EPA-450/3-75-086
December 1975
METHODOLOGIES
FOR COUNTYWIDE
ESTIMATION
OF COAL, GAS,
/
AND ORGANIC
SOLVENT CONSUMPTION
~
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Air and Waste Management
Office of Air Quality Planning and Standards
Research Triangle Park, North Carolina 27711
-------
EPA-450/3-7 5-086
METHODOLOGIES
FOR COUNTYWIDE
ESTIMATION
OF COAL, GAS,
AND ORGANIC
SOLVENT CONSUMPTION
by
Joseph P. Myers and Frank Benesh
Walden Research Division of Abcor, Inc .
201 Vassar Street
Cambridge, Massachusetts 02139
Contract No. 68-02-1410
EPA Project Ofricer: Charles O. Mann
Prepared for
- ENVIRONMENTAL PROTECTION AGENCY
Office of Air and Waste Management
Office of Air Quality Planning and Standards
Research Triangle Park, North Carolina 27711
December 1975
-------
This report is issued by the Environmental Protection Agency to report
technical data of interest to a limited number of readers. Copies are
available free of charge to Federal employees, current contractors and
grantees, and nonprofit organizations - as supplies permit - from the
Air Pollution Technical Information Center, Environmental Protection
Agency, Research Triangle Park, North Carolina 27711; or, for a fee,
from the National Technical Information Service, 5285 Port Royal Road,
Springfield, Virginia 22161.
This report was furnished to the Environmental Protection Agency by
Walden Research Division of Abcor, Inc. , Cambridge, Massachusetts 02139
in fulfillment of Contract No. 68-02-1410. The contents of this report
are reproduced herein as received from Walden Research Division of
Abcor, Inc. The opinions, findings, and conclusions expressed are
those of the author and not necessarily those of the Environmental Protection
Agency. Mention of company or product names is not to be considered
as an endorsement by the Environmental Protection Agency.
Publication No. EPA-450/3-75-086
11
-------
TABLE OF CONTENTS
Section Title Page
I INTRODUCTION 1-1
II SUMMARY 2-1
A. Methodology : . 2-2
B. Data Base 2-28
III DEVELOPMENT OF METHODOLOGY 3-1
A. Residential 3-1
B. Commercial-Institutional 3-17
C. Industrial 3-22
D. Off-Highway Consumption of Gasoline and Diesel
Fuel 3-28
E. Gasoline Consumption by Marine Vessels 3-35
F. Railroad Consumption of Diesel Fuel 3-38
G. Retail Sales of Gasoline 3-38
H. Organic Solvents 3-40
I. Sulfur and Ash Content of Coal 3-48
J. Landing and Take-Off Cycles of Aircraft 3-51
IV COMPUTER PROCESSING 4-1
A. Overview 4-1
B. The ASFA Master File 4-4
C. The ASFA Data Preprocessing System 4-18
D. The Area Source Fuel Allocation Program (ASFAP) . 4-22
E. Results for Selected Test Counties 4-25
V RECOMMENDATIONS 5-1
A. Improvement of Data Base 5-1
B. Improvement of Methodology 5-2
VI REFERENCES 6-1
APPENDIX A - Regression Analysis of Residential Gas Consumption
Patterns A-l
APPENDIX B - Regression Analysis of Commercial Fuel Consumption for
Five Subcategories B-l
APPENDIX C - Analysis of Alternative Data Sources for State Coal,
Gas, and LPG Shipments/Consumption C-l
-------
TABLE OF CONTENTS (continued)
Section Title Page
APPENDIX D - Alternative Methodologies for Allocating Railroad
Use of Diesel Fuel D-l
APPENDIX E - National Use of Organic Solvents E-l
LIST OF FIGURES
Number Title Page
3-1 Four Candidate Equations for Allocation of Residen-
tial Coal Consumption 3-12
3-2 Commercial Allocation Methodology Flow Diagram ... 3-18
3-3 Surface Coatings Industry Estimated Prodction in
1970 3-45
4-1 NEDS Area Source Data Coding Form 4-2
4-2 Computer Processing Flow Diagram 4-3
4-3 ASFAP Program Flow Chart 4-23
A-l Scatter Diagram of Degree Days and Therms per
Customer A-4
B-l Plot of Hotel Rooms vs. Fuel Use B-8
B-2 Plot of Three Hotel Fuel Use Regression Equations . B-9
C-l Questionnaire Used by the Bureau of Mines for Bitu-
minous Coal and Lignite C-4
C-2 Questionnaire Used by the Bureau of Mines for Natural
Gas C-l 7
C-3 Questionnaire Used by the American Gas Association . C-20
C-4 Questionnaire Used by the Bureau of Mines for LPG . C-24
-------
LIST OF TABLES
Number Title Page
2-1 Solvent Types and User Categories 2-21
2-2 National Variables for Allocation Methodologies . . 2-29
2-3 State Variables for Allocation Methodologies . . . .2-31
2-4 County Variables for Allocation Methodologies . . . 2-34
2-5 Sources Required for Input Preparation 2-36
2-6 Contacts for Retail Sales of Gasoline Data 2-41
2-7 Magnetic Tape Requirements 2-42
3-1 Central Heating Load by Degree Days 3-6
3-2 Use of Gas by Residential Appliances 3-8
3-3 Single-Family Dwelling Unit Thermal Efficiencies of
Gas and Coal 3-10
3-4 Industrial Fuel Instensity Ratios 3-24
3-5 Quantity of Fuel Purchased by SIC Group and Fuel
Type by State 3-25
3-6 National Gasoline and Diesel Fuel Consumption by Off-
Highway Sources 3-30
3-7 Usage Rates, Consumption Rates, and Population Dis-
tribution for Heavy-Duty Agricultural Engines Used
for 1973 Update 3-31
3-8 Labor Productivity of Mining, Manufacturing, and
Petroleum Trade 3-33
3-9 Centroid Counties 3-36
3-10 Industrial Solvent Categories 3-41
3-11 National Use of Solvents 3-42
3-12 Distributive Factors for Organic Solvents 3-44
3-13 Distributive Factors for Surface Coatings 3-47
3-14 Sulfur Content of Coal by Production District . . . 3-49
-------
LIST OF TABLES (continued)
Number Title Page
4-1 Area Source Fuel Allocation Master File Formats . . 4-5
4-2 ASFAP Program Subroutine Descriptions 4-24
4-3 Summary of Demographic and Climatological Data for
Selected Test Counties 4-27
4-4 Residential Allocation Test Results 4-28
4-5 Commercial-Institutional Allocation Test Results . . 4-29
4-6 Industrial Allocation Test Results 4-30
4-7 Sulfur and Ash Allocation Test Results 4-31
4-8 Transportation Allocation Test Results 4-33
4-9 Evaporation Losses Test Results 4-34
4-10 Organic Solvent Allocation by User Category .... 4-35
4-11 Delaware State Summary of Test Results 4-37
B-l Employee/Room Ratios for Hotels by State B-10
C-l Distribution of Bituminous Coal in 1971 C-2
C-2 United States Consumption and Exports of Bituminous
Coal C-8
C-3 Estimated Bituminous Coal Consumption by Customer
Class in Eleven Selected State Groupings, 1971 . . . C-10
C-4 Anthracite Shipments in 1972 C-l2
C-5 Estimated Anthracite Coal Consumption by Consumer
Class in Eleven Selected States for 1971 C-l4
C-6 Natural Gas as Reported in the Minerals Yearbook,
1971 C-16
C-7 Gas Utility Industry Sales by State and Class of
Service for 1971 C-18
C-8 Total Gas Sales, AGA Data as Percent of Bureau of
Mines Data, 1960-1972 C-19
VI
-------
LIST OF TABLES (continued)
Number Title Page
C-9 Sales of LPG and Ethane by Use, Excluding Use in
Gasoline Production, by P.A.D. District and State,
1971 and 1972 C-23
f
C-10 Comparison of Bureau of Mines Based Estimates of
Industrial Fuel Use and 1972 Census of Manufactures,
Fuel and Electrical Energy Consumed, for Eleven
Selected State Groupings, 1971 C-26
E-l Percent of Total Consumption Use as Solvent .... E-7
-------
ACKNOWLEDGEMENTS
This work was supported under Contract No. 68-02-1410 by the Environ-
mental Protection Agency (EPA). The assistance and guidance of the EPA
Project Officer, Mr. Charles Mann, was greatly appreciated.
vi n
-------
I. INTRODUCTION
The Environmental Protection Agency has developed an extensive,
nationwide data base of estimated air pollutant emissions from area and
point sources. The data in this National Emission Data System (NEDS)
require updating on a routine basis to provide current information for
the EPA and other branches of government. The data on point sources
is continually being updated by means of legal SIP reporting require-
ments on state agencies. There are no legal requirements, however, for
states to make routine data submittals to update the area source data.
The objective of the current project was to develop methods for EPA to
estimate fuel consumption information on a county-wide basis for area
sources in the NEDS data bank. These techniques were then translated
into computer programs to facilitate the application. The specific ele-
ments of the NEDS file that were considered are:
. Consumption by residential sources of natural gas, liquid petro-
leum gas (LPG), anthracite coal, and bituminous coal
. Consumption by commercial-institutional sources of natural gas,
LPG, anthracite coal, and bituminous coal
. Consumption by industrial sources of natural gas, LPG, anthra-
cite coal, and bituminous coal
. Consumption of gasoline and diesel fuel by off-highway sources
. Consumption of gasoline by marine vessels
. Consumption of diesel fuel by railroads
. Retail sales of gasoline
In addition, the project considered updating information on:
. Consumption of organic solvents
. Sulfur and ash content of anthracite coal and bituminous coal
. Landing and take-off cycles of military, civil, and commercial
aircraft
The project plan was divided into two phases. The first phase was devoted
1-1
-------
to development of the methodology and testing on selected county data. The
second phase included the collection and processing of 1973 data for all
counties in the United States. This report describes the results of the
first phase of the project. The data resulting from the second phase are
available in the form of computer listings and magnetic tapes, as well as
a coding manual [1] and a program documentation [2]. NEDS computer program
listings containing data produced according to the methods described in
this report may be obtained by contacting the Requests and Information Sec-
tion, National Air Data Branch, U.S. Environmental Protection Agency,
Research Triangle Park, North Carolina 27711.
The approach used by Walden to arrive at the county allocation methods
was the development of basic relationships between fuel use and other demo-
graphic and economic factors. Statewide fuel consumption was distributed
among the counties within each state based on the developed relationships.
The resulting county-wide figures will serve to update corresponding figures
of previous years presently available in NEDS format. Up-to-date air pol-
lutant emissions estimates for area sources are calculated from these con-
sumption figures by the NEDS computer programs.
The limitations of the resulting methods are summarized below:
(1) Demographic and economic data for the categories required by this
study were often found to be incomplete or unavailable. Alternate,
but less accurate methods were developed using the available in-
formation, introducing an error factor additional to the inaccuracy
of some of the published data.
(2) The fuel crisis confronted by the United States has resulted in
significant changes in the fuel consumption patterns across the
nation. Lower thermostat settings in homes and businesses, shorter
working hours, and the unavailability of motor fuels will have
altered some of the correlative relationships developed here, based
on historical data.
1-2
-------
II. SUMMARY
A. METHODOLOGIES
Methodologies were developed for determining area source (county)
quantities for the following NEDS area source data items:
Consumption by residential sources of bituminous coal, anthra-
cite coal, natural gas, and LPG
Consumption by commercial-institutional sources of bituminous
coal, anthracite coal, natural gas, and LPG
Consumption by industrial sources of bituminous coal and natural
gas
Retail sales of gasoline
Consumption by off-highway sources of gasoline and diesel fuel
Consumption of diesel fuel by railroads
Marine consumption of gasoline
Organic solvent consumption
Sulfur and ash contentof bituminous and anthracite coal
Aircraft landing and take-off cycles
The approach in developing the methodologies was based on the use of national,
state, and county data items which are readily available, are updated periodi-
cally, and reflect variation of fuel use in time at both national and local
levels. In general, the methodologies were designed to apportion county fuel
use from published state totals. Two basic techniques were used. In some
cases, equations were developed to estimate county consumption values which
are then normalized to the published state totals. In other cases, the pub-
lished state consumption is apportioned directly to the counties according to
distribution of related demographic variables (e.g., population).
Several sources of state fuel use data report only a regional total
for groups of states in some geographic regions. In such cases, related
methodologies apply to the region as if it were a state comprised of the counties
of the states, i.ncluded in the region.
Some data items were not available at the state level. In such
cases, national figures were used, and the totals for counties in all states
normalized against the national level.
2-1
-------
In most instances, available fuel use figures pertain to total
consumption. Thus, it is necessary to subtract corresponding point source
values from the totals to obtain values representative of area sources. In
all cases, the published state, regional, and national consumption figures
are assumed to be correct. Estimated values, therefore, are normalized
against published state totals that have been adjusted to account for point
sources.
The methodologies for allocation of each data item are summarized
in this section. Additional information, including background related to the
derivation of the algorithms summarized in this section, is given in Section
III and the Appendices. The overall scheme for processing of the data for
NEDS is discussed in Section IV.
1. Fuel Consumption by Residential Sources
a. Natural Gas and Liquified Petroleum Gas (LPG)
Consumption of natural gas reported in the NEDS area source
data is actually the combined total of natural gas and LPG. However, separate
methodologies were developed for these two fuels. The results of the LPG
methodology are, therefore, converted to a natural gas equivalent and added
to the natural gas consumption value.
County natural gas consumption by residential sources is cal-
culated by means of the formula 0 588
flL '
T = 47.5 * U * D
0.367
*
U
x
R
0.125
d
where T = county consumption of natural gas in therms
D = annual degree days for the county [3]
U = number of occupied dwelling units in the county using gas [4]
U .= number of occupied dwelling units in the county using gas for space
gh heating [4]
U = larger of the number of occupied dwelling units in the county using
x gas for cooking or hot water [4]
R. = median number of rooms per dwelling unit in the county [5]
2-2
-------
This estimate is then normalized against the reported state consumption:
NG
NGC = Tg * -^
where NG = normalized county natural gas consumption (10 cu.ft.)
fi
NG = published state residential natural gas consumption (10 cu.ft.) [6]
T = sum of calculated county consumption (Ta) for all counties in
5 the state (therms) y
County LPG consumption by residential sources is calculated using the formula
T. = (376 + 0.209D)IL + c" * IL + c~ * U{
where T. = county consumption of LPG by residential sources (therms)
X*
D = annual degree days for the county [3]
IL = total dwelling units in the county using LPG [4]
c~ = regional average water heater consumption (therms) [7]
w
IL = number of dwelling units in the county using LPG for heating water [4]
X/W
c~ = regional average cooking range consumption (therms) [7]
IL = number of dwelling units in the county using LPG for cooking [4]
X* i
No sources were found that report state residential LPG consumption. However,
retail (commercial and residential combined) figures are published [8]. State com-
mercial point source LPG consumption is subtracted from the published state
retail total, yielding state retail area source LPG consumption. The computed
county figures (T.) are converted to kilogallons and summed over the state:
L = 0.00105 T.
C Xy
and
L
- c c
where
0.00105 = conversion factor from therms to kilogallons of LPG
L = estimated county residential LPG consumption (kilogallons)
\f
T. = estimated county residential LPG consumption (therms)
L = estimated state total residential LPG consumption (kilogallons)
2-3
-------
If the estimated state residential LPG consumption exceeds the state retail
area source LPG consumption figure, the county consumption figures are nor-
malized against the state retail area source value. Otherwise, the county
LPG consumption estimate is left unchanged, and the remainder of the state
area source retail LPG is used by the allocation methodology for commercial
LPG consumption.
The total gas equivalent is computed by converting the final county
residential LPG consumption estimate from 103 gallons to a natural gas equivalent
in 10 cu.ft. and adding that value to the normalized county natural gas consumption.
TGEC = NGc + 0.0922 * L'c
where TGE = county residential area source total gas equivalent con-
c sumption (106 cu.ft.)
NG = normalized county residential natural gas consumption (10° cu.ft.)
\f
0.0922 = factor to convert LPG in 10 gal. to natural gas equivalent
in 106 cu.ft. (Ratio of the heat equivalent of 106 cubic feet
of natural gas to the heat equivalent of 1Q3 gallons of LPG)
o
L'c = final county residential LPG consumption (10 gal.)
b. Bituminous and Anthracite Coal
Total county residential consumption of coal (anthracite and
bituminous) is calculated using the formula
1000.
C = 0.00387 * Ucoal * e ' " D
x
'
where C = county consumption of coal (anthracite and bituminous)
U = number of occupied dwelling units in the county using coal
c for space heating
D = annual degree days for the county
Residential consumption of anthracite coal and bituminous coal is derived from
the estimated total coal consumption by
ac = fa * C
bc = (1 - ffl) * C
2-4
-------
where a = estimated county residential anthracite coal consumption (tons)
b = estimated county residential bituminous coal consumption (tons)
C = estimated total county residential coal consumption (tons)
f = fraction of total state coal market that is anthracite coal
a
The county consumption estimates must be normalized against
published state figures. No sources were found that report state residential
bituminous or anthracite coal consumption. Data on state shipments of anthra-
cite and bituminous coal for retail use are available, however [9,10]. A national
retail bituminous coal consumption figure is also obtainable.* A factor for
converting state shipments to consumption is calculated from the national
consumption and the sum of the state bituminous coal shipments, and the
factor is applied to each state coal shipment value. State commercial point
source consumption of anthracite and bituminous coal is subtracted from the
corresponding state retail consumption, yielding state retail area source
fuel consumption values.
One of three conditions will arise at this point. The county
consumption estimates for each coal type are summed over the state. If both
of these estimated state residential consumption totals exceed their respec-
tive retail area source consumption figures, the county consumption estimates
are normalized:
letting a = I a
5 c c
bs = I bc
As
then a1 = a * —
c c a
B
and t>' = b * 7-^-
c c b
where a = estimated county residential anthracite coal consumption (tons)
b = estimated county residential bituminous coal consumption (tons)
AS = state retail area source anthracite coal consumption (tons)[9]
BS = state retail area source bituminous coal consumption (tons)[10]
*See Appendix B.
2-5
-------
a'c= normalized county residential anthracite coal consumption (tons)
b'c= normalized county residential bituminous coal consumption (tons)
For the case where the sum of the estimated county residential consumption figures
exceeds the state retail area source consumption for one coal type and does not
exceed the state retail area source consumption for the other, the excess com-
puted consumption is distributed among the counties and added to the other coal
type as follows:
with X = state retail area source consumption of one coal type (i.e., A or B
o o
Y = state retail area source consumption of the other coal type
x = calculated county residential area source consumption corresponding
to coal type of Xg (i.e., ac or bc)
y = calculated county residential area source consumption corresponding
to coal type of YS
Let x = l x such that x > X
f+
\+
y^ = I yr such tnat y^ < v*
r 03
Ax = xs - Xs
Ay - Ys - ys
If Ax <_ Ay, let Ac = Ax
If Ax > Ay, let Ac = Ay
Then y'c = yc(l + ^)
v"
x
* —
c c
where x' = normalized county residential consumption of coal type x
c (i.e., a'c or b'c)
y1 = adjusted county residential consumption of coal type y
If the sum of county residential consumption estimates is less than the state
retail area source consumption for both coal types, no adjustments are made
(i.e., a'c = ac, and b'c = bc).
2-6
-------
The final county residential area source consumption for
each coal type (a1 and b1 ) is summed over the state. These state totals
L- L»
are subtracted from the corresponding retail area source consumption. The
remainders are the state commercial bituminous and anthracite coal, to be
allocated according to the commercial coal methodologies.
2. Fuel Consumption by Commercial-Institutional Sources
The methodology for determining county fuel consumption of
natural gas, LPG, anthracite coal, and bituminous coal by commercial and
institutional area sources is performed in five stages.
a. Fuel consumption is calculated for each of five commercial-
institutional subcategories using the following formulae:
T, = 126.5 * BEDS + 12.7 * D + 77.4 * E] - 5.72 x 104
T2 = 8.05 x 10~17R0K84D3'"(R0 + 2.84 (R - RQ))
T3 = 165 E3 + 4JO D " 1>81 x ]°4
T4 = 229 E4 + 51.5 D - 2.94 x 105
T5 = 531 Eg - 1.28 x 104
o
where T, = Total fuel consumed in the county by hospitals (10 therms)
o
T£ = Total fuel consumed in the county by hotels (10 therms)
Tg = Total fuel consumed in the county by schools (103 therms)
T4 = Total fuel consumed in the county by universities (103 therms)
T,- = Total fuel consumed in the county by commercial laundries
(103 therms)
D = Annual degree days for the county [3]
BEDS = Number of hospital beds in the county[11]
i R = Number of hotel rooms in the county [12]
R0 = 100 if R > 100
or R0 = R if R <. 100
' E-| = County hospital employment
Eg = County school employment
E. = County university employment
EC = County commercial laundry employment [13]
2-7
-------
The method for determining county employment for hospitals, universities, and school
is explained in Appendix B. Total fuel consumption in the county by the five com-
mercial subcategories (T ) is then computed:
b. The estimated fuel consumption for the five commercial sub-
categories is summed over all counties in the state. This state total is
broken down into natural gas, LPG, and coal consumption according to the
proportion of occupied residential dwelling units in the state using those
respective fuels for space heating:
Letting U. = the number of occupied dwelling units heated by
fuel type i (0 = fuel oil, 1 = natural gas,
2 = LPG, 3 = coal)
X. = published or derived state commercial area source
consumption of fuel type i,
Ts • I Tc
3
and U = UQ + y (Ui : X1 > 0)
then T, c = Tc Ui
1,5 M
where T. = State consumption of fuel type i by five
' commercial subcategories (terms)
Dwelling units using fuel oil for space heating are included in the dwelling
unit total to account for fuel oil consumed by commercial sources in the fuel
consumption total.
c. The state five commercial subcategory consumption for each
fuel type is converted from therms to the standard NEDS units for that fuel
type (106 cu.ft. for natural gas, 103 gal. for LPG, tons for coal). Coal
consumption is then split between anthracite and bituminous according to the
proportions of the state commercial area source consumption of each coal type
2-8
-------
remaining from the input retail value after conversion from shipments and
subtracting normalized residential consumption and commercial point source
consumption (see Section II.A.l.b).
F] = 9.69 x 10"5 T1
F2 = 1.05 x 10~3 T2
F3 = 4.08 x lO'6 (ir-rjr} * T3
o o
where F. = State consumption of fuel type is by the five com-
mercial subcategories in the standard NEDS units of
that fuel type (i = 1 for natural gas, i = 2 for LPG,
i = 3 for anthracite coal, i = 4 for bituminous coal)
a = Actual state commercial area source consumption of
s
anthracite coal (tons)
b = Actual state commercial area source consumption of
s
bituminous coal (tons)
_c 5
9.69 x 10 " = Factor to convert therms to 10 cu.ft. of natural gas
9.05 x 10~3 = Factor to convert therms to 103 gal. of LPG
4.08 x 10 = Factor to convert therms to tons of anthracite coal
3.82 x 10" = Factor to convert therms to tons of bituminous coal
d. Next, state consumption of each fuel type by all commercial
sources other than the five subcategories are computed. Normalization factors
are also computed.
Actual state commercial area source consumption of each
fuel type is derived directly from published values. State commercial
area source natural gas is the total commercial natural gas consumption [5]
less the commercial point source natural gas consumption. State consump-
tion of LPG, anthracite coal, and bituminous coal are the remainder of
state retail fuel consumption after subtacting state commercial point source
consumption and normalized state residential consumption for the respective
fuel types (see Section II.A.I.a and II.A.l.b).
2-9
-------
State fuel consumption of all commercial sources other
than the five commercial subcategories is calculated as follows:
If X1 > F. , then FO.. = Xi - F. and ni = 1
Xi
If X1 < F. , then F0i = 0 and n . = p1
where FO. = State consumption by "other" commercial sources
of fuel type i
X- = Actual state commercial area source consumption
of fuel type i
F. = Calculated state five commercial subcategory
consumption of fuel type i
n.j = Normalization factor for consumption of fuel
type i by five commercial subcategories
e. Finally, the county commercial consumption is calculated,
Fi,c • "iVi.c'lj1 + F°i
L* o
where F. = Normalized county consumption of fuel type i by
1>c commercial area source (in the standard NEDS
units of fuel type i)
n. = Normalization factor for consumption of fuel type
i by the five commercial subcategories
f . = Factor to convert consumption from therms to the
NEDS standard units for fuel type i
T. = Total county fuel consumption by the five commer-
' cial subcategories (therms)
II. = Number of occupied dwelling units in county c using
' fuel type i for space heating [4]
4
Ur = E (Ui c : TT > 0) + U0 C
C • -I 1»C 1 UjU
= Total number of dwelling units in the county
using fuels for space heating that are also
used by commercial sources. This includes
dwelling units using fuel oil. (UQjC).
FO. = State consumption of fuel type i by all
1 commercial sources except the five subcategories
2-10
-------
E = County area source employment for all commer-
cial sources except the five subcategories
E = State area source employment for all commercial
sources except the five subcategories
3. Fuel Consumption by Industrial Sources
The procedure for estimating natural gas and bituminous coal
consumption by industrial area sources involves four steps. Essentially all
anthracite coal consumed by industry is by point sources, so no allocation
is performed for that fuel type. Industrial consumption of LPG is not esti-
mated separately, but is combined with the state natural gas total prior to
apportionment to the county's level.
a. State fuel intensity ratios are computed for each fuel
type by each of SIC categories 20-39 (in this study, SIC category 39 will
represent the combination of SIC 39 and SIC 19), using the most recent
Census of Manufactures fuel use data [14] and employment data for the
corresponding year from the County Business Patterns £13].
where FIR.. = The fuel intensity ratio for fuel type i
J and SIC category j
F. . = State consumption of fuel type i by SIC
1J category [14]
E. = State employment of SIC category j [13]
J
b. The fuel intensity ratios give a measure of fuel use in-
tensity per employee. By applying the state fuel intensity ratios to corres
ponding county area source employment figures, and summing over the 20 SIC
categories, an estimate of county industrial fuel consumption is obtained:
20
where F. = County industrial area source consumption of
1 }C fuel type i
E. = Total employment in county c for SIC j (CBP)
^' minus point source employment in county c for
SIC j (NEDS)
FIR. . = State fuel intensity ratio for fuel type i and
1J SIC category j
2-11
-------
c. Actual state industrial area source consumption of natural
gas (including LPG) and bituminous coal are derived from state totals and
point source consumption figures as follows:
XT = (G - Gp) + f(L - Lp)
and X2 ' fts
-------
4. Consumption of Gasoline and Diesel Fuel by Off-Highway Sources
Off-highway sources have been divided into six categories: farm
equipment, construction equipment, industrial equipment, motorcycles, lawn
and garden equipment, and snowmobiles. Consumption in each category is es-
timated by either of two general approaches:
. Apportionment of a national fuel consumption total to counties
on the basis of employment, population etc.
. Calculation of county or state totals by applying fuel con-
sumption rates to average usage figures and equipment popu-
lations.
Total off -highway consumption of gasoline and diesel fuel is the sum of the
consumption estimates of the individual categories as follows:
6
6 = I Fi *
j=l 1>J
DF = F .
3
where G = Off-highway consumption of gasoline (10 gal.)
q
DF = Off-highway consumption of diesel fuel (10 gal.)
F. .= Consumption of fuel type i (i = 1 for gasoline,
J i = 2 for diesel fuel) by off-highway category j
(j = 1 for farm equipment, j = 2 for construc-
tion equipment, j = 3 for industrial equipment,
j = 4 for motorcycles, j = 5 for lawn and garden
equipment, and j = 6 for snowmobiles)
A description of the methodologies for estimation of fuel consumed by each off-
highway category follows:
a. Farm Equipment
Consumption of gasoline and diesel fuel by farm equipment is
apportioned to individual counties from an estimated state farm equipment con-
sumption on the basis of tractor population. To estimate state fuel consump-
tion by farm equipment, consumption values are calculated separately for each
2-13
-------
of five equipment subcategories: farm tractors, combines, motorized balers.
forage harvesters, and general -purpose large utility engines. The methodology
for estimating state fuel consumption by farm equipment is expressed as:
Fs,i =
where F . = State consumption of fuel type i (i = 1 for gaso-
' line, i = 2 for diesel fuel) by farm equipment (gal.)
f. • = Fraction of farm equipment subcategory j in the
. J state that is powered by fuel type i (j = 1 for
tractors, j = 2 for combines, j = 3 for motorized
balers, j = 4 for forage harvesters, j = 5 for
general -purpose large utility engines) [15]
N. = State population of farm equipment in subcategory j[16]
J
A.. = Average annual usage of equipment in subcategory j
J using fuel type i (hours/year) [15]
FR. • = Average hourly consumption of fuel type i per unit
J of equipment in subcategory j (gals. /hour) [15]
With the exception of the general -purpose large utility engine category, equip-
ment populations are obtained from the Census of Agriculture [16]. The large
utility engine population is estimated from the distribution of tractor popu-
lations in irrigated and non-irrigated areas:
N5 = °-30Nl + °-°5Nl <»)
where N, = Number of tractors in the state
IR = Number of farms in the state in irrigated areas
NIR = Number of farms in the state in non-irrigated areas
The estimated large utility engine population is apportioned by fuel type ac-
cording to the proportions of tractors that are powered by gasoline and diesel
fuel.
County consumption of gasoline and diesel fuel by farm
equipment is determined by:
2-14
-------
where F., = County consumption of fuel type i by farm
equipment
F . = State consumption of fuel type i by farm equip-
Sil ment
N = County tractor population [17]
V*.
N, = State tractor population
b. Construction Equipment
County consumption of gasoline and diesel fuel by construc-
tion equipment are allocated from the state construction equipment consump-
tion on the basis of population. State fuel consumption is estimated by
apportioning national fuel consumption according to total non-building
construction employment, (i.e., employment in heavy construction (SIC 1600)
and special trade (SIC 1700) categories)
Fi 2 = FN i^E~) (p~)
' N s
where F. ~ = County consumption of fuel type i by construction
' equipment (gal.)
F,, , = National consumption of fuel type i by construc-
' tion equipment (gal.) [14]
E = State non-building employment [13]
Ei, = National non-building employment [13]
P = County population [4]
P = State population [4]
c. Industrial Equipment
The methodology for estimating consumption of gasoline and
diesel fuel by industrial equipment is expressed as:
P _ P / c,mmwN
i 1 N i VF '
1>J N" ''N.mmw
where F. 3 = County consumption of fuel type i by industrial
lj equipment
FM . = National consumption of fuel type i by industrial
N>1 equipment [18]
2-15
-------
EM .___. = Total county employment in manufacturing, mining,
li )illlllW j i i ^ t t ^ _ i
and wholesale trade [13]
^N mmw = Total national employment in manufacturing, mining,
N?mmw and wholesale trade [13]
d. Motorcycles
County gasoline consumption by motorcycles is estimated from
state consumption on the basis of population:
P
Fl,4 = F- (M * FR * (flul + f2u2)}
where F, ^ = County consumption of gasoline by motorcycles (gal.)
PC = County population [19]
PS = State population [19]
M = State motorcycle registrations [20]
FR = Motorcycle fuel consumption rate (gal ./mile)[i8]
f^ = Fraction of motorcycles that are off- road [21]
u-j = Average annual usage of off-road motorcycles (miles/year)
f^ ~ Fraction of motorcycles that are combination [21]
u2 = Average annual usage of combination motorcycles [21]
(miles/year)
e. Lawn and Garden Equipment
County lawn and garden equipment consumption of gasoline is
derived from national totals of lawn and garden equipment consumption of gaso-
line and snowthrower consumption of gasoline:
UK FFD Pr S
r r- / _ C_\ I _ C_N . ,, p (J±—\ I C }
Fl,5 = FN,LG^U1J
where F-i c = County consumption of gasoline by lawn and
' garden equipment (gal.)
FN r= National consumption of gasoline by lawn and
N' garden equipment other than showthrowers (gal.) [18]
Ul = Number of dwelling units in single-unit struc-
c tures in the county [5]
U1N = Number of dwelling units in single-unit struc-
tures in the nation [5]
2-16
-------
FFD = Number of freeze-free days (minimum temperature
c > 32°F) in the county [3]
FFDN = I FFD for all counties in the nation
= 0 for counties with annual snowfall < 30 inches
I/
= 1 for counties with annual snowfall > 30 inches
FM CM= National consumption of gasoline by snowthrowers
N'SN [18] (gal
Sc = County snowfall [3]
S = I (S : S > 30 inches) (snow zone snowfall; the
c ^
snow zone is all areas with annual snowfall >
30 inches)
P = County population [19]
pc7 = I (p : s > 30 inches) (snow zone population)
c c
f. Snowmobiles
County consumption of gasoline by snowmobiles is derived from
the national snowmobile gasoline consumption total on the basis of the counties'
share of the snowmobile population. County snowmobile population is estimated
from state snowmobile registrations using one of two formulae. The formulae
compute the fraction of state snowmobiles that are located in the county. The
formula used is determined by the population density of the county:
(1) For counties with population densities that are less than
1,000 inhabitants per square mile,
P S
f = 1.56 * (p^) + 0.0321 (^) - 0.0234 (i)
\f » O
(2) For counties with population densities that are greater
than or equal to 1,000 inhabitants per square mile,
P
f = K ^ [1.5 - 0.0005 pi (ii)
s L
where f = Fraction of state's snowmobiles that are located in
county c
P = County population [19]
\^
PS = State population
2-17
-------
Sc = County snowfall [3]
S = Snowfall at center of the state (centroid
county snowfall)
p = county population density (inhabitants/square
mile) [5]
= 0 for p > 3,000
c
= 1 for 1,000 < p < 3,000
The county snowmobile population is then computed:
c,sm ~ c s,sm I'm)
where NC sm = Number of snowmobiles in county c
N = State snowmobile registrations
The county consumption of gasoline by snowmobiles is then apportioned from
the national snowmobile gasoline consumption total:
NC sm
1,6 N,sm NNjSm
where F, ,, = County consumption of gasoline by snowmobiles
FM = National consumption of gasoline by snowmobiles
N'Sm (gal.) [13]
N~ cm = County snowmobile population
c j sm
NM _m = National snowmobile population
N 5 S IT1
5. Consumption of Gasoline by Marine Vessels
County marine consumption of gasoline is apportioned from state
marine gasoline consumption on the basis of inland water area and coastline:
* mc
-------
where G = County consumption of gasoline by marine vessels (gal.)
W = County inland water area [22]
v*
W = State inland water area [22]
f = Factor for converting coastline to inland water area
L = County coastline
\+
L = State coastline
m = Number of warm months (which promote boating activities).[3]
This is assumed to be the number of months during which
the monthly normal temperatures exceed 45°F for counties
north of 43° latitude, 48°F for counties between 37° and
43° latitude, and 55°F for latitudes south of 37° latitude.
N ,= State inboard boat registrations [23]
FR ,= Average fuel consumption rate of inboard boats (gal/hour)[15]
N 2= State outboard boat registrations[24]
FR 2= Average fuel consumption rate of outboard boats (gal/hour)[15]
6. Consumption of Diesel Fuel by Railroads
County consumption of diesel fuel by railroads is apportioned from
published state consumption on the basis of population distribution:
DFc,r ' DFs,r * <{j>
where DF = County consumption of diesel fuel by railroads (10 gal.)
c ,r o
DF = State consumption of diesel fuel by railroads (10 gal.)[25]
s ,r
P = County population [19]
\*
PS = State population
7. Retail Sales of Gasoline
Some states report retail gasoline sales (volume) by county. For
these states, the reported county figures are used directly. For states that
do not report county retail gasoline sales, a methodology has been developed
that estimates county sales from reported state retail sales of gasoline, re-
ported state aviation gasoline sales, and computed consumption of gasoline by
various off-highway categories.
r 3 LTO
2-19
-------
where VG = County retail sales of gasoline (103 gal.)
Vh = State retail sales of gasoline for highway and
marine use (103 gal.)
rc = Gross receipts of gasoline service stations in county [26]
r$ = Gross receipts of gasoline service stations in state
F.J . = County consumption of gasoline (103 gal.) by off-highway
'J category j (j = 1 for farm equipment, j = 2 for con-
struction equipment, j = 3 for industrial equipment)
o
VQ = State aviation gasoline sales (10 gal.)
LTO = Total landing - take-off cycles in county for military,
civil, and commercial aircraft
LTO. = I LTOr
c c
The county consumption of gasoline by off-highway sources is deter-
mined using the methodologies described in Sections II.A.4.a, b, and c.
The state retail sales of gasoline for highway and marine use are
derived from the reported state total retail sales by subtracting reported sales
for agricultural, commercial, industrial, and aviation off-highway categories:
Vh • Vs - ' j/OH.l'
where V, = State retail sales of gasoline for highway and marine use
V = Total state retail gasoline sales [20]
VQU .= State retail sales of gasoline for off-highway category i
' (i = 1 for agricultural, i = 2 for commercial, i = 3 for
industrial, and i = 4 for aviational)
8. Consumption of Organic Solvents
The methodology for allocating organic solvent consumption by
county consists of apportioning national consumption of seventeen primary sol-
vent groups by major user category according to county population or area source
employment for the individual user categories. Total solvent consumption is
the sum of the consumption value for each of the user categories. Table 2-1
contains a list of the primary solvent groups and corresponding user categories.
Two of the major user categories, surface coatings and other use, are broken down
into subcategories, as shown on the table.
2-20
-------
TABLE 2-1
SOLVENT TYPES AND USER CATEGORIES
User Category ( / designates solvents used)
Solvent Type
Special Naphthas
Penchloroethylene
Ethanol
Trichloroethylene
Toluene
Acetone
Xylene
Fluorocarbons
M.E.K.
1 ,1 ,1-Trichloro-.
ethane
Methyl ene Chloride
Methanol
Ethyl ene Bichloride
Ethyl Acetate
i
1
2
3
4
5
6
7
8
9
10
11
12
13
14
j
S u r
c
1 O
r>
Q.
o
O.
1
/
S
y
/
/
/
/
v/
•
f c
Lf)
CO
UO
(^
O
i — i
OO
2
/
/
/
/
/
/
/
/
/
\ C €
r~.
CO
<_)
oo
3
/
/
/
/
/
/
v/
*/
/
j C
LO
CM
C_>
00
4
•
/
/
/
/
•
/
/
/
: o c
»*
CO
<_)
t/0
5
/
y
/
/
/
/
v/
V
/
/
i t i
o9 IO
CO
LO
CO
o
t/>
6
/
/
/
y
y
/
/
/
/
n c
UD
CM
o
l/>
7
y
•
/
/
/
/
»/
/
/
s
^r
C_> /
/
/
/
CO
r>»
CO
o
^w^,
12
/
/
/
/
/
y
/
/
/
rn
greasin
u>
<->
13
/
/
/
/
/
/
0)
C
C
to
0)
o
>>
S-
r-i
14
/
/
/
en
03 c
•r—
cn^:
C (/>
•r- «r-
4-> r-
C -Q
•r- 3
S- Q-
r>
15
/
/
•
o3 (0
U
CO -M
jn (/>
JQ tO
3 I—
o: a.
16
/
»/
/
i/
/
Other I
i
o
to en
•— M- c
Z3
Q.
O
X
',18
/
/
/
/
/
/
,/
/
/
•
/
/
ro
ro
-------
TABLE ~2-l (continued)
SOLVENT TYPES AND USER CATEGORIES
User Category ( / designates solvents used)
Solvent Type
Cyclohexane
M.I.B.K.
All Other Sol-vents
i
15
16
17
j
S u r
c
1 O
ro -i-
r- 4->
13
CL
Q
o.
1
•
/
f a
LO
•«•>
IT)
O
00
2
/
/
c e
f^I
CO
o
OO
3
/
/
c
LO
CM
O
00
4
v/
/
o a
ti-
ro
o
00
5
v/
»/
t i
o3 ID
CO
LO
CO
0
OO
6
/
/
n 9
UD
C^J
<_)
00
7
/
/
s
v±
<3-
CO CxJ
OO r—
9
/
•
UD
CO
O
OO
10
/
/
l_3 C71_
C C
rC •!-
2: i-
^
r- +J
rO O
•M rd
O 4-
1 —
n
/
/
CO
r-^
CO
c_>
H- 1
L/O
12
/
/
«,
greasin
Ol
r~^
13
/
/
CD
c
c
(D
Ol
(_)
>>
i.
r~i
14
en
o3 C
cn^:
c to
•r— •r™
+J r—
e .a
•r- 3
i- (X
n
15
08 CO
0
J- T-
OJ +->
J3 CO
JQ fO
3 r—
CC. Q.
16
/
Other
i
u
rO CT
i — «*- C
rO 3 -r-
-(-> C $-
O ro 3
1— 2: -p
17
/
/
Use
wl
1 O
rC -r-
r— 4J
3
CL.
0
IX
'.18
v/
•
ro
ro
-------
National consumption of the primary solvent groups is distri-
buted to each of the user categories according to the percentage of solvent
used by the user category:
SNij = Sifij
where S.... = National consumption (tons) of solvent group i
J by user category j (Table 2-1 gives the sub-
script values corresponding to individual
solvent groups and user categories.)
S. = Published total national solvent group i [27,28]
f. . = Percent of solvent group i that is consumed
J by user category j
The county consumption for each solvent group and user category is computed
and summed to give the total county consumption:
S .
J - V
where S = Total county organic solvent consumption (tons)
SN--= National consumption of solvent type i by user
category j (tons)
E . = Number of individuals(employment or population)
in county c in user category j
EN • = Number of individuals in the nation in user
J category j
P . = Number of individuals in county c in point source
J user category j
PN . = Number of individuals in the nation in point
source user category j
9. Sulfur and Ash Content of Coal
Separate methodologies were developed for estimating sulfur and
ash content of bituminous coal and anthracite coal used by area sources in
each county.
a. Bituminous Coal
(1) Determine average sulfur and ash content of bituminous
coal shipped from each production district or production district grouping
for use by retail and industrial sources.
2-23
-------
The Bureau of Mines annually reports sulfur content and
shipments of coal from each of 23 production districts for consumption by
each of five user categories, namely, Electric Utilities, Coke and Gas
Plants, Other Industrial User and Retail Dealers, All Other Users, and
Exports [75]. A weighted average for each production district of sulfur
content for the Other Industrial Users and Retail Dealers and the All
Other Users categories gives a representative value of sulfur content of
bituminous coal shipped for use by retail and industrial sources, ex-
cluding electric utilities and coke and gas plants. Shipments to each
state are reported by the Bureau of Mines for 18 production districts
and two production district groupings [10]. Shipments to- each state
from districts 3 and 6 are reported as one grouping, and shipments to each
state from districts 22 and 23 are reported as another grouping. Ship-
ments from production district 5 to each state are not reported. The
two production district groupings are treated as single districts.
Average sulfur content of coal shipped from each district or district
grouping for consumption by retail and industrial sources (excluding elec-
tric utilities and coke and gas plants) is completed as follows:
w s w s
Wi Wi
where f, . = Sulfur content of coal shipped from production
lfl district i
for use oy retail
and industrial sources (%)
W. = Shipments of bituminous coal from district 1
"h for use by the Other Industrial Uses and Retail
Dealers category [75]
W. = Shipments of bituminous coal from district i for
n2 use by the All Other Uses category [75]
S. = Sulfur content of coal from district i for use
1-l by the Other Industrial Uses and Retail Dealers
category (%) [75]
S. = Sulfur content of coal from district i for
^2 use by the All Other Uses category [75]
2-24
-------
or, for production district grouping i comprised of districts j and k,
VJ1 + \
W. W. W. W.
Jl + Kl + J2 + 2
Ash content of bituminous coal from each production district is computed
by averaging the ash content of coal produced by mines sampled in the
district [47]:
1 n
Ai n Z ak
J n k=1 K
where a, = Ash content of coal from the k — mine in dis
K trict j (%) [47]
n = number of mines sampled in district j [47]
A. = Average ash content of bituminous coal from
J district j (%)
Average ash content of coal from the two production district groupings are
then computed:
F = (Wi Wi } A (\ Wk ) A
r ( J JJ A i K KJ A
W. W. W. W
Jl + J2 + 1 + 2
where f^ - = Ash content of coal from production district
' grouping i composed of districts j and k (%)
For the 18 production districts that are not grouped,
f2,1 ' Ai
(2) State averages of sulfur and ash content are computed
separately for coal shipped for use by retail area sources and coal shipped for
industrial area sources. Coal shipments to each state from each production dis-
trict/district grouping are reported separately for retail users and industrial
users excluding electric utilities and coke and gas plants. Sulfur and ash,
by weight, in coal used by commercial and industrial point sources is available
from the NEDS point source data. Average sulfur and ash content of coal shipped
to each state is calculated by user category as follows:
2-25
-------
20
(°-01 * fikcjk
X,, = 100X ^-!
20
where i = 1 for sulfur, 2 for ash
j = 1 for retail, 2 for industrial excluding
electric utilities and coke and gas plants
k = production district/district grouping number
x.. = Sulfur (i=l) or ash (i=2) content of coal
shipped to the state for use by area sources
in user category j (%)
f-k = Average sulfur or ash content of coal from
production district/district grouping k for
use by retail and industrial services (%)
c.. = Coal shipments to the state from production
J district/district grouping k for user category
j (103 tons) [10]
x . • = Sulfur or ash by weight in bituminous
P* J coal consumed in the state by point sources in
in user category j (ICr tons) (NEDS)
c .. = Bituminous coal consumed in the state by point
p' J sources in user category j (103 tons) (NEDS)
(3) Countywide sulfur and ash content are apportioned from state
sulfur and ash content according to the relative mix of coal used in the county
by retail and industrial area sources:
*" b*
V = — - ---
yi 2
I b' •
J-l CJ
where y- = Sulfur (i = 1) or ash (i = 2) content of bitu-
minous coal used in the county (%)
b' . = County consumption of bituminous coal by
J retail (j = 1) or industrial (j = 2) sources
x.. = Sulfur (i=l) or ash (i=2) content of coal shipped
^ to the state for use by area sources in user
category j (%)
2-26
-------
The value of b1 -, is the sum of the normalized county residential and com-
mercial bituminous coal consumption, which was calculated using the methodo-
logies described in Sections II.A.l.b and II.A.2. The value of b' is the
C£
normalized county industrial consumption calculated using the methodology
described in Section II.A.I.e.
b. Sulfur and Ash Content of Athracite Coal
Because there is only one anthracite producing region in the
country (located in Southeastern Pennsylvania), one value each for sulfur con-
tent and ash content is used for all counties. The values are obtained from
the Bureau of Mines publication, "Distribution of Pennsylvania Anthracite" [g].
10. Aircraft Landing and Take-Off Cycles
County landing and take-off cycles (LTOs) are calculated separately
for civil, commercial, and military aircraft categories. One of two methods
is used to determine county LTOs.
a. For counties with FAA regulated airports and/or military air-
ports, LTOs are derived directly from reported operations:
LTOi = 0.5 x Oi
where LTO. = County LTOs for aircraft category i (i = 1
for commercial aircraft, i = 2 for civil
aircraft, i = 3 for military aircraft)
0. = Total operations in county for aircraft
category i
b. For counties with no FAA regulated airports or miliary airports,
all operations in the county are assumed to involve civil aircraft only. The
number of LTOs for civil aircraft is calculated from the county aircraft regis-
trations:
LT02 = 365 * N
where N = the number of aircraft registered in the county.
2-27
-------
B. THE DATA BASE
The data base for the allocation methodologies is divided into three
major categories:
. National and regional data
. State data
. County data
Whenever possible, Wai den has attempted to use data which are updated annually
or more often. The national and regional data include national and regional
fuel consumption figures, fuel consumption rates, and sulfur and ash contents.
The state data include fuel consumption, socioeconomic, climatological, and
demographic figures. The county data include primarily climatological and
demographic data. Tables 2-2, 2-3, and 2-4 contain a list of all data items
required for the allocation methodologies.
Table 2-5 summarizes the major sources used for obtaining data. Table
2-6 lists the contacts made with the various state highway or tax department
offices throughout the country. Table 2-7 lists sources for data available on
magnetic tape. Other sources are referenced throughout this report, but the
sources in Tables 2-5, 2-6, and 2-7 are essential to prepare the input to the
computer programs that perform the fuel use allocations. State fuel use figures
were primarily taken from Bureau of Mines publications. Other sources used
include Highway Statistics, published yearly by the Federal Highway Administra-
tion, and Synthetic Organic Chemicals, published yearly by the United States
Traffic Commission.
2-28
-------
TABLE 2-2
NATIONAL VARIABLES FOR ALLOCATION METHODOLOGIES
Description of Variable (Units)
Industry sulfur content of coal, production districts 1-23 (%)
Other sulfur content of coal, production districts 1-23 (%)
Industry coal production, production districts 1-23 (tons x 105)
Other coal production, production districts 1-23 (tons x 10 )
Ash content of bituminous coal, production districts 1-18 (%)
Ash content of bituminous coal, production districts 19-23 (%)
Sulfur content of anthracite coal (%)
Ash content of anthracite coal (%)
Annual usage of diesel tractors (hours/year)
Annual usage of gasoline tractors (hours/year)
Annual usage of general purpose—agricultural --equipment (hours/year)
Annual usage of harvesters (hours/year)
Annual usage of bailers (hours/year)
Annual usage of combines (hours/year)
Average gasoline consumption rate, tractors (gallons/year)
Average gasoline consumption rate, general purpose (gallons/year)
Average gasoline consumption rate, harvesters (gallons/year)
Average gasoline consumption rate, balers (gallons/year)
Average gasoline consumption rate, combines (gallons/year)
Percent tractors using gasoline fuel (%)
Percent tractors using diesel fuel (%)
Percent using gasoline, general purpose—agricultural (%)
Percent using gasoline, harvesters (%)
Percent using gasoline, balers (%)
Percent using gasoline, combines (%)
o
Fuel consumption, construction—gasoline (gallons x 10 )
o
Fuel consumption, cdnstruction—diesel (gallons x 10 )
Fuel consumption, industrial—gasoline (gallons x 10 )
o
Fuel consumption, industrial—diesel (gallons x 10 )
2-29
-------
TABLE 2-2 (continued)
NATIONAL VARIABLES FOR ALLOCATION METHODOLOGIES
Description of Variable (Units)
2
Fuel consumption, lawn and garden (gallons x 10 )
3
Fuel consumption, snowthrowers (gallons x 10 )
o
Fuel consumption, snowmobiles (gallons x 10 )
Usage—motorcycles, off road (miles/year)
Usage—motorcycles, combination (miles/year)
Gas mileage, motorcycles (miles/gallon)
Gas usage, inboard boats (gallons/hour)
Gas usage, outboard boats (gallons/hour)
Gas consumption by census region, cooking range (therms/year)
Gas consumption by census region, water heater (therms/year)
Regional percentage, off-road motorcycles (%)
Regional percentage, combination motorcycles (%)
Growth'by census region in LPG heat (%)
Growth by census region in LPG cooking (%)
Growth by census region in coal heat (%}
Solvent consumption by primary solvent group (pounds x 10 )
o
Bituminous coal consumption, steel mills (tons x 10 )
o
Bituminous coal consumption, cement plants (tons x 10 )
o
Bituminous coal consumption, other industrial (tons x 10 )
o
Bituminous coal consumption, retail (tons x 10 )
1971 industrial coal consumption, SIC 20-39, 19, and 39 (tons x 10 )
1971 industrial natural gas consumption, SIC 20-38, 19, and 39 (ft x 10 )
1971 industrial employment, SIC 20-38, 19, and 39
Univeristy employment/enrollment ratio
Solvent point source employment, SIC 26, 26, 27, 30, 24-39, total 19-39,
laundries, 243, 244, 371, 373, 7535, 264, 265
2-30
-------
TABLE 2-3
"STATE. VARIABLES FOR ALLOCATION METHODOLOGIES
Description of Variable (Units)
Current employment, SIC 19-39
Current employment, SIC 701 (hotels)
Current employment, SIC 7211, 7216, 7217 (laundries)
Current employment, SIC 806 (hospitals)
Current employment, SIC 821 (schools)
Current employment, SIC 822 (universities)
Current employment, SIC 60 and 7.0, minus above (other services)
Current employment, SIC 50 (wholesale)
Current employment, SIC 52 (retail)
Current employment, SIC 7215, 2*7216, 7218 (laundries for solvents)
Current employment, SIC 243 (millwork, .plywood, etc.)
Current employment, SIC 244 (wooden containers)
Current employment, SIC 371
Current employment, SIC 373
Current employment, SIC 7535
Current employment, SIC 10
Current employment, SIC 16
Current employment, SIC 264
Current employment, SIC 265
Employment data, SIC 19-39 for year of most recent Census of Manufacturers
Employment data, total, 19-39 for year of most recent Census of Manufacturers
3
Coal consumption data, SIC 19-39 (tons x 10 )
o
Coal consumption data, total, 19-39 (tons x 10 )
Gas consumption data, SIC 19-39 (ft.3 x 106)
Gas consumption data, total, 19-39 (ft.3 x 106)
Farms in irrigated areas
Tractors
Combines
Harvesters (corn huskers)
Pickup balers
2-31
-------
TABLE 2-3 (continued
STATE VARIABLES FOR ALLOCATION METHODOLOGIES
Description of Variable (Units)
Census year population
Gas-heated dwelling units
Coal-heated dwelling units
Elementary and kindergarten enrollment
High school enrollment
Coal shipments—retail total (103 tons)
o
Coal shipments—retail production district groupings 1-2Q (10 tons)
Coal shipments — industrial total (103 tons)
o
Coal shipments—industrial production district groupings 1-20 (10 tons)
Public school employment
Hotel employee/room ratio
Current population
Percent of gas customers with gas heat (%}
Additions to gas heating, each year since census year
Conversions to gas heating, each year since census year
Gas-heated dwelling units (previous year)
q C.
Natural gas consumption, residential (ft. x 10 )
Natural gas consumption, industrial (ft.3 x 10°)
Natural gas consumption, commercial (ft.3 x 106)
Natural gas consumption, other (ft.3 x 106)
q
LPG consumption, industrial (gallons x 10°)
q
LPG consumption, retail (gallons x 10 )
Anthracite coal shipments, retail (tons)
Bituminous coal shipments, industrial (103 tons)
Bituminous coal shipments, retail (104 tons)
Anthracite market share (%)
^
Gasoline consumption, highway (gallons x 10J)
Gasoline consumption, off-highway (gallons x 103)
o
Gasoline consumption, construction equipment (gallons x 10 )
Gasoline consumption, commercial-industrial (gallons x TO3
2-32
-------
TABLE 2-3 (continued)
STATE VARIABLES FOR ALLOCATION METHODOLOGIES
Description of Variable (Units)
Gasoline consumption, agricultural (gallons x 103)
Gasoline consumption, aviation (gallons x 103)
Railroad use of diesel fuel (bbl x 103)
Registrations, motorcycles
Registrations, snowmobiles
Registrations, inboard boats
Registrations, outboard boats
Census region identifier
Centroid county snowfall
Coastline
Coastline area factor
Point Source Data:
Point source employment, SIC 19-39
Point source employment, total, 19-39
Point source employment, SIC 701, (724 + 7216 + 7287), 806, 821, 822,
other sources, 50, 52, (7215 + 2 x 7216 + 7218), 243, 244, 371,
373, 7535, 10, 16, 264, 265
Bituminous coal, commercial consumption (tons)
Sulfur content (tons)
Ash content (tons)
Bituminous coal, industrial consumption (tons)
Sulfur content (tons)
Ash content (tons)
Natural gas consumption, commercial (ft.3 x 10 )
Natural gas consumption, industrial (including LPG) (ft.3 x 106)
LPG consumption, commercial (gallons x 103)
Anthracite coal consumption, commercial (tons)
2-33
-------
TABLE 2-4
COUNTY VARIABLES FOR ALLOCATION METHODOLOGIES
Description of Variable (Units)
Degree days
Number of days with temperatures less than 32°F
Number of "warm" months
Snowfal1
Current employment, SIC 19-39
Current employment, total, 19-39
Current employment, SIC 701 (hotels)
Current employment, SIC 7211, 7216, 7217 (commercial laundries)
Current employment, SIC 806 (hospitals)
Current employment, SIC 821 (schools)
Current employment, SIC 822 (universities)
Current employment, SIC 60 + 70 minus above 5 other services
Current employment, SIC 50 (wholesale)
Current employment, SIC 52 (retail)
Current employment, SIC 7215 + 2 x 7216 + 7218 (laundries for solvents)
Current employment, SIC 243 (mi 11work, plywood, etc.)
Current employment, SIC 244 (wooden containers)
Current employment, SIC 371 (motor vehicles and equipment)
Current employment, SIC 373 (ship and boat building and repairing)
Current employment, SIC 7535 (paint stores)
Current employment, SIC 10 (mining)
Current employment, SIC 16 (heavy construction)
Current employment, SIC 264 (miscellaneous connected paper products)
Current employment, SIC 365 (paper board containers and boxes)
Hospital beds
Hospital employment
Public university enrollment
Population density
Kindergarten and elementary school enrollment
High school enrollment
2-34
-------
TABLE 2-4 (continued)
COUNTY VARIABLES FOR ALLOCATION METHODOLOGIES
Description of Variable (Units)
Year-round housing units
Median rooms per dwelling unit
% Rooms in 1-unit structures
Farms
Farms with sales greater than or equal to $2500
Census year population
Number of occupied dwelling units
Number of occupied dwelling units with gas heat
Number of occupied dwelling units with LP6 heat
Number of occupied dwelling units with oil heat
Number of occupied dwelling units with coal heat
Number of occupied dwelling units with natural gas ranges
Number of occupied dwelling units with LPG ranges
Number of occupied dwelling units with natural gas hot water heaters
Number of occupied dwelling units with LPG hot water heaters
Current population
Tractors
Gross revenues of service stations ($) or retail gasoline consumption
(gallons x 103 )
Air carrier and taxi operations
General aviation operations
Military
Aircraft registrations
Inland water area
Coastline
NEDS Point Source Data:
Point source employment, SIC 19-39
Point source employment, total, 19-39
Point source employment, SIC 701, (7211 + 7216 + 7217), 806, 821, 822, other
services, 50, 52, (7215 + 2 x 7216 + 7218), 243, 244, 371, 373, 7535, 10,
16, 264, 265
Point source employment for solvents, SIC 25-27, 30, 34-39
2-35
-------
i
CO
cr>
TABLE 2-5
SOURCES REQUIRED FOR INPUT PREPARATION, 1973
1.
2.
3.
4.
5.
6.
7.
Source
Coal - Bituminous and Lig-
nite, Annual (preprint),
Bureau of Mines, Washington,
DC 20240
Analysis of Tipple and
Delivered Samples of Coal,
Bureau of Mines, Washington,
DC (Report of Investigations
Series)
Keystone Coal Industry,
McGraw-Hill, Inc.
Use of Gas by Residential
Appliances. American Gas
Association, Arlington, VA
Motorcycle Usage and Owner
Profile Study, prepared
for Motorcycle Industry
Council
Petroleum Statement, Annual
(final summary), U.S. Bureau
of Mines, Washington, DC
20240
Synthetic Organic Chemicals,
U.S. Production and Sales,
Approximate
Date Available
Early January 1975
1973-1974
Quintennial
(Most Recent
1972)
Annual
February, 1975
Late 1975
Application
Sulfur & Ash
Sulfur & Ash
Sulfur & Ash
Residential
LPG
Off-Highway
Solvents
Solvents
Cost Availability
Free Leonard W. Westerstrom
Division of Fossil Fuels,
(703) 557-1350
$ 0.50 U.S. Government Printing
Office, Washington, DC
20402
$60.00 Mining Information Service,
McGraw-Hill, New York, NY
Free Robert Griffiths, Statistics
Department, (703) 524-2000,
X348
Hendrix Tucker Walker,
7447 North Figurroa Street,
Los Angeles, CA 90041
(213) 254-9217
Free Betty M. Moore, Division of
Fossil Fuels, (703) 557-1667
$ 2.40 Chemicals Division, (202)
523-0387
U.S. International Traoe
Commission, Washington, DC
20436
-------
TABLE 2-5 (continued)
SOURCES REQUIRED FOR INPUT PREPARATION, 1973
Source
Approximate
Date Available
Application
Cost
Availability
ro
i
CO
8. Fuels and Electric Energy
Consumed, Census of Manu
factures, U.S. Bureau of
the Census, Washington,
DC 20233
9. County Business Patterns,
$
U.S. Bureau of the Census,
Washington, DC 20233
10. Statistical Abstract of
the United States, U.S.
Bureau of the Census,
Washington, DC 20233
11. Natural Gas Production
and Consumption, U.S.
Bureau of Mines,
Washington, DC 20240
12. Bituminous Coal and Lig-
nite Distribution (annual),
U.S. Bureau of Mines,
Washington, DC 20240
13. Fall 1973 Statistics of
Public School Systems,
U.S. Department of Health,
Education, and Welfare,
Washington, DC
14. Statistics of State School
Systems, 1969-1970, U.S.
Department of Health,
Education and Welfare,
Washington, DC
Quintennial
(Most recent
1972)
Annual
Annual
August 1974
April 1974
1975
Industrial Arthur Horowitz, (301)
763-7666
1973
Industrial
Off-Highway
Residential,
Commercial,
Industrial
Sulfur & Ash
Commercial
$100.00 U.S. Government Printing
Office, Washington, DC
20402
$ 6.30 U.S. Government Printing
Office, Washington, DC
20402
Free Leonard L. Fanelli, Divi-
sion of Fossil Fuels,
(703) 557-1454
Free Leonard W. Westerstrom,
Division of Fossil Fuels,
(703) 557-1350
U.S. Office of Education,
Department of Health,
Education, and Welfare
Commercial U.S. Office of Education,
Department of Health,
Education and Welfare
-------
TABLE 2-5 (continued)
SOURCES REQUIRED FOR INPUT PREPARATION, 1973
Source
Approximate
Date Available
Application
Cost
Availability
i
CO
00
15. Subject Reports, U.S.
Census of Selected
Services, U.S. Bureau
of the Census,
Washington, DC
16. Population Estimates
(Series P-26) , U.S. Bureau
of the Census, Washington
DC
17. Gas House Heating Survey,
Department of Statistics,
American Gas Association,
Arlington, VA 22209
18. Sales of Liquified Petro-
leum Gas and Ethane^
U.S. Bureau of Mines,
Washington, DC
19. Distribution of Pennsylvania
Anthracite for the Calendar
Year, U.S. Bureau of Mines,
Washington, DC
20. Highway Statistics, Federal
Highway Administration,
Washington, DC
21. Sales of Fuel Oil and
Kerosene, U.S. Bureau of
Mines, Washington, DC
Quintennial
(Most recent
1967)
Annual
1974
September 1975
December 1974
1975
September 1974
Commercial
Residential,
Solvents, Off-
Highway,
Rai1 roads
Residential
Residential,
Commercial,
Industrial
Residential,
Commercial
Railroads
U.S. Government Printing
Office, Washington, DC
$30.00 U.S. Government Printing
Office, Washington, DC
Free Robert Griffith, Department
of Statistics, (703) 524-2000
Free Leonard L. Fanelli, Division
of Fossil Fuels, (703)
557-1454
Free Dorothy R. Federoff, Division
of Fossil Fuels, (703) 557-
3562
Retail Sales $ 3.20
of Gasoline
L. French, (202) 426-0180
Free James M. Diehl, Division of
Fossil Fuels, (703) 557-0443
-------
ro
co
i-D
TABLE 2-5 (continued)
SOURCES REQUIRED FOR INPUT PREPARATION, 1973
22.
23.
24.
25.
26.
27.
28.
Source
The Marine Market, 1972,
Marex (International Marine
Expositions, Inc.), Chicago,
IL.
Boating 1972, Marex and
National Association of
Engine and Boat Manu-
facturers
Census of Agriculture, U.S.
Bureau of the Census,
Washington, DC
AHA Guide to the Health
Care Field, American
Hospital Association,
Chicago, IL
The College Blue Book,
MacMillan Information,
New York, NY
Census of Retail Trade,
Area Statistics
FAA Air Traffic Activity,
Approximate
Date Available
April 1973
Quintennial
(Most recent
1972)
Annual
Annual
Quintennial
(Most recent
1972)
February 1974
Application Cost Availability
Off -Highway
Off-Highway U.S. Government Printing
Office, Washington, DC
Commercial John A. Henderson, Director
of Marketing Services,
(312) 645-9400
Commercial ~$10.00 MacMillan Publishing
Company, Inc., 866 Third
Avenue, New York, NY 10022
Retail Sales ~$100.00 U.S. Government Publishing
Office, Washington, DC
LTOs $4.55 U.S. Government Printing
Federal Aviation Administra-
tion, Washington, DC
Office, Washington, DC
-------
TABLE 2-5 (continued)
SOURCES REQUIRED FOR INPUT PREPARATION, 1973
Source
Approximate
Date Available
Application
Cost
Availability
i
-l^
o
29. Military Air Traffic Acti- 1974
vity Report, Federal Avia-
tion Administration,
Washington, DC
30. Census of U.S. Civil Air- 1975
craft, Federal Aviation
Administration, Washington,
DC
31. Area Measurement Reports, 1970
U.S. Bureau of the Census,
Washington, DC
LTOs
LTOs
Free
$2.85
Off-Highway $0.25
Betty Cayce, Office of
Management Systems, (202)
U.S. Government Printing
Office, Washington, DC
U.S. Government Printing
Office, Washington, DC
-------
TABLE 2-6
CONTACTS FOR RETAIL SALES OF GASOLINE DATA
State
Contact
Arizona
Florida
Georgia
Louisiana
Minnesota
New Mexico
Mr. Dave Tweedie
Gas Tax Auditor
1739 W. Jackson
Pheonix, Arizona 85007
State of Florida Gas Bureau
Department of Revenue
Tallahassee, Florida
(904) 488-7417
Curtis B. Modi ing, Director
Motor Fuel Tax Unit
Department of Revenue
318 Trinity Washington Building
Atlanta, Georgia 30334
Richard L. dousing, Supervisor
Special Fuels Tax Unit
Department of Revenue
P.O. Box 201
Baton Rouge, Louisiana 70821
(504) 389-6223
James F. Dagen, Director
Petroleum Division
Minnesota Department of Taxation
Centennial Office Building
Saint Paul, Minnesota 55101
C. Tampin
Bureau of Revenue
State of New Mexico
Baatan Memorial Building
Santa Fe, New Mexico 87501
2-41
-------
TABLE 2-7
MAGNETIC TAPES REQUIREMENTS
Name and Source
Frequency
Cost
Contact
County and City Data Book
Tape
U.S. Bureau of the Census
County Business Patterns
U.S. Bureau of the Census
Census of Housing and
Population
U.S. Bureau of the Census
1009 Name Tape
National Oceanic and
Atmospheric Administration
Monthly Climatological Data
National Oceanic and Atmos-
pheric Administration
Point Source Fuel Consump-
tion
Point Source Employment
CBP-SAROAD-- ,^
NOAA-SAROAD -- ?-Geographic code
GSA-SAROAD"'' index tapes
Quintennial
Latest: 1972
Annual
Decennial
Annual
Annual
Annual
Annual
$ 70
$580
$500
$ 60
$ 60
Helen Tier
(301) 763-5475
Mr. Schieldal
Mr. Norton
(704) 254-0961
Mr. Norton
(704) 254-0961
NADB
NADB
NADB
2-42
-------
III. DEVELOPMENT OF METHODOLOGIES
A. RESIDENTIAL
Development of the methodology for allocation of statewide residen-
tial fuel use within counties is described in this section for each of three
primary fuels: natural gas, LPG, and coal.
1 . Natural Gas
a. Regression Analysis
An algorithm was developed from regression analysis for es-
timating county consumption of natural gas by residential sources. The as-
sumption was made in the analysis that natural gas consumption by residential
users is a function of climatological and housing stock descriptive vari-
ables. The data sample was obtained from a number of gas companies which
distribute gas to about 1,000 communities. Screening the data for complete-
ness reduced the sample size to approximately 300 communities. The candidate
independent variables considered for inclusion in the regression equation
were annual degree days, average wind speed in January, number of dwelling
units using gas for space heating, number of dwelling units using gas for
water heating, number of dwelling units using gas for cooking, percent of
dwelling units in structures built after 1960, number of rooms per dwelling
unit, percent of dwelling units in single-unit structures, percent of annual
growth of gas-heated dwelling units in the state, latitude, and average ele-
vation. Details of the regression analysis results are given in Appendix A.
The resultant algorithm relating natural gas consumption to
the most significant regressors is expressed as
fcn(T) = 3.57 + 0.367 &n(D) + 0.588 £n
u.
Ux
+ 0.125
where an = log to the base e
T = gas consumption per dwelling unit (therms)
D = annual degree days [3]
3-1
-------
U h = number of dwelling units using gas for space heating
a [4]
U = the larger of the number of dwelling units using gas
for cooking or for heating water [4]
Rd = median number of rooms per dwelling unit [5]
The composite variable Uqn/Ux is used to reflect the variation in energy
consumption per dwelling unit between communities with similar climates,
but with different percentages of gas customers using gas for space heating.
The total residential consumption of natural gas for a
county is calculated as follows:
NGc = 9.69xl(T5TngUg
where NG = county residential consumption of natural gas (10
cu. ft.)
_c
9.69 x 10 = factor to convert heat equivalent of gas to
106 cu. ft.
U = total number of dwellings in county that use gas
The total number of dwelling units in the county that use gas is calculated
from the number of occupied dwelling units using gas for space heating and
the fraction of statewide residential gas customers using gas for heating,
viz.,
U9 = 7 ugh
where U = number of occupied dwelling units in the county using
9 gas
f = fraction of statewide residential gas customers with
gas heat [30]
b. Updating the Housing Stock Data
If the year of interest corresponds to the decennial census,
the number of dwelling units using gas for space heating is reported directly
[4]. A method of updating this variable during intervening years was de-
veloped. This procedure is illustrated by the following example:
3-2
-------
AU
gh
to
AP.
A +
Uc
U"
where U1
U
gh
gh
AU
gh
AP_
number of dwelling units using gas for space heating
in the year of interest
number of dwelling units using gas for space heating
in the census year £4]
total increase since the census year (1970) in dwell-
ing units using gas for space heating
increase in county population since the census year
(0 if there was a decrease or no change in popula-
tion)
Ps s
A =
C =
number of additional gas-heated dwelling units in
the state due to new housing starts since the census
year [30]
number of conversions to gas space heating in the
state since the census year [30]
statewide number of dwelling units using gas for
space heating in the census year [4]
An alternate method for distribution of statewide additions of gas-heated
dwelling units was investigated using housing authorized by building per-
mits [31]. At the time of this study, the Construction Statistics Division
of the Census Bureau does not compile these data for all areas of the coun-
try on a routine basis, and so this alternative was abandoned.
The composite housing stock variable U ^/U and the median
number of rooms per dwelling unit are not updated for years intervening be-
tween census years. That is, the value of U ^ used in the composite housing
stock variable is the census year number of housing units using gas for
space heating.
3-3
-------
c. Normalization
Estimates of county residential natural gas consumption are
normalized against the state total published in the Mineral Industry Survey
(M.I.S.): Natural Gas Production and Consumption [6J.
N6C
NGc' = NG^Xs
where NG ' = normalized county residential consumption of
natural gas (10° cu. ft.)
NG = unnormalized county residential consumption of
natural gas
NG. = I NG
c
X = published state total residential consumption of
natural gas (106 cu. ft.) [6]
Because this normalization apportions the published state total to counties
according to the ratio of calculated consumption for individual counties to
total calculated county consumption in the state, the conversion of the es-
timated county consumption in therms to the gas equivalent in 10 cu. ft.
is not necessary.
2. LPG
The algorithm for estimating countywide natural gas consumption
is considered inappropriate for LPG application, due to limitations in the
available data. For example, the variable median rooms per dwelling unit is
reported only for the whole county and is not cross-tabulated by fuel type;
the small percentage of the housing stock using LPG argues against using
this countywide figure. Also, it would be difficult to interpret and apply
the composite housing variable for gas, U n/Ux, to LPG. Consequently, a
simpler estimation procedure was developed based on energy consumption sta-
tistics compiled by the American Gas Association (AGA).
Central heating load, expressed as a function of degree days,
has been compiled by the AGA [7] for the entire country and is reproduced in
3-4
-------
Table 3-1. Using the data for the average load category, the following al-
gorithm for computing energy consumption per dwelling unit was formulated:
Th = 376 + 0.209 D
where T^ = energy consumption for space heating (therms per
dwelling unit)
D = annual degree days
Cooking and water heating requirements are added to this space heating com-
ponent to account for total residential demand. The formula for computing
county residential LPG consumption is
T£ = (376 + 0.209 D) U^h + cJU w + c^
where T{ = county residential consumption of LPG (therms)
D = annual degree days
U.u = number of occupied dwelling units in the county that use
LPG for space heating [4]
c~ = regional average consumption by hot water heating
w (therms/unit) [7]
Uj = number of occupied dwelling units in the county that use
LPG for heating water [4]
c" = regional average consumption by cooking ranges (therms/
r unit) [7]
Up = number of occupied dwelling units in the county that use
LPG for cooking [4]
3
LPG consumption is converted to 10 gallons by the expression
LPGC = 0.00105 T^
where 0.00105 = the factor to convert heat equivalent of LPG
to 103 gallons (TO3 gallons/therm)
The number of occupied dwelling units in each county is available from the
census of housing report for the'most recent census year [4]. These data
are not updated for the intervening years between census years. Regional
average consumption per dwelling unit for cooking and water heating purposes
is derived from data compiled by the AGA on average consumption by residential
3-5
-------
TABLE 3-1
*
CENTRAL HEATING LOAD BY HEATING DEGREE DAYS
(Therms per Year)
Mean Seasonal Minimum Average Maximum
Degree Days Load Load Load
0 70 376 500
500 155 480 612
1,000 240 585 725
1,500 325 689 837
2,000 410 794 950
2,500 495 899 1,062
3,000 580 1,004 1,175
3,500 665 1,109 1,287
4,000 750 1,213 1,400
4,500 835 1,318 1,512
5,000 920 1,422 1,625
5,500 1,005 1,527 1,737
6,000 1,090 1,631 1,850
6,500 1,175 1,736 1,962
7,000 1,260 1,840 2,075
7,500 1,345 1,945 2,187
8,000 1,430 2,049 2,300
8,500 1,515 2,154 2,412
9,000 1,600 2,258 2,525
9,500 1,685 2,363 2,637
10,000 1,770 2,467 2,750
This table was computed from actual househeating load studies and
load estimates of 74 companies located across the United States
during 1971.
* Source: American Gas Association, Department of Statistics
3-6
-------
appliances. The appropriate statistics are from columns 3 and 5 of Table
3-2.
State residential consumption of LPG is not published, but
retail (commercial and residential) consumption of LPG is [8J. If the sum
of the county LPG consumption estimates exceeds the published state total,
the county figures are normalized against the published total. Otherwise,
the county estimates are left unchanged, and the remainder of the published
state retail consumption of LPG is used as the commercial consumption
figure for the state.
3. Coal
The basic approach taken in developing methodologies for al-
location of residential consumption of anthracite and bituminous coal to
individual counties was to establish a functional relationship between coal
consumption per dwelling unit and degree days, to adjust housing data for
secular trends in the number of doal-heated dwelling units, and to disag-
gregate the total coal consumption into anthracite and bituminous components
and normalize the results as necessary.
a. Development of a Relationship between Coal Consumption
and Degree Days
Four approaches were investigated in the search for an ac-
ceptable expression for residential coal consumption in terms of degree days.
(1) The first approach was modeled after the approach used
in developing the methodology for allocation of residential LPG. The method
is based on the average central heating load data compiled by AGA [7]. Res-
idential cooking and water heating uses of coal, however, are assumed to be
negligible and are, therefore, excluded from the algorithm. To correct for
the lower efficiency of coal in relation to gas in space heating systems, a
factor for inflating the therm requirement for space heating by coal was
needed. A factor of 1.33 was derived from the information published by the
Independent Natural Gas Association of America [32] (reproduced in Table 3-3)
3-7
-------
CO
TABLE 3-2
USE OF GAS BY RESIDENTIAL APPLIANCES
(Average Consumption, Excluding Extremes)
(Therms per Year)
United States
1971 Survey
1966 Survey
New England
1971 Survey
1966 Survey
Middle Atlantic
1971 Survey
1966 Survey
East North Central
1971 Survey
1966 Survey
West North Central
1971 Survey
1966 Survey
South Atlantic
1971 Survey
1966 Survey
East South Central
1971 Survey
1966 Survey
West South Central
1971 Survey
1966 Survey
1971
Residential
Customers
23,511,693
27,027,000
1,208,248
NA
4,623,211
NA
6,578,977
NA
1,717,732
NA
1,969,491
NA
807,061
NA
2,661,538
NA
House
Range
105
106
101
101
117
102
101
105
87
101
95
91
127
119
112
122
Apt.
Range
88
74
84
72
81
71
90
64
69
67
77
72
92
107
106
84
Water
Heatef
316
274
242
245
318
282
317
288
341
273
358
241
287
295
319
236
Clothes
Dryer
(Gas Pilot)
75
90
92
91
65
90
69
88
75
86
66
89
60
74
83
84
Clothes
Dryer
(Elec. Pilot)
60
52
76
49
65
52
58
46
52
53
40
49
63
60
63
56
Incinerator
130
138
130
156
119
143
147
144
139
134
158
132
116
114
162
96
Gas
Light
181
183
155
195
174
177
192
189
157
175
202
187
196
174
179
186
A1r
Conditioner
Consumption
Per Ton
283
308
216
209
226
193
200
236
269
284
252
443
343
361
465
479
Gas
Grill
26
29
33
29
28
24
23
23
17
33
32
31
33
22
35
42
Gas Heat
All Types
1,192
NA
1,462
NA
1,313
NA
1,539
NA
1,178
NA
1,022
NA
865
NA
660
NA
-------
CO
I
TABLE 3-2 (continued)
USE OF GAS BY RESIDENTIAL APPLIANCES
(Average Consumption, Excluding Extremes)
(Therms per Year)
' 1971 House Apt.
Residential Range Range
Customers
Mountain
1971 Survey
1966 Survey
Pacific
1971 Survey
1966 Survey
Note: A total of 157
the 39,194,000
1,104,700
NA
2,840,735
NA
103
111
102
118
companies with a total
residential customers
86
73
96
89
of 23
1h the
Water
Heater^
Clothes Clothes
Dryer Dryer
(Gas Pilot) (Elec. Pilot)
261 120
319 107
329 80
278 96
,511,693 residential
United States (1971
77
65
54
46
customers provided
Incinerator
150
136
171
160
data for this
Gas Air Gas Gas Heat
Light Conditioner Grill All Types
Consumption
Per Ton
179
180
178
188
summary.
261 28 1,079
399 26 NA
249 16 841
264 34 NA
This represents 60 percent of
Source: American Gas Association, Department of Statistics, February 1973
-------
TABLE 3-3
SINGLE-FAMILY DWELLING UNIT THERMAL EFFICIENCIES OF GAS AND COAL
Baltimore, Maryland
Boise, Idaho
Brooklyn, New York
Cambridge, Massachusetts
Charles town, South Carolina
Chicago, Illinois
Cleveland, Ohio
Danville, Virginia
Davenport, Iowa
Denver, Colorado
Des Moines, Iowa
Detroit, Michigan
Erie, Pennsylvania
Fort Wayne, Indiana
Grand Rapids, Michigan
Indianapolis, Indiana
Kansas City, Missouri
Louisville, Kentucky
Lowell, Massachusetts
Madison, Wisconsin
Memphis, Tennessee
Milwaukee, Wisconsin
Missoula, Montana
Nashville, Tennessee
Oklahoma City, Oklahoma
Peoria, Illinois
Providence, Rhode Island
Pueblo, Colorado
Richmond, Virginia
Rochester, New York
Salt Lake City, Utah
St. Louis, Missouri
St. Paul , Minnesota
Seattle and Tacoma, Washington
Shelby, North Carolina
Southeast, Michigan
Spokane, Washington
Topeka, Kansas
Tulsa, Oklahoma
Gas
80
75
70
75
80
75
70
80
75
75
80
75
70
75
72
75
75
75
75
80
60
75
75
70
80
80
75
75
80
80
80
71
75
75
80
75
75-78
80
.80
Coal
55
65
50
60
55
50
50
60
60
60
55
55
50
60
60
50
55
50
60
55
50
60
61.4
60
54
60
55
60
60
65
65
57
60
62
45
60
55-60
60
54
3-10
-------
by taking the average of the ratios of gas thermal efficiency to coal
thermal efficiency. With these adjustments, the resulting residential
coal consumption formula becomes:
C = 1.33 * (0.209 D + 376) U * h (i)
where C = county consumption of residential coal (therms)
D = annual degree days [3]
U , = number of occupied dwelling units using
coal for space heating £4]
h = 0.00387 (factor to convert consumption from
heat equivalent in therms to coal in tons)
Comparison of this coal consumption formula with the recent EPA method [18]
represented by
C = 0.0012 * D * UCQal (ii)
showed that method (i) produces higher estimates than the EPA's. The un-
availability of actual consumption data for coal-heated dwelling units im-
posed a constraint on any effort to test the reliability of either equation.
(2) The second method for estimating coal consumption is
based on data from a 1971 survey conducted by the Independent Natural Gas
Association of America (INGAA). The INGAA requested from local gas com-
panies the average coal, oil, electricity, and gas consumption for a typical
house in 64 cities £32] . Coal figures were reported for 38 cities. Accord-
ing to the INGAA, these figures represent actual consumption data obtained
from local coal dealers by the gas distribution company in each city. A
plot of these data points and a regression line (labeled (iii)) are shown
in Figure 3-1. The equation for the line, expressed as
C = 4.15 + 0.00044 * D * U (iii)
indicates a markedly lower slope for coal consumption than either method
(i) or the EPA formula (labeled (i) and (ii) in Figure 3-1, respectively).
Since the INGAA equation was developed from actual coal consumption data,
3-11
-------
FIGURE 3-1. COMPARISON OF FOUR CANDIDATE EQUATIONS FOR AVERAGE RESIDENTIAL
COAL CONSUMPTION
111 *
8*
1000
2000
3000 4000 5000
Annual Degree Days
6000
7000
-------
it is considered to provide closer approximation of coal consumption than
the other two formulas for areas with annual degree days above 3,000. How-
ever, due to the lack of data for areas in the lower degree day range, the
INGAA equation is considered unreliable for counties with annual degree
days below 3,000.
By performing a regression on the INGAA data, but' imposing
the constraint of a zero intercept (i.e., no coal consumption in areas with
zero degree days), and using a reciprocal logarithmic transformation of the
form
fcn y = a - 3 7
a formula that is more realistic for areas with low degree days was obtained:
2.13 -
C = el (iv)
However, this formula, labeled (iv) in Figure 3-1, yielded a rather unsatis-
o
factory coefficient of determination (R ) of 0.267, which prompted further
efforts to resolve the unexplained variation.
(3) The third method involved the specification of addi-
tional regressors. Two dichotomous variables and three housing variables
were used. One of the dichotomous variables was assigned a value of 1 or
0, depending on the type of coal. This variable was included to test for
a different pattern of consumption for anthracite; it was expected to be un-
correlated with consumption. The second dichotomous variable was assigned
a value of 1 for hand-fored and 0 for stoker-fired. This variable was ex-
pected to be positively correlated with consumption.
The housing variables used were percentage of dwelling
units in single-family structures, the median number of rooms per dwelling
unit, and the percentage of dwelling units in structures built before 1950.
For the purpose o.f this analysis, it was necessary to use statistics that
reflect the characteristics of the entire housing stock, because the re-
quired data were not disaggregated by type of fuel [4]. All three variables
3-13
-------
are expected to be positively correlated to consumption.
There was no statistically significant relationship found
between consumption and the variable related to type of coal. With the
exception of the variable related to percentage of dwelling units built
before 1950, the other two housing variables an the variable related to
stoking method had negative regression coefficients and a confidence in-
terval that included zero. Our a priori assumption for positive correla-
tion with consumption was, therefore, not supported by the analysis results
for all but one variable. The inclusion of the percentage of dwelling
units in structures built before 1950 produced the following results:
C = 0.00387 [7.81 - 13
where C = coal consumption per dwelling unit (tons)
p = percentage of dwelling units built before 1950 [4]
D = annual degree days
0.00387 = factor to convert consumption from heat equivalent in
therms to coal in tons
The t-statisties for the coefficients of —I and -i-j are
1.79 and 3.05, respectively; they are both above the 95% one-tailed critical
value of 1.69, allowing the rejection of the null hypothesis of the coeffi-
2
cient equaling zero. The inclusion of this housing variable raised the R
2
from 0.287 to 0.350. The R corrected for the number of degrees of freedom
increased from 0.260 to 0.310.
(4) Analysis for variation in the heating value of coal.
Variation in reported BTU content of coal ranges from 1.0 x 10 to 1.5 x 10
BTU/lb. A regression relating coal consumption in therms to degree days and
percentage of dwelling units built before 1950 resulted in the equation
*n (Tdu) =7.64-1000* (1]
where T. = coal consumption per dwelling unit in therms
This transforms to the equation
3-14
-------
Tdu = 0.00387 e^ u > (v)
The t-statistic for the coefficient of i- is 3.7; the R is 0.287; and the
2 2
corrected R is 0.266. The generally poor R is attributed to the coal data
in the sample. While the inclusion of the percentage of dwelling units in
2
structures built before 1950 did improve the R , the improvement was slight
and, therefore, did not warrant the expenditure of the extra effort associ-
ated with its inclusion. Equation (v) was selected for county allocation
purposes. The formula for total county residential coal consumption de-
rived from this equation is
C =0.00387 T Ucoal e^ ' D > (vi)
b. Disaggregation of Coal Consumption into Anthracite and
Bituminous Components
Total county residential consumption of coal is disag-
gregated into bituminous and anthracite using a state anthracite market
share factor:
ac • fa * C
bc - (1 - fa) * C
estimated coi
sumption (tons)
where a = estimated county residential anthracite coal con-
c
b = estimated county residential bituminous coal con-
sumption (tons)
C = estimated total county residential coal consump-
tion (tons)
f. = fraction of total state coal market that is anthra-
a cite
Appendix C contains an explanation of the development of the anthracite
market share factor.
3-15
-------
c. Adjustment for Secular Trend in Number of Coal-Heated
Dwelling Units
Annual Bureau of the Census estimates for the number of
coal-heated dwelling units will not be available until after 1975 [33].
Until the release of these figures, the secular trend of retail will be
used to adjust the 1970 Census of Housing coal-heated dwelling unit data
[34].
Time series regression analysis of the annual retail coal
shipments [56] for the period 1950-1972, with a lagged dependent variable,
yielded the following results:
Sy = -150.6 + 0.942 * S -, (vii)
o
where S and S -j = annual retail coal shipments (10 tons)
y y~ for years y and y-1, respectively
2
The R is 0.918 and is judged to be acceptable for time-
series application. To test for the possible violation of the assumption
of an uncorrelated error term, the Durbin-Watson statistic was calculated
to test the hypothesis of a non-autocorrelated error term. The resulting
value of 2.037 does not allow the rejection of this hypothesis at the 1%
level. Equation (vii) is used to estimate the decline in coal-heated hous-
ing stock. Neglecting the intercept, the number of dwelling units heated
by coal in any given year in this decade is given by
Ucoal,197x = °'942 Ucoal,1970
where x = the last digit of the year (e.g., x = 3 for 1973)
U -, 1Q7n = number of occupied dwelling units in 1970
coai,iy/u using coal for space neat1ng [4]
d. Normalization
It was assumed that retail coal is used primarily by resi-
dential sources. Therefore, all estimated residential is divided between
3-16
-------
anthracite and bituminous coal types before allotting any remaining retail
coal to commercial users. It was decided that, if estimates of one coal
type exceed the published state total, the excess represents consumption
of the other coal type by residential users. Thus, the excess is distri-
buted to the other coal type by county according to the distribution of
existing consumption estimates of that coal type. If, after any adjust-
ments, estimated consumption of either coal type still exceeds the pub-
lished total, the county consumption is normalized against the state total.
The procedure for the adjustment and normalization of residential coal con-
sumption is given in Section II.A.I.e.
B. COMMERCIAL-INSTITUTIONAL
Area source consumption of fuel by commercial and institutional
sources consists of all fuel burned in stationary sources which is not in-
cluded under the residential sources, industrial sources, power plants, or
commercial point sources. Statewide commercial area source consumption of
natural gas, LP6, anthracite coal, and bituminous coal is determined by sub-
tracting consumption by point sources in the state from published state con-
sumption totals of corresponding fuel type. The methodology developed for
allocating state commercial area source fuel consumption to individual
counties is summarized by the flow diagram in Figure 3-2. The five basic
steps in the methodology for allocation of commercial area source fuel con-
sumption are:
• Determining actual state commercial area source consumption of
each fuel type
• Estimating total fuel used by five major commercial-institutional
subcategories, namely hospitals, hotels, universities, schools,
and commercial laundries
• Apportioning the total fuel used by the five subcategories among
the four fuel types
• Determining normalization factors and statewide fuel use by all
commercial institutions and services other than the five major
subcategories
• Apportioning state consumption of each fuel by these "other" com-
mercial categories and adding it to the corresponding five sub-
category consumption.
3-17
-------
FIGURE 3-2. Commercial Allocation Methodology
rr'"''TY
r ----'I '
r--.
ospitar Bee
Hotel Rooms
POINT SOURCE
FUEL TOTALS
\RE/. SOURCE
Natural Gas
L.P.G.
Anthracite Coal
Bitu-ninous Coal
STATE
COMMERCIAL FUEL USE
/ Natural gas
J L.P.G.
Anthracite Coal
I Bitir.inc.'js Coal
AREA SOURCE STATE
CO:'.:;::-.C:AL FUEL cr
Mc>iui'ti i ysi
l.F.G.
Anthre cite Cc3l
Bituminous Ccal
COUNTY
CONSU.'-'PTH.,', FCP 5
Universities
Schools
Laur.dries
Hospitals
Hotels
ALLOCATlOfP
BY
RESIDENTIAL
RATIO
ARE/VSO'JPCE
COUNTY FUEL USE FOR
Natural Gas
L.P.G.
Anthracite Ccal
Bituninous Ccal
AHEA T SL-vi-CE
STATE COV.-.EP.CIAL FUEL USE
OTHER TH^;: 5 GROUPS
Nature! gas
L.P.G.
Anthracite Coal
Bituriircus Ccal
'ALLOCATIO
BY
(STATE-CC'J.VTY u-
EMPLOY;:E:;T
RATIO
tpf "«TO"frr
COUNTY co':::l:'KcTAl "TUEL US'E
OTHER Tli;1; 5 GR7JPS
Natural c?,s
L.P.G.
Anthracite Coal
Bitur.'.inobs Coal
Universities
Schools
Laundries
Hospitals, & Hotels
Schools
Laundries
Kocpitals
Hotels
POINT SOi'RCE
COl'tiTY
Schools
Laundries
Hospitals
Hotels
PCII;T SOURCE
.EMPLOYMENT
COUNTY RCSIDEf-TIAL
F'JELJJSJ
:-5l Gas
L.P.G.
Anthracite1 Coal
Bituminous Coal
COUNTY
CG.'-WEr.CIAL
Er'iPLOVJ'iEIIT
S'JFTRACT
POINT SOURCE'
- /REA'iGUr.CE
CCM'-'.ERCIAL
CO'j:rY CC'-'-'-'E^ClAL
L!^LOY"EJ,T CV.-'ER
THAN FIVE G.vIUPS
(AREA SOURCE)
COU;;TY APEA rci:r>cf
COMMERCIAL FL'F.L use
Natural Gas
L.P.G.
/•nth'-acite Co?.l
Coal
3-18
-------
1. Determining State Commercial Area Source Fuel Consumption
Xl - Gl + G2 - Gp
X2 = L - (Lr + Lp)
X3 - A - (,Ar + Ap)
X4 = B t (Br + Bp)
where X-| = state commercial area source consumption of natural gas
%2 = state commercial area source consumption of LPG
X- = state commercial area source consumption of anthracite
coal
X* = state commercial area source consumption of bituminous
coal
6-j = statewide sales of natural gas for commercial use [6]
Gg = other gas sales [6]
G = natural gas consumption by commercial-institutional
p point sources
L = statewide retail sales of LPG [8]
Lr = computed state consumption of LPG by residential
sources
L = LPG consumption by commercial-institutional point
p sources
A = shipments of anthracite coal to the state [9]
A = computed state consumption of anthracite coal by resi-
dential sources
A = anthracite coal consumption by all point sources
B = adjusted shipments of bituminous coal to retail dealers
B = computed state consumption of bituminous coal by resi-
dential sources
B = bituminous coal consumption by commercial-institutional
p sources
2. Estimating Total Fuel Consumed by Five Commercial Subcategories
The five major commercial subcategories were defined as hos-
pitals, hotels, universities, schools, and commercial laundries. Regression
3-19
-------
analyses were performed for each of these subcategories to determine any
linear correlation between fuel consumption and degree days, employment
within each category, and other independent variables peculiar to the sub-
category. Details of the analysis are discussed in Appendix B. The final
results are the five equations described in Section II. A. 2. a.
Total fuel consumption by the five commercial subcategories
is then computed by summing the five fuel consumption estimates:
5
where TC = total county fuel consumption by the five commercial
subcategories (therms)
T- = county total fuel consumption for commercial subcate-
gory j (j = 1 for hospitals, j = 2 for hotels, j = 3
for universities, j = 4 for schools, and j = 5 for
commercial laundries)
3. Distributing Total Fuel Consumption by the Five Commercial
Subcategories to Each Fuel Type
The total consumption by the five subcategories must then be
distributed among the four fuel types. Two methods of accomplishing this
were identified:
• Distribution according to the state commercial fuel use
pattern
• Distribution according to the residential fuel use pattern
of each county
Walden elected to use the latter because it reflects variation in fuel use
patterns between urban and rural counties. Relative consumption of the dif-
ferent fuel types is generally reflected by the proportion of dwelling units
using the corresponding fuel types for space heating. Data on the number of
dwelling units by heating fuel type, available from the Bureau of the Census
[4], are used for this apportionment, as follows:
3-20
-------
T -T
'i,c 'c
where T. = consumption of fuel type i in county c (therms)
i »c
U. = number of occupied dwelling units in county c
5 using fuel type i for space heating
Uc=Uo,c+ J, 0>
with X. = statewide consumption of fuel type i by commercial
area sources
The portion of total fuel consumption that is attributed to fuel oil is ac-
counted for by including dwelling units using fuel oil for space heating in
the total used in the denominator (U ).
4. Determining Normalization Factors and Consumption by Commercial
Sources Other Than the Five Major Subcategories
Statewide consumption of each fuel type by commercial sources
other than the five major subcategories is computed by subtracting the cal-
cualted state five-subcategory consumption from the actual state consump-
tion. If, however, the calculated five-subcategory consumption exceeds the
actual state consumption, the five-subcategory estimates must be normalized
against the actual state total. State fuel consumption by "other" commer-
cial sources and normalization factors are determined as follows:
If Xi > F., then F0i = Xi and ni = 1
If Xi ± F., then F0i = 0 and n. = ^-
where X^ = actual state commercial area source consumption of
fuel type i
F. = calculated state five commercial subcategory consump-
tion of fuel type i
FQ.J. = state consumption of fuel type i by "other" commercial
sources
n- = normalization factor for consumption of fuel type i by
five commercial subcategories
3-21
-------
5. Calculating County Area Source Consumption by Commercial-
Institutional Sources
Statewide fuel consumption by "other" commercial sources is
apportioned to the counties on the basis of employment in that category.
Total county consumption of each fuel type by commercial sources is ob-
tained by adding the county consumption by the five commercial subcate-
gories to the consumption by the "other" commercial sources:
'E.
= nifiTi,c
Es
where F. = county commercial area source consumption of fuel
lsC type i
f . = factor to convert fuel type i from therms to ap-
propriate NEDS units
E = county area source employment in "other" commer-
c cial category
E = state area source employment in "other" commercial
s category
C. INDUSTRIAL
A procedure was developed for allocating state industrial area
source consumption of natural gas, LPG, and bituminous coal. Anthracite
coal consumed by industry is almost entirely used by large point sources,
implying that industrial area source consumption of anthracite coal is neg-
ligible. This approach has been adopted here and is consistent with the
EPA Guide [35].
The procedure for determining countywide fuel consumption by in-
dustrial area sources is comprised of four basic steps:
• Developing statewide fuel intensity ratios for each fuel type
by each of twenty 2-digit SIC categories
• Estimating state consumption, reported by, SIC category, to in-
dividual counties on the basis of the fuel intensity ratios
and county area source employment in each SIC category
• Deriving actual state industrial area source fuel consumption
3-22
-------
• Normalizing estimated county industrial area source fuel use
against actual state consumption
1. Developing State Fuel Intensity Ratios
The fuel intensity ratio (FIR) is a measure of the intensity of
fuel use per employee. FIRs are stratified according to twenty 2-digit SIC
categories in order to reflect the large variation in fuel use intensity
among various types of industry. The industries used correspond to SIC
categories 20-39. One set of FIRs is computed for bituminous coal, and one
set is computed for a combination of natural gas and the gas equivalent of
LPG.
In order to account for geographic variation in fuel use inten-
sity, FIRs are calculated separately for each state. A set of national -
level FIRs is also calculated. The national FIRs are substituted in cases
for which the requisite state-level fuel and employment data are unavailable.
Table 3-4 illustrates the dependence of the value of FIRs on
industry type, fuel type, and geographic region.
Fuel consumption data, which are used to estimate FIRs, are
available primarily from two sources, the Annual Survey of Manufactures [36]
and the Census of Manufactures [14], The Annual Survey reports energy con-
sumption as measured by dollars spent by SIC group for the country as a
whole. Table 3-5, extracted from the Census of Manufactures, shows quantity
of fuel purchased by SIC group and by type of fuel for each state. Although
these Census data are available only at 5-year intervals, the stratification
of the fuel data is directly applicable to the proposed methodology. Conse-
quently, the census is the preferred source for fuel use data for FIR deter-
mination. State-level industrial employment by SIC class is obtained from
the County Business Patterns [13]. National employment is obtained by sum-
ming over the states.
The FIRs are calculated by taking the ratio of fuel consumption
by a particular industry to the employment in that industry:
3-23
-------
TABLE 3-4
FUEL INTENSITY RATIOS
SIC Natural Gas
Category California
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39 (includes
19)
0.503
0.061*
0.188
0.004
0.116
0.029
0.677
0.031
1.07
8.16
0.167
0.029*
1.81
0.677
0.132
0.041
0.040
0.066
0.031
0.032
(Ft.3 x 106/Employee)
Pennsylvania National
0.130
0.061*
0.054
0.004
0.132
0.040
0.431
0.022
0.272
13.8
0.114
0.035
0.934
1.04
0.131
0.075
0.063
0.076
0.037
0.097
0.309
0.061
0.111
0.011
0.132
0.044
0.751
0.036
1.67
9.67
0.148
0.029
1.24
0.925
0.123
0.086
0.065
0.085
0.043
0.063
Coal
California
2.89*
2.64*
1.71*
o.m*
0.092*
0.528*
0.192
0.017*
21.5*
2.62*
2.40*
0.415*
0.016
0.066
0.509*
0.008
0.413*
1.57*
1.88*
0.210*
(Tons/Employee) '
Pennsylvania National
2.34
2.64*
0.698
0.050
0.342*
0.966
0.963
0.012
25.2
15.8
1.74
1.13
39.1
9.67
0.388
0.399
0.994
2.06
7.91
0.299
2.89
2.64
1.71
0.111
0.342
0.528
14.9
0.017
21.5
2.62
2.40
0.415
17.1
7.94
0.509
0.697
0.413
1.57
1.88
0.210
* National FIR used because shipments for state not reported separately.
-------
TABLE 3-5
VABLE 4. Quantity and Cost of Purchased Fuels Used for Heat and Power
by State and Industry Group: 19711
"
-
25
13
34
35
20
21
32
23
25
2v
2?
3D
It
3:
34
33
31
39
2;
33
33
3 =
3"
3:
3:
20
?3
23
25
2--.
25
50
3',
3<
34
OS
37
J;
35
Stttt and industry jroup
C-i.-UJll =rJ allied products
..ood ».-tJ t. ^ p
|TJbJC" '11 ** •
i~ext ^.v . 11 -oduc's
** i * t
i. ,r. ., ^.3j . , g P
El-ctncal c-iulps^ftt tnd s'-opi. i«?s
Instr-^.-'its in3 relatrd products
l
i
j prCJcs
. Prt.-iry -.1 • a 1 Ir.rf-js'.riea.
1 Ir*'vr.iv.-.ts arui related products
j
, F^tr ar.1 211 i. 1 f-rod'jr is
! Cr.-.--:itals ...-.- P. 1 1 led proijcts
•Fabricate n^tui products
'Siectr^Vii e-viU^'1^;;;:;
!
1 ;!JM-!.->, t,.t»i
>"••: rnK._.n n-^ c ,-^1 p-0Hucts . .
, '-^jr a-ul ^Ij^t.ci products, n.e.c
''^'^r>tCj"' ''"":J" "••"'' J
1 " jC"- -?r.-.-i'i m-, . f actij rl^R Indus. ries
lignite ar»d a^^racitt
Quant-Y
(I.CCO
short toni)
Col. 1
4,709.1
3,310.6
22.7 1
215.9 ,
1.0S5.S I
1.3
.2
1.7S7.7
4.3
132.6
«7.4
19.6
1.7
65.8
698.4
217.0
2.7
(Z)
1.7
.5
1.6C9.4
3H3.5
87.7
.1
_
76.6
1,046.8
595.6
.3
114. S
2.0
1.3
273.2
104.4
62.9
.1
37.5
37.5
Con
Col.J
41.4
29.7
.2
2.7
7.5
(Z)
24.8
.1
2.2
8.4
.3
(Z)
1.1
9.2
3.3
(Z)
(Z)
(Z)
21. 1
5.0
(Z)
_
1.1
14.0
-
6.5
(Z)
1.8
(Z)
• CZ)
j 3.0
.9
1 .8
(Z)
.3
.3
CoVcandbtMii
Quantity
M.CQO
short tons) |
Ccl. X
(S)
(S)
(S)
(S)
(S)
(S)
(S)
(S)
(S)
(S)
(S)
(S)
(S)
(S)
(S)
(S)
(S)
(S)
(S)
(S)
Cost
dolUn)
Col.L
U.3
.5
(Z)
2.3
(Z)
(Z)
.9
.9
.3
.1
.1
(Z)
.1
.f
(Z)
(Z)
(Z)
.5
(Z)
.1
. i
,1
KMurel gas
Qua«!ity
C
-------
FIR1j - F1J/EJ
where FIR.. = fuel intensity ratio for fuel type i and SIC
category j
F. . = consumption of fuel type i by SIC category [14]
E. = employment in SIC category j [14]
J
2. Estimating County Consumption
The basic algorithm for determining countywide industrial area
source fuel use is represented by the sum of the products of each state FIR
and the corresponding county area source employment:
39
Fi c = I (Ei c ' Pi c} FIRii
1 »c i=20 ** ^
where F. = estimated county industrial area source consumption
lsC of fuel type i
E. _ = county employment for SIC category j (CBP)
J »c
P. _ = county point source employment for SIC category j
J'c (NEDS)
Consistent with the discussion in the previous section, the
state FIR will be replaced by the national analog if state-level data are
not disclosed in the census.
3. Derivation of Actual State Industrial Area Source Fuel
Consumption
a. Natural Gas
Statewide consumption of natural gas (including LPG) by
industrial area sources is derived by deducting point source consumption
of natural gas and LPG from published data on total industrial sales of
natural gas and LPG, as follows:
X1 = (G - Gp) + f(L - Lp)
where X-, = state total gas equivalent consumption by industrial
area sources
3-26
-------
G = total state industrial natural gas sales
G = statewide natural gas consumption by industrial
p point sources
L = total state industrial LPG sales
L = statewide LPG consumption by industrial point
p sources
f = factor to convert LPG to natural gas equivalent
The source of data on sales of natural gas by state and
user category is the Bureau of Mines MIS Natural Gas Production and Con-
sumption [6]. LPG sales data are compiled by the Bureau of Mines and re-
ported annually in the MIS Sales of Liquified Petroleum Gas and Ethane [8].
The Bureau of Mines data are compiled from surveys of producers, pipelines,
and distribution. Detailed analysis of the merits of these data sources
for estimating state natural gas consumption is given in Appendix C.
b. Bituminous Coal
Statewide consumption of bituminous coal by industrial
users is not currently reported. Instead, this is estimated from data on
total shipments to the states and the national distribution by user cate-
gory, using consumption arid shipments reported by the United States Bureau
of Mines [10]. The derivation of state-by-state consumption of bituminous
coal is discussed in detail in Appendix C and is summarized by the equation
below: BM
X2 ' SET
-------
4. Normalization
Estimates of county industrial area source consumption of
natural gas and bituminous coal are normalized against actual state in
dustrial area source consumption as follows:
F'i,c " Fi,c
where F1. = normalized county industrial area source consump-
' tion of fuel type i (i = 1 for natural gas, i = 2
for bituminous coal)
F. = estimated industrial area source consumption of
' fuel type i in county c
F1 =?Fi,c
X. = actual state Industrial area source consumption of
fuel type i
D. OFF-HIGHWAY CONSUMPTION OF GASOLINE AND DIESEL FUEL
The methodologies developed by Southwest Research Institute [15]
for estimating emissions from off-highway sources were examined for their
applicability to estimating off-highway fuel consumption on a county basis.
The off-highway category is comprised of six components, viz., farm equip-
ment, construction equipment, industrial equipment, motorcycles, lawn and
garden equipment, and snowmobiles. Insofar as the present effort is
limited to adapating the Southwest Research Institute (SWRI) methodologies
for direction inclusion in this NEDS area source upgrade, the revisions of
the off-highway methodology are primarily dictated by input data require-i
ments.
In general, the SWRI methodologies for each of the six off-highway
subcategories involve either apportionment of national fuel consumption
total to the counties on the basis of various demographic or economic items
or by direct calculation of county or state totals by applying fuel consump-
tion rates to average usage figures and equipment populations. Total off-
ghiway consumption of each fuel type is the total consumption of the fuel by
the six subcategories. Diesel fuel consumption is assumed to be zero for
3-28
-------
motorcycles, lawn and garden equipment, and snowthrowers.
The national gasoline and diesel fuel consumption totals for con-
struction equipment, industrial equipment, lawn and garden (other than snow-
throwers) equipment, snowthrowers, and snowmobiles are given in Table 3-6.
These consumption estimates were derived from data compiled by SWRI £15] and
NADB [18].
The final off-highway methodology is presented in detail in Section
II.A.4 of this report. The modifications to the original SWRI methodologies
that were found necessary are discussed below.
1. Farm Equipment
The original SWRI methodology estimated consumption for eight
farm equipment categories. Three of these—garden tractors, general-purpose
small utility engines, and lawn and garden small engines—have been grouped
into the lawn and garden equipment off-highway subcategory.
State consumption by the other five equipment categories is
calculated from equipment populations using formula (i) in Section II.A.4.a.
Because equipment population data for large utility engines are not pub-
lished, equation (ii) in the same section, also developed by SWRI, is used
to estimate the state equipment population for the category. Population
for the other four equipment categories is available from the Census of
Agriculture [16].
Total state consumption by the five farm equipment categories
is apportioned to the county level on the basis of tractor population, as
given by equation (iii) in Section II.A.4.a. Farm equipment data used for
the 1973 update are listed in Table 3-7.
2. Construction Equipment
The county consumption of gasoline and diesel fuel by construc-
tion equipment is determined by computing the national total fuel consumption
3-29
-------
TABLE 3-6
NATIONAL FUEL ESTIMATES FOR 1973, GALLONS
Gasoline Diesel
Construction Equipment 423 x TO6 7,833 x 106
Industrial Equipment 944,162 x 103 1,064,705 x 103
Lawn and Garden 583,467 x 103
Snowthrowers 17,504 x 103 —
Snowmobiles 82,593 x 103
Source: National Air Data Branch, U.S. Environmental Protection Agency [18]
3-30
-------
TABLE 3-7
USAGE RATES, CONSUMPTION RATES, AND POPULATION DISTRIBUTION
FOR HEAVY-DUTY AGRICULTURAL ENGINES
USED FOR 1973 UPDATE
i
Annual Fuel Consumption Rate Population Density
Usage (gallons/hour) (percentage)
(Hours/Year) Gasoline Diesel Gasoline Diesel
Combines
Bailers
Harvesters
General Purpose
Tractors
* 490 hours/year
** About 5% is LPG
71
24
120
50
*
diesel ,
, which
2.34
2.34
2.34
3.51
2.28
291 hours/year gasol
is not included here
1.5 57
1.5 100
1.5 0
1.94 50
2.98 65
ine
43
0
100
50
30**
Source: National Air Data Branch, U.S. Environmental Protection Agency [18]
3-31
-------
by construction equipment and apportioning this to the state level accord-
ing to total non-building employment.
The state's share of the national total is allocated to in-
dividual counties on the basis of population. The formula for this method
is given in Section II.A.4.b.
Estimates of national consumption were provided by the Na-
tional Air Data Branch [13] using emission and unit consumption data com-
piled in the SWRI study [15].
3. Industrial Equipment
The SWRI methodology for apportioning national industrial equip-
ment fuel consumption is based on value added, or the combined sales of min-
ing, wholesale trade, and manufacturing industries. Due to the unavailabil-
ity of these data on a county basis, the Wai den method is to replace this
apportioning factor with combined employment for the same categories. This
change will affect the county allocation according to differences in pro-
ductivity of labor between manufacturing, mining, and wholesale trade.
Labor productivity in these industries is shown in Table 3-8. The effect
of this change is to weight the manufacturing category more heavily than
mining and wholesale trade and to weight certain industries within the manu-
facturing category more heavily than others (e.g., textiles, lumber, and
leather products are weighted more heavily than tobacco, petroleum, and
chemicals).
The formula for this apportionment is described in Section II.
A.4.c. Total employment in manufacturing, mining, and wholesale trade is
the sum of the figures reported in the Bureau of Census, County Business
Patterns [13] for division D, division B, and major group 50, respectively.
4. Motorcycles
The SWRI approach for estimating county-level gasoline consump-
tion by motorcycles is based on population distribution, as follows:
3-32
-------
TABLE 3>8
1967 NATIONAL LABOR PRODUCTIVITY
MINING
WHOLESALE TRADE
MANUFACTURING
SIC 19
SIC 20
SIC 21
SIC 22
SIC 23
SIC 24
SIC 25
SIC 26
SIC 27
SIC 28
SIC 29
SIC 30
SIC 31
SIC 32
SIC 33
SIC 34
SIC 35
SIC 36
SIC 37
SIC 38
SIC 39
Value Added or
Sales & Receipts
(x 106)
25848.7
459475.9
261983.8
5584.8
26620.9
2032.0
8153.2
10064.4
4973.4
4169.5
9756.3
14355.1
23550.1
5425.8
6799.5
2626.5
8333.4
19978.2
18042.6
27836.4
24487.3
28173.9
6418.4
4599.4
Employment
(x 103)
567.3
3518.9
19323.2
400.4
1649.6
75.1
929.0
1356.7
554.0
425.3
638.9
1031.0
841.4
141.6
516.7
328.7
589.9
1281.0
1341.8
1864.5
1874.9
1834.1
394.3
423.1
Value Added or
Sales per Employee
(x 103)
45.5
130.6
13.6
19.9
16.1
27.1
8.8
7.4
8.9
9.8
15.3
13.9
27.9
38.3
13.2
7.9
14.1
15.6
13.4
14.9
13.1
15.4
16.3
10.9
Data wereobtained from the following sources:
1967 Census of Manufactures, Volume III, Area Statistics
1967 Census of Manufactures, Volume IV, Wholesale Trade,
Area Statistics
1967 Census of Mineral Industries, Series MIC67(2).
3-33
Illlakkn,
-------
Fm =
M * FR * U
where Fm = county consumption of gasoline by motorcycles (gallons)
PC = county population [19]
PS = state population [19]
M = state motorcycle registrations [20]
U = average annual usage per unit (miles/year) [15]
FR = fuel consumption rate (miles/gallon) weighted by
engine size and corresponding distance traveled [15]
A refinement of this method, separating off-road and combination
motorcycles and weighting the distribution of these two types according to
regional variations, is used for the allocation. The formula used is de-
scribed in Section II.A.4.d.
State motorcycle registration data are available from the
Federal Highway Administration's Highway Statistics [20]. The national
usage rate and usage factors for the two types of motorcycles are extracted
from the Henrix, Tucker, and Walker study [21]. The national fuel consump-
tion rate is estimated to be 0.0235 gallons/mile [15]. The consumption rate
for off-road and combination motorcycles is assumed to be the same.
5. Lawn and Garden Equipment
The original SWRI methodology for allocation of natural lawn
and garden fuel consumption to individual counties is based'on a combina-
tion of the number of single-unit structures, the number of freeze-free
days (i.e., the number of days with a minimum temperature > 32°F), the
fraction of national snow zone population that is in the county (the snow
zone is all areas with an annual snowfall > 30 inches), snowthrower fuel
consumption rate, average snow removal rate, and county snowfall [15]. The
final equation for the lawn and garden methodology is given in Section II.
A.4.e.
Data on population and the number of dwelling units in single-
unit structures are reported in the Bureau of Census Census of Housing [4].
3-34
-------
County snowfall and the number of freeze-free days are compiled from the
Environmental Data Service Climatological Data [3]. Climatological data
from this source are obtained for a selected, representative weather sta-
tion in the county.
6. Snowmobiles
National gasoline consumption by snowmobiles is apportioned to
a county on the basis of the county's share of the snowmobile population.
While state snowmobile populations are generally available from registra-
tion data [15], county breakdown is not. To estimate county-level snow-
mobile population, a set of regression formulations developed by SWRI [15]
is used to relate the percent of state snowmobiles used in the county to
population and snowfall. A distinction is made to reflect the impact of
population density on snowmobile usage. Equations (i) and (ii) in Section
II.A.4.f are the formulae used to estimate the fraction of state snow-
mobile population that is in each county. Equations (iii) and (iv) in Sec-
tion II.A.4.f are then used to compute county snowmobile population and to
apportion the national snowmobile gasoline consumption to the counties.
Centroid counties, used to determine snowfall at the center of the state,
are listed for each state in Table 3-9.
E. GASOLINE CONSUMPTION BY MARINE VESSELS
The SWRI methodology for apportioning state gasoline for marine
uses is based on inland water area, using the formula
Gv = Nv
s
* 10 * MC * FRv
where Gy = county consumption of gasoline by marine vessels
Ny = state boat registrations
WG = county inland water area
ws = ? Wc
M. = number of warm months (which promote boating activi-
c ties)
FRV = average fuel consumption rate (gallons/hour)
3-35
-------
TABLE 3-9
CENTROID COUNTIES FOR EACH SAROAD STATE
SAROAD State Name Centroid County
State SAROAD Number
Number
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
Alabama
Alaska
Arizona
Arkansas
California
Colorado
Connecticut
Delaware
District of
Columbia
Florida
Georgia
Hawaii
Idaho
Illinois
Indiana
Iowa
Kansas
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
Montana
Nebraska
Nevada
New Hampshire
New Jersey
New Mexico
New York
North Carolina
North Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
Puerto Rico
660
620
940
2220
2820
1740
565
60
20
580
5160
140
540
4400
2640
3480
3100
340
2260
595
80
369
1000
2360
1520
3040
460
640
300
440
2980
1200
5260
1840
1060
3440
2180
420
1520
- — •
Centroid County Name
Chi! ton
Yukon-Koyukuk
Yavapai
Pulaski
Fresno
Park
Middlesex
Kent
District of Columbia
Citrus
Twiggs
Honolulu
Custer
Logan
Marion
Story
Rice
Boyd
Coupee
Knox
Anne Arundel
Central Massachusetts
Clare
Morrison
Leake
Miller
Fergus
Custer
Lander
Merrimack
Mercer
Torrance
Otsego
Harnett
Sheridan
Licking
Oklahoma
Crook
Centre
3-36
-------
TABLE 3-9
CENTROID COUNTIES FOR EACH SAROAD STATE
SAROAD State Name Centroid County Centroid County Name
State SAROAD Number
Number
41 Rhode Island 140 Kent
42 South Carolina 460 Calhoun
43 South Dakota 1560 Stanley
44 Tennessee 2960 Rutherford
45 Texas 1000 Coleman
46 Utah 980 Sanpete
47 Vermont 500 Washington
48 Virginia 540 Buckingham
49 Washington 1940 Skagit
50 West Virginia 160 Braxton
51 Wisconsin 4060 Wood
52 Wyoming 460 Natrona
3-37
-------
This methodology was modified by Wai den to allow for accounting
separately for inboard and outboard vessels and to include county coast-
line in the apportioning algorithm. The formula for this final method is
given in Section II.A.5.
Inland water area is reported in the Bureau of Census Area Measure-
ments Reports [22]. State registrations for inboards are as reported in the
Marine Market [23], and outboard data are available from Boating 1972 [24].
Fuel consumption rates for inboards and outboards are assumed to be 3 and
1.5 gallons/hour, respectively [15]. Normal temperatures (based on 1931-
1960 data) are generally unavailable for many of the stations that are used
to represent county climatology. Consequently, the proposed monthly average
temperatures are obtained for representative counties for each station from
NOAA [3]. Limited availability of data precludes any reliable estimates of
the factor to convert coastline to an area equivalent; for the purposes of
the 1973 run, it was assumed to equal 1.
F. RAILROAD CONSUMPTION OF DIESEL FUEL
A number of alternative apportionment schemes were investigated, as
described in Appendix D. As a result of this analysis, the selected method
is to apportion state consumption of diesel fuel by railroads to the county
level on the basis of population distribution. The formula for this meth-
odology is described in Section 11.A.6.
Data on the use of diesel fuel by railroads for each state are ob-
tained from the Bureau of Mines Mineral Industry Survey [25].
G. RETAIL SALES OF GASOLINE
Retail sales of gasoline include all gasoline sold for highway use
and for use by construction equipment, industrial equipment, farm equip-
ment, and aviation off-highway categories. For states which compile and
can make available gasoline sales data on a county basis, the countywide
retail gasles data are used directly. State tax departments are the source
of such data. Only Arizona, Georgia, Louisiana, Minnesota, and New Mexico,
3-38
-------
representing approximately 15% of all counties, currently provide county
breakdowns of retail gasoline sales. For those counties where retail
sales of gasoline are not compiled, sales to the five retail user cate-
gories in the county are estimated separately and summed to give total
county sales.
Consumption by construction equipment, industrial equipment, and
farm equipment is computed using the off-highway methodologies for these
categories described in Sections II.A.4.a, b, and c.
Published state aviation gasoline sales are apportioned according
to total landing and take-off (LTO) cycles of aircraft in the county. LTO
cycles are apportioned to counties according to the methodology described
in Section II.A.10.
State retail sales of gasoline for highway use are derived from
total state gasoline sales by subtracting reported state totals of the four
off-highway category consumption categories. These data are reported in
the Federal Highway Administration (FHWA) publication Highway Statistics
[20]. Total retail sales of gasoline for highway consumption are calcu-
lated by subtracting the highway use of special fuels (Table MF-25*) from
the total highway motor fuel use (Table MF-21*). Use of gasoline by the
four off-highway user subcategories is reported separately in Table MF-24.*
This state component will include sales of gasoline for marine use, since
marine service stations are generally grouped with highway service stations
Similarly, other miscellaneous uses of gasoline are included in this high-
way component. The state highway gasoline sales are apportioned to the
county according to gross receipts of service stations, available from the
Department of Commerce Census of Business [26].
The formula for allocating countywide total retail sales of gaso-
line is given in Section II.A.7.
* Table MF-xx refers to table numbers in Reference 20.
3-39
-------
H. ORGANIC SOLVENTS
1• Identification of Major Solvent Groups and Data Sources
The identification and selection of major organic solvent
groups for the purpose of estimating national consumption and subsequent
countywide allocation are arrived at from the results of a recent systems
study on hydrocarbon pollutants [37]. Table 3-10 shows a compilation of
primary industrial solvents that were considered. Of this list of candi-
dates, the first sixteen solvent types will be considered individually,
while all remaining solvent types will be grouped together as "All Other
Solvents." The category "Special Naphthas," which comprises about two-
thirds of total organic solvent use in the country, includes the aliphatic
naphthas such as V.M. and P. naphthas, stoddard solvents, rubber solvents,
and mineral spirits [38].
Total U.S. production of solvents by type is extracted from a
publication of the U.S. Tariff Commission entitled, Synthetic Organic
Chemicals, U.S. Production and Sales [28]. National production of special
naphthas is taken from the Mineral Yearbook [27]. Data pertaining to usage
patterns for the sixteen most widely used organic solvents are obtained
from two principal sources, viz.,
• The Chemical Marketing Reports publish a weekly "Chemical
Profile" [39] for selected chemicals. These data can be
used to estimate the percentage of the solvent that is
used by various industry groups.
• SRI's Chemical Economic Handbook [38], which publishes
similar profiles on each chemical, but in greater detail.
For perchloroethylene, for example, the data that are
available include (i) producing companies; (ii) histo-
rical production figures; (iii) consumption by markets,
including a description of how and in which processes the
chemical is used; (iv) historical cost figures; (v) U.S.
imports; (vi) foreign producing companies; and (vii) list
of references.
Determination of national usage of each major organic solvent group from
these data is discussed in detail in Appendix E and is summarized in Table
3-11.
3-40
-------
TABLE 3-10
INDUSTRIAL SOLVENTS
Solvent
Special Naphthas
Perchloroethylene
Ethanol
Tri chloroethylene
Toluene
Acetone
Xylene
Fluorocarbons
Methyl Ethyl Ketone
1,1,1-Trichloroethane
Methylene Chloride
Methanol
Ethylene Dichloride
Ethyl Acetate
Cyclohexane
Methyl Isobutyl Ketone
Hexanes
Benzene
n-Butanol
Nitrobenzene
Turpentine
Isopropyl Acetate
Ethyl Ether
Monochlorobenzene
Isopropanol
Diethylene Glycol
Methyl Acetate
Cresols
Phenol
Chloroethane (ethyl chloride)
Carbon Tetrachloride
Pinene
Cyclohexanol
Cyclohexanone
Ethyl Benzene
Isobutyl Alcohol
Chloromethane
n-Butylacetate
Methyl Chloride
Source: Hydrocarbon Pollutant Systems Study, Volume 1, Stationary Sources
Effect and Control, MSA Research Corporation, Evans City, Pennsyl-
vania, October 1972.
3-41
-------
TABLE 3-11
1971 NATIONAL USE OF ORGANIC SOLVENTS
Total Demand
1971
(IbsxlO6)
Special Naphthas
VM&P Solvent
Stoddard Solvent
Rubber Solvent
Mineral Spirits
Penchloroethylene
Ethanol
Trichloroethylene
Toluene
Acetone
Xylene
Fluorocarbons
M.E.K.
1 ,1 ,1-Trichloroethane
Methyl ene Chloride
Methanol
Ethyl ene Bichloride
Ethyl Acetate
Cyclohexane
M.I.B.K.
All Other Solvents
8,711
748
2,023
539
3,422
1,660
3,617
825
491
375
380
749
7,558
159
1,747
190
11,456
Surface
Coatings
26%
4%
5%
9%
6%
65%
31%
0.3%
70%
2%
65%
Degreasing Dry Printing & Rubber & Other
Cleaning Publishing Plastics Miscellaneous
Solvent Use
6% 8% 60%
15% 58% 10%
35%
87% 3%
5%
16%
7%
10% 55%
7%
67% 9% 11%
11% 11% >37%
9%
2% 2.7%
9% 8% 1 0%
2%
25%
5.16%
Solvent Use
As % of Total
Consumption
100%
83%
39%
90%
10%
25%
13%
65%
72%
87%
>90%
9%
5%
97%
4%
90%
5.16%
-------
2. County Apportionment
a. Distributive Factors for Major User Categories
National consumption of organic solvents, distributed by
major user categories, is apportioned to the individual counties on the
*•
basis of applicable SIC employment categories. For example, in degreas-
ing processes use category, total solvent use is allocated to each county
in proportion to the county industrial employment for SIC groups 34-39. It
was estimated that 95% of all degreasing operations occur in these indus-
tries [40]. For dry cleaning applications, the countywide allocation will
be made on the basis of total employment in SIC groups 7215, 7216, and 7218.
In computing the total employment, the employment figure for 7216 will be
inflated by a factor of two, since this SIC group represents establishments
engaged in dry cleaning only, while 7215 and 7218 are for both dry cleaning
and wet laundering [41]. In the category "Other Miscellaneous Solvent Use,"
the distributive factor is made up on one-half by county population and one-
half by total industrial employment. The distributive factors for county
apportionment are summarized in Table 3-12.
b. Secondary Distributions for Surface Coating and
Applications
For the surface coating industry, the total solvent use is
further subdivided according to the stratification of coating application
shown in Figure 3-3. The reported production data by end use (see Figure
3-3) are multiplied by solvent content factors to obtain solvent produc-
tion estimates for surface coating uses. The Boston AQCR hydrocarbon sur-
vey [40] indicates that average solvent content for water-based trade coat-
ings is 3.5%; for solvent-based trade coatings, it is 53%; and for indus-
trial coatings, approximately 67%. When the solvent production estimates
for surface coating uses are applied against the national consumption esti-
mates for all types-of solvent, the secondary distribution percentages and
consumption of the surface coating category are obtained. The results are
exhibited in Table 3-13.
3-43
-------
TABLE 3-12
DISTRIBUTIVE FACTORS FOR ORGANIC SOLVENTS
BY USER CATEGORIES
User Categories
SIC Industry
Distributive Factor
Surface Coatings
Trade Paints-Auto Refinishing
Auto Refinishing (Trade)
Automoti ve
Wood Furniture & Fixtures
Metal Furniture & Fixtures
Metal Containers
Sheet Strip & Coil
Appliances
Machinery & Equipment
Paper
Factory-Finished Wood
Transportation (Non-Auto)
Electric Insulation
Other, Exterior, Interior
Marine
Degreasing
Dry Cleaning
Printing
Rubber and Plastics
Other Miscellaneous Use
7535 (Paint Shops)
371 (Motor Vehicles)
25 (Furniture & Fixtures)
County Population
34 (Fabricated Metal Products)
35 & 36 (Machinery, Electrical
Equipment & Supplies)
26 (Paper & Allied Products)
243, 244 (Millwork, Plywood-
Related Supplies, Wooden
Containers)
37 (Transportation Equipment)
Less 371 (Motor Vehicles) &
373 (Shipbuilding Repair)
36 (Electrical Equipment &
Supplies)
19-39 (Total Manufacturing)
373 (Shipbuilding & Repair)
34-39 (Metal Products, Machinery,
Transportation Equipment, Instru-
ments, Miscellaneous)
2 x 7216, Plus 7215 & 7218 (Dry
Cleaning & Combination with Wet
Laundering)
264, 265, & 27 (Paper Products,
Containers, Printing & Publish-
ing)
30 (Rubber & Plastics)
1/2 by 19-39 Employment
1/2 by Population
3-44
-------
FIGURE 3-3
SURFACE COATINGS
Estimated Production Value and Production in 1970
'Millions of Dollars/Millions of Gallons)
Trade Sales
1550/430
TOTAL
SURFACE COATINGS
2800/830
1250/400
3-45
Interior
755/220
Exterior
605/160
Miscellaneous
170/45 •
Exports
20/5
Product Finishes
1005/330
r.'einter.once Finishes
225/65
Exports
20/5
-------
FIGURE 3-3 (continued)
Solvent-Base
295/75
Water-Base
460/145
Water-Base
295/85
Gloss and Semigloss Enamel
Flat Wall Paint
Varnish
Primer and Sealer
Other
Flat Wall Paint
Semigloss Enamel
Other
House Paint and Other
Automotive Refinishing
125/31
40/10
40/10
30/10
60/15
385/12"
55/15
20/5
Solvent-Base v
310/75
•••,*•/
House Paint
Enamel
Primer and Sealer
Other
125/30
75/20
35/10
75/15
295/85!
110/20!
Traffic Paint
Other
30/15
30/10
Automotive
Wood Furniture and Fixtures
Metal Containers
Mata! Furniture and Fixtures
Appliances
Machinery and Equipment
Paper, Film, and Foil
Sheet. Strip, and Coil
Factory Finished Wood
Transportation (Non-Automotive)
tlectrical Insulation
Other
Exterior
Interior
Marine
130/40
110/501
110/40!
85/25
80/20
75/25
75/251
60/15 i
45/15 i
45/15 j
40/10;
150/50 ;
65/20
35/10
3-46
-------
TABLE 3-13
DISTRIBUTION FACTORS FOR SURFACE COATING SOLVENT USE 1971
Coating Usage Type
SIC Groups
of Surface Coatings
Solvent Use
to
I
Trade Paints-Auto Refinishing
Auto Refinishing (Trade)
Automotive
Wood Furniture & Fixtures
Metal Furniture & Fixtures
Metal Containers
Sheet Strip & Coil
Appliances
Machinery & Equipment
Paper
Factory-Finished Wood
Transportation (Non-Automobile)
Electric Insulation
Other, Exterior, Interior
Marine
Population Distribution
7535 Paint Shops
371 Motor Vehicles
25 Furniture & Fixtures
34 Fabricated Metal Products
35 & 36 Machinery, Electrical
Equipment & Supplies
26 Paper & Allied Products
243, 244 Millwork, Plywood-Related
Supplies, Wooden Containers
37 Transportation Equipment Less
371 Motor Vehicles & 373 Ship-
building Repair
36 Electrical Equipment & Supplies
Total Manufacturing Equipment
373 Shipbuilding & Repair
r\C QOI
C.V.0/0
2.8%
7.1%
13.4%
9.8%
8.0%
4.4%
2.7%
2.7%
1.8%
18.7%
1.8%
TOTAL
100%
-------
The formula used in allocating organic solvent use to
counties is given in Section II.A.8.
I. SULFUR AND ASH CONTENT OF COAL
In developing a methodology for determining sulfur and ash content
of coal, a limited survey was conducted of state and local air pollution
control agencies which may compile data on local point sources as part of
their compliance enforcement activities. It was determined that, due to
incompletion, state and local regulatory agencies do not represent a use-
ful source of data on sulfur and ash in coal. The final methodology that
was developed is divided into the two types of coal discussed below.
1. Bituminous and Lignite
The procedure for estimating sulfur and ash content of bitu-
minous coal at the county level consists of three steps:
• Determine sulfur and ash content of coal associated with
each production district or production district grouping
(see Section II.A.9.a.(1)).
• Compute sulfur and ash content of coal shipped to each
state for industrial users and retail users.
• Compute sulfur and ash content of coal used in each county
according to industrial and retail bituminous coal con-
sumption.
a. Coal Production District Sulfur and Ash Content
As explained in Section II.A.9.a.(1), weighted averages of
sulfur and ash content of coal from each production district must be taken
of industrial coal and coal for other uses. These data correspond to the
"Other Industrial Users and Retail Dealers" and "All Other Users" cate-
gories, respectively, as reported by the Bureau of Mines and published an-
nually [75]. Table 3-14 presents an illustration of these data for the year
1972. This approach excludes accounting for coal shipments to electric
utilities and coke and gas plants. Also, as previously explained, sulfur
and ash data for several production districts must be combined, due to the
3-48
-------
TABLE 3-14
Shlpnents of bituminous coal and lignite by average sulfur content by consumer use In 1972
Quantity (hipped (thousand thort tons)
Average iulfut content (percent)
District
2. Wcst.«rn .Pennsylvania — — —
3. Northern West Virginia
4. Ai I'.'iiums-Okluhoma
Total United States
Electric Coke
Utilities and gun
plantu
20, 35 A 3,290
5,237 14,914
19,104 2.1S6
33.625
8.874
439 6,939
28.765 16,950
43,883
- 44,158 - 2,782
16,335
702
7,982 4,107
376
6,922 58
556
1,871 1,720
10.157
9 , 1 35 34
229 2,152
2,686
668
265,755 55,478
Other
Inciiifl trial
UHC8 And
ret Jill
dealer*
2,506
3,126
2,100
5,633
271
427
5,810
2.144
7,190
4,373
831
1*1
100
16
239 x
10
479
994
>,
29
36,987
All
other
uses
886
2.670
859
1,242
179
274
10,286
1,020
1.459
11
18
193
212
2
23
1
s 28
'^20
26
7
19,416
Export!
overseas and
Canada)
1,628
724
3.216
10.175
7,095
247
210
192
11)
""^ ..
23,602
Total !_/
28,664
26,671
27,435
40,508
9,324
18.254
68,906
47.047
55,589
20.719
720
13.360
777
7,292
574
4,045
10.168
9,676
3.510
4,610
2,700
697
401,246
Electric
Utilities
2.2
2.1
2.7
3.5
3.9
.8
1.1
4.6
3.4
3.4
3.4
1.7
4.9
.5
.6
.6
.6
.7
.9
.9
1.5
2.9
Coke
and gas
plants
0.9
l.S
1.2
.7
.8
.8
M
1.3
.5
.6
.7
.6
1.0
Other
Industrial
uses and
retail
dealers
1.9
1.7
2.4
3.0
3.0
.8
.9
3.7
2.8
3.4
2.0
1.6
4.1
.3
.5
.5
.7
.6
1.0
1.1
.5
2.3
All
other
uses
2.1
1.7
2.0
2.9
2.8
1.0
.9
3.7
2.9
3.9
4.0
1.1
W--
3.8
.3
.7
.5
.6
.7
1.0
.4
1.6
Export*
(overseas, and
Canada)
1.7
2.3
1.8
.7
.8
«M»«
1.3
1.5
.5
.8
1.0
Total
2.0
1.7
2.5
3.4
3.
•
4(
3.
3.
3,
1.
1.
4.
•
•
•
•
•
•
1.
2.3
CO
I
I/ Total shipments by producers reporting sulfur content (67 percent of total U.S. production).
SOURCE U. -S. BUREAU OF MINES
-------
practice of grouping districts when reporting state shipments data [10].
Expressions for counting the weighted average sulfur content are given in
Section II. A. 9. a. (1 } .
Average ash content of coal for each production district,
prior to combining into production district groupings, must be derived
from data reported on individual mines, due to the absence of production
figures for ash. Consequently,
where a,. = ash content of coal from k mine sampled in district
k J [47]
n = number of mines sampled in district j [47]
A- = average ash content of coal from district j
J
b. Average Statewide Sulfur and Ash Content
At the state level, two sets each of sulfur and ash aver-
ages are computed, using coal distribution data reported by the Bureau of
Mines [10]. The first set is for industrial coal consumed in the state,
and the second is for coal shipped to retail dealers in the state. To ob-
tain sulfur and ash contents that are more specific to area source users,
the point source components of the sulfur and ash contents for both cate-
gories are subtracted from corresponding total content. This accounting is
summarized by the expression in Section 1 1. A. 9. a. (2) .
c. Countywide Sulfur and Ash Content
The average sulfur and ash contents of bituminous coal con-
sumed in each county by retail and industrial area sources are weighted
averages of the retail and industrial area sources' consumption in the
county according to the equation given in Section II. A. 9. a. (3) .
3-50
-------
2. Anthracite
Because there is only one anthracite producing region in the
country (located in Southeastern Pennsylvania), the proposed methodology
computes only a single sulfur and ash content for the anthracite from this
region. The averages are computed 'from Bureau of Mines data [47] accord-
ing to the formula:
— 1 n
O ~~ ™~ / ^ i..
n i/;-, k
where k = 1 n is the mine's index, with a total of n
mines in the sample
S^ = sulfur content of coal from the k mine
"S" = average sulfur content of anthracite produced
No attempt is made to weigh this average number according to
amount of coal produced from each mine because this production information
is not available.
The average ash content is computed in an analogous manner.
This methodology implies a uniform sulfur and ash content for all anthra-
cite used by area sources; however, this is not objectionable, due to the
uniformity in the coal. The sulfur content of anthracite is consistently
below 1%. The ash content exhibits greater variation among the coal seams,
but this variability is minimized by the practice of mixing the coal from
various beds prior to shipment [46].
J. LANDING AND TAKE-OFF CYCLES OF AIRCRAFT
Landing and take-off (LTO) cycles for military, civil, and commer-
cial airports are determined for each category from total aircraft opera-
tions reported for each of these categories. An operation is defined by
the FAA as either a take-off or a landing. LTO cycles, therefore, are one-
half the reported operations. Aircraft operations data for each of the
three categories are obtained from the following Federal Aviation Adminis-
tration (FAA) publications:
3-51
-------
. FAA Air Traffic Activity [42]. This publication gives the number
of operations performed by commercial, civil, and military air-
craft at airports with FAA-regulated control towers. These air-
ports will include all the major non-military airfields in the
United States. Operation totals are given both for itinerant
flights, those that terminate at an airport other than the one
at which they originated, and for local flights, those that ori-
ginate and terminate at the same airport. Total operations for
each aircraft category are, therefore, the sum of both itinerant
and local operations.
. Military Air Traffic Activity Report [43]- This publication con-
tains the number of operations performed by military and civil
aircraft at military airfields.
. Census of U.S. Civil Aircraft [44]. This report gives the number
of active civil aircraft for each county in the country. These
data will be used to estimate aircraft activity for those counties
without an FAA-regulated or a reporting utility airport.
For counties with FAA-regulated and/or military airports, the number
of operations for commercial, civil, and military aircraft is set equal to
the sum of the activities reported for each of these categories at airports
located in the county. These activity data are extracted directly from the
FAA Air Traffic Activity report [42] and the Military Air Traffic Activity
report [43]. Airport locations are determined from the Aviation Directory
[45], atlases, and road maps. General aviation activity from non-regulated
airports in counties with regulated airports is assumed to be negligible.
For counties without FAA-regulated or military airports, the number
of LTOs on an annual basis is assumed to be 365 times the number of reported
active aircraft in the county. The number of active aircraft is extracted
from the Census of U.S. Civil Aircraft [42].
3-52
-------
IV. COMPUTER PROCESSING
A. OVERVIEW
In order to facilitate annual updating of the selected NEDS area
source data items (indicated by the darkened fields on the coding form in
Figure 4-1) for all SAROAD counties in the nation, the methodologies described
in the previous sections were programmed for the UNIVAC 1110 system, in
FORTRAN IV code. A system of programs was also developed for preliminary
processing of all data collected for input to the area source fuel alloca-
tion program (ASFAP). A master file comprised of all data required for
input to the ASFAP program is generated by this preprocessing system. A
schematic overview of the entire data processing system is illustrated in
Figure 4-2.
In order to produce an updated file of NEDS area source data items,
the master file that is input to the ASFAP program must be updated first.
All data required for updating the master file is either available on mag-
netic tape or in publications from which it must be coded and keypunched.
A set of preprocessing programs exists for each type of incoming data.
Each set of programs is designed to operate independent of the others, due
to the varying frequency of availability and update requirements of data
from different sources. The final stage of each set of preprocessing pro-
grams is an update file ready to update the corresponding data items on
the current version of the master file. The program that performs the
update can do so for individual update files or for any combination of
update files.
When all data that could be updated have been entered into the
master file, the ASFAP program is run using the updated master file for
input. The ASFAP program performs the allocation calculations and outputs
the results on printed tables and a card-image file formatted according to
NEDS specifications.
4-1
-------
FIGURE 4-1
NATIONAL EMISSIONS DATA SYSTEM (NEDS)
ENVIRONMENTAL PROTECTION AGENCY
OFFICE OF AIR PROGRAMS
AREA SOURCE
Input Form
Date
FORM APPROVED
OMB NO. 151 n009S
NMW of Person
Comptetini Form
SIP EMISSION ESTIMATES (10* lon/yi)
SULFUR CONTENT II)
ASH CONTENT (U)
RESIDENTIAL FUEL
49 50 51 5? 53545956 57 5« 59 6061 626364
COMMERCIAL AMD INSTITUTIONAL FUEL
INDUSTRIAL FUEL
Resid. Oil
10< Gals.
Wood Process G»
102tons
Rend. Oil
104 Gals.
29 30hlb?h' 34
40l41 4?|43!'|.||4r,|46|47 48
OM SITE INCINERATION
Comm. Insl.
ID2 Ions
Residential
102 Ions
Coram'l Instill
1C2 Ions
Litnl Vehicle
IOJ Gals.
Heavy Vehicle
I03 Gals.
Oil Hiway
I03 Gals.
Heavy Vehicle
103 GaU.
54 5556 57l53lfi9 SO 61 626
66 G7 08 F9 70J71 72 73 74 75 7r 77
MEASURED VEHICLE MILES
Commercial
LTD CYC !01
Limited Access Road
Urban Roads
10< Miles
Diesel Oil
10* Gils.
Suburban Roads
104 Miles
^'j7Tjil[is|lf.!l7|lB|inj?(l(7ll27|73|24|25|26
^MMHHfc^^BI^HMI^BMBH^HBaBiMM>MMHMMM^Mr~~*4~~~
4?]4lUll4sUr,l47l.JB!lo|5llt;2 53
II. il Hl.jdl
luvflfd
ID1 tfrlmlf miltl
10
11
12
13
14
15
16
Dill
AirSlnps
LTD CYCLES
17
18
19
20
21
Coniliuclion
Acrci
22
23
24
25
26
27
Mncrlljneout
Wind liunon
103 Ac.ri
28
29
30
31
32
llnd Tilling
10.3 Aem
33
34
35
36
37
FOHCSI waon
38
39
40
41
42
43
44
RCS
Tont/icift
45
46
47
MANAGED BUnN
(Sliih/Priuiibcd Bui
48
49
5051
52
53
NG
ning)
Toni/Acitl
54
55i5G
I
AGRICULTUnAL
BURNING
Aem Burntd
57
58
59
60
61
62
FIELD
Tant/Acres
Buintd
63
64
65
FROST COn
10?
Oicnud
Hrjltil
66 67
68
63
rnoi
Oiyt/Fiird
'TO"
71
72
FIRES
FilFt/V«JI
73!
74l75
76
77
i
7H
7!)
A
Cd
110
5
COWfll NT*
10
11
\/
n
14
IFi
1fi
17
Ifl
1!)
20
21
2?
21
24
25
76
27
2BT29
1
30
31
"2
33
34
35
36
37
38
39
40
41
42
43
44
45
46
J7
48j49
1
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
C9
70
71
72
73
74
75
76
77
S
0
*t
73
79
A
Cd
no
6
-------
FIGURE 4-2.
Area Source Fuel Use Allocation Data Processing System
i
CO
— CBPData J Card Type 7 ^Manual K^°l', «'L' «> * "ata ^_N
VP6V J Processing .|taraiype^ Completion P V— £ ^b%,Spm , ^~
^ ' CNAME Compl etn. Card Typ 7 '" SmXi?*
| CBP | Forms- — -J | tuiuib
1 1 1 ^ 11
i i . .... L . . \ * Y
" Tn ™rmiQf0 1 , County Card State Card ("State Card ( State Card f County Card
CarTlype 12 Types 13 & 14j Types 8 & 9 , | Type 7 Type 10 | Type 15
Complete ^— — _____^^ , __. -^
Forms— J ^- — ^r\^^^"
Manual |
Completion, Manual f State Card
Payrl Tvnp 1? *^_ Codl HQ •> f i r
Ldru iyue it -^^^ ouuiny *. jvoe ]_g
1 ^- ^
^County Card ^^^ J County Card ^
Type 12 \.^ *• Type 11 ^-
l _______ "-\^ I ^^
- -_ _ ^^Sy^ ^^^ " "
f^) ~V"
V J x' \
— Point oQUrCC / \ rnoDci
ivjinu owui vv. , , LUKKtL
X X Data Processes: *~ \ i
/DCCMD \ DCDPMT DOTD T 1 P9RF D1aQ"
f rbtMr i rbKInl, rKLr j \ lour i , m«j
I J \ / ....* nostics |
^ -1 *" -" i 1 V^p. MERGE [^
. - -. ' -—*-- Hi an_
/ . _ . ,^ ^ U lay-
1 Area source /ACIIAftl \ r Mr.HEr.K — * nnstirs — -
[Cards 1,2,43 mm f AS FOAL ^ _J
1 1 I ' 1^ ,,••'
Allocation*1 [_ ..^ ;
r - - f- ~ t Hi ^n_
. Suimiary L_ SYNTAX — ^ uia?. --
uiag- nostics
nostic; [_ J
^
('Part'
Card
ho &
ial data.
Types
15.
T
Manual
Coding
( County Card •«
Type 16
p i :
SEAS
«• FIXC
Seasonal
Degree
Days
^J
1
t
f Corrected
[ Card Types
-J
-------
B. THE AREA SOURCE FUEL ALLOCATION (ASFA) MASTER FILE
The data in the ASFA master file can be divided into three cate-
gories:
. National and regional level data
. State level data
. County level data
The data are stored in card image records according to the formats described
in Table 4-1. Each record consists of an S^digit key field in columns 1-8
and a data'field in columns 9-80. The key field contains a 2-digit state
SAROAD number, a 4-digit county SAROAD number, and a 2-digit card type number.
For national and regional level records, the state and county numbers in the
key field are coded as zero. For state level data, the county number in the
key field is coded as zero. A list of data items on each record and their
formats is contained in Table 4-1.
The format of the ASFA master file records will be referred to as
"ASFA master file format."
The records are stored in the master file in ascending order accord-
ing to the 8-digit key field. The structure of the ASFA master file resulting
from this organization is illustrated below.
National and regional card types 1-99
State level card types 1-45 for SAROAD state 01
County level card types 1-20 for first SAROAD county in state 01
County level card types 1-20 for second SAROAD county in state 01
Cotinty level card types 1-20 for last SAROAD county in state 01
State level card types 1-45 for SAROAD state 02
4-4
-------
TABLE 4-la
NATIONAL INPUT VARIABLES
tn
Card Description of Variables
Number
1 Industry sulfur content of coal, Production Districts
1-23
2 Other sulfur content of coal, Production Districts
1-23
3 Industry coal production, Production Districts 1-23
4 Other coal production, Production Districts 1-23
5 Ash content of bituminous coal, Production
6 Ash content of bituminous coal, Production
Sulfur content of anthracite coal
Ash content of anthracite coal
7 Annual usage of diesel tractors
Annual usage of gasoline tractors
Districts
1-18
Districts
19-23
General purpose - agricultural
Harvesters
Balers
Combi nes
Average gasoline consumption rate, tractors
General purpose
Harvesters
Balers
Combines
* The first 8 digits of each card contain the record
Format?
8X,23F3.1/
8X.23F3.1/
8X.23F3.0/
8X,23F3.0/
8X.18F4.2/
8X,5F4.2/
F4.2
F4.2
8X,F3.0
F3.0
F3.0
F3.0
F3.0
F3.0
F3.2
F3.2
F3.2
F3.2
F3.2
identification key.
Card
Col umns
9-78
9-78
9-78
9-78
i9"80
9-29
30-33
34-37
9-11
12-14
15-17
18-20
21-23
24-27
28-30
31-33
34-37
38-40
41-43
Units
Percent
Percent
Tons x 105
Tons x 105
Percent
Percent
Percent
Percent
Hours/Year
Hours/Year
Hours/Year
Hours/Year
Hours/Year
Hours/Year
Gallons/Year
Gallons/Year
Gallons/Year
Gallons/Year
Gal Ions/ Year
-------
cr>
TABLE 4-1 a (continued)
NATIONAL INPUT VARIABLES
Card Description of Variables
Number
Average diesel fuel consumption rate, tractors
General purpose
Harvesters
Balers
Combines
Percent tractors using gasoline fuel
Percent tractors using diesel fuel
Percent using gasoline, general purpose - agricultural
Harvesters
Balers
Combi nes
8 Fuel consumption, construction, gasoline
Fuel consumption, construction, diesel
Fuel consumption, industrial, gasoline
Fuel consumption, industrial, diesel
Fuel consumption, lawn and garden
Fuel consumption, snowthrowers
Fuel consumption, snowmobiles
Usage, motorcycles, off-road
Usage (off-road), motorcycles, combination
Gas mileage, motorcycles
Gas mileage, inboard boats
Gas mileage, outboard boats
Format *
F3.2
F3.2
F3.2
F3.2
F3.2
F2.0
F2.0
F2.0
F2.0
F2.0
F2.0
8X.F7.0
F7.0
F7.0
F7.0
F7.0
F7.0
F7.0
F3.0
F3.0
F3.1
F2.1
F2.1
Card
Columns
44-47
48-50
51-53
54-57
58-60
'61-62
63-64
65-66
67-68
69-70
71-72
9-15
16-22
23-29,
30-36
37-43
44-50
51-57
58-60
61-63
64-66
67-68
69-70
Units
Gallons/Year
Gallons/Year
Gallons/Year
Gallons/Year
Gallons/Year
Percent
Percent
Percent
Percent
Percent
Percent
Gallons x 103
Gallons x 103
Gallons x 103
Gallons x 103
Gallons x 103
Gallons x 103
Gallons x 103
Miles/Year
Miles/ Year
Miles/Gallon
Gallons/Hour
Gallons/Hour
The first 8 digits of each card contain the record identification ke"y.
-------
TABLE 4-1 a (continued)
NATIONAL INPUT VARIABLES
Card
Number
9
10
11
12
13
14
15
16
17
18
19
20
21
Description of Variables
Gas consumption by census region, cooking range
Gas consumption by census region, water heater
Regional percentage, off-road motorcycles
Regional percentage, combination motorcycles
Growth by census region in LPG heat
in LPG cooking
Growth by census region in coal heat
Solvent consumption
Solvent tonsumption
Bituminous coal consumption, steel mills
Cement plants
Other industrial
Retail
1971 Industrial coal consumption, SIC 20-29
1971 Industrial coal consumption, SIC 30-38, 19 & 39
1971 Industrial natural gas consumption, SIC 20-29
1971 Industrial natural gas consumption, SIC 30-38,
19 & 39
1971 Industrial employment, SIC 20-29
1971 Industrial employment, SIC 30-38, 19 & 30
Census year
Year before update year (or zero if no growth data)
Update year
University employment/enrollment ratio
Format*
8X.9F3.0
9F3.0/
8X,9F3.0
9F3.0/
8X,9F4.1
9F4.1
8XJ4F4.0/
8X.14F5.0/
8X,3FS.O
F5.0
F5.0
F5.0
F5.0/
8XJOF7.0/
8X.10F7.0/
8XJOF7.0/
8XJOF7.0/
8X.10F7.0/
8X.10F7.0/
8X, F4.0
F4.0
F4.0
F6.3
Card
Columns
9-35
36-52
9-44
45-80
9-44
4S-80
9-44
9-78
9-23
24-28
29-33
34-38
39-43
9-78
9-78
9-78
9-78
9-78
9-78
9-12
13-16
17-20
21-26
Units
Therms/Year
Therms/Year
Percent
Percent
Percent
Percent
Percent
Pounds x 106
Pounds x 106
Tons x 103
Tons x 10
Tons x 10
Tons x 103
Tons x TO2
2
Tons x 10
Ft.3 x 108
Ft.3 x 108
* The first 8 digits of each card contain the record identification key.
-------
00
TABLE 4-la (continued)
NATIONAL INPUT VARIABLES
Card
Number
22
23
Solvent
Solvent
Description of Variables
point source employment,
point source employment,
Format*
SIC
SIC
25
39
,26,27,
, Total
30,34-38
19-39,
8X
8X
,9F8.
,9F8.
o/
o/
Card Units
Columns
9-80
9-80
Laundries,243,244,371,373,7535
24 Solvent point source employment, SIC 264,265 8X,2F8.0/ 9-24
REGIONAL INPUT
4
25 Number of regional groupings 8X.I2/ 9-10
26 Fuel index number for first regional grouping 8X,I2 11-12
Number of states in grouping 12 13-14
Retail or residential fuel F6.0 15-20
Commercial fuel F6.0 21-26
Industrial fuel F6.0 27-32
States in grouping (SAROAD numbers) n 12 (n = number of states in grouping)
27-99 Like 26, as many as needed, rest blank except for 8-digit key field in columns 1-8
* The first 8 digits of each card contain the record identification key.
-------
TABLE 4-lb
STATE INPUT VARIABLES
-p.
I
Card
Number
1
2
3
4
5
Description of Variables
State SAROAD number
Number of counties
State name
Census region identifier
Centroid county SAROAD number
Coastline
Coastline area factor
Current employment, SIC 19-27
Current employment, SIC 28-36
Current employment, SIC 37-39
Current employment, Total, 19-39
Current employment, SIC 701 (hotels)
Current employment, SIC 7211+7216+7217 (laundries)
Current employment, SIC 806 (hospitals)
Current employment, SIC 821 (schools)
Current employment, SIC 822 (universities)
Format
12
6X,I3
A14
12
14
F10.0
F10.3/
8X.9F8.0/
8X,9F8.0/
8X,3F8.0
F8.0/
8X.F8.0
F8.0
F8.0
F8.0
F8.0
Current employment, SIC 60+70 minus above(other services)F8.0
Current employment, SIC 50 (wholesale)
Current employment, SIC 52 (retail)
Current employment, SIC 7215+2-7216+7218(laundries
F.80
F8.0
for F3.0/
Card Units
Columns
1-2
8-10
11-24
25-26
27-30
31-40
41-50
9-80
9-80
9-32,
33-40
9-16
17-24
25-32
33-40
41-48
49-56
57-64
65-72
73-80
solvents)
-------
TABLE 4-1b (continued)
STATE INPUT VARIABLES
Card
Number
6
7
8
9
10
11
12
13
14
15
Description of Variables
Current employment, SIC 243 (millwork, plywood, etc.)
Current employment, SIC 244 (wooden containers)
Current employment, SIC 371
Current employment, SIC 373
Current employment, SIC 7535
Current employment, SIC 10—
Current employment, SIC 16
Current employment, SIC 264
Current employment, SIC 265
Employment data, SIC 19-27
Employment data, SIC 28-36
Employment data, SIC 37-39
Employment data, SIC Total, 19-39
Coal consumption data, SIC 19-27
Coal consumption data, SIC 28-36
Coal consumption data, SIC 37-39
Coal consumption data, SIC Total, 19-39
Gas consumption data, SIC 19-27
Gas consumption data, SIC 28-36
Gas consumption data, SIC 37-39
• . ,
Format
8X.F8.0
F8.0
F8;0
F8.0
F8.0
F8.0
F8.0
F8.0
F8.0/
8X,9F8.0/
8X,9F8.0/
8X.3F8.0
F8.0/
8X,9F8.1/
8X,9F8.1/
8X.3F8.1
F8.1/
8X,9F8.1/
8X,9F8.1/
8X,3F8.1/
Card
Columns
9-16
17-24
25-32
33r40
41-48
49-56
57-64
65-72
73-80
9-80
9-80
9-32
33-40
9-80
9-80
9-32
33-40
9-80
9-80
9-32
Units
Tons x 103
Tons x 103
Tons x 10
Tons x 10
Ft.3 x 109
Ft.3 x 109
Ft.3 x 109
-------
TABLE 4-1b (continued;
STATE INPUT VARIABLES
Card
Number
16
17
18
19
20
21
22
23
24
Description of Variables
Coal shipments: retail total, retail production
district groupings 1-13
Retail total, retail production
district groupings 14-20
Industrial total, industrial pro-
duction district groupings 1-6
Coal shipments: industrial production district
groupings 7-20
Public school employment
Hotel employee/ room ratio
Current population
Percent of gas customers with gas heat
Additions to gas heating, each year since census year
Conversions to gas heating, each year since census
year
Gas-heated dwelling units (previous year)
Natural gas consumption, residential
Industrial '
Commercial
Other
LPG consumption, industrial
Retail
Anthracite coal shipments, retail
Bituminous coal shipments, industrial
Retail
Anthracite market share
Format
8XJ4F5.0/
8X,7F5.0
7F5.0/
8X,14F5.0/
8X.F8.0
F8.2
8X,F10.0
8X,F6.1
11F6.1/
8X.11F6.1
F6.0/
8X.F8.0
F8.0
F8.0
F8.0
F8.0
F8.0/
8X,F8.0
F8.0
F8.0
F8.2/
Card
Col umns
9-78
9-43
44-78
9-78
9-16
17-24
9-18
9-14
15-80
9-74
75-80
9-16
17-24
25-32
33-40
41-48
49-56
9-16
17-24
25-32
33-40
Units
Tons x 103
Tons x 10
Tons x 10
Tons x 103
Percent
103 Additions
3
10 Conversions
Ft.3 x 106
Ft.3 x 106
Ft.3 x 106
Ft.3 x 106
Gallons x 103
Gallons x 103
Tons
Tons x 10
Tons x 103
Percent
-------
TABLE 4-1b (continued)
STATE INPUT VARIABLES
ro
Card Description of Variables
Number
25 Gasoline consumption, highway
Off -highway
Construction equipment
Commercial -industrial
Agricultural
Aviation
Railroad use of dlesel fuel
26 Registrations, motorcycles
Snowmobiles
Inboard boats
Outboard boats
27 Farms in irrigated areas
Tractors
Combines
Harvesters (corn huskers)
Pickup balers
28 Census year population
Gas-heated dwelling units
Coal -heated dwelling units
Elementary and kindergarten enrollment
High school enrollment
Format
8X.F8.0
F8.0
F8.0
F8.0
F8.0
F8.0
F8.0/
8X,F8.0
F8.0
F8.0
F8.0/
8X,F8.0
F8.0
F8.0
F8.0
F8.0
8X,F9.0
F8.0
F8.0
F8.0
F8.0/
Card
Col umns
9-16
17-24
25-32
33-40
41-48
49-56
57-63
9-16
17-24
25-32,
33-40
9-16
17-24
25-32
33-40
41-48
9-17
18-25
26-33
34-41
42-49
Units
3
Gallons x 1.0
Gallons x 103
Gallons x 10
Gallons x 103
Gallons x 10
Gallons x 103
Bbl x 103
-------
TABLE 4-lb (continued)
STATE INPUT VARIABLES
Card
Number
POINT
29
30
31
32
33
34
35
36-45
Description of Variables
SOURCE DATA
Point source employment, SIC 19-27
Point source employment, SIC 28-36
Point source employment, SIC 37-39
Total, SIC 19-39
Point source employment, SIC 701, (7211+7216+7217),
806, 821 ,822, other sources,
50, 52, (721 5+2x721 6+721 8)
Point source employment, SIC 243,244,371,373,7535,
10,16,264,265
Bituminous coal, commercial consumption
Sulfur content
Ash content
Bituminous coal, industrial consumption
e Sulfur content
Ash content
Natural gas consumption, commercial < .
Industrial (including LPG)
LPG consumption, commercial
Anthracite coal consumption, commercial
Blank
Format
8X.9F8.0/
8X,9F8.0/
8X,3F8.0
F8.0/
8X.9F8.0/
8X.9F8.0/
8X.F8.0
F8.0
F8.0
F8.0
F8.0
F8.0
F8.0
F8.0
F8.0/
8X,F8.0/
Card
Col umns
9-80
9-80
9-32
33-40
9-80
9-80
9-16,
17-24
25-32
33-40
41-48
49-56
57-64
65-72
73-80
9-16
Units
Tons
Tons
Tons
Tons
Tons
Tons
Ft.3 x 106
Ft.3 x 106
Gallons x 10
Tons
-------
TABLE 4-lc
COUNTY INPUT VARIABLES
I
-pi
Card Description
Number
of Variables
1 County SAROAD number
AQCR
County name
Degree days
Number of days with temperature less than 32°F
Number of "warm" months
Snowfal 1
2 Current employment,
3 Current employment,
4 Current employment,
5 Current employment,
SIC 19-27
SIC 28-36
SIC 37-39
Total, 19-39
SIC 701 (hotels)
Format
2X.I4
2X,I3
A26
F6.0
F3.0
F2.0
F9.0/
8X.9F8.0/
8X.9F8.0/
8X,3F8.0
F8.0/
8X.F8.0
SIC 7211+7216+7217 (commercial F8.0
laundrfes)
SIC 806 (hospitals)
SIC 821 (schools,)
SIC 822 (universities)
F8.0
F8.0
F8.0
SIC 60+70 minus above (other services) F8.0
SIC 50 (wholesale)
SIC 52. (retail)
SIC 7215+2x7216+7218 (laundries for
F8.0
F8.0
F8.0/
Card Units
Columns
3-6
9-11
12-37
38-43
44-46
47-48
49-57 10"2 Inches
9-80
9-80
9-32,
33-40
9-16
17-24
25-32
33-40
41-48
49-56
57-64
65-72
73-80
solvents
-------
TABLE 4-lc (continued)
COUNTY INPUT VARIABLES
01
Card Description of Variables Format
Number
6 Current employment, SIC 243 (millwork, plywood, etc.) 8X
SIC 244 (wooden containers)
SIC 371 (motor vehicles & equipment)
SIC 373 (ship & boat building & repai
SIC 7535 (paint stores)
SIC 10 (mining)
SIC 16 (heavy construction)
SIC 264 (miscellaneous connected
paper products)
SIC 365 (paper board containers &
boxes)
7 Hospital beds 8X
Hospital employment
Public university, enrollment
8 Population density 8X
Kindergarten and elementary enrollment
High school enrollment
Year-round housing units
Median rooms per dwelling units (+10)
% rooms in 1-unit simultaneous (+10)
Farms
Farms with sales >. $2500
9 Census year population 8X
Number of occupied dwelling units
,F8.0
F8.0
F8.0
r)F8.0
F8.0
F8.0
F8.0
F8.0
F8.0/
,F8.0
F8.0
F8.0/
,F9.Q
F9.0
F9.0
F9.0
F9.1
F9.1
F9.0
F9.0/
,F8.0
F8.0
Card Units
Columns
9-16
17-24
25-32
33-40
41-48
49-56
57-64
65-72
73-80
9-16
17-24
25-32
9-17
18-26
27-35
36-44
45-53
54-62 Percent
63-71
72-80
9-16
17-24
-------
TABLE 4-lc (continued)
COUNTY INPUT VARIABLES
Card
Number
Description of Variables
Format
Card
Columns
Units
10 Number of occupied dwelling units with gas heat 8X,F8.0 9-16
Number of occupied dwelling units with LPG heat F8.0 17-24
Number of occupied dwelling units with oil heat F8.0 25-32
Number of occupied dwelling units with coal heat F8.0 33-40
Number of occupied dwelling units with natural gas ranges F8.0 41-48
Number of occupied dwelling units with LPG ranges F8.0 49-56
Number of occupied dwelling units with natural gas F8.0 57-64
hot water
Number of occupied dwelling units with LPG hot water F8.0/ 65-72
11 Current population 8X.F8.0 9-16
Tractors F8.0 17-24
o
£ Gross revenues of service stations or retail gasoline F8.0/ 25-32 $ or (Gallons x 10 )
o? consumption
12 Air carrier and ^axi operations SX.F8.0 9-16
General aviation operations F8.0 17-24
Military F8.0 25-32
Aircraft registrations F8.0/ 33-40
13 Inland water area 8X.F10.0 9-18
Coastline F10.0/ 19-28
14 Point source employment, SIC 19-27 8X.9F8.0 9-80
15 Point source employment, SIC 28-36 8X,9F8.0 9-80
16 Point source employment, SIC 37-39 8X,3F8.0 9-32
Point source employment, Total 19-39 F8.0 33-40
-------
TABLE 4-1c (continued)
COUNTY INPUT VARIABLES
Card Description of Variables Format Card Units
Number Columns
17 Point source employment, SIC 701,(7211+7216+7217), 8X.9F8.0/ 9-80
806,821,822,other services,50,52,
(7215+2x7216+7218)
18 Point source employment, SIC 243,244,371,373,7535, 8X9F8.0/ 9-80
10,16,264,265
19 Point source employment, for solvents, SIC 25-27,30, 8X,9F8.0/ 9-80
^ 34-38
20 Point source employment, for solvents, SIC 39 8X,F8.0/ 9-16
-------
State level card types 1-45 for SAROAD state 52
County level card types 1-20 for first SAROAD county in state 52
County level card types 1-20 for last SAROAD county in state 52
C. THE ASFA PREPROCESSING SYSTEM
The ASFA preprocessing system is conducted in two steps: update file
generation and ASFA-master file updating. There are two types of update files
that are generated by the ASFA preprocessing system. One type is a file of re-
cords in the ASFA master file format that will replace the corresponding records
on the existing ASFA master file. The second type is the format required by
the NOAA UPD8 program for updating the climatological data stored on ASFA master
file county card type 1.
1. Update File Generation
The update file generation is functionally grouped into seven
categories:
. National Oceanographic and Atmospheric Administration (NOAA) Data
Processing
. Census of Housing Data Processing
. County and City Data Book Data Processing
. County Business Patterns (CBP) Data Processing
. NEDS Point Source Data Processing
. Manual Coding and Keypunching
. Card Processing
The flow diagram in Figure 4-2 depicts the interrelationship of
these sets and their interaction with the ASFA master file. Each stage can* be
performed independently of the others. As the necessary input files for each
stage become available, processing of these files can be initiated. The result-
ant output from each stage is an update file that can be entered into the ASFA
master file by the appropriate update program.
4-18
-------
a. NOAA Data Processing
Three programs are used to: (1) extract a set of climato-
logical data for each weather station, (2) replace missing data items with
the corresponding data items from the nearest weather station in the same
climatological district, and (3) assign the data items to the appropriate
county. A fourth program is used to load the NOAA data onto the master file.
Three input tapes are processed by this set of preprocessing
programs. The data extracted from these tapes include the following:
. Station latitude and longitude by station number
. Degree days, snowfall, number of days with the tem-
perature less than 32°F, and the number of months
with the mean temperatures greater than a predeter-
mined temperature which is a function of the lati-
tude of the station
. SAROAD county number and the corresponding NOAA
station number
b. CBP Data Processing
Three preprocessing program are used to process the CBP data.
The CBP data are in a continuous file contained in four tapes. Employment
data are extracted for a number of SIC codes which correspond to the various
industrial and commercial categories used in the allocations. The county
code must be converted from the CBP code to the corresponding SAROAD codes,
and the data for the appropriate SIC codes must be extracted and summed by
county before transferring it into the master file.
c. Census of Housing Data Processing
Three preprocessing programs are used for processing the
Census of Housing data. The data extracted from the Census of Housing tape
include: (1) total number of occupied dwelling units, (2) number of dwelling
units distributed'by fuel use, and (3) number of dwelling units using natural
gas and LPG for cooking as well as for hot water. These data must be assigned
to the corresponding SAROAD county and state numbers and summed by county prior
to entry onto the master file.
4-19
-------
d. County and City Data Book Data Processing
Two programs are used to process the county and city data
book tape. The first program reformats the input tape into card images,
retaining only required data items. The second stage reads the card images,
converts the GSA/FIPS state and county codes to SAROAD equivalents, and out-
puts the following data items, viz.:
. Total number of farms
. Number of farms in class 1-5
. Median rooms per housing unit
. Enrollment in primary schools for age group 3-34
. Enrollment in secondary schools for age group 3-34
. Number of dwelling units which are single-unit structures
e. NEDS Point Source Data Processing
Two magnetic tapes contain the required point source data items
for updating the ASFA master file. The NEDSPS tape contains all data items on
the NEDS point source coding form for each point source. The PSEMP tape con-
tains point source employment by SIC category.
Two programs are necessary to process the two tapes and generate
the required ASFA master file point source data records. One program extracts
the fuel use and sulfur and ash content data from the NEDSPS tape and outputs
the data on a card image file. The second program reads the reformatted fuel
use tape and employment data tape and generates the ASFA master file point
source data records.
f. Manual Coding and Keypunching
The manual coding effort required in the ASFA preprocessing
system can be delineated into five categories:
. National and regional data card types 1-99
. State card types 7-27
. County card type 11
. Input to card processing program (see Section II.B.l.g)
. Revisions resulting from the review of diagnostic files
output by various preprocessors or the update program
4-20
-------
Input requirements for the card processing programs are
described in the documentation for these programs [l ].
Diagnostic files output by the NOAA, CBP, CENSUS, and county
and city data book data processing programs consist of lists of data which
could not be matched with a SAROAD county. These diagnostic messages re-
quire manual review and edit. The review should consist of determining the
SAROAD county to which the data correspond. The complete data for that
SAROAD county required for updating the particular ASFA master file record
must be coded and keypunched for input to the master file update program.
g. Card Processing
The data required for ASFA master file county card types 7
and 12 cannot be coded directly from the source documents. The data are
identified in various documents by county name, but are listed by individual
institution. The data must, therefore, be assigned to the appropriate SAROAD
county, and county totals must be computed. Five programs have been written
to process the data after they have been keypunched on cards. Three of the
programs are involved in reduction of university enrollment, hospital beds,
and hospital employment data by SAROAD county and generating ASFA master file
county card type 7. The other two programs process the aircraft registration
and FAA aircraft operations data by SAROAD county and generate ASFA master
file county card type 12.
2. Updating the ASFA Master File
ASFA master file updating consists of generating an updated version
of the master file by replacing the data on the existing master file with the
corresponding data items in the current update files. The update files created
by the programs described in Sections IV.B.l.b-g are all processed by the UPD8
program. These update files can be processed individually or combined with any
other of these update tiles. The NOAA update file (Section IV.b.l.a) must be
treated exclusively, using a different update program. Output from the update
programs consists of an updated ASFA master file, a list of invalid keys de-
tected, and a disk file containing the records with invalid keys. The invalid
key file must be reviewed and edited, and the corrected records must be entered
onto the master file in a subsequent update. When all required update files
4-21
-------
have been entered onto the master file and all invalid keys resolved, the
master file is ready for input to the ASFAP program.
D. THE AREA SOURCE FUEL ALLOCATION PROGRAM (ASFAP)
The general structure of the ASFAP program is illustrated in
Figure 4-3. This program applies the allocation methodologies to the data
on the ASFA master file and outputs a set of fuel use data by county for
upgrading the corresponding items in the NEDS area source data bank. ASFAP
is comprised of a main program and twelve subroutines. The main program is
responsible for-reading in control cards and master file data and for pass-
ing control to the various subroutines in the proper sequence. The individual
subroutines perform the allocation calculations and output the results. A
list of the subroutines and a brief description of the function of each is
given in Table 4-2.
The main program consists of two state loops, each with a nested
county loop. Prior to the first state loop, arrays are initialized, con-
trol cards are read, and the national and regional level data are read in
from the master file. During the pass through the first state and county
loops, the entire master file is read, but only desired states, as indicated
in the input control cards, are processed. However, certain items are re-
quired from data for each state and county for computing national totals.
For states that are to be processed, various levels of processing for each
methodology are performed by passing control to the appropriate subroutine.
The intermediate calculations for these states and counties are output to
temporary disk files.
In the pass through the second state and county loops, the inter-
mediate calculations are read in from temporary disk, the allocation methodo-
logies are completed, and the county fuel use estimates are normalized to
state totals. As the processing for each state is completed, the allocation
results are output to a print file in summary tables and to a card image
file according to NEDS format specifications. Of the six area source cards
associated with each county, only card types 1, 2, 3, and 4 are affected.
The results of the various allocation methodologies will appear in the
darkened fields shown in the card layouts in Figure 4-1.
4-22
-------
FIGURE 4-3. ASFAP Program Flow Chart
Program Con-
trol Options
Read
Program
Control
Data
1
Read Na-
tional
Level Data
State Loop A
Temporary
Disk for
County
Data
Temporary
Disk for
State Data
4-23
-------
TABLE 4-2
ASFAP SUBROUTINES
Subroutine Name
General Description of Subroutine Function
AREAS
COM
INDUST
LTD
OFFHWY
PRINT
PRMAST
RES
RETGAS
RRVESL
SOLVNT
SULASH
Compute area source employment and consumption by subtract-
ing NEDS point source data from corresponding totals
Perform methodology for allocating commercial use of
natural gas, anthracite coal, and bituminous coal
Perform methodology for allocating industrial use of
natural gas and bituminous coal
Allocate military, civil, and commercial LTO cycles to
counties
Perform methodology for allocating use of gasoline and
diesel fuels by off-highway sources
Print allocation results and punch output for NEDS ASDB
update
Print master file input
Perform methodology for allocating residential use of
natural gas, anthracite coal, and bituminous coal
Perform methodology for allocating retail gasoline sales
Perform methodologies for allocating railroad use of diesel
fuel and gasoline consumption by vessels
Perform methodology for allocating organic solvents
Perform methodology for allocating sulfur and ash content
of anthracite coal and bituminous coal
4-24
-------
Diagnostic messages associated with either input data or inter-
mediate computation are output on a separate print file for further inves-
tigation. A more detailed description of the program is given in the pro-
gram documentation report [1].
E. RESULTS FOR SELECTED TEST COUNTIES
Output from the main allocation program (ASFAP) required verifi-
cation to ensure that all methodologies were being performed properly.
Thirteen counties were selected for performing hand calculations which would
be checked against the computer program results. Several modifications to
the ASFAP program were made before all test requirements were fulfilled, and
all program test results agreed with the manual calculations. A summary of
the results from the final test run of the ASFAP program are contained in
this section.
1. Selection Criteria
A total of thirteen counties was selected for testing of the
computer programs developed to implement the allocation methodologies. Be-
cause the allocation methodologies are generally based on the use of socio-
economic, demographic, and climatological variables to apportion statewide
fuel consumption to the individual counties, extremities in these variables
provide a meaningful base to develop criteria for selecting test candidate
counties.
These criteria were translated into the following set of factors
which were considered in selecting the candidate test counties:
. Total county population
. Degree of urbanization, as measured by the population density
. Predominance of specific type(s) of land use, e.g., heavy
industrial activity or agricultural activity
. High level of employment in service industries, i.e., com-
mercial/institutional use category
. Average annual degree days
. Fuel preference or the predominance of a fuel type
. Extreme amount of inland water area and/or length of coastline
. Absence of either FAA regulated or military airport
4-25
-------
. Population growth rates which are high or negative
. Amount of snowfall
. Complicated geographical or political unit
In addition to data"considerations, a complete state had to be tested in
order to test normalization procedures. Because of the extremely large
amount of processing required for each county, Delaware, which contains
only three counties, was selected for testing normalization procedures.
A list of the selected test counties and a summary of the
demographic and-climatological data used in the testing (and the 1973 update)
are listed in Table 4-3.
2. Summary of County Results
Table 4-4 shows the residential consumption of anthracite coal,
bituminous coal, natural gas, and LPG for the thirteen counties, as computed
by the methodologies described in Section II.A.I. The total gas equivalent
given in the last column represents the sum of natural gas and the term
equivalent of natural gas from the LPG consumption.
Estimated consumption of these four fuels is also compiled for
commercial/institutional use. The results are shown in Table 4-5. These
consumptions refer only to area source use, since point source fuel use has
also been subtracted from statewide fuel use totals and corresponding point
source employment figures have been subtracted from reported county level
employment figures.
Industrial use of coal and gas is given in Table 4-6. These
consumption figures refer to area sources only, since point source contribu-
tions were excluded in a manner similar to the commercial/institutional cate-
gory. No anthracite use is assumed for industrial area sources. The natural
gas equivalent of LPG was added to state residential natural gas consumption
prior to the allocation.
Weighted sulfur and ash contents in coal for the county are given
in Table 4-7. The sulfur and ash contents are reported in percentages for both
anthracite and bituminous coal.
4-26
-------
_c»
ro
TABLE 4-3
SUMMARY OF DEMOGRAPHIC AND CLIMATOLOGICAL CHARACTERISTICS
FOR CANDIDATE COUNTIES FOR TESTING OF METHODOLOGY
County Name
Bristol Bay Division, Alaska
Los Angeles, California
Kent, Delaware
Newcastle, Delaware
Sussex, Delaware
District of Columbia
Honolulu, Hawaii
Ness, Kansas
Population
1,045
7,036,463
81,892
385,856
81,353
765,510
629,176
4,791
Population
Density
Per Sq. Mile
2
1,730
138
883
85
12,402
1,058
4
% Population
Change 1960-
1970
(NA)
16.6
35.9
35.7
8.8
-1.0
26.0
-12.4
% Labor Force
In Manufacturing
0
27.3
24.6
30.5
30.2
4.9
10.3
1.9
% Labor Force
In Services
3.0
9.1
6.2
8.6
6.3
12.6
9.5
5.2
% Dwelling Units
In One-Unit
Structures
67.8
60.6
69.8
74.8
83.2
36.8
58.8
93.6
% of Land
In Farms
(NA)
21.4
57.8
40.4
56.1
0
38.5
100.6
Metropolitan Boston, Massachusetts
Coos, New Hampshire
New York, New York
Schuylkill, Pennsylvania
Jefferson, Texas
National Average or Total
34,291
1,539,233
160,089
244,817
203,212,877
19
66,923
204
259
57
-7.7
-9.4
-7.5
0.3
13.3
43.6
18.1
45.5
28.2
25.9
6.4
13.4
4.5
9.3
7.7
61.2
1.1
80.8
85.8
69.1
7.6
0
20.0
57.5
47.0
1971-1972
Degree Days
12,888
1,193
3,945
4,166
4,340
3,927
0
4,935
5,532
9,312
4,684
6,913
1,150
-------
TABLE 4-4
FUEL USE AND ACTIVITY BY COUNTY 1973
RESIDENTIAL
County Name
Coal (Tons) LPG
Bituminous Anthracite Total (Kilgal.)
Gas (Ft.-* x 10°)
Natural Total Gas Equivalent
ro
oo
Bristol Bay Division, Alaska
Los Angeles, California
Kent, Delaware
Newcastle, Delaware
Sussex, Delaware
District of Columbia
Honolulu, Hawaii
Ness, Kansas
Metropolitan Boston,
Massachusetts
Coos, New Hampshire
New York, New York
Schuylkill, Pennsylvania
Jefferson, Texas
0.0 0.0 0.0 0.404E+01 0.0 0.373E+00
0.0 0.0 0.0 0.181E+05 0.180E+06 0.182E+06
0.550E+02 0.495E+03 0.550E*03 0.266E+04 0.476E+03 0.721E+03
0.636E+03 0.572E+04 0.636E+04 0.354E+04 0.690E+04 0.723E+04
0.211E+03 0.190E+04 0.211E+04 0.538E+04 0.138E+03 0.634E+03
0.439E+05 0,535E+04 0.512E+05 0.174E+05 0.195E+05 0.211E+05
0.0 0.0 0.0 0.188E+04 0.0 0.173E+04
0.0 0.0 0.0 0.135E+04 0.124E+03 0.249E+03
0.995E+04 0.517E+04 0.151E+05 0.123E+05 0.363E+05 0.374E+05
0.586E+01 0.537E+02 0.596E+02 0.509E+03 0.332E+01 0.577E+02
0.343E+05 0.114E+06 0.148E+06 0.843E+04 0.125E+05 0.133E+05
0.208E+05 0.188E+06 0.208E+06 0.252E+04 0.127E+03 0.359E+03
0.0 0.0 0.0 0.268E+04 0.528E+04 0.553E+04
-------
TABLE 4-5
FUEL USE AND ACTIVITY BY COUNTY 1973
COMMERCIAL/INDUSTRIAL
County Name
Ri ttiminou<;
Coal (Tons)
Anthracite
Bristol Bay Division, Alaska o.O
Los Angeles, California
Kent, Delaware
Newcastle, Delaware
Sussex, Delaware
District of Columbia
Honolulu, Hawaii
f" Ness, Kansas
no
^ Metropolitan Boston,
Massachusetts
Coos, New Hampshire
New York, New York
Schuylkill, Pennsylvania
Jefferson, Texas
0.0
0.162E+02
0.189E+03
0.499E+02
0.0
0.0
0.0
0.0
0.221E-05
0.0
0.252E+05
0.0
0.
0.
0.
0.
.0.
0.
0.
0.
0.0
0.0
195E+03
228E+04
603E+03
771E-02
0.0
0.0
384E-03
462E-04
793E-01
507E+05
0.0
Total
0.
0.
0.
0.
0.
0.
0.
0.
0.0
rf.O
212E+03
247E+04
653E+03
771E-02
0.0
0.0
384E-03
485E-03
793E-01
758E+05
0.0
(Ki
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
LPG
logal . }
0.0
566E-02
104E+04
247E+04
243E+04
0.0
862E+04
680E+03
593E-02
443E+03
534E-02
0.0
129E+04
Gas (Ft.3
Natural Total
0.
0.
•>0.
0.
0.
0.
0.
0.
0.
0.
0.
0.0
873E+05
194E+03
290E+04
556E+04
879E+04
0.0
773E+02
212E+05
529E+01
101E+05
453E+02
388E+04
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
x 106)
0.0
873E+05
290E+03
31 3E+04
280E+03
879E+04
794E+03
140E+03
212E+05
465E+02
101E+0.5
453E+02
400E+04
-------
CO
o
TABLE 4-6
FUEL USE AND ACTIVITY BY COUNTY 1973
INDUSTRIAL
County Name
Bristol Bay Division, Alaska
Los Angeles, California
Kent, Delaware
Newcastle, Delaware
Sussex, Delaware
District of Columbia
Honolulu, Hawaii
Ness, Kansas
Metropolitan Boston,
Massachusetts
Coos, New Hampshire
New York, New York
Schuylkill, Pennsylvania
Jefferson, Texas
Bituminous Coal (Tons)
0.0
0.0
0.199E+03
0.830E+04
0.633E+03
0.0
0.0
0.0
0.0
0.203E+02
0.105E+06
0.0
0.428E+01
Total Gas Equivalent (Ft.3 x 10b)
0.0
0.239E+06
0.142E+03
0.266E+04
0.260E+03
0.506E+04
0.206E+03
0.0
0.751E+04
0.119E+02
0.892E+04
0.173E+04
0.161E+06
-------
TABLE 4-7
SULFUR ANU ASH IN COAL BY COUNTY 1973
County Name
Anthracite (%)
flch
Bituminous (%)
fur Ach
i
CO
Bristol Bay Division, Alaska
Los Angeles, California
Kent, Delaware
Newcastle, Delaware
Sussex, Delaware
District of Columbia
Honolulu, Hawaii
Ness, Kansas
Metropolitan Boston,
Massachusetts
Coos, New Hampshire
New York, New York
Schuylkill, Pennsylvania
Jefferson, Texas
0.0 0.0
0.0 Cf.O
0.660E+00 0.112E+02
0.660E+00 0.112E+02
0.660E+00 0.112E+02
0.660E+00 0.112E+02
0.0 0.0
0.0 0.0
0.660E+00 0.112E+02
0.660E+00 0.112E+02
0.660E+00 0.112E+02
0.660E+00 0.112E+02
0.0 0.0
0.0 0.0
0.0 0.0
0.151E+01 0.999E+01
0.162E+01 0.104E+02
0.149E+01 0.991E+01
0.802E+00 0.736E+01
0.0 0.0
0.0 0.0
0.807E+00 0.735E+01
0.164E+01 0.891E+01
0.196E+01 0.101E+02
0.160E+01 0.105E+02
0.293E+01 0.109E+02
-------
Results of countywide transportation fuel use and activity
estimates are shown in Table 4-8. The data reported are off-highway con-
sumption of gasoline and diesel fuel; vessels' use of gasoline; railroad
use of diesel; and aircraft LTOs distributed according to commercial, civil,
and military categories.
Countywide retail gasoline sales and organic solvent consump-
tion are reported in Table 4-9 as evaporation losses. The solvent consump-
tion is disaggregated into special naphthas and all other solvents. Further
distribution of the special naphthas and the total solvents according to the
twelve sub-categories of surface coasting applications and the six primary
categories of solvent users is given in Table 4-10. Each column heading in
this table lists two solvent user categories. For each county, two rows of
values are listed for special naphthas and two rows are listed for total
solvents. In each case, the first row of values correspond to the user
categories in the first row of column headings; the second row of values
correspond to the user categories in the second row of column headings.
3. Summary of Sample State Results
The countywide results given in Table 4-11 are available for
all counties in the United States for 1973 in the form of a computer printout
and NEDS area source punched cards provided to EPA-NADB, Durham, North Caro-
lina. Unlike the above presentation, the county results in the national pro-
cessing are aggregated on a state-by-state basis. Additionally, a fuel use and
activity summary for each state is also presented. An example of the state
summary is given in Table 4-11, showing results for the state of Delaware.
4-32
-------
-P"
I
CO
CO
TABLE 4-8
FUEL USE AND ACTIVITY BY COUNTY 1973
TRANSPORTATION
County Name Gasoline (
Off-Highway
Bristol Bay Division, Alaska
Los Angeles, California
Kent, Delaware
Newcastle, Delaware
Sussex, Delaware
District of Columbia
Honolulu, Hawaii
Ness, Kansas
Metropolitan Boston,
Massachusetts
Coos, New Hampshire
New York, New York
Schuylkill, Pennsylvania
Jefferson, Texas
0.592E+03
0.225E+07
0.110E+05
0.456E+05
0.114E+05
0.432E+05
0.105E+06
0.260E+04
0.197E+06
0.566E+04
0.818E+05
0.258E+05
0.540E+05
Kilogal.)
Vessels
0
0
0
0
0
0
0
0
0
0
0
.136E+02
.350E+04
.824E+03
. 1 04E+04
.234E+04
.395E+04
0.0
0.0
.490E+04
.269E+03
.200E+03
.500E+01
.111E+04
Diesel (Kilogal.)
Off-Highway Railroads
0
0
0
0
0
0
0
0
0
0
0
0
0
.196E+02
.190E+06
.463E+04
.171E+05
.540E+04
.224E+05
.221E+05
.755E+03
.714E+05
.105E+04
.476E+05
.535E+04
.119E+05
0.143E+02
0.121E+06
0.197E+03
0.875E+07
0.188E+03
0.630E+04
0.0
0.199E+03
0.128E+05
0.542E+01
0.675E+04
0.195E+04
0.105E+05
Com.(lOS)
573.5
23,530.0
0.0
753.3
0.0
12,910.0
7,253.0
0.0
0.0
0.0
0.0
0.0
772.9
Aircraft LTOs
Civ.(lOS) Mil.(lOOS) Total (10S)
671.6
90,480.0
222.5
7,411.0
3,769.0
4,031.0
21,410.0
292.0
1,333.0
1,059.0
1,497.0
2,701.0
4,693.0
14.8
360.7
334.0
139.0
0.0
75.2
2,093.0
0.0
255.1
0.0
0.0
0.0
33.6
1
117
3
9
3
17
49
3
1
1
2
5
,393.0
,600.0
,562.0
,554.0
,796.0
,700.0
,590.0
292.0
,884.0
,059.0
,497.0
,701.0
,803.0
-------
TABLE 4-9
FUEL USE AND ACTIVITY BY COUNTY 1973
EVAPORATION LOSSES
County Name
Bristol Bay Division, Alaska
Los Angeles, California
Kent, Delaware
Newcastle, Delaware
Sussex, Delaware
District of Columbia
Honolulu, Hawaii
^ Ness, Kansas
<£ Metropolitan Boston,
Massachusetts
Coos, New Hampshire
New York, New York
Schuylkill, Pennsylvania
Jefferson, Texas
Retail Gasoline Sales
( Kijngal.)
0.293E+03
0.337E+07
0.637E+05
0.207E+06
0.397E+05
0.258E+06
0.222E+06
0.266E+04
0.121E+07
0.161E+05
0.172E+06
0.787E+05
0.146E+06
Sperjial Naphthas;
0.242E+02
0.293E+06
.0.209E+04
0.831E+04
0.181E+04
0.243E+05
0.105E+05
0.561E+02
0.815E+05
0.483E+03
0.927E+05
0.527E+04
0.659E+04
Solvents (Tons)
0.129E+02
0.209E+06
0.119E+05
0.512E+05
0.104E+05
0.113E+05
0.609E+04
0.307E+02
0.579E+05
0.252E+03
0.492E+05
0.363E+04
0.442E+04
Total
0.371E+02
0.501E+06
0.326E+04
0.134E+05
0.285E+04
0.356E+05
0.166E+05
0.868E+02
0.139E+06
0.734E+03
0.142E+06
0.881E+04
0.110E+05
-------
TABLE 4-10
SOLVENTS BY USER CATEGORY
Trade Paint
SIC 36
110 Bristol Bay Div., Alaska
Special Naphthas
Total Solvents
4200 Los Angeles, California
Special Naphthas
Total Solvents
60 Kent, Delaware
Special Naphthas
Total Solvents
180 Newcastle, Delaware
Special Naphthas
Total Solvents
240 Sussex, Delaware
Special Naphthas
Total Solvents
20 District of Columbia
Special Naphthas
Total Solvents
140 Honolulu, Hawaii
Special Naphthas
Total Solvents
0.1628E+01
0.0
0.2632E+01
0.0
0.1028E+05
0.1310E+04
0.1662E+05
0.2118E+04
0.1330E+03
0.0
0.2151E+03
0.0
0.5919E+03
0.1079E+02
0.9572E+03
0.1745E+02
0.1270E+03
0.0
0.2053E+03
0.0
0.1086E+04
0.2242E+01
0.1756E+04
0.3626E+01
0.1004E+04
0.0
0.1623E+04
0.0
Total Man.
SIC 7535
0.0
0.1709E+01
0.0
0.2764E+01
0.7079E+03
0.1097E+05
0.1145E+04
0.1773E+05
0.1048E+01
0.7720E+02
0.1695E+01
0.1248E+03
0.4748E+01
0.2820E+03
0.7678E+01
0.4560E+03
0.7400E+00
0.1125E+03
0.1197E+01
0.1819E+03
0.0
0.2768E+03
0.0
0.4475E+03
0.4748E+01
0.1836E+03
0.7678E+01
0.2969E+03
Marine Surface Coating
SIC 371 SIC 25
.0.0
0.0
0.0
0.0
0.3052E+04
0.7591E+03
0.4936E+04
0.1227E+04
0.0
0.0
0.0
0.0
0.1840E+03
0.0
0.2975E+03
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.3015E+02
0.0
0.4875E+02
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
.0
.3337E+01
.0
.5396E+01
.1263E+05
.7755E+05
.2042E+05
.1254E+06
.1312E+03
.3643E+03
.2122E+03
.5891E+03
.0
.1417E+04
.0
.2291E+04
.0
.2631E+03
.0
.4524E+03
.4322E+02
. 1 588E+04
.6988E+02
.2568E+04
.2484E+03
.1726E+04
.4017E+03
.2791E+04
Degreaslng Dry Cleaning
SIC 34 SIC 35 & 36
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0
0
0
0
6223E+04
0
1006E+04
3378E+05
0
0
0
9762E+01
1430E+03
0
2312E+03
3465E+03
0
0
0
8561 E+02
6237E+02
0
1009E+03
1206E+03
5360E+02
0
8667E+02
1998E+03
0.0
0.0
0.0
0.0
0.5270E+04
0.1080E+OS
0.3714E+04
0.1980E+05
0.3861E+01
0.2332E+03
0.6243E+01
0.4276E+03
0.6217E+02
0.6462E+03
0.1005E+03
0.1185E+04
0.2287E+02
0.0
0.3698E+02
0.0
0.1280E+02
0.3256E+04
0.2070E+02
0.6467E+04
0.2122E+02
0.9601E+03
0.3431E+02
0.1761E+04
Printing Rubber Transportation
SIC 26 SIC 243S244 Other Users
0.0
0.0
0.0
0.0
0.2297E+04
0.1485E+05
0.1248E+04
0.1896E+05
0.0
0.5890E+02
0.0
0.7522E+02
0.1379E+03
0.5414E+03
0.2231E+03
0.6914E+03
0.0
0.4398E+02
0.0
0.5617E+02
0.7999E+02
0.1119E+05
0.1293E+03
0.1429E+05
0.8200E+02
0.6823E+03
0.1326E+03
0.8714E+03
0.0
0.0
0.0
0.0
0.7719E+03
0.0
0.1248E+04
0.1017E+05
0.1795E+02
0.0
0.2902E+02
0.0
0.0
0.0
0.0
0.1891E+03
0.0
0.0
0.0
0.0
0.2459E-I-02
0.0
0.3976E+02
0.0
0.5115E+02
0.0
0.8271E+02
0.4809E+02
0.0
0.1755E+02
0.0
0.2633E+02
0.2329E-I-05
0.1119E+06
0.3765E+05
0.1679E+06
0.0
0.1049E+04
0.0
0.1574E+04
0.0
0.4289E+04
0.0
0.6433E+04
0.0
0.1241E+04
0.0
0.1861E+04
0.0
0.6384E+04
0.0
0.9577E+04
0.4727E+02
0.5455E+04
0.7643E+02
0.8182E+04
-------
TABLE 4r10 (continued)
SOLVENTS BY USER CATEGORY
2580 Ness, Kansas
Special Naphthas
Total Solvents
1291 Metropolitan Boston,
Massachusetts
Special Naphthas
Total Solvents
140 Coos, New Hampshire
Special Naphthas
Total Solvents
7960 Schuylkill, Pennsyl-
vania
Special Naphthas
Total Solvents
•2760 Jefferson, Texas
Special Naphthas
Total Solvents
Trade Paint
SIC 36
0.7399E+01
0.0
0.1196E+02
0.0
0.4101E+04
0.6370E+03
0.6631E+04
0.0
0.5032E+02
0.0
0.8136E+02
0.0
0.2396E+03
0.2273E+01
0.3874E+03
0.3876E+01
-
0.3577E+03
0.0
0.5784E+03
0.0
Total Man.
SIC 7535
0.3762E+01
0.2343E+00
0.6083E+01
0.3789E+00
0.1190E+02
0.3394E+04
0.1924E+02
0.5487E+04
0.0
0.1717E+02
0.0
0.2777E+02
0.0
0.3006E+03
0.0
0.4861E+03
0.1005E+03
0.1547E+03
0.1625E+03
0.2501E+03
Marine
SIC 371
0.(V
0.0
0.0
0.0
0.2851E+03
0.2137E+03
0.4610E+03
0.3456E+03
0.0
0.0
0.0
0.0
0.4438E+02
0.0
0.7176E+02
0.0
0.0
0.7770E+03
0.0
0.1256E+04
Surface Coating
SIC 25
0
0
0
0
0
0
0
0
0
0
0
.0
.1140E+02
.0
.1843E+02
.1098E+04
.1486E+05
.1775E+04
.2403E+05
.0
.6749E+02
.0
0.1091E+03
0
0
0
0
0
0
0
0
.6346E+02
.1098E+04
.1206E+03
.1775E+04
.5723E+02
.1642E+04
.9254E+02
.-2655E+04
Degreaslng
SIC 34
0.0
0.0
0.0
0.0
0.1322E+04
0.0
0.2138E+04
0.9799E+04
0.0
0.0
0.0
0.0
0.6508E+02
0.0
0.1052E+03
C.2947E+03
0.1553E+03
0.0
0.2511E+03
0.5594E+03
Dry 'Cleaning
SIC 35 & 36
0.0
0.0
0.0
0.0
0.2463E+04
0.5367E+04
0.3982E+04
0.9844E+04
0.0
0.0
0.0
0.0
0.1581E+02
0.4T15E+02
0.2556E+02
0.7546E+02
0.2323E+02
0.6507E+03
0.3756E+02
0.1193E+04
Printing
.SIC 26
0.0
0.0
0.0
0.0
0.1184E+04
0.7825E+04
0.1914E+04
0.9993E+04
0.0
0.2495E+02
0.0
0.3186E+02
0.4837E+02
0.1499E+03
0.7821E+02
0.1915E+03
0.0
0.1625E+03
0.0
0.2076E+03
Rubber Transportation
SIC 2434244 Other Users
0.0
0.0
0.0
0.0
0.1509E+03
0.0
0.2441E+03.
0.3840E+04
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.3659E+03
0.0
0.0
0.0
0.0
0.0
0.3330E+02
0.0
0.4996E+02
0.0
0.3859E+05
0.0
0.5789E+05
0.0
0.3226E+03
0.0
0.4839E+03
0.3179E+03
0.2886E+04
0.5141E+03
0.4330E+04
0.1659E+02
0.2494E+04
0.2682E+02
0.3742E+04
-------
TABLE 4-11
FUEL USE AND ACTIVITY SUMMARY 1973
DELAWARE
Coal (Tons)
Bituminous Anthracite
Total
LPG
(Kilogal.)
Gas (Ft.3 x 106)
Natural
Total Gas
Equivalent
Gasoline Diesel Other
(Kilogal.) (Kilogal.)
-p.
i
00
Fuel Combustion - External
Residential 0.902E+03
Commercial/In- 0.255E+03
stitutional
Industrial 0.714E+04
Total Fuel Combustion
0.829E+04
0.812E+04
0.308E+04
0.0
0.112E+05
0.902E+04 0.116E+05 0.751E+04 0.858E+04
0.334E+04 0.594E+04 0.315E+04 0.370E+04
0.714E+04 0.0
0.307E+04 0.307E+04
0.195E+05 0.175E+05 0.137E+05 0.153E+05
State Sulfur and Ash Retail Industrial
Sulfur % 0.969E+00 0.170E+01
Ash % 0.798E+01 0.107E+02
Transportation
Off-Highway
Railroads
Vessels
Transportation Sub-Total
Aircraft (LTO Cycles)
Commercial
Civil
Military
Total Aircraft
Evaporation Losses
Retail Gas SaTes (Kilogal.) and Solvents (Tons)
Retail Gasoline
Solvents
0.680E+05 0.271E+05
0.126E+04
0.420E+04
0.722E+05 0.284E+05
0.753E+04
0.114E+06
0.473E+05
0.169E+06
0.310E+06
0.195E+05
-------
V. RECOMMENDATIONS
A number of recommendations for improving the reliability of the fuel
allocations are presented in the following sections.
A. IMPROVEMENT OF DATA BASE
1. Point Source Employment
Employment at point sources is a critical element of the indus-
trial allocation methodology. It is recommended that point source employment
be routinely collected. If this information is compiled from Dunn and Brad-
street reports, the accuracy of this source should be determined, and its
impact of the accuracy of the industrial allocation methodology assessed.
2. Coastline Area Factor
Total boat registrations are allocated among counties according
to area, computed as the sum of inland water area and the product of coastline
and a coastline area factor. It was not possible within the scope of this
study to estimate the required factor. It is recommended that the value of
this factor be determined or that an alternate method of estimating county
boat populations be developed.
3. Census of Manufactures, Fuel, and Electric Energy Consumed
The Census of Manufactures has previously published a special
report on fuel consumed by industry, showing fuel use by type of fuel bv state
and two-digit SIC code. It may become an annual publication, in which case it
is likely to be an important source of data. The EPA should indicate its in-
terest in seeing this information produced annually.
4. Sulfur and Ash Content Data
The Bureau of Mines annually publishes sulfur content data by pro-
duction districts. Ash content data, however, are not accurately available by
production district. It is recommended that EPA inquire of the Bureau of Mines
about the feasibility of reporting ash data by production district.
5-1
-------
B. IMPROVEMENT OF METHODOLOGY
1. Fuel Use for Five Commercial Subcategories
Regression analysis was used to develop predictive equations for
fuel consumption as a function of the establishment size and climatology for
each of the commercial subcategories:
. Hospitals
. Hotels
. Laundries
. Schools
•. Universities
The data used to develop these relationships were primarily point sources;
therefore, it was implicitly assumed that fuel consumption patterns of what
are generally the largest establishments are indicative of the much larger pro-
portion of smaller establishments in each subcategory that are area sources.
It is strongly recommended that this assumption be closely examined.
2. Fuel Use for Other Commercial Categories^
Predictive equations should be extended to the entire commercial
sector in a separate study which could result in fuel consumption figures by
two-digit SIC for the commercial subcategories in each state.
3. Solvent Use Patterns
This study estimated for each of seventeen solvents consumption
patterns among six application categories:
. Surface Coatings
. Degreasing
. Dry Cleaning
. Printing and Publishing
. Rubber and Plastics
. Other Solvent Use
It is recommended that these consumption patterns be periodically reviewed for
two purposes:
5-2
-------
• The redassification from "other solvent use" to one of
the five specific categories
• The reexamination of all the consumption patterns to
account for petrochemical shortages and technological
change within the industry.
4. Sulfur and Ash Content of Bituminous Coal
The methodology for estimating sulfur and ash content of
bituminous coal resulted in unrealistically large values for states in which
retail consumption is minor and a large portion of industrial consumption is
by point sources. This is attributed to inaccuracies in sulfur and ash
data reported in the NEDS point source file. It is recommended that an
upper limit of sulfur and ash content be set to the highest sulfur and ash
content of coal shipped to the state.
5-3
-------
VI. REFERENCES
1. Myers, J.P., Documentation of Computer Programs for Countywide Allocation
of Coal, Natural Gas, and Organic Solvents, in preparation for Environ-
mental Protection Agency, Contract No. 68-02-1410, Cambridge, Massachusetts,
1975.
2. Benesh, F., Manual for Preparing the Input for Countywide Allocation of
Coal, Natural Gas, and Organic Solvents, in preparation for Environmental
Protection Agency, Contract No. 68-02-1410, Cambridge, Massachusetts, 1975.
3. U.S. Department of Commerce, National Oceanic and Atmospheric Administra-
tion, Climatological Data (Monthly), U.S. Government Printing Office, Wash-
ington, D.C.
4. U.S. Bureau of the Census, Census of Housing (Deciennial), Washington, D.C.
5. U.S. Bureau of the Census, County and City Data Book, Washington, D.C.
6. U.S. Bureau of Mines, Division of Fossil Fuels, Mineral Industry Surveys:
Natural Gas Production and Consumption (Annual), Washington, D.C.
7. American Gas Association, Department of Statistics, Arlington, Virginia.
8. U.S. Bureau of Mines, Division of Fossil Fuels, Mineral Industry Surveys:
Sales of Liquified Petroleum Gases and Ethane (Annual), Washington, D.C.
9. U.S. Bureau of Mines, Division of Fossil Fuels, Mineral^ ^ndustry Surveys:
Pennsylvania Anthracite Pistributton (Annual), Washington, D.C.
10. U.S. Bureau of Mines, Division of Fossil Fuels, Mineral Industry Surveys:
Bituminous Coal^ and Lignite Distribution (Quarterly), Washington, D.C.
11. American Hospital Association, American Hospital Association Guide to the
Health Care Field, Chicago, Illinois, 1973.
12. American Hotel and Motel Association, Hotel and Motel Red Book. New York,
New York.
13. U.S. Bureau of the Census, County Business Patterns (Annual), Washington, D.C.
14. U.S. Bureau of the Census, 1972 Census of Manufacturers ^pecial Report Series:
Fuels and Electric Energy Consumed, Washington, D.C.
15. Exhaust Emissions from Uncontrolled Vehicles and Related Equipment Using In-
ternal Combustion Engines, Final^ Report Part 5: Heavy-Duty Farm, Construc-
tion, and Industrial Engines (EHS 70-108), Southwest Research Institute, San
Antonio, Texas. Prepared for Environmental Protection Agency, Emission Charac-
terization and Control Branch, and National Air Data Branch, October, 1973.
6-1
-------
16. Census of Agriculture - County Data. U.S. Department of Commerce, Washington,
U • U •
17. Census of Agriculture - Area Reports (Annual), U.S. Department of Commerce,
Washington, D.C.
18. Personal Communication from Project Officer, Environmental Protection Agency,
National Air Data Branch, Research Triangle Park, North Carolina, November,
1974. (Calculations based on data from Reference 15).
19. U.S. Bureau of the Census, Current Population Reports, Estimates of the Popu-
lation of Counties (Selected States) (Annual), Series P-26, Washington, D.C.
20. U.S. Department of Transportation, Federal Highway Administration, Highway
Statistics (Annual), U.S. Government Printing Office, Washington, D.C.
21. Motorcycle Usage and Owner Profile Study (Annual), Hendrix, Tucker, and
Walker, Inc., Los Angeles, California.
22. Area Measurement Report: U.S. Summary (GE-20, No. 1), U.S. Department of
Commerce, Bureau of the Census, U.S. Government Printing Office, Washington,
D.C., May, 1970.
23. The Marine Market 1972, MAREX (International Marine Expositions, Inc.),
Chicago, Illinois, April, 1973.
24. Boating 1972, MAREX and National Association of Engine and Boat Manufacturers.
25. U.S. Bureau of Mines, Division of Fossil Fuels, Mineral Industry Surveys:
Fuel Oil Sales, Annual (Annual), Washington, D.C.
26. U.S. Bureau of the Census, Census of Business: Retail Trade Area Statistics
(Annual), U.S. Government Printing Office, Washington, D.C.
27. U.S. Bureau of Mines, Minerals Yearbook (Annual), U.S. Government Printing
Office, Washington, D.C.
28. U.S. Tariff Commission, Synthetic Organic Chemicals, U.S. Production and
Sales (Annual), Washington, D.C.
29. U.S. Bureau of Mines, Division of Fossil Fuels, Mineral Industry Surveys:
Bituminous Coal Annual (Annual), Washington, D.C.
30. American Gas Association, Gas Househeating Survey (Annual), Arlington, Virginia
31. U.S. Bureau of the Census, Housing Authorized by Building Permits, C-40,
Washington, D.C., 1972.
32. Independent Natural Gas Association of American, Comparison of Seasonal House
Heating Costs, 1971 Washington, D.C., 1972.
33. Telephone Communication with Mr. Elmo Beach, U.S. Bureau of the Census, Hous-
ing Division.
6-2
-------
34. U.S. Bureau of the Census, 1970 Census of Housing: Detailed Housing
Characteristics [HC(1)-Bxx], U.S. Government Printing Office, Washing-
ton, D.C., 1972.
35. U.S. Environmental Protection Agency, Guide for Compiling a Comprehensive
Emission Inventory, Research Triangle Park, North Carolina, March, 1973.
36. U.S. Bureau of the Census, Annual Survey of Manufactures, Fuels, and
Electric Energy Used, By Major Industrial Groups (Annual), U.S. Govern-
ment Printing Office, Washington, D.C.
37. M.S.A. Research Corporation, Hydrocarbon Pollutant Systems Study, Vol. 1,
"Stationary Sources: Effects and Control," Evans City, Pennsylvania, 1972.
38. Stanford Research Institute, Chemical Economic Handbook, Menlo Park,
Clifornia.
39. Chemical Marketing Reporter, Chemical Profiles (Weekly), Schnell Pub-
lishing Company.
40. GCA Corporation, "Hydrocarbon Emission Sources in the Metropolitan Boston
Intrastate Air Quality Control Region," Bedford, Massachusetts, 1973.
41. U.S. Office of Management and Budget, Standard Industrial Classification
Manual. Washington, D.C., 1972.
42. U.S. Federal Aviation Administration, FAA Air Traffic Activity, Washington,
D.C., 1973.
43. U.S. Federal Aviation Administration, Military Air Traffic Activity Report,
Washington, D.C., 1971.
44. U.S. Federal Aviation Administration, Census of U.S. Civil Aircraft,
Washington, D.C., 1971.
45. Aviation^ Directory.
46. McGraw-Hill, Inc., Keystone Coal Industry Manual, New York, New York, 1969.
47. U.S. Bureau of Mines, Report of Investigations, Analyses of Tipple and De-
livered Samples of Coal, Washington, D.C., 1967.
48. Couillard, J., Brown's Directory of North American Gas Companies (Annual),
Washington, D.C.
49. U.S. Department of Health, Education, and Welfare, Number and Characteristics
of Employees in Institutions of Higher Learning, U.S. Government Printing
Office, Washington, D.C., 1966-1967-
50. Telephone Communication with Mr. O'Donnell, Higher Education Statistics
Division, U.S. Office of Education.
6-3
-------
51. U.S. Department of Health, Education, and Welfare and U.S. Office of
Civil Rights, Directory of Public Elementary and Secondary Schools in
Selected Districts: Enrollment and Staff by Racial/Ethnic Groups.
Fall, 1970. Washington, D.C.
52. U.S. Department of Health, Education, and Welfare, Statistics of State
School Systems, 1969-1970. U.S. Government Printing Office, Washington,
D.C., 1973.
53. U.S. Office of Education, Fall Statistics of Public Schools (Annual),
Washington, D.C.
54. U.S. Bureau of the Census, Census of Business: Selected Services.
Washington, D.C., 1967.
55. U.S. Bureau of Mines, Division of Fossil Fuels, U.S^ Energy Fac;t Sheets
(Annual), Washington, D.C.
56. National Coal Association, Bituminous Coal Facts (Biennial), Washington,
D.C.
57. National Coal Association, Bituminous Coal Data (Annual), Washington, D.C.
58. U.S. Bureau of Mines, Mineral Industry Survey: Weekly-Coal Report
(Weekly), Washington, D.C.
59. Commonwealth of Pennsylvania, Department of Environmental Resources,
Anthracite Mine Safety Group, Shipments of Anthracite (Monthly), Potts-
vine, Pennsylvania.
60. U.S. Bureau of Mines, Division of Fossil Fuels, Mineral Industry Survey:
Pennsylvania Anthracite Weekly (Weekly), Washington, D.C.
61. Mr. James Tedesco, Lehigh Valley Coal. Personal Communication.
62. American Gas Association, Gas Facts (Annual), Washington, D.C.
63. Mr. Werich, Statistics Department, National Liquified Petroleum Gas As-
sociation, Chicago, Illinois, Personal Communication.
64. U.S. Bureau of Mines, Division of Fossil Fuels, Mineral Industry Survey:
Crude Petroleum. Petroleum Products, and Natural Gas Liquids (Monthly).
Washington, D.C.
65. "Relative Energy Consumption of Rail and Highway Freight Transportation,"
preliminary findings of a Department of Transportation Study, Cambridge,
Massachusetts, 1973.
66. Association of American Railroads, Railroad Mileage by States, December 3.L.
1971, Washington, D.C., December, 1972.
6-4
-------
67. Handy Road Atlas of the U.S., Rand-McNally Corporation.
68. U.S. Department of Transportation, Rail Service for Midwest and North-
east Region, published in two parts, Cambridge, Massachusetts 1974.
69. Personal Communication with Mr. Kenneth Troup, Transportation Systems
Center, Cambridge, Massachusetts.
70. Gadomski, R., David, M., and Blahut, G., Evaluation of Emissions and Con-
trol Technologies in the Graphic Arts Industries, Graphic Arts Technical
Foundation, August, 1970.
71. U.S. Environmental Protection Agency, Office of Air Quality Planning and
Standards, Air Pollution Engineering Manual, John Danielson, Ed., May,
1973.
72. Noble, P., Marketing Guide to the Chemical Industry, Kline, Inc., 1971.
73. Virgil, G., Petroleum Products Handbook. McGraw-Hill, New York, New York,
1960.
74. Personal Interview with Staff Member of the Chemical Marketing Reporter.
75. Walden Research Division of Abcor, Inc., Development of a Methodology to
Allocate Liquid Fossil Fuel Consumption by County, prepared for the En-
vironmental Protection Agency, Research Triangle Park, North Carolina,
March, 1973.
76. Bureau of Mines, Division of Fossil Fuels, Mineral Industry Surveys,
Coal-Bituminous and Lignite (Annual). Washington, DC.
6-5
-------
APPENDIX A
REGRESSION ANALYSIS OF RESIDENTIAL
GAS CONSUMPTION PATTERNS
Regression analysis was used to construct a predictive equation for county
residential natural gas consumption based on climatological and housing stock
variables. The regressions were performed on a community-by-community basis
for two reasons: (1) The majority of gas companies do not individually ser-
vice enough counties to provide a significant sample for a regression analysis;
(2) it is quite common for a utility to service only a portion of a given
county, in which case that utility's sales in the county could not reflect
total consumption. Two sets of regressions were executed:
Regressions using company gas data published in Brown's Directory
of North American Gas Companies [48], and
Regressions using community gas data provided to Walden by a number
of gas companies across the country
A. REGRESSIONS BASED ON BROWN'S DIRECTORY
A sample of 116 gas companies was extracted from Brown's Directory of
North American Gas Companies. One to four companies that had service areas
roughly compatable with the political units reported in the U.S. Census of Popu-
lation and Housing were selected from most states in Brown's Directory. This
led to an unavoidable bias to rural areas and small towns, as large gas com-
panies serving one or more metropolitan regions were systematically excluded
because socioeconomic and demographic variables could not easily be calculated
for their customers.
Regression analysis of average residential gas consumption with degree
days and average rooms per housing units yielded poor results. Over the entire
2
sample, this regression had an R of 0.184. Following this, the sample was re-
duced to 65 companies for which consumption by residential space-heating cus-
tomers could be separated. A similar regression on this smaller sample with a
dependent variable of average residential gas consumption by space-heating
A-l
-------
customers yielded an R of 0.059. These results compare unfavorably with the
regression analyses using the same variables performed previously (R2 = 0.674)
[75].
Several explanations for the results from the Brown's Directory sample
were investigated. It was thought that the gas utilities' practice of report-
ing a multi-family structure as a single customer might have been a factor.
However, the correlation coefficient between the residuals from the regression
and the percentage of the housing units in single-family structures was -0.022,
which tends to indicate that it is not as important as was thought.
Second,- the 65 companies in our sample for which Brown's Directory
gives separate data for house-heating customers are concentrated in the North
Central and Northeastern regions of the country. This limited distribution
narrows the range of degree days observed in the sample. (The mean of degree
days increases from 4,662 in the entire sample to 5,907 in the house-heating
sample as the standard deviation drops from 2.373 in the entire sample to 1,797.)
Finally, we have doubts about the accuracy of the consumption data in
Brown's Directory. By comparison with the data obtained directly from the gas
utilities, differences of as much as an order of magnitude have been^ observed
in residential consumption.
B. REGRESSIONS BASED ON GAS COMPANY DATA
Walden contacted various gas companies across the country to obtain
community residential gas sales figures. The gas distribution companies listed
below provided data.
Company
Year of Data
States Served
San Diego Gas and Electric 1965-1973
Public Service Company of Colo- 1970, 1971
rado
Pacific Gas and Electric
Rochester Gas and Electric
Baltimore Gas and Electric
Boston Gas
East Ohio Gas Company
Peoples Gas Company
Southern Union Gas Company
1973
1972, 1973
1972, 1973
1971, 1972.
1972, 1973
1973
1970, 1971
1972, 1973
1973
California
Colorado
California
New York
Maryland
Massachusetts
Ohio
Nebraska, Iowa, Minne-
sota, Kansas
Arizona, Texas, New
Mexico, Colorado
A-2
-------
The listed gas companies provided Wai den with the total natural gas consumption
and number of customers in approximately one thousand communities. This was
reduced to a sample of 278 cities, towns, and counties which contained a weather
reporting station. A scatter diagram of degree days and therms per housing unit
is shown in Figure A-l. Each occurrence of a letter represents a single obser-
vation (there is no distinction between letters), while a number indicates more
than one observation at that point.
The independent variables considered for inclusion in the regression
equation and the sources of data for these variables are as tabulated below:
Item
Description of Variable
Source of Data
A
B
C
D
E
F
G
H
J
K
Degree days, heating season
Average wind speed, January
Dwelling units with gas heat
Dwelling units with gas hot
water
Dwelling units with gas range
Percent of dwelling units in
structures built 1960 or later
Rooms per dwelling unit
Percent of dwelling units in
single-unit structures
Percent annual growth of gas-
heated dwelling units in state
Latitude
Elevation
Climatological Data, NOAA (July issue)
Climatological Data, NOAA (January issue)
1970 Census of Housing, Bureau of Census
1970 Census of Housing, Bureau of Census
1970 Census of Housing, Bureau of Census
1970 Census of Housing, Bureau of Census
1970 Census of Housing, Bureau of Census
1970 Census of Housing, Bureau of Census
1972 Gas House-Heating Survey, American
Gas Association
Climatological Data, NOAA
Climatological Data, NOAA
The first problem encountered was the unavailability of certain housing-
stock-related variables (C, D, and E in the above table) for cities with a popu-
lation less than 10,000. Accordingly, the possibility of a structuraltshift in
gas consumption patterns between cities above and below 10,000 population was
investigated. Regressions were performed on the entire sample and two classes
(above and below 10,000 population). An F-test was performed contrasting the re-
duction in the residual sum of squares from the restricted to the unrestricted
A-3
-------
FIGURE A-l
SCATTER DIAGRAM OF DEGREE DAYS (X-AXIS) AND
THERMS PER CUSTOMER (Y-AXIS)
Y
P
I'
2500!
2000
UOOC
60CO
poor
inr-nn
i?ocn
2250
2000
1750
1500
1250
1000
750
500
25
.F!
I
I A
B
. B C P .
I A CO. . |
B CC
F F
.C
F. H
... F 1 ....!...
F.
G
F F
Fr r
. . . . I . . .K. K . . .
. HH2.IK K. *
EF 2fHv' 2 K ,K *
FF2GG
-------
regressions. The null-hypothesis of non-homogeneity between the two classes
was rejected at confidence level greater than 99%. We will, therefore, assume
that the sample of cities with populations greater than 10,000 is representa-
tive of the excluded smaller cities.
A similar analysis was performed to test for a shift in gas consump-
tion patterns in 1973 from prior years. The results, however, were inconclu-
sive, since the geographical distribution of the 1973 data is dissimilar from
the earlier data. Additional data from gas companies are required to make this
analysis feasible.
Two different dependent variables were considered, (1) gas consumption
per dwelling unit using gas, and (2) gas consumption per dwelling units using
gas for space heating. After preliminary analysis, we elected to use the former.
Its use allowed a better fit to the data and involved a less cumbersome methodo-
logy. The use of the latter would have necessitated the estimation of gas con-
sumption by gas customers that do not use gas for space heating.
Two models of gas consumption were investigated, namely, an additive
linear model of the form:
Therms/dwelling unit = $Q + $-,X-, + 32xp + ^3^3 • • •
and a log-log transformation of the linear model:
Q Q Q
Therms/dwelling unit = 3Q • X] ] • X2 2 • X3 3 . . .
The first model assumes no interaction between independent variables and a linear
relationship between an independent variable and therms/dwelling unit. There are
indications that neither is the case. For example, the impact of the size of the
dwelling unit on gas consumption would not be expected to remain constant with a
change in degree days. Also, wind speed has a greater impact on construction
with a low heat transmission coefficient than on construction with a high heat
transmission coefficient. If high heat transmission coefficient construction is
primarily in lower degree-day climates, and vice versa, the impact of wind velo-
city should be greater there.
Regressions in these two forms on an identical set of independent vari-
2
ables yielded, with t-statistics in parentheses, followed by the R , corrected
A-5
-------
R2, and F-statistic,
U
Therms/dwelling unit = -473 + 0.103D + 0.468F + 3.05W + 836
"x
(-3.39) (12.01) (5.14) (3.91) (8.85)
R2 = 0.580 CR2 = 0.573 F(4,220) = 75.997
log (therms/dwelling unit) = 2.68 + 0.336 log (D) + 0.088 log (F)
U.
+ 0.225 log (W) + 0.631 log
x
(9.51) (15.16) (4.36) (4.74) (11.46)
R2 = 0.677 CR2 = 0.671 F(4,220) = 100
where D = Degree days
F = Percent of dwelling units in structures built after 1960
W = Wind velocity
U - = Dwelling units using gas for space heating
U = The larger of the number of dwelling units using gas
for hot water or dwelling units using gas for cooking
The ratio (uqn/ux) 1S used as an index to explain the variation in
therms/dwelling unit between observations with similar climates, but different
percentages of gas customers in a community using gas for space heating. On
2
the basis of the higher R for the log-log transformation, Wai den has elected
to use it as a basis of the residential gas allocation methodology.
Further analysis showed that log (rooms/dwelling unit) and log (percent
dwelling units built after 1960) were multicollinear. Considered separately,
they contributed approximately the same to a reduction in the unexplained vari-
ation. We have elected to use log (rooms/dwelling unit) for the selected
methodology. In addition, wind velocity was omitted from the methodology, as
it offered only a small marginal improvement in the regression and was available
only at approximately 300 stations in the nation. This made the process of es-
timating the average wind velocity of a county one of tenuous accuracy. Accord-
ingly, the regression analysis yielded:
U
log (T) = 3.57 + 0.367 log (U ) + 0.588 log (^-) + 0.125 log (F)
(12.60) (16.35) x(12.02) (1.97)
R2 = 0.631 CR2 = 0.626 F93/221) = 100 Std. Error = 0.1653
A-6
-------
where U = Number of occupied dwelling units using gas
D = Annual degree days
U ,= Number of occupied dwelling units using gas
9 for space heating
U = The larger of the number of occupied dwelling units
using gas for cooking or for hot water
F = Median rooms per dwelling units, in tenths
which reduces to
y 0.588
Therms = 47.5 * U * D°'36 * (A * F°'125
x
A-7
-------
APPENDIX B
REGRESSION ANALYSIS OF COMMERCIAL FUEL
CONSUMPTION FOR FIVE SUBCATEGORIES
The basic methodology to determine commercial fuel consumption on a county
basis is to estimate fuel consumption in five separate categories and to dis-
tribute the remainder within the state by commercial employment. The subcate-
gories are used for two reasons:
. Employment in three of the categories (schools, universities, and
hospitals) is not fully reported in County Business Patterns, and
. Fuel consumption in the five categories was assumed to be relatively
homogenous and distinct from other commercial fuel consumption.
Given this assumption, the county allocation can be made more ac-
curately if these five categories are treated separately.
It was, therefore, decided to analyze the relationship between fuel use
and employment for several subcategories in order to determine the fuel use
for these categories in a direct way, based on the number of employees in
each subcategory. We also anticipated that a positive relationship might
exist between the quantity of fuel consumed and degree days [3 ]. Regres-
sion analysis was, therefore, performed in the following two forms for each
of the subcategories:
Therms = a + B (employment)
Therms = a + B, (employment) + B2 (degree days)
Analysis was also done using a log-log transformation
Therms = a (employment) ^ (degree days) 2
2
However, this latter model generally produced smaller R s than the two linear
forms.
Fuel use consumption data of individual companies and institutions' for the
subcategories was extracted from the NEDS point source file, converted to therms,
and analyzed with employment and degree day data. The results of each of these
analyses are discussed below.
B-l
-------
A. UNIVERSITIES
Fuel use data for 102 universities were extracted from the NEDS point
source file. Employment data were obtained from an HEW survey of employees in
institutions of higher learning [50]. Regression analysis yielded (with t-
statistics in parentheses):
Kilotherms = 24,170 + 224 employees
(16.9)
R2 = 0.742 Standard error = 371,561
Kilotherms = -294,100 + 229 employees +51.5 degree days
(17.83) (2.93)
R2 = 0.763 Standard error = 358,191
The second equation for university fuel use by county is a better fit to the
data sample. This equation will, however, estimate a negative therm use in a
small number of counties with low degree days and low university employment.
For example, a county with 1,000 degree days would need 1,060 university em-
ployees in order to have positive therm consumption. In such cases, fuel con-
sumption by universities will be assumed to equal zero.
Enrollment by institution is available annually from the U.S. Office
of Education in both printed and machine-readable format [50]. Enrollment in
in public institutions is summarized by county; this can be converted to employ-
ment by a public employee/enrollment ratio derived below. The result, public
institution employment reported in County Business Patterns [13], can be used
as the employment input variable for the regression equation.
The derivation of a public institution employee/enrollment ratio is
based on the most recent higher education employment data released by the U.S.
Office of Education [50]. Comparing the enrollment and employment data ,for
1967, the public employee/enrollment ratio was 0.178. This ratio is used to
estimate public institution employment from enrollment.
B. SCHOOLS
Fuel consumption data were available for 85 elementary and secondary
schools from the NEDS point source file. Employment data for instructors by
B-2
-------
school were listed in an HEW publication [51]; total school employment figures
were then calculated from a state ratio derived from a second HEW survey that
listed secretarial, plant, and food service personnel [52]. The regression
analysis yielded
Kilotherms = 1,090 + 172 employees
(9.4)
R2 = 0.517 Standard error = 13,011
Kilotherms = -18,200 + 165 employees + 4.10 degree days
(8.98) (2.05)
R2 = 0.540 Standard error = 12,766
o
The second equation provides a higher R and uses two independent variables to
explain fuel consumption in schools. The negative intercept is not large enough
to be a problem in the school equation, as was the case in the university equa-
tion.
County employment in private schools is available annually from County
Business Patterns [13]. Public school employment can be estimated from the
annual state public school employment [53], apportioned to counties by the popu-
lation of 3-34-year-old persons enrolled in regular public elementary and
secondary schools, as reported in Census of Population, 1970 [5 ]. The sum of
the private and public employment is used as the employment input variable for
the elementary and secondary school fuel equation.
C. LAUNDRIES
Fuel data for 16 laundries was extracted from the NEDS point source
file and correlated with employment figures from the Dun and Bradstreet file
and supplemented by a telephone survey. Regression analysis using only employ-
ment data yielded
Kilotherms = -12,800 + 531 employees
(3.4)
R2 = 0.771 Standard error = 28,298
The coefficient of degree days was insignificant in the multivariate form.
B-3
-------
D. HOSPITALS
The fuel use of 99 hospitals was extracted from the NEDS point source
file. Employment data were obtained from the American Hospital Association
Guide to the Health Care Field [11]. Regression analysis yielded
Kilotherms = 30,100 + 126 employees
(5.88)
R = 0.263 Standard error = 128,758
Kilotherms = -45,200 + 120 employees +15.0 degree days
(5.78) (2.87)
R2 = 0.322 Standard error = 124,201
While the coefficients have the expected signs and significant t-statistics,
2
The R s are disappointing. Further analysis showed somewhat better results by
performing separate regressions of the same form for each Census Region.
The regional analysis still produced poor results. In an attempt to
improve the results, fuel consumption was regressed on hospital bed data ob-
tained from the American Hospital Association [11] and degree days:
Kilotherms = -20,400 + 176 beds +13.3 degree days
(6.30) (2.59)
R2 = 0.353 F (2,96) = 26.17 Standard error = 121,307
When the two regressions were run on the four geographical regions
(northeast, South, Central, and Pacific), it was found that employment and
beds produced substantially different results from one region to another. For
example, employment in the Northeast was a far better indicator of fuel con-
sumption than beds (R2 = 0.771 for employment, compared to R = 0.399 for beds).
Conversely, in the South, beds were a better indicator for fuel consumption
than employment (R2 = 0.780 for beds, compared to R2 = 0.363 for employment).
For this reason, it was decided to run a regression using all three independent
variables: beds, degree days, and employment. Multicollinearity between beds
and employment is not a significant problem; the simple correlation between
them is 0.51. This regression analysis yielded
B-4
-------
Kilotherms = -57,200 + 126.5 beds + 12.7 degree days + 77.4 employment
(4.2) (2.6) (3.5)
R2 = 0.430 F (3,95) = 23.84 Standard error = 114,493
p
Some improvement in the R and standard error statistics was observed. Exami-
nation of the results suggested that the fuel consumption data from the Central
region were primarily responsible for the poor overall performance.
This last regression equation can be used to estimate county hospital
fuel consumption as a function of beds, employment, and degree days. The scope
of the study did not warrant further analysis beyond the effort reported above.
Employment data from the County Business Patterns were found to be in-
complete, as the CBP does not include employment for government hospitals. The
American Hospital Association compiles employment and beds data on an annual
basis. These data are available in machine-readable format; however, their cost
was prohibitive for use in the data processing phase of this study.
E. HOTELS
Fuel consumption data by hotels were obtained from the NEDS point source
file and through contact with the Hilton Hotel Corporation. While employment
data for each hotel proved to be unobtainable, rooms in each hotel were available
from the Hotel and Motel Red Book D^]. Regression analysis for therms on degree
days and rooms yielded (with t-statistics in parentheses)
Therms x 10~4 = -39,100 + 25.54 rooms + 5.514 degree days
(7.35) (2.15)
R2 = 0.797 CR2 = 0.773 F(2,17) = 33.3 Standard error = 12,087
In this case, the critical t-statistics for testing a null hypothesis of a co-
efficient equaling zero is 1.740 in a one-tailed test.
Of the 20 hotels included in the sample, four hotels had rooms in excess
of 2,000. Inspection of the data showed that these hotels were using much more
fuel than would be expected from the rest of the sample. With the four hotels
B-5
-------
deleted from the sample, all three coefficients decreased by an order of mag-
nitude:
Therms x 10"4 = -3,000 +2.95 rooms + 0.437 degree days
(4.32) (1.68)
R2 = 0.590 CR2 = 0.527 F(2,13) = 9.34 Standard error = 1,069
The change between the two regressions and inspection of the data tend
to indicate an appreciable change in fuel consumption per room between hotels
of different sizes; the larger hotels in the sample are using much more fuel
per room than the smaller ones. Therefore, a non-linear regression was run
that allowed a curve with a positive and increasing slope to be fitted to the
sample. Using a log-log transformation,
log (therms x 10 ) = -46.3 + 2.84 log (rooms) + 3.99 log (degree days)
(16.31) (4.11)
R2 = 0.944 CR2 = 0.937 F(2,17) = 143.14 Standard error = 0.882
F 95% = 3.59
In the exponential form, this equation becomes
Therms x TO"4 = (8.049 x 10"21) x rooms2'839 x degree days3'991
which, when used to predict the sample, has a sum of squared residuals of 0.283
x 1010, a standard error of 12,918, and an R2 of 0.768. The standard error of
this regression is slightly higher than the standard error of the linear one on
the entire sample, 12,087, while the R2 is slightly less, 0.766 vs. 0.797.
The data sample and the three equations discussed above are plotted in
Figures B-l and B-2, respectively. The exponential function is used to estimate
county hotel consumption of fuel as a function of rooms and degree days:
Tn = Tn + TF MR - Rn)
-------
R = Total rooms in county
RQ= Average number of rooms per hotel for the county
An RQ for the entire country is used; variation in this parameter on
an individual county basis will not be considered.
The number of hotel rooms is not reported on a county basis; instead,
it must be estimated from the county employment [13] and room-employee ratios
derived from the Census of Business (shown in Table B-l) [54].
B-7
-------
FIGURE B-l. Plot of Hotel Rooms vs. Fuel Use
7 8 9 10
0
1 2
Rooms x 10
B-8
-------
FIGURE B-2. Plot of Three Fuel Use Regression Equations
400
800
1200
Rooms
1600
2000
2400
B-9
-------
TABLE B-l
EMPLOYEE/ROOM RATIOS FOR HOTELS BY STATE
State
AL
AK
AZ
AR
CA
CO
CN
DE
DC
FL
GA
HI
ID
IL
IN
10
KA
KY
LA
ME
MD
MA
Ratio
0.28
0.21
0.29
0.39
0.27
0.29'
0.37
0.13
0.58
0.34
0.33
0.56
0.21
0.32
0.30
0.28
0.23
0.30
0.38
0.16
0.29
0.34
State
MI
MN
MS
MO
MT
NB
NV
NH
NJ
NM
NY
NC
ND
OH
OK
OR
PA
RI
SC
SD
TN
TX
Ratio
0.27
0.35
0.28
0.30
0.20
0.26
0.80
0.23
0.27
0.23
0.31
0.22
0.21
0.31
0.22
0.29
0.36
0.36
0.28
0.16
0.28
0.32
State Ratio
UT 0.22
VT 0.26
VI 0.29
WA 0.26
WV 0.29
VII 0.31
WY 0.20
Source: 1967 Census of Business: Selected Services [54]
B-10
-------
APPENDIX C
ANALYSIS OF ALTERNATE DATA SOURCES
FOR STATE COAL, GAS, AND LPG
SHIPMENTS/CONSUMPTION
A large number of publications on coal, natural gas, and LPG use were
surveyed and evaluated. This survey aided in determining the basic fuel use
sources on which to base the state-by-state figures which form the main struc-
ture for the county allocation methods. The various data sources and the
methods derived to obtain state totals are discussed below.
A. BITUMINOUS COAL
1. Sources
The sole source of data available on the distribution of bituminous
coal by state by consumer class is the Bureau of Mines Mineral Industry Survey:
Bituminous Coal and Lignite Distribution DO]- This is identical with the data
later published by the Bureau of Mines in the Mineral Yearbook (27 ] and the
Energy Fact Sheets [55 J, as well as the source for information published by
the National Coal Association in Bituminous Coal Facts [56] and Bituminous Coal
Data [57 J.
The 1971 state-by-state distribution of bituminous coal, as re-
ported in the Mineral Industry Surveys, is listed in Table C-l. These ship-
ment statistics are based on a quarterly survey of producers and wholesalers
who annually handle in excess of 100,000 tons; the survey covers about 91% of
all coal produced. A facsimile of the quarterly coal canvass is shown in
Figure C-l.
Despite the detailed shipment data, the Bureau of Mines does not
report consumption of bituminous coal within each state by consumer class.
This is only done on a national level in the manner shown in Table C-2 [27 , 58].
While these consumption figures are not directly relevant to the needs of this
project, they will be useful for evaluating the accuracy of using the shipment
data as an estimate for consumption after correcting for exports, imports,
and net year-end change in stocks.
C-l
-------
TABLE C-l
DISTRIBUTION OF BITUMINOUS COAL IN 1971
(Thousands of Net Tons)
Region and State
of Destination Electric
Utilities
NEW ENGLAND
Massachusetts
Connecticut
Maine, New Hampshire,
Vermont, Rhode Island
MIDDLE ATLANTIC
New York
New Jersey
Pennsylvania
EAST NORTH CENTRAL
Ohio
Indiana
Illinois
Michigan
Wisconsin
WEST NORTH CENTRAL
Minnesota
Iowa
Missouri
North & South Dakota
Nebraska & Kansas
SOUTH ATLANTIC
Delaware & Maryland
District of Columbia
Virginia
West Virginia
North Carolina
South Carolina
Georgia & Florida
EAST SOUTH CENTRAL
Kentucky
Tennessee
Alabama & Missis-
sippi
122
1,185
877
7,373
2,862
30,273
38,579
21,790
27,930
19,416
10,449
6,403
4,815
11,655
4,718
1,928
6,408
283
5,821
17,458
17,687
4,589
15,763
21,611
16,637
17,761
Shipments to
Coke and Retail All Other
Gas Plants Dealers (Industrial
(Residen-
tial and
Commercial )
4,188
21,760
10,630
11,164
3,347
4,861
405
509
298
4,369
27
4,323
1,660
174
7,310
14
7
54
2
640
1,299
640
1,871
817
1,299
500
113
73
143
41
41
29
407
239
355
219
77
341
549
101
91
86
63
3,981
110
6,309
12,608
5,005
5,141
7,531
3,187
901
1,311
1,332
411
256
781
286
3,003
4,586
1,737
1,411
455
1,978
1,547
2,522
Total
)
227
1,271
947
15,596
2,974
58,982
63,116
38,599
38,289
32,625
15,340
8,313
6,239
13,358
5,272
2,225
11,599
598
9,258
26,606
19,779
6,219
16,295
25,590
18,907
27,694
C-2
-------
TABLE C-l
DISTRIBUTION OF BITUMINOUS COAL IN 1971
(Thousands of Net Tons)
Region and State
of Destination Electric
Utilities
WEST SOUTH CENTRAL
Arkansas, Louisiana,
Oklahoma, Texas
MOUNTAIN
Colorado 3,
Utah
Montana & Idaho
Wyoming 3,
New Mexico 6,
Arizona & Nevada 2,
PACIFIC
Washington & Oregon 1,
California
Alaska
Destinations Not
Reveal able
Sub-Total 333,
DESTINATIONS AND/OR CONSUMER
Great Lakes Movement
Vessel Fuel
U.S. Dock Storage
Railroad Fuel
United Stated Co.
Coal Used at Mines & Sales to
Net Change in Inventory
019
472
782
542
701
184
083
261
580
017
USERS
Empl
Shipments to
Coke and Retail All Other Total
Gas Plants Dealers (Industrial)
(Residen-
tial and
Commercial )
840 4
901 212
1,787 228
299
O£
S-O
^
\
97
c. 1
oc
oo
1,830 3
1 0
i y
195 117
80,578 10,893
NOT AVAILABLE
oyees
TOTAL DISTRIBUTION
TOTAL IMPORTS
43
343
506
267
160
11
113
313
14
468
278*
69,145
FINAL TOTAL DISTRIBUTION
Growth in Consumers
1 Stocks
ESTIMATED U.S. CONSUMPTION
887
4,475
2,993
1,348
3,728
6,713
2,324
1,482
1,847
748
1,170
493,633
713
-263
528
1,483
397
496,491
111
496,602
-1,643
494,959
* Walden estimate based on total export figures published by the Bureau of Mines,
C-3
-------
Figure G-l.Questionnaire Used by the Bureau of Mines
for Bituminous Coal and Lignite
Fern No. 6-1419-Q
(January T)7 1)
UNITED STATES
DEPARTMENT OF THE INTERIOR
BUREAU OF MINeS
WASHINGTON. D.C. 20240
DISTRIBUTION OF BITUMINOUS COAL
AND LIGNITE SHIPMENTS
DURING THE QUOTES.
O.M.B. No. 42-R1296
Approvol expires Morch 1972
INDIVIDUAL COMPANY
DATA—CON FIDBNTIAL
Vnle^s authorization }s i»rnnird in the
section .i!iovr the signature, the iia:a
fu-r.i>rierf in this report will be trrauxl
in connclrnte tiy the Department .if
tKe Interior, rxcnpl thai they m.»y be
disclosed to Uefrnjir
ITEM 1 —NAMc AND ADDRESS OF COMPANY AND NUM3S* OF PRODUCING DISTRICT OF OR/GIN
A. Name of Company -
B. Address of Company -
Please reply to all pertinent ques-
tions on the form and return ant
tofty as promptly as possible in
the enclosed envelope which re-
quites no postage.
C. Producing District number (See definition in instructions).
ITE.M 2—MIN/E O8/G/N Or COAl
A. Coal Pnducid at mines and cleaning plants of company. (If additional space is needed enter under "Remarks").
Ka*-e e/ minftj or cttaamg ptjxlfs}
Tom
Total company coal produced in District specified in Item 1C above
B. Coal Purchased for further shipment. (Distributor and wholesaler companies report the coal marketed).
(If additional space is needed, enter under "Remarks").
of producing nmfiutfTfjf-rtJ
of Mi
Total coal purchased for further shipment (or, in the case of distributors and wholesalers, coal marketed) from
min« in District specified in Irem 1C above.
=*J== ------ ' ' ~
C. Grand total coal produced and purchased (or marketed, in the case of distributors and wholesalers). (Sum of totals
in Icerns 2A and 2B).
ITEM 3— DISTRIBUTION OF
Report the distribution of the tonnage shown in Irem 2C— Grand
total coal produced and purchased (or coal marketed in the case of dis-
trihutors and wholesalers). Railroad weights should be used when they
are available. If your records do not show exact figures, please estimate
the distribution detail requested. AH figures should be reported in
short ton* of 2,000 pounds. Fractions of tons should be omitted.
METHOD Or MOVEMENT
AND DESTINATION
A. Shipments by all-rail method of move-
ment or;:y (ether than railroad fue!):
!. New England:
... Massachusetts . ...
c. M:'.-.ni NT. H.. V: . ind K. I .
2. M,.i.ii» Atlantic-
. , _ '._
? i..-N,,,nOn:ai
,
LIN5
NO.
1
?
^
t
' 5
('•
•,
UTILITIrS
!
COAL SHI?
COXE AND
CAS PLANTS
(c)
•
PED TO —
RETAIL
DEALERS
(d)
.
1
AIL
OTHERS
!
j
I
I
I
-
(b) THROUGH (e)
(0
\
;
i
C-4
-------
d. M:«*-i«n • •• • •
, e •*' ^'.-v:n ...:
4. West N«ii:h Central:
a. •- 12
h
c.
d. North Dakota and South Dakota.
c. Nebraska and JCaruas 16 j
5. South Atlantic:
a. Delaware and Maryland
b. District of Columbia ' 18
c. Virginia 19
d West Virginia 20
c. North Carolina 2_1
f. South Carolina 22
g. Georgia and Florida 23
6. East South Central:
a. Kentucky 24
b. Tennessee
c. Alabama .and Mississippi 26
7. West South Central (Ark., La.,
Okla. and Tex.) 27
. 8. Mountain:
a. Colorado 28
b. Utah 29
c. Montana and Idaho 30
d. Wyoming 31
e. New Mexico 32
f. Arizona and Nevada 33
9. Pacific:
a. Washington and Oregon 34
b. California 35
10. Alaska 36
11. Canada.. 37
12. State «nd use unknown 38
13. Total all-rail shipments 39
B. Shipments via river method of move-
ment only (other than railroad fuel)
LIST STATES:
Total shipments via river
C. Shipments via ex-river method of move-
ment only (other than railroad fuel)
LIST STATES:
TOWJ shipments via ex-river
Pleote continue form ond sign cerlificotion on rev«r«e
Figure C-l (continued)
C-5
-------
ITSM 3—ojsrft/aunoN OF SHIPMENTS
METHOD Or MOVEMENT
' AND DESTINATION
(o)
LINE
NO.
COAL SHIPPED TO—
ELECTRIC
UTILITIES
COKE AND
GAS PLANTS
(0
RETAIL
DEALERS
ALL
OTHcSS
w
TOTALS Of COlS.
(b) THROUGH (.)
D. Shiprr.ents via Great Lakes ports (other
ch2n railroad fuel):
Nev/York..
Pennsylvania 2
Ohio •- 3
Indiana 4
Illinois 5i
Michigan •. 6
Wisconsin 7
Minnesota , 8
Scare and use unknown 9
Canada 10
Shipments to commercial docks:
United States 11
Canada 12
Vessel fuel '- 13
Total shipments via Great Lakes.... 14
E. Shipments via tidewater ports (other
thin railroad fuel):
Massachusetts 1_
Connecticut... .• 2_
Maine and Rhode Island 3_
New York 4
New Jersey 5
Pennsylvania 6
Delaware and Maryland 7
Virginia
Other States (List)
JlO
11
12
State and use unknown 13
Canada I l4
Overseas exports (except Canada)
Shipments to commercial docks..
Bunker fuel '• I7
Total shipments via tidewater. | 18
F. Shipments via truck (other than ra
road fuel). Destination known—
LIST STATES:
_ ^
3
_4
Srat; and use unknown 2.
Total shipments via truck 6
G. S'iirmenCi via ir.trnway. convenor. 3iiJ
pr'-.'.-ue railroad (other th.m railroad
KT!.-LIST STATES: l
2
Total bhiprr.e-nts vi;i tramway, convtyor,
-------
1. $!:•. r-.rr.:s iii J.s;;:L-uio« ini v.-ho!e-
1 ossi Tons .
:. an»l uft ufi».nown) —
I. Proud: belo»- n^mes. Addresses, and tons shipped to distributors And wholesalers when definitions -".-.? uses ;•-"« :.r.\.r.\,
J'or toul tons itp.vred in Item 3J- (It additional space is needed, enter under 'Remark* ').
Address
Ttr:
f..;/
-------
TABLE C-2
United States Consumption and Exports or Bituminous Coal
(Thousands of Net Tons)
Consumed in the United States
Manufacturing and Mining Industries
1933
1933
1939
1940
1941
1942
1943
1944
1945
1945
1947
1948
1949
1950
1951
1952
1953
1954
1955
1956
1957
1958
1959
1960
1961
1962
1963
1964
1965 . ..
1966
1967
1968 ...
1969
1970
1971 . ..
Electric
Power
Utilities
. . 27,088
36.440
42,304
49,126
59,883
63,472
74.036
76,656
71,603
65.743
86.009
95,620
80.610
83,262
. ... 101,898
.... 103.309
. ... 112,283
. ... 115,235
. . . 140,550
.... 154,983
.. 157,398
. .. 152.928
. 165.788
.... 173,882
179,629
.... 190.833
.... 209,038
... 223,032
.... 242,729
264,202
271,784
. ... 294.739
. . 308,461
318,921
326,280
Railroads-
(Class 1)
72.548
73.921
79,072
85.130
97.384
115,410'
130.283
132,049
125,120
110.166
109.296
94.838
68.123
60.969
54,005
37,962
27,735
17,370
15,473
12,308
8,401
3,725
2.600
2,101
0
0
CO
0
(-•)
0
o
(2)
(')
(2)
(2)
Coking
Coal
40.089
46.626
63,514
81,386
93,133
100.850
102,460
105,296
95,349
83.288
104.800
107.306
91,236
103,845
113,448
97,614
112,874
85,391
107.377
105,913
103.020
76,580
79,181
81,015
73,881
74,262
77,633
88.757
94,779
95,892
92,272
90,765
92,901
96.009
82,809
Steel &*
Rolling
Mills
14,129
11,877
13,843
14,169
15.384
14,722
15,864
15,152
14,241
12.151
14.195
14.193
10,529
10.877
11,260
9,632
8,764
6,983
7,353
7,189
6,938
7,268
6,674
7,378
7.495
7,319
7,401
7.394
7.466
7,117
6.330
5,657
5,560
5,410
5,560
Cement
Mills
2,760
4,413
5,194
5.559
6.735
7,462
5,842
3.767
4,203
6,990
7,919
8.546
7,966
7,923
8,507
7,903
8,167
7,924
8.529
9,026
8,633
8,256
8,510
8,216
7,615
7,719
8,138
8,679
8,873
9,149
8,922
9,391
(2)
(2)
(2)
Other Manu-* Retail"1"
facturing Deliveries
& Mining to Other Total U. S.
Industries1 Consumers Consumption
83.675 77.396 317.685
96.506 66.4S8 335.281
103,401 63,770 375,053
110.853 84.537 450,910
125.184 94,402 492.115
135,993 102,141 540,050
145.191 120,121 593.797
134,567 122,112 589.599
129.754 119,297 559,567
120.364 98.684 500.386
127.015 95.657 545.891
112.612 86,794 519,909
98.685 88,389 445.538
97,904 84,422 454,202
105,408 74,373 468,904
95,476 66,851 418,757
96.999 59,975 426,798
78,359 51,793 363,050
91,110 53,020 423,412
94.772 48,667 432.858
88,566 35,712 413,668
82.327 35,619 365,703
74.365 29,138 356,256
77,432 30,405 380,429
78 050 27,735 374.405
79,453 28.183 387,774
83,467 23.548 409,225
83639 19,615 431,116
86 269 19.048 459.164
89,941 19,965 486,266
84,009 17,099 480,416
83,054 15,224 498,830
85,687 14,665 507,275
83,207 12,072 515,619
68,862 11,351 494,862
To
Canada
8.607
9,577
10,043
13,623
18,376
21,099
24,371
24,516
21,767
22,033
26,170
25,998
16,098
23.009
22,823
20.957
19,584
15,910
17,185
20.654
13,445
12.235
12.407
11,639
11,169
11.410
13,762
14,187
15.661
15,829
15,308
16,748
16,788
18,673
17,565
Exports
To All
Other
Countries
430
913
1,547
2.843
2.364
1,844
1,465
1,516
6,189
19.164
42.497
19.932
11,744
2.459
33,899
26,686
14.176
15.131
34,092
47,899
58,001
38,055
24.846
24,870
23,801
27.003
33,316
' 33.782
34,521
33,474
34,220
33,889
39,445
52,270
39,063
Total
Exports
9,037
10,490
11,590
16,468
20,740
22,943
25,836
26,032
27,956
41.197
68,657
45,930
27,842
25,463
56,722
47,643
33,760
31,041
51,277
68,553
76,446
50,291
37,253
36,541
34,970
33,413
47.078
47,969
50,181
49,302
49,528
50,637
56,234
70,944
56,632
Includes bunker fuel. 2 Included in other Manufacturing and Mining Industries.
Source: U.S. Bureau of Mines
* Used in Industrial Category
+ Used in Retail (Commercial and Residential) Category
C-8
-------
2. Distribution Versus Consumption
If the net change in year-end stock and exports [27,58] is
subtracted from the total U.S. shipments of bituminous.coal in 1971 and
the imports [27,58] are added to it, we obtain an adjusted distribution
figure of 494,959 thousand tons. This compares favorably with the figure
of 494,862 thousand tons shown in Table C-2 as total U.S. consumption of
bituminous coal for that year. We estimate state consumption within con-
sumer class by apportioning the difference between total U.S. shipments and
U.S. consumption in each consumer class by the existing distribution ratios
in each state.
3. User^ Categories
The user categories used in the Bureau of Mines publications
(see Table C-l) more or less coincide with the user categories required for
this project. Shipments to "retail dealers" are equivalent to the coal shipped
to residential and commercial users, and the category "all other" is approxi-
mately equal to our industrial category. Shipments to coke and gas plants and
to electric utilities are not included in this project, since the fuel used
at those facilities is reported in the NEDS point source file.
The consumption category, "retail deliveries to other consumers"
(see Table C-2), is equivalent to residential and commercial consumption.
Industrial consumption is the equivalent of the sum of "other manufacturing
and mining industries" and "steel and rolling mills" (see Table C-2). The
category "steel and rolling mills" is primarily consumption by steel and roll-
ing mills in boilers. The use of coal in coking by the iron and steel in-
dustry is included in the category "coking coal" in Table C-2.
The derived state-by-state consumption of bituminous coal is shown
for selected states in Table C-3.
B. ANTHRACITE COAL
1. Sources
There are two sources of data on the shipments of anthracite coal
by state: (L) the U.S. Bureau of Mines Mineral Industry Survey [9 ] entitled,
C-9
-------
TABLE C-3
ESTIMATED BITUMINOUS COAL CONSUMPTION BY CONSUMER
CLASS IN ELEVEN SELECTED STATE GROUPINGS, 1971
(Excluding Coke and Gas Plants, Electric
Utilities, and Vessel Fuel)
Residential
and
Commercial
Alabama & Mississippi
California
Colorado
Florida & Georgia
Maryland & Delaware
Massachusetts
Missouri
New Hampshire, Maine,
Vermont, & Rhode Island
South & North Dakota
Texas, Arkansas, Louisiana,
& Oklahoma
Washington & Oregon
U.S. TOTAL 11
106
3
223
81
43
15
77
7
151
4
91
,351
Industrial
2,725
15
370
492
844
98
1,429
68
444
46
338
77,422
Total
2,831
18
593
573
887
113
1,506
75
595
50
429
88,773
C-10
-------
"Distribution of Pennsylvania Anthracite," and (2) the Pennsylvania Department
of Environmental Resources [59]. The Mineral Industry Survey is the source
for the data later published in the Energy Fact Sheets [55], while the second
source is used for data published in the Mineral Yearbook [27]. Neither these
two sources nor the National Coal Association develops data on consumption within
each state by consumer class. The Bureau of Mines does estimate national con-
sumption by consumer class in the Mineral Industry Survey: Pennsylvania Anthra-
cite Weekly [60], which is later published in the Mineral Yearbook [27].
In 1972, the Bureau of Mines Mineral Industry Survey switched
from reporting on a coal year (April to March) to a calendar year. According-
ly, the two data series will be compared for 1972 instead of 1971.
The national consumption data are compiled from reports on col-
liery, electric utility, cement, coke, sintering, and other industrial users;
the residential-commercial consumption of anthracite is estimated as the re-
mainder of U.S. production minus exports. The distribution data for both series
are from a 100% survey of sales agents, wholesalers, and dock operators.
The information above is summarized in Table C-4. National con-
sumption estimates by user category, not available on a per state basis, are
given at the bottom of this table. It is evident that neither data series on
shipments accounts for all the anthracite consumed. The total distribution
of either series falls short of the estimated U.S. consumption of 5,915 thousand
short tons. The Pennsylvani series does not, however, include dredge coal.
When the annual dredge coal production of 476,792 tons and the colliery con-
sumption of 11,298 tons are added to the total shipments, the total of 6,345,090
is 7% higher than the estimated consumption. The Bureau of Mines shipments,
including colliery consumption, total 5,554,148 tons, or 7% less than the es-
timated consumption.
The Bureau of Mines is not able to explain the short-fall'between
their distribution series and the Pennsylvania series. While each series dif-
fers by approximately the same amount from the estimated consumption, they
differ in their presentation of state data. The Pennsylvania series includes
C-ll
-------
TABLE C-4
ANTHRACITE SHIPMENTS IN 1972
(Short Tons)
State
Connecticut
Maine
New Hampshire
Vermont
Massachusetts
Rhode Island
New Jersey
New York
Pennsylvania
Illinois
Indiana
Michigan
Ohio
Wisconsin
Iowa
Minnesota
Missouri
Delaware
Maryland
District of Col
Virginia
Other States
U.S. TOTAL
(1 ) Included in
Bureau of
M.I.S.
6,795
7,903
5,028
11,126
27,838
1,264
181,699
742,907
3,816,208
51,218
42,639
57,305
128,785
8,525
(1)
10,405
(1)
16,585
37,356
umbia 7,013
3,894
378,357
5,542,850
Other States
* Rail shipments only; truck
Data Source Estimated
Mines Pennsylvania Residential
and Market Share
Mineral Yearbook of Anthracite
2,555*
7,903*
3,519*
10,317*
23,536*
1,066*
174,000
722,000
4,207,000
47,000*
42,000*
49,000*
124,000*
10,000*
31,000*
10,000*
30,000*
20,000
25,000
3,000
3,000*
290,000
5,857,000
shipments included in Other
90
90
90
90
90
90
90
90
90
—
5
5
10
—
—
—
—
90
90
90
—
States.
Estimated
Residential and
Commercial
2,960
Consumption (10
Colliery
11
Tons) (By Bureau of Mines)
Electric Iron and
Utilities Steel
1,584 757
Other Total
603 5,915
C-12
-------
one extra state in its state-by-state breakdown, but, on the other hand, shows
only rail shipments. Any truck shipments to these states are reported under
the category "other states," causing the state figures to be less complete.
Based on these factors, we have decided to use the Bureau of
Mines series for estimating the consumption of anthracite coal by state. In
addition, this publication is somewhat more easily obtained.
2. Distribution Versus Consumption
The difference between total shipments, after adding the colliery
consumption to the Pennsylvania series, and total estimated consumption will
be apportioned to each state by the existing distribution ratio.
3. User Categories
The rightmost column in Table C-4 shows the estimated residential
market share of anthracite coal. These figures were developed from several
telephone interviews with individuals referred to us by the National Coal As-
sociation [61 ]. Residential anthracite consumption can be estimated from
these figures and calculations from the Census of Housing [34]. From the
above-mentioned interviews, it was learned that practically all the anthra-
cite shipped to states other than the fifteen listed in the residential market
share column of Table C-4 is consumed by industry.
In the remaining states, the total residential/commercial con-
sumption can be estimated by subtracting the NEDS industrial point sources
from total shipments. Residential use is estimated as outlined in the pre-
ceding paragraph, with the remainder considered commercial consumption.
The derived state-by-state consumption of anthracite coal is
shown for selected states in Table C-5. This table assumes zero industrial
consumption of anthracite because the NEDS point source data were not 'available.
C. NATURAL GAS
1. Sources
Three sources of data are available on the sales of natural gas
within each state by consumer class.
C-13
-------
TABLE C-5
ESTIMATED ANTHRACITE COAL CONSUMPTION BY CONSUMER
CLASS IN ELEVEN SELECTED STATES FOR 1971
State Residential and Industrial Total
Commercial
Alabama
California
Colorado
Florida
Maryland
Massachusetts
Missouri
New Hampshire
South Dakota
Texas
Washington
Other States
U.S. TOTAL
0
0
0
0
39,783
29,647
0
5,355
0
0
0
0
*
*
*
*
(1)
(1)
*
(1)
*
*
*
402,939
*
*
*
*
37,356
20,647
*
5,355
*
*
*
402,939
5,915,000
* Included in Other States
(1) Assumed to be zero, will be equal to NEDS industrial point sources;
residential and commercial classes will be adjusted accordingly
C-14
-------
(a) The Bureau of Mines Mineral Industry Survey: Natural Gas
Production and Consumption [6] contains three tables which together account
for the disposition of all the natural gas produced or imported in this
country. This information is summarized in Table C-6. It is identical to
the information later published in the Mineral Yearbook [27] and is recon-
cilable with the consumption data in Energy Fact Sheets [55 J. Residential
and commercial consumption are the same in both publications, while industrial
consumption in the Energy Fact Sheets equals "industrial," "other," "pipeline
fuel," and "lease and plant fuel" in the Mineral Industry Survey. The cate-
gory "other" in the Energy Fact Sheets is reported as "transmission loss" in
the Mineral Industry Survey.
The Mineral Industry Survey is compiled from surveys of pro-
ducers, pipelines, and distributors. All producers and pipelines are included,
while about 80% of all distribution of natural gas is covered. This is extra-
polated to cover all distribution by use of the production and pipeline surveys
and the previous year's distribution. A facsimile of the distribution ques-
tionnnaire is shown in Figure C-2.
(b) The American Gas Association annually collects and publishes
sales data by state and by consumer class in Gas Facts [62]. These are shown
in Table C-7. There is a significant difference between the Bureau of Mines
and the A.G.A. data. As shown in Table C-8, the A.G.A. is consistently re-
porting a total sales volume which averages 85% of what the Bureau of Mines
reports. While neither organization can definitely explain this difference,
the A.G.A. does suggest that the principal reason is their non-coverage of
direct sales from producers to consumers. This is supported by their closer
agreement in the residential and commercial categories and their greater spread
in industrial and electric categories, where direct sales would most often
occur. In addition, it is unlikely that the Bureau of Mines sales data are
too high, since they are able to reconcile their sales data with production
data. The Bureau of Mines data seem to be a more complete estimate of natural
gas consumption.
The A.G.A. data are developed from a 96% survey of gas distri-
bution companies, extrapolated to 100% by revenues. Two different questionnaires
are used; a facsimile of the short form, sent to companies with revenues of less
than $3,000,000, is shown in Figure C-3.
C-15
-------
TABLE C-.6. NATURAL GAS AS REPORTED IN THE MINERALS YEARBOOK, 1971 (Million Cubic Ft.)
J.1],., irit Oilier Ui-livi-n-'l lo i-uiisinnorK „
lu! Comiwrria! 1ml ir-lri i! miluiu.* " ciiirTui,!•"••? " —-— riprline fuel ],C;LS<.' unit phuit furl Kxtnirliim IIHW Tul:il
S! ile :ii,.| KVUTI
Nciv 1 In, .l.ii,l.
i 'i.iii,,-. n-'iil
M i:iir N(W ll:illi|i-
J,,r,-. \|.,||,,!HI...
M ,.-. ,.;,,! a-ll-i
1,1.,. lo lsl.mil
Tulal
Midilii' All.iiilir
Ntw Irr-i'V
N,ll K.k
lVni..'.> Iv.iiim
Tulal....
Fast X'.rlli 1'i'iitrnl:
II1.1I.IK....
l:i.ll iiu. . ,.
t'.i.'iu !".'."". ".!"!!
T.ilal 1.
tt'cil Ni.rlli (Vhlriil:
!.,».,
ku,-..i
MlllllI'Mll.l
Mi^., ,t,
N.-|.r,,-k.i
Nurlli Hakclu
Suuth IVil.uU
Tula!
dotilli All.iiilir.
Uilavan-
l-liinil.i
Mjnlni'i, DiMtirl
O ,,(C..!u!i.l,,ii
1 -N, till I'lil.iili,.;
— ' Si, 'ill, ('.iti'lma
O1 Vitkiln:>
Wc.-n Vitn'mia
Tulal
F.usl Sf.mli (Viiltiil:
k.'l IL\V.' '_'."'."'.
MI-.-H.I.I
T.,'.il
\Vr-t Siimli I'rnlriil:
Ark. ,u-. 11 —
l..i'ii-.iaiu
(lklil..iiu:l
•In.i>
T,,|.il
Mii-ltil.mr
\r,..ut. 1
Ci,|. t.iilil
|.| ,1.,,
M,.|,'..lll.l
S, •» \li \:m.
I'l.ili
\V>.iliiilin
T,.l.ll
1'lK.llu .
Ab-,U
I'al.lunn.l
On-full
W.isl, 11'fl.Jll
Tut.vl
T..V •.: ri.ii.-j si.%\»--
3I.S78
5,. vii
St I'll
12,1 (ID
133,003
II I,I'I7
3"i.'.»s5
311 1.127
T'J'.I.OO'J
1112,011
IH2. 7 li
31 1.77:1
li.il. XII)
lil'.i.7JS
5:i'i.7l2
112.231
'.IS. 1)1 I
102. .125
l.'l'l .1.11
:,7.ii7s
8.5.12
12. Mil
525.200
S.I 72
13. '.'12
SX . .1 1 U
S7.017
30. 131
19, 111)7
49, 1Kb
55,1129
352. 5US
5 1 . S'17
SI, 197
10,190
10,'J'J2
22H.2J2
IS. 737
79.VI3
75. 2 IS
2.17. 3*7
4 II. •_'.'>.•>
32. 01!)
.S 1 , Mi 1
S.I55
25,379
7.991
32.T.l'i
III. M9
19. I'll
2111.019
I1..VI3
010. 99S
21.217
33, '>!l
r.Tl.'ll-J
, i.-jn .'.'."
11, HI!
3.0112
.12. 119
1.1X5
51 ,.',78
.r,s.27S
12.1. '.,( IS
111). 501
2M.I1S7
2113.. WS
7il,isu
UI.S2I
IU-..IS2
44,221
03I.5SS
57. 4 19
I'.I.II2
SI, 395
73.10.1
311.017
9,059
S.81U
285,919
3,010
20,511
37,199
35,495
IS, 718
13,713
25,913
1U.KOZ '
171,094 ,
3.1,902
32.097
21,13(1
10.221
129.959
211,715
32.5I7
.17.32.1
97.V02
197,1112
21,392
01,377
0.779
15,731
5,'.i39
Hi. 9i"
S.IS'.l
12., vjl
M '.1.590
7,531
220, S59
12,117
20,412
f -H7 .222
u •J.^.i/i'i
13. 152 S7
2 097 ll'ii
2.r, IV) 9,915
O.I.-.7 2.791
I7.I-.V) I2.9S9
S> IS5 39.9S5
111, 557 US. 57.1
317.7110 9.7W
siii. 7 is I4.s.::oo
llin.1,70 125. H7I
2 VJ.ISII 31,111)
273, LSI 09.777
:;9s.7:u 20, so9
1 55.935 29.210
l,5ir,.!IX5 277.110
101, 10S 72,329
171.7.19 171.019
105,715 .VJ.U92
IO-;,2-|l 07.H72
55.312 IS, 913
1,011 375
5, ISO 3.3I9
552,270 428,03:1
11.297 3,1173
97,9.13 195.511
142,209 03,170
•I7.7SI 9.423
70 CIO 20,927
7U.250 39.N31
IS.2M 4.201
' 80.153 385
5S9.583 3.li,72i
103,370 9.9S7
7I.ISI 8,910
113,0.11 103,353
127.702 IH. 082
5011,1s? 140.332
IHI.I.I7 SI). 129
1,0011.102 1101,339
I2li 507 2 HI,. SI)'.!
1,93.1,233 1.I07.S2I
3,215,759 1..S50.39I
00,177 liS.Ill)
70.I2S 00.171
28 i 29.1
311, SOO 1.075
III 492 37.011
7S..1IO 49,103
5H 017 2.3SI
50,110 2,911
399. 71« 221,225
10 02H 10,2110
023'.UOII 501. 05S
W.liiil 310
95,971
7.2
5.193
7,5.13
OjS
8.759
3,304
755
9,055
4,717
3.S3.H
IS, 905
1,117
31,998
4,150
•11,721
Si , 2.ili
1,212
I.I2S
1 , 1.13
2.375
5.075
10,771
15
!,1S1
32.5111
0,722
12.S20
1 , 1 III
200
iii. MM
I 3110.276
01.391)
I2.0S8
155.790
25,018
251,1112
325,. S97
712.073
771,795
1,810,3(15
1, 205. (IIS
551,030
1*32,951
1,070.151
312,571
•1.005.921
321, SOU
501, 3IS
313,573
419,121
209,101
19,577
31,827
l,hS'.',3SO
20,452
332,191
33&.X79
IS.1.S09
151,132
153.0I1H
130,018
105,273
1,489,055
201, SSI
200,913
3l3,o:l:l
23il,t.:tS
1,021,095
315,195
1.515,209
ISI.5S5
3. 1.10,. SOI
5.7911,153
IS7.1IU
2S 1,271
II.9SO
SI,3iU
011,511
191.159
118,751
Mi. 9112
1.01,1,137
12,037
2,055,317
9I.OSI
150.520
•2. 311. ON*
ID. on. 212
41
G55
24
723
M71
3.417
27.K91I
.12. 1.17
21.252
12,3110
ir,. hio
12,1173
,>.5G1
U9.31S
19,833
74,099
7, SO!)
'J.liH-t
13.235
10
r>
lun.nari
4,233
11,700
2.111
(i.235
3,195
7 , 1 fi3
10,0(14
39,091
211, (i 10
35,205
59, 4 4 it
20 , 049
141,918
I1.7IG
7(1,120
2f,,S71
99,091
2I2.S04
zzz.-.-^. ~ ~-- -: ^=
20,123
1.SG7
5.11G
7 GO
29,095
537
7,117(1
71, 174
17,210
IS.G7U
O.XI3
G.I 13
•1.1. sr.!
i^s=---. : — ~-
--
4(10
2,419
2,879
407
2,145
4,302
G , 85 1
27,972
1,275
13,1)90
43,237
294
451
213
3,043
4,001
-17G
2,212
5.S40
1,524
10,052
0.433
292,589
101,120
7*4,773
1,184,921
M
3,231
0,1 i,S
•IC.lii
2,115
20.348
7.S.OOO
8.459
75,211
S3, 700
T.°-ua,KOO
culm- fi','1) i
55 '
r>r.
13,520
2,013
15.J.33
39.741
599
3 , 592
43,932
180
10,939
lT.119
2X1
6,133
1,217
7, (131
2 . T.G3
195,072
f.5,914
448. 2S8
701.S37
4,152
750
53, HIO
S.K22
12.KU2
70,336
U»
27,fiS5
27.IIX.I
«i<» rri'7~
111 .-1:11
I2.IMS
i:.il,4.'il
25,002
255,11:15
32(1, VllS
7 Hi, 5.10
,S(i2,l(iH
l,Xl5,4»ll
1.242,797
5(1(1.99(1
HW.Mf.
l.lM7,l2fi
31X,i:I2
4,097.1:21;
341,039
(in;. 7:10
351,412
429.105
224 ,273
37.1 illl
31.H32
2,OC5,I90
2ii.4.r>2
:I3(1.9III
312,579
I.SK.37I
II10,M>7
1.1(1,2(13
141.014
1X9.319
1,544,40(1
2x0.254
250,493
379,038
205,011
1,181,290
330,207
2,07,N,99(i
CG7.49G
4, 813, OK!
7.H95.715
213,313
293 . 52 1
50.01111
K9.021
GG.fill
323,178
12!., 225
I2S.IM8
1,2.>8.953
G7,XOri
2, 170. -119
100,197
15ll.U';:|
'i. , OOL1 . J 1 1
' UJ.UTG.r..-- 1
(inacruiitili-il
fnf
1 .455
» ;'t
1 ,913
44
3, 759
14.M1
'2.S.C42
20,3(1J
C3.5M.) 1
K.39.1 1
6,970
-K.OKI
6.039 1
lii;:l
13,232 4
2 . 520
1G.38G
9,152
5,311
-12,341
017
1C2
21.801 2
291
914
11,050
».K5»
3.854
3,142
4.155
-19.V93
I.I .434
12 IIHK
1511.451
255. t;:i5
32C.76-S
7 1C, 5.1(1
S02. K1X
,845,480
2t2 797
5110, 99G
8.12,575
,0*7.126
34H.132
,01)7,026
31l,r,39
610,730
351,442
429.105
221.273
37,1(19
:il.H32
, OG.1. 19U
2i!,452
33(1,901
342,579
I.s.4.:i7l
1 00 , f.07
150,203
141,014
1X9,319
' 7,472 1,611,4(10
-G.7X7
2.5H2
9 , 03!)
X.M.X
14.2.12 I,
1!) OC,3
•J3,OH| 2,
13..S.13
M.h.l'J 4,
Mo.xr.n 7,
4.II3M
3.079
1.45.1
3.V41
1 ,451
1 1 . Sll t
1.7M
2.250
211.1178 1,
3 , f.N2
:ii;,i2i; 2.
i .Mr,
2,71!i
•\\.\\M 2.
:t.1.s.!'u!) 22.
2KG.2.1I
250.493
379.538
2115.011
Ihl ,290
3:li'.,2o7
O'l> .'IlKi
1107, 19(1
hia.oiu
S!i.'.,715
213,313
291. 1.21
MI, 091;
Ml, 021
in;, r. u
323. I7h
125.225
1 2H , 0.1n
2M.D53
07.- 100
1 76. M'J
liili.n'iT
l.ill.Uii:)
f.()J.-Jl4
U76\ 6M
-------
,6-1340-A
fll0U ;6-73|
UNITED STATES
DEPARTMENT OF THE INTERIOR
BUREAU OF MINES
WASHINGTON. D.C. 20240
B25
SUPPLY AND DISPOSITION OF NATURAL GAS
(Non-producing distributor's report)
O.M.B. No. 42-R0052.
Approval expire* November 30, 1977.
INDIVIDUAL COMPANY
D ATA—CONKID ENTIA L
The data furnished in this report vill
be treated in confidence by lh^ Deporr-
ment of the Interior, e»cep» thai lS?y
may be disclosed to Federal d?fwns«
agencies.
Figure £r2.Questionnaire Used by the Bureau of Mines
correct if name or address has changed.)
Please complete the following form and return ONE COPY. Report all gas volumes at the pressure base of 14.73 pounds
per square inch absolute at 60°F. See instructions on reverse side.
;l STATE covered by this report: ___
3. Supply and disposition of natural gas during the year
Item
(D
Code
Quantity
(Million
cubic feet)
(2)
Value
(Thousands
of dollar))
(3)
A. Supply:
1. Received from producing companies in State designated in
Iteml ., 163
2. Received from pipeline companies:
(Name of company)
171
3. Withdrawn from underground storage.,
4 Total suoolv . . - - 199
f. V U%A. Wff .,7 ..
B. Disposition:
1. Delivered directly to:
971
a. Residential consumers ^ ' i
(1) Number of residential
consumers at end of year.
979
b. Commercial consumers •"•'•^
(1) Number of commercial
consumers at end of year I I 37-
27T
c. Industrial consumers *_>_v_
d. Electric utilities .': 274
071;
e. Other consumers..
2. Own company use....
' r>q 1
3. Stored in underground storage r^° '
fj C\Q
4. Unaccounted for
5. Total disposition (should equal A4)..
Title
Date
C-17
-------
TABLE C-7
GAS UTILITY INDUSTRY SALES
BY STATE AND CLASS OF SERVICE FOR 1971
(Trillions of BTUs)
Class ol' S-.T. ice
Division and SlJto
United Staies
New tn.iiland
C»n nee licut
M.iin-
Mjisach'isettj
New Hampshire
Khr.de hJand
Vermont
Middle Atlantic
New Je ray
New York
Pennsylvania
East North Central
Illinois
Indiana
Michigan
Ohio
Wisconsin
West North Central
Iowa
Kansas
Minnesota
Missuuri
Nebraska
North Dakota
South Dakota
South Atlantic
Delaware
District of Columbia
Florida
Georgia
Maryland
North Carolina
South Carolina
Virjrinia
West Virginia
East South Central
Alabama
Kentucky
Mississippi
Tennessee
West South Central
Arkansas
Louisiana
Oklahoma
Texas
Mountain
Arizona
Colorado
Idaho
Montana
NtV.ldil
New Mexico
Utah
Wyoming
Pacific
Alaska
California
Hawaii
Ore/nn
\\3shiPZlKH
Tot:il
16,679.5
259.6
62.9
I.S
157.4
7.2
27.2
3.1
1,714.9
309.1
61 8.8
787.1
•1.072.1
1.167.4
SI 2.9
8*5.3
1.1-16.0
350.6
1.795.5
319.3
491.4
313.6
423.3
199.4
I9.S
28.9
1,367.9
24.6
28.0
22J.6
337.6
154.0
162.6
133.4
142.1
161.9
936.1
275.8
198.4
2SS.3
256.7
3,200.4
30S.S
738.1
363.1
1,793.7
960.7
177.6
271.1
45.3
79.0
70.5
123.8
1 17.6
71.0
2,322.2
14.9
2.0489
3.0
90.6
16J.6
lU'-iJo.-.v.-'l
5.039.7
141.5
32.3
0.5
90 i
3.-»
12 S
1.2
823.7
14-. 6
361. 1
3190
1,572.-;
4ft?. 1
155.:
355.1
481. S
1 U >
529.3
95.5
95.6
10-:. S
150.6
53V
8.7
11.5
359.3
8.4
14 6
17.6
87.2
7-!.!
2S.9
20.1
S.'.l
S-.3
216J
5S.1
S3.0
3i.:
46.1
422.1
47.0
71.7
73 3
225.0
24-i.l
35. 1
B4 S
S.7
2-!. 4
s.o
2o -
43.5
13.0
725.S
:.<>
66:. 9
0.9
;:.»
35 S
Commercial
2,155.5
50.6
1 1.4
0.4
32.1
I.S
4.8
O.S
273.6
60.7
105.7
107.2
669.8
203.2
73.0
1S6.2
191.9
4S.S
257.8
57.0
37.3
47.0
70.8
28.4 .
8.3
9.0
175.6
3.4
10.3
21.6
39.1
18.7
I8.S
J2.9
I6.S
24.3
117.3
28.3
35.4
15.8
37.7
194.8
27.9
24.4
39.0
103.5
144.5
21.3
58.8
6.9
15.9
6.1
12.2
14.7
8.5
271.6
2.8
232.3
0.7
13.4
2 \4
Industrial
8.643.4
57.9
19.1
0.5
26.4
1.3
9.3
1.4
57S.O
97.7
123.7
351.7
1,758.4
4S2.2
2K2.9
36S.S
453.3
171. S
938.4
142.3
346.4
1S3.R
179.2
106.4
2.1
8.1
773.0
11.3
1.0
157.6
209.0
sa.o
106.0
94.2
56.7
79.1
629.5
190.9
72.6
197.2
I6.S.S
2J23.5
228.1
629 4
138.9
1.327.1
539.9
1 10.9
125.5
28. 2
35.6
55. 7
70.9
59.3
47.9'
1.044.7
6.7
876.5
1.5
51.5
105 6
Other
840.9
9.7
O.I
o.\
8.7
O.S
O.3
0.0
3-1.6
2.1
23.3
9.2
71.5
I6.S
0.9
15.4
1S.9
19.4
70.0
24.7
12.1
8.0
13.7
J0.7
0.5
0.4
60.0
1.5
2.2
26. S
2.2
3.2
9.2
6.3
7.5
1.3
23.0
0.6
7.4
11.0
4.0
260.0
2.5
12. S
1 06. 9
13S.1
32.2
10.3
2 0
1.5
3.0
0.7
13.0
0.0
1.7
230.0
2.3
277.2
0.0
0.0
0.0
Source: American Gas Association [62]
C-18
-------
TABLE C-8
TOTAL GAS SALES
Year
1960
1961
1962
1963
1964
1965
1966
1967
1968
1969
1970
1971
1972
Average
A. G. A.
Total
86.83
85.98
86.27
86.24
86.62
85.73
89. 27
88.69
84. 92
83. 95
83.23
83.73
85. 18
85.65
Data as Percent
1960 -
Residential
99.60
99.08
98.61
99.10
99. 11
99.64
97.73
98.06
99.22
98.87
98.62
98.22
97.79
98.68
of Bureau of Mines
1972
Commercial
87.56
89.15
88.02
87.27
90. 16
90.77
87.32
89.05
91.83
93. 19
94.54
96.13
97. 10
91.64
"CJ'iti ^f'otc't \2j.
Data
Industrial
and Electric
73.07
71.51
72.50
72.69
73.31
72. 52
78.70
78.07
71.67
71. 10
69.40
68.89
70. 53
72.25
#•
Gene Robert Abrams "'
Marketing Analyst
Department of Statistics, A. G. A.
C-19
-------
FigureC-3. Questionnaire Used by the American Gas Association
ANNUAL REPORT OF GAS OPERATIONS AS OF
1973
SHORT FOHM-APPLICABLE TO COMPANIES WITH
ANNUAL GAS REVENUES LESS THAN S3.000.000
Please return by April 1 to
DEPARTMENT OF STATISTICS
AMERICAN GAS ASSOCIATION
1515 Wilson Boulevard
Arlington, Virginia 22209
^ Company:
Acklross:
Ccrr-^oondent and i itls:
List States in which Company has gas operations:
INSTRUCT5ONS
GENERAL
If pos-.ib!=, report ail rras quantities m therms; otherwise, report in MCF and so indicate. Utilities acquiring properties during the year by purchase,
merger, e:c., should report data for that property for the entire year, if possible; utilities disposing of properties during the year should omit the statistics
for that property for the entire year. If company records do not contain all of the information requested, please furnish estimates, designated "£".
SCHEDULE A-GENERAL STATISTICAL INFORMATION
Report all financial data on this schedule in actual dollars, to the nearest dollar. Population served should include entire population within territory
served.
SCHEDULE B-GAS UTILITY PLANT AND CONSTRUCTION EXPENDITURES (Dollars)
Pleaia provide actual current year expenditures and estimated expenditures in each of the next three years.
SCHEDULE C-MILES OF MAIN
Companies operating in more than one state should complete this schedule for each s:ate in which they operate.
FIELD AND GATHERING: Include pipe transporting natural gas from individual wells to compressor station, processing point, or main trunk pipe line,
whichever i; located closest to wells on the line system.
TRANSMISSION: Include main trunk pipe lines and branch lines transporting gas to city gates or between retail service areas, as well as subsidiary feeder
lines not included in field and gathering.
DISTRIBUTION: Include mains and pipe transporting gas within retail service areas.
SCHEDULE D-CLASSlFICATiON OF GAS SALES, REVENUES, AND CUSTOMERS
Companies operating in more than one state should complete this schedule for each state in which they operate.
NOTE: Report Data for All Types of Gases EXCEPT BOTTLED GAS.
Customers: Average number of bills rendered should be based on twelve monthly figures. Utilities not reading all meters each month should make allow-
ance for unreod meters of active customers.
Interruption: Include those customers whose service may be interrupted under terms of the gas contract and who purchase under a separate interruptible
rate schedule. Curtailabie customers, where service may be interrupted only during emergencies, are to be considered firm. If any customers in the
residential, "other," or resale categories are interruptible, please list the appropriate therms and dollars on a separate sheet.
O:riir Sa.'?s and Oth-sr Revenues lines 7, 8, & 9, in columns 1 & 2: Include data pertaining to sales and revenues in NARUC Accounts 482 and 484.
ff^sidsntial Hasting Data: Furnish estimates, if figures are not available directly from company records, of all residential customers using principally gas
for house heating (central and space). Sales to, and revenues from, residential heating customers should include amounts associated with other appliance
usage of these customers.
SCHcDULS A-G5NERAL STATISTICAL INFORMATION
1. G;;S Operating Revenue (Dollars) .
2. Gas Operating Expenses (Dollars) ....
3. Gas Operating Income (Dollars)
Total Gns Purchases:
4. Amount (Th2rms) ... . . . .
5. Co;t (Dollars) ...
Population and Territory Served By Gas:
6. Population Served . ... ...
7. S-.;'jf>r;. Miles of Territory Served . . . .
1.2,
51,3,
51.4,
51.5,
il.G,
51,7,
PAGE 1
C-20
-------
IAGE2
Figure C-3 (continued)
ANNUAL REPORT OF GAS OPERATIONS AS OF
1973
State of
PAGE 2
Total System
DO NOT FILL IN - A.G.A. Use Only
Year Rerjion State
HEADING. 1973,
Co. Type Co. Coda Release
:ru;DUL£ B-GAS UTILITY PLANT AND CONSTRUCTION EXPENDITURES (Dollars)
1 Intangible Plant
2 Production
3 L)nd3rground Storage .
4 Other Storage
5 Transmission
6 Distribution ...
8. TOTAL GAS PLANT
50,1,
50,2,
50,3,
50.4.
50,5,
50,6,
50,7,
50,8.
GROSS
UTILITY
PLANT
GAS CONSTRUCT! ON EXPENDITURES
1973
li
li
!
'
1
!|
i
1974
1975
-
1975
SCHEDULE C-fAILES OF MAIN
Miles of Main as of December 31
1. Field and Gathering
3. Transmission
4. Distribution
5. Service Piping
6. TOTAL MILES
26.1., .,
26,2, , , ,
26,3., , ,
26.4. , , ,
26,7. , , ,
26.10,,.,
STEEL
(Including
Fiberglass)
OTHER
SCHEDULE D-CLASSJFICATION OF GAS SALES, REVENUES AND CUSTOMERS
SALES
(Therms)
REVENUES
(Dollars)
CUSTOMERS
Average
RESIDENTIAL
1. With Heating
2. Without Heating
COMMERCIAL
3. Firm
4. Interruptible
INDUSTRIAL
5. Firm
6. Interruptible ;•
OTHER
1- Municipal and Other
Public Authorities-. :
8. Interdepartmental
9. Other
10. TOTAL SALES TO CONSUMERS
H. Other Gas Utilities (for Resale)
!2. TOTAL GAS SERVICE SALES
!3. OtlvT Gas Revenues
K TOTAL GAS OPERATING REVENUES.
1^. Average Btu as Distributed . .
20,1,
20,2,
20,3,
20,4,
20,5,
20.6,
20,7,
20,8,
20.9,
20,10,
20,11,
20,12,
20,13,
20,14.
20,15,
Column 1
xxxxxxxxxxx
xxxxxxxxxxx
Column 2
Column 3
r_cn
-------
(c) The third source of data on natural gas consumption is
Brown's Directory of North American Gas Companies [48], which publishes
annual sales by each company by consumer class. It distinguishes between
heating and non-heating residential customers. However, this information
is incomplete for many of the smaller companies, where sales are only re-
ported as a total. The company sales data in Brown's, when totalled for
each state, fluctuate above and below the state consumption data in the
Mineral Industry Survey, due in part to a difference in the definition of
industrial consumption. Also, when using Brown's, it is difficult to ac-
count for interstate transfers. Due to this problem and the incomplete
reporting by-some companies, BrownJ s is not suitable for developing state
totals of gas consumption.
In conclusion, we consider the natural gas sales reported
by the Bureau of Mines to be the most accurate statewide consumption esti-
mates available. The user categories are as required for this project.
D. LIQUIFIED PETROLEUM GAS (LPG)
1. Sources
The sole generator of data on LPG sales is the Bureau of Mines;
they are reported annually in the M.I.S. Sales of Liquified Petroleum Gas
and Ethane [8 ]. This is the source of data later published by the National
Liquified Petroleum Gas Association [63], The Bureau of Mines data are shown
in Table C-9.
These statistics are compiled from a survey of companies selling
in excess of 100,000 gallons, accounting for 84% of all consumption. The
total is extrapolated from district demand information on the Bureau of Mines
monthly petroleum statement [64]. A facsimile of the survey is shown in
Figure C-4.
2. Sales Versus Consumption
We assume the LPG sales data shown in Table C-9 to equal consump-
tion. We are aware of a small error factor which is introduced here due to
the storage capabilities available to the ultimate consumer. It is felt, how-
ever, that this error is too small to warrant further research into this matter.
C-22
-------
TABLE C-9
SALES OF LIQUIFIED PETROLEUM GASES AND ETHANE BY USE,
EXCLUDING USE IN GASOLINE PRODUCTION,
BY P.A.D. DISTRICT AND STATE: 1972 AND 1971
(Thousand Gallons)
?.A.D. Dint rice
and Stdt«
District I:
Maryland and 01 3 trice of Columbia...
District II:
Total
District III:
Total
District IV:
Wyoiing
Total
District V:
Ilavatl -
Total
United States, total................*.
Residential
aod corarelal
1972
42,242
20,473
279,372
212.685
21,802
57.539
51,780
29,973
45,616
153,852
153,505
100,092
8,176
96.855
23,165
79,011
17.71'.
478,042
358,578
37J.935
218,923
183.771
239,063
375,199
469,607.
197,154
65,142
232,069
289,587
110,716
129,218
311.497
4.104.511
285,694
371,769
149,613
276,473
102,947
758.535
1.947.011
185,203
45,137
54,074
41,474
63.931
110.121
4.641
44,773
223,323
21,739
23,069
19,920
41.313
412.738
8,253,340
1971
38,461
19,310
271,921
208,9-'.S
13,537
50, 5^3
46.536
25,284
40.025
141,753
144,535
92,140
7,231
87.910
19,162
71,197
14.441
431,462
315,739
3U.19A
278,835
169,581
242,411
338,520
446,953
135,053
62,285
202,400
275,075
97,269
115, 260
236.008
3.715,192
260,431
343,912
Ii3 ,693
253,123
92,371
731.307
1.874.1)39
161,345
38,991
44,721
36,692
54.319
316.113
4,379
42,431
266,915
25,993
30,731
46,619
55.947
474.152
7,«M.4U
Internal co-abustian
eagloA fuel
1977
2,763
2,514
22,575
11,564
377
4,776
4,022
628
11,815
15,460
10,447
17,221
3,519
9.101
251
7.290
1.760
58,045
13,374
6,260
41,109
6.829
9,105
11,951
9.105
25,040
235
19,349
59.290
6.968
9,675
10 523
237.113
12,963
99,424
42,870
59,893
34,332
700.146
949.628
24,578
2.221
8.352
952
16.705
52.609
8,569
45 213
2,274
948
2,087
4.2«
63. 155
1,479,190
1971
2,511
1.992
20.736
10.702
303
4,162
3,504
596
11,414
13,690
7,938
15,841
3, If)
7,320
375
6,423
1.113
58,473
12,416
6,604
19,352
6,179
7,922
10,589
9,690
24.568
259
16.237
55,598
5,911
11.107
8 7*9
274,929
11,542
82,807
39,801
52,135
27.796
632.047
844.128
17,871
1,271
5,546
60]
12 207
37.500
67
7,098
37,042
1,963
1,031
2.0)9
4.261
53,501
1,124, I2i
Industrial
I/
1972
20,455
1.3*4
12.794
11. U4
5,114
11,712
11, 9*1
2. 735
51.222
40.696
23.655
57,344
3.346
22.212
1.115
15.616
U.500
73.749
19,756
24,421
15,688
17.648
14.C17
35.516
18.955
13.143
7,543
36.799
23,153
5,635
5.222
33.371
346. ?<>4
15,309
19.230
162.773
24,235
9,291
61.355
292 796
5,865
6,677
8,357
3,261
12 470
13 110
3,463
1,959
112,712
3,377
3,250
6.015
115 -\
1 ,124.253
W71
18,522
1,677
12,651
10,431
3,737
10.276
10, 0><8
2,324
48,215
38,364
20,613
52,455
2,815
19,982
1.315
12,537
14.814
75,788
19,959
22.329
13.135
8.444
14,518
35.265
17, UJ
11,725
7,650
37,333
24,527
4.348
4,973
30.205
327. "32
11,126
16,150
59.910
21.431
6,397
50.391
165^627
5,607
6.052
8,475
2.834
11 571
34,591
2,925
1,915
80,241
2,223
3,353
4,323
4 802
99 807
908,965
Utility
g"
1972
15.458
728
12, 962
4.518
3,618
15,553
19,171
11,742
11,199
3,058
1,019
19,951
2,767
5.514
4,806
9.758
9,128
496
9,126
5,157
5,555
6,130
15,772
13,094
3,273
17,061
7,142
2,975
4 CH7
99,195
311
5.130
5.652
n.toi
2.580
: 530
31,934
9,802
572
42,303
302,481
1971
13.276
474
B, oca
3.621
2,773
10,153
U.3JO
9,609
6.3:4
2,750
1.173
11,531
1.72n
3.302
3,938
6,134
i,153
214
J.132
3,862
3,779
3,170
17,270
5,944
1,111
19,673
3.489
1.972
2.773
7i.«:
116
2,594
2.632
5 612
' 1,628
1 '.?3
20,342
10.271
247
ID RIO
201,773
Miscellaneous
u-«» 1]
1972
1,109
1.190
2.353
25,356
164
1.109
2,781
899
932
4,662
21,5*4
2,945
Ia4
7,351
5.971
31,881
20,69"
13,270
5,669
1,735
2,113
8.626
3,892
3,700
69.V
6,682
1,619
1,715
540
6.822
116 .117*
2,069
12,751
8,922
7,203
2.073
11. 591
46,611
7,580
4,753
300
3,675
1 161
17. S6'
4.562
40,74*
3,830
1.104
52 492
315,5-18
1971
1.477
1,111
3.124
22,843
130
1,140
2,353
735
803
3,237
22.4<,"
2,445
137
6,915
10
4.701
25,194
12,275
17,752
5,210
1.547
1.710
8,517
3,510
2,351
430
4,572
1.575
1.3W
550
5.501
92 J12
1,737
9,779
8,740
5,8*4
1,317
7.P-0
34 7?7
5.810
2.E33
153
2.256
I 091
12 :oi
4,145
13,372
3,658
3.404
50 MO
2i:.94l
Tpeal 3/
1972
81.229
26.499
330,55*
24'., 297
31,079
90,719
89.945
48,013
123.014
222.728
214.590
197,359
17,972
141.261
29.117
117.668
33 »?4
2 717 9'13
652.845
412.902
432.012
301,594
215,340
320, OS8
437.4*2
517.111
252.231
76.942
311,960
375, 4i9
132.257
147,630
363.3*7
5.677.263
316.146
505.22*
364,183
372.939
•148.643
1 539.792
ll.9S2.343
225.80*
59,03»
71.593
51,362
94 520
502.119
8,104
59,863
453.913
39.692
32,107
50,963
57.168
951 S62
21,833,7,00
1971
74,26'>
24.564
316,452
256,54*
25,535
76,52*
76.974
38,349
107,301
1»9.SO»
195.611
m,»u
15.242
125.449
24,800
101 .247
3-1.415
2 Ml. 044
396.073
350. 67J
353,511
286.531
139,811
270/.00
196,0-91
494,626
229,641
71,795
2*0;12O
337,795
112,355
133.872
311.25-5
5,1*0.1'!.
285,17*
452.648
251.664
335,167
127,8^1
1 424.451
I0.l0j.5'i2
192,261
49,197
38.897
41.435
79.243
422. 0<3
7,371
35,610
443.412
40/.S2
35.177
56,709
69.661
979.525
19.131,542
i/ Includes r.)K-ery fuel of 610,890,000 gallons In 1972, and 237, /CO, 000 gallons :s 1971.
2/ Includes secondary recovery.
3_/ District totals do not equal the s«=» of State totals because of the Inclusion la district total! and t>.a exclusion In State totals rf figures Csc cheeilcal and
synthetic rubher, to avoid disclosing company data. Dtta for these uses are sy.ovn In TsSle 8.
N*>t«: District »ale» totat* differ from thn JlSErlct for rail «nd true*
Int«rdl5crlet
C-23
-------
Sales of liquefied petroleum gases and ethane during the year by States of destination and by uses
(See insfruef/ons and definition* on reverse side)
t names of Steles in column Headings)
soi and products
1.033s'5'*
l.OOOgols.
1.0UO gols.
1,000 goli.
A. Direct soles to consumers by your company:
1. Residential and commercial uses:
o. Propane 201
b. Butane 301
c. Butane-propane mixtures401
2. Internal-combustion engine fuel:
a. Propana 202
b. Butane 302
c. Sutane-propane mixtures40..
3. Industrial uses (me/, petroleum refinery fuel):
a. Ethane „ 1 03
b. Propane 203
c. Butane303
d. Butane-propane mixtures
4. Gas distribution companies:
o. Ethane
b. Propane 204
c. Butane304
d. Butane-propans mixtures
5. Raw materials and solvents for chemical plants:
a. Ethcne-ethylene
b. Propane 205
c. Bufane-isobutane305
d. Butane-propane mixtures
6. Substitute (synthetic) natural gas feedstock:
a. Ethane 112
b. Propane 212
c. Butane312
d. Butane-propane mixtures
7. Agricultural uses:
a. Propane
b. Butane 313
c. Butane-propane mixtures ,
8. All other uses:
a. Propane
b. Butane 30£
c. Butane-propane mixtures4Ub
B. Sales to dealers, resellers, producers, and to
refineries for gasoline blending:
a. Ethane-ethylene
b. Propane
c. Butane-isobutane
d. Butane-propane mixtures4UV
C. Total sales (Sum of A and B):
111
a. Ethcne-ethylene ——
71
b. Propane -=±
311
c. Butone-isobutane
d. Butane-propane mixtures
If this company changed ownership during the year, please report name end address of present owner, and date sold:
FigureC-4. Questionnaire Used by the Bureau of Mines for LPG
; N'cn e
Addressi
.'Dote sold;
I,lie
Dote
C-24
-------
3. User Categories
The residential and commercial categories are lumped together
in the LPG sales statistics shown in Table C-9. The residential consumption
of LPG is estimated based on the number of housing units using LPG for heat-
ing in each county [ 4]. The state total use of residential LPG is the sum
of these county estimates and can be subtracted from the reported residential/
commercial category to obtain the commercial use of LPG. The industrial
category is reported separately and can be used as is. The categories "in-
ternal combustion engine fuel" and "utility" and "miscellaneous uses" are
not included here.
E. SUMMARY OF DATA SERIES TO BE USED FOR STATE COAL AND GAS CONSUMPTION
1. Bituminous Coal
U.S. Bureau of Mines, M.I.S. Bituminous Coal and Lignite Distri-
bution Quarterly [10]
U.S. Bureau of Mines, M.I.S. Weekly Coal Report [58]
2. Anthracite Coal
U.S. Bureau of Mines, M.I.S. Distribution of Pennsylvania Anthracite [9]
U.S. Bureau of Mines, M.I.S. Pennsylvania Anthracite Weekly [60]
3. Natural Gas
U.S. Bureau of Mines, M.I.S. Natural Gas Production and Consumption [6]
4. LPG
U.S. Bureau of Mines, M.I.S. Sales of Liquified Petroleum Gas and
Ethane [ 8 ]
F. COMPARISON WITH CENSUS OF MANUFACTURES FUEL DATA
For 1971, the Census of Manufactures published a Special Report on the
use of fuel and energy by manufacturing enterprises [14]. Table C-10 shows a
comparison of the industrial coal and gas consumption data derived by means of
C-25
-------
TABLE C-10
COMPARISON OF BUREAU OF MINES BASED ESTIMATES OF INDUSTRIAL FUEL USE
AND 1972 CENSUS OF MANUFACTURES, FUEL AND ELECTRICAL ENERGY CONSUMED,
FOR ELEVEN SELECTED STATE GROUPINGS, 1971
Natural Gas, McF
Alabama
California
Colorado
Florida
Maryland & District of Columbia
Massachusetts
Missouri
New Hampshire, Maine & Vermont
South Dakota
Texas
Washington
TOTAL, 14 States
Bituminous Coal, 10^ Tons
Alabama & Mississippi
California
Colorado
Florida & Georgia
Maryland & Delaware
Massachusetts
Missouri
N.H., Maine, Vermont & R.I.
S. Dakota & N. Dakota
Texas, Arkansas, Louisiana & Oklahoma
Washington & Oregon
TOTAL, 22 States
Bureau of Mines
163,370
623,006
76,428
97,963
47,781
25,453
108,231
2,697
5,480
1,933,233
95,974
3,179,616
Bureau of Mines
(Bituminous Only)
2,725
15
370
492
844
98
1,429
68
444
46
338
6,869
Census
157,600
467,600
40,800
73,700
38,900
24,700
99,600
4,000
200
1,565,200
90,300
2,562,600
Census
1,916
16
276
633
906
33
1,414
41
60
1,175
125
6,595
Difference
5,770
155,406
35,628
24,263
8,881
753
8,631
-1,303
5,280
368,033
5,674
617,016
Difference
809
-1
94
-141
-62
65
15
27
384
1,129
213
274
Census as I
of BuMines
96.4
75.1
53.4
75.2
81.4
97.0
92.0
148.3
3.6
80.9
94.1
80.6
Census as I
of BuMines
70.3
106.7
74.6
128.7
107.3
33.7
99.0
60.2
13.5
2,554.3
37.0
96.0
C-26
-------
the methods discussed above and the statistics published by the Census of
Manufactures.
There are some fairly large discrepancies between these two sources.
In general, the discrepancies are high for those states in which the parti-
cular fuel is very little used in comparison to other fuels, e.g., gas use
in New Hampshire, Maine, and Vermont; coal use in all the New England states,
South and North Dakota, Washington, and Oregon. This fact seems to indicate
that the Census of Manufactures sample was probably not large enough for those
fuels in those states to arrive at the total statistical universe.
C-27
-------
APPENDIX D
ALTERNATIVE METHODOLOGIES FOR ALLOCATING RAILROAD
USE OF DIESEL FUEL
A. COUNTY APPORTIONMENT ACCORDING TO ENERGY CONSUMPTION
Using the results of a recent Department of Transportation study,
the energy requirements for railroads many be expressed as follows [65]:
e -
where e = Fuel requirement (Ib/ton)
e = Rail energy consumption (HP-hr/ton-mile)
I = Distance (miles)
n = Engine efficiency factor (HR-hr/lb')
r = Fuel consumption rate at idle (Ib/hr)
t = Idling times associated with run of distance I (hours)
w = net load (tons)
Subsequent attempts to find data for the variables identified in the
above expression on a county basis are summarized below.
1 . Railroad Mileage
Railroad mileage statistics are available by state from the Asso-
ciation of American Railroads [66]. However, the A.A.R. does not publish or
collect any county statistics on railroad mileage.
The Rand McNally Handy Railroad Atlas of the U.S. [67] provides a
map of each state indicating all rail lines and the mileage on each section
from station to station. These maps do not show the county boudnaries, but
these can be found in Rand McNally 's Commercial Marketing Atlas. The pro-
cedure would then be to estimate the number of railroad miles within edch
county, using the two atlases. This approach is very time-consuming. Further-
more, mileage information without any density data would not be a useful dis-
tributive factor.
D-l
-------
2. Railroad Traffic Density
A complete survey of the rail services for the Midwest and North-
east regions was undertaken recently by the Department of Transportation [68].
This study subdivides the regions into 323 zones which follow SMSA and/or
county boundaries. Rail traffic density for each line in each zone is given.
Similar studies may be made of the rail service in other regions of the
country in the future [69].
Whereas the density information available for a limited number
of states is .considered a more accurate distributive factor than railroad
mileage, it still contains two major drawbacks:
. Engine idle time is not explained by this variable
. For those counties where the carload variable can be
used, considerable time has to be spent in trying to
match the traffic zone boundaries to the county
boundaries required for this study.
Insofar as there is significant variation in the idle times from one railroad
line to the next and the idling mode can account for as much as 50% of the
total fuel consumption, the inability to quantify this variable would intro-
duce a number of uncertainties into this approach.
3. Ton-Miles
The Census of Transportation compiles ton-miles for railroads for
the census year on a state-by-state basis. Ton-mileage data, however, are not
disaggregated to the county level.
4. Population
The method adopted for allocation of statewide railroad use of
diesel fuel to counties is to apportion the state total according to popula-
tion distribution. This decision is based on the following considerations:
. Distribution of railroad diesel by the Department of Trans-
portation energy algorithm is not viable, due to lack of
data on the county level
D-2
-------
. Apportioning by the time distributive factors considered
above would not significantly improve the accuracy of the
county estimates. Furthermore, the fact that rail fuel
consumption doubles for every degree of rail grade [69 ]
makes the topography of the county an additional variable
to be accounted for
. The time involved in coding rail mileage and carloads by
county would be beyond the scope of this study
Data on the use of diesel by railroads for each state are obtained
from the Bureau of Mines Mineral Industry Survey [25 ].
D-3
-------
APPENDIX E
NATIONAL USAGE OF ORGANIC SOLVENTS
Using the data from the Chemical Marketing Reporter[39] and the SRI
Handbook [38], the national consumption of the primary solvent groups, dis-
tributed by user category, is obtained. These national consumption figures
are used for the countywide allocation and are summarized in Table 3-11. A
description of the basis for arriving at these 1971 consumption estimates is
given below for each of the main solvent groups identified in Section 111.H.I.
Special Naphthas. This group of organic solvents finds wide applica-
tions in industrial and commercial uses and accounts for an estimated 59
percent of total organic solvent consumption.
The use of special naphthas in surface coatings is obtained from SRI
[38]. As shown in Figure E-l, the SRI category of aliphatic hydrocarbons
is comparable to the Bureau of Mines category of special naphthas. The
700 x 10 Ibs of aliphatics reported by SRI agrees fairly well with estimates
by the MSA study [37]. These figures, however, do not account for solvents
and thinners added by the users before application. Since the solvents used
are primarily hydrocarbons, the distribution of chemical types used for sur-
face coatings (Figure E-l) can be used to estimate the additional aliphatic
hydrocarbons that must also be included. Using SRI's 3390 x 10 Ibs of total
solvent use for surface coatings and an assumed equal amount of hydrocarbons
added by the user before application, the additional aliphatic hydrocarbon is
given by
(3390 x 106) * (7007°°80Q) = 1582 x 106 Ibs
When this additional contribution is added to the raw material use, the surface
coatings share of special naphthas works out to approximately 26%.
Use of special naphthas in the printing and publishing industry is' derived
from a 1969 survey of aliphatic solvent use in the graphic arts industries [70 ]•
When the reported aliphatic solvents from the survey are extrapolated to the
E-l
-------
•Hydrocarbons 50 (1500)-
— Aromatic 30 (800)
J— Aliphatic 20 (700)
Solvents
--Oxygenated 185 (1850)
.—Ketones 70 (80)
—Esters 30 (350)
— Alcohols 45 (550)
Glycols
Glycol Ethers
20 (500)
Other Solvents 10 (40)
•Chlorinated Solvents
•Nlitroparafins
— Turpentines and Pine Oil
Figure E-l. Surface Coatings Industry
Raw Materials Used in $ x 10
(And Ibs x 106)
E-2
-------
national level—on the basis of total employment in these industries — an es-
timated figure of 94 x 10 gallons is obtained. This consumption represents
8.25% of total demand for special naphtha for that year and, in the absence
of better data, will be used in the current estimate.
In the dry cleaning industry, the use of special naphthas (e.g., stoddard
solvent and safety solvent) is estimated to be about 500 x 10 Ibs for 1971.
A recent publication [71] indicates that the use of special naphthas in
degreasing applications is negligible. Similarly, no data were obtained to
permit a meaningful estimate of consumption by the rubber and plastics industry.
The end results of the special naphtha use analysis are shown in Table 3-11.
Perch! oroethyl ene . The use of perchloroethylene is derived from data in
the SRI Handbook [38]- Virtually all consumption was for solvent use, as
shown in Table 3-11, except for 17% which was exported.
Ethanol . The SRI Handbook [33] states that 4% of all ethanol consumed in
1971 was for surface coatings, while the Chemical Marketing Reporter [39] states
that total solvent use of ethanol was 39% of total consumption. Ethanol has
been placed in the category "other miscellaneous solvent use."
Trichl oroethyl ene. According to the SRI Handbook [ssL all consumption
of trichloroethylene was for solvent use, except for 10% which was exported.
Only 3% of the total cannot be assigned to a specific industrial process.
Toluene. Only 10% of total toluene consumption was in solvent use, accord
ing to SRI [38]- About half of this use was in the surface coatings industry.
A significant amount was also used in the printing and publishing industry, as
recorded in the M.S. A. Hydrocarbon Pollutants Report [37]. The remaining por-
tion of toluene used as a solvent was assigned to the printing and publishing
industry.
Acetone. According to the Chemical Marketing Reporter [39], 9% of total
acetone consumption was for surface coatings. The Marketing Guide to the
Chemical Industry [72] states that 25% of all acetone consumption was for
solvent use in 1971. Therefore, non-specific solvent use of acetone has been
placed at 16%.
E-3
-------
Xylene. The SRI Handbook [38] states that 13% of total consumption of
xylene in 1971 was used as solvent. The major application of xylene is as
a solvent in surface coatings. Since xylene and toluene are the two basic
aromatic hydrocarbon solvents used in the surface coating industry, total
toluene solvent use in the surface coating industry was subtracted from total
aromatic hydrocarbon consumption in the industry, as given in Figure E-l.
The remainder should be roughly the consumption of xylene in the industry.
Some small amounts of benzene (also an aromatic) are also used in the in-
dustry. However, the use of benzene is so small—due to its toxicity [73]--
that its exclusion here would not significantly affect the present xylene con-
sumption estimate.
Fluorocarbons. According to the Chemical Marketing Reporter, consumption
of fluorocarbons as solvents was 15% of total use in 1971, 10% of which can be
distributed to the rubber industry. The listed total use of fluorocarbons as
solvents at 65% reflects the use of fluorocarbons in manufacturing aerosols.
Insofar as the determination of organic solvent use is for the purpose of es-
timating area source emissions, no distinction is necessary between fluoro-
carbons released from aerosol cans or from solvent use. Therefore, fluoro-
carbons used in manufacturing aerosols are included in the inventory of solvent
use.
Methyl Ethyl Ketone (M.E.K.). The SRI Handbook [38] estimates that 65%
of all M.E.K. consumed in 1971 was for solvent application in the surface
coating industry. Another 7% was consumed as solvent for non-specified uses.
1,1,1-Trichloroethane. According to SRI [38], 65% of all 1,1,1-trichloro-
ethane consumed was used as a solvent in degreasing operations. Another 9%
was used in plastics and rubber production, and 11% was employed in other
non-specific solvent uses.
Methylene Chloride. The Chemical Marketing Reporter [39] states that 31%
of all methylene chloride consumed in 1971 was used as solvent in surface coat-
ing, 11% was used as solvent in degreasing operations, and 11% was used as sol-
vent in the rubber and plastics industry. Another 20% of the total consumption
E-4
-------
was used in the aerosol industry and was, therefore, included in the solvent
use industry. The remaining 27% was listed as miscellaneous use. Further
investigation [74 ] revealed that all of the miscellaneous category was sol-
vent use except for less than 10% which was exported. Total consumption of
methylene chloride for solvent use is, therefore, listed as greater than 90%.
Methanol. The Chemical Marketing Reporter [39] lists 9% of total methanol
consumption in 1971 for non-specified solvent use.
Ethylene Dichloride. According to the Chemical Marketing Reporter [39],
3% of all ethylene dichloride produced in 1971 was consumed as solvent in
surface coating, 2% was consumed in the rubber and plastics industry, and 2.7%
was employed in non-specified solvent use.
Ethyl Acetate. About 70% of the ethyl acetate consumption was for solvent
use in the surface coating industry [39]. The dry cleaning industry used 9% of
total consumption; 8% is attributed to the printing and publishing industry;
and 10% was used as solvent in the rubber and plastics industry.
Cyclohexane According to the SRI Handbook [38], 2% of all cyclohexane
consumed in 1971 was used as solvent in the surface coating industry. An ad-
ditional 2% was used as solvent in the rubber and plastics industry.
Methyl Isobutyl Ketone (M.I.B.K.). The Chemical Marketing Reporter [39]
estimates that 65% of all M.I.B.K. consumed in 1971 was used in the surface
coating industry. An additional 25% was consumed in other non-specified sol-
vent uses.
Other Organic Solvents. The category "other organic solvents" includes
the following chemicals: hexanes, benzene, N-butanol, nitrobenzene, turpentine,
isopropyl acetate, ethyl ether, monochlorobenzene, isopropanol, diethylene
glycol, methyl acetate, and cresols. Their total consumption in 1971 was
11,456 x 106 Ibs [27]. In 1968, about 5.2% of the total consumption of these
chemicals was-used as solvents [37]. Applying this to the 1971 consumption
total, a figure of 591 x 10 Ibs is derived for total solvent use in this
E-5
-------
category. This is less than 4% of the total of all solvent use covered in
this study.
In order to determine any change in the solvent usage pattern, the total
solvent use estimates of M.S.A. Research Corporation [37] for the year 1968
were compared with these estimates for the year 1971, as shown in Table E-l.
The category of "special naphthas" is 100% for both estimates because
both studies use the Bureau of Mines' definition for this category, which is
naphthas used as solvents. The two chemicals which show wide variations in
their usage p-attern between 1968 and 1971 are ethane and fluorocarbon. This
discrepancy is explained by the rapid increase of their use in the production
of aerosols [39]. in general, the current estimates seem to agree fairly
well with those of M.S.A. Research.
E-6
-------
TABLE E-l
PERCENT OF TOTAL CONSUMPTION
USED AS SOLVENT
Solvent
Special Naphthas
Perch loroethylene
Tri chl oroethyl ene
Ethanol
Toluene
Acetone
Xylene
Fluorocarbons
Methyl Ethyl Ketone
Methyl ene Chloride
Methanol
Ethyl Acetate
Cycle Hexane
Methyl Isobutyl Ketone
1 ,1 ,1-Trichloroethane
Ethyl ene Di chloride
All Other Solvents
M.S. A. Research, 1968
100%
89%
94%
25%
10%
30%
9%
42%
71%
93%
7%
94%
8%
83%
93%
5%
5.16%
Wai den, 1971
4
100%
83%
90%
39%
10%
25%
13%
65%
72%
>90%
9%
97%
4%
90%
87%
5%
5.16%
E-7
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-450/3-75-086
3. RECIPIENT'S XCCESSIOWNO.
4. TITLE AND SUBTITLE
Methodologies for Countywide Estimation of Coal,Gas,
and Organic Solvent Consumption
5. REPORT DATE
December 1975
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
Joseph P. Myers and Frank Benesh
8. PERFORMING ORGANIZATION REPORT NO,
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Waiden Research Division of Abcor, Inc.
201 Vassar Street
Cambridge, Massachusetts 02139
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
68-02-1410
12. SPONSORING AGENCY NAME AND ADDRESS
U. S. Environmental Protection Agency
Research Triangle Park, North Carolina 27711
13. TYPE OF REPORT AND PERIOD COVERED
Final Report-4/74 to 10/75
14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
16. ABSTRACT
Methods were developed to calculate the countywide consumption of anthracite
and bituminous coal, natural gas and liquid petroleum gas by residential, industrial
and commercial-institutional consumers. Methods were also developed to determine
countywide consumption of organic solvents, gasoline and diesel fuel use by off-
highway mobile sources, retail sales of gasoline.and aircraft landing and take-off
cycles. A procedure for estimating the sulfur and ash content of coal consumed
by county was also developed. The resulting data are reported in the National
Emissions Data System (NEDS) area source format. The report discusses the method-
ologies which were developed and presents an over-view of the computer processing
schemes used to produce NEDS area source data for 1973.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS C. COSATI Field/Group
Fuel consumption
Coal
Natural gas
Organic solvents
NEDS
Area source
13. DISTRIBUTION STATEMENT
Release unlimited
19. SECURITY CLASS (ThisReport}
Unclassified
21. NO. OF PAGES
20. SECURITY CLASS (This page)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
-------
INSTRUCTIONS
1. REPORT NUMBER
Insert the EPA report number as it appears on the cover of the publication.
2. LEAVE BLANK
3. RECIPIENTS ACCESSION NUMBER
Reserved for use by each report recipient.
4. TITLE AND SUBTITLE
Title should indicate clearly and briefly the subject coverage of the report, and be displayed prominently. Set subtitle, if used, in smaller
type or otherwise subordinate it to main title. When a report is prepared in more than one volume, repeat the primary title, add volume
number and include subtitle for the specific title.
5. REPORT DATE
Each report shall carry a date indicating at least month and year. Indicate the basis on which it was selected (e.g., date of issue, date of
approval, date of preparation, etc.).
6. PERFORMING ORGANIZATION CODE
Leave blank.
7. AUTHOR(S)
Give name(s) in conventional order (John R. Doe, J. Robert Doe, etc.). List author's affiliation if it differs from the performing organi-
zation.
8. PERFORMING ORGANIZATION REPORT NUMBER
Insert if performing organization wishes to assign this number.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Give name, street, city, state, and ZIP code. List no more than two levels of an organizational hirearchy.
10. PROGRAM ELEMENT NUMBER
Use the program element number under which the report was prepared. Subordinate numbers may be included in parentheses.
11. CONTRACT/GRANT NUMBER
Insert contract or grant number under which report was prepared.
12. SPONSORING AGENCY NAME AND ADDRESS
Include ZIP code.
13. TYPE OF REPORT AND PERIOD COVERED
Indicate interim final, etc., and if applicable, dates covered.
14. SPONSORING AGENCY CODE
Leave blank.
15. SUPPLEMENTARY NOTES
Enter information not included elsewhere but useful, such as: Prepared in cooperation with, Translation of, Presented at conference of,
To be published in, Supersedes, Supplements, etc.
16. ABSTRACT
Include a brief (200 words or less) factual summary of the most significant information contained in the report. If the report contains a
significant bibliography or literature survey, mention it here.
17. KEY WORDS AND DOCUMENT ANALYSIS
(a) DESCRIPTORS - Select from the Thesaurus of Engineering and Scientific Terms the proper authorized terms that identify the major
concept of the research and are sufficiently specific and precise to be used as index entries for cataloging.
(b) IDENTIFIERS AND OPEN-ENDED TERMS - Use identifiers for project names, code names, equipment designators, etc. Use open-
ended terms written in descriptor form for those subjects for which no descriptor exists.
(c) COSATI FIELD GROUP - Field and group assignments are to be taken from the 1965 COSATI Subject Category List. Since the ma-
jority of documents are multidisciplinary in nature, the Primary Field/Group assignment(s) will be specific discipline, area of human
endeavor, or type of physical object. The application(s) will be cross-referenced with secondary Field/Group assignments that will follow
the primary posting(s).
18. DISTRIBUTION STATEMENT
Denote releasability to the public or limitation for reasons other than security for example "Release Unlimited." Cite any availability to
the public, with address and price.
19. & 20. SECURITY CLASSIFICATION
DO NOT submit classified reports to the National Technical Information service.
21. NUMBER OF PAGES
Insert the total number of pages, including this one and unnumbered pages, but exclude distribution list, if any.
22. PRICE
Insert the price set by the National Technical Information Service or the Government Printing Office, if known.
EPA Form 2220-1 (9-73) (Reverse)
------- |