment per acre;
5) Yield of recoverable coal remaining after treatment
per acre for each top size and each bed;
6) Percent sulfur in coal after treatment, (by top
size)j
7) Net worth of coal in ground by bed after mining and
treatment, before shipping--for each size grade and
for all size grades;
-------
8) Map of coal sheds about major regions, dependent on
transportation costs, and price for each coal grade;
and,
9) Dynamic nature of coal sheds as we feed in
changes in technology of coal beneficiation, changes
in standards of pollution control at the stack,
subsidized changes in freight rates, changes in
subsidy for expensive technology, and changes
in demand if oil or gas is substituted for coal.
Many of the variables mapped can be compared to sulfur
content to determine quantity of coal less than a certain percent
under varying assumptions. For instance, depth of overburden
can be checked at each character cell to be mapped to see if
mining is feasible. This can be done using the same procedures
as were used to determine quantity for mapping.
I± would appear that the best applications can be made
at the county and town level as in Phases I and II. It is
doubtful that sufficient data can be made available at the mine
level in the near future to satisfy the study area and data
accuracy criteria.
-------
HI. APPENDICES
1. Phases I and II: Appalachian Coai Region
2. Phase III: Allegany and Garrett Counties
3. General References
4. Program Development
-------
1. PHASES I AND II: APPALACHIAN COAL REGION
Phases I and II of the project dealt with a four state
area herein called the Appalachian Coal Region. It covered
an area approximated by counties containing coal in the
Pittsburgh Bed, according to the Bureau of Mines data bank.
Phase I was concerned with data on sulfur content for raw and
washed samples--aggregated by counties--for the Pittsburgh Bed.
Phase II was concerned with data on sulfur content for
raw and washed samples--aggregated by towns--for the
Pittsburgh Bed and raw samples for the Middle Kittanning Bed.
The following sections describe the work that was
done to prepare the data for mapping, including base map
preparation, data bank set up and use, and decisions as to the
actual running of the maps.
1.1. INPUT TO SYMAP
SYMAP, like any other computer program, is simply
a framework in which a user can work. The program provides
a variety of options and the programmer merely chooses which
ones he wishes to employ in order to create the map.
To produce computer maps, input to the computer in
the form of a deck of punched cards must be prepared. This
deck will consist of certain introductory cards (job control
language and a subroutine) and a number of "packages" each
composed of additional cards covering a specific category of
information about the map(s) to be produced.
The number of packages used depends, exclusively, on
the programmer. In this project a total of 6 packages were
set up for SYMAP input:
-------
1. A-OUTLINE--describes the outline of the
entire Pittsburgh Coal Bed by specifying the x-y
coordinate locations of each significant directional
change along that polygon curve. The shape of the
bed is thus defined in the program by connecting each
vertex to the next to form a series of straight line
segments.
2. B-DATA POINTS--describes the point locations
to. which your data is to be related, by specifying
their x-y coordinate locations. These data points may
be either the points for which data is available (towns
in Phase II), or the centers of areas (counties and/or
bed outcrops in counties in Phase I), for which data
is available. A printed listing of the data points is
found in Table III-l and III-3 for Phase I and Tables
III-5 and HI-7 for Phase II.
3. C -LEGENDS - -causes certain supplementary in-
formation, namely legends, to appear on the face of
an output map (e. g. , for Phases I and II this meant the
graphic scale, legend box, and all the state names) by
specifying their coordinate locations and content.
4. C-OTOLEGENDS--describes line legends such as
rivers, bridges, roads, and county or state lines which
must be adjusted if the size (scale) of the map is altered.
Appropriate deletions or additions to these line seg-
ments are made in accordance with change in scale.
This package adjusts with scale identically to the
A-OUTLINE package and with only minor alterations
in card order (and not in card format) one package can
-------
easily be converted to the other. In both Phases I
and II, the C-OTOLEGENDS package contained the
significant state and outside county boundaries of
the four-state region.
5. E-VALUES --contains the values or quantitative
information applicable to each data point; one value for
each data point. It is wisest to label each card for
future reference. The order of the cards is not im-
portant--any convenient way of arranging them is
possible as long as these two packages are in the same
order. In Phase I the counties were ordered alpha-
betically by state and in Phase II alphabetically by
state, county, and town. This is the same order in
which the data was received. A printed listing of
tke values is found in Table III-2 and III-4 for Phase
I and Tables III-6 and III-8 for Phase II.
6. F-MAP--is the map instruction and execution
package. The user specifies an appropriate title to
appear below the map as well as electives concerning
the size or scale, number of levels, level breaks, map
symbolism, minimum and maximum values to be regarded;
search radius, etc. A sample F-MAP package is in-
cluded in Table 111-12.
The above brief package descriptions were only intended
to present a general discussion of the workings of SYMAP. For
a more detailed explanation of the program and its operations,
see Volume Two: General Documentation.
-------
1.2. BASE MAP PREPARATION
The first step in producing a computer map is for
the user to obtain a source map of the area under study.
The map must be prepared in such a form that it is digestable
to the computer. It therefore becomes necessary to code a
base map that has all the information on it that you wish to
show.
For this beginning stage in the project a base map at a. scale
of 1:250,000 was prepared from the Army Map Service series of the
same scale. This particular series was chosen because of the scale
clarity and its availability--the Laboratory has the complete set for
the United States. By using this scale, the detail is preserved in
reduction and it becomes possible to select portions of the map to
blow up for close scrutiny (see maps D-3 thru D-7). These maps
also have Universal Transverse Mercator grids (UTM s) marked on
the map. This becomes an important factor for Phases II and III
since all of the data is located by UTM coordinates.
The entire study area conveniently (and deliberately)
rests within one UTM grid zone--zone 17. It was decided
that work would be greatly facilitated if the study were to
restrict itself to only one UTM zone. This was done because
of problems inherent in the projection which make it somewhat
difficult to align adjacent UTM zones. Some of the counties were
not in zone 17 and were deleted and others were not in the
chosen study area. The area was chosen so as to be within
the Appalachian Coal Region and contain all of the Pittsburgh
bed.
-------
The base map was prepared by tracing 84 conterminous
county outlines, in sections, onto the working base map of
69. 7" by 77. 7" (x-y). Points representing raw and washed
data for the Pittsburgh Bed were placed in their respective
counties. The coordinates for the outer edge (or outline of
the 4 state and county area), the data points, and state lines
were all punched in inches and generated by the use of an
electronic digitizer.
The use of this machine is imperative in facilitating
the work of locating coordinates and eliminates the chance
for human error. The digitizer was employed in this project
because of its ability to quickly and accurately produce a great
number of coordinates in card form. The procedure involved
in using this machine is quite simple. The user tapes the
base map to the table in such a way that the x and y axes on
the map are true to the horizontal and vertical, respectively.
The origin is then set by the user to 0, 0 in the top left hand
corner of the map. The machine is instructed to measure the
coordinates in inches by an interchangable circuit board within
the machine. The coordinates are then taken from the map by
lining up the cross-hairs on an adjustable arm over the point
that one wishes to digitize. By then pressing a button on the
arm, the coordinate is taken and transferred to a keypunch which
is interfaced to the machine. The keypunch then punches the
point in card form giving its x-y coordinates in inches to two
decimal places, left-justified to columns 11 and 21 and also
punches a unique reference number in columns 73 and 80.
The numbers were assigned to these column fields by the use
of a drum card in the key punch machine. The resultant card
output is in a format compatible to SYMAP input with only
minor alterations in terms of the regrouping of cards.
-------
For smaller areas with fewer vertices, this pro-
cedure is done by hand with a ruler. The SYMAP program
will not accept any outline that has an excess of a hundred
vertices in the A-OUTLINE package. Since more than 500
cards were punched in the outline of the conterminous study
area, it was necessary to break up the cards into "islands"
or sections of 100 vertices or less. In so doing, the appear-
ance of the map is not affected whatsoever. After this pro-
cedure was followed, a number of maps were run of the low
sulfur Pittsburgh data using different level breaks and symbol-
ism shadings at a scale of 1:633,600 or 1 inch = 10 miles.
These maps were run with the study area county outlines as
an A-OUTLINE; the remaining work in Phases I and II was
done with the bed outcrop as the A-OUT LINE at the same
scale.
A 1938 base map at a scale of 1:506,880 or 1 inch = 8
miles, of the Pittsburgh Coal Bed was sent to us by the Bureau
of Mines. It was almost exactly half the scale of the base
map with which we had been working. However, because of
incompatibility of the map projections between our base map at
a scale of 1:250, 000 and the 1:506, 880 map it did not seem feasible at the
time to enlarge the map by a factor of two. Therefore the
map of the Pittsburgh Coal Bed was enlarged by eye by the
similar grid method to fit the 1:250,000 base map. In retro-
spect, it would have been a lot less time consuming to have
photo-enlarged the map anyway, since the error factor would
have been so slight in the photo-enlargement and less than
enlarging the map by eye. After the outline of the Pittsburgh
Bed had been delineated on the base map, it was then digitized
and made into 71 different "islands". Each island represents
one outcrop or part of an outcrop in the case of the major
-------
seams of the Pittsburgh Bed. The time involved in digitizing
the base map was three hours. Because of the sensitive and
delicate nature of the digitizer, it tends to be prone to errors.
These errors, however, are very easily discovered upon
running the first base map. From the nearly 2000 cards
created, about 200 (or 10%) had validity checks because of
a defunct transistor in the machine. This kind of error pro-
duces a multiple punch on the card which is unreadable to
the computer and therefore unacceptable for processing. This
mistake was easily corrected in a matter of a few hours.
The old outline of the study area (the county outline)
was then easily converted into the C-OTOLEGENDS package.
The state boundaries were also coded as C-OTOLEGENDS in
a similar manner. Legends were also made for state names,
the graphic scale, and legend box which were input by row
and column coordinates taken from the output map. These were
all included in the C-LEGENDS package. The C-OTOLEGENDS
package was used for all maps in Phase I and II because these
legends can adjust in scale. The C-LEGENDS package was
changed for the D-series maps.
1.3. PHASE I DATA BANK
A small data bank was sent to us by the Bureau of
Mines. The deck consisted of 66 cards for 59 counties in
the four state region and is broken down as follows, for the
Pittsburgh Bed counties mapped:
# Counties Raw Data Washed Data
Maryland 2 0
Ohio 8 3
Pennsylvania 7 5
West Virginia 16
33 19
-------
From the original data bank, the values for the
Middle Kittanning Bed were deleted. From the remaining
values--for the Pittsburgh Bed--six counties were found to
be outside the study area and were deleted: Columbiana,
Mahoning, and Stark, Ohio, and Center, Clinton, and Tioga,
Pennsylvania. This left the above 33 counties used for
mapping. The card format was slightly altered from the
original to facilitate the identification of the fields containing
values:
Number of Analyses Columns 1-5
Low Sulfur Value Columns 6-10
Mode Value Columns 11-15
High Value Columns 16-20
Bed Code Columns 50-52
Raw or Washed Column 53
Point Sequence Number Columns 55-56
Three-Digit State Code Columns 61-63
Three-Digit County Code Columns 64-66
Two-Digit State Name Columns 69-70
County Name Columns 73-80
As was noted in an earlier section, the values and
the data points are in a one to one correspondence to each
other. Hence, the first value will be associated with the
first data point. After all of.the values are read into the com-
puter, they are associated with their respective data points
and contouring takes place, based on the SYMAP contouring
algorithm.
1.4. PHASE I DATA POINT PLACEMENT
There are 33 raw and 19 washed data points in the
study area. One data point is assigned to represent an entire
-------
xO
:3 ~r • « 7*
."3 a . * o
4 3 » SO
3 1 .OS
53.93
31.15
27.09
26.38
46.10
40 .04
25.74
32.40
34.01
20.57
32.66
28.98
27.09
45.89
53.44
27.40
50.33
42.90
47.11
39.60
34. 13
49.67
42.15
37.21
29.61
42.20
56.27
43.90
47.69
f5 f=S . A f-t
es A • ^ s
1 A . *SA
^ 7. Bh
1 O . 3 9
23.7 7
27.99
31.04
1 3.80
14 .99
39.93
45.41
37.69
49.61
54.38
37.31
46.41
39.39
29.95
33.66
30.49
35.46
33. 13
36.45
32.36
13.73
53.00
38.08
33.27
43.04
1 6.36
39.67
38.49
fj I T K
t- I T «
H I TK
H I T*
HI TW
HITR
HITR
HITK
HITR
HITK
HITR
HITK
HITR
HITK
HITK
HITR
HITK
PITK
HITR
HITR
PITR
HITR
HITR
HITR
HITR
HITR
HITR
HITR
HITR
HITR
HITR
HITR
o '<-.
o r-(
OA
OS
Oft
O/
u«
09
1 0
1 1
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
i ** < > < i ,- •<
-s ^ (MM**
3 'I O 1 ) \ 1
•i/* r> ns -i
3 4 O O S ^
3 4 0 0 e> /
34 OOM 1
3401 OS
34 0 1 IS
372003
37205 1
37205V
372063
372 1 1 1
372125
372129
47000 1
470O07
470009
470021
470033
47004 1
470049
470051
470053
470057
470061
470069
470077
470079
470091
470097
M 1 -.
( ) II
Oil
nn
I1H
nn
(HI
OH
OH
HA
HA
HA
HA
HA
HA
HA
WV
WV
WV
WV
WV
WV
WV
WV
WV
WV
WV
WV
WV
WV
WV
WV
O A 1-: ^ 1^£ "T T
A THH; IMS
RF?l_ MDN T
GALL I A
GUEKNSFY
HARR I SON
JEFFERSO
MF I CiS
MORGAN
ALLFGHEN
FAYETTE
GREENE
INDIANA
SOMERSET
WASHINGT
WESTMORE
RARROUR
RRAXTON
RROOKE
GILMER
HARRISON
LEWIS
MAR ION
MARSHALL
MASON
MINERAL
MONGAL I A
OHIO
HRESTON
HUTNAM
TAYLOR
UHSHUR
Key:
y and x locations
Table III-l
-------
I CSO .
1 .
1 1 .
322.
7.
7.
318.
368.
1 1 .
2.
431 .
551 .
61 .
109.
69.
525.
552.
361 .
13.
44 .
106.
2448.
89.
404.
19.
70.
1 .
1659.
29.
5.
43.
257.
12.
O . T
0 . H
3. ft
2.0
1 .9
4 . 0
1 .9
0.9
3.0
4.5
0.7
0.5
1 .3
1 . 1
0.7
0.8
0.6
1 .7
2.4
1 .2
1 .7
1 .4
2.7
0.8
4.2
2.1
0.7
1 .5
2.8
O.9
2.3
1 . 1
2.3
0 . <-)
0.8
3. ft
3.8
1 .9
4.8
3. 1
2.9
3.6
4.5
1 .6
1 .2
2.8
1 .2
1 .8
2.0
1 . 1
3.3
2.5
2.5
2.3
3.1
3.3
1 .6
4.6
2.8
0.7
2.9
3.7
1 .1
3.0
2.1
3. 1
i . f->
o . H
5.0
5.9
4.8
5.7
5.5
4.0
5.3
4.5
4.0
3.0
3.5
2.4
2.3
5.0
3.5
5.9
3.2
5.5
3.4
5.3
4.3
4.0
5.6
4.1
0.7
4.4
5.0
1 .8
3.4
4.4
5.8
1— I T K
H I TK
H ITK
PITR
PITR
HI TR
HITR
PITR
PITR
PITR
PITR
PITR
PITR
PITR
PITR
PITH
PITR
PITR
PITR
PITR
PITR
PITR
PITR
PITR
PITR
PITR
PITR
PITR
PITR
PITR
PITR
PITR
PITR
0 1
l)^
O3
04
OS
Ot>
()/
Ob
09
10
1 1
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
1 ^* O O O 1
1 ^-> O O ^ .-(
1,4 OOOv»
3 1\ O O 1 3
1 1\. 0 O S 3
34 (105 V
340 Of> 7
34 OOH 1
3401 05
3401 15
372003
372051
372059
372063
372 1 1 1
372125
372129
47000 1
470007
470009
470021
470033
470041
470049
470051
470053
470057
470061
470069
470077
470079
47009 1
470097
M 1 >
M 1 1
Mil
(in
OH
OH
OH
OH
OH
OH
PA
PA
PA
PA
PA
PA
PA
WV
wv
WV
wv
wv
wv
wv
wv
wv
wv
wv
wv
wv
wv
wv
wv
Al_l_c- Ci /\ IVJY
UAKRE T T
ATHENS
H (^ I . MO N' T
GALL I A
GUEKNSFY
HARk I SON
JEFFEWSO
ME I G S
MORGAN
ALLEGHEN
F AYETTE
GREENE
IND I ANA
SOMERSET
W ASHI NGT
W-E-STMORE
BARROUR
RRAXTON
BROOKE
GI LMER
HARR I SON
LEW I S
MARION
MARSHALL
MASON
MINERAL
MONGAL I A
OHIO
PRESTON
PUTNAM
TAYLOR
UPSHUR
Key:
number . low . mode . high
Table III-2
-------
Ul
I
31 .0-5
27. O
-------
* OS .
1 1 7.
65 .
353.
4.
3.
100.
71 .
245.
48.
28.
50.
619.
1 .
563.
2.
29.
709.
26.
2.4
1 . 7
1 .3
1 .3
1 . 1
1 .5
0.9
1 .0
1 .8
1 .0
1 .6
1 .0
0.9
3.0
0.7
4.5
2.7
1 .7
2.7
r-i . «
?> .5
2.7
1 .8
1 .4
1 .5
2.8
1 .0
2.5
1 .3
1 .9
1 . 1
3. 1
3.0
2.5
4.5
3.3
2.4
3.6
•-> . o
."-i . -5
3.7
2.6
2.6
1 .7
4.4
2.7
3.7
2.8
2.7
1 .8
5. 1
3.0
3.5
4.7
3.4
3.6
4.5
^ i i '.V
H 1 1 >fl
H 11 W
HITW
HITW
K 1 TW
I-1 I T W
H 1 TW
H I T W
H I TW
HI TW
H I T '.V
PITW
HITW
MITW
HI TW
HITW
HITW
HITW
f> i
o ,'
() •!
04
OS
O6
0 f
OM
Ol-»
1 0
1 1
12
13
14
15
1 6
17
18
19
*'i O O 1 7»
1 ^ o n t> 7
3 4 O O rt 1
^72003
^ / 2 0 5 1
372111
3/2125
372 129
4 ^0001
470007
4/0009
47002 1
470033
47004 1
470049
47005 1
470053
470061
470069
n M
OH
OH
PA
HA
HA
HA
HA
WV
WV
WV
WV
WV
WV
WV
WV
WV
WV
WV
^f= l_MO IN1T
HARRIS (DM
JEF^ERSO
ALLRGHEN
F AYFTTE
SOMFRSET
W ASH I NGT
W F S T MO H E
RARHQUK
RRAXTON
BROOKE
GILMEH
HARRI SON
LFW IS
MAR ION
MARSHALL
MASON
MONGAL I A
OHIO
Key:
number . low - mode . high
Table III-4
-------
county. As in the case of the outline of the Pittsburgh Bed,
the coordinates for the data points were digitized in inches.
The rationale for the placement was two-fold:
1. If a map from the Report of Investigation
series was available, the point was placed at the
center of gravity of the reserves in seams of 28"
or more in thickness;
2. If no map existed, then the point was deter-
mined by using the center of gravity of the existing
Pittsburgh seam for that county based on the 1938
map of 1:506,880 scale.
All vertical coordinates were punched as a decimal
number in Columns 11-15 and all horizontal coordinates were
punched as a decimal number in Columns 21-25. Columns
50-80 are in the same format as the E-VALUES package-
A listing of the data points appears in Table III-l (raw) and
III-3 (washed) and of the values in Table III-2(raw) and III-4
(washed).
1.5. PHASE II DATA BANK
In every respect, except for the number of data points
and value's, many of the maps created in this phase have the
identical SYMAP input format as in the first phase. Some of
the maps are blow-ups of certain areas of the region which
may merit further study (maps D-3 thru D-7).
An additional coal bed was mapped in Phase II-the
Middle Kittanning. There were no base maps available for
this bed so contouring was only done within a 5 and 10 mile
-------
radius of each town (see maps C-2 thru C-7). It was felt
that contouring beyond these limits was perhaps unjustified
and therefore misleading. Maps of the Pittsburgh Bed were
created in the same way for the purpose of: (1) comparison
with the Middle Kittanning Bed and (2) obtaining uniform accuracy
over the areas mapped. These maps are in series B-16 thru
B-21.
The data as received from the Bureau of Mines were
arranged by bed (Pittsburgh followed by Middle Kittanning),
alphabetized by state, county, and town, and sequenced by mine
number within the towns. This same order was preserved
and used in the data bank that was created.
A number of steps were necessary in order to arrive
at the appropriate set of values for the Phase II data bank.
The first step concerned removing those towns from
the data listing which were not in the 84 county study area.
Only towns in the following counties of the Middle Kittanning
Bed were deleted: Columbiana, Mahoning, and Stark, Ohio;
and Center, Clarion, Clinton, and Tioga, Pennsylvania. In
a number of cases, the same town appeared more than once
for the same bed. When this occured, the 'town1 which had
the fewer mines was transfered to the 'town1 which had the
greater number. These towns for the Pittsburgh Bed were:
Hopedale, Ohio, Clarksville, Pennsylvania, Point Marion,
Pennsylvania, Burnsville, West Virginia, Worthington, West
Virginia; and for the Middle Kittanning bed were: Crooksville,
Ohio, New Straitsville, Ohio, Osceola Mills, Pennsylvania,
and Philipsburg, Pennsylvania. This procedure was necessary
because SYMAP can only associate one value per data point
location.
-------
The majority of the towns on the data listing had
more than one mine (value) associated with them. As the
second step, it was therefore necessary to arrive at one
value per town by some sort of a mathematical process. The
following procedure was used after advisement by the Bureau
of Mines:
Value Procedure Adopted
Number of Analyses Mathematically totalled by town
Low Sulfur Value The lowest sample taken was
used for the low value.
Mode Sulfur Value The most frequent or the mode
value was used
High Sulfur Value The highest of the high values
was used.
All addition for the Number of Analyses was checked and re-
checked on an adding machine. The low, mode, and high
values were verified by eye three separate times for accuracy.
After the non-pertinent towns were deleted and the
mine values were totalled by town, the third step involved
data analysis. The data was run thru a small statistical
computer program called SPSS (Statistical Package for the
Social Sciences). This package program, amongst other
things, computes the mean, median, mode, variance, standard
deviation, skewness, range, minimum and maximum of the data.
This kind of information is a most helpful tool to anyone who
is making a map because it leads to a better visualization of
the data and wiser decisions can be made on how to best display
this information.
The SPSS program was used to analyze all the data
for Phases I and II. However after meetings with APCO and
-------
the Bureau of Mines it was decided to use equal value
range intervals for all maps.
As in Phase I, the cards were punched in a set
format for easy identification of the fields as follows:
Number of Analyses Columns 1-5
Low Sulfur Value Columns 6-10
Mode Value Columns 11-15
High Value Columns 16-20
Bed Code Columns 50-52
Raw or Washed Column 53
Point Sequence Number Columns 54-56
Three-Digit State Code Columns 60-62
Three-Digit County Code Columns 63-65
Five-Digit Town Code Columns 66-70
Town Name Columns 73-80
A card listing for the values of the raw and washed data may
be found in Tables III-6 and III-8, respectively, for the
Pittsburgh Bed and Table III-10 (raw values) for the Middle
Kittanning Bed. The corresponding tables for data point
locations are Tables III-5 and III-7 for the Pittsburgh Bed
and III-9 for the Middle Kittanning Bed.
1.6. PHASE II DATA POINT PLACEMENT
It was decided to aggregate the mine data to the
nearest town. This was done because: (i) many of the mine
locations were not yet referenced to the UTM grid, and (ii)
such small-scale mapping would create a great number of
superimposed data points that would negate the effort expended
in doing a map of such detail at the scale selected. Even at
the town level, there were 8 superimposed towns occuring at
-------
377-
391 .
391 .
381 .
384.
389.
395.
361 .
367.
422.
426.
439.
430.
435.
440.
441 .
444.
4.15 .
^45.
440 .
438.
441 .
430.
412.
436.
424.
427.
311 .
295.
424.
458.
450.
472.
463.
468.
458.
448.
450 .
457.
452.
465.
454 .
457.
450.
484.
482.
462.
461 .
456.
457.
468.
446.
478.
469.
316.
322.
374.
471 .
463.
455.
670.
680 .
677.
673.
675.
675.
658.
417.
394.
480.
485.
513.
521.
515.
51 1 .
505.
494.
521 .
48H.
498.
523.
510.
515.
516.
508.
521 .
490.
403.
396.
474.
500.
477.
504.
508.
498.
505.
483.
506.
506.
510.
515.
519.
531 .
519.
529.
527.
533.
51 1 .
513.
518.
532.
525.
533.
525.
408.
402.
410.
588.
582.
601 .
026K001
026R002
026R003
026R004
026R005
026R006
026R007
026R008
026R009
026R010
026R011
026R012
026R013
026R014
026KO15
026RO16
026R017
026R018
026R019
026R020
026R021
026R022
026R023
026R024
026R025
026R026
026R027
026R028
026R029
026R030
026R031
026R032
026R033
026R034
026R035
026R036
026R037
026R038
026R039
026R040
026R041
02bR042
026R043
026R044
026R045
026R046
026R047
026R048
026R049
026R050
026R051
026R052
026R053
026R054
026R055
026R056
026R057
026R058
026R059
19000100800
19000104900
19000106000
19000109500
19000110100
19000110120
19002307000
34 000901600
34000955200
34001304180
34001304700
34001304850
34001305800
34001307970
34001319480
34001325400
34001326400
34001329950
34001335700
34001339750
34001347800
34001348730
34001355000
34001367500
34001372000
34001374100
34001377650
34005315350
34005328100
34 005967900
34006712100
34006727500
34006729050
34 006736000
34006738300
34006738920
34006765230
34006775350
34006782450
34008100600
34008108410
34 008109350
340081 10100
34008122200
34008124900
34008139630
34008152000
34008163550
34008165370
34008175900
34008178300
34008180800
34008181400
34008190800
34010550000
34010571700
34011507940
37200304080
37200306520
37200307580
026R060
Table III-5
II Pittsburgh Bed - Raw Data Points
BARTON
E C K H A R T
FROSTRUR
LGNACONI
MIDLAMD
MIDLOTHI
GRANTSVI
AMESV I LL
NELSONVI
R A I L E Y M
BARNESVI
BARTON
BELLA I RE
RLAINE
CRESCENT
F A IRPOIN
FLUSHING
GLEN ROR
HOLLOW AY
LAFFER TY
MARTINS
M A Y N A R D
NEFCS
POWHA TAN
ST. CLAI
SHADYSID
SPEIDEL
CHESHIRE
GALLIPOL
QUAKER C
CADI Z
FREEPORT
GERMANO
HOPEDALE
J F W E T T
KENWOOD
PIEDMONT
SHORT CR
UNIONVAL
A DEN A
RLOOMING
BRADLEY
BR I I_LI AN
DILLONVA
EMPIRE
KNOXVILL
MINGO JC
PARLETT
PINEY FO
SMITHF I E
STEUBENV
TILTONSV
TORONTO
W INTERSV
MIDDLEPO
RUTLAND
RISHOPVI
RAL DWIN
BETHEL P
-------
467-
472.
478.
477.
460.
449.
475.
477.
467.
491 .
471 .
483.
472.
453.
470.
429.
438.
439.
398.
415.
406.
411 .
404.
420.
428.
410.
419.
398.
406.
436.
417.
401 .
^24.
400.
420.
401 .
487.
487.
495.
420.
414.
408.
401 .
459.
441 .
459.
470.
435.
456.
466.
428.
461 .
456.
470.
469.
451 .
468.
461 .
441 .
463.
575,
577
563,
577
582,
592
602,
574
572,
604
567,
6O2
569,
601
566,
594
620,
599
597,
596
593
594
593
598
632
608
609,
594
60 1
605
608
587
581 ,
592
580,
590
629
637
639
674
666
669
664,
545
584,
557
551 ,
594
568,
544
585,
559
566,
554
565,
565
559,
567
594,
565
026R061
026R062
026R063
026R064
026R065
026R066
026R067
026R068
026R069
026R070
026R071
026R072
026R073
026R074
026R075
026R076
026R077
026R078
026R079
026R080
026R081
026R082
026R083
026R084
026R085
026R086
026R087
026R088
026R089
026R090
026R091
026R092
026R093
026R094
026R095
026R096
026R097
026R098
026R099
026R100
026R101
026R102
026R103
026R104
026R105
026R106
026R107
026R108
026R109
026R110
026R111
026R112
026R113
026R114
026R115
026R116
026R117
026R118
026R119
026R120
37200308800
37200335800
37200338600
37200338800
3 7200344700
37200354000
37200354150
37200354700
37200354850
37200358800
37200359900
37200360350
37200361600
37200377770
37200382150
37205109700
37205126900
37205128200
37205141740
37205146730
37205149170
37205149400
37205158540
37205159300
37205159950
37205162390
37205162400
37205166800
37205177800
37205180900
37205186900
37205907600
37205914000
37205919950
37205949600
37205966850
37206303800
37206313770
37206391200
3721 1 106200
3721 1 13 1400
3721 1 150900
3721 1 1 73300
37212503400
372 12506000
37212509980
37212510000
37212510400
3/212510700
37212525030
372 12530300
37212536290
37212537500
37212539970
37212546900
37212549900
37212551600
37212556500
37212580100
37212587580
RRIDGFVI
HF I DF.LBU
I MPERI AL
INGRAM
LIRRARY
MONDNGAH
MONROFVI
MOON RUN
MORGAN
NFW KENS
NORLFSTO
NORTH RF
OAKDALF
SMITHDAL
STURGEON
BROWNSVI
FVERSON
FAYETTE
LAKE L Y N
MCCLELLA
MARTIN
MASONTOW
MEW GENE
NFW SALE
NORMALVI
OLIPHANT
OLIVER
POINT fv'A
SMI THE I E
STAR JCT
UN IONTOW
ROB TOWN
CLARKSVI
DILLINER
MATHER
POLAND M
AVONMORE
CLARKSRU
WEST LER
BERLIN
GARRETT
MEYERSDA
SAL ISBUR
AVELLA
BENTLEYV
BULGER
BURGFTTS
C A L I F O R N
CANMONSR
FLDFRSVI
FRFDF.R I C
H I CKORY
HOUSTON
JOFFRE
MCDONALD
MEADOWLA
MIDWAY
MUSE
SPEERS
VENICE
Table III-5
-------
447.
431 .
463.
463.
469.
457.
465.
474.
457.
464.
463.
456.
451 .
465.
472.
467.
455.
442.
460.
439.
479.
445.
452.
448.
460.
450 .
452.
338.
341 .
325.
342.
333.
325.
300.
290.
468.
464.
460 .
472.
458.
303.
348.
347.
358.
363.
358.
353.
334.
358.
355.
340.
345.
360.
354.
349.
347.
329.
32.1 .
370.
373.
St>4.
595,
614,
627
616,
612
644,
616
636,
609
637,
650
630,
612
633,
642
631 ,
597
607,
620
625,
607
627,
598
653,
610
612
579
573
576
575
583
574
529
522
538
534
540
534
533
524
564
556
550
562
556
557
556
555,
558
556
558
560,
558
551 ,
545
551 ,
546
573,
564
026R121
026R122
026R123
026R124
026R125
026R126
026R127
026R128
026R129
026R130
026R131
026R132
026R133
026R134
026R135
026R136
026R137
026R138
026R139
026R140
026R141
026R142
026R143
026R144
026R145
026R146
026R147
026R148
026R149
026R150
026R151
026R152
026R153
026R154
026R155
026R156
026R157
026R158
026R159
026R160
026R161
026R162
026R163
026R164
026R165
026R166
026R167
026R168
026R169
026R170
026R171
026R172
026R173
026R174
026R175
026R176
026R177
026R178
026R179
026R180
37212588800
37212590000
37212900250
37212907830
37212913600
37212919070
37212919800
37212927200
37212937470
37212939000
37212942900
37212944800
37212947830
37212948100
37212957500
37212958150
37212966350
37212968000
37212970350
37212974500
37212977300
37212978000
372 12986940
372129894 1 0
37212994060
37212995110
37212996600
47000102080
47000103250
47000104550
47000109660
47000121100
47000126840
47000703600
47000708370
47000905850
47000909300
47000926760
47000927300
47000927500
47002110080
47003303 POO
47003305400
47003306950
47003308280
47003311680
47003312030
47003315200
47003315300
47003317330
47003318600
47003319600
47003324300
47003325000
47003328850
47003328980
47004113100
47004127900
4 7004908400
47004908800
WASHINGT
WFST HRO
ADAMSRUR
BOVARD
CLARIDGE
DARRAGH
DFRRY
EXPORT
HOSTETTE
I RW IN
LATRDRE
LIGONIER
MAMMOTH
MANOR
NEW ALEX
NEW DERR
PLEASENT
PRICEDAL
RILL TON
SCOTTDAL
SLICKVIL
SMITHTON
UNI TED
WEBSTER
W I L P E N
W Y A N O
YUKON
RERRYRUR
RROWNTON
CENTURY
GALLOWAY
PHILIPPI
VOLGA
RURNSVIL
EXCHANGE
COLL IERS
POLLAMSB
VIRGINVI
WE IRTON
WELLSRUR
GILMER
RRIDGFPO
CLARKSRU
DOLA
EMTERPRI
H A Y W O O D
HFPZIRAH
LOST CRE
LUMRERPO
MEADOWRR
MOUNT CL
NUTTER F
SHINNSTO
SPFLTER
W I LSOMRU
WOLF SUM
JANE LEW
WFSTON
FA IRMONT
FARMING!
Table III-5
-------
3 7 H .
36b.
367.
374.
376.
366.
429.
41S.
314.
364.
393.
386.
389.
441 .
435.
360.
262.
346.
35*.
3^6.
346.
3^9.
315.
670.
570.
567.
560.
575.
563.
522.
522.
408.
661 .
587.
589.
584.
525.
526,
598.
426.
575.
583.
572.
578.
577.
567.
026R181
026R182
026R183
026R184
026R185
026R186
026R187
026R188
026R189
026R190
026R191
026R192
026R193
026R194
026R195
026R19b
026R197
026R198
026R199
026R200
026R201
026R202
026R203
4 7004910800
47004913940
47004918000
47004922050
47004923200
47004929100
47005102000
47005118500
47005327550
47005724110
47006115960
47006118400
47006120050
47006927130
47006928300
47007718900
47007901290
47009109200
47009110600
47009123460
47009124380
47009125720
47009703400
GRANT TO
K INGMONT
MONONGAH
RACHFL
R I V F S V I L
WORTHI NIG
RENWOOD
MOUNDSVI
WEST COL
SHAW
MA IDSVIL
MORGAMTO
nSAGE
w ARWOOO
WHEELING
NEWPURG
RANCROF T
FLE MI N,GT
GR A= TON
ROSEMONT
S I MPSON
WEN DEL
R U C K H A N N
and x locations
Table III-5
(Cont'd)
-------
29.
3.
2.
48.
1.
3.
1.
6.
2.
1.
18.
2.
9.
1.
2.
24.
1.
1.
"* «
?.
7.
1.
1.
13.
1.
1.
1.
1.
3.
3.
ft7.
2.
6.
P.
S.
2.
1.
1.
3.
?..
19.
1.
2.
3.
1.
1.
1.
3.
1ft.
7.
2.
1.
2.
2.
2.
2.
2.
2.
1.
1.
O.B
0.9
1.4
O.R
0.9
1.0
O.R
3.6
4.0
5.2
3.6
4.6
3.7
4.4
3.6
3.3
4.9
3.9
2.0
4.2
3.6
3.0
4.Q
3.8
5.3
4.3
4.4
4.4
3.6
4.R
1.9
3.R
3.6
1 .1
2.R
3.5
3. 1
3.6
4.0
4.0
1.5
3.0
4.0
3.6
2.6
4.2
3.2
1.1
2.4
1 .0
3.4
4.9
2.7
2.4
3.0
3.5
4.5
1 .3
1.3
1 .2
0.9
0.9
1 .4
0.9
0.9
1 .0
0.8
3.6
4.0
5.2
3.6
4.6
3.7
4.4
3.6
3.6
4.9
3.9
4.3
4.2
3.8
3.0
4.9
3.8
5.3
4.3
4.4
4.4
3.6
4.8
2.4
3.8
3.9
3.3
3.0
3.5
3. 1
3.6
4.0
4.0
3.3
3.0
4.0
3.7
2.6
4.2
3.2
1 . 1
3.0
2.4
3.4
4.9
2.7
2.4
3.0
3.5
4.5
1 .3
1 .3
1 .2
1 .2
1 .2
1 .6
1 .0
0.9
1 .0
0.8
3.8
5.0
5.2
5.7
5.6
4.5
4.4
4.5
3.9
4.9
3.9
5.3
5.2
4.7
3.0
4.9
4.9
5.3
4.3
4.4
4.4
4.8
5.7
4. 1
3.8
4.3
3.5
4.0
5.5
3. 1
3.6
*.6
4.1
3.9
3.0
4.0
4.5
2.6
4.2
3.2
3.6
3.0
3.8
4.0
4.9
4.1
3.3
3.3
5.3
4.5
1 .4
1 .3
1.2
Table III-6
Phase II Pittsburgh Bed
026R001
026R002
026R003
026R004
026R005
026R006
026R007
026R008
026R009
026RQ10
026R011
026R012
026R013
026R014
026R015
026R016
026R017
O26RO18
026R019
026R020
026R021
026R022
026R023
026R024
026R025
026R026
026R027
026R028
026R029
026R030
026R031
026R032
026R033
026R034
026R035
026R036
026R037
026R038
026R039
026R040
026R041
026R042
026R043
026R044
026R045
026R046
026R047
026R048
026R049
026R050
026R051
026n!052
026R053
02&R054
026R055
026R056
026R057
026R058
026R059
026R060
- Raw Values
19000100800
19000104900
19000106000
1 9000109500
19000110100
19000110120
19002307000
34000901600
34000955200
34 001304180
340013 04700
34001304850
3400 1 305800
34 001307970
34001319480
34001325400
34001326400
34001329950
34001335700
34 001339750
34001347800
34001348730
34001355000
34001367500
34001372000
34 001374100
34001377650
34005315350
34005328100
34005967900
34006712100
34006727500
34006729050
34 006736000
34006738300
34006738920
34006765230
34 006775350
34006782450
34 008 1 0-0600
3400810841n
34008109350
34008110100
34 008122200
34008124900
34008139630
34008152000
34008163550
34008165370
34 008 1 75900
34008178300
34 008 1 80800
34008181^00
34008190800
3401 0550000
34010571700
3401 1507940
37200304080
37200306520
37200307580
RARTDN
FCKHART
FRDSTRUR
L n N A C n N I
M I Dl_ AMD
M I DLO THI
GRANTS V I
A "F SV I LL
NF|_Sn'-:V I
P A I L - Y v
RARNESV I
RARTOM
RELLA I RF
RLA INE
CRFSC.FNT
F A I RJ-TI I M
FLUSH i \'G
GLEN- -O-
LAFFFfv T Y
MART I >••!<=>
M A Y M A R i )
N p F •- S
P O W H A T A M
ST. CLA I
SHADY^ I D
SPF I HFL
CHF SH I R F
GALL It-OL
QIJAKFk C
CADI Z
JFWFTT
K F N v.; n n -. '•
PIEDMONT
SHORT CRJ
UN i n N v AI_
A DEN A
RLHO M i NG
RRAHLFY •
RR ILL I Af-/
D I L L n N V A
FMP I RE '
KNOXV I LL
MINGD JC-i
P A R L P T T I
P I M F r r T
SVI T-- 1 =[
~TE „- E 'i v.
TILTH' S V
THRO IVTQ
WINTERSV
MI DDLEMO)
RUTLAND ^
RI SHDPVI
R A L DV" I N
RFTHEL P
-------
1.
1.
IB.
1.
2.
12.
7.
1.
1.
2.
3.
1.
2.
2.
1.
9.
l.
5.
31.
2.
1.
3.
4.
1.
7.
1.
4.
14.
106.
1.
126.
3.
4.
4.
1.
2Q.
1.
7.
100.
33.
2.
31.
1.
o.
3.
2,
7.
1«.
1.
1.
3.
1.
4.
2.
1,
3.
1.
I.
•5.
2.
3.1
0.9
0.7
2.2
1.1
1.0
0.9
2.4
2.1
2.7
1.2
1.5
1.6
0.9
1.6
1 .3
0.7
0.9
2.1
1.9
2.7
0.9
0.9
1.0
1.7
1.3
1.0
2.1
0.5
1.2
0.8
2.5
0.9
2.4
1 .3
1.5
2.4
1 .1
1.1
1 .0
1.5
0.9
0.7
1 .8
1.2
1 .6
2.0
0.8
2.2
3.S
1.0
2.4
1.4
2.6
3.2
1.5
1.8
2.0
1. 1
2.9
3.1
0.9
2.0
2.2
1.1
1 .0
0.9
2.4
2.1
2.7
1.2
1 .5
1 .6
0.9
1 .6
1 .4
0.7
0.9
3.2
1 .9
2.7
0.9
0.9
1 .0
1.7
1 .3
1.0
2.1
1.2
1 .2
0.9
2.5
1 .3
2.5
1 .3
3.0
2.4
1 . 1
1 .2
1 .8
1.5
1 .9
0.7
2.8
1 .2
1 .6
2.0
0.8
2.2
3.5
1.0
2.4
1 .4
2.6
3.2
1 .5
1 .8
2.0
1 . 1
2.9
3.1
0.9
3.0
2.2
1 .5
1 .1
1 .0
2.4
2. 1
3.2
1 .4
1 .5
2.1
0.9
1 .6
1 .4
0.7
1 .2
3.6
2.0
2.7
1 .9
2.2
1 .0
1 .7
1.3
1.1
3.5
1 .9
1 .2
1 .1
3.5
1 .7
3.2
1 .3
3.3
2.4
2.0
1 .6
2.3
1 .8
2.3
0.7
4.4
1 .4
1 .6
4.5
1 .3
2.2
3.5
1 . 1
2.4
1 .9
4.5
3.2
1 .9
1 .8
2.0
1 .1
3.2
026R061
026R062
026R063
026R064
026R065
026R066
026R067
026R068
026R069
026R070
026R071
026R072
026R073
026R074
026R075
026R076
026R077
026R078
026R079
026R080
026R081
026R082
026R083
026R084
026R085
026R086
026ROH7
02t>R08b
026R089
026R090
026R091
026R092
026R093
026R094
026R095
026R096
026R097
026R098
026R099
026R100
026R101
026R102
026R103
026R104
026R105
026R106
026R107
026R108
026R109
026R110
026R111
026R112
026R113
026R114
026R115
026R116
026R117
026R118
026R119
026R120
37200308800
37200335800
37200338600
37200338800
37200344700
37200354000
37200354 1 50
37200354700
37200354850
37200358800
37200359900
37200360350
37200361600
37200377770
37200382150
37205109700
37205126900
37205128200
37205 14 1 740
37205146730
37205149170
37205149400
3720515854 0
37205159300
37205159950
37205162390
37205162400
37205166800
37205177800
37205180900
37205186900
37205907600
37205914000
37205919950
37205949600
37205966850
37206303800
37206313770
37206391200
3721 1 106200
37211131400
3721 1 150900
37211173300
37212503400
37212506000
37212509980
37212510000
37212510400
37212510700
37212525030
37212530300
37212536290
37212537500
37212539970
37212546900
37212549900
37212551600
37212556500
37212580100
37212587580
RRIDGEVI
HE I DFLBU
I MPFR I AL
INGRAM
L I RR ARY
MONOr K, A H
M ON I-")- / t
Mf")^ r- A ' ,
MEW KE'.S
NHBLF c TO
NORTH -, E '
n AKD ALE
S M I T h 0 A L
STURGEON
R R O W N S V I •
F fiYETTE
LAKE L Y r-.'
M C C L P L L A
MARTI \'
MASON TOW l
NFW GENE
NEW SALE '
NORM A|_V I
H L I P H A f^ T
OLlVFk
POINT MA'
SMI THE I E
STAR JCT
UN IONTOW
RORTO'JV1 N
CL ARKSVI
D ILL I NER
MATHER
POLAND M
AVONMORF
C L A R K S H U
WEST L F B
PFRL I M
G A R P F T T
MEYERS DA
SAL I SPUR
AVELL A
REN TLFY v
BULGER
RURGFTTS
C A L I F O 4 N
C ANNO MS H
ELDERS V I
FRFDEH I C
HICKORY
HOUSTON
JOE FRF
MCDONALD
MEADOWLAi
M I DW A Y
MUSF
SPFERS
VEN I CF-
Table III-6
-------
ft.
4.
1.
3.
3.
1.
2.
2.
2.
4.
3.
3.
-1.
1!.
4.
1 .
63.
1.
3.
6.
19.
3.
1.
1.
3.
3C.
3.
4.
?P.
4.
1.
14.
7.
3.
; t
1.
10.
1.
H.
S.
29.
3?.
44.
30.
1.
2.
r>.
°.
1A.
35.
!i.
S.
71.
2?.
7.
9.
9.
7fi.
1.
A.
1.9
0.9
2.0
1.2
0.7
1.4
2.4
1 .5
0.9
1 .8
0.9
1.0
0.9
0.7
0. -
1 .2
0.9
1 .0
0.9
1.2
0.7
1 .0
1 .0
1 .0
1.4
1 .0
0.6
3.2
1 .7
4. .0
3.0
2.1
3. 1
1 .7
2.7
3.5
1 .2
1 .9
2.3
1 .6
2. 1
2.1
2.5
3.2
3.6
3.7
3.2
3.1
2.2
1 .4
3.0
3.2
1.4
2.9
3. 1
3.1
3.2
2.7
1.5
1 .0
2.3
0.9
2.0
1 .2
1 .7
1 .4
2.4
1 .5
0.9
2.0
0.9
1 .0
0.9
0.9
0.?
1 .2
1 .0
1 .0
0.9
1 .2
1 .2
1 .0
1.0
1 .0
1 .5
1 .0
0.6
3.2
1 .8
4.3
3.0
2.2
3.5
2.5
2.7
3.5
1 .7
1 .9
2.3
1 .6
2.3
3.3
2.9
4.0
3.6
3.7
3.3
3.8
4.0
2.4
3.8
3.2
2.5
3. 1
3.8
3.7
3.2
3.3
1 .5
1 .4
2.5
1 .2
2.0
1 .3
1 .8
1 .4
2.5
1 .7
1 .3
2.2
0.9
1 .3
1 .5
1 .6
0.9
1 .2
1 .2
1 .0
1 .0
1 .4
2.0
1 .2
1 .0
1 .0
1 .6
1 .3
1 .3
3.5
4.5
4.5
3.0
5.2
5.4
3.0
2.7
3.5
3. 1
1 .9
2.7
4.6
3.4
4.4
4.5
4.9
3.6
3.9
3.4
4.5
4.6
3. 1
4. 1
4.8
4.2
3.7
4.6
4.0
4.3
4.0
1 .5
2.8
026R121
026R1??
026R123
026R124
026R125
026R126
026R127
02t>H 1 28
026R129
026R130
026R131
026R132
026R133
026R134
026R135
026R136
026R137
026R138
0 2 6 R 1 3 9
026R140
026R141
026R142
026R143
026R144
026R145
026R146
026R147
026R148
026R149
026R150
026R151
026R152
0 2 6 R 1 5 3
026R154
026R155
026R156
0 2 6 R 1 5 7
026R158
026R159
026R160
026R161
026R162
026R163
026R1b4
026R165
026R166
026R167
026R168
026R169
026R170
026R171
026R172
026R173
026R174
026R175
026R176
0 2 6 R 1 7 7
026R178
026R179
026R180
37212588800
37212590 0 0 0
372 12900250
37212907830
37212913600
37212919070
372 12919800
37212927200
37212937470
37212939000
37212942900
37212944800
37212947830
37212948100
37212957500
37212958150
37212966350
37212968000
372 12970350
3 7212974500
37212977300
37212978000
37212986940
37212989410
37212994060
37212995110
37212996600
47000102080
47000103250
47000104550
47000109660
47000121100
47000126840
47000703600
47000708370
47000905850
4 7000909300
47000926760
47000927300
47000927500
47002 1 10080
47003303200
470033054 00
47003306950
47003308280
47003311680
47003312030
47003315200
47003315300
47003317330
47003318600
47003319600
47003324300
47003325000
47003328850
47003328980
47004113100
47004127900
47004 908400
47004908800
W A SH ING T
W F S T H R f)
AOAMSHUR
RfJV A HO
CLAR I }•)(-> f-.
D A R R A G H
QFR^Y
P X P n ~> T
HO STF TTF
IRW I \
LAT3O-£
LIGD^IFR
MAMMO TI-I
f/ AMOR
N P W ALEX
M F W D F R R
PLEASANT
PR I OF DAL
R I LLTOri
SOOTTOAL
SL I CK" VI L
SMITHTON
UNITFO
W F. P S T F w
W I LHF>
W Y A M O
YUKON
RFRRYRUR
RROW NTON
CFNTURY
CALLOW AY
P h I L I P H I
VOLGA
R'JRMSV I l_
FX CH AMG^
COLL I FR
-------
\\?> l-fi
/1 , 0 . 9
7, °«R
I1"!. I'4
H. 0.0
A. 1.1
1. 4.5
S, 4.5
4. 2.4
1 . 0.7
l?fl. 1.5
SO. l.«
}.• 2.5
A. 2.8
1. 4.4
3. 0.9
11. 2.5
1^. 1 .-
•i. 1.5
r . 1.3
3 . 4-
1". 1 .4
^. 3. 1
1 .6
1 .2
O.R
1 .4
1 .R
2.0
4.5
4.6
2.4
0.7
2.8
2.5
2.5
2.8
4.4
0.9
2.5
1 .6
1 .5
2.3
3.4
2.5
3. 1
2.4
3.2
2.8
3.7
2.7
2.9
4.5
5.3
3.6
0.7
4.4
3.4
3,1
4.3
4.4
1 .8
2.6
^.4
3.7
3.6
3.6
2.5
4.2
02GR181
0 2 6 R 1 8 2
02t>h!l 83
026hil«4
026R185
026R186
0 2 6 R 1 8 7
026K1H8
026W189
026R190
026H191
026R192
026R193
026R194
026R195
026R196
026R197
026R198
026R199
026R200
026R201
026R202
026R203
47004910800
4700491394 0
47004918000
47004 922050
47004923200
47004929100
47005102000
4 70051 18500
4700532 7550
4 7005724 1 1 0
47006115960
47006118400
47006120050
47006927130
47006928300
47007716900
47007901290
47009109200
47009110600
4 7009123460
47009124380
47009125720
47009703400
GRANT TO
K ING MO NT
MONO MG AH
RACHFL
R I V E S V I L
WORTH I MG
RPMWOH O
MDl JM ,r I I
MAIDS V I L
MQRGA '-.'TO
OSAGF
W ARWOOD
WHEEL I NG
A-
SI V
'j .1 P N D F L
RUCK HA • :M
iber . low . mode . high
Key:
Table III--6
(Cont'd)
-------
430.
435.
445.
423.
430.
412.
436.
458.
449.
456.
471 .
466 .
478.
40? .
398.
401 .
420.
459.
441 .
470 .
466.
463 .
447.
477.
464.
467.
342.
333.
300.
472.
303.
347.
358.
3o3 .
334.
353.
359.
349.
321 .
373.
370.
378 .
367.
374 .
376.
3b6.
418.
314.
380.
37Q.
393.
386.
391 .
432.
433.
346.
521 .
515.
521.
507.
515.
516..
508.
500.
51 6.
513.
588.
571 .
563.
606.
597.
590.
674.
545.
584.
551 .
544.
565.
564.
647.
640.
627.
575.
583.
529.
534.
524.
556.
550.
562.
556.
555.
563.
551 .
546.
564 .
559.
570.
567.
560.
575.
563.
522.
408.
581 .
580.
587.
589.
584.
529.
531 .
575.
026W 01
026W 02
026W 03
026W 04
026W 05
02bW 06
026W 07
026W 08
026W 09
026W 10
026W 11
026W 12
026W 13
026W 14
026W 15
026W 16
026W 17
026W 18
026W 19
026W 20
026W 21
026W 22
026W 23
026W 24
026W 25
026W 26
026W 27
026W 28
026W 29
026W 30
026W 31
026W 32
026W 33
026W 34
026W 35
026W 36
0 2 6 W 37
026W 38
026W 39
026W 40
026W 41
026W 42
026W 43
026W 44
026W 45
026W 46
026W 47
026W 48
026W 49
026W 50
0 2 6 W 51
02bW 52
026W 53
026W 54
026W 55
026W 56
34001305800
34001307970
34001329950
34001337560
34001355000
34001367500
34001372000
34006712100
34008122770
34008165370
37200304080
37200318120
37200338600
37205127500
37205141740
37205966850
37211106200
37212503400
37212506000
37212510000
37212525030
37212587580
37212588800
37212907000
372 12908350
37212934680
47000109660
47000121100
47000703600
47000927300
47002110080
470033054 00
47003306950
47003308280
47003315200
47003315300
47003320120
47003328850
47004127900
47004908800
47004909450
47004910800
47004918000
47004922050
47004 923200
47004929100
47005118500
4 7005327550
47006 107650
47006108350
4700M 15960
47006118400
47006121920
47006908160
47006926300
47009109200
BELLA I WE
RLA IMF
GLEN ROR
JACnRSBU
NEFF S
POWHA TAN
ST. CLAI
CADIZ
DUNGLEN
p i N E Y F n
RALOWiN
CUDDY
IMPERIAL
F A i RCHAN
LAKE LYN
POLAND M
RFRLIN
A V F L L A
RENTLFYV
RURGETTS
F L D E R S V I
VFNICE
WASH IMGT
RLAIRSVI
RRADENVI
HANMASTO
GALLOWAY
PHILIPPI
RURNSVIL
W F I R T O N
GILMER
C L A R K S H U
on LA
ENTFRPRI
LOST CKE
LUMRFRPO
OW INGS
W I L S O N RIJ
WFSTON
F ARMIMGT
FOUR STA
GRANT TO
MONO MGAH
RACHEL
R I V F S V I L
WORTH ING
MOUNDSVI
WEST COL
FDNA
EVFRETTV
MA IDSV IL
MORGAMTO
PURSGLOV
ELM GROV
TRIADELP
FLE f>"I NGT
Table III-7
Phase II Pittsburgh Bed - Washed Data Points
-------
3.
2.
2.
T,
?,
11.
2.
2.
1.
4.
1.
1.
1.
2.
2.
1.
T.
3.
1.
7.
1.
2.
1.
2.
1.
56.
11.
1.
a.
1.
1.
20.
12.
1.
2.
1.
i.
1.
1.
i.
11.
1.
2.
A.
2.
1.
1.
1.
'<"••
2.
4.
U.
?•
2.
3.
1.
3.7
4. 1
3.6
4.9
A. 2
3.8
4.2
2.0
3.2
2.0
1 .2
2.0
1 .7
1. 1
2.5
3.0
1 .5
1.2
1 .0
2.9
3.7
2.6
l.Q
2.6
2.1
1.0
2.8
3.2
1.2
2.7
1 .3
2.4
3.1
2.8
3.P
3.S
1 .9
3.4
3.0
0.9
1 .5
2.7
0.7
1.6
1 .2
2.1
4 .S
3.3
2.7
3.3
1 .8
2.3
2.1
3.3
3.2
2.6
3.7
4. 1
3.6
4.9
4.2
4.3
4.2
2.0
3.2
2.3
1 .2
2.0
1 .7
1 . 1
2.5
3.0
1 .5
1 .2
1 .0
3.1
3.7
2.6
1 .9
2.6
2. 1
1.0
2.8
3.2
1 .6
2.7
1 .3
2.5
3.6
2.8
3.8
3.5
2.2
3.4
3.0
1 .8
2.7
2.7
0.7
1.7
1 .2
2. 1
4.5
3.3
2.7
3.3
2.5
2.5
2. 1
3.3
3.2
2.6
3.9
4.1
4.2
5.0
4.4
4.8
4.3
2.8
3.2
3.3
1 .2
2.0
1 .7
1 .4
2.6
3.0
1 .7
4.2
1 .0
3.3
3.7
3. 1
1 .9
2.7
2.1
1 .2
3.2
3.2
2.8
2.7
1 .3
3. 1
3.8
2.8
4.8
3.5
2.6
3.4
3.0
2.6
3.2
2.7
1 .4
2.3
2.4
2.1
4.5
3 . 3
2.7
3.6
2.7
3.3
2.8
4.2
4. 1
2.6
026W 01
026W 02
026W
026W
026W
026W
026W
026W
026W
026W
026W
026W
026W
026W
026W
026W
026W
026W
026W
026W
02 6W
026W
026W
026W
026W
026W
026W
026W
026W
026W
026W
026W
026W
026W
02fc>W
026W
026W
026W
026W
0 2 6 W 40
026W 41
0 2 6 W
026W
0 2 6 W
02bW
026W
026W
026W
026W
026W
026W
026W
026W
026W
026W
026W
03
04
05
06
07
oa
09
10
11
12
13
14
15
1 6
1 7
1 8
19
20
21
22
23
24
25
26
27
2H
29
30
31
32
33
34
35
36
37
3 a
39
42
43
44
45
46
47
4H
49
50
51
52
53
54
55
56
34001305800
34001307970
34 001329950
34001337560
34001355000
34001367500
34001372000
34006712100
34008122770
34008165370
37200304 080
372003181?0
372O0338600
37205127500
37205141740
37205966850
37211106200
37212503400
37212506000
37212510000
37212525030
37212587580
37212588800
37212907000
37212908350
372 1 2934680
47000109660
47000121100
47000703000
4 7000927300
47002110080
470033054 00
47003306950
470033 08280
47003315200
47003315300
47003320120
47003328850
47004127900
47004908800
47004909450
47004910800
47004918000
47004922050
47004923200
47004 929100
47005118500
4 7005327550
47006107D50
4 700^1 08.3 50
47.0061 15960'.
470061 184 00
4700612 1920.
47006908160
47006926300
47009109200
RELLAIRE
RL A I NE
GLEN ROR
J A CO R S R U
NFFFS
POWHATAN
ST. CLAI
CADIZ
DUNGLFN
P I NFY F n
R A LOW I N
CUDDY
IMPFRIAL
F A I R C <-i A \
L A K F L \ K'
POLAND ^
PER LI \
AVE LL A
»ENTLEY V
BURGFTT?
F L D E 3 ? V I
VFNJ I CF
WASHI vGT
RL A I RSVI
RHADENV I
HANNASTO
G A L L O 'A A Y
P H I L I >-• '-'• I
R L i R N' S V I L
WF I RTP\
GIL ^ «
C L A R K S -1 1 '
DHL A
E N T E R •-" K I
LO?T
O W IMG?
W I LSn\>^!J
FARMING!
FOUR c T A
GRANT TO
M n N O N G A ^
RACHEL
R I V E ^ V I L
W O R T H I ,v (5
MOUNDS V I
WEST Cni_
P DN A
EVFRF T r v
MAI DSV IL
MORGAN TO
PURSGLOV
ELM GROV
TR I ADF-'LP
FLEM i MG T
Table III-8
Phase II Pittsburgh Bed - Washed Values
-------
Phase
404.
390.
402.
406.
406.
405.
394.
399.
469.
41 8.
423.
427.
437.
436.
374.
399.
382.
440.
509.
532 .
405.
410.
419.
416.
434.
408.
415.
405.
415.
394 .
403.
406.
400.
393.
403.
402.
396.
393.
406.
401 .
395.
388.
459.
468.
474.
461 .
44 7.
465.
368.
548.
545.
550.
720.
592.
593.
589.
607.
572.
579.
592.
Table III -9
II Mitldls Kittanning Bed
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
07QR
070R
070R
070R
070R
070R
070R
- Raw
01
0?
03
04
05
06
07
08
09
1 0
1 1
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Da
34000903600
34000913230
34000915100
34 000930200
34000937500
34000951350
34000955200
340009621?0
34001951700
34003108150
34003118300
34003119200
34003127620
34003187600
34007353890
34007354500
34007378250
340075044 00
34008107200
34008178300
3401 1519800
3401 152 1020
34011900250
34011929380
34011962550
3401 1970900
340 1 1978540
34 01 1989360
34011992600
34012710140
34012718350
34012718900
34012734500
34012745330
340127455PO
34012754330
34012757000
34012758100
34012768700
34 012772850
34012774600
34012776200
34015722400
34015750300
34015756450
34015757700
34015779100
34015782000
34016332400
37200719050
37200726260
37200758500
37201301300
37201907990
37201910200
37201926650
37201928350
37201967350
37201977500
37201992290
ATHENS
CARRONDA
CHAUNCEY
GLOU5TER
JACKSONV
MILLFIFL
NELSONVI
ORRISTON
M I M E R A L
PL ISSF I E
CONESVIL
COSHOCTO
FRESNO
WFST LAF
MOUNT PL
MURRAY C
STARR
BALTIC
RFRGHOLZ
STEURENV
CROOKSVI
OFAVER TO
ADAMS MI
GILRFRT
OTSEGO
ROSEVILL
STOVERTO
WHITE CO
ZANFSVIL
RR I STOL
CONGO
CORN IMG
HE MLOCK
MCCUNEVI
MCL1 JNFY
MQXAHAL A
NEW LEX I
NEW STRA
RENDVILL
SALTILLO
SHAWNEE
SOMERSET
DOVER
MIDVALE
NEW C U M R
NEW PHIL
SUGAR CR
TUSCARAW
HAMDEN
DARLIMGT
ENON VAL
NEW G A LI
ALTOOMA
ROYFRS
RUTLFR
FUCLID
FENELTON
PORTERSV
SLIPPERY
-------
503.
538.
523.
553.
522.
555.
541 .
546.
522.
536.
525.
531 .
560.
549.
549.
541 .
576.
572.
551 .
540.
564.
443.
442.
441 .
333.
248.
369.
297.
310.
292.
687.
735.
723.
733.
723.
742.
738.
691 .
716.
733.
730.
734.
746.
697.
721.
722.
693.
709.
677.
547.
540.
685.
686.
674.
583.
490.
613.
583.
590.
567.
070R
070R
070R
070R
070R
070R
070R
070k
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
37202 104300
37203301P50
37203309000
37203330920
37203337600
37203340440
37203341640
37203346170
37203347530
37203354880
37203362920
37203365500
37203366940
37203371150
37203375880
37203394630
37204708570
37204789430
37206569700
37207324710
37208575500
37211110300
37211111800
37211181700
47000121100
47001502420
47007714100
4 7008304410
47008319550
47009700450
RARNSHDR
ALLPORT
RR i SRIN
FRENCHVI
HOUTZDAL
KARTHAUS
KYLERTOW
LUTHERSR
MADERA
MORRISDA
C1SCEOLA
PHILIPSR
POTTERSD
ROCKTON
SHAW SVIL
WOO PL AND
BRANDY C
WFEDV ILL
RFYNOLP5
E PINRURG
SHARON)
cAiUNRRO
CENTRAL
STOYSTOW
PHILIPPI
P ICKMORE
KINGWOOD
CASSITY
NORTON
ALEXANOPE
y and x locations
Table III-9
(Cont'd)
-------
1.
30.
B.
1.
1.
1.
P.
2.
1.
2.
1.
11.
1.
1.
1.
7.
2.
1.
: l.
' '.
" i
1«
•;,
i.
3.
q'
3.
IK"
1.
1.
A.
1.
13.
29.
1.
1.
9.
IS.
2,
1.
26.
1.
2.
1.
1.
2.
1.
1.
1.
^
1.
1.
2.
2.
H-
1.
2.
3.
?..
1.
1.7
1.4
0.8
2.2
1.0
1.5
1.1
1.4
4.8
3.2
B.4
1.1
3.0
3.8
3.6
0.6
2.1
3.2
2.i-
2.0
2 * "3
3.-
2.7
3.0
3.9
4.0
4.3
2.8
4.0
3.3
0.6
4.3
1 .6
2.8
3.1
4.4
2.5
2.2
2.5
4.6
1 .0
4.2
3.4
5.0
3.5
3.8
4.1
3.9
2.4
0.6
1 .2
0.9
1 .8
0.9
1 .3
1.6
3.3
4.1
3.7
3.3
1 .7
1 .8
1 .0
2.2
1 .0
1.5
2.9
1 .4
4.8
3.2
5.4
1 .9
3.0
3.8
3.6
2.4
2.1
3.2
2.4
2.6
3.8
3.5
2.7
3.0
4.3
4.4
4.3
2.8
4.0
3.3
1.0
4.3
2.0
3.7
3.1
4.4
2.7
2.2
2.5
4.6
3.0
4.2
3.4
5.0
3.5
3.8
4. 1
3.9
2.4
0.6
1 .2
0.9
1 .8
0.9
1 .3
1 .6
3.3
4. 1
3.7
3.3
1 .7
2.1
1 .8
2.2
1 .0
1 .5
2.9
3.2
4.8
3.9
5.4
4.6
3.0
3.8
3.6
2.9
2.1
3.2
2.4
3.5
4.8
3.5
3.8
4.2
4.5
4.6
5.5
4.7
4.0
3.3
2. 1
4.3
2.6
4. 1
3. 1
4.4
5.3
2.9
2.7
4.6
4.3
4.2
3.6
5.0
3.5
4.1
4. 1
3.9
2.4
4.5
1 .2
0.9
3.7
4.0
3.7
1 .6
3.4
6.2
4. 1
3.3
07QR
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
070R
01
02
03
04
05
06
07
Ob
09
1 0
1 1
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
34000903600
34000913230
34000915100
3400093 0200
34000937500
3400095 1350
34000955200
340009621 20
3400195 1 700
34003108150
34003 1 18300
34003 1 19200
34003127620
34003 1 87600
34007353890
34007354500
34 007378250
34007504^00
340081 07200
340081 78300
3401 1 5 1 9800
3401 1521020
3401 1900250
340 1 1 929380
3401 1962550
3401 1 970900
34 01 1 97854 0
340 1 1989360
34 0 1 1 992600
3401271 0140
340127 18350
34012718900
34012734500
340 12745330
34 01 2745520
340 1 2754330
3 40 12 75 7000
340 1 2 758 1 00
34012768700
340 12 77? 850
34 0 1 2774600
34012776?00
34 0 1 57224 00
34015750300
3401 5756450
340 1 5757700
34 0 1 5779 1 00
340 15782000
34 0 1 6332400
3720071 9050
37200726260
37200758500
37201301300
3 72 01 907990
37201910200
37201926650
37201928350
3720 1 967350
37201 977500
3720 1 992290
A THENS
CARRONDA
CHAUNCEY
GLOUSTFR
JACKSONV
MI LI_F I EL
NFLSONVI
ORRI STOM
MINERAL
RL ISSFI E
CONESVIL
COSHDCTO
FRESNO
W F ^ T |_ A F
MOUNT PL
MURRAY C
STA-&
RALT I r.
ppRGnr LZ
STFl jc^NV
CPQOK" =• V I
DE AVFk TO
A D A f." ^ f/i I
G I L R F R T
O TSFGO
ROSFV I LL
STOVER TO
WHI TE CO
Z ANFSV I L
RR ISTOL
CONGO
CORN I NG
HE MLOCK
MCCUNEVI
MCL UK'~ Y
MO X A H A L
-------
3.
1.
].
2.
1.
8.
P.,
1.
2.
1.
ft,
2,
2,
1.
S.
K •
1.
1.
7.
2.
1.
1.
1.
].
1.
4,
2.
3.
5.
1.
0.6
2.7
1.8
2.1
1.6
1.3
0.5
1.6
1.2
3.4
2.3
1.8
2.5
2.?
0.5
1 . 1
1.8
3.9
0.6
2.5
1 .6
1.0
2.8
1.2
i .=;
0.8
2.8
2.4
0.8
0.6
0.6
2.7
1 .8
2. 1
1 .6
2.9
0.5
1 .6
1 .2
3.4
2.3
1 .8
2.5
2.5
0.5
2.2
1 .8
3.9
0.6
2.5
1 .6
1.0
2.8
1 .2
1 .5
0.8
2. ft
2.4
1 .7
0.6
1 .6
2.7
1 .8
2.5
1 .6
2.9
1 .0
1 .6
i .a
3.4
4.2
1 .9
3.2
2.5
1 .8
4. 1
1 .8
3.9
0.7
3.9
1 .6
1 .0
2.8
1 .2
1 .5
1 .1
4.8
2.8
1 .7
0.6
070R bl
070k 62
070k
070R
070k
07QR
070R
070R
070k
070k
070k
070K
070R
070k
070k
070k
070k
070k
070k
070k
070k
070R
070k
070k
070k
070k
070k
070k
070k
070R
63
64
65
6b
67
6b
69
70
71
72
73
74
75
76
77
78
79
dO
81
82
83
84
85
86
67
88
89
90
37202104300
37^03301250
37203309000
3 7203330920
37203337600
37203340440
37203341640
37203346170
37203347530
37203354880
37203362920
37203365500
37203366940
37203371150
37203375880
37203394630
37204 708570
37204 789430
37206569700
37207324710
37208575500
37211110300
37211111800
372 1 1 181 700
47000121100
470015024 20
47007714100
47008304410
47008319550
47009700450
BARNFSBO
ALLPORT
R R I 5 R I N
FkENCHVI
HOUTZDAL
KARTHAUS
KYLER TOW
LUTHERSR
MADERA
MOkk I SOA
OSCFOLA
P H I l_ I P S R
POTT-PSO
ROC^ TON
S H A vl S w I i_
W O Q D L A N n
R R A M D Y C
WFFOV I LI.
kFYNHI OS
P rj I N H i j H> f,
s H A R n r-'
CAIRNPkO
CFN TR A I.
9 TO YS T'J ,-)
PHILIR^I
R I CK vrj^F
K i N G w n n n
CASSITY
NORTON
A L E X A M I") F
»ber . low . mode . high Key:
Table III-10
(Cont'd)
-------
4 locations for the Pittsburgh Bed and 2 at 1 location for
the Middle Kittanning Bed. In the case where two data
points are so close that they are superimposed then the program
associates the mean of the values with that location. The
character symbol slash (/) was used to represent this
phenomenon on the map.
There are 203 raw and 56 washed data points in the
Pittsburgh Bed and 90 raw and 15 washed data points in the
Middle Kittanning Bed. Since there was so little washed
data available in the latter bed, this information was not
mapped.
For mapping it is necessary to locate each town by
some kind of an x-y coordinate system. SYMAP will accept
any unit of measurement as long as one specifies to the pro-
gram what measurement system is being used. The program
expects coordinates to increase in magnitude in both the
horizontal and vertical directions, unlike UTM s which increase
in magnitude in the horizontal but decrease in the vertical
direction. This incompatibility can be overcome in SUBROUTINE
FLEXIN. In READ statement 90 (see Table III-ll), the vertical
(T(l)) and horizontal (T(2)) UTM coordinates are read in from
the cards in the B-DATA POINTS package. The difference
between these coordinates and the recorded UTM s at the
0,0 origin (626,322) is then divided by the number of UTM s
per linear base map inch. These transformed coordinates
are in inches and are now compatible as SYMAP input. The
resultant town placement is accurate to within one and a half
miles. If the coordinates were read into the program unchanged,
then the resultant map would have appeared true in the
horizontal but upside down in the vertical direction.
-------
It was decided to use UTM measurements since all the coal data
at the Bureau of Mines is located by these units for the United States.
Thus the four-state region could also serve as a test area to see how
SYMAP could be used for coal data anywhere in the United States.
The procedure for locating the 200 plus towns involved using
three different map scales. Before any towns could be located, it
was necessary to devise a standardized system for the easy removal
of coordinates. To this end, a 10x10 kilometer acetate grid overlay
was created.
As many towns as possible were located by this method on th,e
Army Map Service 1:250,000 series. All towns are situated within
zone 17 of this series. Approximately one tenth of the towns could
not be found because of their small size. It was therefore necessary
to use USGS maps of a larger scale—namely the 1:62,500 and
1:24, 000 map series. The towns were generally located within a county
by using the 1968 Rand McNally Commercial Atlas and Marketing Guide
and specifically located by examining large scale quadrant maps of the
two aforementioned scales. Points representing the towns were located
on the 1:250,000 maps by a combination of two methods: (1) accurate
latitude and longitude and (2) visual comparison by identifiable physio-
graphic features for further locationai accuracy. The UTM coordinates
were then easily removed from the map by using the acetate grid.
All UTM coordinates were read to the nearest one kilometer cell.
Decisions had to be made concerning two different types of locationai
situations: (1) if a town was larger than one UTM cell, then the cell
nearest to the central business district was used and (2) if a small
town (shown by a dot on the map) fell on either the horizontal or
vertical line or both then the cell nearest to a mine symbol was used.
-------
Since all the towns were within one UTM zone it was
only necessary to record the last three UTM numbers onto
cards. All vertical coordinates were punched as a decimal
number in Columns 17-20, and all horizontal coordinates were
punched as a decimal number in Columns 27-30. Columns
50-80 are in the same format as the E-VALUES package.
All of the data points were sorted by machine in
ascending order by their x and y coordinates as a verification
in key punching, and in the locational accuracy of the towns.
It is a much simpler task to recheck the town locations on the
map quadrangles if they are grouped by UTM s. By sorting
on both the x and y column field, one sorted Listing provides
a cross-check for the other. Since each card has a uniquely
associated point sequence number, they may be sorted back
into the proper order if dropped or out of sequence. It
thus becomes important to have the cards numbered arid to
have a card sorter at your disposal. There are computer
utility programs which give a sorted card listing and do not
disturb the card order.
1.7. SUBROUTINE FLEXIN
As previously mentioned in the section INPUT to SYMAP,
the input deck consists of a few introductory cards which
preceed the various packages. Three of these cards are the
skeletal framework of an optional sub-program known as SUB-
ROUTINE FLEXIN. Within this framework of instructions
(written in FORTRAN IV computer Language) the user, commen-
surate to his programming ability, has considerable latitude
and flexibility in the amount and kind of data he can exact
from the computer. Even if no data or program manipulation
is desired, SUBROUTINE FLEXIN must be provided in a
-------
SUBROUTINE FLEX IN 7X,2A4 )
RETURN
20 READ (5,200) T
200 FORMAT (5X,F5.1,62X,2A4)
RETURN
30 READ (5,300) T
300 FORMAT ( 1 OX,F5.1 ,57X,2A4 )
RETURN
40 READ (5,400) T
400 FORMAT ( 15X ,F5.1 , 52X , 2A4)
RETURN
50 READ (5,500) T
500 FORMAT (F5.0,67X,2A4)
RETURN
60 READ (5,600) T
600 FORMAT (5X , F5,1 , 62X , 2A4)
RETURN
t!o 70 READ (5,700) T
QO" 700 FORMAT ( 1 0 X , F5 . 1 , 5 7X , 2 A4 )
1 RETURN
80 READ (5,800) T
800 FORMAT (15X,F5.1,52X,2A4)
RETURN
90 READ (5,900) T(1),T(2)
900 FORMAT (10X,2F10.2)
T( 1 ) = (626.-T( 1 ) )/6.37
T(2)=(T(2)-322.)/6.37
RETURN
FND
Table III-11
-------
"dummy" form since it is an integral part of the SYMAP
deck. Such a "dummy" FLEXIN consists of the following three
cards: (1) SUBROUTINE FLEXIN (IFORM, T, FIRST) (2)
RETURN (3) END. This simple format tells the computer
that there are no special manipulations to take place, e. g. ,
data bank retrieval, map axis rotation, card identification,
number incrementation, etc., and for the computer to continue
reading and executing the successive cards.
In the case of this project, the instructions required
to retrieve the data from the E-VALUES package were placed
in this subroutine. The values package is essentially a data
bank since each card consists of more data than is applicable
for any given map, so the necessary retrieval instructions to
obtain the data pertinent to a single map are contained within
the SUBROUTINE FLEXIN framework.
A single SUBROUTINE FLEXIN was used in both
Phases I and II, a listing of which appears in Table III-ll.
As may 'be noted in this table there are nine different READ
statements, or one READ statement for each execution. How-
ever, in Phase I, only six of these executions were performed
by the computer; READ statements 20, 30 and 40 produced
maps A-2, A-3, and A-4 and READ statement 60, 70 and 80
produced maps A-6, A-7, and A-8. READ statements 10 and
50 were later made obsolete by a separate subroutine to produce
the base maps for the raw and washed data for both phases of
the project. READ statement 90 is used to manipulate the UTM
coordinates into map inches in Phase II and was discussed in
that section under Data Point Placement.
An example of SUBROUTINE FLEXIN will be discussed
here as used in Phase I. If one wishes to produce a map of
-------
raw sulfur content (low value) then it is necessary to put the
number 2 in column 5 and the number 33 in columns 9 and
10 of the second card in the E-VALUES package. This first
number, 2, causes the second execution in the SUBROUTINE
FLEXIN to take place. The FORMAT statement which follows
tells the program which columns to read in order to get the
pertinent values to be used in the map and to print the name
of the county in Columns 73 through 80. The second number on
the card in the E-VALUES package, 33, refers to the number
of subsequent cards to be read, i. e. , the number of counties
on file.
1.8. MAP EXECUTION
The F-MAP package is the instruction and execution
package of the SYMAP program. For most of the maps in
Phases I and II only changes in the data banks and subroutine
instructions arid in certain electives of the F-MAP package were
required from map to map. In general, maps were run in
series of six or more maps at a time. The maps run and
the changes from run to run are described below. A sample
F-MAP package is shown in Table III-12.
For maps A-2 thru A-4 and A-6 thru A-8 the same
electives in the F-MAP package were used, with only the
titles being changed. These electives were numbers 1 thru 9
which were used as follows:
Elective Number Description
Elective 1 Output size--specified to produce maps
at a 26" width.
Elective 2 Extreme points--specified in terms of
map inches from an origin of 0, 0. The
vertical dimension of the source map
is 77. 7 inches and the horizontal dim-
ension is 69. 7.
-------
F-MAP
FOUR-STATE COAL REGION
PITTSBURGH BED - RAW SULFUR CONTENT (LOW)
EXPRESSED AS A PERCENT BY COUNTY
1 26.0
0.0 77.7 69.7
tv)
I—"
I
2
3
4
5
6
7
,-=+*XXHN
-=-*-= HZ
= 1-
IN
8
9
99999
0.0
9.
0.5
5.0
0.5
0.5
123456789
LOH
X
A
V
0.5 0.5 0.5 0.5 0.5
0.5 0.5
Table III-12
-------
Elective 3 Number of levels--nine for the purposes
of contouring.
Elective 4 Value Range Minimum--specified at
0.5; values below this level shown in
a unique symbolism.
Elective 5 Value Range Maximum--Specified at 5.0;
values above this maximum shown as
unique symbolism.
Elective 6 Value Range Intervals--specifies the
ranges of each of the class intervals.
Elective 7 Symbolism--user-specified symbolism
in 9 separate levels.
Elective 8 Contour Lines--suppresses the appearance
of white bands between the levels on the
map.
Elective 9 Histogram Bars--suppresses the appear-
ance of a histogram below the map.
For Phase II the same electives in the F-MAP package
were used for maps B-2 thru B-4 and B-9 thru B-ll. For
maps B-5 thru B-7 and B-12 thru B-14 elective 3 was changed
to 1 level to show the critical isoline of 0.5% and 1.0% sulfur
content; elective 7 was changed accordingly and elective 5 was
changed to 1.0.
For maps B-16 thru B-21 and C-2 thru C-7 the same
electives as in Phase I were used, with the addition of elective
35 specifying the maximum search radius at either 5 miles
(1.340 inches) or 10 miles (2.681 inches). In addition, the
outline package was removed and converted to a C-OTOLEGENDS
package for maps B-14 thru B-21. No outline existed for the
Middle Kittanning Bed.
For maps D-4 thru D-7 the same electives as in Phase
I were used, with the exception of elective 2--extreme points
-------
of (17.6,27.0) and (48.4,53.0) for D-4 and D-6 and
(17.6,40.0) and (33.0,53.0) for D-5 and D-7. For
map D-3 elective 1 was 13" and elective 2 had extreme points
of (35.9, 47..9) and (45.9, 57.6); otherwise the electives were
similar to maps B-2 thru B-4 for the left hand map and B-5
thru B-7 for the right hand map. The boxes in maps D-4 and
D-6 showing the portion of the study area to be blown up were
put in with a C-OTOLEGENDS package. The same legend was
used in the blow ups (maps D-5 and D-7) but the legend was
overriden by the map border.
The base maps—Figures A-l, A-5, B-l, B-8, B-15, and
C-l--were produced in the same manner as maps A-2, A-6, B-2,
B-l6, and C-2 with the following exceptions. First of all,
different symbolism was employed so that the area outside the
study area would be shaded and the data point locations would
stand out as black squares. Second, the number of analyses
was shown as a C-OTOLEGEND above each data point. These
legends were created by a special data handling program.
A newly created s ubroutine--Subroutine MANIP--was
used in conjunction with the user Subroutine FLEXIN to produce
maps D-l and D-2, The maps for the mode value of sulfur
content (both raw and washed) by county and by town (maps
A-3, A-7, B-3, and B-10) were run and the values for each
character cell stored on disk. These values were read in and
the two surfaces compared at locations on a grid defined by
the user in Subroutine FLEXIN; maps A-3 and B-3 were com-
pared to produce D-l and A-7 and B-10 to produce D-2. Table
IH-13 shows the Subroutine FLEXIN used. Read statements 31
and 32 refer to the data points for maps D-l and D-2 respectively,
and READ statement 40 to the values to be mapped. The same
-------
SUBROUTINE FLEX IN { I FORM , T. F I RST )
DIMENSION T(2)
LOGICAL FIRST
GO TO (31,40,32 ,40),IFORM
31 CONTINUE
IF (FIRST) CALL MAN IP(5,4,1,12,11t77.7,16.0,0.C,0.0,0)
C ************************************* *^*T(c*lfr**^*-*-**-********************
C* ARGUMENTS - IGR IDX,IGRIDY,MAPNO,ITP,ITPSKR,XYBASE,XYMAP *
C* YOMAP,XOMAP,IBACK. " *
C* IGRIDX = ACROSS, IGRIDY = DOWN GRID SPACING FOR CALCULATING FROM *
C* MAP ON UNIT 8. 1/2 INCH" GRID T5~y~X~~57 JSE~~I K. I FOR GRID *
C* MAPNO = NUMBER OF OUTPUT MAP USING SUBROUTINE - IF ONLY MAP, 0 *
C* IF FIR ST OR OTHETT~M~AlS~T. ^BTTFTAST MAP, 9. ~ ~*~
C* ITP = UNIT NUMBER OF OUTPUT MAP, USUALLY 12. *
C* ITPSKR = UNIT NUMBER OF SUBROUTINE "SCRATCH TAFE, USUALLY 11. *
C* XYBASE = X OR Y DIMENSION OF BASE MAP, IN SOURCE UNITS *
C* XYMAP = CORRESPONDING DIMENSION OF HSP'WUNTT ~B~,~ IN MAP INCHES *
C* YOMAP AND XOMAP = X AND Y DISPLACEMENT OF MAP ON UNIT 8 FROM 0-0 *
C* IBACK = 0 STANDARD. IF > OT^TUL WRITF D"UT-&ATKGRDUND CILLS~AS"~ *
C* INPUT TO GRID PROGRAM. *
READ(12) ITEST,T(2) ,T(1)
IF(ITEST.EO. 99999) IFORM=99999
RETURN
32 CONflNUE
IF(FIRST) CALL MANI P( 5 ,4,9, 12 , 11 , 77.7, 16. 0,0. 0 ,0 .0,0 )
READ(12) TTESTfT(2),T(l)
IF( ITEST.EQ. 99999) IFORM=99999
RETURN "
40 CONTINUE __ _____ _______
REWIND~T2~~
READC12I ITEST,X,Y,V1,V2
IF( ITEST.EQ. 99999) IFORM=99999
T(l)= V1-V2
RETURN
END
Table III-13
-------
outline and legends packages were used as in the original
maps; the data point and values packages were created
internally by the program. The F-MAP package employed
different electives for minimum and maximum values, number
of levels, value range intervals, and symbolism. The same
comparison procedure can be made with the GRID program as
discussed in the section on Phase III. However, for maps
of this size where the study area outline remains the same,
the use of the subroutine in SYMAP requires less programming--
only one to two additional cards in Subroutine FLEXIN.
This subroutine is described in section 4, Program Development.
-------
PHASE III: ALLEGANY AND GARRETT COUNTIES
Phase III of the project dealt with a two county area
in Maryland, chosen because of the relatively good data
available for sulfur content, bed thickness, and washabiiity
characteristics. In Stage One of this phase maps were
produced with the SYMAP program showing sulfur content
and bed thickness by seam for the Upper Freeport and Upper
Baker stown Beds. Maps of quantity of coal in the ground by
seam were made using the GRID program.
In Stage Two maps were produced only for the Upper
Freeport Bed. Maps showing the sulfur content to which coal could
be washed --using various screen thicknesses for washing--
at certain yields were made with the SYMAP program. The
GRID program was used to map the quantity of coal that could
be obtained at these yields as well as the difference in sulfur
content depending upon the screen size used for washing.
The following sections describe the work that was
done to prepare the data for mapping, including base map
preparation, data bank set up and use, and decisions as to
the actual running of the maps.
-------
1. A-OUTLINE--specif led the same as in Phases
I and II.
2. B-DATA POINTS--the UTM coordinates for
mine or sample location for the data on sulfur content<
bed thickness, or washabiiity; specified the same as
in Phases I and II.
3. C-LEGENDS--specified the same as in Phases
I and II.
4. C-OTOLEGENDS--specified the same as in
Phases I and II.
5. E -VALUES--specified the same, in data banks,
as in Phases I and II.
6. F-MAP--specified the same as in Phases I
and II.
The user subroutine FLEXIN, was used in the same way to
read in the appropriate values from the data bank (E-VALUES
package) and to transform the coordinates (B-DATA POINTS
package) from the UTM system to source map inches.
The corresponding instructions for Phase III with the
GRID program were as follows:
1. Outline--the GRID program reads in a
rectangular matrix of values created by SYMAP; Subroutine
FLEXIN of GRID is used to check each cell to see if it is
outside the study area. The user does not need to code any
outline package.
-------
-------
the symbolism for background; a new outline for
the area to be mapped is therefore created by the
test for sulfur content. No separate values package
is needed.
b. Map package--title cards and electives are
coded in the same way as with the SYMAP program;
many of the electives are exactly the same.
From the preceeding discussion it should be evident
that once the SYMAPs had been generated and the matrices
of values stored in data banks, it was necessary to specify
only the user subroutine and the MAP package to produce
maps using the GRID program.
2. 2. BASE MAP PREPARATION
Maps of both Garrett and Aliegany counties were
obtained from the Maryland Geological Service at a scale of
1 to 6Z500 or approximately 1" = 1 mile. The map of
Garrett County showed more detailed information pertaining
to the coal beds than that of Aliegany County. The Garrett
County map (reprinted 1965 on 1959 base) shows in section
and in diagram the relative composition of the various geologic
formations. It also shows the outcrop line of each of the
coal seams existing in the area, along with symbols keyed
to the names of the various beds. This detailed information
is lacking on the geological map of Aliegany County (I95b).
It was therefore difficult to define the extent of the outcrop
lines for both the Upper Freeport and Upper Bakerstown
beds.
-------
A second series of maps at i" = Z miles was
obtained showing the various outcrops and mineable reserves
in the coal seams. This is a series o± maps that had
been prepared by John T. Boyd & Assoc. , Mining Engineers,
March 1964. These maps show well-defined outlines for the
strip and deep coal reserves, including the deep reserve limit
for the Upper Bakerstown and the Upper Freeport seams.
They also show the portion of each seam greater than Z8"
thick.
It was decided to use the Boyd maps as a basis for
determining the outlines for the computer maps for the two
coal seams under study, since they were more complete.
However, the clarity of information at this scale was poor.
Accordingly, the Boyd maps were enlarged and the information
from them transferred to a composite map at a scale of
1" = 1 mile. In this way information on the Garrett and
Ailegany County maps (also at I" =1 mile) could be used
and the scale of the base map would be such that portions
could be enlarged without losing detail on the final computer
maps.
The enlarging was done using a double grid technique.
This process consisted of producing gridded overlays for both
the small map (l" = Zmiies) and the large map (1" = 1 mile).
These overlays were directly proportional to the scale
differences between these two source maps. Using the horizontal
and vertical lines of the grids, the outline was translered quite
easily from one scale to the other. Likewise, information was
taken from the larger scale so as to produce composite overlay
outlines for the Upper Freeport and Upper Bakerstown Coal
seams. Periodic adjustments had to be made so that the
information would conform to the general outlines of the county
geological maps.
-------
A separate overlay was produced lor each oi the
ioliowing:
-- 1. Upper Freeport total outcrop;
-- £. Upper Freeport area over Z8" thick;
-- 3. Upper Bakerstown total outcrop;
--4. Upper Bakerstown area over Z8" thick.
These four overlays served as the source maps for developing
the computer base map outline packages. (Overlays for four
or five other beds were produced, but these were discarded
because sufficient data was lacking for these beds. )
Each of the contiguous areas representing part of a
particular coal seam is known as an island, which must be
described in terms of the x-y coordinates that define the
boundary. These x-y coordinates can be recorded by using
a digitizer as in Phases I and II or by hand. The latter pro-
cedure was used here based on the relatively small number
of vertices to be recorded.
This procedure involved the use of special graph pape:
with the x-y coordinates printed along one side and the top.
The overlay was placed on top of the graph paper and the
coordinates for each point were written on coding forms,
proceeding from point to point in a clockwise manner, and
repeating the starting point to 'close the outline'. This
information is then punched onto IBM cards--one pair of co-
ordinates per card. Each island is a separate deck with the
decks for ail islands forming the A-OUTLINE package. These
packages were prepared for each of the four computer base
maps.
-------
Each of the base maps was run and any discrepancies
with the source maps were noted and corrected. When the
Upper Freeport total outcrop base map was tested together
with the data point locations, it was found that the 20 data
point locations all fell in the two main islands. No data
points existed for the island which was in the upper left hand
corner of the study area as well as for the small islands on
the westernmost side of the study area. It was felt that any
information interpolated from the data points to these islands
would be invalid and misleading. Therefore, these islands
were removed from the outline package and only the two areas
shown in Figure E-l were used for mapping. For the Upper
Bakerstown total outcrop base map the data point locations
were reasonably well distributed for all islands; accordingly
no islands were removed from the outline package. (See
Figure E-5)
The two base maps for the areas over 28" thick were
prepared directly from the Boyd source maps and are shown
in Figures E-l and E-5. During the testing of these outlines
together with the appropriate data points, it became apparent
that the location of data points did not correlate very well
with the outline defining the areas 28" and greater in thickness;
many of the data points fell outside these outlines. However,
all of these areas were included if they had been used for the
total outcrop map.
The size of the area for mapping in Stage Two for the
Upper Freeport Bed was determined by the location of the
values for washability. Most of the data points fell within
the center portion of the two-county study area used for Stage
One. Therefore, it was decided to map only this center area.
This area was defined for mapping by changing Elective 2 in
-------
SYMAP. Figure E-l shows the relationship of the center
area to that of the total two county study area. Figure
G-l shows the base map for the Stage Two area. No changes
in the A-OUTLINE packages were necessary to produce
base maps at this smaller scale.
Four separate legend packages were developed, two
for each stage of the mapping. The C-OTOLEGENDS packages
for each stage contained identical information, but the
C-LEGENDS packages differed according to the scale of the
output map. The information provided in each of the
C-OTOLEGENDS packages consisted of the state outline show-
ing the border between Maryland and West Virginia and the
county border separating Allegany and Garrett Counties. In
the C-LEGENDS packages point legends were used for the
names: Maryland, West Virginia, Allegany County, Garrett
County; the legend block explaining the symbolism used to
delineate county and state outlines; and the graphic scale.
The coordinates of the various outlines were located
on the source map in the same manner as for the outline packages.
Slight adjustments were made to produce a high quality, single
character line defining the state and county boundaries. This
was necessitated by the reduction in size from the original
scale of the source map to the size of the final computer map; a
clustering of symbols occured because of the reduction in size .
To eliminate this problem the x-y coordinates for the points
to be removed were recorded from the output map; a blank
symbol was assigned to each of these points, in effect
removing the unwanted symbols. These blank symbols were
coded as a part of each of the C-LEGENDS packages, but
-------
their effect was to override instructions specified in the
C-OTOLEGENDS package. This allowed us to develop a
pleasing and graphically consistent legend.
With the exception of these blank symbols the uses
of the C-LEGENDS and C-OTOLEGENDS packages were the
same as in Phases I and II.
2. 3. PHASE III DATA BANKS
For Stage One the data banks included sulfur content
for both the Upper Freeport and Upper Bakerstown beds
as shown in Tables 111-14 and 111-16. The values shown
are the original data provided to us by the Bureau of Mines.
The 20 locations for the Upper Freeport Bed included two
taken from the Lower Freeport Bed; all locations for Bakers -
town coal were in the Upper Bakerstown Bed. As in Phases
I and II information on number of analyses, low, mode, and
high values for sulfur content were included in the data bank.
Although the number of analyses at a location was as high
as 78, it was generally 1* accordingly, only the mode values
were mapped.
The data banks and data point locations for bed thickness
(Stage One) for both beds are shown in Tables 111-15 and 111-17,
and for washability (Stage Two) for the Upper Freeport bed in
Table IH-18.
2.4. PHASE in DATA POINT PLACEMENT
All of the data points for both the Upper Freeport and
Upper Bakerstown beds and for sulfur content, bed thickness,
and washability were provided in terms of UTM coordinates.
The coordinates corresponded to mine or sample location,
or to the best estimate when the coordinates were not available.
Often the estimated location would be the nearest town. A
procedure was developed to convert these points from the
-------
8— DATA
1
TS
20
(Jl
I
99999
E-VALUES
2
2.
1.
1.
1.
15.
1.
78.
1 •
1 •
2*
3.
1 .
1.
1.
1.
31.
17.
1.
1 .
1 .
20
2.0
3.7
0.5
1.1
1.2
0.8
0.9
1.1
2.2
i.o
2.7
3.1
2.5
0.7
0.5
0.7
1.8
2.1
1.4
1.8
3769.
3726.
3550.
3537.
3961.
3957.
3953.
3961.
3957.
3953.
3900.
3851.
3623.
3645.
3520.
3407.
3608.
3662.
3969.
3957.
X
2.0 2.3
3.7 3.7
0.5 0.5
1.1 1.1
1.3 1.5
0.8 0.8
1.1 1.7
1.1 !•!
2.2 2«2
1.0 1.3
2.7 5.8
3.1 3.1
2.5 2.5
0.7 0.7
0.5 0.5
1.2 2.2
1.9 2.4
2.1 2.1
1.4 1*4
1.8 1*8
6614.
6604.
6440.
6390.
6558.
6560.
6558.
6566.
6550.
6566.
6558.
6558.
6574.
6531.
6470.
6311.
6548.
6564.
6848.
6581.
99999
071R 01
Q71R 02
071R 03
Q71R 04
071R 05
071R 06
071R 07
071R 08
071R 09
071R 10
071R 11
071R 12
071R 13
Q71R 14
071R 15
Q71R 16
071R 17
071R 18
074R 19
074R 20
071R 01
071R 02
071R 03
071R 04
071R 05
071R 06
071R 07
071R 08
071R 09
071R 10
071R 11
071R 12
071R 13
071R 14
071R 15
071R 16
071R 17
071R 18
074R 19
074R 20
Table III-14
Upper Freeport Bed - Sulfur Content
02301550197
023Q1550248
02304200454
023069500^0
02307000039
02307000067
02307000270
02307000272
02307000464
02307000690
02308330488
02308330489
02308700258
023087Q0491
02310400769
0231348Q370
02314290877
02315270660
00100750570
02307000112
02301550197
02301550248
02304200454
02306950020
02307000039
02307000067
02307000270
02307000272
02307000464
02307000690
02308330488
02308330489
02308700258
02308700491
02310400769
02313480370
02314290877
02315270660
00100750570
-------
B-DATA POINTS
1 7
99999
E-VALUpS
2
99999
3707.
3603.
3958.
3649.
3952.
3814.
3620.
37.
48.
30.
38.
43.
40.
48.
6383.
6473.
6580.
6571.
6317.
6345.
6538.
01
02
03
04
05
06
07
01
02
03
04
05
06
07
Table III-15
-------
B-OATA POINTS
99999
E-VALUES
2
1.0
2.0
1.0
1.0
1.0
1.0
1.0
99999
7
0.6
1.3
1.8
3.7
0.6
3.3
3.3
Key: Number, low
i
t\>
u>
3961 .
3961.
3819.
3558.
3579.
3880.
3869.
X
0.6 0.6
1.3 1.7
1.8 1*8
3.7 3.7
0.6 0.6
3.3 3.3
3.3 3.3
. mode, high
6817.
6817.
6523.
6467.
6505.
6558.
6558.
062R 01
062R 02
062R 03
062R 04
062R 05
062R 06
062R 07
2
3
4
5
6
7
00110700423
00110700460
02301450042
02304200454
02304200580
02308330136
02308330333
423
460
042
454
580
136
333
Table III-16
-------
POINTS X
16
3939.
3960.
3810.
3498.
3633.
3591.
3680.
3591.
3645.
3591.
3742.
3729.
3612.
3794.
3761.
3652.
6856.
6818.
6510.
6368.
6524.
6476.
6567.
6476.
6531.
6476.
6368.
6372.
6566.
6515.
6660.
6570.
99999
E-VALUES
2 16
OJ
00
38.
37.
38.
52.
35.
30.
42.
48.
35.
30.
48.
45.
35.
18.
30.
38.
01
02
03
04
05
06
07
08
09
10
11
12
13
14
15
16
1
02
03
04
05
06
07
08
09
10
11
12
13
14
15
16
99999
Table 111-17
Upper Baker slow n Bed
-------
B-OATA POINTS
3939. 6856.
3705. 6652.
3603. 6473.
3633. 6532.
3652. 6570.
E-VALUES X
2 5
60 43939 6856 1400 1200 2.79 1
60 43705 6652 1300 1200 3.67
60 43603 6473 2750 2600 3.83 1
60 43633 6532 2750 2600 4.25
60 43652 6570 1970 1938 3.0^ 1
99999
.41
.76
• 18
• 60
.05
1.49
.76
1.24
.60
1.10
1.59
.75
1.29
.61
1.16
1*69
• 78
1.49
• 62
1.30
01
02
03
04
05
.59
• 16
.40
.05
• 16
.70
• 16
• 53
• 07
• 29
• 81
• 16
.69
.08
• 46
• 93
• 20
• 89
• 10
• 68
Key:
x
50 60 70 80
(yield) total sulfur
50 60 70 80
(yield)pyritic sulfur
B-OATA POINTS
1 5
99999
E-VALUES
2 5
60 £>3939
60 43705
60 43603
60 43633
60 43652
3939.
3705.
3603.
3633.
3652.
X
6856 1400
6652 1300
6473 2750
6532 2750
6570 1970
6856.
6652.
6473.
6532.
6570.
1200 2.79
1200 3.67
2600 3.83
2600 4.25
1938 3.04
01
02
03
04
05
0.00 2*03 2*13 2*19 0.00
0.08
1.30 1.40 1.47
0.12 0.00 0.24
0.74 0.74 0.00 0.83
1.15 1.30 0.00 1.72 0.47 0.63 0.00 1.05
0.64 0.66 0.69 0*00 0*12 0.16 0.19 0*00
1.00 1.15 0.00 1.57 0.37 0.53 0.00 0.95
99999
Table III-18
-------
UTM coordinates to the coordinates of the Phase III
output maps. An algorithm similar to the one used in Phase
II was used for both Stage One and Stage Two.
A separate B-DATA POINTS package (and correspond—^;
E-VALUES package) was prepared for each computer run.
This consisted of a separate card for each data point to be
mapped. On each card were punched the UTM coordinates of
the point, a reference number, and a code related to the mine
or sample location.
For both the Upper Freeport and Upper Bakerstown
coal seams there were two superimposed data points for
the sulfur content values. For the thickness and washability
information, there were no superimposed data points; a super-
imposed data point corresponding to a top bench value was
removed from the washability data bank. Table III-1V shows
each of the data banks, the number of data points, and the
number of superimposed data points.
Several of the seams that were to have been mapped
could not be meaningfully analyzed due to the high number oi
superimposed data points; the effective number of data points
was too few for mapping. The SYMAP program handles super-
imposed data points by summing and calculating a mean for
ail values that are located at the same point; thus, a new value
is calculated for use in the interpolation procedures. Taking
the mean of two values that are themselves modal values based
on different numbers of analyses may well skew the results in
this case.
-------
DATA BANK ORIGINAL # DATA POINTS TOTAL # EFFECTIVE
SUPERIMPOSED FOR MAPPING
Upper Freeport - Sulfur Content 20 2 19
Upper Freeport - Thickness 7 07
Upper Bakerstown - Sulfur Content 7 26
Upper Bakerstown - Thickness 16 0 16
Upper Freeport - Washability 5 05
ro
Table III-19
-------
2.5. SUBROUTINE FLEXIN
Subroutine FLEXIN is used as a part of both the
SYMAP and GRID programs. FLEXIN was used in the
SYMAP program to:
1. read in the data points and convert them
from UTM coordinates to source map inches. The
algorithms following READ Statement 1 in Table III-ZU
(a sample SYMAP FLEXIN for Phase HI) show the
conversion made for both Stage One and Stage Two;
and,
Z. read in all the values for the variables in the
appropriate data bank; these values are written out
onto a scratch disk file; for each map in a run, the
file is rewound and read and the appropriate variable
is returned to the main program for mapping.
This form of FLEXIN was used for maps E-Z thru E-4, E-6
thru E-8, E-9 thru E-ll, F-l and F-Z, G-Z thru G-5, and
G-8 thru G-10.
FLEXIN was used in the GRID program in a slightly
different way. When maps were run with SYMAP the matrices
of map values were written out onto a disk to create two data
banks--one containing all the SYMAPs for Stage One and the
other containing all the SYMAPs for Stage Two. These were
stored as real values and the entire data bank read in by
Subroutine FLEXIN of GRID. Table III-Z1 shows the Sub-
routine for Stage One.
In most cases FLEXIN was used in GRID to read in
values for sulfur content and thickness; if the cell were inside
-------
SUBROUTINE FLEX IN(I FORM,T,FIRST)
DIMENSION T(3) , SK)
LOGICAL FIRST
GO TO (1, 10,20,30),IFORM
1 READ (5,900) T( 1),T(2)
900 FORMAT (10X.2F10.2)
T(l)=(3986.-T<1))716. 0428
T(2)=(T(2)-6285.)/16.0428
RETURN
10 IF(FIRST) REWIND 11
READ (5,100) S,T(2),T(3)
100 FORMAT (4F5.0,52X,2A4)
WRITE (11,101) S,T(2),T(3)
101 FORMAT (4F5.1.2A4)
T(l)=S(2)
RETURN
23 IF (FIRST) REWIND ll
READ ( 11,200) T
200 FORMAT (1 OX,F5.1,5X,2A4)
RETURN
30 IF (FIRST) REWIND 11
READ (11,300) T
300 FORMAT (1 5X, P5. 1, 2A4)
RETURN
END
Table III-20
Subroutine FLEXIN - SYMAP
-------
SUBROUTINE FLEX I N( I FCRy , T , F I RST )
LOGICAL FIRST
DIMENSION Z(16),IZ(16)
EQUIVALENCE(Zd) ,IZ(1 ))
READ (12) Z
GO TO (1 ,2t3t4,5,6) , IFCPM
CONTINUE
IF(Z(2).GE.l .0) GO TO 10
IF(IZ(7). EQ.25) GO TO 10
T=1676.0*(Z(7)/12.0)
RETURN
CONTINUE
IF(Z(5) .GE.1.0) GO TO 10
IF(IZ(8).EQ.25) GO TO 10
T=1676.0*(Z<8)/12.0)
RETURN
CONTINUE
IF(Z( 10) .GE.1.0) GO TO 10
IFUZU5I.EQ.25) GO TO 10
T=1676.0*(Z( 151/12.0)
RETURN
CONTINUE
IF(Z(13) .GE.1.0) GO TC 10
IF( IZ(16).EQ.25) GO TO 10
T=1676.0*(Z(16)/12.0)
RETURN
CONTINUE
IF(Z(2) .GE.1.0. AND. Z( 10). GE.1.0) GO TO 10
IFUZI7) .EQ.25. AND. IZ(15) .EQ.25) GO TO 10
A=1676.0*(Z<7)/12.0)
6=1676. 0*
-------
the study area and the sulfur content was less than 1%,
then a value for quantity would be returned to the main program
for mapping. The value would be proportional to the bed
thickness and was calculated as follows:
Q = 1676. U x T x M x Y
TZTO
where: Q = quantity in tons per acre
T = thickness in inches
M = adjustment for recovery:
1. 00 for reserves and
0. 50 for amount mineable.
Y = yield, varying from 0.50 thru
0. 80 for the washability data, and
1.00 for reserves.
Since the scale of the map is known and the area in
acres of each grid cei.1 can be determined, only a few statements
would need to be added to the Subroutine to calculate the total
quantity of coal. This would involve summing the quantity in
tons per acre for all cells being mapped and multiplying the
sum times the acres per grid cell. An approximate answer
can be obtained directly from the printout accompanying the
maps. Below the chart showing the symbolism the number of
cells is printed for each level. For any level, the number
of cells can be multiplied by the average quantity in tons per
acre for that level and the number of acres per grid cell.
The results can be summed for all levels to give the total
quantity of coal. Maps F-3 thru F-5 and H-l thru H-5 were
produced using this type of FLEXIN.
-------
Subroutine FLEXIN was used in GRID in another way.
For Maps G-6, G-7, and H-7 values were read in, the check
for background cells was made (cells outside the study area),
and the difference in values calculated. This is similar to
the use of Subroutine MANIP in Phase II.
-------
Z. 6. MAP EXECUTION
For most oi the maps in Phase III only changes in
the data banks and subroutine instructions and in certain
eiectives of the F-MAP or MAP package were required from
map to map. The decks were set up so that a series of maps
could be produced in one run.
Most maps employed the same eiectives in the
F-MAP package for SYMAP and the MAP package for GRID:
Elective SYMAP GRID
1. 13" 100 rows by 12V columns
for Stage One;
76 rows by 1Z9 columns for
Stage Two.
Z. extreme points IFORM number related to
from source the execution statements
map in FLEXIN
3. nine levels nine levels
4,5,6,7, same way as Same way as Phase I & II
Phase I & II
8. same way as (not used. )
Phase I & II
Maps E-Z thru E-4, E-6 thru E-8, E-9 thru E-ll, F-l and
F-Z, G-Z thru G-5, and G-8 thru G-10 were all run with
SYMAP with changes occurring only in eiectives 4 and 5, and
in elective 1 from Stage One to Stage Two. Maps F-3 thru
F-b in Stage One and maps G-6, G-7 and H-l thru H-7 in
Stage Two were run with GRID. Changes occured in eiectives
Z, 4, and 5, and in elective 1 from Stage One to Stage Two.
-------
Maps E-l, E-5, and G-l were base maps; these were run
with special symbolism and used only a 'dummy' subroutine.
-------
2- 7. INPUT TO SYMVU
Two three-dimensional views were produced for
each of four surfaces using the SYMVU program. These
surfaces had been produced by SYMAP and the matrix
of values stored on disk files for later use by the GRID
program; they are stored and read in by SYMVU in a
similar way. No changes were necessary in the present
version of the SYMVU program to produce these eight
plots.
The user of SYMVU must specify a number of
electives, similar in nature to the electives of the F-MAP
or MAP packages of SYMAP and GRID. These are ail
contained on two control cards which follow the title care].
Table 111-22 shows these three cards for the two views
of the surface of thickness for the Upper Freeport Bed
for areas over 28" thick (the right hand map in Figure F 1)
For example the '100' and '129' at the beginning of card
2 specify the number of rows and columns, respectively,
on the matrix of values and, correspondingly, on the
SYMAP-produced map. If we examine Figure F-l we will
find that the numbers on the borders show us that there
are, indeed, 100 rows and 129 columns. These numbers
were specified according to the instructions for electives
2-1 and 2-2, respectively; the other numbers on cards 2
and 3 were specified according to similar instructions in
the computer program manual.
The next 31 cards are used to specify the north
arrow which is composed of 31 small five-pointed stars.
At the present time all legends to be shown on the plot
-------
must be punched up separately with a card for each
character cell. In Table 111-22 the first and second
integer (92 and 110, respectively) on card 4 are the
row and column locations as they would be measured on
the SYMAP-produced map; in this case a new legend was
created for SYMVU so these points were not measured
from the map in Figure F-l. The integer number,
14, is a code for the five-pointed star symbol.
The last three cards are the title card and two
control cards for the second of the two views for this
surface. The card deck for the views of the other three
surfaces were exactly the same, except for the two title
cards, and the elective specifying the file number of the
data set.
-------
I
ro
On
U. FREEPORT
100 129
60. C .
92
93
93
93
94
94
94
94
94
95
95
95
95
95
95
95
96
96
96
96
96
96
96
96
96
97
97
97
98
98
98
U. FREEPQRT
100 129
45. 309.
BED, THICKNESS
2 0
6. 3.
lie
109
110
111
108
109
110
111
112
107
108
109
lie
111
112
113
106
107
108
109
110
111
112
113
114
109
110
111
109
110
111
BED, THICK. NESS
4 1
6. 2.
JF AREAS >28" ,GARRE TTG A L L EGA NY C OUNTY , MARYLAI
0 81 0 31 4
.04
14
14
14
14
14
14
14
14
14
14
l^
14
14
14
14
14
14
14
14
14
14
14
14
14
14
14
14
14
14
14
14
OF AREAS >28",GARRETT£ALLEGANY COUNTY ,MARYLAf
2 81 3 31 4
28.0 .04
Table III-Z2
-------
3.
GENERAL REFERENCES
3.1. APPALACHIAN COAL REGION REFERENCES
U.S. Bureau of Mines Report of Investigations #
"Estimate of Known Recoverable Reserves of Coking Coal
in County"; for the following counties
Report #
4970
5003
4807
5143
4757
4998
5109
4803
5160
5267
5207
5233
5171
maps for the
State
MD
PA
PA
PA
PA
PA
PA
PA
WV
wv
WV
wv
wv
Pitt sburgh Bei
County
Allegany
Allegheny
Fayette
Greene
Indiana
Somerset
Washington
Westmoreland
Brooke
Marion
Marshall
Monogalia
Ohio
Unpublished current data frQ>m U.S. Bureau of Mines
data banks.
-------
3. 2. ALLEGANY AND GARRETT COUNTY REFERENCES
'Amsden, T.W., 1953, Geologic Map of Garrett County,
Maryland Geological Survey. (Reprinted 1965). Scale 1:62,500.
'Anon. , 1962, Maryland Coal Reserves, Maryland Bureau of
Mines, Westernport, Maryland. Mimeo.
'Anon., 1967, Calendar Year 1967 forty-fifth annual report of
the Maryland Bureau of Mines. Westernport, Maryland.
'Anon. , 1968, List of Publications of the Maryland Geological
Survey, Latrobe Hall, Johns Hopkins University, Baltimore,
Maryland 21218.
'Berryhill, H. L. Jr., G. W. Colton, W. de Wilt, Jr., J. E.
Johnston, 1965, Geologic Map of Allegany County, U.S.
Geological Survey and State of Maryland Dept. of Geology, Mines,
and Water Resources. Scale 1:62,500.
"Boyd, J. T. and Associates, 1964, Remaining Coal Reserves
of Allegany and Garrett Counties, Maryland, Report issued by
the company to the State of Maryland. 27p. plus maps.
"Crentz, W. L. , and T. Fraser, 1947, "Preparation Character-
istics of Maryland Coals," U.S. Bureau of Mines Technical
Paper 701.
'Snyder, N. H. , and S- J. Aresco, 1953, "Analyses of Tipple
and Delivered Samples of Coal (collected during the fiscal years
1948 to 1950 inclusive)," U.S. Bureau of Mines Bulletin 516.
-------
'Toenges, A. L. , L, A. Turnbull, L. Williams, H. L.
Smith, H. J. O'Donnell, H. M. Cooper, R. F. Abernethy,
and K. Waage, 1952, "Investigation of Lower Coal Beds in
Georges Creek and North Part of Upper Potomac Basins,
Allegany and Garrett Counties, Md. , " U.S. Bureau of Mines
Technical Paper 725.
'Toenges, A. L. , et al. , 1952, "Castleman Basin, Garrett
County, Md. , " U.S. Bureau of Mines Bulletin 507.
-------
3. 3. OTHER REFERENCES
'Beaumont, Edward C. , 1963 (August), "A Procedure for
Determining Strippable Coal Reserves11, reprinted from Coal
Age.
' Blaylock, D.W , et al. , 1955, "Estimate of Known Recoverable
Reserves of Coking Coal in Clearfield County, Pennsylvania, "
U.S. Bureau of Mines Report of Investigations #5166.
'Crentz, W. L. , A. L. Bailey, and J. W. Miller; 1952,
"Preparation Characteristics of Coal from Clearfield County,
Pa. , " U.S. Bureau of Mines Report of Investigations #4894
"DeCarlo, Joseph A., E. T. Sheridan, and Z. E. Murphy,
1966, "Sulfur Content of United States Coals, " U.S. Bureau
of Mines Information Circular #8312
'Deurbrouck, A. W. , and E. R. Palowitch, 1966, "Survey of
Sulfur Reduction in Appalachian Region Coals by Stage Crushing, "
U.S. Bureau of Mines Information Circular 8282.
'Edmunds, Wm. E. , 1968, "Geology and Mineral Resources of
the Northern Half of the Houtzdale 15-Minute Quadrangle, Pa.,"
Pennsylvania Geological Survey Bulletin A85ab
"Englund, K. J. , 1969, "Availability of Low-Sulfur Coal in
Wyoming County, West Virginia, " U.S. Geological Survey
Administrative Report.
'Gluskoter, Harold J. and J. A. Simon, 1968 "Sulfur in Illinois
Coals," Illinois State Geological Survey, Urbana, Illinois
-------
4- PROGRAM DEVELOPMENT
The computer mapping programs as used at the
Laboratory for Computer Graphics and Spatial Analysis are
constantly undergoing revision and improvement. Accordingly,
it is important to specify which version of each program was
used for the project:
SYMAP Version 5. 15
GRID Version 3
SYMVU Version 1. 0.
A number of significant changes in the SYMAP and
GRID programs were made during the project to facilitate
the mapping of the coal data, particularly data having to
do with quantities of coal. In the SYMAP program changes
were made in the printout of the textual information below
the map. In addition, a new subroutine was written to be
used in conjunction with the normal user subroutine for com-
paring surfaces having the same study area outline. Since
the GRID program can generally be used more efficiently for
comparison of surfaces, this SYMAP subroutine is only useful
for the user who wishes to perform the comparison with one
program and one submission.
The major changes to the GRID program involved the
rewriting of the legends routine that had been included in
previous versions of the program run on the IBM7094. This
enables the same kind of legends used with the SYMAP program
to be displayed on a map produced by GRID. Formally, it
was not possible to include legends with the GRID program.
With the inclusion of these changes, the two programs are now
completely compatible in terms of graphic output.
-------
The changes made to both the SYMAP and GRID
programs were not incorporated into the existing standard
versions of the programs. Instead, the specific changes to
the standard programs necessary to duplicate the work of
this project are documented separately, as requested.
A number of other important changes in the mapping
procedure were undertaken as part of the program development;
they do not specifically involve changes to the mapping programs.
These include:
1. The use of statistical routines incorporated in
user-oriented programs to aid in the choosing of the
levels or contour intervals for mapping.
2. Data handling procedures used to facilitate the
mapping particularly when a large series of maps is
involved.
3. The use of a simple FORTRAN program to
create legends showing the number of analyses associated
with each data point. These legends can be displayed
directly on the base map.
Most of the material in the following s ections is of
a technical nature aimed at the computer programmer interested
in using the programs.
4. 1. SYMAP PRINT OUT STATEMENTS
Two changes were made in the way that textual infor-
mation following the maps is printed out by the SYMAP program.
-------
Since the clock subroutine is presently inoperable for runs on
the IBM 360-65 at the Harvard Computing Center all state-
ments specifying TIME= were deleted from the printout.
Secondly, the way in which the ranges of data values assigned
to each of the data value levels has been specified in the past
has been unclear to many users of the program. Therefore,
the lines stating "absolute value range applying to each level,
(maximum included in highest levels only)", were deleted from the
printout. The explanation of the value ranges specified which
follows this statement was changed as follows:
Value range changed from
level 1. 2.
Minimum 1. 00 2. 00
Maximum 2. 00 3. 00 etc.
to
level 1. 2.
Minimum 1.00 2.00
Maximum 1.99 2.99 etc.
In the normal version of the program, it was not clear which
level the value 2. 00 would be assigned to; in the new print-
out of the level ranges, this confusion has been eliminated.
The procedure for specifying the portion of the value
range to be included in each level in elective 6 remained
unchanged; therefore, these minor program modifications have
no effect on the way the user inputs his information. They
only affect the clarity of his output.
-------
These changes were made only for the Phase I
and II Appalachian Coal Region maps; the standard version
of SYMAP was used for the Phase III Allegany and Garrett
County maps.
-------
4.2. SYMAP SUBROUTINE MANIP
As discussed in later sections of the report, data
for sulfur content for the Pittsburgh Bed was mapped in a
four state area of the Appalachian Region. In one series of
maps the sulfur content data was aggregated and mapped by
counties; in a second series of maps, the data was aggregated
and mapped by towns. The surfaces resulting from these
two data aggregation procedures were compared to determine
what scale of aggregation would be most useful, in relation
to the time involved in obtaining and manipulating the data.
To compare the two surfaces directly in one submission
with the SYMAP program required the writing of a new sub-
routine entitled Subroutine MANIP. MANIP is called by the
normal user subroutine FLEXIN. The parameters to be
specified are clearly spelled out in comment statements
contained in subroutine FLEXIN; these include the size and
form of the surfaces to be compared, the desired size and
form of the output map, and the spacing of the grid on which
the comparison will be made. Subroutine FLEXIN as used
with Subroutine MANIP is shown in Table III-13.
Normally, comparison is not made for each character
cell as is done with the GRID program; this is an expense-
saving procedure. The use of subroutine MANIP with SYMAP
is therefore advisable for comparing large maps where accuracy
at each character cell is not necessary. On the other hand,
when small maps are being compared and accuracy at each cell
is important--as with the determination of the quantity of coal
in later phases of the mapping--the use of the GRID program
is advisable for surface comparison.
-------
4. 3. LEGENDS WITH GRID
Earlier versions of the GRID program--run on the
IBM 7094 at the Harvard Computing Center--had provided
for a legend package on cards and tape which would allow
the inclusion of legends similar to those used with SYMAP.
Considerable programming was necessary to make these
changes compatible with the current versions of SYMAP and
GRID run on the IBM 360-65.
The final solution involved the following changes from
the current versions of the programs:
(i) changes in one subroutine of SYMAP so that
all legends appearing on the output map are written
out by row and column onto a file stored on disk; and,
(ii) changes in one subroutine in GRID so that these
legends can be read in by rows and columns and shown
on the map output by GRID in the same way they were
shown by SYMAP. The maps must be for the same
area and at the same size*
For the user the only considerations are using the
versions of SYMAP and GRID edited to perform this task.
First of all, the special version of SYMAP is used with only
the legends and map packages. In the F-MAP package electives
1 and 2 should be specified as usual, elective 3 as one level,
and elective 7 with some symbol not used in the legends for both
the level and background symbolism. The legends package should be
-------
specified as usual. This separate run of SYMAP will create
the necessary legends file on disk.
The edited version of the GRID program is used,
for mapping. No other changes are necessary from a normal
run with the GRID program.
4.4. USE OF STATISTICAL ROUTINES
A user-oriented statistical program called SPSS
(Statistical Package for the Social Sciences) was initially used
to aid in the choosing of contour intervals for mapping. This
package program can compute the mean, median, mode, variance
standard deviation, skewness, range, minimum and maximum of
the data. In particular, the user can request a frequency dis-
tribution of the data values.
Too often the user of SYMAP does not know the
frequency distribution of the data to be mapped; consequent.'.y
several test maps are made before the optimum value range
intervals are determined for the final maps. By using a
statistical program such as SPSS to determine the value range
intervals, considerable time and money can be saved.
SPSS was used to determine the frequency distribution
of all the sulfur content data for Phases I and II. Tentative
value range intervals were set accordingly. As a result of
review with APCO and the Bureau of Mines, two level break-
downs were selected for use with the Phase I and II maps:
1. nine equal value range intervals from 0.5% to
5. 0% sulfur, and
-------
2. critical value level breaks at 0. 5% and 1%
sulfur; this serves to highlight those areas of the
study area with less than 1%, or less than 0.5%,
sulfur content coal.
The number of data points available for mapping in
Phase III did not warrant the use of SPSS to determine
frequency distributions. The same level breakdown was used
for sulfur content as in Phases I and II; level breakdowns
for bed thickness and quantity were determined by inspection
of the data, as were the levels for the washability sulfur content
data.
For any project where a large volume of data is to
be mapped and several value ranges exist, a program such
as SPSS used prior to mapping will prove very useful and
time saving.
4.5. DATA HANDLING PROCEDURES
The data handling procedures used for Phases I and
II were mainly an extension of prior procedures. Considerable
time was spend setting up separate computer decks for each
run or series of maps. This was done primarily to keep ail
of the component packages needed to make specific runs in
their proper order. The mapping could have been done by
using one basic set of A-OUTLINE, C-OTOLEGEND, and
C-LEGEND packages for all runs and interchanging the appro-
priate B-DATA POINTS, E-VALUES and F-MAP packages.
However, experience has shown us that it is worth the extra
effort at the beginning of a project to duplicate outline and
legend packages and set up separate decks for each series of
maps to be run. Otherwise, considerable confusion can occur
from constantly switching and alternating the various packages.
-------
When maps for a study area are to be run continuously
over a long period of time it would be advisable to store the
outline and legends on a tape or disk file, store the data banks
of data points and values on other files, and submit card decks
for each run which contain only the map package, user
subroutine, and job control language.
In Phase I one series of maps run was the sulfur
content for washed' samples for the Pittsburgh Bed. The deck
set-up to produce the three maps for low, mode, and high
values was as follows:
1. job control language.
2. user subroutine FLEXIN, used to read in the
appropriate values for mapping.
3. outline package.
4. data points package, containing the x-y locations
for the washed samples.
5. legends packages.
6. data bank or values packages, containing the
data for the low, mode, and high values for each
county.
7. map execution packages.
In this case all data were on cards; the deck for the
raw samples would differ only for the data points, data bank
values, and the title cards of the map packages. Similar
procedures were used in Phase II.
In Phase III more sophisticated data handling procedures
had to be developed. Considerable thought was given at the
-------
outset to the most efficient and logical means for mapping
and displaying the information pertaining to this phase of
work. Before any mapping was done, the total mapping
program was laid out and evaluated. It became apparent that
two problems had to be solved in order to produce the
required maps efficiently. Both problems are caused by
the very inefficient way in which SYMAP is used to take
output from one map and use it as input to another:
1. developing a technique to store the output of
each computer run; and,
2. developing a technique to retrieve and use
this output as input to further computer runs.
For example, the first run might map the mode values
for sulfur content and the second run, the thickness of the
bed. A third run is then desired to calculate the quantity of
coal with less than 1% sulfur content, using the first and
second runs. It is necessary to have an efficient technique for
storing the matrices of values created by the first two runs
and then retrieving them to produce the map in the third run.
Most of the previous work dealing with SYMAP has
used tapes as a means of storing output maps. This procedure
works quite well when the number of tapes the user is handling
is small, but as the number of output tapes increases so
does the confusion and difficulty. Consequently, a disk storage
device was used to store all of the output maps that were run
in Phase III. The procedures involved are almost identical to
those used with tapes. Minor modifications were made in the
SYMAP program to add the necessary job control language
cards and to specify the needed parameters for writing on
disks.
-------
The following steps were necessary to set up this
procedure:
1. Acquiring a 2314 disk pack from the Harvard
Computing Center to be used for storing the basic
information. The SYMAP source program was also
stored on this disk; although this was not an essential
step, it was found to be more efficient to have the
mapping program on the same disk, instead of using one
disk for data and a separate system disk which normally
has SYMAP stored on it.
2. Developing the job control language (JCL)
needed for each run. This is a fairly straight-
forward procedure; these cards specify the disk that
is to be mounted, the block size factor, the record
format factor, and the space allocation parameters.
As SYMAP is currently set up to write its output on
channel 8 for tape, this channel was used to write
output onto disk. The data file names are specified
for each of the output files on the job control cards.
These names were carefully selected and organized
to eliminate confusion later on in the project.
The GRID program was selected as a second mapping
program because of its ability to very efficiently read the
previously-stored output files from SYMAP and to manipulate
these data to calculate new variables such as quantity. Only
slight modifications had to be made in the program for this
purpose. It was necessary to write a pre-program that
would combine various output files from the SYMAP runs
stored on disk into one data bank containing 22 variables.
-------
The standard user subroutine of GRID was then employed
to read in specific values from the data bank and manipu ate
them according to various algorithms developed for this
purpose. The results of these calculations were then
temporarily stored and graphically portrayed by GRID.
For example, a series of maps was run in Stage One of
Phase III which displayed the quantity of coal with less
than 1% sulfur content. These maps required the use of the
GRID program. The user subroutine developed performed
the following tasks:
1. Read the matrix of sulfur content values that
were stored on a data file on the disk by the SYMAP
program;
2. Determine if the value for a particular ceil
was less than 1%;
3. For those areas where sulfur content was less
than 1%, read the thickness data file; also created by
the SYMAP program;
4. Calculate the quantity of coal in tons per acre;
and,
5. Temporarily store the values for quantity prior
to mapping.
Our maps of quantity of coal were displayed in terms
of tons per acre. To determine for the particular outcrop or area
over 28" thick just how much coal in tons would be available
it is necessary to multiply this quantity mapped times the number
of acres; the number of acres can be determined by counting
up the computer character locations or cells within the study
area outline and multiplying by the scale factor for the
-------
acreage per character cell. It is possible to
subroutine to do this task within the mapping program
Due to the questionable accuracy of the data used and the
quantities determined, as well as the arbitrary study area
boundaries chosen, our results were presented only in terms
of tons of coal per acre.
4.6. DISPLAYING NUMBER OF ANALYSES ON BASE MAPS
The data banks of sulfur content used in this project
generally contained four variables: the low, mode, and high
value for sulfur content, and the number of analyses on which
these values were based. On the base maps for the study area
in Phase I and II the data points, outlines, and legends are
displayed, but no values are shown. We wished to display
the number of analyses adjacent to each data point to aid in
the evaluation of the accuracy of the data for various parts of
the study area.
The number of analyses for each data point is displayed
on the base map as a point C-OTOLEGEND. A small data
handling program was written which performed the following
tasks:
1. Read in the data point locations; transform
these if the source map units used for data points and
legends are different;
2. Read in the corresponding values for the
number of analyses at each data point;
3. Punch out cards in the standard C-OTOLEGENDS
package when base maps are run.
-------
This program can easily be generalized to produce
other legends such as mine codes, town names, year samples
taken, etc. to be printed on the base maps at or adjacent to
the data points. If large scale maps are used or there are
few data points, these legends can be used on all maps in a
series; because they often obscure the symbolism they are
generally used only with base maps.
------- |